Mohammed A. Al-Tayyar*
| Saif W. Mohammed Ali | Hussam H. Jabbar
© 2026 The authors. This article is published by IIETA and is licensed under the CC BY 4.0 license (http://creativecommons.org/licenses/by/4.0/).
OPEN ACCESS
The current study proposes a solution to the thermal fatigue that caused failure in the Brazed Aluminum Heat Exchanger (BAHX) located in the Natural Gas Line (NGL) producer train plant or the Enterprise Products Midstream gas plant (PGP), which was installed in Pascagoula, Mississippi, USA, to fit the Iraqi environmental conditions. Thermal Energy Storage (TES) system would be integrated into the BAHXs to store any implausible increase in the heat that can cause thermal fatigue. Simultaneously, part of the stored heat is used in the distiller to avoid any excessive heat for the separation process. The study was performed under different mass flows of NG and different inlet temperatures. The effect on the purity of Compressed Natural Gas (CNG) and thermal properties was taken into consideration, too. The degradation of other product components, like Liquid Petroleum Gas (LPG), was also studied in the current work. Different Phase Change Material (PCM) types were investigated to be used in the TES systems, where a novel parameter was differentiated called the PCM parameter to characterize the PCM physical properties. It was found that 6346 kW of heat can be stored at half the processed flow rate, while this amount can approach 25386 kW of heat storage if the flow rate is doubled. Stream temperature and mass flow rate increase as maximum possible heat storage increases, but at different inclinations and gradients. The best PCM parameter was found to be placed within (1.8-3.6 ℉/min) of the Rate of Change (ROC), where the values range between (2788 kJ/K – 2777 kJ/K). 17.6% is the maximum LPG mole fraction form 1th stage of the distiller.
natural gas processing, natural gas distillation, compressed natural gas and liquid petroleum gas processing, thermal fatigue
The growing reliance on fossil fuel-based energy, as opposed to renewable sources, is anticipated to have detrimental environmental and health consequences. To mitigate these effects and comply with regulatory limits on toxic pollutants such as CO₂, H₂S, SO₂, and particulate matter associated with natural gas processing, effective control measures are essential. One viable approach is to enhance natural gas treatment processes through strategies such as acid gas removal and thermal management of products. These improvements not only enhance byproduct sustainability and equipment efficiency while reducing exergy destruction but also contribute to improved economic performance.
Poe and Mokhatab [1] outlined key principles for understanding the complex systems involved in natural gas processing, emphasizing integrated design, advanced control systems, and real-time optimization to ensure safe, efficient, and profitable operations. Stewart and Arnold [2] provided a comprehensive guide to natural gas processing and sweetening, focusing on acid gas removal (H₂S and CO₂), dehydration, Natural Gas Liquid (NGL) recovery, and overall process design.
AlNouss and Al-Sobhi [3] assessed the sustainability of recent acid gas treatment technologies—specifically targeting CO₂ and H₂S in Liquefied Natural Gas (LNG) production using Aspen HYSYS. The study introduced innovations such as Acid Gas Enrichment (AGE) and Tail Gas Treatment Units (GTU), designed to lower emissions and enhance sulfur recovery. These advancements were found to benefit the environment and contribute to improved sustainability.
A proposed optimized design for Acid Gas Removal Units (AGRUs) in natural gas production to meet environmental regulations while minimizing energy consumption and utility costs is presented by Cho et al. [4]. Their study included a sensitivity analysis, revealing that structural modifications can enhance acid gas removal. Key parameters such as solvent flow rate, temperature, condenser duty, and heat recovery were found to influence both energy efficiency and removal performance.
Given the critical role of moisture content in natural gas production and the limitations of online sensors that often necessitate manual measurement, Ren et al. [5] suggested a machine learning-based approach to predict the water dew point in dehydration units. By automatically collecting process parameters, the method enables real-time prediction using existing Supervisory Control and Data Acquisition (SCADA) data, aiming to enhance both safety and operational efficiency.
Zarezadeh et al. [6] presented an optimized approach for pretreating natural gas, specifically sweetening and dehydration before liquefaction, aimed at reducing energy and exergy losses. For the sweetening unit, key optimization variables included MDEA concentration, piperazine concentration, and the operating pressures of the absorber and regenerator. In the dehydration unit, inlet and regeneration temperatures were identified as critical parameters influencing performance.
To enhance system efficiency, reduce power consumption, and minimize equipment requirements, Ghorbani et al. [7] proposed an innovative and optimized integrated process that combines NG, NGL recovery, and Nitrogen Rejection Unit (NRU). Exergy analysis revealed that the major losses occurred in components such as air coolers, heat exchangers, compressors, expanders, valves, and columns. Sensitivity analysis identified the recycle stream temperature, flow rate, and NRU operating pressure as key parameters affecting energy consumption.
Petropoulou et al. [8] focused on reducing energy consumption while evaluating the economic feasibility of a natural gas dehydration unit using triethylene glycol (TEG). The study involved simulating and optimizing the TEG process by comparing two thermodynamic models. Seven operational parameters were analyzed, namely, stripping gas rate, reboiler temperature, regenerator pressure, regenerator feed temperature, surge drum temperature, contactor temperature, and contactor pressure. The findings highlighted the importance of accurate thermodynamic modeling for designing efficient dehydration systems. Demonstrating that optimization strategies can effectively lower energy usage and operating costs without increasing capital investment.
A turboexpander with a Mechanical Refrigeration (TEMR) scheme was used to enhance the energy efficiency of a Natural Gas Processing Unit (NGPU) [9]. The Energy, Economic, Recovery, Quality, Safety, and Production were evaluated as key criteria of this study. The optimized criteria at the operational conditions reduced energy consumption and enhanced economic expenses. However, these improvements come with trade-offs, including minor decreases in NGL recovery and elevated safety risks.
Shoghl et al. [10] examined a three-stage Joule–Thomson (JT) throttling process in an NGL recovery unit to maximize NGL recovery, improve exergy efficiency, and evaluate phase behavior. The proposed configuration comprised three JT valves, three separators, and a heat exchanger, and its performance was assessed under seven different pressure-drop scenarios. The study found that optimal design should prioritize a balance between recovery efficiency and energy savings, rather than relying solely on cooling capacity. Jourablou et al. [11] examined the nitrogen expansion cycle used in mini-LNG plants, focusing on its sensitivity to ambient temperature and feeding gas pressure. Their study found that stable LNG production is maintained between 5–15 ℃ but declines significantly above 15 ℃ due to reduced cooling efficiency and lower turbine power output. Additionally, feed gas pressures above 35 bar support stable production, while pressures below 30 bar led to sharp declines due to pressure drop effects. The study also concluded that water cooling is more efficient and recommended for nitrogen expansion units, as the cooler is the primary source of exergy losses. Abubakar et al. [12] simulated a system designed to simultaneously recover CO₂ and produce sweet, carbon-neutral natural gas from wet natural gas. The study found that discharge pressure and feeding temperature impact the compressor's power requirements. Additionally, the performance of the heat exchanger was influenced by temperature and pressure, which directly affect vapor fraction and heat duty.
The corrosion caused by CO2 on carbon steel was studied by Hamad and Mohammed [13]. The effect of CO2 depends on several factors, such as temperature and the duration of exposure to corrosive media. The study identified that the corrosion rate increases as the temperature increases.
The natural gas processing stages, including acid gas removal, require specialized equipment capable of handling fluctuations in temperature and pressure, as well as managing byproducts. However, the absence of online monitoring for certain parameters and the lack of a reliable safety process often led to incidents, such as the one in Pascagoula, Mississippi [14]. The incident report identified thermal fatigue in the brazed aluminum heat exchanger (BAHX) as the root cause, resulting from temperature fluctuations. These fluctuations caused small cracks that eventually developed into hydrocarbon leaks. The unrepaired presence of such cracks can contribute to the catastrophic failure of the equipment.
To overcome this problem, a novel design will be presented in this paper that utilizes the Phase Change Material (PCM), which is capable of storing excess energy during the acid gas removal process. The Thermal Energy Storage (TES) unit is proposed to be integrated into BAHXs to store the excessive heat. A novel parameter is derived from the current study called the PCM parameter to characterize the proper PCM properties. The study is performed under different inlet conditions and different ROCs. The effect of utilizing part of the energy storage on the purity of CNG product and liquefied petroleum gas (LPG) degradation is considered in the present work too.
BAHXs are designed to operate under higher pressure that approaches 100 bar and a temperature range of -269 to 204 ℃ [15]. Literally, PCM material is used to be integrated with these kinds of heat exchangers to recover any excessive heat. The main issue with these PCM-heat exchanger integrated units is the lower layer efficiency; therefore, a number of projects were studied to enhance the thermal performance, including different configurations of fins; plain, triangular, perforated, and serrated are among these shapes. Løvseth et al. [16] made different approaches to create the most efficient and feasible LNG, including modeling and optimizing the heat exchanger processes. Skaugen et al. [17, 18] utilized dynamic flow sheet simulation tools to emulate the thermodynamic processes in BAHXs. It was found that static instability and process efficiency are traded off in such a way that obtaining higher instability requires lower process efficiency. Skaugen et al. [17, 18] also used the ASPEN PLUS tool to process the BAHXs in the LNG system. Optimization models were developed by previous studies [19-22] to obtain the most efficient and feasible geometries of heat exchangers with regard to recording best thermal performance for these types of heat exchangers. Entropy and exergy analysis are the base theories that were used to analyze the overall performance and the optimization algorithm viability. Integrating PCM as a thermal storage material has been a fashionable trend when it is associated with a heat exchanger in cryogenic systems. A mixture of Propane-Methanol was proposed by Li et al. [23] under low-temperature conditions and high heat capacity. It was found that in the case of transferring heat between gaseous and solid phases, the required storage volume becomes 10 times that of liquid and gaseous phases. Therefore, utilizing liquid phases is more feasible and compact economically for heat exchangers, as stated by Araki et al. [24].
The Midstream gas processing plant, which is shown in Figure 1, was built virtually via dynamic flow sheet software called COFE version 3.9.0.5. The unit mainly consists of seven units. The first unit is to reduce the inlet temperature of the feed NG stream to about 290 K before entering the first stage of BAHX. Another reduction in the exit stream temperature would be processed before entering the second stage of BAHX to approach 251 K. Both heat exchangers represent the second unit to process the NG stream, where it would exchange heat with the final produced pure gas stream (97% of methane) to raise its temperature to 287 K. The demethanizing processing unit is the third unit, which is used to extract the Methane gas stream mentioned above. On the other hand, the residues would be processed in the last four units to extract Ethane, Propane, and two stages of Butane extraction, respectively.
Figure 1. Dynamic flow sheet represents the Midstream gas processing unit [25]
Table 1 illustrates the components' mole fractions and flow rates (kmol/h) of the processed feed NG stream to the unit. The maximum mole fraction in the feed NG stream is the fraction of Methane (CH4 = 0.864), representing (4553.416 kmol/h) of the processed gas, while the minimum is the fraction of (Heptane = 0.001), representing (7.9035 kmol/h) of the total processed gas. The remaining percentages represent all other gas components.
Table 1. The feed stream components of the processed natural gas (NG) [25]
|
NG Feed Components |
Mole Fraction |
Compounds Flow (kmol/h) |
|
Methane (CNG (C1)) |
0.864 |
4552.416 |
|
Ethane (C2) |
0.0647 |
340.9043 |
|
Propane (LPG(C3)) |
0.0287 |
151.2203 |
|
Butane (C4) |
0.0072 |
37.9368 |
|
n-Butane (nC4) |
0.0082 |
43.2058 |
|
Pentane (C5) |
0.0041 |
21.6029 |
|
n-Pentane (nC5) |
0.0031 |
16.3339 |
|
n-Hexane (nC6) |
0.0031 |
16.3339 |
|
Heptane (C7) |
0.0015 |
7.9035 |
|
N7 |
0.0154 |
81.1426 |
Table 2 shows the standard specification limits of the BAHX that would be used in the NG processing unit, after which the BAHX is exposed to the danger of thermal and structural fatigue [26-28]. As it is shown in Table 2, the maximum allowable pressure for BAHX ranges between (1200–480 Psig), while (1050–570 Psig) represents the most frequent normal range that can be reached. It was shown that heat can be transferred at a rate equal to 108 ℉ per hour or at a normal range between (1.8–3.6 ℉/min). On the other hand, if ROC passes the (7 ℉/min) threshold, this will subject the BAHX to a possible thermal shock. The expected inlet temperature of the NG feed stream ranges between (250–290 K), where any increase above this range refers to thermal fatigue danger.
Table 2. Brazed aluminum heat exchanger (BAHX) specifications and feed properties [26-28]
|
Property |
Specification Limits Range |
Frequent Repeated Value |
|
Over pressurization limits |
1200 – 480 Psig (Ibf/in2) |
1050 – 570 Psig (Ibf/in2) |
|
Rate of Change (ROC ℉/min) |
1.8 -7 ℉/min |
1.8 – 3.6 ℉/min |
|
Inlet stream temperature (K) |
250-290 (K) |
250-290 (K) |
3.1 Economic feasibility
Regarding the economic feasibility of integrating such system, it was demonstrated that adopting solutions to avoid thermal fatigue in BAHXs unit would cost much lower price than the operation and maintenance costs of these kinds of heat exchangers as was stated by Vicente and Jeffery [29] that these projects would save about \$3.5 millions over ten years and also extent the life span of BAHXs units where the cost of each unit requires more \$10 million.
3.2 Validation
To validate the obtained values of ROC (F/min) of the temperature model developed in the current study, the range of obtained data was compared against the data recorded by Decker [30]. It was shown that the data obtained from the current work are very close to the range of ROC (F/min) gotten by Douglas D., as explained in Table 3 below.
Table 3. Validation of the current work data
|
ROC (F/min) Form Current Study |
ROC (F/min) Form Douglas [13] |
Error % |
|
1.8 to 3.8 with maximum value 8 |
2-5 with maximum value 9–10 |
15 % to 24% |
4.1 Assumptions
The presented mathematical model would be developed under the following assumptions:
The present theoretical work was governed by three systems of equations. The governing equations for the cooling and heat exchange of the NG processing were the first system of equations, which were used in the early stages of NG processing. Secondly, the PCM material in the TES system was solved theoretically using heat balance in a storage system. Finally, inside the gas distiller column, the product separation between column stages was treated using mass and heat balance as described below.
4.2 Heat balance inside the cooling and heat exchanging process
In the early stages of NG processing, there are two working fluids that are operated on: the hot fluid stream and the cold fluid stream. Thus, the equations below represent the above-mentioned processes [31, 32]:
$Q_{h-c}= \dot{m}_{ {cold }}\ C p_{ {cold }}\left(T c_{ {out}}-T c_{ {in}}\right)$ (1)
where, $Q_{h-c}, \dot{m}_{{cold}}$, and $C p_{ {cold }}$ are the heat transfer, mass flow rate, and heat capacity of the cold gas stream, while $T c_{{out}}$ and $T c_{ {in}}$ represent the outlet and inlet cold stream temperatures. On the other hand, on the hot side of the cooling and heating process, the equation would be as below:
$Q_{c-h}=\dot{m}_{{hot }}\ C p_{ {hot }}\left(T h_{ {out }}-T h_{{in }}\right)$ (2)
where, $h$ refers to the hot stream, and $c$ refers to the cold stream.
As thermal fatigue happens in the heat exchanger, the temperature jumps to its highest value. When this process occurs, there would be a possibility of storing the residual heat in the TES unit using PCM material. The maximum allowable energy storage ($Q_{storage}$) that can be processed in this stage can be calculated as:
$Q_{{storage }}=Q_{{heat }-j u m p-w i t h ~ f a t i g u e ~-\left(Q_{c-h}\ {or}\ Q_{h-c}\right)_{ {normal-without\ fatigue }}}$ (3)
4.3 The energy equation of Phase Change Material and Phase Change Material parameters
Regarding PCM modeling with Heat exchanger details and according to Zhang and Ge [33], the energy equation can be represented as:
$\begin{aligned} \rho \frac{\partial H}{\partial t}=\frac{\partial}{\partial n}\left(k \frac{\partial T}{\partial n}\right) & -\frac{\partial}{\partial n}(\rho u H) = Q_{\ {charge-or-Discharge }}\end{aligned}$ (4)
where, $H=C * T$.
$\eta=\frac{E_{ {storage }}}{\int_0^t Q d t}$ (5)
$Q_{{charge-or-Discharge }}=\frac{E_{ {storage }}}{v({ volume })}$ (6)
$Q_{{storage }}=E_{ {storage }}=Q_{ {charge }- { or-Discharge }} \times v$ (7)
$\rho c v \frac{\partial T}{\partial t}=Q_{ {storage }}$ (8)
Thus, the energy stored in the PCM material was simply governed by the equation below [34]:
$Q_{ {storage }}=\frac{d T}{d t} \rho c v$ (9)
As $\frac{d T}{d t}$ is the ROC of temperature with time, while (ρ), (c), and (v) are the density, heat capacity, and volume of the energy storage material. A novel parameter was differentiated in the current work called the PCM parameter, to characterize the proper PCM material that can be integrated in the TES unit. This parameter can be calculated as below:
PCM parameter $=\rho c v=\frac{Q_{{storage }}}{\frac{d T}{d t}}\left(\frac{k J}{K}\right)$ (10)
4.4 Mathematical modeling of the distiller column
Generally, the distiller column is operated under simple mass and heat balance. However, as the column consists of a number of stages in addition to using an extra amount of energy to separate between phase products and breaking down the chemical bonds, this may seem to increase the complexity of the equation system somehow. Thus, the mathematical modeling of a distiller can be derived as below [35]:
$\begin{aligned} V_{j+1} Y_{i, j+1}+L_{j-1} X_{i, j-1}+F_j Z_i & -\left(1+r_j^v\right) V_j Y_{i, j}-(1 \left.+r_j^L\right) L_j X_{i, j} =0\end{aligned}$ (11)
$\begin{aligned} V_{j+1} q_{j+1}^v+L_{j-1} q_{j-1}^L+F_j q_j^F & -\left(1+r_j^v\right) V_j q_j^v-(1 \left.+r_j^L\right) L_j q_j^L -Q_j=0\end{aligned}$ (12)
where,
$r_j^v=\frac{U_j}{V_j}, r_j^L=\frac{W_j}{L_j}$ (13)
$j$ is the stage inside the column, while $X_i$ and $Y_i$ are the rates ($i$) of the product compositions of ($L$) phase and ($V$) phase, respectively. $F_j$ and $Z_i$ are the added streams to the column at the stage $j$ and its product rate, respectively. $W_j$ and $U_j$ are the streams at the column exit. Finally, $q$ represents the phase enthalpy while $Q_j$ is the heat added for the separation process.
5.1 Effect of inlet conditions
Figure 2 shows the effect of feed stream inlet conditions on the maximum allowable heat storage in the proposed integrated system. The figure demonstrates that as the inlet temperature increases, the maximum allowable energy storage increases linearly for all studied flow rates. The system also investigates the maximum allowable amount of heat storage at different rates of NG processing discharge: 0.5, 1, 1.5, and 2 of the processed NG discharge. As illustrated, as the flow rate increases, the maximum allowable storage amount increases too. However, the slope and gradient for higher discharge are steeper than the slopes for lower discharge. This seems to be dependent on the temperature difference of the heat exchanger.
Figure 2. The maximum possible energy storage at different inlet conditions
5.2 The calculated Phase Change Material parameter
Figure 3 shows the calculated PCM parameters at different rates of change (ROC) values and different inlet conditions, respectively. It was shown in Figure 3(a) that as (ROC) increases, the calculated PCM parameter is reduced linearly. Mathematically, this comes from the fact that ROC is placed in the denominator of the PCM parameter equation. However, as was stated in Table 2, the most frequently repeated normal value of ROC ranges from (1.8–3.6 ℉/min); thus, the most appropriate PCM parameter ranges from (2788 kJ/K – 2777 kJ/K). Therefore, even the proposed PCM parameter would not be feasible if the PCM parameter is not placed within the above-mentioned range, as any reduction below this range still exposes the system to sudden thermal fatigue. Figure 3(b) demonstrates the effect of inlet mass flow rate and inlet temperature on the required PCM parameter to stabilize the system and avoid sudden thermal fatigue. It was shown that increasing the processed NG flow rate and feed inlet temperature requires storage material that has a higher PMC parameter. However, as the processed NG rate increases, the slope and magnitude of the PCM parameter increase too.
Figure 3. The calculated phase change material (PCM) parameters (a) at different rates of change (ROC) and the processed flow rate, (b) at different inlet conditions and the highest maximum ROC (8 ℉/min)
5.3 Effect of the processed energy on the product's purity
The effect of utilizing part of the stored energy on the produced phases’ mole fractions along the distiller column’s stages, the temperature of the exit CNG stream, the purity of the exit CNG stream, and the degradation of the LPG stream is shown in Figure 4, respectively. As the stages of distiller column are numbered for (1) to (31) starting from the distiller top until distiller bottom, the highest CNG mole fraction (0.96889) would be at the top exit stream and starts decreasing along the column till reaching its minimum values at the bottom of the column (stage 31) where all other products phases fractions would be processed in another column distillers. This explains why LPG mole fraction behaves in contradictory trends with CNG behavior, where LPG mole fraction increases from the top at its minimum value to reach its maximum value at the distiller's bottom stages, as was shown in Figure 4(a).
Figure 4. Produced components properties, (a) Compressed natural gas (CNG) and liquid petroleum gas (LPG) products along column stages, (b) CNG exit temperature at different utilized energy, (c) CNG phase fraction at different utilized energy, (d) LPG degradation at different utilized energy
Figure 4(b) illustrates the effect of utilized energy on the CNG exit stream temperature. It is shown that as utilized energy increases, an exponential increase is noticed in the exit temperature. However, the increase in temperature is insignificant, as most of the energy utilized is consumed to separate between stream phases rather than the temperature increase. The effects of the utilized energy on the purity of the CNG exit stream and degradation of the LPG stream are shown in Figure 4(c) and (d), respectively. As shown in the figures, increasing the processed energy in the column distiller unit leads to exponential reduction trends in the CNG mole fraction and exponential increase trends in the LPG mole fraction. This comes from the fact that any increase in the processed energy will affect the strength of chemical bonds between components, leading to the breaking down of these bonds. In addition, the increase in the utilized energy would lead to an increase in the vaporization rate of the other phases, causing degradation in the LPG phase and other products.
A proposed solution for the failure that occurred at the Enterprise Products Midstream gas plant (PGP), Pascagoula, Mississippi, USA, was investigated in the present study. The failure was caused by thermal fatigue and residues in the BAHX unit of the PGP plant. The current study proposed integrating a TES system into the BAHX unit. The maximum allowable heat storage was investigated under different processing mass flow rates and inlet temperatures. In addition, a novel parameter, referred to as the PCM parameter, was developed to facilitate the selection of a suitable PCM for the TES system. The effect of utilizing part of the stored heat in the CNG distillation column on CNG product purity and LPG degradation was also investigated. The remarks below were concluded from the current study:
The authors of the present work are very grateful to the University of Kufa, the University of Karbala, and the University of Baghdad for supporting their work.
[1] Poe, W.A., Mokhatab, S. (2017). Introduction to natural gas processing plants. In Modeling, Control, and Optimization of Natural Gas Processing Plants, pp. 1-72. https://doi.org/10.1016/b978-0-12-802961-9.00001-2
[2] Stewart, M., Arnold, K. (2011). Gas sweetening. In Gas Sweetening and Processing Field Manual, pp. 1-140. https://doi.org/10.1016/b978-1-85617-982-9.00002-8
[3] AlNouss, A., Al-Sobhi, S. (2022). Sustainable analysis of recent acid gas treatment schemes for LNG production. Computer Aided Chemical Engineering, 51: 589-594. https://doi.org/10.1016/b978-0-323-95879-0.50099-0
[4] Cho, H., Binns, M., Min, K.J., Kim, J.K. (2015). Automated process design of acid gas removal units in natural gas processing. Computers & Chemical Engineering, 83: 97-109. https://doi.org/10.1016/j.compchemeng.2015.05.030
[5] Ren, H., Yin, A., Dai, Z., Liu, X., Tan, Z., Zhang, B. (2023). Parameter screening and optimized Gaussian process for water dew point prediction of natural gas dehydration unit. Process Safety and Environmental Protection, 170: 259-266. https://doi.org/10.1016/j.psep.2022.12.019
[6] Zarezadeh, F., Vatani, A., Palizdar, A., Nargessi, Z. (2022). Simulation and optimization of sweetening and dehydration processes in the pretreatment unit of a mini-scale natural gas liquefaction plant. International Journal of Greenhouse Gas Control, 118: 103669. https://doi.org/10.1016/j.ijggc.2022.103669
[7] Ghorbani, B., Hamedi, M.H., Amidpour, M. (2016). Development and optimization of an integrated process configuration for natural gas liquefaction (LNG) and natural gas liquids (NGL) recovery with a nitrogen rejection unit (NRU). Journal of Natural Gas Science and Engineering, 34: 590-603. https://doi.org/10.1016/j.jngse.2016.07.037
[8] Petropoulou, E.G., Carollo, C., Pappa, G.D., Caputo, G., Voutsas, E.C. (2019). Sensitivity analysis and process optimization of a natural gas dehydration unit using triethylene glycol. Journal of Natural Gas Science and Engineering, 71: 102982. https://doi.org/10.1016/j.jngse.2019.102982
[9] Franco, S.V.D.A., da Cunha Ribeiro, D., Meneguelo, A.P. (2020). A comprehensive approach to evaluate feed stream composition effect on natural gas processing unit energy consumption. Journal of Natural Gas Science and Engineering, 83: 103607. https://doi.org/10.1016/j.jngse.2020.103607
[10] Shoghl, S.N., Naderifar, A., Farhadi, F., Pazuki, G. (2021). Thermodynamic analysis and process optimization of a natural gas liquid recovery unit based on the Joule–Thomson process. Journal of Natural Gas Science and Engineering, 96: 104265. https://doi.org/10.1016/j.jngse.2021.104265
[11] Jourablou, M.M., Salimi, M., Amidpour, M. (2025). A comprehensive analysis of the nitrogen expansion process for liquefied natural gas industry from the operational point of view. International Journal of Thermofluids, 27: 101200. https://doi.org/10.1016/j.ijft.2025.101200
[12] Abubakar, A.M., Umdagas, L.B., Arowo, M.N., et al. (2024). Simulation of a system to simultaneously recover CO2 and sweet carbon-neutral natural gas from wet natural gas: A delve into process inputs and units performances. Cleaner Energy Systems, 9: 100156. https://doi.org/10.1016/j.cles.2024.100156
[13] Hamad, H.S., Mohammed, K.A. (2024). Localized corrosion behavior of carbon steel as a function of surface temperature and water condensation rate at the top of oil and gas pipelines. Journal of Petroleum Research and Studies, 14(2): 88-100. https://doi.org/10.52716/jprs.v14i2.866
[14] Tzoumis, K.A. (2020). US Chemical Safety and Hazard Investigation Board (CSB). In Toxic Chemicals in America: Controversies in Human and Environmental Health, pp. 645-646.
[15] Thulukkanam, K. (2013). Heat Exchanger Design Handbook, 2nd ed. CRC Press.
[16] Løvseth, S.W., Snustad, I., Brunsvold, A.L., Skaugen, G., Wahl, P.E., Lervåg, K.Y. (2015). From droplets to process: Multilevel research approach to reduce emissions from LNG processes. Energy Procedia, 64: 3-12. https://doi.org/10.1016/j.egypro.2015.01.003
[17] Skaugen, G., Gjøvåg, G.A., Nekså, P., Wahl, P.E. (2010). Use of sophisticated heat exchanger simulation models for investigation of possible design and operational pitfalls in LNG processes. Journal of Natural Gas Science and Engineering, 2(5): 235-243. https://doi.org/10.1016/j.jngse.2010.10.003
[18] Skaugen, G., Kolsaker, K., Walnum, H.T., Wilhelmsen, Ø. (2013). A flexible and robust modelling framework for multi-stream heat exchangers. Computers & Chemical Engineering, 49: 95-104. https://doi.org/10.1016/j.compchemeng.2012.10.006
[19] Peng, X., Liu, Z., Qiu, C., Tan, J. (2014). Passage arrangement design for multi-stream plate-fin heat exchanger under multiple operating conditions. International Journal of Heat and Mass Transfer, 77: 1055-1062. https://doi.org/10.1016/j.ijheatmasstransfer.2014.06.058
[20] Chen, Q., Wang, M., Pan, N., Guo, Z.Y. (2009). Optimization principles for convective heat transfer. Energy, 34(9): 1199-1206. https://doi.org/10.1016/j.energy.2009.04.034
[21] Chen, L., Chen, Q., Li, Z., Guo, Z.Y. (2009). Optimization for a heat exchanger couple based on the minimum thermal resistance principle. International Journal of Heat and Mass Transfer, 52(21-22): 4778-4784. https://doi.org/10.1016/j.ijheatmasstransfer.2009.06.012
[22] Wang, Z., Han, F., Sundén, B., Li, Y. (2017). A thermal design method for the performance optimization of multi-stream plate-fin heat exchangers. Journal of Mechanical Science and Technology, 31(6): 3017-3024. https://doi.org/10.1007/s12206-017-0545-y
[23] Li, Y., Cao, H., Wang, S., et al. (2014). Load shifting of nuclear power plants using cryogenic energy storage technology. Applied Energy, 113: 1710-1716. https://doi.org/10.1016/j.apenergy.2013.08.077
[24] Araki, H., Nakabaru, M., Chino, K. (2002). Simulation of heat transfer in the cool storage unit of a liquid–air energy storage system. Heat Transfer—Asian Research, 31(4): 284-296. https://doi.org/10.1002/htj.10035
[25] Luyben, W.L. (2013). Control of a train of distillation columns for the separation of natural gas liquid. Industrial & Engineering Chemistry Research, 52(31): 10741-10753. https://doi.org/10.1021/ie400869v
[26] Chart Energy & Chemicals, Inc., Chart Lifecycle, Inc. (2025). Installation, Operation, and Maintenance Manual for Chart Brazed Aluminum Heat Exchangers (BAHX) and Core-in-Kettle™ Assemblies. https://files.chartindustries.com/BAHX-IOM.pdf.
[27] Energy API. (2018). Publications Programs Services 2018. https://www.api.org/~/media/files/publications/2018_catalog/2018_pubs_catalog_final_sm.pdf.
[28] ASME. (2013). Code Cases: Boilers and Pressure Vessels. https://www.scribd.com/document/628551220/Code-Cases-Boilers-and-Pressure-Vessels.
[29] Vicente, F.F., Jeffery, S. (2022). Braskem combining staff knowledge with System 1 decision support to increase plant safety & operational excellence. Orbit Newsletter Digital Publication, Q3/2022, Baker Hughes.
[30] Decker, D. (2025). Brazed aluminum heat exchangers operation and maintenance for asset integrity. Inspectioneering Journal - Asset Intelligence, 31(3).
[31] Xu, Z., Yu, Z., Ning, X., Zhao, C., Nie, F. (2025). Configuration optimization for offset strip plate-fin heat exchanger using a method of PID-based search algorithm driving design indicators mathematical model. Case Studies in Thermal Engineering, 68: 105901. https://doi.org/10.1016/j.csite.2025.105901
[32] Yang, H., Wen, J., Gu, X., et al. (2017). A mathematical model for flow maldistribution study in a parallel plate-fin heat exchanger. Applied Thermal Engineering, 121: 462-472. https://doi.org/10.1016/j.applthermaleng.2017.03.130
[33] Zhang, X.Y., Ge, Y.T. (2026). Experimental and numerical investigation of a microchannel PCM heat exchanger for space heating. Applied Thermal Engineering, 302: 131850. https://doi.org/10.1016/j.applthermaleng.2026.131850
[34] Garg, A., Singhal, G. (2022). Mathematical modeling for transient thermal performance analysis of phase change material-based heat exchanger. Applied Thermal Engineering, 216: 119029. https://doi.org/10.1016/j.applthermaleng.2022.119029
[35] Mohammed Ali, S.W., Jabbar, H.H., Alshaibany, A.K. (2025). Technical and economic analysis of retrofitting concentrated solar collector to generate the thermal energy required for the crude oil products distillation process. Kufa Journal of Engineering, 16(2): 197-214. https://doi.org/10.30572/2018/kje/160212