Drilling Deeper in Shallow Geoexchange Heat Pump Systems—Thermogeological, Energy and Hydraulic Benefits and Restraints

Abstract

In the last decade, due to climate change concerns and new environmental regulations in the EU, there was a tremendous rise in installed heat pump systems in new homes and buildings. The majority of these installed units are related to air-source heat pumps, as they offer a good trade-off between capital and operating expenses. However, when analysing heating and cooling heat pump systems from the primary energy consumption and ecological aspects, groundwater and shallow geothermal heat pump systems offer superior efficiency, compared to all market-available thermo-technical systems today. In the last decade, ground-source systems have seen some technological improvement by employing new borehole heat exchanger designs, such as piping with internal fins and a wider diameter (so called Turbocollector) to enhance the heat transfer between fluid and rock, as well as to reduce the pressure drop in the system. Furthermore, the process of drilling deeper offers higher ground temperatures and consequently higher seasonal performance factors in the heating cycle, due to the effect of the geothermal gradient. Nevertheless, although deeper boreholes provide better heat extraction rates per meter during the heat pump heating cycle, at the same time, it reduces heat rejection rates during the heat pump cooling cycle. The objective of this paper is to analyse and evaluate benefits and downsides of a new approach in the heat pump system design with deeper borehole heat exchangers of up to 300 m, comparing it to the traditional design of double-loop exchangers with 100 m depth. The geothermal borehole grid design simulation model, along with heat extraction and rejection, is performed on a yearly basis. The results are showing that the benefits of shallow geothermal boreholes, from the hydraulic and thermodynamic point of view, still dominate over deeper solutions.

1. Introduction

In 2019, 40% of the final energy consumption belonged to the heating sector in Europe [1]. One way to diversify the heating and cooling sector is to employ more heat pumps, as seen in Europe, with a total of installed units around 20 million in 2022 [2]. Ground-source heat pumps (GSHP) have obvious advantages, from the aspect of the coefficient of performance (COP), when compared to air-source heat pumps due to the fairly constant temperature of the ground [3]. Shallow geothermal energy exploitation, via borehole heat exchangers (BHE), is well underway in Europe, especially in countries like Sweden, Switzerland, and Germany [4].

Walch et al. [4] addressed the issue of dense BHE fields in urban areas and the consequent cooling of the subsurface. The case study was carried out in an urban area in Switzerland and it showed that to prevent subcooling of the ground, the cumulative BHE depth should be under 2 km/ha, for minimal 5 m BHE spacing and maximum 200 m of length per BHE.

Lee et al. [5] investigated the BHE field geometry and its influence on the thermal interference between the boreholes. Single, compact, and irregular-type BHE field arrays were proposed. The study showed that the compact geometry had a decreased heat transfer rate due to higher heat interference, especially with the increase in the number of BHEs. Irregular configurations had better results than the compact schematic, owing to the lesser thermal interference, and those with irregular geometry had higher heat transfer rates. Letizia Fascì et al. [6] proposed a new model to evaluate thermal interference in areas with dense, hydraulically independent BHE installations, suitable for conduction-dominated conditions. The model couples the existing FLS model with new boundary conditions seen as different thermal loads on different BHEs with a uniform temperature along each BHE. The result of the model is that while temperature is uniform along each borehole, it is different for each BHE, depending on the various thermal loads. Besides designing various BHE field geometries, the influence of the thermal interference on BHE performance can be decreased by drilling deeper thus decreasing the number of BHEs needed. Zhang et al. [7] studied the influence of increasing the BHE depth and fluid flow rate on thermal resistance of the single-U (1 U) BHE. The depth changed from 50 to 200 m. The analytical method for calculating mean fluid temperature over the entire BHE length was presented. The effect of the depth and volumetric flow rate on the relative deviation between borehole thermal resistance and effective borehole thermal resistance was calculated. The results showed that for a volumetric flow of 0.15 L/s, the borehole and effective borehole thermal resistance values increased from 4.2 to 29.7% with the increasing depth. Optimal volumetric flow rates were found for different BHE depths, where borehole thermal resistance almost matches effective borehole thermal resistance, with depth change. Also, the study showed that the effect of heat rate per meter of BHE on borehole thermal resistance is limited, however the volumetric flow rate had more effect on borehole resistance with higher thermal conductivity.

Kurevija et al. [8] studied the influence of different depth and geometrical settings of BHEs. They compared the single-U (1 U) and double-U (2 U) BHEs with smooth and ribbed inner pipes. The study showed that the deepest studied BHE, 1 U DN45 with a ribbed inner pipe and 150 m of depth, was the optimal design for the investigated field. This was due to a positive geothermal gradient, i.e., a higher initial BHE temperature. Also, due to the inner wall of the pipe being ribbed, the turbulent flow was maintained. Casasso and Sethi [9] conducted a sensitivity analysis of 1 U closed-loop BHEs to observe the impact of different parameters on GSHP efficiency. The study was on BHEs with varying depths from 50 to 100 m. The study showed that with an increased depth, the efficiency of the GSHP rises. However, it was also concluded that the optimal depth for each project has to also be determined from the techno-economical standpoint. Jun et al. [10] showed that with the increase in the fluid flow velocity, in the studied 1 U heat exchanger, there is a decrease in the total borehole thermal resistance, and an increase in the heat exchange rate.

Even though the deeper BHEs, up to 300 m, are becoming a standard in BHE installation, especially in countries where geoexchange systems are a standard in urban areas [11], there is a lack of the research involving the benefits, or drawbacks, of going deeper with different geometry settings of the BHEs. Rybach [12] observed that the customary boundary between shallow and deep geothermal exploitation is at 400–500 m, however, Piiponen et al. [13] noted that in Nordic countries, shallow geothermal systems are up to 300 m in depth. It was also noted that the term medium–deep BHE for the Nordic countries considers the range of 600–3000 m. In Croatia, drilling up to 200 m for BHE is considered as shallow, due to different geological conditions as well as commonly used drilling technology. The benefits and drawbacks of different BHE configurations in the range of 100–300 m depth are evaluated in this paper for a case study in Croatia.

2. Methods

Efficient shallow geothermal designing process is dependent upon numerous geological, thermodynamical, engineering, economical, and climate input data. In general, modelling of borehole heat exchanger systems is generally based on two principle analytical methods: the method based on Kelvin’s theory of linear heat transfer and the method based on the theory of cylindrical heat transfer. Based on these methods, numerous numerical models have been developed for the sizing and operation prediction of the borehole heat exchangers, depending on the thermogeological characteristics of the ground and annual extraction/rejection of heat energy. Each of these models that are applied for the purpose of sizing the borehole heat exchanger must be efficient in order to simulate the effects of variable input parameters over a long operational time. The use of an analytical solution for modelling a borehole exchanger is mathematically relatively simple and approximately correct, but the main problem is that the installed U loops of the exchanger in reality are not coaxial with the borehole itself, and a number of materials with different thermal properties are used in installation. Therefore, certain simplifications are made in the calculations. Hence, the most important is the assumption of equivalent diameter, whereby the two loops of the U exchanger are considered as one tube coaxial with the well, which is the basic foundation of the analytical solution of cylindrical heat transfer. The geometric profile of the borehole can be further simplified and considered as an infinitely dimensionless long line as a source of heat, which is the basis of the line source theory. Each of these analytical solutions is based on the Fourier law of heat conduction and is dependent on the experimental measurement of the thermal conductivity of the ground using the thermal response test (TRT).

In the case of applying borehole heat exchanger in wells at greater depths than common sizing practices in Europe (depths exceeding approximately 150 m), special attention should be made to hydraulics of the systems and optimization of borehole thermal skin/resistances. Since borehole extraction heat power grows with depth, differences in leaving source (LST) and entering source temperature (EST) of the heat exchanger should maintain at maximum Δ5 °C for a heat pump to operate optimally. In commonly developed commercial software that are used to size shallow geothermal systems (like GLD—Ground Loop Design; EED—Earth Energy Designer; GLHEPro; etc.), efficiency of design depends on obtaining accurate data about thermogeological properties of the ground (mean thermal conductivity, initial static temperature, and skin/resistance) with in situ measurements (TRT), as well as reliable building heating/cooling loads for an entire year. For shallow geothermal system to be considered efficiently sized, over multiple decades, operational entering source fluid temperature (EST) should not fall below 0 °C, even at peak heating load conditions. Hence, minimum temperature criteria that can appear at evaporator side of the heat pump in system analysis should not drop below −5 °C during peak loads. Most heat pump manufacturers state that an allowed technical temperature minimum is −10 °C, but from the system modelling point of view, this is unacceptable.

Therefore, for a deeper borehole heat exchanger flow should be adjusted to maintain Δ5 °C temperature difference at peak conditions. This would obviously raise a pressure drop at heat exchanger and include implementation of higher power circulating pumps, which could negatively influence a system’s seasonal coefficient of performance (SCOP). After screening representative European borehole heat exchanger manufacturer products, it could be seen that up to 300 m in length exchangers can be ordered for dimensions of double-loop (2 U) D40, single-loop (1 U) D45, and 1 U D50. These three exchangers are subject of heat extraction simulation in this paper’s analysis. Another important issue in hydraulic analysis of the BHE extraction efficiency is the implementation of glycol–water mixtures as antifreeze working fluid.

In colder regions, installations require the use of water and certain glycols mixture to lower fluid-freezing temperature in the system, of which the most often used are propylene–glycol and ethylene–glycol mixtures. Ethylene-based mixtures generally have better rheological properties, but propylene–glycol is more environmentally friendly due to significantly lower toxicity, which is very important in groundwater area. If borehole fluid is oversaturated with glycol (for example 30% vol.), as it is often conducted in practical simple installations to avoid design risks and freezing of the fluid, during the peak consumption in the heating cycle of heat pump (subcooling of the ground), there is a switch from turbulent to laminar flow regime in pipes. This effect leads to additional rise in thermal skin/resistance of the system as a result of a working liquid coating formation on the pipe walls. Such design causes hydraulic problems for fluid flow, as to achieve turbulent flow regime, higher flow rates are required. If there is laminar flow regime inside the pipe, heat transfer gets worse as thermal resistance grows.

Furthermore, as source temperature drops near 0 °C during peak load conditions, or even below, there is an exponential growth in dynamic viscosity of the water–glycol mixtures. Higher viscosity in the system has to be avoided at all costs since it leads to significantly higher pressure drop in PE pipes, appearance of laminar flow, higher thermal skin/resistance to heat transfer, and at the end, lower extraction rates from the borehole heat exchanger. This negative phenomenon can lead to complete malfunction of the system and activation of protection alarms in the heat pump. This problem is especially seen in domestic installations where smaller size heat pumps usually have factory pre-installed circulation pumps, whereas efficient flow cannot be achieved anymore due to higher pressure drops and high viscosity in the system.

To demonstrate this effect, entire pressure drop calculations for three types of heat exchanger were carried out using SF Pressure Drop 8.0 software. Equivalent borehole thermal skin/resistances were obtained with software Earth Energy Designer v3.22 (EED) for three borehole options; 150 mm drilling diameter with 2 U D40, 1 U D45, and 1 U D50 probe installation.

Simulation model was conducted for borehole heat exchangers completed on site with standard bentonite–cement mixture and physical properties of glycol for different volume share at temperature of °C calculated using software SecCool Properties v1.33 and ASHRAE standardized data.

3. Case Study Project Data and Thermogeological Environment

The analysed location is situated in northwestern Croatia, in the City of Zabok. Zabok is located within the Hrvatsko zagorje County, which is bordered with Samoborsko gorje, Medvednica, Kalnik, Ivanščica, and Macelj Mountains. The majority of the Hrvatsko zagorje comprises mostly Negoene seidments. The area of the City of Zabok consists of alluvial and Pleistocene sediments as seen in Figure 1. The Pleistocene deposits consist of sand and gravel with sparse deposits of clay and sandy marl. The alluvial sediments, deposited by the rivers Krapina and Krapinčica, comprises gravel, sands, and clays in various thicknesses.

From the hydrogeological standpoint, the Krapina catchment area is defined by deposits of low-to-very low permeability [16]. With morphological characteristics of the area, this results in surface runoff and poor infiltration of the precipitations. The numerous streams go through occasional ephemeral flooding. The main aquifer locations are in fractured limestones and dolomites. The springs in the Krapina catchment area can give from 1 to 70 L/s, with highest values occurring around the Ivanščica mountain banks. Drilled water wells usually give less than 5 L/s of groundwater. The groundwater in the analysed location was encountered at the depth of around 110 m, within a thin layer of sand deposits. However, the layer is too thin to be of significance to the advective effect on the BHE performance. During hydrocarbon deposits research in the last century, a deep exploration well was drilled in the area of the Zabok city, within 1 km of analysed location, with a depth of 2044 m. The geological monitoring was carried out every 5 m. The section between 0 and 600 m is characterized as Rhomboidea beds. The roof of the formation was characterized by plastic, grey-greenish clays, sandy clays, clayey sands, and thin layers of differently coloured gravels. The bottom of the sedimentary stratum is defined by grey clayey marls and grey sandy marls. Between the roof and the bottom of the sedimentary stratum, deposits are characterized as micaceous sands, clays, and sandy clays of different thickness. The geothermal gradient was calculated according to the novel geothermal gradient map [16] from the data obtained during exploration well drilling in Croatia and was determined at 0.036 °C/m for the investigated area.

The BHE field was drilled within the city area, in order to supply thermal energy for a building that is going through reconstruction, as well as two new buildings. The reconstructed building was formerly part of a bigger industrial complex for a textile company. The abandoned building was donated and will be transferred to a modern centre of urban culture “reGENERATOR”. It will be used as a cultural, creative, and civil association centre for the Hrvatsko zagorje area. During drilling operations, geological observations were made. They confirmed the previously described deposits. In Figure 2, the initial design of the borehole field can be observed. The geoexchange system comprised 24 boreholes, each 130 m in depth, with a total depth for heat exchange of 3120 m. The drilled pattern was a 8 × 3 well grid with 7 m separation distances between adjacent boreholes. Inside the boreholes, D45 single-loop heat exchangers were installed.

Building energy data are shown in Figure 3, as monthly heating and cooling loads. Total annual heating energy required for building is projected to be 223 MWh, while cooling energy is estimated at 118 MWh. The peak power of the geothermal system is at 200 kW_t for heating loads and 195 kW_t for cooling loads.

4. Results

4.1. Results of Performing the Thermal Response Test (TRT) as the Project Groundwork

The thermal response test (TRT) is performed on a principle of heat rejection into the ground to determine the basic thermo-geological parameters. The basic interpretation of the TR test is based on the observation of temperature responses as a function of time and constant thermal power. The method was developed on the basis of flow tests during hydrodynamic testing of the oil and geothermal wells. The analogy of the method follows from the similarity of Darcy’s and Fourier’s laws that describe the flow of non-compressible fluid in the porous medium, or the conduction of heat through the non-permeable medium, using the same principle as the differential diffusion equation for the infinite reservoir. The on-site test procedure is governed by the guidelines of the International Ground Source Heat Pump Association (IGSHPA), which implies a minimum of 48 h of testing time to satisfy accuracy. Interpretation of TRT results gives an accurate insight into main thermodynamic parameters of heterogenous rock, such as thermal conductivity and extraction/rejection rates, as well as borehole thermal resistance. Different approaches in interpretation procedure are well known and described in various research papers [17,18,19].

The idealized TR test with constant thermal power in the infinite medium can describe the temperature change at a given radius using the solution of the function of the exponential integral from the diffusion equation:

$$T(r,t)=T_i-\frac{q^{\prime }}{4\pi \lambda }\left(Ei\left(-\frac{r^2}{4\alpha t}\right)\right)$$

(1)

By replacing the exponential integral with the function of the natural logarithm for the value x < 0.01, the temperature change at the radius r is obtained:

$$T\left(r,t\right)=T_i+\frac{q^{\prime }}{4\pi \lambda }\ln \left(\frac{e^{\gamma }{r}^2}{4\alpha t}\right)=T_i+\frac{q^{\prime }}{4\pi \lambda }\left(\ln \frac{\alpha t}{r^2}-0.80907\right)$$

(2)

However, in real tests, it is necessary to include the size of damage of the grout zone, which is referred to as a skin factor in hydrodynamics and in the indirect exploitation of shallow geothermal energy it appears in the form of equivalent thermal resistance to the heat flow from the ground to the fluid in the pipes. Grouted space between the outer wall of the built-in plastic pipes and the borehole wall is a thermal resistance, due to the thermal conductivity of the mixture, which is often lower than the thermal conductivity of the ground (in the case of energy piles, the grout is a classic concrete mix). Also, the thermal resistance between the fluid and the ground also contributes to the low thermal conductivity of the polyethylene material through which the fluid flows. The thermal resistance value can also be expressed through the temperature difference of the working fluid and the temperature on the borehole outer wall, in a continuous mode of operation.

$$s=\frac{1}{2}\ln \left(\frac{e^{\gamma }{r}_b^2}{4\alpha t}\right)-\frac{\left(T-T_i\right)2\pi \lambda }{q^{\prime }}$$

(3)

$$\Delta T_{skin}=s\left(\frac{q^{\prime }}{2\pi \lambda }\right)$$

(4)

$$R_b=\frac{\Delta T_{skin}}{q^{\prime }}$$

(5)

$$T=T_i+\frac{q^{\prime }}{4\pi \lambda }\left(\ln \left(\frac{e^{\gamma }{r}_b^2}{4\alpha t}\right)-2s\right)=\frac{q^{\prime }}{4\pi \lambda }\left(\ln \left(\frac{\alpha t}{r_b{}^2}\right)-0.80907\right)+q^{\prime }\cdot R_b+T_0$$

(6)

For the determination of thermal conductivity, it is crucial to first determine the period of time after which the semi-steady-state heat transfer begins, i.e., the initial duration of the unsteady-state period. The usual method of determining the beginning of this semi-steady-state period, in which the thermal conductivity of the ground is defined, is by means of a formula that includes the value of thermal diffusion of the soil, which is usually assumed on the soil composition obtained by drilling.

$$t>\frac{5r^2}{\alpha }$$

(7)

In geological environments of marly clays, like in the discussed project, this usually corresponds to a time of 12–24 h.

On site (Figure 4), a TR test was performed for the duration of almost six days. The test was realized as a specialized thermal response test with multiple heat steps involved, as a novel method developed and elaborated in detail as part of the earlier research [18,19,20,21]. In short, the first heat step condition is performed until the temperature stabilization is nearly reached, the so-called steady state. After stabilization of the temperature, the heat step is changed until the stabilized temperature is again reached. The procedure is repeated three to four times with the recommendation that the maximum heat requirement is two to three times higher than the minimum. In the low conductivity ground, for the high-heat step, it is often necessary to have an extremely long time to reach the steady state, so a close point of the semi-steady is often used as the approximation of the steady state. Also, to decrease the test time it is advisable to use a reversible way of performing the step test where the heat step decreases from the highest to the lowest. The test is used to precisely determine the ground heat exchanger yield in W/m for the default inlet and outlet temperature from the borehole exchanger (EST—entering source temperature; LST—leaving source temperature). The application of this step test in applied thermogeology can contribute to the understanding of the behaviour of borehole fluid where the geothermal system works with a peak load for a longer period of time. Such an analysis is key to understanding minimum temperatures at the borehole field without any further significant sub-cooling in the function of time.

At the first drilled borehole on the site, a single-loop exchanger was inserted in the BHE-1 well (classification of the pipe: TC45 turbocollector PE100 SDR11 PN16 RC). According to the empirical knowledge about the thermal properties of the drilled material (clay and marl) and the borehole depth of 130 m, the TR machine heater power of 2.5 + 1.5 kW@240 V was used. The test started with a circulation of 0.61 L/s, without turning on the heaters to determine the mean effective soil temperature along the length of the borehole. The circulation lasted for twenty minutes and the effective static soil temperature of 13.9 °C was recorded. After that, electric heaters with an average power of 3.87 kW were turned on, with monitoring of voltage and electric current to determine the thermal conductivity coefficient in this first heat-step interval. This period was tested for a duration of 97.2 h without significant voltage and current fluctuations during the day/night period. Temperature at the outlet from the well (EST) was 22.3 °C, which is 8.4 °C higher than the static initial rock temperature of 13.9 °C. In a heating cycle with a heat pump (inverse curve in the diagram), such an identical temperature difference would mean the temperature of the circulating medium; outlet (LST) from BHE-1 would be +5.5 °C. During this condition, 345 kWh of thermal energy was rejected into the rock.

In the second heat step of TRT, the heat power was lowered to observe the cooling of the well after the first condition. This approach is used to determine the point of outlet temperature stabilization and thermal energy transfer steady state, which will be some of the input data to determine the possible extraction capacity of the exchanger. The heat step was set at 2.54 kW and lasted for 42 h and after that time, the temperature drop stabilized to an almost constant value. The entering source temperature from BHE-1 was 20.1 °C, which was 6.2 °C higher than the static initial rock temperature. In an inverse curve, like for the case when the ground is similarly cooled, such an identical temperature difference would mean the temperature of the circulating medium; EST from the BHE-1 = +7.7 °C. During this condition, 83.5 kWh were stored in the soil.

In last heat step condition, the thermal power to the borehole exchanger is additionally reduced in order to observe the cooling of the well and to determine the third point of stabilization. The power of the electric heater was 1.6 kW and this period lasted 48 h, and after that time, the temperature drop stabilized to an almost constant value EST = +18.4 °C, which was 4.5 °C higher than the static initial rock temperature. In the inverse curve, such an identical temperature difference would mean the EST = +9.4 °C. During this condition, 37.9 kWh was stored in the soil.

The entire TR test procedure and temperature evolution during all heat steps are shown in Figure 5.

In order to determine the coefficient of thermal conductivity of rock, it is necessary to plot the data of the mean temperature of the circulating fluid from the first condition as a function of the natural logarithm of time ln(t), as seen on Figure 6. After the heater is switched on, the mean temperature in the borehole exchanger starts to rise as a function of the thermal conductivity of the rock, and in this area of the diagram, it is necessary to determine the period of time when the temperature rise becomes linear as a function of the logarithm of time but after a fixed time of 24 h, as a period is necessary to completely stabilize the heat transfer, i.e., to overcome the thermal resistances of the well and the near-well zone. The period from which the coefficient of thermal conductivity was determined is 24–96 h and represents the end of the first period where the average thermal power is 3875 W.

The slope of the interpolated trendline from Figure 6 is 1.3673, the depth of the well is 130 m, and the average power of the heater is 3.876 kW, therefore using the equation of Eklof and Gehlin [22], the thermal conductivity coefficient of the marly clay environment is 1.73 W/m K.

$${\lambda }_{}=\frac{Q}{4\pi H\kappa }$$

(8)

With the calculated coefficient of thermal conductivity and knowledge of the geometric characteristics of the borehole (length and diameter), the calculation of the thermal resistance of the borehole was performed for 5 min time step intervals. An integral part of this parameter is the thermal resistance to the transfer of thermal energy from the working fluid to the plastic wall of the pipe (convection heat transfer), thermal energy transfer through the wall of the PE pipe (conduction), and the thermal energy transfer from the wall of the PE pipe to the cement filling and further through to the surrounding rocks (conduction). Using known Equations (3)–(5) for calculation of skin factor and thermal resistance, the mean value of 0.105 m K/W was obtained, as seen from Figure 7.

Since IGSHPA guidelines dictate that at least 48 h of TR test should be recorded so the test would be representative, Figure 8 shows the analysis of the thermal conductivity as a function of elapsed time. It is obvious that the results clearly become more unified and representative after 72 h of interpretation. For example, if the same test would have been performed for any given time under 72 h, the results would have a higher degree of uncertainty.

To determine maximum extraction/rejection heat rates that can be obtained by borehole heat exchanger as a function of chosen constant entering source temperature (EST), it is necessary to set the steady-state temperature in each of the test conditions as a separate point. As can be seen from Figure 9, it is possible to construct the extraction/rejection yield diagram (W/m) as a function of the desired outlet temperature (EST). In the first heat step condition, temperature after 96 h of the TR test is chosen as the first point of stabilization, although no actual steady-state heat transfer has been achieved (minor error is accepted since the BHE rarely operates under a maximum load for such a long period of time). The other two points of stabilization correspond to reached temperature at the end of second and third period.

If a temperature of 0 °C is set in the heating cycle as the minimum permissible on the outlet of the exchanger during long-term operation, with a constant heat load applied, Figure 9 presents the method to establish the heat exchanger yield. With the graphically obtained equation of the steady-state heat transfer (y = −0.3207x + 13.9), estimation of the heat exchanger steady-state extraction rate is 43.3 W/m, or 5.63 kW for this 130 m deep BHE.

4.2. Sensitivity Analysis for Various BHE Depths

On the basis of a technical solution for the Zabok reGENERATOR project, as well as with results of conducted thermal response test measurements, a sensitivity analysis was performed for various borehole depths. For three borehole heat exchanger solutions, as seen from

Table 1. Hydraulic calculated data for 2 U D40 SDR11 PEHD borehole heat exchanger.

Propylene Glycol Vol. Conc., %	15.0	Depth, m	Re	pD, Bar	ΔT, °C	Extraction, kW	Flow, l/s Both Pipes	Flow, l/s per Pipe
Pipe sdr11	2U D40/32.7 mm	100	2360	0.083	3.3	5.2	0.400	0.200
Specific heat, kJ/kg °C	3.986	150	2573	0.144	5.0	8.7	0.437	0.218
Density, kg/m3	1019.8	200	3847	0.380	5.0	13.0	0.652	0.326
Viscosity, mPas	3.37	250	5110	0.771	5.0	17.3	0.866	0.433
Freezing point, °C	−5.3	300	6573	1.427	5.0	22.2	1.114	0.557
Propylene glycol vol. conc., %	20.0	Depth, m	Re	pD, bar	ΔT, °C	Extraction, kW	Flow, l/s both pipes	Flow, l/s per pipe
Pipe sdr11	2U D40/32.7 mm	100	2367	0.120	2.8	5.2	0.480	0.240
Specific heat, kJ/kg °C	3.929	150	2367	0.180	4.6	8.7	0.480	0.240
Density, kg/m3	1025.8	200	3264	0.414	5.0	13.0	0.662	0.331
Viscosity, mPas	4.05	250	4330	0.836	5.0	17.3	0.878	0.439
Freezing point, °C	−7.3	300	5572	1.546	5.0	22.2	1.130	0.565
Propylene glycol vol. conc., %	25.0	Depth, m	Re	pD, bar	ΔT, °C	Extraction, kW	Flow, l/s both pipes	Flow, l/s per pipe
Pipe sdr11	2U D40/32.7 mm	100	2345	0.222	2.1	5.2	0.650	0.325
Specific heat, kJ/kg °C	3.861	150	2345	0.342	3.5	8.7	0.650	0.325
Density, kg/m3	1031.0	200	2431	0.472	5.0	13.0	0.673	0.337
Viscosity, mPas	5.57	250	3225	0.951	5.0	17.3	0.894	0.447
Freezing point, °C	−10.0	300	4148	1.753	5.0	22.2	1.150	0.575

,

Table 2. Hydraulic calculated data for 1 U D45 SDR11 PEHD borehole heat exchanger.

Propylene Glycol Vol. Conc., %	15.0	Depth, m	Re	pD, Bar	ΔT, °C	Extraction, kW	Flow, l/s per Pipe
Pipe sdr11	1U D45/36.8 mm	100	2726	0.074	5.0	5.2	0.261
Specific heat, kJ/kg °C	3.986	150	4582	0.269	5.0	8.7	0.437
Density, kg/m3	1019.8	200	6837	0.714	5.0	13.0	0.652
Viscosity, mPas	3.37	250	9081	1.458	5.0	17.3	0.866
Freezing point, °C	−5.3	300	11,681	2.710	5.0	22.2	1.114
Propylene glycol vol. conc., %	20.0	Depth, m	Re	pD, bar	ΔT, °C	Extraction, kW	Flow, l/s per pipe
Pipe sdr11	1U D45/36.8 mm	100	2322	0.082	5.0	5.2	0.265
Specific heat, kJ/kg °C	3.929	150	3882	0.292	5.0	8.7	0.443
Density, kg/m3	1025.8	200	5802	0.775	5.0	13.0	0.662
Viscosity, mPas	4.05	250	7695	1.576	5.0	17.3	0.878
Freezing point, °C	−7.3	300	9903	2.928	5.0	22.2	1.130
Propylene glycol vol. conc., %	25.0	Depth, m	Re	pD, bar	ΔT, °C	Extraction, kW	Flow, l/s per pipe
Pipe sdr11	1U D45/36.8 mm	100	2372	0.159	3.6	5.2	0.370
Specific heat, kJ/kg °C	3.861	150	2891	0.333	5.0	8.7	0.451
Density, kg/m3	1031.0	200	4314	0.877	5.0	13.0	0.673
Viscosity, mPas	5.57	250	5724	1.779	5.0	17.3	0.893
Freezing point, °C	−10.0	300	7372	3.300	5.0	22.2	1.150

and

Table 3. Hydraulic calculated data for 1 U D50 SDR11 PEHD borehole heat exchanger.

Propylene Glycol Vol. Conc., %	15.0	Depth, m	Re	pD, Bar	ΔT, °C	Extraction, kW	Flow, l/s per Pipe
Pipe sdr11	1U TC50/40.9 mm	100	2462	0.059	5.0	5.2	0.261
Specific heat, kJ/kg °C	3.986	150	4123	0.164	5.0	8.7	0.437
Density, kg/m3	1019.8	200	6152	0.434	5.0	13.0	0.652
Viscosity, mPas	3.37	250	8171	0.884	5.0	17.3	0.866
Freezing point, °C	−5.3	300	10,510	1.640	5.0	22.2	1.114
Propylene glycol vol. conc., %	20.0	Depth, m	Re	pD, bar	ΔT, °C	Extraction, kW	Flow, l/s per pipe
Pipe sdr11	1U TC50/40.9 mm	100	2366	0.061	4.4	5.2	0.300
Specific heat, kJ/kg °C	3.929	150	3493	0.178	5.0	8.7	0.443
Density, kg/m3	1025.8	200	5220	0.471	5.0	13.0	0.662
Viscosity, mPas	4.05	250	6876	0.945	5.0	17.3	0.872
Freezing point, °C	−7.3	300	8910	1.778	5.0	22.2	1.130
Propylene glycol vol. conc., %	25.0	Depth, m	Re	pD, bar	ΔT, °C	Extraction, kW	Flow, l/s per pipe
Pipe sdr11	1U TC50/40.9 mm	100	2365	0.115	3.3	5.2	0.410
Specific heat, kJ/kg °C	3.861	150	2601	0.203	5.0	8.7	0.451
Density, kg/m3	1031.0	200	3882	0.533	5.0	13.0	0.673
Viscosity, mPas	5.57	250	5151	1.081	5.0	17.3	0.893
Freezing point, °C	−10.0	300	6633	2.003	5.0	22.2	1.150

, complete hydraulic calculations have been made, which are necessary for a 30-year analysis of the borehole field operation. Three subvariants were taken into account, in the form of different glycol mixture volumes; 15, 20, and 25% volume concentrations. For a given extraction rate for each borehole depth variant, the flow was adjusted according to the specific heat of working fluid and maximum allowed temperature difference at the wellhead of 5 degrees. According to the calculated adjusted necessary flow, a pressure drop was calculated with a criteria of at least turbulent-transition flow regime (minimum Reynolds number = 2300). From the analysis of the three available different heat exchangers, a pressure drop for the deepest solution of 300 m was quite significant, especially for high viscous mixture options. This is due to the limitation in the available diameter piping on the market, as well as the necessity to achieve at least 5 degrees temperature difference at the borehole wellhead.

5. Discussion

This project of retrofitting a cultural building was completed with a total of 3120 m drilled, in the form of 24 boreholes of 130 m depth each. The separation distance between adjacent boreholes is 7 m, which indicates that the certain thermal interference between boreholes would be present after multi-decade operation. Also, since there is an imbalance in extracted heat during winter months and rejected heat into the ground during the summer months, some subcooling of the ground would occur, as well as a slight fall in the seasonal performance factor. This is in fact expected for climate areas where heating is dominant over cooling needs. The simulation software takes this effect into account over a 30-year period and adds up the length of the total geothermal grid to compensate for the subcooling effect. The principle of the exchange model is therefore for the lock peak load temperature after 30 years of operation not to surpass 0 degrees.

Therefore, to identify how much borehole depth is playing a role on the heat pump system seasonal coefficient of performance, detailed modelling was conducted using EED v3 and GLD2009 software. Operation time of the shallow geothermal system was set at 240 months or 20 years of operation. Looking at the modelling results from Figure 10, it can be seen that three different temperature evolutions are monitored; (1) source temperature during short peak heating loads in winter months as a blue curve; (2) source temperature during short peak cooling loads in summer months as a red curve; and (3) black curves representing mean source temperature for each month. Examining results from variants A to E (100 m to 300 m of depth, respectively), all variants show subcooling of the ground with a small but persistent lowering of the source temperature for each new heating/cooling season over 20 years. The simulation was carried out on a principle to keep constant heating and cooling loads but to reduce the number of boreholes with each new deeper variant keeping the criterion that the heating peak load source temperature after 20 years does not fall below 0 °C.

After that, a simulation was conducted for 150, 200, 250, and 300 m depths taking into account the geothermal gradient and temperature rise with depth. Since extraction rates are raising because of temperature increases with depth, for each case, the total number of boreholes was reduced until approximately the peak temperature in heating was achieved as a model for 100 m depth boreholes. As can be seen from Figure 11, the model would give the same peak heating conditions after 30 years for the initial case of 30 boreholes with depths of 100 m (total 3000 m of drilling), as it would for just 7 boreholes with 300 m in depth (2100 m in total). Therefore, for the same extraction scenarios in the heating cycle, for 300 m borehole depths, there would be 900 m less drilling involved in total, and 23 less boreholes in the field.

The positive side of this effect is less surface expenses for the gathering system, and a smaller area required for the geoexchange system. On the other hand, the downsides are a degraded cooling efficiency due to higher rock temperature and an inability to reject the same amount of heat to the ground for the same conditions as the shallow model. Also, drilling expenses are expected to rise significantly after 150 m because mud circulation would be required, instead of air compression, to bring drilling bits to the surface. On some occasions, this expense can even double, as opposed to the shallow model. Furthermore, the pressure drop is rising enormously with depth for all commercially available heat exchangers, which leads to significantly higher operating costs for a circulating pump. This effect can lower the overall SCOP of the heat pumps. As seen from Figure 11, if heating and cooling SCOP are added together, then it is obvious that going deeper does not offer a more efficient system, due to a cooling efficiency degradation.

6. Conclusions

An analysis of different borehole heat exchanger depths was carried out for a current developed project in Croatia (building expected to be finished and ready to use in fall/winter 2023). This projected served as a base case scenario to make a sensitivity analysis of different solutions that could be applied during the project design phase. Numerous simulations were carried out for a 30-year project with a heat extraction cycle from the ground and a heat rejection cycle into the ground. Results clearly shows that due to the temperature increase with depth (according to the location geothermal gradient), heat extraction is rising for deeper solutions, but at the same time, the heat rejection during the cooling cycle is degrading. The SCOP analysis shows that there is no cumulative gain with deeper boreholes applied, as opposed to standard solutions with 100 m. Further research is required in the field of techo-economics, as although a significantly smaller amount of boreholes is required for the same heating and cooling loads for deeper solutions, there is an unexpected rise in drilling costs per meter for depths greater than 150 m because of more complex technology required. Also, due to enormously high pressure drops (even for wider D50 loops) at 300 m of depth, there is a very high rise in operational costs to sustain the flow of the entire geothermal field.