*4.1. OTEC System Location*

Initially, selecting an appropriate place to locate the OTEC facility is discussed. According to NASA data presented in Table 1, San Andrés Island has the appropriate sea and weather conditions to host an OTEC facility. Table 1 presents the air temperature

(AT), relative humidity (RH), daily solar radiation (DSR), wind speed (WS) and earth temperature (ET).


**Table 1.** Climatological data of San Andrés Island, average of the last ten years. Source NASA.

The main requirement to meet is the condition Δ*T* ≥ 20 ◦C between the surface and deep layers of the ocean. Conveniently, the Δ*T* is achieved as superficially as possible to minimise the water pumping cost.

It can be seen in Figure 1 that the Δ*T* at San Andrés Island is greater than 20 ◦C throughout the year, between the surface and 1000 m depth. In the period of highest temperature, which occurs in September, it is observed that the Δ*T* is around 22.5 ◦C. In the season of lowest temperature, which occurs in March, the Δ*T* is around 20 ◦C. It should be noted that the blank spaces in Figure 1 mean that the depth is less than 1000 m in those places. Consequently, it is convenient to select locations for the OTEC facility where the access to 1000 m depth is easy. To this end, the bathymetry of the San Andrés Island is presented in [14].

**Figure 1.** Differential on water temperature between surface and 1000 m depth around San Andrés Island on March **(left**) and September (**right**).

Figure 2 shows the bathymetric profile for seven points at San Andrés Island. The seven points selected are Airport, Barrio Obrero, Hans Dive Shop, La Loma, Playa Rocky Cay, San José, and Tana. The specific location of these points can be found in [14]. The bathymetric and temperature conditions of the seven points are presented in Table 2. Where *sizemp* is the distance from the coast to where the platform break is generated, *depthpb* is the depth where the platform break starts, *distancef c* is the distance from the coast to reach 1000 m depth, Δ*Tmin* is the minimum temperature differential and Δ*Tmax* is the maximum temperature differential. It can be seen that the maximum width is around 4 km.

**Figure 2.** Bathymetric profiles around San Andrés Island, showing how high depths are reached near the coast.


**Table 2.** Continental platform break in 7 points of San Andrés Island.

Based on the climatological data, the San Andrés Island has an ideal temperature to implement an OTEC plant, in addition to the other necessary characteristics, such as a short distance between 1000 m deep and the coast, with a steep slope and a seabed without many reliefs, the location must have low waves (height less than 3.7 m), and ocean surface currents less than 1.5 m/s, and a low incidence of natural phenomena such as storms, earthquakes, hurricanes, among others [54].

It is concluded that the best location for the OTEC facility is at Tana Point, at coordinates 81º44 W and 12º29 N. Tana presents the 1000 m depth at 2.49 km; another advantage is that the slope decreases rapidly from the break in the platform at 0.35 km until the first kilometre. After that, it descends gently, being an advantage when establishing the mooring system. On the other hand, the Δ*Tmin* is 19.91 ◦C, and the Δ*Tmax* is 22.24 ◦C, being an ideal temperature for the optimal efficiency of an OTEC system. Therefore, this will be the location chosen to analyse the economic feasibility of the OTEC plant.

*4.2. Technical Conditions of the OTEC System*

The technical conditions of the proposed system are described below.

#### 4.2.1. Development of the Energy Model

According to the calculations made in [55], they determined that about 4 m3/s of surface water and 2 m3/s of deep water are required with <sup>Δ</sup>*<sup>T</sup>* ≈ <sup>20</sup> ◦C for each MW of net electricity generated. An average speed of around 2 m/s is required for the seawater to circulate through the pipes from deep water to the surface; in this way, pumping losses more significant than 30% of the gross power are avoided [56]. Taking as a reference [57], which states that 10% of the steam that enters the turbine can be converted into desalinated water, this value is taken as a reference point in the calculations presented here.

An open-cycle OTEC of 2MW is chosen for the design, representing approximately 9.4% of the yearly electricity demand at San Andrés Island (160–187 GWh/year). Table 3 shows the variables used in the energy model.


**Table 3.** Summary of the variables with their respective values calculated for a 2 MW OTEC.

Equations (3)–(10) present the calculations for the energy model; it should be noted that these calculations are based on recommendations made by [55]. The values were adjusted to meet the 2 MW objective of the proposed plant.

$$HWF = 4 \text{ m}^3/\text{s} \cdot 2 = 8 \text{ m}^3/\text{s} \tag{3}$$

$$\text{CWF} = 2 \text{ m}^3/\text{s} \cdot 2 = 4 \text{ m}^3/\text{s} \tag{4}$$

$$\mathcal{WS} = \mathbf{2} \text{ m/s} \tag{5}$$

$$IDHWP = \sqrt{\frac{4 \cdot 8 \text{ m}^3/\text{s}}{2 \text{ m}/\text{s} \cdot \text{\textdegree B}}} = 2.25 \text{ m} \tag{6}$$

$$IDCWP = \sqrt{\frac{4 \cdot 4 \text{m}^3/\text{s}}{2 \text{ m}/\text{s} \cdot \text{\textdegree B}}} = 1.59 \text{ m} \tag{7}$$

$$SFT = 8 \text{ m}^3/\text{s} \cdot 1000 \text{ kg/m}^3 \cdot 1.17\% = 93.75 \text{ kg/s} \tag{8}$$

$$MFDW = 93.75 \text{ kg/s} \cdot 10\% = 9.375 \text{ kg/s} \tag{9}$$

$$DWF = 9.37 \text{ L/s} \tag{10}$$

In Figure 3, it can be seen the system sketch, along with the amounts of water in both liquid and gaseous states, which are calculated for each processing part. Table 4 presents the values calculated in the RetScreen software, taking into account the thermodynamics of a steam turbine [58]. A 2 MW plant can produce up to 15.8 GWh/year.

**Figure 3.** A schematic diagram showing values for an open-cycle OTEC system of 2 MW.


**Table 4.** Operating Specifications for Steam Turbine.

#### 4.2.2. Sketch of the Power Plant

Due to the island's conditions, an offshore floating OTEC is chosen. An offshore installation is selected instead of an onshore one since San Andrés Island has a high density and limited onshore space. On the other hand, the 1000 m depth is obtained near the coast, reducing transmission costs to the coast of both the electrical energy and the water produced. By the calculations shown above, the pipes' sizes and the plant itself are dimensioned. A preliminary design is created and shown in Figure 4.

#### 4.2.3. Electrical Transfer Scheme

Since many companies and businesses have diesel plants, these could continue to be used if the OTEC plant needs to stop for maintenance or it cannot provide the required energy. For these cases, the electrical design is presented in Figure 5.

This transfer circuit would allow switching between the OTEC and the Diesel generation to the grid. The change occurs automatically when the power of the OTEC network is not available, in which case it will take eight seconds, and the backup system will be activated. When the OTEC power returns, the backup system will be automatically deactivated, eight seconds will pass, and the central system will re-enter.

**Figure 5.** Electrical transfer circuit. In case of a failure, the electric transfer circuit is in charge of switching between energy coming from the OTEC to a backup diesel plant.

#### 4.2.4. Analysis of Emissions

The emissions analysis gives a favourable result for the OTEC plant since, if it compares the same generation of energy made with the existing diesel plants, an annual reduction of 2404 tons of CO2 can be obtained, equivalent to 1,032,932 L of gasoline not consumed.

In Table 5, it can be seen the broken down of this data, all of them calculated using the RetScreen software by using statistical data on the production of greenhouse gases in power generation plants.



#### *4.3. Result of Economic Analysis*

Apart from generating electricity, the OTEC power plant offers another essential advantage: potable water production. In this paper, two possible scenarios for the OTEC plant at San Andrés Island are examined, the first is without potable water production, and the second is if potable water is produced and sold as a public service.

In both cases, an initial investment of 50% of the total project value is assumed by the company that builds the OTEC system, while the other 50% is requested from external financing with a 15-year loan. According to the current OTEC plants, it is estimated that the cost of construction of one of these plants is, on average, USD 15,000 per kW, without counting the transmission lines or the adjustments that must be made to the terrain. It should be noted that the economic data for the construction of the plant and the energy production itself were taken based on the OTEC built by the Natural Energy Laboratory of Hawaii Authority (NELHA) [59], also taking into account importation costs and Colombian labour.

#### 4.3.1. Scenario 1: Without Production of Potable Water

Table 6 presents the information from the financial analysis. In this case, the total initial cost is USD 39,245,257 and has an annual cost of USD 1,528,675 for salaries, maintenance and payments for the external financing. The only income is the sale of electric energy, approximately 15,945 MWh per year, which generates an average annual income of USD 2,657,472. Figure 6 shows the behaviour of the cash flow of scenario 1. The project starts to be profitable after year 13; then, the profit exceeds the annual costs.


**Table 6.** Initial and annual costs, and annual income, for Scenario 1.

**Figure 6.** Return on investment, and long-term cash-flow, for Scenario 1.

Using Equations (1) and (2) a LCOE of 0.22 USD/kWh is obtained for scenario 1. Despite this value may be slightly high compared to [35], whose LCOE is around 0.2 USD/kWh, it is observed that the proposed system has a cost of 36% lower than diesel production, which at this time is around 0.3 USD/kWh. Consequently, installing an OTEC facility without desalination is economically viable on San Andrés Island.

#### 4.3.2. Scenario 2: Potable Water as Public Service

In this case, the total initial cost is USD 42,395,257, as can be seen in Table 7. It is more expensive than the first scenario because equipment must be purchased, and rooms must be conditioned to convert the desalinated water into potable water (it is necessary to adjust the pH to 7.7 and eliminate all types of pathogenic organisms by chlorination, perchlorination or ozonation).


**Table 7.** Initial and annual costs, and annual income, for Scenario 2.

The annual cost is about USD 1,930,641 for salaries, maintenance, payments to the bank, and potable water production. It has two incomes:

The sale of electric energy, approximately 15,945 MWh per year, which generates an average annual income of USD 2,657,472.

The sale of potable water as public service, approximately 295,492 m<sup>3</sup> per year, with a standard price of 1 USD/m3. It represents an additional annual income of USD 295,492.

The two incomes add up to an annual total of USD 2,952,964. The project starts to be profitable after year 15. After that, some profits greatly exceed the annual costs, as shown in Figure 7.

**Figure 7.** Return on investment, and long-term cash-flow, for Scenario 2.

For scenario 2, the LCOE obtained is 0.26 USD/kWh, which remains competitive concerning diesel production and makes installing an OTEC facility with desalination economically viable at San Andrés Island.

#### **5. Conclusions**

San Andrés Island has the ideal condition for the location of an OTEC power plant since it meets all the technical and environmental conditions required, and a power plant of this type would bring significant benefits to the island, such as an electrical system more ecological, economic and stable, together with a more continuous potable water service.

The results show that the operation of an OTEC plant at San Andrés can be viable. Of course, it requires a high initial investment, but given that it is a clean technology that does not consume fuels and that can cogenerate associated products, in the long term, the investment can be recovered and eventually give benefits.

However, it must be taken into account that the values obtained in this paper are based on a theoretical analysis; these values would change when implementation is carried out since it is possible that environmental factors and the type of soil, among others, impact the costs of the plant. In this sense, it is recommended to carry out more detailed studies regarding the installation and moorings costs in different parts of the island, which may be lines of future research. However, the analysis performed in this work provides the positive overall conclusion that it is worthy to seriously explore the installation of such a facility in San Andrés since the system is economically feasible.

**Author Contributions:** Conceptualization, J.H., S.S. and A.F.-H.; investigation, S.S. and J.H.; writing original draft preparation, J.H., H.H.-H., S.S., N.A., A.F.-H. and A.I.; writing—review and editing, J.H., H.H.-H., S.S., N.A., A.F.-H. and A.I.; funding acquisition, J.H., A.F.-H. and A.I. All authors have read and agreed to the published version of the manuscript.

**Funding:** This work was partially supported by the ERANET-LAC project, which has received funding from the European Union Seventh Framework Programme, and has been funded by the Ministry of Science, Innovation, Universities through the Spanish Research Agency (PCI2019-103376, ERANet17/ERY 0168) and by the Faculty of Natural Sciences and Engineering of the Universidad Jorge Tadeo Lozano.

**Institutional Review Board Statement:** Not applicable.

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** Not applicable.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **Abbreviations**

The following abbreviations are used in this manuscript:

