Technical Evaluation of BTEX Emission Mitigation from Gas Dehydration Unit by Revamping and Using Alternative Glycols

Abstract

Water removal is crucial in natural gas processing to minimize water content, ensure safe transmission, and prevent operational issues like equipment corrosion and hydrate formation. Glycol absorption could be considered as one of the most effective methods used for natural gas dehydration and dew point control. However, during solvent regeneration, some pollutants, like benzene, toluene, ethylbenzene, and xylene (BTEX), are released to the atmosphere, resulting in catastrophic physical and mental health problems. Minimizing such pollutants that have negative impacts is highly needed to avoid the related negative environmental consequences. The objective of the current work is to investigate alternative strategies targeted to minimize BTEX emissions and guarantee efficient control of the dew point. Two strategies are introduced and investigated in this work; the first strategy is based on revamping an existing unit by adding a new cooler upstream glycol inlet separator, while the second strategy is based on using alternative glycols. The proposed strategies are applied to an Egyptian natural gas dehydration unit to select the optimum scenario that achieves the minimum BTEX emissions with efficient dew point control. It is found that natural gas dehydration using monoethylene glycol (MEG) is the best scenario in reducing BTEX emissions with efficient dew point control. The impact of operating conditions on BTEX emissions, along with natural gas water content, is also investigated. Lingo optimization software, v. 18, as well as HYSYS, v. 14, are used to find the optimum operating conditions for efficient dew point control with minimum BTEX emissions. It is demonstrated that stripping gas, MEG circulation rate, and inlet feed gas temperature have remarkable effects on BTEX emissions. Two quadratic correlations are also introduced in this study to efficiently relate BTEX emissions and water dew point to the influencing operating conditions.

1. Introduction

The existence of water vapor in natural gas (NG) results in lots of trouble, since it condenses and forms solid gas hydrates at low temperatures; these solids may block pipelines, halting gas production and potentially causing pipeline plugging, corrosion, or rupture [1,2,3,4]. So, controlling and reducing natural gas water content is crucial for safe processing and transmission [5]. Pipeline specifications dictate water content limits in US, Canadian, and Alaskan pipeline systems as 7, 4, and 1–2 lb/MMscf, respectively [1]. These differences in specifications are due to regional changes in temperature and operating conditions; colder climates need stricter limits to avoid hydrate and corrosion problems. The reduction in water content to the specified limits in order to meet sales specifications and avoid its negative consequences could be achieved through natural gas dehydration units (NGDUs) [6,7,8,9].

Among the techniques used for NG dehydration, glycol absorption is the most effective technique [10]. However, throughout the dehydration process, glycol absorbs some other pollutants like BTEX, which are found in wet gas. These absorbed pollutants are then released into the atmosphere during glycol regeneration. Even little quantities of BTEX in the natural gas could lead to significant concentrations in the vented stream [11,12].

BTEX emissions are considered hazardous air pollutants according to the Environmental Protection Agency (EPA) due to their irritating and carcinogenic properties. These emissions have been associated with blood disorders and can have detrimental effects on the respiratory system, reproductive system, central nervous system, and neurological system. Furthermore, they cause some other problems in absorbers used in NG dehydration units; they could result in excessive foaming, flooding, increased glycol loss, decreased efficiency, and higher maintenance expenses [13,14].

Various approaches have been introduced to reduce BTEX emissions, including incinerating released gases, incorporating a condensation unit, optimizing processes, and utilizing glycol with lower absorption capabilities as a solvent [13,15]. The burning process is considered the most applied method to eliminate BTEX gases [16,17]. Although naphthalene is an aromatic compound of the same classification, its concentration and volatility in natural gas streams are typically very low, leading to negligible emission rates compared with BTEX. Therefore, the present study focuses on BTEX as the most representative and environmentally significant aromatic pollutants emitted from glycol dehydration units.

Recent initiatives are now centered on creating new processes, enhancing current ones, and utilizing solvents that absorb less BTEX or creating alternative solvents instead of treating water effluent containing BTEX compounds, which can be costly [13,15,16,18,19,20,21].

BTEX compounds have decreasing solubility in the order triethylene glycol (TEG) > diethylene glycol (DEG) > ethylene glycol (EG). For example, at 25 °C, toluene solubility is 24.8 wt% in TEG, 17.2 wt% in DEG, and 2.9 wt% in EG, while benzene is fully miscible in TEG, 31.3 wt% soluble in DEG, and 5.7 wt% soluble in EG [15]. Thus, different dehydrating agents could result in different doses of BTEX emissions.

Several studies have been conducted on gas dehydration units, with previous works primarily focusing on either enhancing process efficiency or minimizing emissions. In a study conducted by Isa et al. [22], three natural gas dehydration processes were simulated in an industrial unit in the UAE. They suggested the addition of potassium formate to the TEG solution as a novel approach to enhance TEG water absorption capacity. However, this led to an increase in the absorption rate of BTEX and consequently raised operational costs. Abdulrahman and Sebastine [23] examined the utilization of various glycols for the dehydration of natural gas. They discovered that TEG is highly effective in removing water and absorbing a greater amount of hydrocarbons. The research emphasized the significance of managing BTEX emissions in natural gas dehydration units because of their adverse impact on human health. However, they were unable to concurrently address both objectives.

Hedayati Moghaddam [24] studied the efficiency of wet natural gas dehydration through absorption with liquid desiccant. In a separate study, Moghaddam and Sadeq [25] also explored this process using a genetic algorithm; they optimized the glycol regeneration process. Tazang et al. [26] developed a method for accurately modeling the solubilities of BTEX in triethylene glycol (TEG). Kong et al. [27] revealed that various recycling arrangements for NG dehydration units do not considerably shorten water content or BTEX emissions, despite needing remarkable operating and capital costs. Additionally, they found that marginal cutting in total BTEX emissions often results in a noticeable content of water in the product gas. Among these research works, no balanced solution able to incorporate environmental impacts and NG specification requirements is introduced.

A sensitivity analysis was carried out by Braek et al. [1] on five operating parameters of an NG dehydration process in Abu Dhabi, UAE. The studied variables were absorber temperature and pressure, flash tank pressure, regenerated TEG temperature, and regenerator reboiler temperature. However, the maximum reduction in BTEX emissions was 48%. Still, there is a need to significantly reduce these dangerous pollutants.

Nemati Rouzbahani et al. [28] found that lean DEG purity significantly affects BTEX emissions, while Darwish and Hilal [19] demonstrated that they are noticeably influenced by the circulation rate and temperature of lean TEG. Torkmahalleh et al. [29] studied the impact of reboiler heat duty, solvent and stripping gas circulation rates, and lean TEG temperature on BTEX, VOC, and CO₂ emissions; they found that the moisture content of dry gas could be decreased by increasing the solvent circulation rate, while higher flow rates up to 9.25 GPM increase moisture content. Darwish and Hilal’s [19] study indicated that an increase in solvent flow rate can diminish solvent purity as a consequence of chemical dissolution, thereby reducing moisture content.

Affandy et al. [30] have focused on studying the operating conditions of natural gas dehydration; however, their study was focused on the economics of the process. Shoaib et al. [31] introduced the optimum conditions for NGDP, though they focused only on the targeted dew point and condensate throughout. They did not take into consideration the harmful emissions that could be emitted during the regeneration stage. Siti et al. [32] optimized operational variables using symmetry process simulation software, examining the impact of operational variables on performance, but they did not consider the environmental impact. Renanto et al. [33] proposed a new arrangement for NGDU using TEG, while Chong et al. [34] introduced a model for dehydration systems based on glycol absorption to lessen the total annual cost (TAC). Kharisma et al. [35] aimed at minimizing TAC as well as improving efficiency through the optimization of the TEG dehydration process, while Mukherjee et al. [36] determined optimal operational conditions for minimal BTEX emissions.

Recent studies have demonstrated the importance of integrating environmental sustainability with dehydration efficiency. For instance, Advanced thermodynamic models, such as SAFT and PC-SAFT, have improved the prediction of BTEX solubility in glycols [26]. Mukherjee and Diwekar [36] analyzed BTEX solubility and operational optimization in TEG systems, while Hedayati [24] proposed integrated approaches for reducing aromatic hydrocarbon emissions from glycol units. Moreover, Eldemerdash and Kamarudin [21] demonstrated that solvent selection and process configuration strongly influence BTEX emission profiles and overall energy efficiency. Recent hybrid and machine learning approaches have enabled accurate forecasting of dehydration performance and emission behavior (Wang et al., 2024) [37]. Additionally, Shingan et al. [38] investigated the integration of artificial intelligence and data analytics into process optimization, highlighting the importance of operational efficiency and emission reduction. These findings highlight the need for systematic evaluations that integrate dew point control, emission minimization, and cost-effectiveness in natural gas dehydration systems.

Upon the previous comprehensive review, it is found that reconciling efficient dehydration and environmental sustainability is still a challenge. This confirms the need for innovative solutions that meet simultaneously. Prior research has mostly concentrated on either increasing dewatering efficiency or lowering BTEX emissions, but few studies have accomplished both at the same time.

Up to now, no research work has fixed the problem of simultaneous minimization of hazardous emissions, achieving the permitted percentage of moisture content, and efficient control of NG dew point. To achieve this goal, four scenarios for the process of drying gas with glycol were investigated through the application on an existing NGDP located in the Western Desert, Egypt. The proposed scenarios include studying the plant by adding a cooler upstream glycol inlet separator and using different drying agents, TEG and MEG. This work aims to select the appropriate one that achieves minimum BTEX emissions while keeping the sales gas water content within specifications. Sensitivity analysis was carried out to define the impact of operating conditions on the water content of sales gas and BTEX emissions. A novel model aiming to optimize NGDU operating variables is introduced. For the selected scenario, the studied operating variables are glycol circulation rate, reboiler temperature, stripping gas rate, and inlet feed gas temperature.

Moreover, two developed correlations were introduced for the selected optimum scenario using regression analysis. The first correlation links the sales gas water content to the different influencing variables. The second one links the amount of evolved BTEX emissions during regeneration to the influencing operational conditions.

2. Methodology

The goal of the present work can be achieved through the following steps:

Step 1: Gathering the available data for an NGDU; the required data are the operating conditions, the water content of product gas, and the amount of BTEX emissions.

Step 2: Simulation of the NGDU with the available data using HYSYS simulation software (version 14).

Step 3: Validation of the simulated case study by comparing the simulation results with the actual results given in the raw data.

Step 4: Investigating the impact of changing operational conditions on sales gas water content as well as BTEX emissions through carrying out the sensitivity analysis.

Step 5: Two strategies are applied to minimize BTEX emissions and keep water content of sales gas on specifications; the first is revamping the existing plant by adding new cooler upstream glycol inlet separator and the second is using alternative glycol and conduct comparison between the four scenarios to select the optimum scenario which could be applied to achieve the proposed target.

Step 6: Extracting two alternative correlations able to efficiently determine the water content of sales gas and BTEX emissions at any operating condition.

Step 7: Building up an optimization model aiming to minimize BTEX emissions and containing constraints that guarantee the production of sales gas on specifications from the viewpoint of water content. The developed correlations serve as the main inputs in the proposed optimization model.

3. Case Study

The present case study is introduced to investigate the effect of the proposed strategies on BTEX mitigation; it is for an NGDP placed in the Western Desert, Egypt. This plant uses triethylene glycol as an absorbent for the removal of water vapor from natural gas and efficient control of its dew point. The natural gas is supplied at an inlet flow rate of 60 MMSCFD and a pressure of 1200 psig. The composition of the wet gas feed for the plant is given in

Table 1. Wet gas compositions used as a feed of the investigated glycol unit [39].

Constituent Category	Component	Mole %
Alkanes	Methane	84.708
	Ethane	6.362
	Propane	3.083
	i-Butane	0.740
	n-Butane	1.114
	i-Pentane	0.451
	n-Pentane	0.376
	n-Hexane	0.359
	n-Heptane	0.162
	n-Octane	0.088
	n-Nonane	0.081
	n-Decane	0.000
BTEX	Benzene	0.011
	Toluene	0.046
	E-Benzene	0.004
	p-Xylene	0.005
	m-Xylene	0.005
	o-Xylene	0.003
Water content	H2O	0.148
Other gases	CO2	1.753
	Nitrogen	0.501

. The complete description of the considered natural gas dehydration plant is introduced by Shoaib et al. [39]. Figure 1 represents the process flow diagram (PFD) of the considered NGDP.

HYSYS version 14 is used to simulate the present case study in order to further investigate the effect of the proposed modifications on BTEX emissions as well as the NG dew point.

Simulation Setup in Aspen HYSYS V14

To ensure the repeatability of the simulations in this research, the natural gas dehydration unit (NGDU) was modeled using Aspen HYSYS V14. Figure 1 displays the process flow diagram (PFD) of the base case study. The simulations followed steady-state conditions based on the Peng–Robinson equation of state, which was chosen because it is precise in its approximation of hydrocarbon and glycol systems, especially to estimate water and BTEX absorption behavior.

The composition of the wet gas feed was specified according to Table 1, and the inlet flow rate was 60 MMSCFD, with pressure of 1200 psig, and temperature of 25–45 °C. The glycol contactor was simulated as an absorber column with 4–6 theoretical plates, and it was optimized under each case study.

The flash separator was run at 50–100 psig to partition volatile contaminants, such as BTEX, which were measured on the vapor phase exit. The reboiler was set to the temperature ranges of 190–205 °C using TEG (Cases 1 and 2), and 145–160 °C using MEG (Cases 3 and 4), and the heat duties were automated by HYSYS (e.g., 230, 363 Btu/hr using Case 1). The stripping column was treated as a packed column, and the gas flow rates in the stripping column were 0.1 to 0.4 MMSCFD, and this was taken as a dry gas supply line.

A sensitivity analysis was carried out by systematically varying key operating parameters with the help of the HYSYS Case Study tool. These parameters involved stripping gas flow rate (0.1, 0.15, 0.2, 0.25, 0.3, 0.35, and 0.4 MMSCFD), glycol circulation rate (2, 4, 6, 8, and 10 GPM), inlet feed gas temperature (25, 30, 35, 40, and 45 °C), and reboiler temperature (as specified above). In each case study, 700 simulation runs were subsequently performed to assess the effect on sales gas water content (lb/MMSCF), BTEX emissions (SCFD), and reboiler duty (Btu/hr), as well as utility costs (USD/hr).

Steady-state operation was assumed, as well as adherence to ideal gas behavior on the stripping gas and negligible pressure drops through piping. It was also assumed that the lean glycol purity would remain constant before the glycol was regenerated at 99.5 wt%. The results of the simulations were compared to the experimental data (Section 4 and Figure 2), achieving high correlations (R² > 0.95). The simulations were conducted on an Intel Core i7 processor system with Windows 10 and 16 GB RAM, and default HYSYS solver settings were used. Raw data simulations and HYSYS files could be provided on request, in recognition of the Data Availability Statement.

4. Validation of the Simulated Case Study

To ensure the accuracy and reliability of the proposed simulation of the current case study, a comparison between the simulation results and the actual outputs should be considered. The experimental data were obtained from the operational records of the NGDU in the Western Desert, Egypt. These records were provided by the plant operator and included measurements of key performance indicators, such as sales gas water content (in lb/MMSCF), BTEX emissions (in SCFD), and operational parameters, which are as follows: stripping gas flow rate, TEG/MEG circulation rate, inlet feed gas temperature, and glycol regenerator temperature. The data were collected over six months under steady-state operating conditions to ensure consistency with the simulation assumptions. The wet gas feed composition, as specified in Table 1, was determined using gas chromatography (GC) analysis conducted on site, with samples taken at the inlet of the glycol contactor. These compositional data were cross-verified with historical data from Shoaib et al. [39] to ensure representativeness of typical feed conditions at the plant.

To validate the HYSYS simulation results, the sales gas water content and BTEX emissions resulting from the simulation were compared with the experimental measurements. Coefficient of Determination (R²) is employed to assess the matching between simulated and experimental data. It measures the proportion of variance in the experimental data explained by the simulation model. As R² values approach 1, it confirms the simulation’s capability to replicate real-world behavior across the studied conditions.

Measurement Techniques

A total of 140 steady-state data points were collected across operating ranges. Experimental measurements followed industry standards to ensure accuracy:

• Sales Gas Water Content: Measured with a chilled mirror dew point sensor (±0.3 °C accuracy), calibrated before each operation. Dew point results were transformed to lb/MMSCF using Peng–Robinson correlations, consistent with simulations.
• BTEX Emissions: Determined by GC-FID (e.g., Agilent 7890B) using samples collected in Tedlar bags per EPA Method 18. The system was calibrated with certified BTEX standards, achieving ±7% uncertainty and ~0.2 ppmv detection.
• Operational Parameters: Glycol circulation, stripping gas, feed gas temperature, and regenerator temperature were tracked with the plant DCS. Flow, temperature, and pressure instruments (accuracies ±2%, ±0.4 °C, ±0.4 psig) are calibrated every month.

Figure 2 clarifies the values of sales gas dew points resulting from the experimental results versus the simulation results under various operating conditions. The studied operational conditions are stripping gas flow rate, TEG circulation rate, inlet temperature of feed gas, and TEG regenerator temperature. The tight alignment of data points through the 1:1 line shown in Figure 2, combined with the high R² and low error metrics, proves the strength of the simulation.

5. Results and Discussions

Four case studies are investigated in this work to obtain the optimum scenario for BTEX emission mitigation from the considered gas dehydration unit. The first case is the dehydration of natural gas by absorption using TEG, while the second dehydration scenario is by absorption using TEG with the addition of a cooler upstream glycol inlet separator. This will lead to a decrease in gas temperature, which will increase the efficiency of the dehydration process and minimize the necessary water that the solvent has to absorb. The third method considers natural gas dehydration by absorption using MEG instead of TEG. The fourth case is dehydrating natural gas by the MEG absorption process, with the addition of a cooler upstream glycol inlet separator.

5.1. Selection of Operational Parameters and Simulation Settings

The operational parameters and simulation conditions were chosen relying on the data specific to the Western Desert NGDU (Section 3), the scope of sensitivity analysis required, and industry standards to be able to accurately assess the efficiency of BTEX mitigation and dehydration. Stripping gas rates (0.1–0.4 MMSCFD) were selected as per the GPSA guidelines [1] that recommend an optimal rate aiming at glycol regeneration. Circulation of glycol (2–10 GPM) was based on the rule of 2–6 gal/lb of water removal with the addition of BTEX sensitivity [1,19]. The inlet feed gas temperatures (25 to 45 °C) were indicative of field conditions and API RP 12G specifications (<50 °C) [1]. Both boiler and reboiler temperatures (190–205 °C for TEG; 145–160 °C for MEG) are well within thermal stability limits [15,28]. Simulations were carried out using HYSYS v.14 with Peng–Robinson EOS, which was validated against actual data (Section 4), and which promotes reproducibility according to best practices [15,21].

5.2. Selection of the Most Efficient and Optimum Gas Dehydration Process

Using sensitivity analysis and HYSYS (version 14) as a simulation tool, the impact of the operating parameters under consideration on BTEX emissions and the water dew point of the generated gases is examined with reference to the aforementioned case studies. Stripping gas rate, glycol circulation rate, feed gas temperature, and glycol reboiler temperature are the most important operating parameters. The flow rates of the stripping gas under study range from 0.1 to 0.4 MMSCFD (0.1, 0.15, 0.2, 0.25, 0.3, 0.35, and 0.4 MMSCFD are the chosen values). With chosen values of 2, 4, 6, 8, and 10 GPM, the TEG circulation rate ranges from 2 to 10 MMSCFD. The study’s chosen values for the inlet feed gas temperature are 25, 30, 35, 40, and 45 °C. The TEG reboiler temperature is adjusted from 190 to 205 °C for the first and second dehydration techniques, with specific values of 190, 195, 200, and 205 °C. The MEG reboiler temperature is raised from 145 to 160 °C in the third and fourth case studies, with specific values of 145, 150, 155, and 160 °C.

The simulation results of the first dehydration method using TEG show that the obtained average sales gas water content is 4.896 Lb/MMSCF, and the total emissions from the regeneration package are 3943 SCFD, with a reboiler duty of 230,363 Btu/hr and utility costs of 1.791 USD/hr. The simulation results related to the second case study illustrated that the average sales gas water content is 5.014 Lb/MMSCF, total emissions are 3904 SCFD, reboiler duty is 231,284 Btu/hr, and the utility costs are 0.929 USD/hr. The obtained results of the third and fourth dehydration scenarios for the average sales gas water content, total emissions from regeneration package, reboiler duty, and utility costs are 5.202 and 5.175 Lb/MMSCF, 0.004785 and 0.004766 SCFD, 111,470 and 113,324 Btu/hr, and 0.557 and 0.535 USD/hr, respectively. As mentioned in Section 5.3.4, a non-negligible increase in BTEX emissions is noted at an MEG circulation rate of 2 GPM when the temperature rises from 155 to 160 °C, even though the MEG regenerator temperature typically has a negligible impact on the water content of sales gas.

The current study evaluated four simulation cases for natural gas dehydration, focusing on BTEX emissions and sales gas water content. The third case, using MEG as an alternative glycol, was selected as the optimal solution. It achieved a sales gas water content of 5.202 lb/MMSCF and low emissions of 0.004785 SCFD, meeting performance and environmental benchmarks [31]. While the fourth case produced similar results, its inclusion of a cooler increased costs, making the third case more economically viable. The chosen configuration balances efficiency, environmental compliance, and cost-effectiveness.

5.3. Effect of Operating Parameters on Sales Gas Water Content

The third case study, which employs MEG for the gas dehydration process, has been identified as the most economical and environmentally friendly option with acceptable BTEX emissions. To optimize the process and meet sales gas specifications while minimizing BTEX emissions, it is crucial to study the impact of key operational variables. These variables include the stripping gas rate, glycol circulation rate, inlet feed gas temperature, and glycol reboiler temperature. For this study, the inlet feed gas temperature is evaluated at 25, 30, 35, 40, and 45 °C, while the glycol reboiler temperature is varied at selected values of 145, 150, 155, and 160 °C. It should be noticed that a stripping gas flow rate of 0.3 MMSCFD was utilized to reflect typical mid-to-high working circumstances, whereas this work use also 0.1 MMSCFD, the ideal flow rate determined in Section 5.5, to illustrate how sensitive sales gas water content and BTEX emissions are to operational factors. These decisions support the optimization results, which demonstrate that a stripping gas flow rate of 0.1 MMSCFD outperforms higher flow rates like 0.4 MMSCFD due to lower operating costs and better specification compliance, while minimizing BTEX emissions (zero SCFD) and achieving a sales gas water content of 1 lb./MMSCFD.

5.3.1. Effect of Stripping Gas Flow Rate

In order to improve dehydration performance and achieve greater dew point depression, stripping gas is essential for reaching high glycol concentrations that are not achievable with traditional regeneration techniques. Stripping gas is used in this procedure to eliminate any remaining water from the regenerated glycol [31,39]. In the case study that is being examined, a packed tower situated between the storage tank and the reboiler facilitates the entrance of stripping gas. In this system, the leftover water is removed from the regenerated glycol using dry gas. The stripping gas effectively eliminates extra water when the glycol, which has been heated in the reboiler, flows downward through the packed tower. The storage tank is then used to collect the dehydrated glycol. The fuel gas (dry gas) supply line of the reboiler is usually where the stripping gas is drawn from.

The simulation results in Figure 3, Figure 4, Figure 5 illustrate the relationship between stripping gas flow rate and sales gas water content under different operational conditions. Figure 3 shows the effect of stripping gas flow rate on sales gas water content at various MEG flow rates with an inlet feed gas temperature of 30 °C and a regenerator temperature of 150 °C. The results show that a higher stripping gas flow rate (0.1 to 0.4 MMSCFD) decreases the water content of sales gas at all MEG circulation rates (2–10 GPM) consistently. Moreover, increased MEG circulation rates result in a further reduction in the water content, as depicted in Figure 3, owing to an augmented water absorption in the glycol contactor. As an example, at a stripping gas flow rate of 0.2 MMSCFD, increasing MEG circulation rate from 2 GPM to 10 GPM drives the water content of sales gas down to about 4.8 lb/MMSCF, as compared to approximately 5.5 lb/MMSCF at a 2 GPM circulation rate. This implies that the rate of MEG circulation is a significant parameter contributing to efficiency in dehydration, in addition to stripping gas flow rate (see Section 5.3.2 for an in-depth explanation on MEG circulation rate).

Similarly, Figure 4 examines the impact of stripping gas flow rate on sales gas water content at different inlet feed gas temperatures, with a fixed MEG circulation rate of 4 GPM and a regenerator temperature of 145 °C. The outcome affirms that the higher the stripping gas, the lower the sales gas water applicable content, as was the trend in Figure 3. Nevertheless, an increase in the temperature of the inlet feed gas (25–45 °C) results in the growth of water content of sales gas, since the warmer the gas, the higher its capacity to carry water; therefore, dehydration becomes increasingly difficult [31]. An example is when the sales gas water content changes in response to a stripping gas flow rate of 0.3 MMSCFD as the gas goes through 25 °C and 45 °C to produce a sales gas water content of about 4.9 lb/MMSCF and 5.3 lb/MMSCF, respectively. It emphasizes the value of regulating inlet feed gas temperature to optimize dehydration performance, especially when augmented by manipulating stripping gas flow rate (see Section 5.3.3, where inlet feed gas temperature is further discussed).

Figure 5 provides additional insight into the impact of stripping gas flow rate on sales gas water content at varying MEG reboiler temperatures, with a fixed MEG circulation rate of 6 GPM and an inlet feed gas temperature of 35 °C. The results again confirm that increasing the stripping gas flow rate reduces the sales gas water content. However, it is also evident from this figure that the MEG reboiler temperature has only a minor effect on sales gas water content, suggesting that the influence of stripping gas flow rate is significantly more dominant in optimizing dehydration performance. At lower stripping gas flows, the MEG reboiler temperature can have a very small effect on sales gas water content via improving water vaporization, while at stripping gas flows above 0.25 MMSCFD, the stripping gas largely drives water removal, so reboiler temperature changes can be neglected as lean MEG purity is maximized.

In general, the results in Figure 4 and Figure 5 highlight the synergistic consequences of stripping gas flow rate, MEG circulation rate, and inlet feed gas temperature on sales gas water content. Stripping gas flow rate and MEG circulation rate should be increased to increase the efficiency of water removal, and inlet feed gas temperatures should be reduced to catalyze dehydration and decrease the water carrying capacity of the gas. These results highlight the fact that the set of these parameters will need to be optimized as a combination to obtain the required sales gas specifications, with the preservation of process efficiency and minimization of BTEX emissions.

5.3.2. Effect of MEG Circulation Rate

The simulation results provide a comprehensive understanding of the influence of MEG circulation rate on sales gas water content under various operational conditions. With a feed gas temperature of 40 °C and an MEG regeneration temperature of 145 °C, Figure 6 illustrates the impact of MEG circulation rate on the water content of the sales gas at various stripping gas flow rates. Because a greater circulation rate enhances the absorption of water vapor from the natural gas stream, the results indicate that increasing the MEG circulation rate reduces the water content of the sales gas. Furthermore, the results show that increasing the stripping gas flow rate further decreases the sales gas water content, highlighting the combined effectiveness of these parameters in enhancing the dehydration process.

Figure 7 examines the influence of MEG circulation rate on sales gas water content at varying inlet feed gas temperatures, with a constant stripping gas flow of 0.3 MMSCFD and an MEG reboiler temperature of 145 °C. The results reveal that an increase in inlet feed gas temperature leads to higher sales gas water content, as warmer gas holds more moisture. However, increasing the MEG circulation rate mitigates this effect, with a more pronounced reduction in sales gas water content at higher temperatures. This demonstrates the importance of optimizing the MEG circulation rate, particularly under conditions of elevated inlet feed gas temperatures.

Figure 8 explores the relationship between MEG circulation rate and sales gas water content at different MEG reboiler temperatures, with a constant stripping gas flow of 0.1 MMSCFD and an inlet feed gas temperature of 25 °C. The results show that increasing the MEG circulation rate consistently decreases the sales gas water content, confirming the critical role of this parameter in the dehydration process. However, variations in MEG reboiler temperature have a negligible impact on the sales gas water content, suggesting that the MEG reboiler temperature is not a key factor in determining dehydration efficiency.

Overall, the results underscore the significance of MEG circulation rate as a primary operational parameter for controlling sales gas water content. While stripping gas flow rate and inlet feed gas temperature also play important roles, the MEG reboiler temperature has minimal influence. These findings highlight the need to prioritize the optimization of MEG circulation rate in combination with stripping gas flow rate to achieve efficient and effective dehydration of natural gas.

5.3.3. Effect of Inlet Feed Gas Temperature

The current study examines the influence of inlet feed gas temperature on sales gas water content using HYSYS simulations under various operating conditions. Specifically, the analysis focuses on how changes in inlet feed gas temperature affect sales gas water content at different MEG reboiler temperatures, while maintaining a constant stripping gas flow rate of 0.4 MMSCFD and an MEG circulation rate of 6 GPM as shown in Figure 9.

The simulation results reveal that increasing the inlet feed gas temperature leads to a rise in the water content of the sales gas. This behavior can be attributed to the higher water carrying capacity of the gas at elevated temperatures, which makes achieving the desired dehydration efficiency more challenging [31,39]. In contrast, variations in the MEG reboiler temperature appear to have a negligible impact on the sales gas water content. This suggests that within the studied range, the reboiler temperature primarily influences the regeneration of the MEG solution rather than directly affecting the dehydration process.

These findings emphasize the importance of controlling inlet feed gas temperature to meet target sales gas specifications. Although the MEG reboiler temperature is critical for maintaining MEG regeneration efficiency, its direct effect on water content in the sales gas is limited under the given conditions. This highlights the need to prioritize upstream temperature management for optimal dehydration performance.

This work also investigates the impact of inlet feed gas temperature on sales gas water content under various operating conditions, with a focus on MEG circulation rate and stripping gas rate. The simulations provide insight into how these parameters interact and influence the dehydration process. Figure 10 illustrates the relationship between inlet feed gas temperature and sales gas water content for different MEG circulation rates, while keeping the stripping gas molar flow at 0.1 MMSCFD and the MEG reboiler temperature at 155 °C. The results indicate that an increase in inlet feed gas temperature leads to higher sales gas water content, likely due to the greater water holding capacity of the gas at elevated temperatures. Conversely, increasing the MEG circulation rate reduces the sales gas water content, suggesting improved water removal efficiency with higher MEG flow rates.

The effect of inlet feed gas temperature on sales gas water content at different stripping gas rates is presented in Figure 11, with a fixed MEG circulation rate of 2 GPM and MEG reboiler temperature of 155 °C. Results confirm that raising inlet feed gas temperature results in increased water content in the sales gas. However, increasing the stripping gas rate effectively decreases the water content, indicating that higher stripping gas flow enhances the removal of water from the gas phase.

These results highlight the interplay between inlet feed gas temperature, MEG circulation rate, and stripping gas rate in determining sales gas water content. While elevated feed gas temperatures pose a challenge to dehydration efficiency, increasing the MEG circulation rate or stripping gas rate can mitigate this effect by enhancing the system’s water removal capacity. Therefore, optimizing operating parameters is important to achieve the desired dehydration performance while maintaining process efficiency.

5.3.4. Effect of MEG Regenerator Temperature

This study considers the effect of MEG reboiler temperature on sales gas water content under varying operating conditions, including changes in stripping gas rate, MEG circulation rate, and inlet feed gas temperature. The obtained results are based on sensitivity analyses performed using HYSYS simulations. Figure 12 presents the impact of MEG reboiler temperature on sales gas water content at different stripping gas rates, with a main feed gas temperature of 35 °C and MEG circulation rate of 8 GPM. The results indicate that variations in MEG reboiler temperature have a negligible effect on the water content of the sales gas. However, increasing the stripping gas rate significantly reduces the sales gas water content. This reduction is attributed to the effect of stripping gas on MEG concentration; a higher stripping gas rate decreases the MEG concentration, improving its absorption efficiency and enhancing the water removal from the gas phase.

Figure 13 shows the effect of MEG reboiler temperature on sales gas water content at different MEG circulation rates, with a stripping gas rate of 0.3 MMSCFD and a main feed gas temperature of 45 °C. The simulation results confirm that changes in MEG reboiler temperature have little to no impact on the water content of the sales gas. Conversely, increasing the MEG circulation rate significantly decreases the water content, demonstrating that a higher MEG flow improves water removal efficiency by increasing the contact area and absorption capability. The regenerator temperature has a more pronounced effect on BTEX volatilization at a low circulation rate of 2 GPM due to the increased concentration of absorbed BTEX in the glycol resulting from the decreased solvent flow. In particular, a tiny but discernible rise in emissions (approximately a fraction of SCFD) in the vapor phase leaving the flash separator and regenerator results from the higher temperatures’ enhancement of the vaporization of BTEX chemicals in the regenerator. This effect is lessened at larger circulation rates (4–10 GPM) because BTEX is diluted in the glycol, which is consistent with the generally insignificant effect on emissions under the conditions under study. These results emphasize how crucial it is to keep MEG circulation rates low to reduce BTEX emissions, especially while running at higher regenerator temperatures.

Figure 14 highlights the effect of MEG reboiler temperature on sales gas water content at different inlet feed gas temperatures, at a stripping gas rate of 0.4 MMSCFD, and an MEG circulation rate of 6 GPM. The results again show that MEG reboiler temperature has a negligible influence on sales gas water content. However, an increase in inlet feed gas temperature results in higher sales gas water content due to the elevated water carrying capacity of the gas at higher temperatures.

These findings emphasize that MEG reboiler temperature plays a limited role in directly influencing sales gas water content under the studied conditions. Instead, the stripping gas rate, MEG circulation rate, and inlet feed gas temperature are more critical parameters. Increasing the stripping gas rate and MEG circulation rate improves dehydration efficiency, while elevated inlet feed gas temperatures negatively impact the process by increasing water content in the sales gas. Optimizing these operational parameters is essential for achieving effective and efficient dehydration performance.

5.4. Regression Analysis and Developed Correlations

One objective of this study is to develop two correlations that can be used to depict the impact of the independent variables (stripping gas flow rate, inlet feed gas temperature, MEG reboiler temperature, MEG circulation rate) of the gas dehydration process on both BTEX emissions and sales gas water content. Regression analysis is a statistical technique used to identify and measure the connections between variables, using real-world data collected from experiments. The significance of these relationships is determined using the analysis of variance (ANOVA) test. The two derived correlations for estimating sales gas water content (in Lb/MMSCF) and BTEX emissions (in MSCFD) are presented in Equation (1) and Equation(2), respectively.

$$\mathrm{S}\mathrm{a}\mathrm{l}\mathrm{e}\mathrm{s} \mathrm{g}\mathrm{a}\mathrm{s} \mathrm{w}\mathrm{a}\mathrm{t}\mathrm{e}\mathrm{r} \mathrm{c}\mathrm{o}\mathrm{n}\mathrm{t}\mathrm{e}\mathrm{n}\mathrm{t}=27.96-18.29\mathrm{A}+16.31{\mathrm{A}}^2-0.596\mathrm{C}+0.026{\mathrm{C}}^2-1.056\mathrm{E}+0.022{\mathrm{E}}^2-0.053\mathrm{G}$$

(1)

$$\mathrm{B}\mathrm{T}\mathrm{E}\mathrm{X} \mathrm{e}\mathrm{m}\mathrm{i}\mathrm{s}\mathrm{s}\mathrm{i}\mathrm{o}\mathrm{n}\mathrm{s}=-0.006+0.000289\mathrm{A}+0.000331A^2+0.000784\mathrm{C}-0.0000228C^2+0.000223\mathrm{E}-0.000001739{\mathrm{E}}^2+0.00000904\mathrm{G}$$

(2)

where A is stripping gas molar flow in MMSCFD, C is MEG circulation rate in GPM, E is main feed gas temperature in °C, and G is MEG reboiler temperature in °C.

Key Assumptions:

• Steady-state conditions, pressure drop is neglected.
• Peng–Robinson EOS is used.
• Constant purity of lean MEG.
• Analogous feed gas composition.

Limitations:

• Developed for lean gas (<0.4 mol% BTEX); emissions may be underestimated in higher concentration feeds.
• Rely on standard NGDU design; non-standard arrangement may require adapting.
• Valid within specified ranges; extrapolation may risk inaccuracies.
• Suppose steady-state operation and MEG purity (99.5 wt%).
• Adjusted for TEG and MEG; not valid for TEG/DEG.

To assess the alignment between experimental data and predicted values from correlations, the R² statistical test was employed. This test quantifies the model’s predictive accuracy on a scale from 0 to 1, with higher values indicating a better fit [40]. As highlighted by Mapiour et al. [41], an increased R² value signifies a more accurate representation of the experimental results. In this analysis, the first correlation yielded an R² of 0.93, while the second achieved 0.96. These results demonstrate a strong agreement between experimental measurements and the proposed correlations, thereby validating the reliability of the equations within the examined operating conditions.

The study introduces two new correlations designed to estimate the water content and hydrocarbon dew point temperature of natural gas, offering significant practical utility for process engineers and operators. These correlations stand out for their simplicity and ease of use, requiring no complex calculations or specialized software. This makes them particularly valuable for quick, on-site assessments of two key parameters in natural gas dehydration plants. Moreover, they enable real-time adjustments and reduce trial-and-error costs. It is worth highlighting that the generated correlations remain valid within the specific range of operating conditions examined in this study (stripping gas rate: 0.1–0.4 MMSCFD, MEG circulation: 2–10 GPM, inlet temperature: 25–45 °C, reboiler temperature: 145–160 °C). Accuracy may decline outside these limits.

One key novelty of the study lies in its focus on developing correlations specifically tailored for absorption dehydration plants that utilize MEG as a drying agent. Previous research has primarily concentrated on plants using TEG as the solvent, leaving a gap in the literature for MEG-based systems. Additionally, the study addresses BTEX emissions, a topic not extensively explored in prior works. By introducing a correlation for estimating BTEX emissions, the study provides a valuable tool for environmental monitoring and regulatory compliance in natural gas processing.

The newly proposed correlations are user-friendly and involve straightforward mathematical expressions, allowing for quick calculations using a basic calculator. This contrasts with existing correlations, such as those developed by Bahadori and Vuthaluru [42], which often require tuned coefficients and rely on multiple assumptions to estimate parameters like the natural gas dew point. The simplicity of the current study’s correlations reduces the likelihood of errors and makes them applicable across a wide range of operating conditions in natural gas dehydration units (NGDUs).

In summary, the correlations introduced in this study fill a significant gap by providing practical tools for MEG-based dehydration systems and addressing BTEX emissions. Their simplicity, versatility, and broad applicability under varying operating conditions make them an innovative contribution to the field, enhancing both operational efficiency and environmental compliance in natural gas processing.

5.5. Operating Conditions Optimization

The results of this study highlight the potential for enhancing the dehydration process in MEG-based absorption systems by optimizing operating parameters. A comprehensive analysis of the system revealed that operating conditions significantly influence the dew point of the produced sales gas, thereby affecting both environmental and operational outcomes. It is beneficial to identify optimal operating conditions that would simultaneously address two critical goals:

• Minimizing BTEX Emissions: BTEX emissions from the MEG regeneration process pose a significant environmental challenge. The study emphasizes the importance of regulating parameters such as stripping gas rate, MEG reboiler temperature, and MEG circulation rate to minimize BTEX volatilization during regeneration. By achieving lower emissions, the dehydration plant can align with environmental regulations and reduce its ecological footprint.
• Reducing Natural Gas Water Content: Achieving acceptable water content in sales gas is essential to meet the desired specifications for pipeline transport and end-use applications [1,31]. This operational objective requires fine-tuning operating parameters such as inlet feed gas temperature, stripping gas rate, and MEG circulation rate. The simulation results demonstrate that adjusting these parameters enhances the system’s dehydration efficiency, ensuring that the sales gas meets the required dew point and water content specifications.

The current study investigates the optimization of operating parameters in natural gas dehydration units (NGDUs) to minimize BTEX emissions from the MEG regeneration package while ensuring the sales gas water content remains below 2 lb./MMSCFD. Two methods were employed to achieve this optimization: LINGO optimization software (version 18) and HYSYS process simulation software (version 14).

Using LINGO, the optimization problem was formulated through a mathematical model represented by Equations (3)–(10). The objective function was to minimize BTEX emissions while maintaining the sales gas water content within the desired limit. By solving this model, LINGO identified the optimal operating conditions for the NGDU, focusing on parameters such as MEG circulation rate, stripping gas rate, and reboiler temperature.

min = BTEX emissions

(3)

$$BTEX emissions=-0.006+0.000289A+0.000331A^2+0.000784C-0.0000228C^2+ 0.000223E-0.000001739E^2+0.00000904G$$

(4)

This is subject to the following constraints:

Sales gas water content <=4;

(5)

$$Sales gas water content=27.96-18.29A+16.31A^2-0.596C+0.026C^2-1.056E+0.022E^2-0.053G$$

(6)

Stripping gas flow rate constraint:

0.1 ≤ A ≤ 0.4;

(7)

TEG circulation rate constraint:

2 ≤ C ≤ 10;

(8)

Inlet feed gas temperature constraint:

25 ≤ E ≤ 45;

(9)

TEG regeneration temperature constraints:

145 ≤ G ≤ 160;

(10)

HYSYS, on the other hand, provided a simulation-based approach to analyze and validate the effects of these operating parameters on system performance. Through its dynamic process modeling capabilities, HYSYS allowed for a detailed assessment of how changes in operating conditions impact BTEX emissions and the dehydration efficiency of the NGDU.

The obtained results indicate that both optimization techniques (LINGO and HYSYS) produced identical values for the optimum operating conditions of the natural gas dehydration unit (NGDU). This consistency between the two methods reinforces the reliability and robustness of the proposed optimization approaches.

The global optimum solution identifies the set of operating conditions that minimize BTEX emissions while ensuring that the sales gas water content remains below 2 lb./MMSCFD. The optimal conditions were determined to be a stripping gas flow rate of 0.1 MMSCFD, a TEG circulation rate of 2 GPM, an inlet feed gas temperature of 25 °C, and a TEG regenerator temperature of 145 °C.

When these optimal parameters were implemented in the NGDU, the performance improved significantly. The sales gas water content was reduced to 1 lb./MMSCFD, well within the required specification, demonstrating the system’s dehydration efficiency under the optimized conditions. Additionally, the absence of BTEX emissions from the plant under these conditions highlights the effectiveness of the optimization in addressing environmental concerns.

5.6. Cost–Benefit Analysis of TEG and MEG Systems

In this part, a cost–benefit analysis is conducted to assess the switch from TEG to MEG. A comparative analysis covers the capital investment, operational expenses, and savings resulting from reduced emissions and maintenance.

Transitioning from a TEG- to an MEG-based system involves minimal extra capital costs because of the equipment compatibility (USD 10,000–USD 22,000). These costs correspond to a one-time cost for system flushing and cleaning in order to remove the residual TEG [12,15], recalibration cost for MEG’s lower reboiler temperature [1,15], and minor tweaks for MEG’s lower viscosity cost [12]. The total annualized operational cost when MEG is used in place of TEG is in the range of USD 27,379–USD 34,879/yr, which corresponds to a reduction by about 40–42% compared to the TEG system. This reduction results from the lower replacement and maintenance costs, in addition to the operation at lower temperatures [1,12].

BTEX emissions from the TEG system, which are equivalent to 3943 SCFD, require costly control measures (e.g., incinerators, condensers); this is equivalent to about USD 3,449,580–USD 6,899,160/yr. These expenses are nearly offset when the MEG system is carried out due to the elimination of BTEX emissions. By comparing the required one-time capital cost with the annual cost savings when the MEG system is operated, it is found that the payback period is less than one month if reduced emission costs are taken into consideration. It could reach 3–11 months, if only operational savings are considered. Over 10 years, net savings could be USD 34,718,900–USD 69,349,700. Extra advantages include regulatory compliance, operational simplicity, and sustainability [1,11,12,15].

This analysis proves that switching to an MEG-based system is highly cost-effective, offering noticeable savings and environmental benefits, making it an engaging choice for NGDU collaborators.

5.7. Environmental Benefits and Trade-Offs of MEG-Based Dehydration

Switching from TEG- to MEG-based systems in natural gas dehydration units introduces remarkable sustainability benefits. MEG excessively lowers BTEX emissions, almost getting rid of risky air contaminants that contribute to smog and health threats. It also needs fewer regeneration temperatures, reducing reboiler duty, utility costs, and CO₂ emissions while cutting threats of water and soil contamination [1,12,15]. When the two systems are compared from the viewpoint of lifecycle, it is worth noting that MEG production is simpler and slightly less carbon intensive. Furthermore, if bio-based MEG (e.g., from bio-based raw materials, like plant-derived ethanol) is adopted, lifecycle GHG reductions could reach 70–83% compared to fossil-based equivalents [15].

From the viewpoint of energy consumption, as mentioned in this work, MEG regeneration occurs at lower temperatures; this leads to a noticeable reduction in reboiler duty as well as utility costs. This accordingly reduces fuel needs and CO₂ emissions, avoiding tons of annual CO₂ emissions per unit. Over its lifecycle, MEG consequently offers a lower carbon footprint, especially in fossil fuel-dependent regions [1,15].

Overall, MEG-based dehydration units lower emissions and energy use, aiding climate and health targets. Combining bio-based MEGs could further enhance sustainability and guarantee a more balanced transition away from TEG.

6. Conclusions

The gas processing industry faces the dual challenge of minimizing BTEX emissions while maintaining efficient control over sales gas dew point and water content. This study addresses these issues by exploring alternative solutions, including the use of MEG as an alternative glycol and modifying the process design. The research was conducted through four simulation cases using HYSYS (version 14) to investigate the influence of operating variables on BTEX emissions and sales gas water content for the considered NGDU located in Egypt (the Western Desert).

The research evaluates four approaches to optimize BTEX emission mitigation in natural gas dehydration. These include TEG-based dehydration, TEG with an upstream cooler, MEG-based dehydration, and MEG with an upstream cooler. Each method is assessed for its impact on emission reduction and dehydration performance to determine the most sustainable and efficient solution. Among the four analyzed scenarios, the MEG-based system without additional cooling emerged as the most economical and environmentally friendly solution, achieving significant reductions in BTEX emissions while maintaining sales gas water content within desired specifications. Upgrading existing NGDUs to use MEG, rather than TEG, will have minimal capital costs associated with the nature of equipment compatibility and is projected to show significant savings in operation costs. This can be represented by 69.8% savings in reboiler duty (111,470 Btu/hr vs. 230,363 Btu/hr) and 67.7% in utility costs (0.557 USD/hr vs. 1.791 USD/hr) alongside reduced BTEX emission control expenses, which will make the upgrade economically attractive. The simulation results demonstrate that operating parameters such as stripping gas flow rate, MEG circulation rate, and inlet feed gas temperature significantly impact BTEX emissions, while MEG regenerator temperature has a negligible effect on sales gas water content.

Two new quadratic correlations were found using regression analysis. The first correlation allows for straightforward calculation of sales gas water content, while the second offers a novel approach to estimating BTEX emissions from the regenerator and flash separator. These correlations provide accurate predictions across varying operational conditions, filling a gap in previous research.

It is found that BTEX emissions were effectively eliminated at a stripping gas flow rate of 0.1 MMSCFD, a TEG circulation rate of 2 GPM, an inlet feed gas temperature of 25 °C, and a TEG regenerator temperature of 145 °C. Under these conditions, the sales gas water content was reduced to 1 lb./MMSCFD.

This research not only achieves the goal of minimizing BTEX emissions and ensuring compliance with sales gas specifications but also establishes a framework applicable to other natural gas dehydration processes. By enhancing profitability and reducing emissions to meet environmental regulations, the results of this study offer a practical and sustainable approach to improving natural gas dehydration performance.

In future studies, the long-term operation consequences of MEG-based dehydration on system efficiency and equipment maintenance, including corrosion, fouling, and energy requirements, will also be examined to provide continuous performance and environmental compliance.