1. Introduction

Nowadays, various research activities and policy initiatives are being undertaken towards the achievement of climate neutrality and zero greenhouse gases emissions by 2050. One of the most significant of these initiatives is the European Grean Deal [1]. The achievement of the targets set by the European Grean Deal requires the maximization of renewable energy sources utilization. In this context, geothermal energy can play a pivotal role, as it is a clean energy source that appears to have the highest technological potential compared to other renewable energy sources [2] since it is renewable, environmentally friendly, and has an exceptional availability (capacity factor >90%) [3].

In terms of availability, only a relatively small percentage of geothermal fields, e.g., ~10–15% in Southern Europe, have temperatures exceeding 80 °C. Conversely, a vast geothermal potential can be found in low-enthalpy geothermal fields (at a maximum temperature in the range of 60-75 °C), which remains mostly unexploited despite being typically located at shallow depths where drilling costs are considerably lower. This is due to a number of complex factors, including limited regulatory and market support frameworks and the prioritization of other renewable energy sources such as wind and solar power. These low-enthalpy geothermal fields account for almost ~40% of the total geothermal supply in Southeast Europe [4]; consequently, their utilization for power generation represents an attractive opportunity. Efforts for electricity power generation from low-enthalpy geothermal fields have previously been reported in countries such as USA (Chena, Alaska, 400 kW), Mexico, the Philippines, Indonesia, Austria, Germany, Turkey (Atca Project, Denizli, 32.8 MW) and Iceland [2,5,6]. However, in most developing countries, low enthalpy geothermal power generation has not yet received significant attention for electricity production, as there are concerns about its economic viability for commercial- scale electricity generation [2,5].

Recent geostrategic developments along with the resulting rising energy costs and uncertainty over future energy supplies, have improved the attractiveness of low-enthalpy geothermal energy resources, which can be exploited for electricity generation using binary cycles [7–9]. For the utilization of low-enthalpy geothermal fields, binary cycle power plants represent the most promising option, as they have the capability to exploit lower-temperature level sources and binary thermodynamic cycles [4]. This technology has virtually no greenhouse gas emissions (GHG) into the atmosphere [2,5,7] and is also promising due to its simplicity and relatively limited number of components/parts, most of which are commercially available. As a result, binary cycles power plants are the fastest-growing category of geothermal energy utilization systems [10].

However, to sustainably meet future energy requirements, the use of geothermal energy must be both technically and economically viable. Thus, the effective utilization of low-enthalpy geothermal energy requires careful consideration of cost-related economic factors [11]. Various factors affect the performance of binary-cycle power plants: the temperature level of the external source, the working fluid, the type of turbine, the condensation temperature, and the operating conditions in general. Therefore, the configuration of a typical binary-cycle power plant varies significantly, depending on the specific application, and no standard design has been established [12–14] as the absolute optimum.

The first low-enthalpy binary-cycle power plants were constructed in 1952, in the Democratic Republic of Congo, 200 kW (geothermal water temperature 91 °C) and in 1967 in Kamchatka in the USSR, (670 kW, geothermal water temperature 85 °C) [13]. These plants demonstrated that it is possible to generate electricity profitably if the condensation temperature of the working fluid is kept sufficiently low, as evidenced by the case of the plant in Chena, Alaska (210 kW, geothermal water temperature ~74 °C, average condensation temperature ~5 °C) [15] and the case of the LOW-BIN project, “Efficient Low Temperature Geothermal Binary Power” [6,16].

Greece is particularly rich in low-enthalpy geothermal fields and has a long experience in geothermal energy utilization, mainly for medical and therapeutic purposes. However, until now the use of geothermal energy for electricity generation has not been possible, with the only noteworthy effort being the 2 MW geothermoelectric pilot plant that operated in Milos for only ten months in the late 1980s. This plant, which utilized a high enthalpy field, was not fully accepted by the local community, mainly due to a lack of awareness concerning the environmental and economic advantages of geothermal energy. On the other hand, despite the existence of many low-enthalpy geothermal fields, in other Greek areas (e.g., Lesvos, Evros and Sidirokastro), these are currently used almost exclusively for greenhouses’ heating purposes [17,18]. The growing demand for clean, efficient, and renewable energy sources renders low-enthalpy geothermal energy exploitation as one of the most promising sectors in the future energy mix, particularly due to its abundance in countries such as Greece. In this context, binary-cycle power plants, typically based on organic Rankine cycles (ORC), have emerged as a particularly effective technological solution for electricity generation from medium and/or low-temperature geothermal fields, which are widespread in Greece. Unlike the classical Rankine cycle that uses water as the working fluid, the organic cycle uses organic fluids (e.g., R134a) with a lower boiling point. This allows for efficient conversion of thermal energy into electrical energy even when the available temperatures are below 150 °C. This makes ORC technology ideal for exploiting low and medium enthalpy geothermal resources.

In a binary cycle ORC power plant, the selection of an appropriate working fluid can be of critical importance especially when environmental parameters such as GWP- Global Warming Potential or ODP - Ozone Depletion Potential are considered. Working fluids such as R134a, R236ea, R245fa, and R142b are suitable for low enthalpy applications in the temperature range of 50-100 °C [19]. However, when more strict environmental concerns are considered the use of these fluids can become potentially problematic, since their GWP and ODP values are usually relatively high or present safety level concerns. In addition, taking into account the planned hydrofluorocarbon (HFC) phase-down in Europe in 2030 [20], due to the fact that HFCs may account as responsible for the majority of fluorinated greenhouse gases emissions, the substitution of HFCs with cleaner, environmentally friendlier, and non-toxic working fluids has become increasingly important. Consequently, the use of less conventional fluids such as carbon dioxide, with a GWP of 1 and an ODP of 0, is gaining attention, making the quantification of their thermodynamic performance and component costs highly relevant.

Such an effort is undertaken in the current work which investigates the thermodynamic performance and the components cost of a low-enthalpy geothermal binary-cycle power plant using supercritical carbon dioxide as its working fluid. Supercritical carbon dioxide binary-cycles gained substantial attention during the past decade due to their potential for achieving higher thermodynamic efficiency compared to conventional steam-based Rankine cycles. The low-enthalpy geothermal field under investigation is located in Sidirokastro region (Serres area, Greece), shown in Figure 1, where recent studies conducted by the Hellenic Authority for Geological and Mineral Research identified a geothermal fluid reservoir with a maximum temperature of 78 °C at a depth of over 98 meters, with flow rates exceeding 120 m³/h [21].

Figure 1. The Sidirokastro region in Serres and the estimated location of the low-enthalpy geothermal field (indicated by the blue dot, source: Google Earth).

2. Materials and Methods

2.1. Thermodynamic cycle model

The present work is investigating the utilization of a low-enthalpy geothermal field in the Sidirokastro area for the generation of approximately 250 kW of electrical power. Initially, to estimate the thermodynamic performance of this specific geothermal field, a numerical thermodynamic model was developed, modelling a transcritical Rankine cycle using sCO₂ as the working fluid.

The use of sCO₂ as the working fluid in a binary geothermal power plant based on a transcritical Rankine cycle has many advantages due to its unique properties. More specifically, carbon dioxide is an environmentally friendly, pollution free, non-flammable and widely available working fluid, which is also characterized by low-cost, low toxicity and low corrosivity. Furthermore, sCO₂ is a high-density working fluid enabling the use of highly compact turbomachinery components and heat exchangers. Supercritical CO₂ has also very good heat transfer properties (density, viscosity, thermal conductivity, heat capacity) and critical temperature similar to ambient conditions. Consequently, effective thermal matching can be achieved in the transcritical Rankine cycle during heat transfer processes. Finally, CO₂ has a GWP equal to 1 and an ODP equal to 0, values that provide significant environmental advantages in relation to other working fluids for similar applications (see Table 1).

Table 1. Properties of various working fluids (adapted data from [22-25]).

The geothermal thermodynamic cycle examined in this study is a simple Rankine cycle comprising a geothermal water heat exchanger, an sCO₂ turbine, a condenser, and a pump, with sCO₂ used as the working fluid. The model was developed using the Cape Open to Cape Open (COCO) simulator [26] incorporating the most recent available data on geothermal source properties [27] and typical performance-characteristics of the main thermodynamic cycle components. The performance of the COCO simulator was validated in relation to commercial software, such as ASPEN, in various studies, e.g. in [28], showing close agreement. The geothermal power plant model is shown in Figure 2. The modelling equations and assumptions are presented in Table 2.

Figure 2. Geothermal power plant model developed in the COCO simulator.

Table 2. Thermodynamic equations and assumptions applied.

For the estimation of the sCO₂ thermophysical properties, the Peng-Robinson (1976) state-equation [29] was used based on the conclusions of the works of [30–32] where the Peng-Robinson state-equation was employed for modelling the sCO₂ thermophysical properties in ASPEN software.

(1)

P=RTVm−b−aaVm2+2bVm−b2

(2)

a=0.45724R2Tc2Pc

(3)

b=0.07780RTcPc

(4)

a=(1+(0.37464+1.54226ω−0.26992ω2)(1−Tr0.5))2

(5)

Tr=TTc

Concerning the geothermal plant thermodynamic model, it includes a geothermal heat exchanger where sCO₂ at 100bar pressure is heated by a low-enthalpy geothermal water source of maximum temperature of 78 °C. Then, the sCO₂ expands in the turbine, generating power and afterwards is condensed inside a condenser with the use of cooling water of 8 °C temperature and 4bar pressure. Finally, the sCO₂ exits the condenser in the form of liquid to properly feed and protect the pump, where the pressure is increased back at 100 bar. Regarding the selected operating characteristics of the cycle components, an isentropic efficiency of 80% was selected for both the sCO₂ turbine and the pump and an effectiveness of 95% for the geothermal heat exchanger and the water condenser heat exchanger.

2.2. Components cost model correlations

In the present study, a preliminary assessment of the purchase costs of the main geothermal power plant components was conducted based on dedicated correlations from international literature which relate the main power plant components to the key-operational characteristics, such as the power output and the UA parameter. These cost-correlations are mainly based on the works of Weiland et al. [33], Carlson et al. [34] and McCollum and Ogden [35]. Furthermore, where necessary, additional correlations from Towler and Sinnott [36], Blecich and Blecich [37] and Shamoushaki et al. [38] were used. A summary of the adopted cost-components correlations is presented in Table 3.

Table 3. Geothermal power plant cost-components correlations.

To ensure the accuracy of the calculations of the UA parameter in the power plant’s heat exchangers, detailed sub-models of the geothermal heat exchanger and the water-condenser heat exchanger were utilized, in which the overall heat transfer process in the heat exchangers was divided into various sub-units. At the next step, a sub-unit independency study was performed, considering the heat exchangers consisted of varying numbers of sub-units (1,5,10,15 and 20), in order to effectively capture the influence of the sCO₂ thermophysical property variations, mainly of enthalpy and specific heat capacity (C_p), on the calculation of the logarithmic mean temperature difference and thus, to the calculation of the overall UA parameter value in each heat exchanger. These sub-unit models were developed using the COCO simulator, applying four scenarios to both the geothermal water and cooling water condenser heat exchangers, with configurations of 5, 10, 15, and 20 sub-units, as illustrated in Figure 3. Then, the total UA value was calculated as the sum of the UA values of all sub-units of the heat exchangers.

Figure 3. Indicative view of the heat exchanger sub-models for the 5, 10, 15 sub-units.

3. Results

In the next step a thermodynamic analysis of the binary-cycle power plant was conducted taking into account the thermodynamic conditions and components performance characteristics as described in Table 2. The selected operating conditions were determined based on the findings of the work of Karypidis et al. [27]. The main results of the binary cycle power plant are summarized in Table 4. The thermodynamic cycle of the binary plant under investigation is presented in Figure 4.

Table 4. Binary Cycle Main Results.

Figure 4. Thermodynamic cycle of the geothermal power plant in T-s chart.

Furthermore, a sub-unit independency study using 1, 5, 10, 15 and 20 sub-units was performed for the geothermal heat exchanger and the condenser. The aim was to accurately estimate the total UA value for each heat exchanger of the geothermal power plant, as illustrated in Figure 5 and Figure 6, by taking into proper consideration the thermophysical properties variation inside the heat exchangers during the heat transfer process. As one may observe, the total UA values are almost constant after the use of 15 sub-units in order to describe the heat exchange process.

Figure 5. UA vs number of heat exchangers sub-units for geothermal heat exchanger (counterflow alignment is assumed).

Figure 6. UA vs number of heat exchangers sub-units for condenser heat exchanger (counterflow alignment is assumed).

In the subsequent step, based on the thermodynamic cycle results and the estimated UA values of the heat exchangers, the purchase costs of the main components of the geothermal power plant were evaluated using the dedicated correlations from the international literature, as presented in Table 3. The results are illustrated in Figure 7, which presents the percentage distribution of the cost-components.

Figure 7. Percentage distribution of the cost – components.

Additionally, to evaluate the performance of the geothermal power plant under varying condensation temperatures and alternative working fluids, such as R134a, supplementary simulations were performed using the open-source DWSIM software [39], which incorporates the CoolProp library for calculating the thermophysical properties of R134a. The results, presented in Figure 8, correspond to maximum temperature conditions similar to the ones applied of the Sidirokastro geothermal field, taking also into account the conclusions from the works of Karypidis et al. [27] for sCO₂, and Aneke et al. [15] for R134a. A more detailed comparison of the performance of the low enthalpy geothermal field power plant of Chena, Alaska corresponding to the specific conditions described in Aneke et al. [15], carried out using Coolprop free properties library is presented in Table 5, where the presented results are compared to the ones of Aneke et al. [15] which were carried out with the commercial Simulation Software SimTech IPSEpro [40]. The total process that was followed in the analyses in this work is presented in Figure 9.

Table 5. Components cost percentage distribution net values.

Figure 8. Thermal efficiency variation vs condensation temperature for sCO₂ and R134a.

Figure 9. Analysis flow chart.

4. Discussion and Conclusions

In the present study a thermodynamic model was developed in COCO simulator to represent the conditions of the low enthalpy geothermal field located in the region of Sidirokastro near the city of Serres, Greece, incorporating the most recent available data on the geothermal source properties. In the thermodynamic model, the performance characteristics of the geothermal power plant components were defined using benchmark data from the international literature on comparable applications. According to these results, a preliminary analysis of the geothermal power plant efficiency and cost-components distribution was performed. More specifically, as can be seen in Table 4, the performance of the geothermal power plant is characterized by a relatively low thermal efficiency value of 5.46%. This value can be mainly attributed to the combination of (a) relatively low geothermal source temperature, i.e., 78 °C, (b) low turbine pressure ratio, i.e., ~1.925, as a result of the condensation temperature value and (c) relatively high sCO₂ pump consumption. As a result, the geothermal power plant produces ~250 kW since a large part of the turbine generated power of ~425 kW is consumed for the operation of the sCO₂ pump, which requires ~175 kW.

Regarding component costs, which were estimated using the most recent correlations from the international literature, the water condenser and the geothermal heat exchanger were identified as the most cost-intensive elements, together accounting for over 55% of the total equipment purchase cost. the water condenser represents the highest share, accounting for 32.44% of the total power plant component purchase cost, while the geothermal heat exchanger follows as the second most expensive component, corresponding to 23.05% of the total. The cost of the cooling tower, the sCO₂ pump, the sCO₂ turbine and the gearbox cost are also significant corresponding to 12.26%, 11.21%, 7.17% and 5.04% respectively. Finally, the cost of the generator, motors and the water pumps are considerably lower ranging from 2.39% to 0.57%.

For the estimation of the heat exchangers costs, i.e. the water condenser and the geothermal heat exchanger costs, which depend on the UA parameter (the product of the overall heat transfer coefficient and the heat exchanger surface area), detailed sub-models were employed to accurately capture the influence of CO₂ thermophysical property variations on the calculation of the logarithmic mean temperature difference (LMTD) and, consequently, on the UA value. In the cost calculations, the final UA values corresponding to the 20 sub-units configuration, as presented in Figure 5 and Figure 6, were used as the most accurate values. These values are considered stabilized beyond 15 sub-units, showing only about 0.1% and 0.7% deviation for the geothermal heat exchanger and the water condenser, respectively, compared to the 15 sub-units results.

In comparison with one of the most well-documented low-enthalpy geothermal applications in the international literature—the Chena, Alaska power plant using R134a as the working fluid, described by Aneke et al. [15] and Holdmann [41]—the performance of the Sidirokastro low-enthalpy geothermal power plant using sCO₂ is approximately 4% lower in terms of net thermal efficiency. However, the components purchase total cost is more than 40% higher, based on the best available data reported by Holdmann [41] which indicate total project expenses of $2,007,770 by the end of 2006. The cost comparison was performed by adjusting all component costs to 2017 USD values using the Chemical Engineering Plant Cost Index (CEPCI), as reported by Weiland et al. [33], where the CEPCI 2017 index was 567.5.

However, it should be mentioned that a part of this cost difference can be attributed to the fact that the Chena, Alaska power plant operates with a combination of water condenser and air cooler. Owing to Alaska’s low ambient temperatures, the use of an air cooler can be thermodynamically efficient during certain periods, resulting in an approximately 10% reduction in net cost compared to systems utilizing only a water condenser. Furthermore, the use of R134a in the Chena, Alaska power plant facilitates the use of commercially widely available components and heat exchangers, as mentioned in Holdmann [41], resulting in lower cost in relation to sCO₂, for which the technological maturity has not yet reached the same level.

As can be seen also in Figure 8, in all cases the use of R134a results in improved thermodynamic cycle performance in relation to sCO₂. However, taking into consideration the planned hydrofluorocarbon (HFC) phase-down in Europe for 2030 the use of carbon dioxide as working fluid becomes more interesting. Thus, further efforts are required to enhance the performance of sCO₂-based thermodynamic cycles, given the significant environmental advantages associated with replacing HFCs with carbon dioxide.

It is also important to note that future studies should incorporate more detail and up-to-date data in component cost correlations and extend these correlations to a broader range of power outputs (<1 MW), while also considering the latest technological developments and advancements in the field. In addition, detailed thermoeconomic models for the heat exchangers should be developed, in order to correlate the heat exchangers design geometrical and performance characteristics to the achieved thermal effectiveness and imposed pressure drop and thus, to the effect on the binary cycle thermodynamic efficiency. Furthermore, material selection and manufacturing limitations should also be taken into consideration to the heat exchangers purchase cost and the binary cycle power plant economic performance. These actions will increase the cost correlations range of validity, reduce the analysis uncertainty degree, and provide a significant design tool for the optimization of the power plant, whilst also taking into account both thermodynamic and cost-related technoeconomic parameters. Last but not least, future efforts will focus on the evaluation of the thermodynamic and technoeconomic performance of the low-enthalpy geothermal field combined with supplementary renewable energy sources (mainly solar energy). This will allow to increase the temperature of the sCO₂ before the latter enters the turbine, thereby enhancing the power generation potential and the cycle’s thermal efficiency, providing a more cost-effective and environmentally friendly solution.

The present work is a follow-up analysis of previous works concerning the low enthalpy geothermal field located in the region of Sidirokastro in Serres, Greece, this time focusing on a detailed analysis of the binary cycle components cost, which was not so far included. To the authors’ best available information and knowledge this is the first time such an analysis has been presented for low enthalpy geothermal applications specifically refering to Greece’s geothermal fields. Key aspects differentiating the present work in relation to most existing literature works are: (a) the use of supercritical carbon dioxide as working fluid combined with the use of a low enthalpy heat source in a binary cycle, (b) the use of only open source and free software for the modelling approaches and (c) the development of detailed heat exchangers models specifically targeting the accurate calculation of UA values taking into account the working fluid thermophysical properties variation during the heat transfer process.

It worths also to mention that targeting the efficiency improvement of power plants and energy systems, special interest has been presented in international literature on hybrid systems utilizing recuperative and regenerative cycles [42] and power plants combining more than one heat sources, more notably solar and geothermal energy sources, as presented in [43,44]. All these combinations can present enhanced opportunities for more efficient utilization of energy sources as opposed to standalone systems and power plants, especially when low enthalpy geothermal sources are considered as the prime energy source, and thus, are planned to be strongly considered in the near future for low enthalpy geothermal applications as the one presented in this work.

Declarations

Abbreviations

The following abbreviations are used in this manuscript:

References