Advancing Hybrid Cryogenic Natural Gas Systems: A Comprehensive Review of Processes and Performance Optimization

Abstract

Recent research in the liquefied natural gas (LNG) industry has concentrated on reducing specific power consumption (SPC) during production, which helps to lower operating costs and decrease the carbon footprint. Although reducing the SPC offers benefits, it can complicate the system and increase investment costs. This review investigates the thermodynamic parameters of various natural gas (NG) liquefaction technologies. It examines the cryogenic NG processes, including integrating NG liquid recovery plants, nitrogen rejection cycles, helium recovery units, and LNG facilities. It explores various approaches to improve hybrid NG liquefaction performance, including the application of optimization algorithms, mixed refrigerant units, absorption refrigeration cycles, diffusion–absorption refrigeration systems, auto-cascade absorption refrigeration processes, thermoelectric generator plants, liquid air cold recovery units, ejector refrigeration cycles, and the integration of renewable energy sources and waste heat. The review evaluates the economic aspects of hybrid LNG systems, focusing on specific capital costs, LNG pricing, and capacity. LNG capital cost estimates from academic sources (173.2–1184 USD/TPA) are lower than those in technical reports (486.7–3839 USD/TPA). LNG prices in research studies (0.2–0.45 USD/kg, 2024) are lower than in technical reports (0.3–0.7 USD/kg), based on 2024 data. Also, this review investigates LNG accidents in detail and provides valuable insights into safety protocols, risk management strategies, and the overall resilience of LNG operations in the face of potential hazards. A detailed evaluation of LNG plants built in recent years is provided, focusing on technological advancements, operational efficiency, and safety measures. Moreover, this study investigates LNG ports in the United States, examining their infrastructures, regulatory compliance, and strategic role in the global LNG supply chain. In addition, it outlines LNG’s current status and future outlook, focusing on key industry trends. Finally, it presents a market share analysis that examines LNG distribution by export, import, re-loading, and receiving markets.

1. Introduction

Natural gas (NG) extracted from underground reservoirs often contains various impurities, such as carbon dioxide (CO₂), hydrogen sulfide (H₂S), nitrogen (N₂), and other contaminants (e.g., suspended particulates, wastewater, and condensates) [1,2,3]. Unrefined (unprocessed) NG is sourced from three types of wells: oil reservoirs, gas deposits, and condensate zones [4,5]. When NG is extracted alongside crude oil, it is referred to as associated gas. This gas can exist independently as free gas within the reservoir or be dissolved within the crude oil [6]. Also, NG obtained from gas and condensate zones, where crude oil is minimal or absent, is called non-associated gas. Gas wells primarily yield NG, while condensate zones produce NG and liquid hydrocarbon condensates. Regardless of its origin, once NG is separated from crude oil, it typically exists alongside other hydrocarbons such as ethane (C₂H₆), propane (C₃H₈), butane (C₄H₁₀), and pentane (C₅H₁₂) [7,8]. NG is typically treated before it can be delivered to markets. Gas processing (treatment) encompasses a series of industrial techniques to purify raw gas by eliminating or reducing impurities, contaminants, and heavier hydrocarbons [9]. NG processing involves condensate/water (H₂O) elimination, acid gas extraction, dehydrationremoval of moisture, mercury extraction, N₂ removal, and natural gas liquid (NGL) recovery [10,11].

In addition to these steps, installing scrubbers and heaters at or near the wellhead is often essential. Scrubbers eliminate sand and other large-particle impurities, while heaters prevent the NG temperature from dropping too low, which could form hydrates with the H₂O present in the gas stream [12]. Natural gas hydrates (NGHs) are crystalline, ice-like solids or semi-solids that can obstruct the NG flow through valves and pipelines [12,13]. Figure 1 presents a typical NG flow processing diagram. Following an initial scrubbing process to remove particulates, the first step in NG processing involves the removal of condensate (oil) and H₂O [12]. This is accomplished by regulating the temperature and pressure of the inlet stream from the well, as depicted in Figure 1. The gas that is separated in this unit proceeds to acid gas recovery; the recovered condensate or oil is typically sent to a refinery for further processing, while the H₂O is either disposed of or treated as wastewater [10]. Following sweetening and dehumidification processes, the treated and dried gas must meet specific transmission standards to be eligible for integration into the pipeline system and subsequent injection into the national grid [14]. This ensures that the gas maintains its single-phase (gaseous) state throughout the entire transmission route, thereby preventing the liquefaction of heavier hydrocarbons and, consequently, the occurrence of two-phase flow within the pipeline [15].

A decrease in pressure and temperature typically accompanies gas transmission through pipelines. These conditions can create an environment conducive to separating heavier hydrocarbons from the gas stream [16]. Heavier hydrocarbons, which possess a higher calorific value and are valuable in various conversion processes, are useful for domestic consumption and potential export commodities [17]. The feed gas is generally removed from heavier hydrocarbons (e.g., C₂H₆, C₃H₈, and C₄H₁₀, collectively referred to as NGLs) [18,19]. Efforts have been made to integrate the NGL recovery process with the liquefied NG (LNG) unit to minimize redundant equipment and leverage the benefits of a shared refrigeration system [19,20,21,22,23,24,25]. Following this treatment, the NG undergoes liquefaction through multiple stages of pre-cooling and sub-cooling, typically within compact but high-capacity plate-and-frame heat exchangers. The resultant methane-rich (CH₄-) gas is then transferred to cryogenic storage tanks and maintained at or near atmospheric pressure, in preparation for transportation [26,27,28].

British chemist Michael Faraday was the first to conduct experiments on natural gas liquefaction. In 1873, a German engineer, Carl von Linde, constructed the earliest compression-based refrigeration apparatus [29,30,31]. The initial experimental facility for storing LNG was developed in Virginia, United States of America (USA), in 1912. By 1941, the first commercial natural gas liquefaction plant was established in Cleveland, Ohio, USA, featuring storage tanks operating at atmospheric pressure [32,33]. In 1959, Methane Pioneer, the inaugural experimental LNG carrier, started on its journey from the Louisiana Gulf to the United Kingdom (UK) [34,35]. In 1964, the UK and France received the first commercial delivery of Algerian LNG, with Algeria’s Arzew facility recognized as the initial large-scale liquefaction terminal [33,34,36,37]. By 1969, further LNG trades were conducted between Algeria and France, Libya and Italy, and Spain and Cook Bay in Japan, initiating the first LNG project in the Pacific area [33]. From 1972 to 1978, the USA debuted in the LNG market by launching four regasification facilities. In 1973, Japan began importing LNG from Brunei, signifying the start of the Pacific region’s critical role in the LNG industry, with Korea and Taiwan later emerging as key importers [33]. Royal Dutch Shell plc. has a significant role in the industry, with stakes in over 30% of global LNG production [38]. While Shell’s proprietary dual mixed refrigerant (DMR) technology was used primarily at the Sakhalin-2 LNG plant, most of Shell’s LNG operations utilize technologies from companies like Linde plc., ConocoPhillips, and Air Products and Chemicals, Inc. (APCI) [38]. Linde plc. oversees small- to mid-scale LNG plants using a single mixed refrigerant (SMR) technology or a N₂ double-expansion cycle, with a capacity of 6.7 million tons per annum (MTPA). Major facilities in Norway and Venezuela use Linde’s mixed fluid cascade (MFC) process, each producing approximately 4.3 MTPA [38]. ConocoPhillips’ optimized cascade (CPOC) process is implemented in large-scale LNG trains, delivering 113.9 MTPA by 2024 [39].

Among the liquefaction facilities that commenced operations in 2023, Tangguh LNG Train 3, located in Bintuni Bay, Indonesia, utilized Air Products’ propane pre-cooled mixed refrigerant/mixed component refrigerant (AP-C3MR/SplitMR) technology. Figure 2 compares the global capacity of the various liquefaction technologies installed and approved from 1961 to 2029. Air Products’ liquefaction technologies currently lead the market in liquefaction processes, with approximately 67% of the total operational capacity in 2023. Among these, AP-C3MR and AP-C3MR/SplitMR technologies contribute about 57% of the global operational capacity [39]. The installed and approved liquefaction technologies in LNG processes from 1961 to 2029 included Shell Technologies, BHGE Technologies, Linde Technologies, CPOC, Air Products Technologies, AP-X, AP-C3MR/SplitMR, AP-C3MR, and others (e.g., DMR and MFC) [39,40,41,42,43].

LNG production, similar to other liquefaction processes, is highly energy-intensive. Current research in the LNG industry primarily focuses on reducing specific power consumption (SPC) during production, ultimately lowering project costs and increasing economic viability [19,20,44,45,46,47]. Due to the high costs associated with cryogenic systems and the complex interactions within the ultra-cold integrated NG processes, extensive research has been conducted on optimizing and integrating these systems with the core production process [48,49,50,51]. Integrating LNG, NGL, and NRU units eliminated reboilers and condensers within the distillation columns and used shared refrigeration systems for utility supply [23,52]. A technique was introduced for extracting CH₄, N₂, and NGL from an NG feed stream, with the extraction process conducted under high-pressure conditions by incorporating a N₂ removal unit (NRU) [53]. This process comprised a distillation column powered by a heat pump and a dedicated stage for handling NGLs. N₂ removal occurred over a broad concentration range, from 1% to 80% of the input stream. N₂ was separated from NG through cryogenic distillation by McNeil et al. [54]. The feed entered a system of two columns at similar pressures. CH₄-rich liquid from the primary column was partially vaporized, and condensed N₂ vapor was returned for reflux. In the secondary column, CH₄-rich liquid was vaporized to produce more CH₄, and condensed N₂ vapor was returned for cooling. Extensive research has investigated optimization analysis and pinch techniques for heat recovery in cryogenic natural gas systems, with selected studies [14,15,22,24,27,51,55,56,57,58,59,60] available in the Supplementary Materials.

Efforts to reduce SPC in LNG units are frequently desirable because they can significantly improve the efficiency and the economic viability of large-scale NG storage and transmission. Lowering SPC directly translates into lower operational costs and a more competitive LNG production process. In addition, reducing SPC decreases the carbon footprint of LNG operations, which is crucial for meeting global net-zero emission targets. Despite these advantages, most research on NG liquefaction systems has primarily concentrated on reducing the SPC. During efforts to optimize the performance of LNG units, key parameters such as capital and operational costs, system complexity, and environmental impacts are frequently disregarded. Adopting a more comprehensive approach with the growing global focus on reaching net-zero emissions is important. This approach should prioritize SPC reduction while also incorporating a comprehensive assessment of capital and operating costs, system design simplicity and scalability, and environmental impacts. To the best of our knowledge, based on an extensive literature review, no comprehensive review has been conducted to enhance the performance of cryogenic NG processes by considering factors such as SPC, energy/exergy efficiency, capital and operating prices, capacity, complexity, and associated emissions. In more detail, this review examines various NG liquefaction technologies (e.g., APCI, AP-X, CPOC, SMR, DMR, MFC, and Axens Liquefin). It systematically analyzes integrated cryogenic NG processes such as LNG/NRU, LNG/NGL, LNG/NGL/NRU, and LNG/helium recovery. It details technologies aimed at improving the performance of cryogenic NG processes, including using optimization algorithms, mixed refrigerant (MR) systems, absorption refrigeration cycles (ARCs), diffusion–absorption refrigeration cycles (DARCs), auto-cascade absorption refrigeration (ACAR) units, thermoelectric generators (TEGs), liquid air cold recovery, ejector refrigeration cycles, and the integration with renewable energy sources and waste. This review explores the financial, safety, and environmental considerations associated with different methods used in LNG systems. Moreover, it assesses the current developments in LNG technologies and identifies potential future advancements.

2. Liquefied Natural Gas Technologies

Due to the non-uniform global distribution of NG, with its reservoirs being concentrated in specific geographic regions, NG transmission is a crucial industry. It significantly contributes to the broader utilization of NG by transporting energy [61,62]. The costs, which are further determined by the source-to-destination distance, largely influence the selection of the most suitable transmission method among traditional options such as pipelines, LNG, liquid hydrocarbons, and electricity [28,63]. Transporting NG through pipelines results in lower greenhouse gas (GHG) emissions compared to LNG. However, this benefit diminishes as the distance of transportation increases. LNG becomes a more cost-effective option for onshore and offshore pipelines exceeding 4800 and 1600 km, respectively [61]. The energy usage and GHG emissions of onshore pipelines and LNG become comparable at transport distances of 13,000 and 7500 km, respectively. The limitations of using gas pipelines include inflexibility in route options, reliance primarily on long-term supply agreements, and a fixed supply capacity determined by the pressure differential within the pipeline [64,65].

The primary distinction between LNG and compressed NG (CNG) is in their storage methods. CNG is created by compressing NG to a high pressure exceeding 200 bar while maintaining ambient temperature [66]. Constructing pressure vessels for large-scale and long-distance gas transportation can be prohibitively expensive and sometimes unfeasible. Thus, the most cost-effective solution for large-scale gas exports is to liquefy and transport the gas using marine tankers [67,68].

Natural gas that is composed primarily of paraffinic hydrocarbons such as CH₄, C₂H₆, C₃H₈, and C₄H₁₀ may contain small amounts of higher hydrocarbons, CO₂, H₂S, and N₂ [69,70]. It is converted to LNG by cooling it to −162 to −164 °C, which turns it into a liquid and reduces its volume by over 600 times, allowing it to be transported globally in specially designed ships [70,71,72,73,74]. LNG is a clear, odorless, non-toxic, non-corrosive cryogenic liquid with a 0.4–0.5 kg/L density, depending on conditions. LNG floats and evaporates quickly when spilled on H₂O, eliminating the need for cleanup. It evaporates quickly without residue and only burns when its concentration in the air is 5–15% [75]. Removing acid gases such as CO₂ and H₂S before liquefaction is important for producing pure CH₄. Table S1 in the Supplementary Materials compares the physical and chemical properties of LNG, diesel, gasoline, and liquefied petroleum gas [70]. LNG has a high auto-ignition temperature of 540 °C and an extremely low flash point of −187 °C, in contrast to diesel and gasoline which have lower ignition temperatures and different flash points. LNG’s boiling point is much lower at −160 °C, and it has a flammable range of 5–15%, whereas diesel does not have a specified flammable range [70]. LNG is stored under atmospheric pressure, is non-toxic and non-carcinogenic, and poses no health risks, unlike diesel and gasoline, which are toxic and carcinogenic and can cause health issues such as eye irritation [70,76].

LNG can be categorized into three types based on density: heavy, medium, or light [69,77,78,79]. The composition of each type is shown in Table S2 of the Supplementary Materials [80,81]. Heat transfer to LNG through insulated spaces causes it to boil, creating boil-off gas (BOG) [82,83]. Most BOG forms during cargo transport due to temperature differences, LNG spraying for tank cooling, and fluid movement inside the tank (especially in rough weather) [81,82,84]. LNG processes are typically categorized based on their core process technologies, such as expander-, MR-, or hybrid-type systems [8,85]. However, a more refined classification takes into account several additional factors. These include the number of refrigeration processes involved, the type of refrigerant used (i.e., mixed or pure), the arrangement of the refrigeration process (i.e., cascade or parallel), whether an expander is employed, the nature of the refrigerant (i.e., flammable or non-flammable), the presence or absence of pre-cooling, and the type of heat exchanger utilized (e.g., spiral-wound and plate-frame exchanger) [8]. Figure 3 depicts the classification of refrigeration units in LNG processes. The classification of refrigeration systems used in LNG processes is illustrated in three primary categories [24]: single stage, cascade, and multi-stage.

The following section investigates the most common gas liquefaction technologies, including SMR, APCI, Air Products’ advanced mixed refrigerant (AP-X), DMR, CPOC, MFC, and Axens Liquefin processes, which have been effectively implemented in land-based liquefaction plants. Also, a detailed examination of technological advancements, thermodynamic efficiencies, and main challenges associated with each process is provided in this section, along with a comparative assessment of their suitability across different operational conditions and performance constraints.

2.1. SMR Processes

The SMR process encompasses the Black & Veatch Pritchard PRICO process, the Technip/Air Liquide TEALARC process, the APCI single mixed refrigerant process (AP-SMR), the Linde Multi-stage mixed refrigerant process (LIMUM), and the Kryopak Pre-cooled mixed refrigerant process (PCMR) [64]. Black & Veatch Company initially introduced the PRICO process in the 1950s. Featuring just a single throttle valve, this system is known for its simplicity, compact design, and low capital costs, making it a popular choice for small- to mid-sized LNG plants [86]. Figure 4 presents a simplified schematic of the PRICO process, which is patented by Black & Veatch. This process operates with a single cooling cycle, utilizing a refrigerant mixture of N₂, CH₄, C₂H₆, C₃H₈, and C₅H₁₂. It has been implemented in older LNG plants in Algeria, with each unit having a capacity of around 1.3 MTPY. The refrigerant fluid undergoes liquefaction after two-stage condensation and inter-stage cooling and then becomes colder and two-phased through flash evaporation. The vapor and liquid are separated and further cooled in PFHE, with the composition of the cooling fluids at different stages controlled by adjusting the liquid levels in the separators. The existing Algerian unit uses axial compressors and gas turbines [87].

However, due to equipment size constraints, large-scale LNG production using the PRICO process requires parallel systems, which increases fixed and operating costs and reduces its competitiveness compared to processes with two or three cooling cycles. As a result, the PRICO process, along with other single-stage liquefaction processes, is no longer favored for large LNG production and export units [86,87]. In addition, constructing low-capacity LNG units for gas export has become economically unfeasible with the advent of newer, higher-capacity processes. As a result, the PRICO process is primarily utilized for LNG production and storage for injection into gas supply networks during peak consumption periods.

Figure 5 illustrates the SPC characteristics of various SMR-based LNG structures. A indicates that the SPC for LNG cycles using the SMR process ranges from 0.22 to 0.48 kWh/kg LNG [86,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106]. In addition, studies report that the exergy efficiency of SMR-based LNG processes varies between 30% and 67.8% [86,97,98,100,101,107].

2.2. APCI Technologies

The APCI processes are the most widely used LNG production method worldwide and are considered an American technology [115,116,117]. Since its initial deployment in Brunei LNG in 1972, the AP-C3MR technology has held a leading position in liquefaction, representing nearly 57% of global operational capacity (including the SplitMR variation) by 2023. The expansion of AP-C3MR’s market share was largely driven by the projects undertaken by QatarGas (now QatarEnergy). Beginning with the launch of QatarGas 1 Train 1 in 1996, approximately 30 MTPA capacities have been achieved. The Damietta LNG facility was the first to implement the C3MR/SplitMR technology, which enhances the AP-C3MR system by optimizing its mechanical configuration, resulting in higher turbine efficiency [39]. Even today, it remains the preferred choice among buyers and operators. Figure 6 presents a simplified and detailed schematic of the APCI C3MR process for NG liquefaction. This process uses two cooling cycles: the C₃H₈ cycle for pre-cooling and the MR cycle for liquefaction and supercooling [115]. C₃H₈ undergoes multi-stage compression to reach cooling temperatures of −35 to −40 °C while also cooling the MRs. In the MR cycle, a refrigerant mixture undergoes compression, cooling, and liquefaction before being directed to a spiral-wound heat exchanger (SWHE) for additional gas cooling processes. The C3MR process is popular in the LNG industry for its reliability and ease of operation despite its lower thermodynamic efficiency due to using pure refrigerants [118,119,120]. Figure 7 presents the SPC characteristics of several C3MR-based LNG structures. According to the literature, the energy consumption of LNG cycles utilizing the C3MR process has been documented to be 0.2–0.41 kWh/kg LNG [14,24,38,56,88,103,106,110,121,122,123,124,125,126,127,128,129,130,131,132,133]. Also, studies report that the exergy efficiency of C3MR-based LNG processes ranges from 29.2% to 65.2% [38,56,110,126,129,131,134,135,136,137,138,139,140]. Furthermore, technical reports indicate that the energy consumption of C3MR-based LNG cycles is 0.292–0.3 kWh/kg LNG [64].

2.3. AP-X Technologies

In the past decade, APCI introduced a modification that has increased LNG plant capacities to over 6 MTPA (e.g., several liquefaction trains in Qatar, commissioned between 2008 and 2011, exceed 7.8 MTPA). The AP-X process adds a third refrigerant cycle, utilizing a N₂ expander, to handle LNG sub-cooling outside the main cryogenic heat exchanger [118]. This strategy preserves the main exchanger’s size by transferring the sub-cooling workload to the N₂ cycle, enabling liquefaction capacities of up to 10 MTPA without requiring a larger main exchanger. [118]. This larger train capacity is achieved by incorporating a N₂ expander cycle for sub-cooling, which reduces C₃H₈ flow by 20% and MR by 40% [143]. As an American innovation, the ConocoPhillips technology is a widely used method for LNG production. First implemented in 1969 at the Kenai LNG plant in Alaska, the process has since been modified and applied in Egypt, Australia, Equatorial Guinea, Angola, Australia, and the USA. The process uses three separate cooling cycles in a cascade system for NG pre-cooling, liquefaction, and supercooling. Each cycle uses a pure refrigerant in two or three pressure stages [144]. Air Products’ nitrogen expansion process (AP-N), as a compact version of Air Products’ advanced mixed refrigerant (AP-X) supercooling technology, is implemented on Petronas’ PFLNG Satu and PFLNG Dua in Malaysia, while Coral South FLNG in Mozambique utilizes the AP DMR process [39]. AP-N is the sole expander-based (EXP) technology used in offshore projects, offering simplicity and reduced equipment requirements that are advantageous over the MR process. The Golar Gimi FLNG, a retrofitted moss-type LNG carrier, will employ Black & Veatch’s PRICO technology [39].

2.4. ConocoPhillips Technologies

The optimized cascade method developed by ConocoPhillips was initially applied at Kenai LNG in the late 1960s and came back with the launch of Atlantic LNG Train 1 in 1999 [39]. Air Products’ market share in liquefaction technology dropped from over 90% in the 1980s and 1990s to 67.3% in 2023, due to the rise in CPOC technology. Now used in a 113.9 MTPA capacity, it is the second leading technology. New technologies are expected to emerge from 2024 to 2029, driven by demand for smaller LNG trains [39]. Figure 8 illustrates the process diagram of the CPOC-based LNG process. In the first cycle, C₃H₈ is used to cool the gas to −35 °C. The second and third cycles use a plate-fin heat exchanger (PFHE), with ethylene in the second cycle cooling the gas to −100 °C, and CH₄ in the third cycle for supercooling. In the 1990s, ConocoPhillips improved this process by converting the third cycle to a two-stage open cycle [145]. In this process, feed gas passes through C₃H₈ and ethylene chillers in stages. Air or cooling H₂O removes the heat and condenses C₃H₈, which in turn cools and condenses ethylene. Heavier components (i.e., NGL) are removed after some chilling stages, and the CH₄-rich feed is then cooled in the CH₄ refrigeration system. If the CH₄ refrigerant has N₂, a portion is used as fuel to prevent inert buildup. The LNG from the final flash drum is pumped to storage tanks at around 70 mbar above atmospheric pressure and −161 °C. Each refrigeration circuit operates with two 50% compressors sharing common process equipment. Brazed aluminum and core-in-kettle heat exchangers, being simpler and widely available, are used instead of the more complex proprietary SWHEs in the C3MR process [118]. The findings in the literature reveal that the energy consumption of LNG cycles utilizing the CPOC process is 0.205–0.341 kWh/kg LNG [64]. Furthermore, technical reports suggest that the energy consumption of CPOC-based LNG cycles is 0.324–0.384 kWh/kg LNG [64].

2.5. Dual-Stage Mixed Refrigerant Processes

This modified version of the C3MR process, licensed by Shell, blends American and European technologies. In the DMR process, the C₃H₈ cycle is replaced with an MR cycle (typically C₂H₆ and C₃H₈ or sometimes with C₄H₁₀) [147]. SWHEs are used instead of PFHEs, providing enhanced durability, higher efficiency in handling cryogenic conditions, and improved reliability for large-scale LNG production. The second cooling cycle is similar to that of the C3MR, using a mix of N₂, CH₄, C₂H₆, and C₃H₈. This process, used in the Sakhalin Island LNG project in Russia (4.8 MTPA), requires three SWHE converters for capacities above 5 MTPA and splits the liquefaction cycle into two parallel cycles [118]. Figure 9 shows the process flow diagram of the two-stage MR cycle. The DMR cycle features two seawater (SW) coolers, three compressors, four heat exchangers, four valves, a tee, a phase separator, and two common headers. It operates with two cycles, substituting a valve for an expander, and uses two types of MRs for NG pre-cooling, liquefaction, and sub-cooling. The pre-cooling refrigerant, which is a mixture of CH₄, C₂H₆, C₃H₈, and C₄H₁₀, cools the NG, the primary refrigerant, and itself as it circulates through the pre-cooler cold box. The main refrigerant, consisting of N₂, CH₄, C₂H₆, and C₃H₈, is further cooled in the pre-cooler cold box. This refrigerant then liquefies and sub-cools the NG while also cooling itself [148]. The NG undergoes cooling, liquefaction, and sub-cooling via heat exchangers 1 through 4, with the refrigerant reaching −160.1 °C. As the NG flows through valve 5, it expands, separating into liquid and vapor (flash gas), with the liquid being sent to storage as LNG [149]. Shell’s DMR technology is set to be implemented at CLNG Canada. In colder climates, the DMR process is particularly advantageous, as pre-cooling with MR bypasses the low-temperature pressure limits of C₃H₈. Novatek’s Arctic Cascade process, designed for cold environments, is currently in use at Yamal LNG T4 (0.9 MTPA) and will also be employed at Ob LNG (planned for 2028, 5 MTPA) [39].

Figure 10 displays the SPC characteristics of various DMR-based LNG systems. A literature review suggests that the energy consumption of LNG cycles utilizing the DMR process is 0.212–0.414 kWh/kg LNG [15,22,24,122,149,151,152,153,154,155,156,157,158,159,160,161]. Moreover, research findings indicate that the exergy efficiency of LNG processes employing the DMR method is 28.2–62.3% [15,24,155,162].

2.6. Mixed Fluid Cascade Processes

The MFC system developed by Statoil/Linde is illustrated in Figure 11. The Statoil/Linde LNG technology partnership was formed to create alternative base-load LNG plants specifically designed for challenging environments [163]. The first of these plants was deployed at the Snohvit facility on Melkoya Island, located off the coast of Hammerfest in the northern North Sea at 71° north latitude in Norway. This facility remains Europe’s only base-load gas liquefaction export plant [118].

In this process, purified NG is pre-cooled, liquefied, and sub-cooled using three separate MR cycles [64]. The pre-cooling cycle transfers cold to the NG through two PFHEs, while the liquefaction and sub-cooling cycles utilize two SWHEs with different refrigerants. The refrigerants comprise selected components such as CH₄, C₂H₆, C₃H₈, and N₂. The three refrigerant compression systems can operate independently with separate drivers or be integrated into two compression strings. This process is specifically designed for large LNG trains with capacities greater than 4 MTPA [164]. Unlike traditional cascade processes, the MFC system replaces single-component refrigerant cycles with MR cycles, improving thermodynamic efficiency and offering greater operational flexibility.

Figure 12 illustrates the SPC characteristics of various MFC-based LNG processes. According to the literature, the energy consumption of MFC-based LNG cycles is 0.196–0.423 kWh/kg LNG [18,24,51,58,154,165,166,167,168,169,170,171,172,173]. Also, a number of studies report that the exergy efficiency of MFC-based LNG processes is 51.8–62.8% [18,24,51,162,168].

2.7. Axens Liquefin Processes

The Liquefin process, developed by the French company Axens, is a two-stage MR process. The refrigerant used in each stage is a mixture of CH₄, C₂H₆, C₃H₈, C₄H₁₀, and N₂. In the first stage, the refrigerant vaporizes at three different pressure levels during heat exchange with the process gas, cooling the NG to between −50 and −80 °C and fully liquefying the refrigerant for the second stage. The second stage handles gas liquefaction and supercooling. Figure S1 in the Supplementary Materials shows a simplified schematic of the Liquefin process in LNG technologies [174]. This process uses PFHE for all heat exchange operations. It maximizes liquefied petroleum gas (LPG) recovery and reduces investment and operational costs compared to previous technologies like the C3MR process. This cost reduction is achieved using fewer and non-proprietary heat exchangers, requiring less space due to the compact unit design [8]. The Axens liquefaction process is similar to Shell’s DMR process but is more cost-effective due to simpler cooling cycles and non-exclusive equipment.

Several key characteristics distinguish the Liquefin process. The Liquefin process utilizes two MRs composed of CH₄, C₂H₆, C₃H₈, C₄H₁₀, and N₂. This design allows for easy adjustment of pre-cooling temperatures, making the process adaptable to significant variations in feed gas composition and pressure. The efficiency of the process remains consistent across a wide range of production rates (3 to 8 MTPA) [8]. Also, the process is streamlined, requiring only a limited number of control loops, simplifying operation and control.

Table 1. Comparison of main LNG production technologies: technologies, refrigeration stages, advantages, disadvantages, and performance metrics.

Technologies [References]	Refrigeration Stages	Advantages	Disadvantages	Performance Metrics
SMR [39,86,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,175]	▪ Pre-cooling: MR ▪ Liquefaction: - ▪ Sub-cooling: -	▪ Efficient layout ▪ Lower investment cost ▪ Compact structure	▪ Lower efficiency compared to multi-refrigerant systems ▪ Restricted expansion potential ▪ Use of flammable refrigerants	▪ SPC (kWh/kg LNG): 0.22–0.48 ▪ Exergy efficiency: 0.3–0.678 ▪ Capacity (MTPA): 0.6–2.5
C3MR [14,24,38,39,56,88,103,106,110,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,175]	▪ Pre-cooling: Propane ▪ Liquefaction: MR ▪ Sub-cooling: -	▪ High efficiency and widely used ▪ Better energy consumption than SMR ▪ Well-established technology	▪ Less suitable for offshore use ▪ Significant propane storage requirement ▪ Higher capital cost than SMR	▪ SPC: 0.2–0.41 ▪ Exergy efficiency: 0.292–0.652 ▪ Capacity (MTPA): 1.2–5.5
DMR [15,22,24,39,122,149,151,152,153,154,155,156,157,158,159,160,161,162]	▪ Pre-cooling: MR ▪ Liquefaction: MR ▪ Sub-cooling: -	▪ Eliminating constraints on C3 compressors ▪ Enhanced capacity ▪ Higher efficiency than SMR and C3MR ▪ Good for cold environments	▪ More complex operation and control than SMR and C3MR ▪ Requires high capital investment	▪ SPC (kWh/kg LNG): 0.212–0.414 ▪ Exergy efficiency: 0.282–0.623 ▪ Capacity (MTPA): 3.4–4.8
MFC [18,24,39,51,58,154,162,165,166,167,168,169,170,171,172,173,175]	▪ Pre-cooling: MR ▪ Liquefaction: MR ▪ Sub-cooling: MR	▪ Large throughput ▪ Optimized performance ▪ High efficiency and low SPC ▪ Suitable for modular designs ▪ Lower operational costs	▪ Significant investment requirements ▪ Complex process with multiple refrigerants ▪ Requires advanced process control	▪ SPC (kWh/kg LNG): 0.196–0.423 ▪ Exergy efficiency: 0.518–0.628 ▪ Capacity (MTPA): 1.1–4.3
CPOC [39,64,118,145,146,175]	▪ Pre-cooling: Propane ▪ Liquefaction: Ethylene ▪ Sub-cooling: Methane	▪ Simpler than C3MR ▪ Lower footprint and capital costs ▪ Suitable for medium-scale LNG	▪ Lower efficiency than C3MR and DMR ▪ Limited scalability for large plants	▪ SPC (kWh/kg LNG): 0.205–0.341 ▪ Capacity (MTPA): 3–5.2
AP-X [8,39,174,175]	▪ Pre-cooling: Propane ▪ Liquefaction: MR ▪ Sub-cooling: N2	▪ High efficiency ▪ Suitable for mega-scale LNG production ▪ Reduces SPC	▪ Complex and expensive ▪ Requires extensive process expertise	▪ Capacity (MTPA): 7.8

provides a comparative analysis of major LNG production technologies, including their refrigeration stages, advantages, disadvantages, and key performance metrics.

3. Thermodynamic Analysis

The findings from thermodynamic analyses of low-temperature NG processes are presented through established indicators or criteria. One type of thermodynamic analysis is energy analysis, where energy consumption is quantitatively assessed, and its efficiency is evaluated. Key quantitative metrics in this analysis include the SPC and coefficient of performance (COP) [126,176,177]. Additional indicators assessing energy efficiency in the process include temperature–entropy (T-S) diagrams, pressure–enthalpy (P-H) diagrams, and combined curves [178,179,180]. This section examines key criteria and tools for thermodynamic analysis of low-temperature NG processes, including SPC, COP, T-S, and P-H diagrams, pinch technology, exergy analysis, and equations of state.

3.1. Specific Power Consumption

Specific power consumption is calculated as the amount of energy used by process equipment (such as compressors, pumps, and air coolers) in kWh, divided by the quantity of LNG produced in kilograms. A lower value of this index indicates greater process efficiency. Vatani et al. [162] indicated that the MFC-Linde process has the lowest SPC, making it the most energy-efficient option studied. The DMR-APCI and C3MR-Linde processes also demonstrate strong energy efficiency, while the SMR-APCI and SMR-Linde processes, though less efficient, still have acceptable energy consumption relative to the others. The SPC (kWh/kg LNG) for the actual LNG processes of SMR-Linde and SMR-APCI is 0.3–0.4 kWh/kg LNG, while for C3MR-Linde, DMR-APCI, and MFC-Linde, it is less than 0.3 [181,182].

3.2. Coefficient of Performance

The performance of cooling cycles is measured by the COP, an index indicating the efficiency of cooling cycles and heat pumps [183,184]. In the gas industry, COP is typically 0.8 to 4, which represents the ratio of heat extracted from the cold source to the work consumed [162]. For NG liquefaction, a higher COP means more heat is removed from the gas for the same power input, indicating greater efficiency. The Carnot cycle has the least wasted work and is the ideal benchmark for evaluating a cycle’s thermodynamic efficiency [185]. Vatani et al. [162,182] demonstrated that the COP for MFC and DMR processes’ pre-cooling cycles is the highest compared to others. Also, in processes with multiple refrigeration cycles, the COP is generally higher for pre-cooling stages due to the larger temperature difference between the refrigerant and NG during heat transfer.

3.3. T-S and P-H Diagrams

These diagrams illustrate the changes in entropy and enthalpy of the MRs as they relate to temperature and pressure throughout the cycle [186,187]. The degree of deviation from the ideal state is assessed, with greater deviation directly linked to irreversibility in the process. All real-world cooling cycles experience some level of irreversibility. Contributing factors to these irreversibility outcomes include friction, heat transfer between specific temperatures in the evaporator, compressor, condenser, and transmission lines, and pressure drops in expansion valves [51,182].

3.4. Pinch Technology

Pinch analysis is a robust and efficient technique for evaluating thermodynamic processes and optimizing industrial heat exchanger networks. It allows for the planning and optimizing of thermal systems through two key methods [188,189]: designing from the base and retrofitting existing heating networks.

Retrofitting existing systems focuses on energy savings, cost reduction, and increasing unit capacity. Tjoe et al. [190] introduced the first systematic approach for optimizing thermal systems. Generally, improving heat exchanger networks using pinch analysis involves two stages: setting goals and designing solutions. Pinch technology using tools such as composite charts (CCs) and grand composite curves (GCCs) helps set goals to optimize and improve processes [191,192]. GCC charts can be specifically used to determine the required auxiliary services and heat levels. However, since CC and GCC diagrams are based on enthalpy–temperature data, they are only effective for setting thermal load targets and are not suitable for systems such as refrigeration units or turbines that involve mechanical work [193,194]. Exergy analysis is used alongside pinch technology to evaluate thermal and work-related energy. The key advantage of combining pinch technology with exergy analysis is that it overcomes pinch technology’s limitations in systems involving power or mechanical work. This approach allows for a comprehensive analysis of heat and work within the system, setting goals to improve efficiency, increase production, or reduce energy consumption before detailed design. Combined pinch and exergy analysis (CPEA) such as pinch technology involves two stages: targeting and design. The targeting phase uses tools to identify necessary improvements, while the design phase focuses on implementing these enhancements [195,196,197]. The main tools in CPEA are exergy composite curves (ECCs) and exergy grand composite curves (EGCCs), which replace the temperature axis in CC and GCC diagrams with the Carnot coefficient [198,199,200]. The area between the curves in these diagrams represents the system’s required mechanical work or production power. Heat transfer efficiency improves as the gas cooling and refrigerant heating curves come closer together. However, the smaller the minimum approach temperature in a heat exchanger, the more surface area is required for effective heat transfer, increasing the overall size and complexity of the exchanger’s design [188,201]. Therefore, this parameter cannot be reduced indefinitely. The minimum approach temperature is typically set between 1 and 3 °C [197]. In LNG technologies, the minimum temperature difference in multi-stream heat exchangers has been considered in various studies within the ranges of 0.5–2.5 °C [25,55,113,202,203,204], 2.5–3 °C [49,91,101,112,152,155,205,206,207,208,209,210,211], and >3 °C [107,212].

3.5. Exergy Analysis

The indices of lost work and exergy efficiency are employed to better compare the performance of different equipment within each process [213,214,215,216]. By simultaneously evaluating the lost work and exergy efficiency of various equipment in a process, it becomes possible to identify specific areas where energy is not utilized efficiently and is consequently wasted, which is the primary goal of exergy analysis [26,217,218,219,220]. Upon identifying these inefficiencies, strategies such as adjusting operational conditions, replacing equipment with more suitable alternatives, or altering the process design can be proposed to reduce irreversibility and enhance energy efficiency [221,222].

3.6. Equations of State

Thermodynamic models are crucial for accurately modeling and simulating LNG processes, providing detailed insights into fluid behavior, phase equilibria, and thermodynamic properties under various conditions. Among the widely used models, the Peng–Robinson (PR) [14,20,22,24,72,86,100,102,139,211,223,224,225,226,227,228,229,230,231,232,233,234,235,236,237,238] and Soave–Redlich–Kwong (SRK) [239,240,241,242,243] equations of state are notable for their balance between accuracy and computational efficiency, particularly when applied to hydrocarbon systems. While the ideal gas law [244,245,246] is effective in high-temperature and low-pressure scenarios, it is inadequate for the highly non-ideal conditions typical of LNG processes. The Antoine equation [247,248] is commonly utilized for vapor pressure calculations but is often supplemented by more robust models for handling complex systems. The Benedict–Webb–Rubin (BWR) equation [223] provides accurate predictions for scenarios demanding high precision, though it results in greater computational complexity. In designing LNG processes, the simulation errors from the PR, SRK, and BWR equations of state (EOSs) were compared, with the PR EOS being identified as the most appropriate thermodynamic model based on comparison of modeling and experimental values [249]. Furthermore, the National Institute of Standards and Technology (NIST) REFPROP database [245,250,251] is extensively used for its comprehensive thermophysical property data, employing advanced models like the Helmholtz energy equation to achieve high accuracy.

Vidal [252] reported that the PR EOS demonstrates good accuracy in predicting the properties of pure substances such as N₂, CH₄, C₂H₆, and C₃H₈. However, due to its greater complexity and additional experimental constants, the Lee–Kesler EOS was found to be more accurate than the PR EOS, making it reliable for calculations in the liquid and vapor phases of pure substances. Chen et al. [253] found the average error in calculating bubble point pressure, the vapor composition percentage in vapor–liquid equilibrium, and the specific volume of liquid for various hydrocarbon binary mixtures using the PR EOS to be 4.12, 2.08, and 3.41%, respectively. Remeljej et al. [254] and Gong et al. [255] used the PR EOS to analyze phase equilibrium and predict the enthalpy and entropy of MRs in the LNG process, which involves a mixture of hydrocarbons (C1 to C5) and N₂. Cao et al. [256] utilized the PR EOS for phase equilibrium analysis and the Lee–Kesler EOS to predict the enthalpy and entropy of the MR. Danesh et al. [257] reported that complex experimental EOSs (e.g., the BWR equation) do not offer greater reliability than cubic EOSs for phase equilibrium calculations of materials. Considering the computational simplicity of the PR EOS compared to the Lee–Kesler EOS, several studies [55,217,258,259,260,261,262,263,264] utilized the PR EOS to calculate the vapor–liquid phase equilibrium of the multi-component refrigerant, while the Lee–Kesler EOS was used to predict the enthalpy and entropy.

The thermodynamic analysis of low-temperature NG processes evaluates key performance indicators, including SPC, COP, T-S/P-H diagrams, pinch technology, exergy analysis, and EOSs. SPC quantifies energy efficiency in LNG production, with lower values indicating better performance. COP assesses cooling cycle efficiency, where higher values reflect improved effectiveness. T-S and P-H diagrams depict entropy and enthalpy variations, highlighting process irreversibility outcomes. Pinch technology, utilizing CC, GCC, ECC, and EGCC, optimizes heat exchanger networks and is often combined with exergy analysis to assess heat and work interactions comprehensively. Exergy analysis identifies inefficiencies through irreversibility outcomes and exergy efficiency, guiding modifications to enhance energy utilization. EOSs, such as PR and SRK, are widely used for phase equilibrium and thermodynamic property calculations, with PR being the most preferred due to its balance of accuracy and computational efficiency.

4. Integrated Process Structures

Heavy compounds in NG can serve as feedstock for downstream units, but due to the low temperatures involved in the process, they may also lead to solid formation. Therefore, separating these heavy compounds is essential and can yield valuable products. Additionally, NG typically contains some amount of N₂, which reduces its calorific value [15,51]. Given the negative impacts of N₂ on fuel’s calorific value, the size of transmission lines, and the capacity of gas pressure-boosting stations, N₂ is usually removed from NG when its concentration exceeds 4 mol% at high flow rates. The N₂ concentration in sales gas is typically maintained at around 2 mol% [51]. The required equipment and energy consumption can be minimized by designing and optimizing the integrated process structure of NGL recovery, NG liquefaction, and N₂ removal. Gas pre-treatment is essential for conditioning feed gas to meet the precise specifications required for LNG production. This process involves carefully controlling the gas composition to maintain N₂ levels below 1.2%, reduce H₂S to under 4 ppmv, and keep CO₂ concentrations below 50 ppmv [15,265]. Additionally, the total sulfur content must be limited to less than 10 ppmv, H₂O must be reduced to below 1 ppmv, and Hg levels must be kept at or below 0.01 µg/Nm³. Removing heavy hydrocarbons is also essential to prevent freezing issues in cryogenic heat exchangers. The gas pre-treatment process provides high-quality LNG production and efficient facility operation by complying with appropriate specifications [265]. This section explores hybrid cryogenic NG systems, including nitrogen separation, LNG/NRU hybrid units, LNG/NGL hybrid units, LNG/NGL/NRU hybrid units, and hybrid helium recovery systems.

4.1. Nitrogen Separation from Natural Gas

NG typically contains some level of N₂, which lowers its calorific value but generally does not pose significant issues. However, high N₂ concentrations can substantially impact heat capacity in certain gas deposits. In such cases, three potential strategies can be employed: (1) blending with a higher-calorific-value gas to maintain overall energy content, (2) accepting a lower market price along with associated financial risks, or (3) removing nitrogen to meet the required sales specifications [266]. Options (1) and (2) are practical and sensible solutions, but their viability depends on the specific characteristics of the gas processing and production site [266].

NRU technologies employ various methods to separate N₂ from NG. Cryogenic fractionation utilizes low temperatures for separation, while membrane technology relies on selective permeability. The molecular gate technique traps N₂ on a solid surface, whereas solvent absorption removes it by dissolving it in a liquid. N₂ sponges, made from advanced porous materials such as metal–organic frameworks and zeolites, feature precisely engineered pore structures that selectively adsorb nitrogen while allowing other gases to pass through. Cryogenic lean oil absorption operates at low temperatures, using absorbent liquids to selectively capture heavier hydrocarbons while allowing N₂ and lighter gases to remain in the vapor phase [267,268].

Pressure swing adsorption (PSA) relies on alternating pressure cycles to selectively adsorb and desorb gases. Specialized adsorbents, such as zeolites or activated carbon, trap N₂ at higher pressures while releasing it at lower pressures, enabling efficient separation [269,270,271]. Chelating solvent absorption chemically binds N₂ for the removal [272]. In other words, chelating solvent absorption uses chelating agents to selectively bind and remove N₂ from NG. This process relies on solvents that preferentially absorb N₂, facilitating its separation from CH₄ and other hydrocarbons. A major advancement in chelating solvent absorption for N₂ separation is the use of transition metal complexes [268,273]. Li [273] introduced a Ru-based transition metal complex aqueous solution that selectively bonds N₂ over CH₄. It offers key benefits such as selective bonding (0.5 mole N₂ per mole Ru²⁺), reversibility confirmed by desorption tests, and moderate absorption energy (20–70 kJ/mol) for efficient regeneration. Recent studies have investigated deep eutectic solvents combined with other materials to improve CO₂/N₂ separation [274]. Although not primarily aimed at N₂ capture, this approach highlights the potential of advanced solvent systems in gas separation processes. Hybrid systems combining absorption with membrane separation improve N₂ removal efficiency [268]. A high-pressure absorption column separates hydrocarbons, while membrane technology recovers the remaining components, maximizing hydrocarbon recovery and reducing N₂ content.

Table 2. A comprehensive summary comparison of the NRU technologies, modified from Ref. [272].

NRU Technology	Key Features	Applications/Limitations	Remarks
Cryogenic fractionation	▪ Utilizes J-T or expander methods, cooling, and distillation at cryogenic temperatures. ▪ Needs recompression. ▪ Uses a cold box with BAHE.	▪ Suitable for a wide range of inlet gas pressures and flow rates. ▪ May be less efficient for low gas throughput. ▪ Handles very low CH4 concentrations (100 PPM to 1.5%) in N2 vent streams.	▪ Can recover >99% HC. ▪ Requires pre-treatment including compression, AGRU, molecular sieve dehydration, and MRU, with further cryogenic distillation and recompression. ▪ Widely used in commercial applications. Some companies such as APCI, Bechtel, Linde, and KBR can provide EPC services globally.
Membranes	▪ Uses single or multiple membrane modules to separate N2 from HCs. ▪ May require recompression, especially for multiple membrane installations.	▪ Design pressure capped at 85 barg. ▪ Maximum design capacity is 100 MMSCFD per train.	▪ Recovers about 90% HC, depending on N₂ concentration and pressure. ▪ Pre-treatment is typically not needed, though CO2 removal might be necessary based on CO2 concentration in the feed.
Molecular gate	▪ Similar to molecular sieve adsorption technology. ▪ Recompression is likely required.	▪ Maximum design pressure of 85 barg. ▪ Operates optimally at 17–41 barg.	▪ Recovers around 90% HC. ▪ Requires pre-treatment, including inlet handling, AGRU for CO2 removal, and molecular sieve dehydration. Can remove N2 and CO2 in a single step.
Solvent absorption	▪ Separates HCs from N2 using a solvent. ▪ HCs are released from the solvent by pressure reduction in multiple gas decompression steps.	▪ Operates at a maximum pressure of 70 barg. ▪ Maximum capacity is 5 MMSCFD per train.	▪ Recovers > 99% HC. ▪ Some commercial success with N2 contents as high as 50 mol%.
N2 sponge	▪ Absorbs H2O and N2.	▪ Design pressure limited to 4 barg. ▪ Throughput limited to 5 MMSCFD per train.	▪ Recovers > 92% HC. ▪ Pre-treatment is generally not required except for inlet handling.
Cryogenic lean oil absorption	▪ CH4 is absorbed into cryogenic lean oil.	▪ No information on commercial programs.	▪ Requires high recompression pressure for HC products.
Pressure swing adsorption	▪ Adsorbs HCs at high pressure and releases them at low pressure.	▪ No information on commercial programs.	▪ Can tolerate a wide range of feed gas pressures.
Chelating solvent absorption	▪ A chelating solvent absorbs N2 selectively.	▪ No information on commercial programs.	▪ No commercial usage; still in R&D.

AGRU: acid gas removal unit; MRU: mercury removal unit; EPC: engineering, procurement, and construction; KBR: Kellogg Brown & Root; J-T: Joule–Thomson; BAHEs: Brazed Aluminum Heat Exchangers; MMSCFD: million standard cubic feet per day; and HCs: hydrocarbons.

provides a detailed comparative summary of NRU technologies. While multiple NRU methods exist, cryogenic fractionation remains the most widely adopted for large-scale applications due to its efficiency and adaptability to varying N₂ concentrations.

For large-scale operations with capacities exceeding 500,000 standard cubic meters per day, cryogenic N₂ extraction has become the favored method among current NRU technologies due to its superior adaptability to different N₂ levels in the feed gas. Numerous N₂ extraction systems are available, each offering distinct capacities for NRU [275]. Depending on the specific requirements of the gas mixture being processed, configurations utilizing one, two, or even three columns can be implemented [276,277]. Several leading EPC companies, such as APCI, Bechtel/IPS, Linde, KBR, BCCK, and Costain, provide these solutions globally [272]. Distinct methodologies for distillation have a long history of application in isolating N₂ from NG mixtures. This technique achieves over 99% hydrocarbon recovery, with CH₄ as the primary component, across various N₂ concentrations in the input stream and is particularly effective for high feed-gas flow rates [272]. An illustration of an NRU block flow is depicted in Figure S2 of the Supplementary Materials [278]. The typical structure of this technology involves five fundamental stages: reception and compression of the input gas, necessary pre-treatment processes, refrigerating through J-T expansion or an expander, low-temperature fractional distillation, and the final compression phase [278].

The configuration of the N₂ separation unit is intricately tailored based on the specific N₂ content present in the NG stream. A heat pump cycle is employed for cases where the N₂ concentration is below 20%, as depicted in Figure S3a of the Supplementary Materials [278,279]. However, a notable drawback of this method is the requirement for a compressor to operate the heat pump. Modern designs typically incorporate two columns and an initial separation stage to address these challenges. In cases where the N₂ concentration is high, a dual-column system is implemented to effectively isolate the N₂ (as illustrated in Figure S3b of the Supplementary Materials) [278,279]. This design offers considerable flexibility, being applicable even when N₂ levels exceed 50%. Advanced configurations may involve using two or three columns, reducing the construction cost of the distillation column’s cold box within the NRU by 25% [266]. When the cryogenic distillation technique is chosen as a process option to remove N₂, it is often combined with LNG and NGL recovery units [14].

Hamedi et al. [280] examined N₂ rejection for sub-quality gas and enhanced recovery, where N₂ concentration ranged from 5% to 70%. They optimized single- and multi-column processes (double-, three-, and two-column structures) using particle swarm optimization (PSO) and exergy analysis. It was found that single-column processes consumed more energy when N₂ content was below 55 mol% but less when it exceeded this level. Two-column systems performed similarly to multi-column processes when N₂ concentration was lower than 20%. Multi-column processes reduced exergy loss at all levels, with further reductions in heat exchangers and compressors below 50% N₂. Due to higher equipment costs, maintenance, and complexity, multi-column systems may be less suitable for N₂ rejection in oil/gas recovery programs. However, an economic analysis is needed to determine the optimal design under relevant energy regulations. MacKenzie et al. [273] conducted a comparative study on four N₂ removal processes, evaluating their efficiency based on compression power, CH₄ recovery, and process complexity. Their analysis highlighted that single-column processes require higher compression power when the N₂ content in the inlet gas is below 35%, making them less efficient for low-N₂ streams. Also, double-column processes perform well across a wide range of N₂ concentrations, offering reduced power consumption and higher CH₄ recovery. Two-column and three-column processes show no significant difference in compression power requirements; however, the three-column process provides better CH₄ recovery. It was concluded that selecting the optimal process depends on balancing energy efficiency, CH₄ recovery, and N₂ separation effectiveness, with the double column being the most suitable for high-N₂ content feeds.

4.2. LNG/NRU Hybrid Units

Including N₂ in NG diminishes its calorific value, necessitates larger transmission pipelines, and increases the capacity requirements of gas pressure-boosting stations [51]. Therefore, N₂ is extracted from NG when its concentration exceeds 4 mol% under high current intensities [14]. Figure 13 presents a streamlined arrangement of an LNG unit with an enhanced NRU. In stream 1, NG is prepped for the feed system by reducing the concentrations of compounds like H₂O and CO₂ to prevent freezing. The feed gas stream usually contains 5–15 mol% N₂ and is under 25–130 bar. Some of the N₂ output may be utilized as a utility N₂ system for tasks such as cooling in heat exchangers and absorber regeneration. The CH₄ content in the N₂-separated output stream can be reduced to less than 2 mol%, allowing this process to achieve a CH₄ recovery rate of approximately 99.8% [54].

An integrated structure combining NRU and LNG units was developed by Mehrpooya et al. [281]. The C₃H₈ refrigeration cycle is utilized for pre-cooling, while the multi-component refrigerant cycle is employed for liquefying the incoming feed. Figure 14 presents the process flow diagram of this cryogenic NG process. The analysis showed that the process achieved 41.27% exergy efficiency, with an exergy destruction rate of 89.90 MW. Exergy destruction and investment costs were categorized as avoidable–unavoidable and endogenous–exogenous [281]. The results revealed that both were mainly driven by internal factors, with minimal impact on exergy efficiency from component interactions. Exergy destruction costs in compressors were avoidable, while their investment costs were not. In contrast, exergy destruction in heat exchangers and air coolers was unavoidable, but their investment costs were avoidable.

An innovative combined system comprising an LNG plant, an NRU system, an ACS process, a power generation facility, and a biomass gasification process was developed [17]. The proposed system employed a cascade arrangement integrating an ARC-compression refrigeration system (-CRS) and MR cycles to supply cryogenic energy. The liquefaction process had an SPC of 0.7673 kWh/kg LNG, while the integrated system achieved an overall thermal efficiency of 54.29%.

Integrating LNG and NRU units reduces energy consumption and lowers operating costs. This integration decreases the need for some utility-providing equipment, yet it also increases system complexity, making capital costs a critical factor [14,17]. Effective operation demands precise process control, synchronization of refrigeration cycles, and optimal heat exchanger performance, further adding to the system’s complexity. In addition, process disturbances, variations in feed gas composition, and fluctuations in operating conditions can impact system stability and performance and necessitate advanced control strategies.

4.3. LNG/NGL Hybrid Units

In all processes for producing LNG, it is essential to remove heavier hydrocarbons. The value of products such as C₂H₆, which serves as feedstock for downstream operations, is well recognized, making it increasingly important to extract these compounds from NG [22,282]. One of the benefits and uses of this method is the ability to produce LNG with a customizable calorific value. This feature allows producers to adjust the proportion of heavier hydrocarbons in LNG based on market demand, economic factors, and the selling price of LNG and NGL [227,283]. Modifying the unit’s operating conditions enables the adjustment of calorific value and price to align with market needs, optimizing profitability [14]. Moreover, the NGL produced serves as feedstock for downstream facilities such as olefin or aromatic production units, where further processing and separation can generate significant added value [188,283]. Products derived from NGLs include C₂H₆, LPG, and pyrolysis gasoline, which are critical feedstocks for petrochemical industries, particularly for aromatic and olefinic units. Therefore, separating heavier hydrocarbons is necessary and enables the production of valuable by-products. Designing the units and integrating their processes reduces the need for additional equipment and minimizes energy consumption [17]. To achieve the integration of these units, ConocoPhillips, CH.IV, APCI, and Ortloff have each proposed plans [284,285,286]. John Mack introduced a process that underwent modifications in 2008 [287]. This method uses an absorption tower alongside a distillation tower to separate NGLs and incorporates an external closed-loop refrigeration cycle for the NG liquefaction. The core element of this process is an absorption tower with a reflux system and a distillation tower that operates at a pressure higher than that of the distillation column. This setup enables the production of cold gas at an appropriate pressure, containing only minimal heavy hydrocarbons. The CH₄-rich gas exiting from the top of the tower is compressed and directed to the liquefaction unit, utilizing the energy generated by the expansion of the incoming gas. The degree of separation is controlled by regulating the temperature within the towers, achieved through adjusting the flow split and designing the separation section in two distinct parts. One of the proposed solutions for maximizing C₂H₆ recovery from the overhead stream of a demethanizer column is to provide the necessary reflux stream at the lowest possible temperature [288]. Figure 15 illustrates some of the simplest available processes for achieving maximum C₂H₆ recovery with minimal cost. These cryogenic processes include residue recycle (RR), gas sub-cooled process (GSP), cold residue recycle (CRR), and multiple reflux streams.

Table 3. Technical characteristics of selected LNG systems integrated with NGL and NRU processes.

References	SPC (kWh/kg LNG) Unless Indicated Otherwise	Exergy Efficiency	Remarks
Sabbagh et al. [223]	0.3473	0.5512	▪ C3MR unit (LNG/NGL) ▪ PR EOS ▪ ASPEN PLUS package ▪ Controlled NSGAII ▪ Multi-objective (annualized profit, exergy efficiency, and SPC) ▪ Multi-criteria decision-making (LINMAP and TOPSIS)
Sabbagh et al. [224]	0.347	-	▪ C3MR unit (LNG/NGL) ▪ PR EOS ▪ ASPEN PLUS/EDR package ▪ GA method ▪ Single objective (annualized profit)
Sabbagh et al. [139]	0.3626	0.7179	▪ C3MR unit (LNG/NGL) ▪ PR EOS ▪ ASPEN PLUS package ▪ GA method ▪ Single objective (SPC)
Khan et al. [20]	0.3863	-	▪ SMR unit (LNG/NGL) ▪ PR EOS ▪ ASPEN HYSYS package ▪ Knowledge-based methodology ▪ Single objective (SPC)
Vatani et al. [22]	0.414	-	▪ DMR units (LNG/NGL) ▪ PR EOS ▪ Pinch approach (CC) ▪ ASPEN HYSYS package
Uwitonze et al. [225]	-	-	▪ DMR unit (LNG/NGL) ▪ PR EOS ▪ Pinch approach (CC) ▪ ASPEN HYSYS package ▪ Knowledge-based methodology ▪ Single objective (power consumption)
Ghorbani et al. [14]	0.359	0.6162	▪ C3MR unit (LNG/NGL/NRU) ▪ Pinch approach (CC) ▪ ASPEN HYSYS ▪ NSGAII method ▪ Multi-objective (period of return and SPC)
Ghorbani et al. [191]	0.339	0.6614	▪ DMR unit (LNG/NGL/NRU) ▪ PR EOS ▪ Pinch approach (CC and ECC) ▪ ASPEN HYSYS package
Ghorbani et al. [51]	0.33–0.343	0.6282	▪ MFC unit (LNG/NGL/NRU) ▪ Pinch approach (CC and ECC) ▪ ASPEN HYSYS package ▪ NSGAII method ▪ Single objective (SPC)
Ghorbani et al. [18]	MCF: 0.4231 ARC-MR2: 0.2722	MCF: 0.5606 ARC-MR2: 0.4893	▪ MFC-ARC/MR2 units (LNG/NGL) ▪ Pinch approach (CC) ▪ ASPEN HYSYS package ▪ Prime cost of the product: MFC (0.305) and ARC-MR2 (0.237 USD/kg LNG)
He et al. [226]	0.44 kWh/Nm3	0.47	▪ C3MR unit (LNG/NGL) ▪ PR EOS ▪ Pinch approach (CC) ▪ ASPEN HYSYS package ▪ GA algorithm ▪ Single objective (SPC)
He et al. [25]	0.371 kWh/Nm3	-	▪ SMR unit (LNG/NGL) ▪ PR EOS ▪ Pinch approach (CC) ▪ ASPEN HYSYS package ▪ GA algorithm ▪ Single objective (SPC)
Mehrpooya et al. [24,227]	MFC: 0.364 C3MR: 0.391 DMR: 0.375	MFC: 0.59 C3MR: 0.56 DMR: 0.55	▪ MFC-C3MR-DMR units (LNG/NGL) ▪ PR EOS ▪ Pinch approach (CC) ▪ ASPEN HYSYS package
Mehrpooya et al. [281]	-	0.4127	▪ C3MR unit (LNG/NRU) ▪ Peng–Robinson–Stryjek–Vera equation ▪ ASPEN HYSYS package
Ebrahimi et al. [17]	0.7673	-	▪ ARC-CRS-MR unit (LNG/NRU) ▪ ASPEN HYSYS package ▪ Integrated with power plant and biomass gasification unit
Mak et al. [289,290]	-	-	▪ Pure MR ▪ C2H6 recovery: 25–85% ▪ Pinch approach (CC)

GA: genetic algorithm; NSGAII: non-dominated sorting genetic algorithm II; LINMAP: linear programming technique for multidimensional analysis of preference; and TOPSIS: technique for order of preference by similarity to ideal solution.

lists the technical characteristics of selected LNG liquefaction systems integrated with LNG and NRU processes. Mehrpooya et al. [24] developed three new integrated configurations for the simultaneous production of NGL and LNG, based on the C3MR, DMR, and MFC refrigeration cycles. These configurations achieved exergy efficiencies of 55%, 56%, and 59%, respectively, with SPCs of 0.391, 0.375, and 0.364 kWh/kgLNG. The hybrid process for producing NGLs and LNG is designed based on the C3MR, DMR, and MFC refrigeration cycles, as depicted in Figure 16.

When designing and developing multi-component refrigerant cycles for low-temperature applications (−50 to −163 °C), engineers encounter various options regarding the composition of refrigerant components and the selection of operating pressures, temperatures, and cycle configurations [55,201,260]. These factors complicate such cycles’ designs and optimization processes [27]. Ghorbani et al. [283] designed an integrated system for cogenerating the NGLs and LNG utilizing the C3MR refrigeration cycle. Figure S4 in the Supplementary Materials shows the process flow diagram for the C3MR-based NGL/LNG hybrid structure [283]. This system achieved an exergy efficiency of 61.17% and a refrigeration cycle performance factor of 1.551.

4.4. LNG/NGL/NRU Hybrid Units

Figure 17 demonstrates a simplified hybrid low-temperature NGL/LNG/NRU system arrangement. In integrated NG supercooling processes, the cascade system typically comprises a cascade refrigeration section (section 10), a heavy liquid separation section for NG (NGL, section 11), a supercooling and liquefaction section for NG (LNG, section 12), and a N₂ removal unit (NRU, section 26) [52]. Generally, the first, second, and third cascade refrigeration cycles (sections 13, 14, and 15) can cool some of the NG flow entering the liquefaction unit. These refrigeration cycles can operate in closed-loop, open-loop, or hybrid configurations. In this patented design, the first and second cycles (sections 13 and 14) operate in a closed loop, while the third cycle (section 15) functions as an open loop. When configuring the third cycle as an open loop, the NG liquefaction unit features an integrated N₂ removal section (section 26).

A combined system for NGL/LNG/NRU was developed, incorporating an ARC for initial cooling and an MFC for further cooling and liquefaction [15]. The system was designed to achieve 91.9% C₂H₆ recovery and 81% N₂ rejection. These processes’ simultaneous design and integration allow for eliminating the reboiler and condenser in the demethanizer and N₂ rejection columns. The SPC for the three cycles based on C3MR, DMR, and modified AR-MR1 are 0.359, 0.35, and 0.250 kWh/kg LNG, respectively. The exergy efficiency for the three hybrid cycles based on C3MR, DMR, and modified AR-MR1 are 61.6, 62.3, and 58.1%, respectively. By replacing the first stage of the pre-cooling cycle in the C3MR process with an ammonia–water (NH₃/H₂O) refrigeration cycle, efficiency is reduced by 5.6%, and SPC decreases by 32.4%.

4.5. Hybrid Helium Recovery Systems

The primary sources of helium as a noble gas are atmospheric air and NG. Among these, NG stands out as the most abundant source of helium. Although extracting helium from the air is technically viable, it is not considered economically justified. Helium found in NG is often associated with N₂ [291,292]. Helium recovery from NG using a low-temperature separation technique, known as fractional distillation, is considered cost-effective [293]. This method has been developed and refined by companies such as Air Products (USA), Linde (Germany), and Air Liquide (France). Linde Co. was the first to introduce a technique for helium production. In this approach, the initial step produces a volumetric flow containing at least 80% helium [294,295,296]. Cryogenic methods for helium recovery can be categorized into three types: those based on flash evaporation, those relying on distillation, and hybrid processes that combine both flash evaporation and distillation [19,293,297,298]. A common method for extracting helium involves utilizing the gas from the final stage of flash evaporation in LNG production units [299,300,301,302]. The gas, consisting of a mixture of helium (i.e., crude helium) and N₂, is then sent to another facility for further liquefaction and purification. For helium extraction to be economically viable, the helium content in NG needs to be at least 0.1–0.5 mol% [297]. Crude helium gas typically contains at least 50% helium and other substances such as N₂, argon, and CH₄ [303]. Helium production can be seamlessly integrated into the final cooling stage of LNG processing for effective extraction. This integration can be designed to minimize energy consumption and reduce the complexity of the overall design [304,305]. To achieve this, numerous oil and gas companies have developed optimized, integrated processes for recovering helium from the final flash evaporation stage within LNG cooling systems, aiming to lower energy usage during crude helium production. NGs that naturally contain helium typically also have a significant amount of N₂. The presence of N₂ lowers the calorific value of the NG, which can cause complications during the transportation of LNG. Consequently, ensuring that the N₂ content in LNG remains below a certain threshold is essential to avoid these issues [61,306]. The waste stream exiting the PSA unit contains N₂ and trace amounts of helium. This stream is recompressed and reintegrated into the unit’s inlet flow to recover the helium. A high concentration of N₂ in the incoming stream to the PSA unit will increase the residual flow, thereby raising both the initial and operational costs associated with compressing the flow [303,307]. Therefore, it is recommended that the N₂ concentration in the gas entering the PSA unit be maintained between 5 and 15% by volume [308]. Figure S5 in the Supplementary Materials illustrates an integrated helium production unit and targets for N₂, hydrogen (H₂), and CO₂ management [303].

A cryogenic technique was developed to generate a helium-rich vapor stream, a CH₄-enriched stream, and a liquid stream containing hydrocarbons heavier than CH₄ [309]. The temperature and pressure of the input feed stream for this process typically ranged from −40 to −10 °C and from 30 to 50 bar, respectively. In this setup, a distillation column was utilized to separate C1 (i.e., CH₄) and NGLs. The helium concentration in the helium-enriched stream was approximately 50 mol%.

APCI developed a hybrid NG/NRU process integrated with a distillation-based helium extraction unit (HeXU) featuring a condenser [310]. In this setup, two N₂ removal stages (i.e., 23 and 25) are connected with the HeXU, which includes a heat exchanger (i.e., 70) and a separator (i.e., 72). Streams 41, 42, and 50 correspond to pressure-reduced LNG, sub-cooled LNG, and the final LNG product, respectively. This process is integrated with an LNG plant, as shown in Figure S6 of the Supplementary Materials [310]. LNG is extracted from the liquefier (i.e., 18) at a relatively high temperature (i.e., 17), brought to medium pressure, and then passed through the first NRU stage (i.e., 23), a distillation column that separates helium and N₂ [310]. The bottom product is returned to the liquefier for further cooling and depressurization to near atmospheric pressure. The decompressed LNG then enters the second NRU stage (i.e., 25), a phase separator that also serves as a condenser for the first column. The final LNG product (i.e., 50) is produced after providing the necessary cooling for the NRU. The two-stage system connects a high-pressure stripping column (i.e., 23) with a flash tank (i.e., 25), each operating at different pressures [310]. N₂-rich gas and helium (i.e., 26) are collected from the stripping column and sent to the HeXU, where crude helium (i.e., 73) is extracted. Unlike flash-based helium extraction, this helium-rich gas (i.e., 46) contains minimal CH₄. The condensed liquid (i.e., 75) in HeXU, nearly pure N₂, can be vented to the atmosphere after being used as a refrigerant (i.e., 77).

Flash-based processes require fewer pieces of equipment and are easier to configure compared to other methods. The SPC of this process is lower than that of other approaches and offers an efficient rate of helium extraction [291]. Two modified process configurations were presented and compared with the conventional processes developed by Linde and APCI [291]. Figure 18 and Figure 19 illustrate the block and process diagram of the APCI/Linde cycles utilizing flash evaporation for the co-production of helium and LNG. J-T expansion and phase separation are among the most commonly used techniques to separate helium from NG. The pressure drop across the dew point expansion valves reduces the concentration of lighter components, such as helium, which then vaporizes and exits as a vapor product in what is known as the end flash. The helium extraction rate for the modified Linde-based process is higher than that of the modified APCI-based process [291].

Table 4. Technical specifications of chosen LNG systems combined with helium recovery processes.

References	Helium Extraction Rate	Power Consumption Ratio (kWh/kmol He) Unless Indicated Otherwise	Remarks
Zaitsev et al. [170]	0.951	132.9	▪ MFC unit and ARC-MR2 unit (LNG/He recovery) ▪ PR EOS ▪ ASPEN HYSYS package and M-file code ▪ Pinch approach (CC and ECC) ▪ SPC: MFC (0.265) and ARC-MR2 (0.1849) kWh/kg LNG ▪ Exergy efficiency: MFC (0.8966) and ARC-MR2 (0.8896) ▪ Prime cost of the product: MFC (0.2069) and ARC-MR2 (0.1939) USD/kg LNG
Mehrpooya et al. [304]	0.9142	-	▪ Multi-stage flash unit, modified APCI (LNG/He recovery) ▪ PR EOS ▪ ASPEN HYSYS package and M-file code ▪ Integrated with fuel cell, ARC, and steam power system ▪ SPC: 0.2086 kWh/kg LNG ▪ Exergy efficiency: 0.9491
Mehrpooya et al. [291]	▪ Linde (flash): 0.96 ▪ APCI (flash): 0.91	▪ Linde (flash): 388 ▪ APCI (flash): 227	▪ Multi-stage flash unit, modified APCI (LNG/He recovery) ▪ PR EOS ▪ ASPEN HYSYS package and M-file code ▪ Pinch approach (CC and ECC)
Donghoi Kim [311]	▪ APCI (flash): 0.9 ▪ Linde (flash): 0.95 ▪ APCI (distillation): 0.9 ▪ Technip (distillation): 0.63 ▪ Reboiled (distillation): 0.95 ▪ Linde (integration): 0.95 ▪ Exxon (integration): 0.95	▪ APCI (flash): 87 kWh/Sm3 He ▪ Linde (flash): 139 ▪ APCI (distillation): 93 ▪ Technip (distillation): 87 ▪ Reboiled (distillation): 99 ▪ Linde (integration): 28 ▪ Exxon (integration): 27	▪ LNG/He recovery ▪ PR EOS ▪ ASPEN HYSYS package
Ansarinasab et al. [299]	0.958	-	▪ Multi-stage flash unit, modified Linde process (LNG/He recovery) ▪ PR EOS ▪ ASPEN HYSYS package and M-file code ▪ Integrated with fuel cell, ARC, and steam power system
Shafaei et al. [293]	▪ Modified Linde (integration): 0.97 ▪ Modified ExxonMobil (integration): 0.95 ▪ Linde (integration): 0.9 ▪ ExxonMobil (integration): 0.9	▪ Modified Linde (integration): 1249 ▪ Modified ExxonMobil (integration): 605	▪ Hybrid method, modified Linde and ExxonMobil processes (LNG/He recovery) ▪ PR EOS ▪ ASPEN HYSYS package ▪ Pinch approach (CC)

depicts the technical specifications of chosen LNG systems combined with helium recovery processes.

Donghoi Kim [311] evaluated various process configurations for helium extraction from LNG end-flash gas, determining that the optimal technology depended on specific project conditions. The Linde flashing process had achieved the highest helium extraction rate while the reboiled distillation process had emerged as a robust option for consistent helium extraction and energy efficiency, especially with helium-rich feed gas. Although the ExxonMobil and Linde integration processes demonstrated superior energy efficiency, they required more complex equipment [311]. Economic analysis had suggested that integrating helium extraction into LNG plants was increasingly viable, particularly as helium prices were expected to rise. Despite some drawbacks, the reboiled distillation process was a generally favorable choice due to its balance of high helium recovery, energy efficiency, and simplicity [311].

This section examines integrated process structures for NG treatment, focusing on LNG, NGL, and NRU processes to enhance efficiency and reduce costs. Cryogenic fractionation remains the preferred NRU method for large-scale operations due to its efficiency and adaptability to various N₂ concentrations. Multi-column configurations improve exergy efficiency and process optimization but lead to higher equipment costs and complexity, which may limit oil and gas recovery feasibility. The GSP method is widely employed for cost-effective C₂H₆ recovery by extracting and cooling side streams via multi-stream heat exchangers before reinjection. This integration eliminates the need for reboilers and condensers in demethanizers and nitrogen removal towers, reducing energy consumption but increasing process complexity and requiring detailed economic analysis. Helium recovery can be incorporated into LNG production, with flash-based methods providing a cost-effective alternative to distillation by demanding fewer components and consuming less power.

5. Solutions to Improve LNG System Efficiency

The support for reducing SPC in cryogenic NG processes is a common practice due to its potential to enhance the efficiency and economic feasibility of large-scale NG storage and transmission. Lowering SPC minimizes energy consumption, directly reducing operational costs and making the LNG production process more competitive [45,156,216,277,312]. Furthermore, a reduction in SPC is essential to decrease the carbon footprint of LNG operations and achieve global net-zero emission goals. This section covers technologies designed to enhance the performance of cryogenic NG processes, such as the implementation of optimization algorithms, MR systems, ARCs, DARCs, ACAR, TEG, liquid air cold recovery systems, ejector refrigeration cycles, and the integration with renewable energy sources and waste heat recovery.

5.1. Process Enhancement Through Various Algorithms

Designing an LNG facility is highly intricate and encompasses many factors. Employing optimization methodologies can ensure that designs are optimized, provided that robust and reliable models are constructed [313]. A well-designed LNG plant is characterized by minimal capital cost and reduced energy usage [314]. Several studies have explored optimization methods for LNG facilities [101,107,148,233,245,315,316,317,318,319]. Research shows that several objective functions have been employed in optimizing the LNG process, including minimizing power consumption [103,110,123,125,149,154,226,236,237,238,320,321,322,323,324,325,326,327,328,329,330,331], maximizing exergy efficiency [56,106,124,127,129,223], increasing production [332,333,334], reducing operating costs [124,335,336,337], lowering capital expenditures or total annual costs [56,66,122,129], improvement of economic parameters (e.g., profit and net present value) [21,338], and addressing multi-objective functions [56,106,122,124,129,133,223,339,340,341]. Environmentally friendly refrigerant mixtures were identified using mathematical programming techniques to minimize environmental impact [342]. This process involved solving a mixed-integer nonlinear programming (MINLP) model that aimed to achieve specific properties for the MR, with environmental considerations integrated as constraints. The study revealed that MR mixtures possess superior characteristics compared to single-component refrigerants. Another method developed for MR systems relies on nonlinear programming (NLP) combined with thermodynamic principles to solve an NLP problem by leveraging CCs and fine-tuning the refrigerant composition [107]. Their strategy focused on optimizing the composition of the refrigerant mixture, specifically C1-C4, and N₂, under specific pressure conditions and flow rates. In cases where no temperature crossover occurs within the heat exchanger, they suggested adjusting the refrigerant flow rates and pressure settings based on heuristics, expert insights, or optimization strategies. They assessed three distinct objective functions: reducing individual crossovers, minimizing the cumulative sum of crossovers, and minimizing compressor energy consumption.

Five distinct liquefaction processes (i.e., SMR, C3MR, DMR, MFC, and AP-X) were explored by simulating them in ASPEN HYSYS and optimizing their energy consumption using the particle swarm optimization algorithm implemented in MATLAB version R2020a programming [154]. The outcomes from the economic and energy analyses revealed that despite being the most energy-intensive at 0.2561 kWh/kg LNG, the small-scale SMR process was the most financially advantageous, generating a profit of 20.41 million USD/year. Also, the APCI AP-X process, the most energy-efficient at 0.2367 kWh/kg LNG, was ranked as the second least profitable, with an annual profit of 20.01 million USD/year. Variations were noted in two commonly assumed trends: that higher liquefaction process complexity leads to reduced energy consumption and that energy-efficient processes lower total annual costs. For large-scale facilities, the C3MR and MFC processes were identified as the least economically viable, while for small-scale operations, the C3MR and APCI AP-X processes were deemed the least feasible [154].

An in-depth analysis was conducted to compare the performance of the PSO, GA, and BOX algorithms in optimizing the SMR, DMR, C3MR, and MFC processes, focusing on energy consumption, exergy improvement, and overall algorithm efficiency [133]. The study found that the lowest SECs for the SMR, DMR, C3MR, and MFC processes were achieved through PSO, with values of 0.3233, 0.2351, 0.2489, and 0.2427 kWh/kg LNG, respectively, indicating that PSO provided the best optimization results. Also, PSO demonstrated higher convergence accuracy and stability compared to GA, though GA had a faster computation time. PSO and GA demonstrated superior performance compared to the BOX algorithm, with PSO maintaining a slight advantage over GA. Moreover, the exergy efficiency of the SMR, DMR, C3MR, and MFC processes through PSO was enhanced to 35.34, 48.59, 45.90, and 47.07%, respectively [133].

Graphical targeting was employed to improve the SMR process by identifying the LNG heat exchanger as the primary source of energy loss in liquefaction. Given the multiple local optima in MR systems, a novel approach was necessary to effectively overcome this challenge [233]. This approach involves solving the NLP using a GA algorithm, with the objective function designed for adaptability, incorporating factors such as compressor power, total cost minimization, and capital investment reduction, and allowing the objective function to be modified based on the specific design goals. A similar optimization of SMR systems focused on minimizing energy consumption as the objective function and was also addressed using a GA algorithm [319]. Improvements in certain critical parameters indicated the underlying causes of reduced irreversibility. Across all these methods, one of the most challenging aspects was formulating the objective functions, which required substantial time and effort. Previous studies have demonstrated that various tools such as AIMMS [343], AMPL [344], ASPEN HYSYS/PLUS [25,97,101,124,314,329,345,346,347,348,349], Engineering Equation Solver [350], FICO Xpress Optimization Suite [351], GAMS [238,243,352], UniSim [100,102], MATLAB [56,110,127,154,314,320,341,353], and modeFRONTIER [354] have been utilized for optimizing the LNG process. LNG processes are complex from a thermodynamic perspective, featuring high levels of interaction and nonlinearity, and creating significant optimization challenges. Further complications in optimization arise due to conflicting objectives [236]. For instance, increasing the pressure difference and flow rates within the heat exchanger boosts the heat transfer rate. Nevertheless, this also escalates energy consumption, as more power is required by the compressors, which represent a major component of operational costs (OPEXs). Also, decreasing pressure differentials and flow rates necessitates a larger heat transfer area, leading to higher capital costs (CAPEXs) [236].

Table 5. Various optimization methods applied in some natural gas liquefaction plants.

Reference	Optimization Approach	Objective Function	Remarks
Alabdulkarem et al. [329]	GA	Power consumption	▪ C3MR unit ▪ PR EOS ▪ Pinch approach (CC) ▪ ASPEN HYSYS package ▪ SPC for optimized plant: 5.14 kWh/kmol LNG ▪ Second law efficiency: 49.97%
Lee et al. [345]	SQP	Power consumption	▪ C3MR unit ▪ PR EOS ▪ Pinch approach (CC) ▪ ASPEN HYSYS package ▪ Energy reduction vs. baseline: 17.74%
Taleshbahrami et al. [346]	GA	Power consumption	▪ C3MR unit ▪ PR EOS ▪ Pinch approach (CC) ▪ ASPEN HYSYS package and M-file code ▪ Energy reduction vs. baseline: 23%
Hwang et al. [72,149]	GA and SQP	Power consumption	▪ DMR unit ▪ PR EOS ▪ Pinch approach (CC) ▪ ASPEN HYSYS package and M-file code ▪ Energy reduction vs. baseline: 7.45% [72] ▪ Energy reduction vs. baseline: 34.5% [149]
Xu et al. [97]	GA	Power consumption	▪ SMR unit ▪ PR EOS ▪ Pinch approach (CC) ▪ ASPEN PLUS package ▪ SPC for optimized plant: 1003 kJ/kg LNG ▪ COP: 0.782 ▪ Exergy efficiency: 43.9%
Xu et al. [86]	GA	Power consumption	▪ SMR unit ▪ PR EOS ▪ Pinch approach (CC) ▪ ASPEN PLUS package ▪ SPC for optimized plant: 1013 kJ/kg LNG ▪ COP: 0.762 ▪ Exergy efficiency: 39.6–42.3%
Khan et al. [102]	SQP	Power consumption	▪ SMR unit ▪ PR EOS ▪ Pinch approach (CC) ▪ UniSim package ▪ Energy reduction vs. baseline: 4.5%
Kamath et al. [243]	NLP	Power consumption	▪ SMR unit ▪ SRK EOS ▪ Pinch approach (CC) ▪ GAMS package ▪ Energy reduction vs. baseline: 12%
Khan et al. [100]	PSO	Power consumption	▪ SMR unit ▪ PR EOS ▪ Pinch approach (CC) ▪ UniSim package and M-file code ▪ Exergy efficiency: 42% ▪ Energy reduction vs. baseline: 7.7% ▪ Exergy improvement vs. baseline: 5%
Sun et al. [133]	GA, BOX, and PSO	Power consumption and exergy efficiency	▪ SMR, C3MR, DMR, and MFC units ▪ PR EOS ▪ Pinch approach (CC) ▪ ASPEN HYSYS package and M-file code
Furda et al. [38]	GA	Levelized costs and CO2 emissions	▪ SMR, C3MR, DMR, and MFC units ▪ PR EOS ▪ Pinch approach (CC) ▪ ASPEN PLUS package and M-file code
He et al. [25]	GA	Power consumption	▪ SMR unit for NGL/LNG ▪ PR EOS ▪ Pinch approach (CC) ▪ ASPEN HYSYS package and M-file code ▪ Payback period: 3.84 years ▪ Energy reduction vs. baseline: 9.64%
Pereira et al. [154]	PSO	Power consumption	▪ SMR, C3MR, DMR, MFC, and AP-X units ▪ PR EOS ▪ Pinch approach (CC) ▪ ASPEN HYSYS package and M-file code
Santos et al. [341]	GA	Power consumption and overall heat transfer coefficient	▪ SMR and C3MR units ▪ PR EOS ▪ Pinch approach (CC) ▪ ASPEN HYSYS package, M-file code, and GAMS

BOX: box constraints; CC: composite curve; GA: genetic algorithm; NLP: nonlinear programming; PSO: particle swarm optimization; and SQP: sequential quadratic programming.

lists the various optimization methods used in some NG liquefaction plants. Previous studies have shown that several optimization algorithms, including GAs [25,49,50,51,56,89,129,139,140,172,191,207,223,226,314,323,324,353,355,356,357,358,359,360], NLP [243], sequential quadratic programming (SQP) [102,345], evolutionary search [361], particle swarm optimization [100,133,154,362], simulated annealing [104,363,364], Tabu search [101,365], knowledge-based optimization [20,90,106], and other approaches such as sequential coordinate random search [103], modified coordinate descent [105], multivariate Coggins optimization [111,152,165], shuffled complex evolution [166], vortex search [110,366], teaching–learning self-study optimization [153], invasive-weed optimization [151], and neural network algorithms [367], have been applied to LNG facilities. Also, recent advancements in production data analysis methods, such as flowing material balance equations and Monte Carlo simulation techniques, have significantly improved the estimation of gas in place and reservoir performance, particularly in non-volumetric and naturally fractured gas condensate reservoirs [368,369]. These advancements contribute to optimizing LNG production by ensuring a more accurate assessment of feed gas availability and enhancing supply chain reliability.

5.2. Multi-Component Refrigerant System in Natural Gas Liquefaction

The notion of employing multi-component refrigerants has a long history. In 1936, Podbielniak [370] introduced the initial multi-component refrigerant cycle. Afterward, Kleemenko [371], and Gammer and Newton [372] further refined and applied this concept across various fields. More recently, the issue of ozone layer damage has brought the replacement of Chlorofluorocarbon-based refrigerants into focus, prompting extensive investigations into the application of multi-component refrigerants in household and industrial contexts. Notable research includes work by Stead [373], Lamb et al. [374], and Diodi and Achni [375]. Moreover, several studies have concentrated on using these refrigerants to reach extremely low temperatures [376,377]. These investigations consistently utilized mathematical programming techniques to establish the optimal refrigerant mixture proportions. One key application of multi-component refrigerant systems is in the LNG industry, where NG is cooled to −160 °C. In 1970, APCI was the first to implement a simple SMR for liquefying NG [378]. Recent studies on improving multi-component refrigerant systems have utilized innovative design strategies and expert knowledge, followed by mathematical optimization, to determine the optimal refrigerant composition and operating conditions [334,379]. Furthermore, Lee [99] presented a structured approach to designing NG liquefaction systems, focusing on choosing the optimal refrigerant mix via an NLP model. Over the past years, MR processes have become the leading technology for liquefaction, accounting for more than 75% of current LNG systems [380]. The importance of MR processes lies in their adaptability, enabling precise regulation of hot and cold temperature profiles by modifying the refrigerant mixture [113,122,127,381,382,383]. In addition, MR processes utilize fewer compressors and heat exchanger components, resulting in lower capital requirements [236]. However, employing MR processes significantly complicates the design and operation due to heightened thermodynamic interactions, making process control and optimization considerably more difficult [381].

5.3. Absorption Refrigeration Process

The surplus heat generated from different industrial sectors can be utilized to improve the efficiency of the liquefaction process [329]. One effective method to generate cooling from this waste heat is implementing ARCs. Unlike traditional refrigeration methods that rely on electrical power, ARCs utilize medium-temperature thermal energy to produce refrigeration. This distinct feature of ARCs makes them well suited for effectively utilizing waste heat within industries [384,385,386]. Within the LNG industry, there has been research into the indirect application of ARCs to enhance cooling performance [44,387,388]. However, an alternative approach that has not been extensively explored is the direct substitution of CRCs with ARCs. The feasibility of utilizing ARCs as a substitute for the pre-cooling process in an LNG system was evaluated [389]. The findings indicated that utilizing the waste heat from a 9 MW gas turbine could conserve approximately 1.9 MW. ARCs have been extensively employed across various sectors, including the food and chemical industries [390]. These refrigeration systems operate using a solution as the working fluid, with the most commonly used fluid pairs being H₂O–lithium bromide (typically used for temperatures above 0 °C) and NH₃/H₂O (generally used for temperatures below 0 °C). The energy demand of the cycle is primarily limited to the power required for the pump (or the fan power in air-cooled heat exchangers, when air serves as the medium-temperature reservoir), which is significantly lower compared to that of CRCs [44,388]. ARCs operate quietly and have lower maintenance costs due to the absence of large rotary equipment. Figure 20 depicts the utilization of the ARC units for pre-cooling in LNG processes. The thermal energy required to power the generator is sourced from renewable energy and waste heat from industrial processes [44,188,199].

The NH₃/H₂O-based ARC system is composed of several essential components, each contributing to the cycle’s efficient operation, as depicted in Figure 21a. The system begins with the generator or separator, which has an important role in separating H₂O from ammonia, ensuring the proper composition of the working fluids [391]. The purification unit cleans the ammonia stream, effectively removing residual H₂O droplets to maintain a pure NH₃ flow. The condensation unit converts the purified ammonia vapor into a liquid state, a necessary step for the subsequent processes [58,197,392]. The evaporative unit generates the cooling effect by evaporating ammonia, a key function of the refrigeration cycle. The absorption component is essential in dissolving ammonia into the H₂O, creating the NH₃/H₂O mixture required for the cycle’s operation [393,394]. Pressure management devices including pumps and pressure relief valves are critical for maintaining the appropriate pressure levels and facilitating the absorption and desorption processes within the system [199,391].

ARCs with smaller capacities often encounter difficulties related to the performance of the solution pump, leading to less-than-ideal efficiency [390]. DARCs were introduced to address these challenges. This advanced technology combines three distinct working fluids: H₂ (or helium), H₂O, and ammonia. In this setup, ammonia typically serves as the refrigerant, H₂O is the absorbent, and H₂ (or helium) is an auxiliary inert gas. The concept of DARCs originated in the 1920s and has since undergone significant advancements [395,396]. These systems are adaptable, as they can be powered by various energy sources, including liquid gas, NG, or kerosene, offering flexibility in their operation. A key benefit of DARCs is eliminating a pump between the absorber and generator, which reduces mechanical complexity and increases system reliability. Moreover, these systems are lightweight, highly safe, and economical, making them well suited for applications with minimal refrigeration requirements [397]. A new process involving an SMR-based LNG process, pre-cooled by a DARC system, was introduced and extensively analyzed through 4E evaluations and 3D parametric studies [398]. The DARC unit is based on waste heat as its energy source was engineered to provide cooling down to −29.62 °C. The economic analysis revealed that the levelized cost of production was 0.198 USD/LNG. Figure 22 presents the schematic of the DARC designed for domestic use, which consists of four main components: the boiler, condenser, evaporator, and absorber. The integration of geothermal energy into the ARC for liquefaction has been explored through three methods [399]: direct use within the liquefaction facility, pre-cooling within the absorption cycle, and splitting geothermal functions between pre-cooling and power generation. The results showed that using geothermal energy in the ARC significantly reduces the energy needed for liquefaction, proving to be more efficient than direct geothermal application in the liquefaction process.

Figure 21. Schematic of the ARCs (a) and DARCs (b) designed for application in the LNG systems, modified from Refs. [58,392,396,400].

Figure 21. Schematic of the ARCs (a) and DARCs (b) designed for application in the LNG systems, modified from Refs. [58,392,396,400].

He et al. [323] developed innovative NG liquefaction systems that incorporate waste heat from gas turbine exhaust to decrease energy consumption in liquefaction plants. The systems utilize two specific methods: NH₃/H₂O-based ARC and the organic Rankine cycle (ORC). The NG industry’s widely used C3MR and cascade processes were simulated and optimized using HYSYS software and GA. Integrating waste heat utilization with the C3MR process reduced power consumption by 16.23% with the NH₃/H₂O-based ARC and by 20.57% with the ORC, achieving exergy efficiencies of 39.39% and 41.86%, respectively. Overall, the integrated systems demonstrated significant potential, including a reduction in CO₂ emissions by 4000 to 6000 tons annually.

An NH₃/H₂O-based ARC was utilized to replace the MR refrigeration cycle within the MFC-based LNG process [58]. The strategy aimed to reduce the electrical energy consumption of the processes by harnessing the waste heat produced within the facility. The results demonstrated that electricity usage can be reduced by 31% in the modified process. Furthermore, there was potential for a 30% reduction in cold box surfaces employed in the main factory equipment and improved H₂O utility efficiency. An integrated LNG/NGL system was developed using an ARC for pre-cooling and two MR processes for cooling and liquefaction [18]. SPC for the base and modified MFC cycles was 0.423 and 0.272 kWh/kg LNG, respectively. The exergy efficiency dropped by 12.72% when replacing the first stage of the pre-cooling cycle with an NH₃/H₂O-based ARC system. Implementing a combined MR and ARC reduced the product’s prime cost by 22% and the annualized system cost by 4.32%, increasing the net annual benefit by 6.2%.

He et al. [45] investigated the ACAR system as a pre-cooling system in the LNG production process. Their investigation revealed that the ACS system, achieving a refrigeration temperature of −52.9 °C and operating with a COP of 0.35, was well suited for the pre-cooling stage in LNG production. The study highlighted that employing various forms of ACARs can effectively provide the necessary cooling temperatures while lowering electricity consumption [401]. A cutting-edge ACAR, featuring a COP of 0.226 and an evaporating temperature of −54.62 °C, was introduced and thoroughly evaluated using an energy analysis method [402]. The results showed that the proposed ACAR could be an excellent candidate for a pre-cooling mechanism in LNG production processes.

Table 6. Technical features of selected LNG liquefaction systems utilizing ARCs, ACARs, and DARCs.

Reference	SPC in Modified Unit (kWh/kg LNG)	Remarks
Mehrpooya et al. [58]	0.172	▪ ARC-MR2 unit (MFC-based LNG) ▪ PR EOS ▪ Pinch approach (CC) ▪ ASPEN HYSYS package and M-file code ▪ Energy reduction vs. baseline: 31% ▪ Cold box surface reduction vs. baseline: 30% ▪ COP for the ARC: 0.48
Ansarinasab et al. [59]	0.207	▪ ARC-MR1 unit (C3MR-based LNG) ▪ PR EOS ▪ Pinch approach (CC) ▪ ASPEN HYSYS package and M-file code ▪ Energy reduction vs. baseline: 20.38% ▪ COP for the ARC: 0.49
Ghorbani et al. [18]	0.2722	▪ ARC-MR2 unit (MFC-based LNG/NGL) ▪ PR EOS ▪ Pinch approach (CC) ▪ ASPEN HYSYS package ▪ Exergy efficiency: 0.4893 ▪ Energy reduction vs. baseline: 35.66% ▪ Exergy reduction vs. baseline: 12.72% ▪ Prime cost of product reduction vs. baseline: 22% ▪ Annualized cost reduction vs. baseline: 4.32%
Ghorbani et al. [15]	0.25	▪ ARC-MR1 unit (DMR-based LNG/NGL/NRU) ▪ PR EOS ▪ Pinch approach (CC) ▪ ASPEN HYSYS package and M-file code ▪ Exergy efficiency: 0.581 ▪ Energy reduction vs. baseline: 12.6% ▪ Exergy reduction vs. baseline: 6.7% ▪ Capital cost reduction vs. baseline: 25% ▪ Increase in annual net profit vs. baseline: 27%
Ghorbani et al. [15]	0.25	▪ ARC-MR1 unit (C3MR-based LNG/NGL/NRU) ▪ PR EOS ▪ Pinch approach (CC) ▪ ASPEN HYSYS package and M-file code ▪ Exergy efficiency: 0.581 ▪ Energy reduction vs. baseline: 18.4% ▪ Exergy reduction vs. baseline: 5.6% ▪ Capital cost reduction vs. baseline: 23% ▪ Increase in annual net profit vs. baseline: 12%
Yin et al. [403]	0.7878	▪ ARC-CRS (N2 expansion-based LNG) ▪ PR EOS ▪ Pinch approach (CC) ▪ ASPEN HYSYS package and M-file code ▪ Exergy efficiency: 0.3657 ▪ Energy reduction vs. baseline: 11% ▪ Increase in exergy efficiency vs. baseline: 12.6% ▪ Increase in COP vs. baseline: 11.8% ▪ Increase in capital cost vs. baseline: 11.75% ▪ Operating cost reduction vs. baseline: 0.222%
Shariati Niasar et al. [60]	0.1988	▪ ARC-MR2 unit (MFC-based LNG) ▪ Pinch approach (CC) ▪ ASPEN HYSYS package and M-file code
Mehrpooya et al. [404]	0.189	▪ ACAR-MR1 unit (DMR-based LNG) ▪ PR EOS for the LNG process ▪ PSRK EOS for the ACS process ▪ Pinch approach (CC and ECC) ▪ ASPEN HYSYS package and M-file code ▪ Prime cost of the product: 0.1959 USD/kg LNG ▪ COP of the ACS: 0.268
Ghorbani et al. [197]	0.179	▪ ARC-MR2 unit (MFC-based LNG/NGL/NRU) ▪ PR EOS ▪ Pinch approach (CC and ECC) ▪ ASPEN HYSYS package and M-file code ▪ Exergy efficiency: 0.5811 ▪ Energy reduction vs. baseline: 38.94% ▪ Exergy reduction vs. baseline: 4.7% ▪ Capital cost reduction vs. baseline: 31.9% ▪ Prime cost of the product reduction vs. baseline: 15.31%
Zaitsev et al. [170]	0.1849	▪ ARC-MR2 unit (MFC-based LNG/He recovery) ▪ PR EOS ▪ Pinch approach (CC and ECC) ▪ ASPEN HYSYS package and M-file code ▪ Exergy efficiency: 0.8896 ▪ SPC reduction vs. baseline: 30.22% ▪ Exergy reduction vs. baseline: 0.781% ▪ Operating cost reduction vs. baseline: 6.51% ▪ Increase in capital cost vs. baseline: 31.21% ▪ Increase in prime cost of the product vs. baseline: 6.283%
Lu et al. [405]	-	▪ ARC-CRS unit (CRS-based LNG/NGL) ▪ ASPEN PLUS package ▪ COP of ARC: 0.29–35 ▪ SPC of the modified cycle: 0.28 kWh/Nm3 ▪ SPC reduction vs. baseline: 30% ▪ Operating cost reduction vs. baseline: 9.4% ▪ Annual natural gas consumption cost reduction vs. baseline: 17.3%
Taghavi et al. [173]	0.172	▪ ARC-MR2 unit (MFC-based LNG) ▪ PR EOS ▪ ASPEN HYSYS and PVsyst packages ▪ Energy reduction vs. baseline: 30% ▪ Cold box surface reduction vs. baseline: 31% ▪ COP for the ARC: 0.48
Mehrpooya et al. [406]	0.225	▪ DARC-MR1 unit (SMR-based LNG/NGL/NRU) ▪ ASPEN HYSYS package and M-file code ▪ Energy reduction vs. baseline: 19.36% ▪ CO2 emission reduction vs. baseline: 17.85%

ACS: air cooling system; ARC: absorption refrigeration cycle; C3MR: propane pre-cooled mixed refrigerant; CC: composite curve; COP: coefficient of performance; CRS: cascade refrigeration system; DMR: dual mixed refrigerant; ECC: exergy composite curve; He: helium; LNG: liquefied natural gas; MFC: mixed fluid cascade; MR1/MR2: mixed refrigerant cycle; N₂: nitrogen; NGL: natural gas liquid; NRU: nitrogen rejection unit; PR: Peng–Robinson; PSRK: Predictive Soave–Redlich–Kwong; SPC: specific power consumption; and SMR: single mixed refrigerant.

lists the technical features of selected LNG liquefaction systems utilizing ARCs, ACARs, and DARCs. The findings indicate that employing ARCs, ACARs, and DARCs instead of compression refrigeration cycles for pre-cooling decreases SPC. However, conducting exergy and economic analyses is essential before making any conclusive conclusions.

5.4. Thermoelectric Generator Units in LNG Processes

Natural gas liquefaction is an important and energy-intensive process within the industry. Due to its significance and the high costs associated with production, enhancing the efficiency of the liquefaction process is paramount [131,154,407]. Implementing innovative strategies to recover waste energy can significantly improve these systems’ performances [408]. The TEG modules combined with the LNG production process were a promising option [409]. Because TEGs provide numerous benefits, their use in converting heat directly into electricity has been increasing significantly [410,411,412,413]. Some primary advantages of TEGs include the absence of moving parts, a long operational lifespan, a simple design, and their environmentally friendly nature. However, a notable limitation of TEGs is their low efficiency [412,414,415]. Two locations for TEGs in the LNG cycle were identified: one between the condenser and evaporator and the other between the condenser and ambient air. The results indicated that placing TEGs between the condenser and evaporator was not recommended, as it reduced the COP of the integrated system [409]. However, different outcomes were observed when TEGs were placed between the condenser and ambient air. The results showed that, in this configuration, the COP increased as certain values were optimized. Also, it was noted that the levelized cost of TEGs increased from 2.3 to 47 USD/year as the price of TEGs rose from 1 to 20 USD/W.

5.5. Liquid Air Cold Recovery in LNG Processes

During NG liquefaction, about 800 kJ/kg LNG of exergy destruction occurs in compressor electricity use, and 400 kJ/kg LNG of exergy destruction occurs in regasification, primarily through seawater heat exchange. Studies focus on reducing energy in liquefaction and capturing energy during regasification [137]. Figure 23 depicts the typical layout for liquid air cold recovery in LNG processes.

The C3MR process, recognized as highly efficient, involves a pre-cooling C₃H₈ cycle and an MR unit [325]. As shown in Figure 24, liquid air supports the MR cycle in several configurations, each designed to improve energy efficiency by optimizing cold energy transfer and minimizing refrigerant flow. The results revealed that incorporating liquid air into the LNG supply chain brings several notable advantages [137]. It reduces energy consumption by more than 25% during the liquefaction of NG. Also, the transportation of liquid air is almost cost-free. There is also the opportunity to sell liquid air as N₂ and oxygen. Moreover, this approach helps mitigate environmental impacts by decreasing the amount of ballast H₂O needed for the return transfer. In addition, it capitalizes on the readily available air supply from the environment.

Park et al. [137] illustrated that by returning liquid air to the LNG production facility, the liquefaction and regasification phases could be combined, which leads to a substantial decrease in the energy needed for NG liquefaction. The suggested MR integrated with the liquid air cold recovery process demonstrated an energy requirement of 917.20 kJ/kg LNG, which is 26.1% lower than the SMR process and 6.7% less than the C3MR process. Also, the techno-economic evaluation indicated that the proposed processes are economically viable, reducing total costs by 18.6 to 22.6%. Ghorbani et al. [418] introduced an innovative method for the simultaneous production of liquid CO₂ and LNG. The primary concept of this approach was to use air for energy storage and production, while simultaneously transferring cold energy to facilitate the LNG process. This was achieved through a mixed refrigeration cycle, and recovery of cold energy from liquid air was employed to liquefy NG. Also, a CO₂ liquefaction cycle and a combined cooling and power cycle were integrated to enhance the overall system performance by utilizing the liquefaction of CO₂ and power generation.

5.6. Ejector Refrigeration Cycles in LNG Processes

MRs achieve a high COP, improving energy efficiency, but their intricate composition and thermodynamic properties complicate control, stability, and operational optimization. In these systems, maintaining a constant ratio of stream compositions becomes difficult in the case of a leak [199]. Also, due to the significant power requirements of CRS cycles, incorporating compression–ejector refrigeration systems can reduce energy consumption. An ejector is a device used for suction, transfer, compression, or mixing of gases, vapors, liquids, and even solid particles, functioning similarly to a compressor [419]. A nozzle transforms the potential energy of a high-pressure driving fluid into kinetic energy, allowing it to draw in secondary material by lowering the static pressure [420,421,422]. Therefore, the driving fluid and the secondary material are mixed and condensed in the diffuser. Due to the pressure and velocity changes caused by the fluid’s movement within the divergent and convergent channels, this device is classified as a type of fluid compressor [423,424,425]. Rooholamini et al. [147] introduced an innovative hybrid design for LNG production, incorporating a two-stage ejector refrigeration system and an MR cycle. The hybrid structure was assessed through energy and exergy analyses, pinch methodology, and sensitivity evaluations. The compression–ejector refrigeration cycle was employed for the pre-cooling NG, achieving temperatures as low as 170.5 K. This cycle demonstrated a COP of 0.8635, while the exergy efficiency of the integrated system was 36.42%.

6. Economic, Safety, Risk, and Environmental Perspectives on Liquid Natural Gas

From January 2023 to February 2024, LNG prices have experienced significant volatility across major markets. The Japan Korea Marker (JKM) reached its highest price of 23.90 USD/MMBtu on 3 January 2023. This increase reflected strong demand or supply constraints in Asia. However, by 26 February 2024, the price had dropped to a low of 7.98 USD/MMBtu, indicating increased supply or reduced demand [39]. Also, the Dutch Title Transfer Facility (TTF), a key European benchmark, peaked at 23.15 USD/MMBtu on 3 January 2023. By 28 February 2024, it had declined to 7.23 USD/MMBtu, indicating market stabilization in Europe. In the USA, the Assumed Henry Hub Term Contract Price peaked at 7.55 USD/MMBtu on 4 January 2023, before declining to 4.56 USD/MMBtu on 20 February 2024 [39]. Henry Hub, located in Louisiana, is the primary pricing point for NG in the United States and serves as the benchmark for New York Mercantile Exchange (NYMEX) natural gas futures [426,427].

LNG projects involve significant expenses, encompassing OPEXs and CAPEXs. Modifications become challenging and expensive after construction, minimizing OPEXs and CAPEXs during the essential design phase [236]. Reducing CAPEXs in hybrid cryogenic NG liquefaction often involves downsizing heat exchangers, which may lower initial costs but can lead to operational inefficiencies. A smaller heat exchanger area means less surface for heat transfer, necessitating a larger temperature difference between the hot and cold streams. This inefficiency leads to increased irreversibility and greater demands on compressors, which must operate at lower temperatures. Therefore, efforts to lower CAPEXs often result in higher OPEXs. While CAPEXs benefit from larger gaps between the hot and cold temperature profiles, OPEXs are optimized when these temperature profiles are closer together [428]. A substantial portion of initial and operational investments in these industries is allocated to the refrigeration cycle [128,204,417,429,430].

The cost metrics for LNG liquefaction plants vary depending on the scope and location of the facility (i.e., Middle East, Europe, Asia Pacific, Southeast Asia, America, and Africa). For a complete facility, the costs ranged from 1000 to 1200 USD/TPA (based on 2014) in areas with standard construction costs, while in high-cost locations, the expense increased to between 1400 and 1800 USD/TPA (based on 2014) [431]. For facilities focusing only on the liquefaction process, the costs were lower at 600–800 USD/TPA (based on 2014) in typical locations. However, in high-cost areas such as remote regions, the cost of liquefaction facilities increased to 1000–1200 USD/TPA (based on 2014) [431]. Capital cost estimates found in academic sources (125–1285 USD/TPA based on 2018) were lower than those presented in technical reports (220–2255 USD/TPA based on 2018) [64]. Most plants mentioned in technical reports were within a small capacity range (3–5.5 MTPA), influenced by the standard size of industrial gas turbines and heat exchangers. However, their specific capital costs showed significant variation [64]. Figure 25 depicts the impact of liquefaction unit capacity on capital costs worldwide. The liquefaction train accounts for approximately two-thirds of the total cost of building a complete plant. All costs referenced in this figure have been adjusted to 2024 Q2 levels using the IHS Upstream Capital Costs Index (UCCI) [432]. In studies, the reported price of LNG varied between 0.3 and 0.7 USD/kg [433].

Table 7. Economic characteristics of various hybrid LNG units as reported in the literature.

Liquefaction Technology	Specific Capital Cost 2024 (USD/TPA)	LNG Price (USD/kg)	Capacity (MTPA)	Reference
MCFC-based LNG/NGL	708.6	0.3670	0.8360	Ghorbani et al. [18]
ARC/MR2-based LNG/NGL	591.7	0.2851	0.8360	Ghorbani et al. [18]
C3MR-based LNG/NGL/NRU	621.6	0.3464	1.656	Ghorbani et al. [15]
DMR-based LNG/NGL/NRU	643.8	0.3835	1.656	Ghorbani et al. [15]
ARC/MR1-based LNG/NGL/NRU	476.4	0.3216	1.656	Ghorbani et al. [15]
ACAR/MR1-based LNG	-	0.2519	1	Mehrpooya et al. [404]
C3MR-based LNG	836–930	0.219–0.229	3	Wang et al. [122]
DMR-based LNG	826–1184	0.205–0.235	3	Wang et al. [122]
MFC-based LNG/NGL/NRU	932.4	0.4115	1.656	Ghorbani et al. [191]
MFC-based LNG/He recovery	173.2	0.2649	31.52	Zaitsev et al. [170]
ARC/MR2-based LNG/He recovery	251.8	0.2483	31.52	Zaitsev et al. [170]

provides the economic characteristics of different hybrid LNG units as documented in the literature. The results illustrate that the capital costs in academic references (173.2–1184 USD/TPA based on 2024) were lower than those presented in technical reports (486.7–3839 USD/TPA based on 2024). Also, LNG prices computed in academic examinations (0.2–0.45 USD/kg based on 2024) were lower than those presented in technical reports (0.3 to 0.7 USD/kg based on 2024).

LNG is widely regarded as a cost-effective energy source compared to traditional fossil fuels, but its competitiveness against renewable energy sources varies. NG typically has a lower energy cost per kilowatt-hour than oil and coal [435]. For instance, LNG yields 40–50% fuel savings compared to oil on Oahu and 22–44% on the neighboring islands [436]. Also, LNG produces about 25% of the CO₂ emissions of oil and roughly half that of coal, leading to potential cost savings through lower carbon taxes [435,436]. However, renewable energy technologies (e.g., solar and wind) are becoming increasingly competitive. The levelized cost of electricity for utility-scale solar and onshore wind is, on average, lower than that of gas-fired power stations [437]. Key factors affecting LNG’s cost-effectiveness include transport distances exceeding 3000–5000 km, price volatility linked to global oil markets, and high infrastructure costs. Carbon pricing mechanisms and advancements in energy storage may further impact LNG’s long-term economic viability [437,438].

LNG offers a promising alternative to coal for reducing GHG emissions and particulate pollution in China’s power generation, industrial processes, and district heating systems. Three separate research teams from Canada and the USA conducted independent life cycle assessments (LCAs) of an identical LNG supply chain that was planned to transport LNG from Canada to China [439]. The findings revealed that GHG emissions associated with Canadian LNG used in power and heat production in China range between 427 and 556 gCO₂eq/kWh and between 81 and 92 gCO₂eq/MJ. Compared to coal-fired power generation in China, LNG usage reduced emissions to 291–687 gCO₂eq/kWh, representing a 34–62% decrease in emissions [439]. The LNG sector has experienced fewer incidents than other facilities in the petrochemical industry [440]. An analysis of LNG-related incidents in

Table 8. Review of important LNG accidents: locations, consequences, and economic impact; extracted data from [440].

Location	LNG Facility Type	Deaths	Hospital Admissions	Projectile	Cost (million USD)
Cleveland, Ohio, USA (1944)	LNG containment tank (peak demand facility)	128	200–400	No
La Spezia, Italy (1971)	LNG receiving port	-	-	No
New York, USA (1973)	LNG peak demand facility	37	-	No
Skikda, Algeria (1975)	LNG cryogenic processing facility	-	-	-	-
Cove Point, Maryland, USA (1979)	LNG processing terminal	1	1	No	3
Pinson, Alabama, USA (1985)	LNG peak demand facility	0	6	Yes
Skikda, Algeria (1989)	LNG delivery	0	0	-
Thurley, United Kingdom (1989)	LNG peak demand facility	0	2	No
Indonesia (1993)	Liquefaction process	-	-	No
East of the Strait of Gibraltar (2002)	LNG marine vessel, Norman Lady	-	-	No
Catalonia, Spain (2002)	LNG transport vehicle incident	1	2	Yes
Skikda, Algeria (2004)	Liquefaction process	27	80	No
Nigeria (2010)	LNG Edo	-	-	No
Plymouth/Benton County, WA, USA (2014)	LNG storage vessel	0	1 (Injuries: 6)	Yes	45.74 (2014)

reveals that such events are infrequent, and only a small fraction of these rare occurrences have involved projectile-related incidents.

LNG is a high-risk substance that can impact human health, the environment, and the economic stability of communities. Converting NG into this liquid state significantly decreases its volume, facilitating easier transportation and storage in specialized insulated containers, such as those found on LNG carriers, LNG-fueled vessels, stationary storage tanks, and bunker trucks [441]. The low temperature of LNG introduces severe risks, as it can cause material degradation, such as the cracking of tank walls and ship structures, and can inflict frostbite on individuals who come into contact with it. To mitigate these dangers, all storage tanks, pipelines, and valves that come into contact with LNG must be constructed from specially designed cryogenic materials capable of enduring such low temperatures [442].

The primary dangers associated with LNG storage and bunkering are the possibility of fires and explosions, which leaks or spills can trigger if an ignition source is present. In the absence of ignition, LNG will evaporate, disperse, and potentially form a vapor cloud that can spread through the atmosphere [441]. If an ignition source is introduced, four distinct hazardous scenarios could unfold: (a) a flash fire caused by the ignition of the vapor cloud, (b) a jet fire resulting from a pressurized release, (c) a pool fire that arises from LNG pooling and igniting on a surface, and (d) an explosion of the vapor cloud [61,443]. The severity of the consequences from such fires and explosions is influenced by the initial temperature and composition of the LNG and the diameter of the resulting pool fire [444,445].

The risk assessment methods were classified as qualitative [446,447,448], semi-qualitative [449,450,451,452], quantitative [453,454,455,456,457,458], and further delineated as deterministic [459,460,461,462] or probabilistic [463,464,465,466], as well as hybrid deterministic/probabilistic [467,468,469]. The analytical tools employed in risk assessment studies such as event tree analysis (ETA), fault tree analysis (FTA), failure modes and effects analysis (FMEA), and Bayesian networks were examined to understand their application in LNG risk contexts [470]. Additionally, the outcomes of these assessments were linked to strategic frameworks like risk-based inspection (RBI), risk-based maintenance (RBM), risk-based integrity management (RBIM), and facility siting decisions. Furthermore, the applications of these risk assessments were explored across different LNG domains, including LNG carriers and LNG-fueled ships, LNG terminals and stations, offshore floating units, and LNG plants [470].

In the USA, multiple federal agencies are responsible for approving and monitoring LNG facilities’ safety and operational integrity. These include the Federal Energy Regulatory Commission (FERC), the Pipeline and Hazardous Materials Safety Administration (PHMSA) under the Department of Transportation (DOT), the USA Coast Guard, and various other federal bodies, depending on the site’s location, the type of materials handled, and the quantities involved. Additionally, industry groups like the National Fire Protection Association (NFPA) have established standards related to projectile hazards. Also, NFPA 59A [471] provides specific guidelines for membrane tanks in this context. Moreover, recommendations regarding the thickness of local impact-resistant materials, with considerations for penetration, perforation, and scabbing, are provided for stationary LNG storage tanks based on the guidelines outlined in CEB 187, a report titled Concrete Structures Under Impact and Impulsive Loading [472]. In addition, other high-risk facilities governed by agencies such as the Nuclear Regulatory Commission (NRC), the Department of Energy (DOE), and the Federal Aviation Administration (FAA) under the DOT also assess projectile risks and offer further guidance that can be adapted or enhanced as necessary [440]. The most crucial aspects derived from the regulations and guidelines governing LNG operating ports include the following key points:

(a) LNG storage facilities must adhere to the Seveso III Directive, as well as the EN 1473 and NFPA 59A standards. These regulations ensure compliance with stringent safety and design criteria [441]. The EN 1473:2021 standard outlines guidelines for designing, constructing, and operating onshore LNG installations over 200 tons. Updated in May 2021, it includes pressurized storage and considerations for climate change impacts [473,474].

(b) LNG bunker trucks are required to operate under the ADR agreement. Buffer vessels must be constructed following the IGC code and must also satisfy the safety conditions outlined in the ADN agreement [441]. The IGC code establishes an international standard for safely transporting liquefied gases by sea, specifying design and construction requirements for ships engaged in this operation [475]. Additionally, they must comply with ISO 28460 and EN 1532 standards.

(c) Ships powered by LNG should be designed and constructed to fully comply with the safety and operational standards set by the IGF code [441]. The IGF code sets mandatory safety requirements for vessels using gas or low-flashpoint fuels. It provides safe design, risk reduction, hazard protection, leak prevention, and proper fuel management. The code also mandates reliable piping systems, containment measures, and regular inspections to maintain high safety standards [476].

(d) The ISO 18683, 16901, and 20519 standards should be meticulously followed to ensure a high standard of safety during LNG bunkering operations [441]. ISO 18683 sets safety guidelines for LNG bunker transfer, covering shore-to-ship, mobile-to-ship, and ship-to-ship operations [477]. ISO 16901 focuses on risk assessment for onshore LNG facilities, addressing hazards at terminals and storage sites [478]. ISO 20519 defines requirements for LNG bunkering systems, including hardware, procedures, and personnel training [479].

The global LNG supply chain also faces risks from potential disruptions at key maritime chokepoints like the Panama and Suez Canals, which could lead to long-term changes in shipping patterns [480,481,482].

Geopolitical and economic risks create volatility in LNG supply chains, disrupting investments and altering global energy strategies. Conflicts and sanctions, such as the Russia–Ukraine conflict (2014–present) and USA–Iran tensions, have forced Europe to diversify suppliers and restricted Iran’s LNG exports. Infrastructure risks (e.g., blockades and sabotage) threaten LNG terminals, as seen in the Qatar diplomatic crisis (2017–2021) and Europe’s growing reliance on LNG after the 2025 halt of Russian gas transit through Ukraine [483,484]. Supply chain fragmentation due to sanctions has redirected LNG flows, with Russia shifting exports to Asia while Europe turns to USA and Qatar supplies [484,485]. Economic risks further challenge LNG markets through supply–demand imbalances, cost competitiveness issues, and stranded assets. While demand growth is slowing (<2% by 2025), supply is set to surge by 250 bcm by 2030, led by the USA and Qatar. Countries such as the EU, Japan, and South Korea are reducing LNG imports due to renewables and nuclear power [483,486]. Qatar’s low-cost production makes it a dominant player, while high-cost producers such as Canada struggle. Moreover, oversupply could render Canadian LNG projects unviable, risking USD 16 billion in taxpayer-funded infrastructure [486]. To mitigate these risks, diversification [483,484], infrastructure security [484], and long-term contracts [485,487] with hybrid pricing models are essential. As global markets shift, investors favor low-cost producers, while high-risk projects face increasing scrutiny.

7. Current Status and Future Outlook of Liquid Natural Gas

Global LNG trade reached a record 401.4 MT in 2023, connecting 20 exporting and 51 importing markets, with 21 markets involved in re-exports. The 8.4 MT increase was driven by falling LNG prices, encouraging spot market purchases, especially in Asia. Although global LNG trade grew in 2023, the annual growth rate of 2.1% was lower than the 5.6% recorded in 2022 [39]. Over the next 5–6 years, global LNG export capacity is expected to grow significantly from around 400 to 700 MTPA, driven by rising demand in Asian economies, particularly China. The reduction in Russian gas exports to Europe led to an increase in European LNG purchases, which drove global prices to unprecedented peaks [488]. The USA became the largest LNG producer, exporting 84.53 MT in 2023, followed by Australia at 79.56 MT, and Qatar at 78.22 MT [39]. Russia and Malaysia remained the fourth and fifth largest exporters, with 31.36 and 26.75 MT, respectively. The Asia Pacific was the largest exporting region with 134.8 MT, followed by the Middle East at 94.69 MT and North America at 84.53 MT. The Asia Pacific also led in imports with 155.32 MT, though this was down by 3.47 MT from 2022, followed by Europe at 121.3 MT and Asia at 105.5 MT. Asia saw the most significant import growth, driven by lower LNG prices [39,489]. China regained its position as the world’s largest LNG importer, with total imports reaching 71.2 MT. This represented a significant increase of 7.6 MT compared to the previous year, driven by rising energy demand and efforts to secure supply. Japan imported 66.1 MT, although this marked a decrease of 6.9 MT from 2022 [39,490]. The intra-Asia Pacific trade route continued to dominate as the largest global LNG trade flow, accounting for 95 MT of trade. Additionally, India imported 1.9 MT more than in 2022, bringing its total to 22 MT, while European imports remained steady at 121.3 MT [39]. Figure 26 depicts market share analysis and quantities by export, import, re-loading, and receiving markets in 2023.

Table 9. Natural gas liquefaction plants built in past years, modified from Refs. [39,493,494,495,496,497].

Middle East
Country	Infrastructure Start Year	Liquefaction Technology	Liquefaction Plant Train	Liquefaction Capacity (MTPA)
Qatar	1996	AP-C3MR	Qatargas 1 T1	3.2
Qatar	1996	AP-C3MR	Qatargas 1 T2	3.2
Qatar	1996	AP-C3MR	Qatargas 1 T3	3.2
Qatar	1999	AP-C3MR	Rasgas 1 T1	3.3
Qatar	1999	AP-C3MR	Rasgas 1 T2	3.3
Qatar	2004	AP-C3MR/SplitMR	Rasgas 2 T3	4.7
Qatar	2005	AP-C3MR/SplitMR	Rasgas 2 T4	4.7
Qatar	2005	AP-C3MR/SplitMR	Rasgas 2 T5	4.7
Qatar	2009	AP-X	Rasgas 3 T6	7.8
Qatar	2009	AP-X	Rasgas 3 T7	7.8
Qatar	2009	AP-X	Qatargas 2 T4	7.8
Qatar	2009	AP-X	Qatargas 2 T5	7.8
Qatar	2010	AP-X	Qatargas 3 T6	7.8
Qatar	2011	AP-X	Qatargas 4 T7	7.8
UAE	1977	AP-C3MR	Adgas LNG T1	1.15
UAE	1977	AP-C3MR	Adgas LNG T2	1.15
UAE	1994	AP-C3MR	Adgas LNG T3	3
Oman	2000	AP-C3MR	Oman LNG T1	3.55
Oman	2000	AP-C3MR	Oman LNG T2	3.55
Oman	2006	AP-C3MR	Oman LNG T3 (Qalhat)	3.3
Yemen	2009	AP-C3MR/SplitMR	Yemen LNG (T1 + T2)	6.7
Southeast Asia
Country	Infrastructure Start Year	Liquefaction Technology	Liquefaction Plant Train	Liquefaction Capacity (MTPA)
Malaysia	1982	AP-C3MR	MLNG Satu T1-T3	8.4
Malaysia	1995	AP-C3MR	MLNG Dua T4-T6	9.6
Malaysia	2003	AP-C3MR	MLNG Tiga T7-T8	7.7
Malaysia	2017	AP-C3MR/SplitMR	MLNG T9	3.6
Malaysia	2017	AP-N	Petronas FLNG Satu (PFLNG1)	1.2
Malaysia	2021	AP-N	Petronas FLNG Rotan (PFLNG2)	1.5
Indonesia	1983	AP-C3MR	Bontang LNG TC-TD	5.6
Indonesia	1989	AP-C3MR	Bontang LNG TE	2.8
Indonesia	1993	AP-C3MR	Bontang LNG TF	2.8
Indonesia	1988	AP-C3MR	Bontang LNG TG	2.8
Indonesia	1999	AP-C3MR	Bontang LNG TH	2.95
Indonesia	2009	AP-C3MR/SplitMR	Tangguh LNG T1	3.8
Indonesia	2009	AP-C3MR/SplitMR	Tangguh LNG T2	3.8
Indonesia	2009	AP-C3MR/SplitMR	Tangguh LNG T3	3.8
Indonesia	2015	AP-C3MR	Donggi-Senoro LNG T1	2
Brunei	1972	AP-C3MR	Brunei LNG T1-T2	2.88
Brunei	1973	AP-C3MR	Brunei LNG T3-T4	2.88
Brunei	1974	AP-C3MR	Brunei LNG T5	1.44
Asia Pacific
Country	Infrastructure Start Year	Liquefaction Technology	Liquefaction Plant Train	Liquefaction Capacity (MTPA)
Australia	1989	AP-C3MR	North West Shelf LNG T1	2.5
Australia	1989	AP-C3MR	North West Shelf LNG T2	2.5
Australia	1993	AP-C3MR	North West Shelf LNG T3	2.5
Australia	1993	AP-C3MR	North West Shelf LNG T4	4.6
Australia	2006	CPOC	Darwin LNG T1	3.7
Australia	2008	AP-C3MR	North West Shelf LNG T5	4.6
Australia	2012	Shell Propane Pre-cooled Mixed Refrigerant	Pluto LNG T1	4.9
Australia	2015	CPOC	GLNG T1	3.9
Australia	2015	CPOC	Queensland Curtis LNG T2	4.25
Australia	2016	CPOC	GLNG T2	3.9
Australia	2016	AP-C3MR/SplitMR	Gorgon LNG T1	5.2
Australia	2016	AP-C3MR/SplitMR	Gorgon LNG T2	5.2
Australia	2016	AP-C3MR/SplitMR	Gorgon LNG T3	5.2
Australia	2016	CPOC	Australia Pacific LNG T1	4.5
Australia	2016	CPOC	Australia Pacific LNG T2	4.5
Australia	2017	CPOC	Wheatstone LNG T1	4.45
Australia	2017	CPOC	Wheatstone LNG T2	4.45
Australia	2018	AP-C3MR/SplitMR	Ichthys LNG T1	4.45
Australia	2018	AP-C3MR/SplitMR	Ichthys LNG T2	4.45
Australia	2019	Shell DMR	Prelude FLNG	3.6
North America
Country	Infrastructure Start Year	Liquefaction Technology	Liquefaction Plant Train	Liquefaction Capacity (MTPA)
USA	2016	CPOC	Sabine Pass T1-T2	9–10
USA	2017	CPOC	Sabine Pass T3-T4	9–10
USA	2018	AP-C3MR	Cove Point LNG T1	5.25–5.3
USA	2019	CPOC	Sabine Pass T5	5
USA	2019	AP-C3MR/SplitMR	Cameron LNG T1	4.5
USA	2019	Shell MMLS	Elba Island T1	0.25
USA	2019	Shell MMLS	Elba Island T2	0.25
USA	2019	Shell MMLS	Elba Island T3	0.25
USA	2019	Shell MMLS	Elba Island T4	0.25
USA	2019	CPOC	Corpus Christi T1	4.52–5
USA	2019	CPOC	Corpus Christi T2	4.52–5
USA	2019	AP-C3MR	Freeport LNG T1	5.1
USA	2020	AP-C3MR/SplitMR	Cameron LNG T2	4.5
USA	2020	AP-C3MR/SplitMR	Cameron LNG T3	4.5
USA	2020	Shell MMLS	Elba Island T5	0.25
USA	2020	Shell MMLS	Elba Island T6	0.25
USA	2020	Shell MMLS	Elba Island T7	0.25
USA	2020	Shell MMLS	Elba Island T8	0.25
USA	2020	Shell MMLS	Elba Island T9	0.25
USA	2020	Shell MMLS	Elba Island T10	0.25
USA	2020	AP-C3MR	Freeport LNG T2	5.1
USA	2020	AP-C3MR	Freeport LNG T3	5.1
USA	2021	CPOC	Corpus Christi T3	4.52–5
USA	2022	CPOC	Sabine Pass T6	5
USA	2022	BHGE SMR	Calcasieu Pass LNG T1	0.56
USA	2022	BHGE SMR	Calcasieu Pass LNG T2	0.56
USA	2022	BHGE SMR	Calcasieu Pass LNG T3	0.56
USA	2022	BHGE SMR	Calcasieu Pass LNG T4	0.56
USA	2022	BHGE SMR	Calcasieu Pass LNG T5	0.56
USA	2022	BHGE SMR	Calcasieu Pass LNG T6	0.56
USA	2022	BHGE SMR	Calcasieu Pass LNG T7	0.56
USA	2022	BHGE SMR	Calcasieu Pass LNG T8	0.56
USA	2022	BHGE SMR	Calcasieu Pass LNG T9	0.56
USA	2022	BHGE SMR	Calcasieu Pass LNG T10	0.56
USA	2022	BHGE SMR	Calcasieu Pass LNG T11	0.56
USA	2022	BHGE SMR	Calcasieu Pass LNG T12	0.56
USA	2022	BHGE SMR	Calcasieu Pass LNG T13	0.56
USA	2022	BHGE SMR	Calcasieu Pass LNG T14	0.56
USA	2022	BHGE SMR	Calcasieu Pass LNG T15	0.56
USA	2022	BHGE SMR	Calcasieu Pass LNG T16	0.56
USA	2022	BHGE SMR	Calcasieu Pass LNG T17	0.56
USA	2022	BHGE SMR	Calcasieu Pass LNG T18	0.56
Trinidad and Tobago	1999	CPOC	Atlantic LNG T1	3
Trinidad and Tobago	2002	CPOC	Atlantic LNG T2	3.3
Trinidad and Tobago	2003	CPOC	Atlantic LNG T3	3.3
Trinidad and Tobago	2005	CPOC	Atlantic LNG T4	5.2
South America
Country	Infrastructure Start Year	Liquefaction Technology	Liquefaction Plant Train	Liquefaction Capacity (MTPA)
Peru	2010	AP-C3MR/SplitMR	Peru LNG T1	4.45
Former Soviet Union
Country	Infrastructure Start Year	Liquefaction Technology	Liquefaction Plant Train	Liquefaction Capacity (MTPA)
Russia	2009	Shell DMR	Sakhalin 2 T1	4.8
Russia	2009	Shell DMR	Sakhalin 2 T2	4.8
Russia	2017	AP-C3MR	Yamal LNG T1	5.5
Russia	2018	AP-C3MR	Yamal LNG T2	5.5
Russia	2018	AP-C3MR	Yamal LNG T3	5.5
Russia	2019	Air Liquide Smartfin	Vysotsk LNG T1	0.66
Russia	2021	Novatek Arctic Cascade	Yamal LNG T4	0.9
Russia	2022	Linde LIMUM	Portovaya LNG T1	1.5
Africa
Country	Infrastructure Start Year	Liquefaction Technology	Liquefaction Plant Train	Liquefaction Capacity (MTPA)
Algeria	1978	AP-C3MR	Arzew GL1Z T1-T6	7.9
Algeria	1981	AP-C3MR	Arzew GL2Z T1-T6	8.4
Algeria	2013	AP-C3MR/SplitMR	Skikda GL1K T1 (Rebuild)	4.5
Algeria	2014	AP-C3MR/SplitMR	Arzew GL3Z (Gassi Touil) T1	4.7
Nigeria	1999	AP-C3MR	NLNG T1	3.3
Nigeria	1999	AP-C3MR	NLNG T2	3.3
Nigeria	2002	AP-C3MR	NLNG T3	3.3
Nigeria	2005	AP-C3MR	NLNG T4	4.1
Nigeria	2006	AP-C3MR	NLNG T5	4.1
Nigeria	2007	AP-C3MR	NLNG T6	4.1
Egypt	2005	AP-C3MR/SplitMR	Damietta LNG T1	5
Egypt	2005	CPOC	Egyptian LNG (Idku) T1	3.6
Egypt	2005	CPOC	Egyptian LNG (Idku) T2	3.6
Libya	1970	AP-SMR	Marsa El Brega LNG	3.2
Equatorial Guinea	2007	CPOC	EG LNG T1	3.7
Papua New Guinea	2014	AP-C3MR	PNG LNG T1	3.45
Papua New Guinea	2014	AP-C3MR	PNG LNG T1	3.45
Cameroon	2018	Black & Veatch PRICO	Cameroon FLNG	2.4
Congo	2024	Black & Veatch PRICO	Tango FLNG	0.6
Mozambique	2022	AP-DMR	Coral South FLNG	3.4
Angola	2013	CPOC	Angola LNG T1	5.2
Europe
Country	Infrastructure Start Year	Liquefaction Technology	Liquefaction Plant Train	Liquefaction Capacity (MTPA)
Norway	2007	Linde MF	Snohvit LNG T1	4.3

presents LNG plants built across different regions in recent years, considering the infrastructure, start years, liquefaction technology, plant train details, and liquefaction capacities. Although projects from countries such as Qatar, Iran, and Russia have the potential to offer lower-cost LNG, various challenges may hinder their progress. As a result, USA projects are such to become key suppliers in the global market, maintaining the long-term market clearing price at approximately 7 USD/MMBtu [491]. Over 50 LNG projects are vying to meet the projected supply–demand gap, which is expected to reach around 30 to 60 MTPA from 2040 onward. This gap will be managed by LNG projects with cost economics below 8–9 USD/MMBTU, primarily from North America. Considering the anticipated growth in LNG demand through 2050, the 2030–2040 period may represent the final opportunity for developing new LNG projects [492].

Since 2016, the USA has significantly expanded its LNG infrastructure and emerged as a major force in the global LNG market. The country’s growth in LNG capacity has been driven by multiple large-scale projects, reflecting a major leap in its liquefaction capabilities. The most important LNG facilities in the USA include Sabine Pass LNG, Cove Point LNG, Corpus Christi LNG Texas, Elba Island LNG Georgia, Cameron LNG Louisiana, Freeport LNG Texas, and Calcasieu Pass LNG Louisiana [494,498]. Figure 27 depicts the LNG facilities of Sabine Pass LNG, Cove Point LNG, Corpus Christi LNG, and Elba Island LNG in the USA.

The Sabine Pass LNG facility, located in Louisiana near the Texas border, is a modern infrastructure within the vibrant Louisiana wetlands and offers impressive views of the Gulf of Mexico. This facility began operations in February 2016 and has six fully operational liquefaction units, known as “trains”, each with a production capacity of approximately 5 MTPA LNG, totaling 30.6 MTPA [493]. Sabine Pass, owned by Cheniere Energy, is located less than four nautical miles from the Gulf of Mexico and provides convenient vessel access [502]. The facility has access to various regional and national pipelines, which ensures a stable and diverse NG supply. Sabine Pass has five storage tanks and three berths, which enables it to handle substantial LNG export volumes with efficiency [493].

Cove Point LNG, located near Lusby on the western shore of the Chesapeake Bay in Southern Maryland, south of Washington, D.C., is a complete facility for LNG import, export, and liquefaction [503]. The facility, which occupies 131 acres of a 1000-acre site, initially began importing LNG in 1978, though operations were paused in 1980 due to rising gas prices. Construction of the liquefaction facility commenced in October 2014, while the Cove Point LNG facility in Maryland, with a capacity of 5.25 MTPA, began operations in March 2018 [496,504]. Cove Point LNG is operated by BHE GT&S, a subsidiary of Berkshire Hathaway Energy, and is jointly owned by Dominion Cove Point LNG LP (50%), Berkshire Hathaway Energy (25%), and Brookfield Asset Management (25%) [39,504]. The facility is well connected to the major Mid-Atlantic gas transmission networks; it features infrastructure such as one LNG liquefaction train, seven storage tanks, and state-of-the-art equipment like GE Frame 3 and Solar Titan turbines [504].

The Corpus Christi Liquefaction (CCL) facility in South Texas, the first greenfield LNG export site in the USA, reflects Cheniere’s commitment to efficiency by completing infrastructure ahead of schedule while ensuring safe and dependable energy production [505]. CCL, located on over 1000 acres in San Patricio County, consists of three liquefaction trains, each with a capacity of approximately 5 MTPA of LNG, completed on time and within budget. The facility includes three storage tanks (160,000 m³ each) and two berths designed to accommodate the world’s largest LNG carriers [497]. With access to abundant, low-cost NG and a fully permitted expansion project, CCL is well positioned for Cheniere’s growth as global demand for cleaner energy solutions increases.

The Cameron LNG project is a 13.5 MTPA LNG export facility under construction in Louisiana, USA [506]. Valued at USD 10 billion, the project is being developed by a consortium led by Sempra LNG and Midstream (50.2%), with Total, Mitsui, and Japan LNG Investment each holding 16.6% shares [507]. Located in Hackberry, near the Calcasieu Pass LNG terminal, the facility covers 203.2 hectares. The initial stage of the Cameron LNG project, focused on processing and exporting LNG, is set to feature three liquefaction units, each with a capacity of 4.5 MTPA. This phase will also incorporate a recently constructed LNG storage tank with a full-containment design and a capacity of 160,000 m³, alongside three pre-existing storage tanks of comparable size originally designed for the Cameron LNG import terminal. Moreover, the facility will include storage tanks for refrigerants and condensate products and a designated area for truck-loading operations [506]. The proposed second expansion stage will involve adding two more LNG production units, each with a capacity of 4.98 MTPA. This development will also introduce another storage tank with a capacity of 160,000 m³, along with two compressor units for handling BOG, two tanks for storing liquid N₂, and separate tanks for storing diesel and condensate [506].

The Freeport LNG facility on Quintana Island near Freeport, Texas, is an NG liquefaction and export site with a 20 MTPA capacity. Currently, three of its four liquefaction trains are operational (i.e., 15 MTPA), with the first starting in December 2019, the second in January 2020, and the third beginning operations in May 2020 [508,509,510]. Initially developed for LNG import and regasification, the site still maintains much of its original infrastructure, including storage tanks and pipeline connections. The Freeport LNG export terminal is about 10 km from the sea buoy of Freeport Port and includes a ship channel that is 600 feet wide. The facility, located close to extensive pipeline networks, employs Air Products’ C3MR process and operates with General Electric 75 MW motors equipped with variable frequency drives, establishing it as one of North America’s first large-scale electric LNG plants [508].

Venture Global LNG fully owned the Calcasieu Pass LNG project, a 10 MTPA LNG export terminal built by its subsidiary in Louisiana, USA [511]. Situated on a 930-acre site in Cameron Parish, near Lake Charles, where the Calcasieu Ship Channel meets the Gulf of Mexico, the facility included nine liquefaction blocks, each with a capacity of 1.2 MTPA, and two LNG storage tanks of 200,000 m³ [512]. The project utilized a modular construction approach, with each block housing two LNG trains of 0.56 MTPA [39]. In addition, the site featured a marine terminal on its western side that can accommodate LNG carriers with capacities of up to 185,000 m³ [512].

The Elba LNG export project, also called the Elba liquefaction project, focuses on converting an existing LNG import and regasification terminal at Elba Island, Chatham County, Georgia, into a 2.5 MTPA LNG export facility. The project primarily involves the installation of ten liquefaction units at Southern LNG (SLNG) Company’s existing Elba Island LNG terminal, situated on Elba Island in the Savannah River, Chatham County, Georgia, USA [501]. The project, executed in two phases, is managed by Elba Liquefaction Company (ELC), a joint venture of Kinder Morgan (51%) and EIG Global Energy Partners (49%), and operates using Shell’s MMLS liquefaction technology [513].

By 2028, most of the new LNG capacity will be concentrated in the USA and Qatar. This shift is expected to push Australia, the leading LNG exporter in 2021 and 2022, to a distant third among global suppliers. At the same time, significant new LNG capacity is being developed in Russia, Canada, and several African countries [494].

Table 10. Global LNG expansion: key projects and emerging challenges by 2028, extracted from Ref. [494].

Country	Key LNG Projects (MTPA)	Total Capacity (MTPA)	Additional Details
USA	▪ Plaquemines LNG (18) ▪ Golden Pass LNG (16) ▪ Rio Grande LNG (15) ▪ Port Arthur LNG (12) ▪ Corpus Christi LNG Expansion (10)	71	USA LNG export capacity is set to increase from 94 MTPA today to 172 MTPA by 2028. A 1.4-MTPA project in Mexico using USA gas has recently begun operations, with an additional 6 MTPA in the pipeline.
Qatar	▪ North Field project (first phase expected online by 2025/2026, 48 MTPA by 2028, another 16 MTPA by 2030)	64	Qatar has the lowest LNG production costs globally, attributed to its vast, low-cost, and liquid-rich gas reserves.
Russia	▪ Arctic LNG 2 (initial phase of 20 MTPA)	20	The project has encountered delays due to international sanctions with later phases potentially facing further setbacks.
Canada	▪ LNG Canada (14 MTPA) ▪ Woodfibre LNG (small-scale, significant cost increases)	14	Canada’s first commercial LNG plant is expected to come online by 2025/2026, although delays and cost overruns have hampered the country’s broader LNG ambitions.
Africa	▪ Floating LNG projects in Republic of Congo (2 projects) ▪ Mauritania–Senegal the floating LNG ▪ Nigeria LNG expansion and new train, small project in Gabon	14	Several LNG projects have been proposed in Mozambique, but local opposition, social unrest, and security risks for project personnel have slowed progress.

provides an overview of global LNG expansion, highlighting key projects and anticipated challenges by 2028.

Although the global LNG production capacity is set to expand, operational and supply challenges will hinder actual output. One-third of LNG plants have faced significant issues, including Yemen’s and Libya’s mothballed terminals [494] and the Freeport LNG explosion in 2022 [514]. Gas supply issues have also affected plants in several countries (e.g., Angola, Algeria, Indonesia, Australia, Nigeria, and Malaysia). These disruptions caused the global LNG fleet to operate at 87% capacity in 2023, though strong demand led some projects to exceed their rated output [494].

8. Key Thermodynamic Insights on LNG Units

Key technical insights derived from this review are summarized below:

• The academic literature review on various LNG liquefaction processes provides significant variability in SPC across different technologies. SMR processes were widely utilized in onshore liquefaction plants, with the SPC ranging from 0.22 to 0.48 kWh/kg LNG. Also, the C3MR processes as common technology showed SPC values between 0.2 and 0.41 kWh/kg LNG. The CPOC processes demonstrated an SPC of 0.205–0.341 kWh/kg LNG, although technical reports suggested a slightly higher consumption of 0.324–0.384 kWh/kg LNG. The DMR processes exhibited an SPC of 0.212–0.414 kWh/kg LNG, while the MFC processes recorded an SPC of 0.196–0.423 kWh/kg LNG.
• The MFC process had the lowest SPC, which made it the most energy-efficient option. The DMR and C3MR processes also showed strong energy efficiency, while the SMR processes, though less efficient, maintained acceptable energy consumption compared to others. The SPC for LNG processes indicated that SMR cycles ranged between 0.3 and 0.4 kWh/kg LNG, whereas C3MR, DMR, and MFC cycles had SPCs below 0.3 kWh/kg LNG. Energy and economic analyses revealed discrepancies in expected relationships: more complex liquefaction processes did not necessarily result in lower energy consumption, and energy-efficient processes did not always lead to lower total annualized costs. Although the SMR process consumed the most energy, it remained the most cost-effective.
• Exergy efficiency is an important metric for assessing the performance of LNG liquefaction processes, and it varies among the different technologies. SMR-based LNG processes had reported exergy efficiencies of 30–67.8%, indicating a broad range of performance depending on operational conditions. The C3MR processes that are widely adopted in the industry showed exergy efficiencies of 29.2–65.2%. Also, the DMR processes had documented exergy efficiencies of 28.2–62.3%. Meanwhile, the MFC processes demonstrated a higher efficiency (i.e., 51.8–62.8%), making it one of the more effective options in terms of energy conversion. These findings highlight the importance of optimizing exergy efficiency to enhance overall process performance and reduce operational costs in LNG production.
• The PR, ideal gas law, SRK, BWR, REFPROP database, and Lee–Kesler EOSs were utilized in cryogenic NG liquefaction. The study demonstrated that the PR EOS achieved greater accuracy than the other EOSs. It is particularly effective for phase balance analysis and predicting MRs’ enthalpy and entropy trends. According to the literature findings, the components of the refrigerant mixture should be chosen from a combination of several pure substances with low and high boiling points to cover the wide temperature range required by the process effectively. As the cooling temperature decreases, a mixture with components that have a lower boiling point (and lower molecular weight) should be employed. The design and development of cryogenic cycles using MRs for low-temperature processes involve several important decisions regarding the composition percentages of refrigerant components, operating pressures, temperatures, and the overall arrangement of the cryogenic cycle. Analysis results indicate that the power consumption of refrigeration cycles with a specific arrangement is highly dependent on the operating pressures and the composition percentages of refrigerant components while being less influenced by the sub-cooling temperature parameter. The modified combined and exergy diagrams provide a detailed graphical indication of the cryogenic cycle’s arrangement and its deviation from the optimal configuration, serving as a quality indicator. These diagrams assist in adjusting the structure and arrangement of equipment within the cryogenic cycle to improve efficiency.

9. Summary and Conclusions

Natural gas reserves are unevenly distributed globally, with significant concentrations in specific regions, which makes its transportation an important industry for widespread utilization. Liquefaction offers the most efficient and cost-effective alternative to pipelines. Liquefying NG significantly reduces its volume, which allows it to be transported by ships to various destinations, where it is later regasified. However, the main challenge in expanding LNG usage is the high energy demand of the liquefaction process. Current research focuses on reducing energy consumption in LNG production, which lowers operating costs and enhances system efficiency. This review investigates the thermodynamic aspects of different LNG technologies, covering cryogenic processes, NGL recovery, N₂ rejection, helium recovery, and LNG systems. In addition, various methods to enhance the performance of hybrid liquefaction processes are examined, including optimization algorithms, MR systems, ARCs and DARCs, ACAR, TEG, liquid air cold recovery, ejector refrigeration cycles, and the integration of renewable energy and waste heat. The main outcomes of the review study on hybrid cryogenic natural gas systems are summarized below:

• The GSP is commonly employed to maximize C₂H₆ recovery at minimal cost. Integrating LNG, NGL, and NRU processes involves removing the reboiler and condenser from the demethanizer and N₂ removal towers. In this approach, side streams are extracted from various trays of the towers, cooled using multi-stream heat exchangers, and then reintroduced into the tower. This integration of low-temperature NG processes leads to reduced energy consumption and lower operating costs. While this integration simplifies the system by eliminating additional equipment such as reboilers, condensers, and separate refrigeration cycles, it also increases the complexity of the overall structure. Therefore, a detailed economic analysis is essential to assess the investment required for hybrid cryogenic NG liquefaction.
• Helium production processes can be effectively integrated with the final cooling stage in LNG production to enable effective extraction. This integration was designed to minimize energy consumption and simplify the overall system design. The results illustrated that flash-based processes required fewer pieces of equipment and were simpler to configure compared to other methods, such as distillation or a combination of flash evaporation and distillation. This process had a lower SPC and provided an efficient rate of helium extraction.
• Different approaches were employed to improve the performance of cryogenic NG processes. These approaches included using optimization algorithms, MR systems, ARCs, DARCs, ACAR, TEG, liquid air recovery systems, ejector refrigeration cycles, and integration with renewable energies and waste heat recovery. Optimization strategies for LNG units focused on various objectives, including minimizing power consumption, maximizing exergy efficiency, increasing production, reducing operating costs, lowering total annual costs, minimizing environmental impact, improving economic parameters (e.g., profit and net present value), and addressing multi-objective functions. The results illustrated that using ARC, ACAR, DARC, and liquid air recovery cycles instead of CRC cycles for pre-cooling in hybrid cryogenic NG processes reduced SPC and operating costs. However, it is necessary to consider exergy and initial investment cost analyses. Integrating renewable thermal energy (e.g., solar and geothermal) and industrial waste heat with hybrid cryogenic NG process units was possible through three methods: using waste heat in ARC/DARC units, power generation systems, and a combination of ARC/DARC and power generation cycles. Using renewable and industrial waste heat in ARC cycles was found to be more efficient than the other two methods.
• Capital cost estimates found in academic sources (173.2–1184 USD/TPA based on 2024) were considerably lower than those presented in technical reports (486.7–3839 USD/TPA based on 2024). Also, LNG prices calculated in academic studies (0.2–0.45 USD/kg based on 2024) were considerably lower than those presented in technical reports (0.3–0.7 USD/kg based on 2024).
• LNG has significant risks to human health, the environment, and economic stability due to its hazardous properties and the complexities of its storage and transportation. The conversion of NG to LNG reduces its volume, enabling easier transport in specialized insulated containers, but the extremely low temperatures of LNG can cause material degradation and severe injuries. To mitigate these risks, storage and transport infrastructure can use cryogenic materials designed to withstand such conditions. Major risks associated with LNG include fires and explosions that can occur due to leaks and spills, with the severity depending on various factors such as LNG composition and temperature. Different qualitative and quantitative risk assessment methods were employed to evaluate these risks and inform strategic safety frameworks such as risk-based inspection and maintenance. In the USA, multiple federal agencies, including the FERC, PHMSA, and the United States Coast Guard, regulate LNG facilities’ safety and operational integrity. These efforts are supported by industry standards such as NFPA 59A and the Seveso III Directive, which set stringent safety criteria for LNG storage and bunkering operations.
• While Qatar, Iran, and Russia could offer lower-cost LNG, challenges may limit their progress, positioning USA projects as key suppliers and keeping the long-term price around 7 USD/MMBtu. Over 50 LNG projects are competing to fill a 30–60 MTPA supply gap from 2040 onward, with North American projects, costing below 8–9 USD/MMBtu, expected to lead. The 2030–2040 period may be the last chance for new LNG developments due to growing demand through 2050.
• To further this research, examining environmental parameters in the NG transmission supply chain, including different methods such as LNG, CNG, NGH, and gas-to-liquid (GTL) conversion, would be beneficial. Furthermore, updating and reviewing codes and standards related to the physical storage of NG (particularly for LNG and CNG) can be considered. The use of multi-component refrigeration cycles with lower SPC compared to C3MR and CPOC in industrial applications could also be explored.