Chinese Journal of Chemical Engineering 40 (2021) 179–186

Contents lists available at ScienceDirect

Chinese Journal of Chemical Engineering

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

Full Length Article

Feasibility analysis and process simulation of CO2 dehydration using

triethylene glycol for CO2 pipeline transportation
Shupanxiang Chen, Guangying Chen, Xiaolan Zhao, Xiao Luo ⇑, Hongxia Gao, Wensheng Li, Zhiwu Liang ⇑
Joint International Center for CO2 Capture and Storage (iCCS), Provincial Hunan Key Laboratory for Cost-effective Utilization of Fossil Fuel Aimed at Reducing
Carbon-dioxide Emissions, Department of Chemical Engineering, College of Chemistry and Chemical Engineering, Hunan University, Changsha 410082, China

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

Article history: The operation of dehydration is very important in the process of gas transportation. This study aims to
Received 14 October 2020 evaluate the application feasibility of CO2 dehydration using triethylene glycol, which is also called
Received in revised form 25 November 2020 TEG for short. Aspen Plus software was used to simulate the dehydration process system of CO2 gas trans-
Accepted 5 December 2020
portation using TEG dehydration. Parameter analysis and process improvement were carried out for the
Available online 23 January 2021
simulation of dehydration process. At first, a sensitivity analysis was conducted to analyze and optimize
operating conditions of conventional CO2-TEG dehydration process system. Subsequently, a recycle unit
Keywords:
was introduced into the conventional CO2-TEG dehydration process system, it can be found that the
Carbon dioxide
Dehydration
improved process system with the recycle unit has a higher CO2 recovery rate which was about 9.8% than
Simulation the conventional one. Moreover, the improved process system showed excellent operation stability
Triethylene glycol through the comparison of simulation results of several processes with various water contents in their
Process system feed gases. Although the energy consumption is increased by about 2%, the improved process was eco-
nomically and technically feasible for the long-term availability of CO2 pipeline transportation. The sim-
ulated results showed that the improved CO2-TEG process system has promising application prospects in
CO2 dehydration of CO2 gas transportation with high stability.
Ó 2021 The Chemical Industry and Engineering Society of China, and Chemical Industry Press Co., Ltd. All
rights reserved.

1. Introduction pose, the range of water concentration in the transported CO2

stream can be smoothly larger, which could be no more than
Pipeline transportation of carbon dioxide (CO2) plays an impor- 500 mg
L 1 [5]. According to the research result of Abbas et al.
tant role in the deployment of carbon dioxide capture and storage [6], a water content of 50 mg
L 1 was considered as the indication
(CCS) with large transportation volume, little environmental of full dehydration for the overall range of applications. And this
impact, low operational costs, enclosed conveying and low product content is also regarded as the restricted target in this study.
losses [1,2]. It allows only extremely low amount of water in the Up to now, there are several commonly used dehydration meth-
CO2 fluid in case the formation of hydrate and free water, which ods for acid gas stream, including compression and cooling, solid
can plug valves and fittings in the process of CO2 pipeline trans- adsorption, absorption, and so on [7–9]. Among them, compression
portation and also react with CO2 to cause electrochemical corro- and cooling is one of the most widely used methods since it can
sion. Therefore, it is urgently needed to find proper approaches compress and cool the gases at the same time. However, it is not
to remove water from the CO2 gas fluid to be transported. suitable for deep dehydration [6]. Solid adsorption dehydration
Up to now, there is no standard criterion to accurately define using adsorbent, such as silica gel, molecular sieve, activated alu-
the allowable water content in the CO2 gas stream to be trans- mina, activated carbon, etc., can be applied to remove water vapor
ported. For different application purpose, the extent of dehydration under various temperature, pressure and flow rate conditions,
is different. For the CO2 stream being transported from the CCS which is also a mature enough method. However, considering the
projects, the water content is usually lower than 50 mg
L 1 [3]. economic and technical benefits, the absorption via solvent is the
When the CO2 stream is used for enhanced oil recovery (EOR), its most commonly adopted method [10]. Up to now, it is also consid-
limitation is also 50 mg
L 1 [4]. While for geological storage pur- ered to be the most attractive method to dehydrating water deeply
from almost pure gas fluid [11].
⇑ Corresponding authors at: Fax: +86 731 88573033 (Z. Liang). Glycol is one kind of good dehydrants for gas dehydration due
E-mail addresses: x_luo@hnu.edu.cn (X. Luo), zwliang@hnu.edu.cn (Z. Liang). to its molecular structure and physical property. The molecular

https://doi.org/10.1016/j.cjche.2020.12.025
1004-9541/Ó 2021 The Chemical Industry and Engineering Society of China, and Chemical Industry Press Co., Ltd. All rights reserved.
180 S. Chen et al. / Chinese Journal of Chemical Engineering 40 (2021) 179–186

structure of glycol contains hydroxyl and ether bond, which can rations or optimize operating conditions in the attempt to improv-
form hydrogen bond with water and perform a strong affinity to ing operating performance and minimizing energy consumption
water so that it has a high dehydration depth. TEG, which dehy- [16,19,21,23,24]. Aspen Plus has found widespread application in
drates water by means of its hydrophilicity, is thermal-stable and the simulation of absorption/regeneration process and therefore
easy to regenerate, and has relatively low vapor pressure and high can be used for CO2 dehydration owing to its rich database of prop-
moisture-absorption performance [12,13]. It has been widely used erty parameters (contains Aspen CD and DIPPR) and comprehen-
for various gases dehydration, especially for natural gas (NG). sive unit operation modules. Moreover, Aspen Plus can be used
Grosso [14] studied the gas dehydration process using the glycol to perform sensitivity analysis to evaluate the effects of key param-
solvent, and discovered that glycol is the good choice for gas dehy- eters on the CO2 dehydration performance, and finally, to optimize
dration due to its molecular structure and physical property. Gand- operational conditions.
hidasan [15] dealt with the design analysis of NG-TEG dehydration In view of the above mentioned issues, this paper attempts to
process and briefly discussed the effects of various operating evaluate the potential of TEG in CO2 dehydration by simulating
parameters (such as pressure, temperature, and TEG circulation its performance in two process configurations, namely the conven-
rate) on the design of this dehydration process. Piemonte et al. tional process and the improved process. Both process configura-
[16] studied the TEG regeneration process in NG dehydration tions were simulated using Aspen Plus simulation software.
plants to meet the regeneration level of TEG (98.5%–99.0%) through Moreover, sensitivity analysis was performed to investigate the
the measurements of boiling temperatures for binary mixtures of effect of several key parameters on the CO2 dehydration efficiency
TEG and water and the simulation of a natural gas dehydration in both process configurations and finally to obtain optimized
process using the fitted NRTL parameters in the Hysys process sim- operating conditions. And then, the feed gas streams with different
ulator. Besides, some mathematical methods, such as advanced water content were dehydrated using the improved process to
equation of state, correlation and artificial neural network model, evaluate the dehydration performance of TEG solvent and to vali-
were developed to analyze and simulate the whole NG-TEG dehy- date the stability of the improved CO2-TEG dehydration process.
dration process or optimize the key operational parameters
[13,17–19].
2. Process Descriptions
When it comes to CO2 dehydration, firstly, its process is so sim-
ilar with that of natural gas; secondly, impurities in the highly con-
2.1. The dehydration mechanism of TEG solvent
centrated CO2 gas stream from CO2 capture unit and that in the
natural gas stream are also similar; and finally, even though CO2
Due to the existence of hydroxy and ether bond in the TEG
is more soluble in TEG than methane (CH4), the amount of dissolved
molecular structure, water molecules can be fixed on TEG mole-
CO2 accounts for quite a small proportion of total CO2. In view of the
cules through the intermolecular hydrogen bonds. TEG has a strong
reasons mentioned above, TEG is expected to be suitable and has a
affinity for water, and therefore possesses a high dehydration
great potential for CO2 dehydration. As well known, evaluation of
depth. Moreover, TEG is easy to regenerate, and has a large dew
TEG performance in CO2 dehydration requires comprehensive sim-
point depression and relative low operation cost, which makes it
ulation at various operating conditions. Yet so far, there are only
widely used in the field of gas dehydration.
several references in the open literature about the simulation of
CO2 dehydration using TEG. Grynia et al. [20] have investigated,
by means of two methods (i.e., by compression and cooling alone 2.2. Feed gas and product specifications
and by TEG dehydration), the dehydration process of an acid gas
mixture including H2S, CO2, and a small amounts of other compo- In general, the CO2 stream to be transported is rarely pure and
nents. They discussed the design considerations of application for always has some contaminants such as N2, H2O and hydrocarbons.
both methods and concluded that TEG is very feasible for CO2 dehy- These impurities may influence or even hinder the cost-effective
dration. This concept was also accepted by Abbas et al. [6]. In their transportation of CO2 stream when their amount over a certain
research, they mentioned that TEG can be used as dehydrant for CO2 specification, especially the H2O [25]. Therefore, reducing H2O con-
dehydration with a few modification in the process. Moreover, in tent to a certain value is the primary work for CO2 gas purification.
2014, Øi and Fazlagic [21] applied Aspen HYSYS to consider the In this study, the specifications of various components in the CO2
application possibility of a traditional glycol dehydration process gas to be deeply dehydrated, as shown in Table 1, are based on
and more advanced CO2 dehydration processes for CO2 dehydration the stream specifications of CO2 product from post-combustion
process using different equilibrium models. Their simulated results capture technologies proposed by Abbas et al. [6]. Post-
demonstrated that TEG is a reasonable alternative for CO2 dehydra- combustion capture is an energy intensive and expensive process
tion to reduce water levels lower than 5 mg
L 1. and it can be done using a variety of methods, including chemical
As it is well known, the water removal efficiency of NG-TEG solvents, sorbents, membranes and distillation. Here, CO2 is
dehydration process is mainly influenced by absorber temperature, removed from flue gases of power plants or from other fossil
absorber pressure, stage number of absorber (or height of packing), fuel-based large point sources [26,27]. And for the CO2 pipeline
TEG circulation rate, TEG purity in the lean TEG solution, etc. As
similar with it, in this study, these parameters are also considered
as the investigation objects of CO2-TEG dehydration process to look Table 1
The specifications of CO2 stream compositions from post-com-
for the optimum operation parameter values. Besides, due to the
bustion capture technologies [6]
high mutual solubility of TEG and CO2, the loss of TEG solvent
and CO2 increase, which result in larger TEG makeup rate and Specifications/%
CO2 92–97
higher reboiler duty and costs [22]. Consider of this, the loss of
H2O 2.8–7.3 [28]
TEG solvent and CO2 also need to be analyzed, and some improve- N2 0.02–0.13
ments need to be done in the CO2-TEG dehydration process to Ar 0.00001–0.000025
reduce the losses and reboiler duty. O2 0.001–0.03
In the aspects of CO2 transportation and dehydration, a number SO2 0.001
Gas feeding temperature/°C 40–42
of process commercial software, such as Aspen Plus, Aspen HYSYS, Gas feeding pressure/MPa 0.1–0.285
Pro/II and PROSIM, have been applied to simulate process configu-
 S. Chen et al. / Chinese Journal of Chemical Engineering 40 (2021) 179–186 181

transportation purpose, the constrained limit of H2O in CO2 stream that most of the dissolved CO2 in the H2O-rich TEG stream is
is usually 50 mg
L 1, therefore it is also be the dehydration crite- flashed out through a flash tank. Since the majority of the flash
rion used in this work. This study assumes the total gas stream gas is CO2, if it is emitted to the atmosphere, there will be a loss
flow rate to be 3  104 kmol
h 1, and the CO2 recovery rate is great of CO2 in the dry gas stream and the total CO2 recovery rate would
than 98%. Based on these assumptions, a series of process simula- reduce. On the contrary, if the flash gas is recycled directly with the
tions about CO2 gas streams with different water content, which main dry gas (out of the top of the absorber), the final CO2 dehy-
are within the range of 2.8%–7.3%, are conducted to evaluate the dration efficiency will be decreased due to the relative high mois-
dehydration efficiency of TEG solvent and to validate the stability ture content in the flash gas. Given this, we consider to recycle the
of the improved CO2-TEG dehydration process. flash gas (which contains some water) back to the dehydration
process for further simulation. Fig. 2 showed the improved CO2-
2.3. Process configuration of the CO2-TEG dehydration process TEG dehydration process with the recycle unit. As it was shown
in Fig. 2, the flash gas is recycled back to the mixer, and mix with
The basic CO2-TEG dehydration process in this work consists fresh feed gas to form a new gas stream (Stream 1). Then the new
mainly of two sections: a dehydration unit and a regeneration unit, gas stream is introduced at the bottom of the absorber. Recycling
as shown in Fig. 1. Detailed description about the dehydration and CO2 in the flash gas stream would results in a significant reduction
regeneration operational unities presented as follows: of CO2 loss, and increases the total CO2 recovery rate as a
Dehydration: The feed CO2 gas stream is introduced at the bot- consequence.
tom of the absorber, while the TEG solvent is injected at the top of Table 2 shows the thermodynamic model and model setting
the absorber. Through the countercurrent contact of feed wet gas parameters of the improved CO2-TEG dehydration process.
and TEG solvent in the absorber, water is removed from the gas
stream. Then the water-rich TEG stream drains out from the bot-
tom of absorber, and the dried CO2 gas ejects from the top of the 3. Results and Discussion
absorber. After that, the H2O-rich TEG stream is introduced into a
flash tank to flash off some of the dissolved CO2 before it heated In this work, Aspen Plus was applied to simulate the CO2-TEG
and sent to the regenerator. dehydration process, so as to evaluate the application feasibility
Regeneration: The H2O-rich TEG stream (from the flash tank) is of CO2 dehydration using TEG. At first, we conducted the dehydra-
preheated in a cross-flow heat exchanger by the lean TEG (returned tion process simulation of one CO2 gas stream (which is composed
from the bottom of the reboiler) to a desired temperature before of 94.9% CO2, 5 % H2O, and 0.1% other infinitesimal impurities)
being pumped into the top of the regenerator. In the regenerator, using the basic CO2-TEG dehydration process. And sensitivity anal-
absorbed water is stripped out from the TEG solvent by elevating ysis was performed to evaluate the effects of several key opera-
temperature under atmospheric pressure. After the H2O stripping tional parameters on the dehydration performance. These
step, the already-regenerated lean TEG is cooled by the lean/rich operational parameters include absorber pressure, absorber tem-
heat exchanger first and a lean cooler second, and then pumped perature, stage number of absorber, TEG circulation rate and reboi-
back to the absorber for recycling. ler temperature. Secondly, the performance of the improved CO2-
TEG dehydration process with the recycle unit was evaluated and
2.4. Process configuration of the improved CO2 -TEG dehydration with compared with the basic one using the same gas stream composi-
the recycle unit tion. Finally, a range of process simulations were done to dehy-
drate CO2 gas streams with different water content, i.e. 3%, 4%,
Since TEG is a physical solvent for CO2 absorption, some amount 5%, 6% and 7%, respectively, and to evaluate the dehydration effi-
of CO2 will inevitably dissolve into the TEG during the absorption ciency of TEG solvent and to validate the stability of the improved
process, especially at higher pressures. It can be seen from Fig. 1 CO2-TEG dehydration process.

Water

Condencer
Lean cooler Fresh TEG
Recycle
Absorber
Regenerator
Flash gas
Feed gas

Pump
Rich TEG
Flash tank

Rebolier
Rich TEG
Lean/rich heat
exchanger

Lean TEG

Fig. 1. Process flowsheet of the CO2-TEG dehydration process.

182 S. Chen et al. / Chinese Journal of Chemical Engineering 40 (2021) 179–186

Water

Lean cooler Fresh TEG Condencer

Recycle
Absorber

Feed gas Regenerator

Stream 1
Mixer
Flash gas Pump

Rich TEG
Flash tank
Rebolier
Pump 2 Rich TEG Lean/rich heat
exchanger

Heater
Lean TEG

Fig. 2. Process flowsheet of the improved CO2-TEG dehydration process with the recycle unit.

Table 2
The model setting parameters

Model name Property method Model Temperature/℃ Pressure/MPa

Heater PENG-ROB Heater 40 2
Lean cooler PENG-ROB Heater 40 2
Heat exchanger PENG-ROB Heater 40 0.01
Pump PENG-ROB Pump 40 2
Pump 2 PENG-ROB Pump 50 2
Flash tank PENG-ROB Flash 2 50 0.1
Model name Property method Model Number of stages Condenser pressure
Absorber PENG-ROB RadFrac 7 2
Regenerator PENG-ROB RadFrac 4 0.01

3.1. Simulation and optimization of the basic CO2-TEG dehydration pressure is greater than 2.0 MPa, the change range of water content
process in dry gas becomes very small. Even though increasing the absor-
ber pressure would improve the dehydration performance, it will
increase the operational cost as well. As mentioned in Section 2.2,
CO2 can dissolve into TEG solvent, and the solubility of CO2 in TEG
3.1.1. Effect of absorber pressure on the CO2-TEG dehydration at higher pressure can’t be ignored. Therefore, the CO2 loss during
efficiency and CO2 loss rate the absorption process needs to be discussed. It can also be seen
Absorber pressure has significant effect on the CO2-TEG dehy- clearly from Fig. 3 that CO2 loss rate increases with the increase
dration efficiency and CO2 loss rate. As can be seen in Fig. 3, with of absorption pressure, which means that a higher pressure may
the increase of absorber pressure, water content in dry gas first lead to a considerable amount of CO2 loss. Consider of these discus-
decreases dramatically and then becomes horizontal. When the sion above and in order to meet the product specifications that
water content in dry gas considerably below 50 mg
L 1, the opti-
mum pressure is recommended at 2.0 MPa.
10
600 Water content
Water content in dry CO2/mg·L-1

9
CO2 loss rate 3.1.2. Effect of absorber temperature on the CO2-TEG dehydration
500 8 efficiency and CO2 loss rate
Absorber temperature is an important factor for the CO2-TEG
CO2 loss rate/%

400 7
dehydration process, since it can affect the solubility of H2O in
6 TEG and the fluidity of TEG in the absorber (due to the change of
300
5 TEG viscosity). However, the viscosity of TEG is high at low tem-
200 perature, which may worsen the mobility and distribution of TEG
4
in the absorber and even block pipelines. Therefore, it is important
100 3 to select an appropriate absorber temperature.
The effects of absorber temperature on the dehydration effi-
2
0 ciency and CO2 loss rate are shown in Fig. 4, in which the water
1 content in the dry gas increases monotonously with the rise of
1.0 1.5 2.0 2.5 3.0 3.5 4.0
absorber temperature while the CO2 loss rate changes slightly dur-
Absorber operating pressure/MPa ing this temperature range (CO2 loss rate varied from 3.98% to
Fig. 3. Effect of absorber pressure on the dehydration efficiency and CO2 loss rate
4.12%). It can be seen from Fig. 4 that water content is less than
(absorber temperature: 40 °C; TEG circulation rate: 2500 kmol
h 1; stage number of 50 mg
L 1 at 40 °C, and CO2 loss rate is about 4.02%; therefore, this
absorber: 5). temperature is quite reasonable for the absorption process.
 S. Chen et al. / Chinese Journal of Chemical Engineering 40 (2021) 179–186 183

90 4.14 Considering above, the optimal stage number of absorber is 7 in

this process.
Water content in dry CO2/mg·L-1

80 Water content 4.12

CO2 loss rate
70 4.10
3.1.4. Effect of TEG circulation rate on the CO2-TEG dehydration

CO2 loss rate/%

4.08 efficiency and reboiler duty
60
4.06 TEG circulation rate relates closely to the dehydration degree of
50 TEG solvent and energy consumption for heating and pumping.
4.04 The increase of TEG circulation rate is beneficial to the dehydration
40
4.02 performance. Fig. 6 exhibits the effect of TEG circulation rate on the
30 dehydration efficiency and reboiler duty. As shown in this figure,
4.00
water content in dry gas decreases with the increase of TEG circu-
20 3.98 lation rate, indicating that increasing the TEG circulation rate is
10 3.96 beneficial to the absorption of moisture and then improving the
10 20 30 40 50 60 TEG dehydration efficiency; while as the TEG circulation rate
TEG feed temperature/ exceeds 2250 kmol
h 1, the change trend gradually slows down,
and the amount of water be removed increases inconspicuously.
Fig. 4. Effect of absorber temperature on the dehydration efficiency and CO2 loss
rate (absorber pressure: 2.0 MPa; TEG circulation rate: 2500 kmol
h 1; stage
However, a high TEG circulation rate requires a corresponding
number of absorber: 5). large energy input to heat the H2O-rich solvent to the proper tem-
perature for TEG regeneration, which will raise the operational
cost. The required energy consumption of reboiler for each TEG cir-
Besides, since the feed gas temperature after the CO2 capture pro- culation rate is also displayed in Fig. 6. It can be seen clearly that
cess is about 40 °C (as shown in Table 1), keep the dehydration more energy is required to obtain a higher TEG circulation rate,
temperature at the same level could make an energy balance and and their line is linear and steep. According to the simulation
achieve a high dehydration efficiency. Based on these discussions, results and the desired dehydration degree, the most economic
the optimum absorber temperature is 40 °C. TEG circulation rate is 2500 kmol
h 1.

3.1.3. Effect of stage number of absorber on the CO2-TEG dehydration 3.1.5. Effect of the temperature of reboiler on the CO2-TEG dehydration
efficiency and CO2 loss rate efficiency and purity of the regenerated TEG
The stage number of absorber, which directly related to the col- The temperature of reboiler has a great effect on the regenera-
umn height, has a significant effect on the CO2-TEG dehydration tion performance, the higher the reboiler temperature, the better
performance. Fig. 5 shows the investigation of the effect of stage the TEG regeneration efficiency, and therefore, the better the
number of absorber on the water content in the dry gas and CO2 TEG-CO2 dehydration efficiency. As it is known that TEG has a ther-
loss rate. It can be seen that the water content in dry gas decreases mal decomposition temperature of 206.7 °C [29], therefore, the
as the increase of absorber stage number. This is because increas- temperature of reboiler could not be more than 204 °C. Fig. 7
ing the number of stages leads to the increases of both contacting shows the changes of the water content in dry CO2 gas and the pur-
area and residence time for gas and liquid phases. Both these ity of the regenerated TEG (i.e. the purity of lean TEG solution) with
effects result in the full usage of the TEG solvent and thus requiring the increase of reboiler temperature. As can be seen from Fig. 7, the
less amount of TEG solvent. With respect to the loss rate of CO2 gas, water content in dry CO2 gas decreases with the increase of the
it varies quite slight with the addition of absorber stage number temperature of reboiler, and the purity of TEG increases accord-
(which is about 4%). However, with the increase of the stage num- ingly, which is agreement with the fact that the higher the purity
ber of absorber, the facility and operational costs increase accord- of TEG lean solution lead to a higher dehydration efficiency for
ingly. Therefore, the best choice is using less number of stages to wet feed gas. Consider of this, in order to get a desirable TEG purity
achieve the desired dehydration efficiency (i.e. the mole fraction (usually 99.9%) in continuous operation process, the temperature
of water in dry gas less than 50 mg
L 1). As shown in Fig. 5, water of reboiler should be as high as possible. However, it is generally
content in dry gas is less than 50 mg
L 1 at a seven-stage absorber. known that the higher the temperature of reboiler, the higher

350 100
4.06 60000
Water content in dry CO2/mg·L-1

Water content
Water content in dry CO2/mg·L-1

300 90 Heat duty 55000

4.04
250 80
CO2 loss rate/%

50000
Heat duty/kW

200 4.02
70 45000
150 Water content
4.00 60 40000
CO2 loss rate
100 35000
3.98 50
50 30000
3.96 40
0 25000
4 5 6 7 8 9 10 1500 2000 2500 3000 3500
The number of stages TEG circulation rate kmol/h
Fig. 5. Effect of stage number of absorber on the dehydration efficiency and CO2 Fig. 6. Effect of TEG circulation rate on the dehydration efficiency and reboiler duty
loss rate (absorber pressure: 2.0 MPa; absorber temperature: 40 °C; TEG circulation (absorber pressure: 2.0 MPa; absorber temperature: 40 °C; stage number of
rate: 2500 kmol
h 1). absorber: 7).
184 S. Chen et al. / Chinese Journal of Chemical Engineering 40 (2021) 179–186

pressure and flash temperature on the CO2 recovery rate and CO2
120 Water content 100.00
Water content in dry CO2/mg·L-1

purity in flash gas, and (ii) the analysis and comparison of total
Purity
energy consumptions of the improved process and the basic one.
100

The purity of TEG/%

99.95

80 3.2.1. Effect of the flash pressure and temperature on the CO2 recovery
99.90
rate and CO2 purity in flash gas
60 For flash tank, its main role is to provide a space for the high
99.85
pressure fluid to rapid evaporation and realize gas–liquid separa-
40 tion, and this separation can be achieved by reducing the flash
99.80 pressure or increasing the flash temperature. Therefore, this sec-
20 tion is to search for the most suitable flash operation conditions
99.75 to get the objective goal. The effects of the flash pressure and tem-
0 perature on the CO2 recovery rate and CO2 purity in flash gas are
185 190 195 200 205
Reboiler temperature/ shown in Fig. 8. As can be seen from it: (i) CO2 recovery rate
increases with the increase of flash temperature and the decrease
Fig. 7. Effect of the temperature of reboiler on the dehydration efficiency and purity of flash pressure, while the purity of CO2 reduces as the rise of flash
of the regenerated TEG (absorber pressure: 2.0 MPa; absorber temperature: 40 °C; temperature and the decline of flash pressure, and (ii) at lower
stage number of absorber: 7; TEG circulation rate: 2500 kmol
h 1).
pressure (e.g. 0.1 MPa), with the increase of flash temperature,
CO2 recovery rate is relative large and increases gently, while the
purity of CO2 reduces largely; at larger pressure (such as 1 MPa),
the energy consumption accordingly. Thus, it is important to select
the results are reversed. These results are mainly because of the
a fit temperature in order to get the optimal TEG purity and meet
fact that the boiling point of a fluid is increased with the increment
the requirement of low-energy. According to the specification of
of pressure, when the pressure reduces, fluid evaporates rapidly
the purity of TEG and the desired dehydration degree, the suitable
with the decrease of its boiling point. At this status, the dissolved
temperature of reboiler should be 200 °C.
substance could transform from a saturated or unsaturated state
into a super-saturation state, and then separated from the main
3.1.6. Overview of the conventional CO2-TEG dehydration process body of fluid. For CO2-TEG-H2O system (i.e. the H2O-rich TEG
Based on the sensitivity analysis mentioned above, the appro- stream), that is to say, a high CO2 recovery rate can be achieved
priate operational parameters of the conventional dehydration at a low flash pressure (as shown in Fig. 8). And it can be deduced
process are obtained. These parameters and their values are pre- that pressure is almost no effect on the solubility of water in TEG,
sented in Table 3. After the CO2-TEG dehydration process, the namely that little amount of water will go into the flash gas, which
water content in dry gas is 47.3 mg
L 1, the CO2 recovery rate lead to the high CO2 purity in the flash gas. On the other hand, the
and CO2 purity in the dry gas is about 95.97% and 99.87%, respec- increase of flash temperature could enhance the evaporation of
tively. And this process can achieve a very high TEG purity in the CO2 since it offers more energy for the CO2-TEG-H2O fluid to form
recovered lean TEG solution (which is 99.93%) with a reboiler duty a new gas–liquid equilibrium to evaporate more CO2. But due to
of 5.53104 kW. It can be seen from Table 3 that the total dehydra- the temperature dependence of solubility of H2O in TEG, more
tion performance of TEG is quite good except for a little high CO2 water will go out of the rich-H2O TEG stream to the gas phase,
loss rate of 4%. which makes an obvious decrease in the CO2 purity in the flash
gas. As shown in Fig. 8, at a certain flash pressure, there is a balance
3.2. Simulation and optimization of the improved CO2-TEG point between CO2 recovery rate and CO2 purity. After correlating
dehydration process with the recycle unit the corresponding curves of each flash pressure, we can get the
optimal flash operation conditions. And at a pressure of 0.1 MPa
Since the problem of low CO2 recovery rate exists in the basic and a temperature of 13 °C, both the CO2 recovery rate andCO2 pur-
CO2-TEG dehydration process (more than 4% of CO2 is lost), the ity can achieve 94.5%, which is similar with the water content in
improved CO2-TEG dehydration process increased the recycle unit,
and the flashed gas goes into the absorber for recycling aim to
increase the CO2 recovery rate. It is generally accepted that CO2
recovery rate should be great than 98%. Therefore, the operational 1.0
CO2 recovery rate and purity

parameters (i.e. flash pressure and flash temperature) of flash tank

have great influences on the CO2 recovery rate and CO2 purity in 0.8
the flash gas. So this part mainly discusses: (i) the effect of flash

0.6
Table 3
The optimized operational parameters of the conventional CO2-TEG dehydration The purity of CO2/0.001MPa
process 0.4 CO2 recovery rate/0.001MPa
Item Parameter values The purity of CO2/0.005MPa
Pressure of absorber/MPa 2.0
0.2 CO2 recovery rate/0.005MPa
Absorber temperature/°C 40
The purity of CO2/0.01MPa
Stage number of absorber 7
TEG circulation rate/kmol
h 1 2500
0.0 CO2 recovery rate/0.01MPa
Reboiler temperature/°C 200
1
The water content in dry gas/ mg
L 47.3 0 20 40 60 80 100
CO2 recovery rate/% 95.97 Flash drum temperature/
Purity of the dry CO2 gas/% 99.87
Purity of the lean TEG solution/% 99.93
Fig. 8. Effect of the flash pressure and temperature on the CO2 recovery rate and
Reboiler duty/kW 5.53104
CO2 purity in flash gas.
 S. Chen et al. / Chinese Journal of Chemical Engineering 40 (2021) 179–186 185

Table 4
Simulation results of the improved CO2-TEG dehydration process at various water contents

Parameter Water content in the feed gas/%

3 4 5 6 7
Water content in dry gas/mg
L 1 59.2 53.2 47.3 42.4 38.4
Total CO2 recovery rate/% 99.79 99.78 99.78 99.77 99.76
Purity of the dry CO2 gas/% 99.88 99.88 99.87 99.87 99.87
Purity of the lean TEG solution/% 99.93 99.93 99.93% 99.93 99.93
Reboiler duty/kW 4.97104 5.25104 5.53104 5.82104 6.11104
Energy consumption of the recycle unit/kW 670.7 791.3 895.4 984 1058.4

the feed gas. And the total CO2 recovery rate in dry gas can be lations are conducted at the same operation conditions, in other
increased to 99.78%. Therefore, the optimal flash conditions are words, the operating parameters of absorber, regenerator and recy-
0.1 MPa and 13 °C, respectively. cle units remain unchanged at every process simulation. Based on
these preconditions, the final simulation results were obtained (as
3.2.2. The analysis and energy consumption calculation of the shown in Table 4).
improved process From Table 4, it can be seen that: (i) at the same operation con-
It should be noted that in the improved process, the recycled ditions, water content in the dry gas decreases with the increase of
CO2 gas from the flash tank is mixed with the fresh wet gas in water content in the feed gas, but all of them have relative good
the mixer directly and then be injected into the absorber. For this dehydration performances (all their dehydration efficiencies are
unit, there are two questions should be pay attention to: (i) when around 50 mg
L 1); (ii) all of these five processes have high CO2
mixing the flashed CO2 gas with the fresh one, due to the difference recovery rates and CO2 gas purities near 99.8%; (iii) the purities
in their inlet temperatures, some additional heat is needed to of TEG in lean solution after regenerating are greater than 99.9%;
heating the flashed CO2 gas to the feed gas inlet temperature (iv) the energy consumption (including the reboiler duty and the
(i.e. heated from 13 °C to 40 °C); (ii) since there is a difference of energy for CO2 recycle) of each process is different with the vary
water content between the feed gas and the flashed CO2 gas, it of water content in the feed gas and increases with the rise of
may have an effect on the composition of the gas enter into absor- water content. All these results indicated the excellent dehydration
ber (i.e. Stream 1). For the first issue, the energy consumption of performance of our improved process, since the final dehydration
the recycle unit is calculated (i.e. 895.4 kW), which is pretty small efficiency basically unchanged along with the fluctuation of water
compared to the reboiler duty of regenerator and can be neglected. contents in the feed gas, and the increments of the total energy
For the second question, the difference of water content in the feed consumptions are not too much. Based on the process simulation
gas and the flashed CO2 gas is slight (5% and 5.5%, respectively), and results discussion, it can be concluded that the improved pro-
and the water content in the final mixing gas is 5.02% (which is cess is quite suitable for the dehydration of CO2 gas using TEG
almost 5%), therefore, water contents in the feed gas basically dehydration with high stability.
remain unchanged with or without the recycle unit, and the
improved dehydration process can be considered as steady 4. Conclusions
simulation.
In this paper, Aspen plus software was used to simulate the
3.2.3. Overview of the improved process dehydration process of CO2 gas using TEG dehydration. Two dehy-
In this work, the improved CO2-TEG dehydration process with dration processes were conducted for this purpose, include the
the recycle unit is simulated. Through the sensitivity analysis, the conventional CO2-TEG dehydration process and the improved
optimal operating parameters were obtained: the temperature and CO2-TEG dehydration with the recycle unit. Through the simula-
pressure of flash tank are13 °C and 0.1 MPa, respectively. The sim- tion and technical analysis of these two processes, the optimal val-
ulation results showed that the final CO2 recovery rate is 99.78%. ues of several main operating parameters of the CO2-TEG
Compared with the conventional process, the improved CO2-TEG dehydration process were obtained. And it can be found that the
dehydration has a higher CO2recovery rate (increase by 3.81%), improved process with the recycle unit has a higher CO2 recovery
which is beneficial for the greenhouse effect in the long run. It also rate (about 99.8%) than the conventional one. Moreover, the
should be noted that, due to the recycle of flashed CO2 gas, some improved process showed excellent operation stability through
more energy is need to mixing it with the fresh feed gas. However, the comparison of simulation results of several processes with var-
the energy consumption is quite small and has slight effect on the ious water contents in their feed gases. Although the energy con-
total energy consumption of the improved process. Therefore, it sumption is increased (about 2%), the improved process was
can be said that the improved process is better than the basic economically and technically feasible for the long-term availability
one with more CO2 recovery rate and almost the same energy of CO2 pipeline transportation.
consumption.
Declaration of Competing Interest
3.3. Simulation and performance evaluation of the improved CO2-TEG
dehydration process with various water contents The authors declare that they have no known competing finan-
cial interests or personal relationships that could have appeared
It can be known from Table 1 that water contents in the cap- to influence the work reported in this paper.
tured CO2 stream are usually vary from 2.8% to 7.3%. Therefore,
in order to get a more clear understanding about the performance Acknowledgements
and stability of the improved CO2-TEG dehydration process, a ser-
ies of processes with various water content (i.e. 3%, 4%, 5%, 6% and This work was supported by the National Natural Science Foun-
7%, respectively) were simulated and analyzed. And in order to dation of China (21536003, 21776065 and 21978075) and the Nat-
maintain the consistency of each simulation, all the process simu- ural Science Foundation of Hunan Province in China (2019JJ20006).
186 S. Chen et al. / Chinese Journal of Chemical Engineering 40 (2021) 179–186

References [15] P. Gandhidasan, Parametric analysis of natural gas dehydration by a

triethylene glycol solution, Energy Sources 25 (2003) 189–201.
[16] V. Piemonte, M. Maschietti, F. Gironi, A triethylene glycol-water system: a
[1] M. Anheden, A. Andersson, C. Bernstone, S. Eriksson, J. Yan, S. Liljemark, C.
study of the TEG regeneration processes in natural gas dehydration plants,
Wall, CO2 quality requirement for a system with CO2 capture, transport and
Energy Sources Part A: Recov. Utiliz. Environ. Effects 34 (2012) 456–464.
storage, Greenhouse Gas Control Technol. 7 (2005) 2559–2564.
[17] A. Bahadori, H.B. Vuthaluru, Rapid estimation of equilibrium water dew point
[2] L. Shan, H. Li, B. Meng, Y. Yu, Y. Min, Improvement of CO2 capture performance
of natural gas in TEG dehydration systems, J. Nat. Gas Sci. Eng. 1 (2009) 68–71.
of calcium-based absorbent modified with palygorskite, Chin. J. Chem. Eng. 24
[18] A. Bahadori, H.B. Vuthaluru, Simple methodology for sizing of absorbers for
(2016) 1283–1289.
TEG (triethylene glycol) gas dehydration systems, Energy 34 (2009) 1910–
[3] S.M. Forbes, P. Verma, T.E. Curry, S.J. Friedmann, S.M. Wade, Guidelines for
1916.
Carbon Dioxide Capture, Transport and Storage, World Reseurces Institute,
[19] N.A. Darwish, N. Hilal, Sensitivity analysis and faults diagnosis using artificial
Washington DC, USA, 2008.
neural networks in natural gas TEG-dehydration plants, Chem. Eng. J. 137
[4] R. Anantharaman, O. Bolland, N. Booth, E. Van Dorst, C. Ekstrom, E. Sanchez
(2008) 189–197.
Fernandes, F. Franco, E. Macchi, G. Manzolini, D. Nikolic, European best
[20] E.W. Grynia, J.J. Carroll, P.J. Griffin, Dehydration of acid gas prior to injection,
practice guidelines for assessment of CO2 capture technologies, CAESAR Project
Acid Gas Inject. Related Technol. (2010) 107–127.
FP7—ENERGY (2007) 1.
[21] L.E. Øi, M. Fazlagic, Glycol dehydration of captured carbon dioxide using Aspen
[5] D.-C. Stream, Implementation of Directive 2009/31/EC n the Geological Storage
Hysys simulation, in: Proceedings of the 55th Conference on Simulation and
of Carbon Dioxide, Guidance Document (2011).
Modelling (SIMS 55), Modelling, Simulation and Optimization , Linköping
[6] Z. Abbas, T. Mezher, M.R. Abu-Zahra, CO2 purification. Part I: Purification
University Electronic Press, Aalborg, Denmark (2014) 167–174.
requirement review and the selection of impurities deep removal technologies,
[22] S. Kumar, O. Zarzour, G. King, Design of CO2 dehydration and compression
Int. J. Greenhouse Gas Control 16 (2013) 324–334.
facilities, Abu Dhabi International Petroleum Exhibition and Conference,
[7] M. Netusil, P. Ditl, Comparison of three methods for natural gas dehydration, J.
Society of Petroleum Engineers, 2010.
Nat. Gas Chem. 20 (5) (2011) 471–476.
[23] P.S. Roy, M.R. Amin, Aspen-HYSYS simulation of natural gas processing plant,
[8] M. Netušil, P. Ditl, Natural Gas Dehydration, INTECH Open Access Publisher,
in: J. Chem. Eng. 26 (2012) 62–65.
2012.
[24] M. Bahmani, J. Shariati, A.N. Rouzbahani, Simulation and optimization of an
[9] X. Huang, Z. Li, Y. Tian, Process optimization of an industrial acetic acid
industrial gas condensate stabilization unit to modify LPG and NGL production
dehydration progress via heterogeneous azeotropic distillation, Chin. J. Chem.
with minimizing CO2 emission to the environment, Chin. J. Chem. Eng. 25
Eng. 26 (2018) 1631–1643.
(2017) 338–346.
[10] M. Salman, L. Zhang, J. Chen, A computational simulation study for techno-
[25] B. Guo, A. Ghalambor, Natural Gas Engineering Handbook, Elsevier, 2014.
economic comparison of conventional and stripping gas methods for natural
[26] M.A. Zamarripa, J.C. Eslick, M.S. Matuszewski, D.C. Miller, Multi-objective
gas dehydration, Chin. J. Chem. Eng. 28 (2020) 2285–2293.
optimization of membrane-based CO2 capture, Comput. Aided Chem. Eng. 44
[11] M. Stewart, K. Arnold, Gas dehydration field manual, Gulf Professional
(2018) 1117–1122.
Publishing, Elsevier, 2011.
[27] M. Abu-Zahra, A. Sodiq, P. Feron, Commercial liquid absorbent-based PCC
[12] M. Huffmaster, Gas dehydration fundamentals, Proceedings of the Laurance
processes, in: Absorption-Based Post-combustion Capture of Carbon Dioxide,
Reid Gas Conditioning Conference, University of Oklahoma OUTREACH, 2004.
Elsevier (2016) 757–778.
[13] S.A. Fazel, A. Safekordi, M. Jamialahmadi, Pool boiling heat transfer in dilute
[28] T. Sanpasertparnich, R. Idem, I. Bolea, D. deMontigny, P. Tontiwachwuthikul,
water/triethyleneglycol solutions, Chin. J. Chem. Eng. 17 (2009) 552–561.
Integration of post-combustion capture and storage into a pulverized coal-
[14] S. Grosso, Glycol choice for GAS dehydration merits close study, Oil and Gas J
fired power plant, Int. J. Greenhouse Gas Control 4 (2010) 499–510.
76 (7) (1978) 106–110.
[29] C. Group, Inc., Triethylene Glycol-Kinematic Viscosity Data, 2012.