GeoThermag: A geothermal magnetic refrigerator

GeoThermag: A geothermal magnetic refrigerator

Accepted Manuscript Title: GeoThermag: a geothermal magnetic refrigerator Author: Ciro Aprea, Adriana Greco, Angelo Maiorino PII: DOI: Reference: S01...

1MB Sizes 0 Downloads 0 Views

Accepted Manuscript Title: GeoThermag: a geothermal magnetic refrigerator Author: Ciro Aprea, Adriana Greco, Angelo Maiorino PII: DOI: Reference:

S0140-7007(15)00221-2 http://dx.doi.org/doi:10.1016/j.ijrefrig.2015.07.014 JIJR 3102

To appear in:

International Journal of Refrigeration

Received date: Revised date: Accepted date:

4-5-2015 8-7-2015 14-7-2015

Please cite this article as: Ciro Aprea, Adriana Greco, Angelo Maiorino, GeoThermag: a geothermal magnetic refrigerator, International Journal of Refrigeration (2015), http://dx.doi.org/doi:10.1016/j.ijrefrig.2015.07.014. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

GeoThermag: A geothermal magnetic refrigerator Ciro Apreaa,1, Adriana Grecob, Angelo Maiorinoa,2,* a

Department of Industrial Engineering, University of Salerno, Via Giovanni Paolo II 132, 84084, Fisciano (SA), Italy b Department of Industrial Engineering, University of Naples Federico II, P.le Tecchio 80, 80125, Napoli, Italy

HIGHLIGHTS     

The new concept of GeoThermag has been introduced A test apparatus has been described Gadolinium spheres have been used as refrigerant Experimental results under various conditions have been performed The maximum cooling capacity, the temperature span and the COP have been measured

GRAPHICAL ABSTRACT

Abstract 1

Page 1 of 15

Magnetic refrigeration has played an important role in the scientific landscape in recent years and could potentially be an alternative to vapour compression technology. Although new prototypes have recently been developed and new material refrigerants have been tested, there is still no equipment that can be used in the refrigeration industry. In the present work, we demonstrate how magnetic refrigeration can be applied to climate system environments by combining the technology of magnetic refrigeration with that of low temperature geothermal energy. Thus, we introduce the concept of GeoThermag. After describing the GeoThermag system in detail, we report experimental results that demonstrate the validity of this application. These experimental tests were obtained by using a magnetic refrigerator that was connected to a geothermal probe. Using 1.20 kg of gadolinium, we found that the GeoThermag system configured in this manner is capable of providing cold water to feed a radiating panel and to develop a cooling capacity of 190 W with a COP of 2.20. Keywords: magnetic refrigeration; renewable energy; geothermal; experimental; permanent magnet; regenerator

_____________ * Corresponding

author: Tel. +39 (0) 89 964002; fax +39 089 964037; E-mail: [email protected] (A. Maiorino). 1 Member of IIR-IIF Commission E2. 2

Junior Member of IIR-IIF.

Nomenclature Acronyms AMR

active magnetic regenerator

COP

coefficient of performance

GSHP

ground source heat pump

TRT

thermal response test

Symbols 2

Page 2 of 15

thermal diffusivity

[m2 s-1]

∆Tspan

temperature span

[K]

fAMR

cycle frequency

[Hz]

coefficient defined in Eq. 3

[K]

Euler's constant

[-]

thermal conductivity of the ground

[W m -1 K -1]

mrf

mass of regenerating fluid

[g]

p

coefficient defined in Eq. 4

[K]

Pin

pressure at the inlet of the pump

[Pa]

Pout

pressure at the outlet of the pump

[Pa]

linear thermal power

[Wm-1]

maximum thermal power exchanged with the ground

[W]

cooling capacity

[W]

radius of the hole

[m]

equivalent thermal resistance of the couple probe-filling material

[W m-1 K-1]

temperature of the fluid measured during the TRT

[K]

TH

temperature of heat rejection

[K]

THFi

temperature of the water outgoing from the geothermal probe

[K]

THFo

temperature of the regenerating fluid outgoing from the hot side of the refrigerator

TLFo

temperature of the regenerating fluid outgoing from the cold side of the refrigerator [K]

rb

temperature of the undisturbed ground

[K]

tm

magnetization or demagnetization period

[s]

tf

fluid flow period

[s]

Utilization factor

[-]

volumetric flow rate

[l min-1]

electrical power absorbed by the electric heater

[W]

electrical power absorbed by the pump

[W]

electrical power absorbed by the electric motor

[W]

angular velocity of the magnets

[rpm]

Vfr

Ω

[K]

1. Introduction In recent years, the refrigeration industry has undergone major renewal processes that are driven by the need to reduce the environmental impact of current refrigerators as much as 3

Page 3 of 15

possible. In particular, attention has been directed more towards fluorinated refrigerants that are widely used in the commercial refrigeration sector; these refrigerants are greenhouse gases. Severe restrictions have been introduced by international regulations. For example, the new European F-Gas Regulation that was approved in March 2014 has placed significant limitations on the use of existing fluorinated refrigerants. Since January 1, 2015, domestic refrigerators have been forbidden to use refrigerants with a GWP greater than 150. This limitation will become effective for other applications starting in 2020 (European Commission, 2014). This scenario implies that most of the refrigerants that are currently in use will be banned; therefore, a challenge has been presented for identifying refrigerants that can satisfy the required limits. An answer to the problem has been advanced for some years by various researchers who have proposed the adoption of natural fluids, such as propane, isobutene and carbon dioxide. Although these fluids are equipped with a small direct impact, they have limitations. Specifically, because of their flammability, hydrocarbons can only be used in equipment that needs a few grams of fluid and can only be used in countries that permit their use (Balsan et al., 2011), confining this technical solution to a niche in the cold industry. Although the use of carbon dioxide has no direct problems because it is a non-flammable fluid and has a GWP of one, its application remains limited as a result of the high working pressures and the obtainable reduced energy performance (Llopis et al., 2015; Aprea et al., 2015). Additionally, in recent years, a growing number of researchers have focused on a new technology known as magnetic refrigeration. This technology was particularly interesting because its application is based on solid refrigerants that are considered to be environmentally friendly and are therefore not subject to any limitations as a result of current international regulations. As demonstrated by several scientific papers that have been published to date (Kitanovski et al., 2015; Lozano et al., 2013; Lozano et al., 2014; Tura and Rowe, 2011; Yu et al., 2010;), magnetic refrigeration has reached a good level of maturity. However, the magnetic refrigerators that have been built to date, together with the refrigerants used, have shown some technical limitations that demonstrate that magnetic refrigeration remains at the experimental level. Considering the data presented in the literature, it appears that current magnetic refrigerators are capable of ensuring low values of ΔTspan, which is defined as the temperature difference that the refrigerating machine is able to establish between the cold and hot source. In addition, the ΔTspan of a magnetic refrigerator tends to significantly decrease with increasing thermal load. According to the latest experimental results (Jacobs et al., 2014; Engelbrecht et. al., 2012; Ericksen et al., 2014) in room temperature applications, the maximum ΔTspan settles down to approximately 19 K when the thermal load is close to 200 W. If one considers the most common refrigeration applications, such as domestic refrigerators or air conditioning systems, for which the required ΔTspan could be 50 K and for which the cooling capacities range from a few tens of watts to thousands of watts, it may be difficult to consider the existing magnetic refrigerator prototypes as possible competitors for vapour compression systems. However, in our article, we propose a technical solution that allows the use of magnetic refrigeration as a viable alternative 4

Page 4 of 15

to vapour compression systems for some applications. We have named this solution GeoThermag. GeoThermag was designed to be the first air conditioning system that is free of F-gases; at the same time, GeoThermag was designed to be able to exploit a renewable energy source. From the experimental tests conducted, we obtained proof of the applicability of the GeoThermag technology, and in this article, we report the experimental results in terms of ΔTspan, cooling capacity and COP. 2. The concept of GeoThermag It is well known that a refrigerating machine used for air conditioning has to operate with a high ΔTspan (approximately 45K) because heat rejection generally occurs by means of the external air. If heat rejection takes place by means of water that exchanges heat with the ground and is composed of systems that are generally known in the literature as ground source heat-pumps (GSHP), the ΔTspan can be appreciably reduced. Consequently, as suggested by Omer (2008), using a renewable energy source, such as geothermal energy at low temperature, a GSHP can reach COP values that are far superior to those of conventional air-to-air systems. Nevertheless, subject to reduced ΔTspan and to variable thermal load, vapour compression systems that are equipped with the most common expansion device at a fixed point are not technically convenient; thus, expensive technical solutions (i.e. expansion valve) are an easier technical alternative, but they can reduce some of the benefits provided by the geothermal source. However, if it is possible to significantly reduce CO2 emissions remains the problem that is associated with the use of harmful substances, such as the F-Gas, for which the use of geothermal energy, together with vapour compression technology, would not allow overcoming the restrictions imposed by recent international regulations. With this in mind, we combined magnetic refrigeration technology with that of low temperature geothermal energy, thus introducing the concept of GeoThermag. Fig. 1 shows a basic scheme that helps in understanding what we are introducing. Assuming the need to cool an environment (e.g., a room), it is possible to consider an AMR-type magnetic refrigerator in which the regenerating fluid flows into a heat exchanger that is embedded in the ground (ground heat exchanger) to reject the heat drawn in the cold exchanger, as represented by a fan coil or by a radiant floor. To make the most of the characteristics of the geothermal source, it is possible to realize a ground heat exchanger by one or more vertical probes with a depth that exceeds the neutral zone (approximately 20 meters, as indicated by Omer, 2008). The water temperature at the exit of the geothermal probe depends on the thermal characteristics of the ground and on the depth of the heat exchanger; the heat exchange with the ground takes place at relatively low temperatures compared with those of the external environment. Thus, if we consider that the regenerating fluid can be used directly within the cold heat exchangers, it is intuitive to understand that ∆Tspan can be particularly low. In fact, considering that the average supply temperature of the typical cooling elements for domestic use is between 280.0 K (fan coils) and 287.0 K (radiant floors) and 5

Page 5 of 15

considering the temperature levels that a fluid can reach at the geothermal probe exit (Ozyurt and Ekinci, 2011; Zarrella et al., 2013), a working system with a ∆T span of 10 K can be expected. It is clear that the technology is completely invertible. Indeed, to employ a GeoThermag device for winter heating, the cold side should be reversed with that of the hot, as is the case for traditional GSHP. In this case, if we consider that the average supply temperature of the typical lowtemperature heating elements for domestic use is between 298.0 K (radiant floors) and 318.0 K (fan coils) and if we consider the temperature levels that a fluid can reach at the geothermal probe outlet (Omer, 2008), a device working at a ∆Tspan ranging from 7.0 K (radiant floors) to 25.0 K (fan coils) is conceivable. In addition to the air conditioning field, the GeoThermag system could easily be used to power the cooling systems that are present in some industrial sectors, such as in the food sector. A typical example is that of wine preservation (wine cooler room) where it is necessary to maintain a temperature of between 283.0 K and 287.0 K.

3. Proof of concept To demonstrate the technical feasibility of the GeoThermag technology, we developed an experimental apparatus and tests aimed at assessing its energy performance. 3.1. Experimental facility In Fig. 1, we show an elementary scheme of the experimental device that we used to demonstrate the GeoThermag technology. The magnetic refrigerator that was employed is 8Mag (Fig. 2), a prototype that we extensively discussed in our previous article (Aprea et al., 2014). 8Mag is a magnetic refrigerator that is characterized by a group-rotating of permanent magnets realized through a Halbach array configuration that was modified and is able to guarantee a maximum magnetic field of 1.25 T in two areas with a high magnetic field and a magnetic field of 0.01 T in two areas with low magnetic field when the free air gap (magnetization area) is 43 mm. The magnetocaloric material is housed in 8 regenerators with an available volume of 31.5 cm3. The cycle frequency (fAMR) is determined by rotating the magnets; specifically, for each rotation of the magnets, each regenerator experiences two AMR cycles. A hydraulic system obtained by the combination of a rotary valve and a vane pump ensures the proper distribution of the regenerating fluid in each component of the apparatus in accordance with the AMR cycle phases.

Compared to the configuration shown in our previous article (Aprea et al., 2014), 8Mag has undergone a slight modification to the piping system. To reduce the overall pressure losses, we adopted pipes with larger diameters. Specifically, we replaced the pipes that had internal diameters of 4 mm with pipes that had an inner diameter of 6 mm, with the exception of the connection sections of the valve-regenerator. The cold heat exchanger is realized by the combination of electric resistance with a thermally insulated pressure vessel. A variable voltage supply feeds the electrical resistance to provide a thermal load that is variable from 0 to 500 W. 6

Page 6 of 15

A vertical geothermal probe acts directly from the hot heat exchanger. For our research purposes, we took advantage of a geothermal probe that was intended to be used with a traditional GSHP. The probe is double U-shaped and is made with PE-XA pipes that have an external diameter of 32 mm and a total length of 100 m. It is inserted into a hole with a diameter of 152 mm. The volume between the probe and the ground is filled with a mixture of bentonite and sand (thermal conductivity equal to 1.8 W m-1K-1). The probe length was defined according to the ASHRAE method (Philippe et al., 2010) by setting the thermal design power exchanged with the ground to 5.0 kW. However, to thermally characterize the coupled probe-ground, we conducted a Thermal Response Test (TRT); the results are reported in the next paragraph.

3.2. Thermal characterization of the coupled probe-ground Before connecting the geothermal probe to 8Mag, we conducted a series of tests in accordance with the UNI 11467: 2012 to determine the thermal power that is actually exchanged from the probe towards the ground, the temperature of the undisturbed ground and the thermal resistance of the coupled probe-ground. The tests were conducted using a special device that consists of a hydraulic circuit, a boiler and a measurement system (Tab. 1). Following the instructions given in the literature (Borinaga-Treviño et al., 2013; Lee et al., 2012), we considered the model of the linear thermal source for which: (1) where:  is the average temperature of the fluid at the inlet and outlet of the probe as a function of time t;  is the temperature of the undisturbed ground  r b is the radius of the hole that housed the probe  is the linear thermal power transferred by the probe through the boiler that is measured by knowing the electrical power supplied and the total length of the probe, which is interpreted as the depth of the hole  is the thermal diffusivity of the ground estimated from the average values available in the literature (Zarrella et al., 2013; Paoletti et al., 2015)  is Euler's constant and is equal to 0.5772  is the thermal conductivity of the ground  is the equivalent thermal resistance of the coupled probe-filling material.

7

Page 7 of 15

In eq. (1), all of the values are determined in situ by experimental measurements, except for the thermal conductivity, and the equivalent thermal resistance, . In particular, was measured by circulating the fluid in the test apparatus and keeping the test apparatus off the water heater and waiting until the temperature of the incoming fluid was close to that of the fluid in the output from the probe. Knowing , we proceeded with the moment by moment measurement of by keeping the water heater on. Next, we measured the by feeding the electrical resistances of the boiler with a variable voltage. Starting with a value of 100 W, we increased the electrical power until was not in steady state. For values higher than 4500 W (q = 45 W m-1), tended to not reach steady state; we chose that value as the maximum value for the thermal power exchanged by the fluid through the geothermal probe. Consequently, we analysed the profile obtained with a linear power of q = 45 Wm-1 and considered equation 1 to be rewritten in the following form: (2) with (3) .

(4)

Interpolating the experimental data, we obtained the values of the coefficients and p; subsequently, by eq. (3) and eq. (4), we obtained the values of and . In Tab. 2, we report the thermal characterization of the geothermal probe.

4. Experimental investigation To test the energy performance and the temperature levels reached by the GeoThermag system, we obtained measurements of the temperature span, volumetric flow rate of the regenerating fluid, pressure drop, thermal load, electrical power and COP. 4.1. Measurement method Using PT100 thermo-resistances placed between the input and the output of each component, we measured the temperature of the regenerating fluid; in particular, with reference to Fig. 1:  the temperature span (∆Tspan) is the average time between the temperature of the regenerating fluid exiting the hot side (THFo) and the temperature of the regenerating fluid exiting the cold side (TLFo);  the temperature of heat rejection (TH) is the average time between the temperature of the water exiting the hot side of the magnetic refrigerator (THFo) and the water exiting the geothermal probe (T HFi), both measured under steady state condition. By using a magnetic flow meter placed at the exit of the geothermal probe, we measured the volumetric flow rate of the regenerating fluid. To estimate the pressure drop of the entire device, we employed two piezoelectric pressure sensors that were placed between the input and the 8

Page 8 of 15

output of the pump. With reference to Fig. 1, we calculated the total pressure loss as the average time over the difference between the pressure at the outlet (Pout) and the pressure at inlet of the pump (Pin). To measure the cooling capacity developed by the device, we thermally isolated the outer surface of the cold heat exchanger. In this way, we can consider the following equation to be valid: (5) where is the electrical power that is absorbed by the electric heater inserted into the exchanger and measured using a watt transducer. Using a PT100, we performed temperature measurements on the outer surface of the insulation coat of the exchanger and verified the quality of the insulation level. In addition, using a two-watt transducer, we measured the electrical power absorbed by the pump ) and the electrical power absorbed by the electric motor used for magnet rotation following equation:

. For the evaluation of the COP, we referred to the (6)

In Tab. 3, we report the characteristics of the sensors used and the accuracy of the measurements performed.

4.2. Test method For the following work, we used gadolinium as the magnetocaloric refrigerant and distilled water as the regenerating fluid. Specifically, we loaded each regenerator with 150 g of gadolinium spheres with a diameter in the range from 400 to 500 microns, for a total of 1.20 kg of refrigerant. To reduce the influence of the temperature of the air surrounding the device, we insulated the connecting pipes between the heat exchangers and the magnetic refrigerator. In addition, as suggested by Engelbrecht et. al. (2012), during all of the tests, we introduced the magnetic refrigerator in a climatic room, where the air temperature was maintained in a small range between 293.0 and 298.0 K. Although the fluid flow rate affects the performance of a magnetic refrigerator (Tušek et al., 2013), for this first experimentation, we couldn’t easily change the flow rate, because we would preserve the bubble air growing across the geothermal probe and, at the same time, reduce the pressure drop. Consequently, keeping the fluid flow rate fixed at 5.0 l min-1, we carried out a set of tests at AMR cycle frequencies and for various thermal loads. In Tab. 4 we report some useful operating conditions adopted during the tests and measured in accordance with Aprea et. al (2014).

Because the temperature of the hot side was not under control, as imposed by the heat exchange conditions between the geothermal probe and ground, we chose to conduct the 9

Page 9 of 15

experiments entirely during the summer season; all of the tests were carried out from June to September. Starting from a condition of no load, for each AMR cycle frequency, we increased the thermal load step by step. For each test, we waited to achieve the following expected steady-state conditions: the temperature span change was less than the measurement accuracy for a time longer than 300 s. In doing so, we increased the load until the temperature span that was reached was greater than or equal to zero. 4.3. Results and discussion In the various test conditions, the temperature of the external environment showed excursions in the range from 290.1 to 305.3 K, while the rejection temperature, TH, underwent small changes that were contained in the range of measurement accuracy, settling at a value of 289.5 K next to the temperature of the undisturbed ground Tm. This result confirmed the first fundamental characteristic of the GeoThermag technology, the possibility of having a temperature TH that is approximately constant and lower than the external environment temperature. Because we held the volumetric flow rate of water constant for all of the tests, the pressure drop showed values that were approximately constant and equal to 6.85 bar. Specifically, excluding the hydraulic circuit of the magnetic refrigerator, we performed circulation tests to estimate the pressure drop resulting from the geothermal probe. The tests showed that along the geothermal probe, the pressure drop settled to 0.96 bar. In agreement with the literature (Jacobs et al., 2014), in Fig. 3, we report the change of ∆Tspan as a function of the thermal load for different frequencies f AMR. The maximum refrigeration power reached by GeoThermag is equal to 239.8 W, with a ∆Tspan close to 0 K and a frequency f AMR of 0.38 Hz; the maximum ∆Tspan is equal to 10.9 K in the absence of thermal load and with a frequency f AMR of 0.77 Hz.

From Fig. 4, we can see that for each thermal load value, it is possible to identify a value of the frequency fAMR for which the ∆Tspan is at a maximum. This trend of the temperature span obtained experimentally is in agreement with published results for other conditions (Tura and Rowe, 2011; Tusek et al., 2011). As noted by Nielsen et al. (2011), at lower frequencies there is a great influence of the longitudinal thermal conduction and the regenerator utilization becomes too high. As observed experimentally in Russek et al. (2010), we can envision that at lower frequencies the temperature span increases with increasing frequency, until a certain optimum frequency from which the irreversible losses become significant and the heat transfer is affected, so the regenerator is not capable to maintain a high temperature span.

However, to define the applicability of the GeoThermag as an air conditioning system, we show the performance of the temperature TLFo (feeding Fluid Flow temperature) at different frequencies fAMR and for different values of the thermal load in Fig. 5. It is well known that the terminals in the hydronic systems require an input of water temperature ranging from 280 to 10

Page 10 of 15

282 K for the fan coil and 286 to 291 K for the radiant panels. Consequently, when feeding one or more fan coils through the GeoThermag device, the efficiencies would not be satisfactory; the system could produce a reduced thermal load (60 W) only for two values of the frequency fAMR. Instead, one may notice that GeoThermag shows good compatibility with radiant floor cooling systems. In fact, the system is able to ensure a flow temperature that is appropriate for the operation of radiant floors under any condition in terms of the thermal load. Furthermore, it can be noted that for high thermal loads (140-190 W), each of the experimental frequencies fAMR represents a valid operating point for the GeoThermag system coupled to the radiant floor. This result confirms the second essential characteristic of the GeoThermag technology, the possibility of feeding commercial cooling terminals.

By analysing Fig. 6, it should be noted that for each thermal load condition, an increase of the frequency fAMR results in a reduction of COP. Because the volumetric flow rate was held constant in all tests, this result is attributable to the increase of the mechanical work that was necessary for the rotation of the magnets; the reasons for this increase can be either of a mechanical nature (bearing friction) or the nature of thermo-magnets (magnetization work and eddy currents). However, additional research is required to allocate a weight to either reason; this is the subject of future work. In addition, in Fig. 6 it can be seen that for each frequency fAMR, the value of the COP increases with increasing thermal load. This is solely related to the increase in cooling capacity at the expense of reducing the ∆Tspan. These results allow for the optimum operating conditions to be defined for the GeoThermag system at a lower frequency; specifically, the system has high-energy performance (COP = 2.20) for a frequency fAMR equal to 0.26 Hz with a cooling capacity of 190 W and a flow temperature of 287.9 K. Although the maximum value of COP that was obtained was found to be lower than that obtainable with any vapour compression refrigerator that is currently on the market, it should be noted that the GeoThermag system is able to work without harmful substances. It should be noted that the use of different refrigerants, including gadolinium, that are characterized by lower Curie temperatures, could greatly improve the performance of the GeoThermag systems in terms of both cooling capacity and ∆Tspan. In this regard it would be interesting to experiment with some novel materials that were recently analysed, as LaFeCoSi-based alloys or to adopt regenerators multi layer made of Gd-based alloys. Furthermore, the reduction of the work of handling magnets would bring a great advantage in terms of COP. For instance, by acting on the mechanical drag of the magnets it can reduce the friction work, thus reducing a quote of the work necessary for the magnetization. Consequently, we can confirm the third feature, the possibility to obtain an eco-friendly air conditioning system.

5. Conclusions

11

Page 11 of 15

In this paper, we presented a new refrigeration concept, GeoThermag, which is a combination of magnetic refrigeration technology with that of low temperature geothermal energy. To demonstrate the applicability of the GeoThermag technology, we developed a pilot system that consists of a 100-m deep geothermal probe; inside the probe, water flows and is used directly as a regenerating fluid for a magnetic refrigerator. After thermally characterizing the geothermal probe and the heat source at high temperatures, we conducted experimental tests aimed at defining the performance of the GeoThermag system. In this regard, with a fixed volumetric flow rate of the regenerating fluid, we experienced different operating conditions in terms of frequency fAMR and in terms of the thermal load. For each test, we measured the ∆Tspan, cooling capacity, energy consumption and COP. From a qualitative point of view, we can say that the GeoThermag technology allows:  having a temperature T H that is approximately constant and lower than the temperature of the external environment.  feeding commercial cooling terminals  obtaining an eco-friendly air conditioning system. The above statements are justified by the experimental results. In particular, the GeoThermag system showed the ability to produce cold water even at 281.8 K in the presence of a heat load of 60 W. In addition, the system has shown the existence of an optimal frequency f AMR, 0.26 Hz, for which it was possible to produce cold water at 287.9 K with a thermal load equal to 190 W with a COP of 2.20. Observing the temperature of the cold water that was obtained in the tests, the GeoThermag system showed a good ability to feed the cooling radiant floors and a reduced capacity for feeding the fan coil systems. Although these results can be easily overcome by vapour compression systems, it should be noted that the use of different refrigerants, including gadolinium, that are characterized by lower Curie temperatures, could greatly improve the performance of the GeoThermag systems in terms of both cooling capacity and ∆Tspan. Furthermore, the reduction of the work of handling magnets would bring a great advantage in terms of COP. In future studies, we will focus on the analysis of the energy losses.

References Aprea, C., Greco, A., Maiorino, A., Mastrullo, R.,Tura, A., 2014. Initial experimental results from a rotary permanent magnet magnetic refrigerator, Int. J. Refrigeration, 43, 111-122 Aprea, C., Greco, A., Maiorino, A., 2015. The application of a desiccant wheel to increase the energetic performances of a transcritical cycle, En. Conv. and Man., 89, 222-230. Bansal, P., Vineyard, E., Abdelaziz, O., 2011. Advances in household appliances – A review, App. Th. Eng. 31, 3748-3760 Borinaga-Treviño, R., Pascual-Muñoz, P., Castro-Fresno, D., Blanco-Fernandez, E., 2013. Borehole thermal response and thermal resistance of four different grouting materials measured 12

Page 12 of 15

with a TRT, App. Th. Eng., 53(1), 13-20 Engelbrecht, K., Eriksen, D., Bahl, C.R.H., Bjørk, R., Geyti, J., Lozano, J.A., Nielsen, K.K., Saxild, F., Smith, A., Pryds, N., 2012. Experimental results for a novel rotary active magnetic regenerator, Int. J. Refrigeration. 35 (6), 1498-1505. Eriksen, D., Engelbrecht, K., Bahl, C.R.H., Bjørk, R., Nielsen, K.K., Pryds, N., 2014. Design and initial testing of a compact and efficient rotary AMR prototype, 6th IIF-IIR International Conference on Magnetic Refrigeration, Victoria BC, Canada, 7-10 September 2014 European Commission, 2014. Regulation (EU) No 517/2014 of the European Parliament and of the Council of 16 April 2014 on fluorinated greenhouse gases and repealing Regulation (EC) No 842/2006. Jacobs, S., Auringer, J., Boeder, A., Chell, J., Komorowski, L., Leonard, J., Russek, S., Zimm, C., 2014. The performance of a large-scale rotary magnetic refrigerator, Int. J. Refrigeration, 37 (1), pp. 84-91. Kitanovski, A., Tusek, J., Tomc, U., Plaznik, U., Ozbolt, M., Poredos, A., 2015. Overview of existing magnetocaloric prototype devices, in Magnetocaloric energy conversion, Springer International Publishing, 269-330 Lee, C., Moonseo Park, M., Nguyen, T.B., Sohn, B., Choi, J. M., Choi, H., 2012. Performance evaluation of closed-loop vertical ground heat exchangers by conducting in-situ thermal response tests, Ren. En., 42, 77-83 Llopis, R., Cabello, R., Sánchez, D., Torrella, E., 2015. Energy improvements of CO2 transcritical refrigeration cycles using dedicated mechanical subcooling, Int. J. Refrigeration. http://dx.doi.org/10.1016/j.ijrefrig.2015.03.016. Lozano, J.A., Engelbrecht, K., Bahl, C.R.H., Nielsen, K.K., Eriksen, D., Olsen, U.L., Barbosa, Jr. J.R., Smith, A., Prata, A.T., Pryds, N., 2013. Performance analysis of a rotary active magnetic refrigerator, Appl. Energy, 111, 669-680 Lozano, J.A., Engelbrecht, K., Bahl, C.R.H., Nielsen, K.K., Barbosa, Jr. J.R., Smith, A., Prata, A.T., Pryds, N., 2014. Experimental and numerical results of high frequency rotating active magnetic refrigerator, Int. J. Refrigeration, 37, 92-98 Nielsen, K.K., Tušek, J., Engelbrecht, K., Schopfer, S., Kitanovski, A., Bahl, C.R.H., Smith, A., Pryds, N., Poredoš, A., 2011. Review on numerical modelling of active magnetic regenerators for room temperature applications. Int. J. Refrigeration, 34(3), 603-616 Omer A. M., 2008. Ground-source heat pump systems and applications, Ren. and Sust. En. Rev., 12, 344-371 Ozyurt, O., Ekinci, D.A., 2011. Experimental study of vertical ground-source heat pump performance evaluation for cold climate in Turkey, App. En., 88, 1257-1265 Paoletti, V., Langella, G., Di Napoli, R., Amoresano, A., Meo, S., Pecoraino, G., Aiuppa A., 2015. A tool for evaluating geothermal power exploitability and its application to Ischia, Southern Italy, Appl. En., 139, 303-312 Philippe, M., Bernier, M., Marchio, D., 2010. Vertical geothermal borefields, Ashrae Journal, July, 20-28

13

Page 13 of 15

Russek, S., Auringer, J., Boeder, A., Chell, J., Jacobs, S., Zimm, C., 2010. The performance of a rotary magnet magnetic refrigerator with layered beds. In: Proc. 4th Int. Conference on Magnetic Refrigeration at Room Temperature, Baotou, Inner Mongolia, China, 339-349 Tura, A., Rowe, A., 2011, Permanent magnet magnetic refrigerator design and experimental characterization, Int. J. Refrigeration 34 (3), 628-639. Tušek, J., Kitanovski, A., Prebil, I., Poredoš, A.. 2011 Dynamic Operation of an Active Magnetic Regenerator (AMR): Numerical Optimization of a Packed-Bed AMR. Int. J. Refrigeration, 34 (6): 1507–17. Tušek, J., Kitanovski, A., Zupan S., Prebil, I., Poredoš, A.. 2013. A comprehensive experimental analysis of gadolinium active magnetic regenerators. App. Th. Eng., 53(1), 57-66 Yu, B., Liu, M.,Egolf, P.W.,Kitanovski, A.,2010. A review of magnetic refrigerator and heat pump prototypes built before the year 2010, Int. J. Refrigeration, 33(6)1029-1060 Zarrella, A., Capozza, A., De Carli, M., 2013. Performance anlysis of short helical borehole heat exchangers via integrated modelling of borefield and a heat pump: a case study, App. Th. Eng., 61, 36-47

Fig. 1 Elementary scheme of the experimental device Fig. 2 Picture of 8Mag Fig. 3 Performance of ∆T span with varying heat loads at different AMR cycle frequencies, f AMR Fig. 4 ∆Tspan as a function of the f AMR at different thermal loads Fig. 5 Change of the flow temperature at different frequencies f AMR for different thermal loads Fig. 6 Change of COP at different f AMR for different thermal loads

Table 1. Measurement instruments used for the TRT Transducer Characteristic Accuracy Flowmeter Electromagnetic  0.5% RTD PT100  0.15 [K] Pressure gauge Piezoelectric  75 [kPa] Table 2 TRT results Quantity Result 4500 [W] Tm 290.2 [K] 2.56 [W m -1 K -1] 0.164 [KW m -1] 14

Page 14 of 15

Table 3 Characteristics of the sensors used and the accuracy of the measurements performed Quantity Characteristic Accuracy Temperatures RTD PT100  0.1 [K] Pressure Piezoelectric Pressure  75 [kPa] gauge Volumetric flow rate Electromagnetic  0.5% Flowmeter Rpm / Frequency Optical Encoder  0.01 [° s -1] Mass of magnetocaloric Electronic Balance  0.2 [g] material Electrical power Electromagnetic  0.2% Wattmeter COP Eq. 5 & Propagation 0.35 Error Analysis Table 4. Operating conditions of 8Mag during the tests. In accordance with Aprea et al.(2014), the resulting accuracy of the utilization factor ( ) is ±5.3% Vfr fAMR tf = tm mrf Ω -1 [l min ] [rpm] [Hz] [s] [g] [-] 5 7.7 0.26 0.98 36.17 2.65 5 11.5 0.38 0.65 24.15 1.77 5 15.4 0.51 0.49 18.02 1.32 5 23.1 0.77 0.32 12.00 0.88 5 38.4 1.28 0.20 7.23 0.53

15

Page 15 of 15