Experimental and computational investigation of a

Thermal Science and Engineering Progress 7 (2018) 87–98

Contents lists available at ScienceDirect

Thermal Science and Engineering Progress journal homepage: www.elsevier.com/locate/tsep

Experimental and computational investigation of a latent heat energy storage system with a staggered heat exchanger for various phase change materials

T

⁎

Maria K. Koukoua, , Michalis Gr. Vrachopoulosa, Nikolaos S. Tachosa, George Dogkasa, Kostas Lymperisb, Vassilis Stathopoulosc a

Technological Education Institute of Sterea Ellada, Department of Mechanical Engineering, Energy and Environmental Research Laboratory, 344 00 Psachna campus, Evia, Greece Z&X Mechanical Installations Ltd, 12 Agapinoros Street, 8049 Paphos, Cyprus c Technological Education Institute of Sterea Ellada, Department of Electrical Engineering, Laboratory of Chemistry and Materials Technology, 344 00 Psachna campus, Evia, Greece b

A R T I C LE I N FO

A B S T R A C T

Keywords: Heat exchanger PCM Energy storage Experimental Simulation Convection

This work reports the operation of a Latent Heat Thermal Energy Storage system (LHTES) utilizing a staggered heat exchanger (HE) and using various organic Phase Change Materials (PCMs). In a LHTES test rig set measurements regarding energy storage and release were performed in the working temperature range of each Phase Change Material. Nominal melting temperatures of the PCMs used were 40–53 °C. Computational Fluid Dynamics (CFD) simulation was applied to follow the operation of the test rig. The test rig consisted of a compact insulated tank, filled with PCM, a staggered heat exchanger to supply or extract thermal energy by the PCM and a water pump to circulate water as a Heat Transfer Fluid (HTF). Different HTF flow rates affect charging (melting) and discharging (solidification) processes but more significant was the effect of heat transfer mechanisms occurring. The latter was confirmed by inserting buoyancy currents created due to convection in a CFD simulation program where melting time was reduced compared to the same conditions with only conduction occurring. The suggested LHTES configuration is a promising compact unit despite the PCMs thermal resistance and solidification hysteresis phenomena, as well as the heat transfer mechanism strongly affecting the energy storage process.

1. Introduction Energy economy demands cost effective solutions for storing heat either in case of a mismatch among production and demand or by heat waste streams. Thermal Energy Storage solutions have been extensively applied on buildings in order to utilize the excess of for example solar energy in the form of sensible energy in a hot water storage tank. Thus solar energy systems for residential cooling/heating and Domestic Hot Water (DHW) production systems require thermal energy storage as a reliable mean to store energy for later use, improving the reliability and the performance of the system and reducing the mismatch between supply and demand of the available solar power [1,2]. Considerably more efficient way to store thermal energy is latent heat thermal energy storage (LHTES) via the integration of PCM, exploiting high values of latent heat [1]. LHTES can store 5–14 times more heat per unit volume than sensible storage materials, such as water [2]. This is due to the

⁎

utilization of the high storage capacity of PCMs as they release or absorb large amounts of thermal energy during their liquid to solid and vice versa phase transition. This occurs in a small volume, much smaller than sensible storage technical solutions and it is highly related to each PCM molecular physical and chemical properties. In a LHTES a heat exchanger is used to capture or transfer the thermal energy from a PCM to a heat transfer working fluid (HTF) during discharging (solidifying) as well as the opposite during PCM charging (melting). The energy stored can be recovered in a constant temperature of the HTF either a bit higher or lower than the melting temperature of the PCM. This is easily utilized by heat pumps. There are many phase change materials in the market and the selection of the appropriate PCM to be used in a certain application or temperature operation must take into consideration technical requirements, material properties as well as cost [3]. Some of these characteristics are the melting temperature which should meet the

Corresponding author. E-mail addresses: [email protected], [email protected] (M.K. Koukou).

https://doi.org/10.1016/j.tsep.2018.05.004 Received 22 September 2017; Received in revised form 16 May 2018; Accepted 20 May 2018 2451-9049/ © 2018 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/BY-NC-ND/4.0/).


M.K. Koukou et al.

condenser in small air conditioning units, shows the highest ratio of heat transfer area to external volume. It has the highest average thermal power, with values above 1 kW for both charging and discharging tests at cases with larger temperature differences between PCM and water. For lower temperature differences, average thermal power values are threefold higher than the highest power values achieved by the second best heat store. Agyenim et al. [25] worked with a horizontal shell and tube heat exchanger in contact with a medium temperature PCM with a melting point of 117.7 °C. Two experimental configurations were studied. One used a heat transfer tube and the second used a multi-tube unit with four heat transfer tubes. Results showed that heat transfer extended in the radial and angular directions during the phase change in both configurations, showing two-dimensional heat transfer in the systems. A high temperature pilot installation able to test different types of thermal energy storage systems and materials was designed and built by Oro et al. [26]. It composed by a heating system, a cooling system, and different storage tanks. A synthetic oil was circulated as HTF working in temperature range from 100 °C to 400 °C. Two different PCMs with different phase change temperatures and heats of fusion were selected and studied using different operating conditions for a solar cooling application. Hydroquinone and d-mannitol were the PCM selected. It was found that the energy stored by d-mannitol was higher than that for hydroquinone; in particular, the enhancement was about 30% and 20% during the charging and the discharging processes even though the enhancement of the latent heat was only 10 and 16%, respectively. Rathod et al. [27] conducted an experimental study on melting and solidification enhancement achieved in a shell-and-tube latent heat thermal energy storage system using longitudinal fins. The results showed a 12.5% and 24.52% reduction in the melting time for HTF inlet temperature of 80 °C and 85 °C respectively. Paria et al. [28] designed and developed a thermal storage unit and studied melting and solidification of a paraffin wax. They used a horizontal shell-tube heat exchanger having radial fins and applied various Reynolds numbers at the laminar flow regime. The results generally showed that the increase in the fin density increased the amount of charging and discharging thermal energy during melting and solidification processes, respectively. An 86% increase on the energy extracted was shown by Khalifa et al. [35] when using finned heat pipes in high temperature latent heat thermal energy storage systems. The heat pipes effectiveness also increased by 24%. Amini et al. [31] used embedded finned water containing heat pipes in contact with a PCM in order to study the capability of such a LHTES system to store and release energy in a small volume compared to sensible heat TES system The PCM used was PLUSICE S89 with melting temperature of 89 °C. The operation was not optimized but the system was successfully applied for the recovery and reuse of wasted heat. It was observed that the PCM low thermal conductivity was also a significant issue mainly during the discharging stage. Authors suggested further studies towards better heat transfer either based on the design of the HE, or the PCM properties. Promoppatum et al. [32] designed a cross flow shell-and-tube heat exchanger with staggered tube array containing PCM to absorb the heat from the warm air stream ejected from building HVAC, and to release heat into the cold air stream entering building. Frazzica et al. [33] tested two heat exchangers, a fin-and-tubes custom made HE and a commercial asymmetric plate heat exchanger, for the application with PCMs. The two devices were tested in a custom made testing rig using a commercial paraffin (Plus ICE A82), for heat storage temperatures in the range 80–100 °C. The results show that for the tested material the plate heat exchanger assures a better exploitation of the heat stored inside the material. Merlin et al. [34] developed a test bench to characterize the performances of five exchangers for an industrial application with various configurations. The first one is a tubular exchanger with the PCM in the annular part with a melting temperature between 55 and 60 °C. For two configurations, the heat exchange surface is increased

temperature range of the application focused, latent heat value, chemical stability and compatibility avoiding corrosion issues [3,4]. High values of latent heat as well as small density variation with phase transition will reduce the volume of the thermal energy storage tank. Chemical stability and compatibility will provide flexibility in tank material design and eventually cost. In principle any phase transition from solid-to-solid, solid-to-liquid, solid-to-gas and liquid-to-gas can be of thermal storage use. Despite that solid-to-gas and liquid-to-gas can store high amounts of energy they exhibit extreme volume changes prohibiting their practical use in TES. From a different point of view such condensation effects are very effectively utilized in waste heat recovery on a different technical approach [5–7]. PCMs with solid-to-liquid transition show much smaller or minimum change in volume and they find application in TES systems, especially in building energy storage systems [8–13]. PCMs used, are classified according to their chemical nature in organic, inorganic and eutectics and their applications as well as their thermo-physical properties are a subject of interest [4,8,10,13]. The organic PCMs, as these used in this study are in general hydrocarbon molecules based blends, of technical grade to minimize cost, they exhibit chemical stability and high latent heat of fusion. Their melting point is suitable for heating/cooling systems for the building sector and household applications and they show negligible corrosiveness to metals being by far more environmentally friendly than inorganic PCMs. On the other aspect view they can be of questionable compatibility with engineering polymers, they show low thermal conductivity and they are flammable. The design of commercial thermal energy storage units, requires further investigation to successfully implement PCM. The proper design of HEs used in such units must take into account desired storage rate, transient PCM heat exchange and required energy storage capacity. Such design concept considers the PCM characteristics [3] including the low thermal conductivity, a major drawback of organic PCMs. In order to overcome low conductivity, various engineering approaches have been suggested like the use of fins to maximize heat transfer as well as materials approaches introducing conducting additives i.e. nano-particles in the fluid medium [3]. Research on the development of various heat exchangers used in LHTES units has been attracting interest since the first approach of Shamsundar and Srinivasan four decades ago [14]. Various configurations have been studied including tubular heat exchangers, shell-and tube heat exchangers, plate and frame heat exchangers and heat pipe technology [15–36]. A multi-layer latent TES system, with vertically parallel PCM plates, was reported to reduce the mismatch between energy supply and demand [15]. Later Banaszek et al. studied the behaviour of PCMs during charging and discharging periods in a spiral TES system [16,17]. Ismail and Henriquez suggested the design of a storage tank, integrating a working fluid circulation and encapsulated PCM [18]. Different fin configurations have been suggested [19–35] as they proved to significantly enhance the heat transfer during the phase change process. Dimension, type and number of fins are selected based on a compromise between maximizing HE contact area for increasing the heat transfer surface versus the available PCM mass towards higher energy storage capacity. The heat transfer enhancement in the latent heat thermal energy storage heat exchanger using an internally finned tube was studied by Zhang and Faghri [29]. Jamal and Baccar [30] showed the effect of natural convection on the PCM solidification time and the influence of fins number on heat transfer rate. This work was performed in a PCM-air heat exchanger system using internally and externally finned tubes. Medrano et al. [22] compared different types of heat exchangers operating with RT35 by Rubitherm GmbH as PCM and water as HTF. Performance evaluation was based on the required time to fully melt a certain PCM quantity with different types of heat exchangers. Results indicate that compact tube finned (staggered) heat exchanger made of aluminum fins and copper tubes widely used as evaporator or 88


M.K. Koukou et al.

with fins. Conductivity enhancement is also tested with dispersed graphite powder in one case and an Expanded Natural Graphite (ENG) matrix in the other case. The ENG matrix was concluded as the best configuration to enhance performances of the LHTES. The measured thermal conductivity of the ENG/PCM composite was found about 100 times greater than the conductivity of the PCM alone. Gasia and colleagues [36] investigated the influence of the addition of fins and the use of two different heat transfer fluids (Water and silicone Syltherm 800) in four latent heat thermal energy storage systems, based on the shell-and-tube heat exchanger concept. A paraffin RT58 PCM was used. Results showed that finned designs introducing 4.7–9.4 times more heat transfer surface improved performance by up to 40%. On the contrary, for the same design, water of 3 times higher specific heat and 4.9 times higher thermal conductivity than Syltherm 800, yielded results up to 44% higher. In this work, a test rig has been developed in order to generate data about the performance of heat exchangers when they are used in LHTES applications. More specifically, a staggered heat exchanger has been installed and four different organic PCMs with different melting profile were tested. Thus, a range of operation temperatures was monitored collecting data by four types of PCMs based on the requirements of solar thermal energy storage for residential building applications. The testing temperature range was 30–70 °C and the PCMs nominal melting points used were 40–53 °C. Such type of exchanger has not be tested in such a temperature range. Therefore limited data are available on using such type of HE in comparison with other configurations [22]. The main target of this work was to make an initial assessment of the working performance of the developed thermal energy storage (TES) unit in a temperature range of 30–70 °C suitable for residential heating and domestic hot water production systems. Data were collected by applying two different HTF flow rates on the melting and solidification time of the four PCMs. The thermal properties of the PCMs used were also investigated by means of Differential Scanning Calorimetry (DSC) and the results were related to the data retrieved by the operation of the unit providing useful operational information. The role of the heat transfer mechanisms was identified by experimental results and CFD. The experimental rig was simulated by CFD and the results extracted were used to understand the process duration and the effect of the HTF flow rate along with the heat transfer mechanism acting in both melting and solidification processes.

base, 2 × 9213 16ch cards, LabVIEW® data logger) and a personal computer. The rig allows easy replacement of the phase change material and the heat exchanger as well. Thermal insulation has been applied on the outer side of the tank glass walls. The space between the tubes, the fins and the tank is filled with PCM. The hot HTF supplied from hot water boiler, delivers its thermal energy through HE and melts the PCM causing charging process. HTF inlet temperature is above the melting point of PCM in use. Charging is complete when PCM melts and an amount of HTF heat is now stored in the molten PCM. During the solidification (discharging) process, HTF supplied by the cold water buffer tank in a low temperature according to the PCM used. Cold HTF flows through the tube and solidifies the molten PCM absorbing the heat of the exothermic phase change. The stored energy is released from the PCM as it solidifies and discharges. To analyse the HE thermal performance and the charging and discharging process of the LHTES test rig, temperature measurements are performed on the water inlet and outlet and also within the PCM mass. Positioning of the thermocouples is indicated in Fig. 1c.

2. Experimental methodology

2.3. Experimental procedure and conditions tested

2.1. Experimental set-up description

In each experimental condition studied, prior PCM loading, the rig was tested for leakages. The tank was de-aired under warm conditions after filled by melted PCM in order to avoid any air inclusions. The PCM was then solidified by cooling to room temperature. The experimental procedure was initiated following a charging – discharging sequence in a temperature range related to product properties according to manufacturer (Table 2) and as described below:

2.2. PCM’s properties The selection of the PCM to be used is an important parameter regarding the design of a LHTES as the amount of energy that may be recovered is directly related to the nature of the material. In this work, the selection of the PCM is based on the consideration of a low temperature solar thermal energy storage application. The materials tested are commercial products of PCM PRODUCTS® [37] and their detailed thermal and physical properties provided by the company are summarised in Table 2. The nominal phase change points of A40, A44, A46 and A53 used were 40–53 °C allowing identification of the working performance of the developed thermal energy storage (TES) unit in a temperature range of 30–70 °C suitable for residential heating and domestic hot water production systems. Further information on the thermal properties of the A40, A44, A46 and A53 was collected by thermal analysis measurements [38,39]. Differential Scanning Calorimetry tests were carried out using a Setaram® STA unit under air. DSC results were collected up on heating and cooling in the range 5–70 °C with 0.8 °C/min. Integration of thermal phenomena peaks for the required enthalpy calculation, was performed using Calisto® Processing software.

Fig. 1 shows the test rig with an experimental set-up of the following main components: an insulated heat storage glass tank, a heat exchanger, a boiler, a water buffer tank, a circulation pump, a flow meter, thermocouples, a data logger and a computer. An insulated heat storage glass tank is used allowing, if required, the visual observation and recording the melting and solidification steps. The dimensions of the tank were: 600 × 120 × 80 mm (Length × Width × Depth). A staggered heat exchanger (KLIMALLCO S.A.) (Fig. 1) was immersed in the tank. Its technical data are given in Table 1 and a 3D drawing in Fig. 1b. Hot water was supplied by an 80 l boiler with a 4 kW electric heater and cold water was supplied by a 200 l water buffer tank connected to an air to water heat pump. The experimental rig also included a circulation pump (Grundfos® ALPHA 2 32–60 180 inverter pump), a three-way temperature control valve (Belimo® LR24A-SR valve and Vector® TCI-W11 controller), a flowmeter (rotameter with measuring range of 6–60 l/h with ± 4% accuracy), thermocouples (T-type, with ± 0.5% accuracy), and the necessary piping and valves to regulate water flow. The data recording for the melting/solidification processes, namely PCM temperatures, HTF inlet/outlet temperatures and mass flow rate, was accomplished by a data logger (National instrument® cDAq – 9174

(i) Charging (melting): Up on initiation HTF (hot water) was circulated within the piping at a temperature of about 8 °C higher than the nominal melting temperature of the PCM used (Table 2). HTF temperature inlet was controlled by mixing water of the boiler and the buffer tank using a three-way mixing valve. Charging (melting) process ended when the temperature of the thermocouples at the edges of the exchanger (thermocouples K, L, M, N) exceeded nominal melting temperature as defined by the PCM producer by more than 3 °C. Then, discharging (solidification) step was initiated. (ii) Discharging (solidification): Subsequently, a discharging experiment started. Cold water from the buffer tank was circulated with a temperature of about 8 °C less than the melting temperature of the PCM. Temperature inlet was controlled similar as in charging step by a three-way mixing valve. Solidification process ended when the 89


M.K. Koukou et al.

Fig. 1. (a) Test – rig main parts layout, (b) Staggered heat exchanger 3D drawing, (c) Positioning of thermocouples within the tank. Table 1 Staggered heat exchanger characteristics. Number of tubes Inner tube diameter (mm) Outer tube diameter (mm) Tube material Fins material Fin thickness (mm) Fin length (mm) Fin spacing (mm)

Table 2 Thermal and physical properties of organic PCMs used in the experiments [37]. 12 7.75 9.525 Copper Aluminium 0.3 68 5.0

Freezing point (°C) Melting point (°C) Density solid (kg/m3) Density liquid (kg/m3) Thermal conductivity solid (W/m·K) Thermal conductivity liquid (W/m·K) Specific heat solid (kJ/kg·K) Specific heat liquid (kJ/kg·K)

temperature of the thermocouples at the edges of the exchanger (thermocouples K, L, M, N in Fig. 1c) was measured about 3 °C or more, depending on the case, less than the nominal solidification temperature defined by the PCM manufacturer product specifications (Table 2).

A40

A44

A46

A53

40 39 830 775 0.24 0.24 2.4 1.8

45.5 44 830 775 0.24 0.24 2.4 1.8

47 45 830 775 0.24 0.24 2.4 1.8

52.6 51.2 830 775 0.22 0.22 2.4 1.8

(0.0083 kg/s) and 60lt/h (0.0166 kg/s).

3. Experimental results and discussion

All experiments were performed under two HTF flow rates: 30lt/h

Melting and solidification process of a material is related to the 90


M.K. Koukou et al.

During solidification, all PCM exhibited the same behaviour. Similar behaviour was observed for all PCMs tested. In order to follow the operation of the TES unit the temperature difference between ΗΕ inlet and outlet was monitored and recorded. At the beginning of each experiment, and according to Eq. (1) heat flow rate was maximum due to high the temperature difference. It reduces as temperatures converge. Apart from flow rate, a key role regarding melting and solidification processes of a PCM is the temperature variation between HTF inlet temperature and PCM temperature. The above are based on the assumption that the tank containing the material is adiabatic. Plots below, show time variation for both charging and discharging processes of A40, A44, A46 and A53. Τhe mean value of recorded temperature by the thermocouples placed into the PCM tank (Fig. 1c) versus time is shown in Figs. 3–6. Fig. 3 shows charging and discharging regimes for A40 loaded LHTES unit working under two different flow rates. At the beginning of the charging regime the accumulation of energy in PCM is quite rapid (Regime Ma) until PCM reaches liquid state. This point is evident in Fig. 3 as a shoulder in the temperature curve at 40–41 °C for both flow rates (Point M in Fig. 3). The procedure is faster for the case of the high flow rate as expected. As it is shown in A40 DSC heating curve (Fig. 7) the thermal effect of latent heat absorption during melting is completed by 43.9 °C with a maximum at 39.2 °C. Subsequently, charging is associated with steady state heat transfer rate to the liquid PCM. In discharging regime a rapid reduction of PCM temperature is observed releasing initially sensible heat (Regime Sa) and then latent heat when phase change takes place (Regime Sm). The phase transition from liquid to solid is by far different than the solid to liquid phase change for A40 in Fig. 3. In order to identify the reason thermal analysis was carried out recording the thermal effects of A40 and all the rest PCM melting and solidifying behavior. Results are shown in Fig. 7. It has been reported that organic PCMs may show a temperature hysteresis on their solidification, identified also as super cooling or sub cooling effect [40]. The temperature hysteresis upon cooling is enhanced over faster cooling rates widening the temperature gap among melting and solidifying [40]. In such case phase change latent heat release may not be quantitatively captured by the HTF if LHTES operation is not optimized well within the temperature window of PCM liquid to solid transition. Therefore DSC measurement were conducted on heating and cooling cycle in the range 5 to 70 °C under a constant heating and cooling rate of 0.8 °C/min. Results in Fig. 7 show a temperature hysteresis for A40 solidification despite the narrow temperature variation reported by the producer (Table 2). Before that and upon melting main thermal solid to liquid phase change effect initiates at about 29 °C which coincides with the end of sensible heating regime Ma

energy provided (charging) or extracted (discharging) by the HTF. The rate of heat transfer depends on the HTF inlet temperature and flow rate via the equation:

q = ṁ HTF cP, HTF (Tin−Tout )

(1)

ṁ HTF = HTF mass flow rate (kg/s), cP, HTF = HTF specific heat capacity (kJ/kg·K), Tin = HTF inlet temperature (K), Tout = HTF outlet temperature (K), q = rate of heat transfer (kJ/s) In the other end LTHES systems can store energy in sensible and latent forms as shown below: Tm

Q = mαm QL, PCM +

Tf

∫ mCp dT + ∫ mCp dT Ti

Tm

= m [αm QL, PCM + Csp (Tm−Ti ) + Clp (Tf −Tm)]

(2)

where m, is the fraction of the melted phase change material, QL,PCM, is the latent heat per unit mass (J/kg), Tm, is the melting temperature (°C), Csp, is the average specific heat between the initial and melting temperatures (J/kg.K) and Clp is the average specific heat between melting and final temperatures (J/kg.K). By Eq. (1) it is then expected that the larger the value of HTF flow rate (for the same constant inlet temperature) the larger the rate of heat transfer by the HTF to the PCM hence the shorter the period to complete the process of charging (melting). Discharging process period is also decreased, respectively. Experiments were conducted using A40, A44, A46 and A53. When warm HTF starts to circulate within the HE pipes, the PCM starts to melt due to conduction between pipes and PCM. The liquid fraction, i.e. the ratio of the liquid to the overall mass, increases as time progresses. Due to buoyancy within the liquid state PCM, convection heat transfer occurs. This phenomenon is expected to increase heat transfer rate. Actually, conduction is the mechanism to transfer heat from liquid to solid PCM however convection is expected to facilitate the process. As PCM phase state changes from solid to liquid, energy equivalent to the respective latent heat is expected to get stored to the PCM without altering its temperature. Finally, when phase transition to liquid is achieved the PCM acts as a sensible heat storage medium by storing heat and increasing its temperature. During cooling down at the discharging step the first region to solidify was the volume of the PCM in contact with the HE walls (Fig. 2). The formation of the solidified PCM in the axial direction around HE pipes had a significant effect on total discharging process. The formation of that layer reduced convection heat transfer between HE walls and the liquid PCM of higher temperature. Conduction heat transfer is now the only possible mechanism to complete solidification.

Fig. 2. Pictures capturing different melting (upper row) and solidification (lower row) stages of A44. 91


M.K. Koukou et al.

Fig. 3. Mean temperature profiles of A40 for charging and discharging and two inlet feed rates.


Similar behavior is obtained for the rest PCMs. Fig. 4 shows charging and discharging regimes for A44 and for two flow rates. In the beginning of the charging the increase in PCM temperature is again quite rapid (Regime Ma, Fig. 4) until PCM initiates phase transition to liquid at 39–40 °C for both flow rates (Point M in Fig. 4). This is confirmed by the A44 DSC heating curve (Fig. 7) showing melting effect starting at 39 °C with a maximum at 44.4 °C. The latent heat absorption during melting is completed at about 47 °C. Subsequently, charging is associated with steady state heat transfer rate to the liquid PCM. During discharging a rapid reduction of PCM temperature is

identified in the monitored LHTES unit (Fig. 3). Maximum of thermal effect is found at 39.4 °C which is very close to point M of Fig. 3. The latent heat emission for A40 begins at about 37 °C (Regime Sm, Fig. 3) and it is completed at temperatures below 30 °C. This is a significant deviation from product properties information data, expected to affect the TES operation and efficiency. Similar data are recorded, in the case of the high flow rate (60 l/h) and the liquid to solid phase change has also been merely concluded. However, for both rates lower amount of energy is expected to be transferred in the LHTES unit due to the hysteresis effect of A40 PCM.

92


M.K. Koukou et al.



delivered. In this case the super cooling feature of this product is expected to affect the LHTES operation and efficiency when A44 is used. Fig. 5 shows charging and discharging regimes for A46 under two flow rates of HTF. Similar to A40 and A44 rapid increase is observed during charging (Regime Ma, Fig. 5) for both flow rates reaching phase transition (Point M in Fig. 5) at 46–47 °C. This is clearer for 60 l/h flow rate data and in good agreement with the results of the respective DSC heating curve (Fig. 7) for A46. The thermal melting effect is completed at about 51 °C. In discharging section a rapid reduction of PCM temperature is

observed releasing sensible heat (Regime Sa) and then latent heat when phase change takes place (Regime Sm). The region of phase change transition from liquid to solid is also different from solid to liquid in A44 curve (Fig. 4). A hysteresis phenomenon occurs also in A44 (Fig. 7) but different thermal phenomena are observed. In A44 DSC cooling curve (Fig. 7) the two recorded thermal effects show that latent heat emission occurs in two discrete phenomena in the range of 35–39 °C (first peak) where latent heat emission begins (Regime Sm) and a second sharp peak at 31 °C is only completed at about 28 °C. Thus the phase change is completed and the stored energy as latent heat may be

93


M.K. Koukou et al.

3

Table 3 Heat transfer data of LHTES.

(a)

0

PCM

ΔHC (kJ/kg)*

1.298·103 1.029·103

92.82

0.394·103 0.382·103

−93.73

A44

30 l/h 60 l/h

2.036·103 1.839·103

201.48

1.336·103 1.151·103

−199.84

A46

30 l/h 60 l/h

1.136·103 1.066·103

121.24

0.602·103 0.509·103

−110.44

A53

30 l/h 60 l/h

1.713·103 1.459·103

115.06

0.890·103 0.888·103

−116.68

15 5

QC (kJ)

30 l/h 60 l/h

-6

(b)

ΔHH (kJ/kg)*

A40

-3

10

QH (kJ)

* DSC data; calculated by the major peak; both peaks used for A44.

Heat Flow (mW)

0 at 26.9 °C is found. A twin peak was observed up on heating with maximum 32.7 °C. The small peaks recorded in DSC at lower temperatures associate to solid to solid phase transitions of a much lower thermal content. The calculated amounts of transferred heat through HTF are shown in Table 3. It is clear that there is a relationship between the latent heat value of each PCM and the amount of heat that is delivered through the HTF. This is a common feature observed for both 30 and 60 l/h flow rates with A44 having the highest power intensity among PCMs tested due to its higher latent heat absorbed or released in a narrow temperature window. The deviation from theoretical values may be related to thermal losses and during discharging to the hysteresis effect. Therefore it is clear that a careful tuning and operation control of LHTES is needed taking into account not only the super cooling effect of each PCM used but also the variations in the temperature window of each phase transition capable for thermal energy capture or release. In Table 3 the heat power that is stored QH and retrieved QC by the system through the HTF is calculated. Although latent heat shows minor deviations the QH/QC ratio is as high 3.3 for A40. Such deviation among charging and discharging Q values is expected due to the nonoptimized working temperature window of the LHTES tested according to the PCM super cooling features. Furthermore heat loss due to insulation issues cannot be excluded. These issues may not be identical up on heating and cooling as different heat transfer conditions may apply. Up on solidification a thin layer of PCM is formed on the HE fins and tubes. Thus convection is limited. Conduction is expected to be the main heat transfer mechanism making system much less responsive and sensitive to thermal gradients. This may explain also the different shape of the cooling part of the curve of Figs. 3–6 for all PCMs immediately after the liquid to solid transition initiates. Thus although response was fast on Regime Sa recording a fast decrease in all PCMs as soon as the solidification temperature point is reached (Regime Sm) and a PCM solid film forms around HE including the thermocouples, the system is less responsive.

-5 -10 3

(c)

0 -3 -6 3

(d)

0 -3 -6 0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 o

Temperature ( C) Fig. 7. DSC result of PCMs A40 (a), A44 (b), A46 (c) and A53 (d) on heating and cooling cycle in the range 5–70 °C under 0.8 °C/min,

observed releasing initially sensible heat (Regime Sa) and then latent heat when phase change takes place (Regime Sm). The region of phase transition from liquid to solid is less clear for A46 as well as closer to Regime Ma. However, similar to all previous PCMs super-cooling is found for A46 (Fig. 7). In A46 DSC cooling curve (Fig. 7) the latent heat emission occurs in the range of 30.6–41 °C as the major peak where latent heat emission begins (Regime Sm). The smaller peaks that are also recorded in DSC results at much lower temperatures associate to solid to solid phase transitions with much small thermal content. In Fig. 6 charging and discharging temperature are shown for A53. At the beginning of the charging energy accumulation in PCM is rapid (Regime Ma, Fig. 6) until A53 reaches liquid state. This region is not as evident as for the rest PCMs but for the 60 l/h flow rate Point M in Fig. 6 is identified at approximately 49 °C. In the respective A53 DSC heating curve (Fig. 7) he latent heat absorption ends at 57 °C. In discharging section a rapid reduction of PCM temperature is observed releasing initially sensible heat (Regime Sa) and then latent heat when liquid to solid phase change takes place (Regime Sm). Up on cooling the liquid to solid phase change area for A53 as it shown from DSC cooling curve (Fig. 7) exhibits a super cooling effect. It begins at about 50 °C and is completed at temperatures at about 36 °C (Regime Sm). A second exothermic peak of lower thermal content is observed with a maximum

4. Computational approach 4.1. Computational set-up description The heat exchanger in contact with PCM system was simulated in two dimensions (2D) using the ANSYS double precision Fluent CFD Solver [41]. The geometry of the computational domain is presented in Fig. 8. The problem is set as axisymmetric. Due to computational requirements, the length of the water tube was limited to 0.5 m. The water enters the tube with 0.25 m/s and at a constant temperature of 54 °C. The PCM fills the space between and over the aluminum fins, which have a length of 12.5 mm. The side and upper boundaries are considered adiabatic and the volume of the PCM is constant. The mesh was created by ANSYS grid generator and it consisted of 95,071 cells (Fig. 9). The analysis of the solid–liquid phase change is based on Voller’s 94


M.K. Koukou et al.

Fig. 8. Geometry of computational domain (boundary conditions included).

enthalpy-porosity technique [42]. The computational grid is fixed and does not change dynamically to follow the movement of the solid–liquid interface. So the melting and solidification model was enabled and the default mushy zone parameter value, M = 106 was selected. To illustrate the buoyancy effect on the convection heat transfer mechanism, the Boussinesq approximation was applied. With the Boussinesq approximation the density of the PCM can change with the temperature, but because the domain of the PCM does not change its volume, the computation expense is not prohibitive. Coupling of pressure with velocity was applied by the SIMPLE algorithm. Regarding gradients, least squares cell based option was chosen, spatial discretization of the pressure was chosen by PRESTO! scheme and regarding momentum and energy, the second order upwind was used. For the step discretization the first order implicit method was used. The latter method offers unconditionally stable solution with respect to the value of the time step. The set-up of the problem is presented in Table 4. PCM properties were those of A46 organic PCM.

Table 4 Solver settings and initial conditions. Solver Type Algorithm (Pressure-Velocity Coupling)

Pressure-Based SIMPLE

Spatial Discretization Gradient Pressure Momentum Energy Time Step Discretization

Least Squares Cell Based PRESTO! Second Order Upwind Second Order Upwind First Order Implicit

occurring as PCM liquefies and how it affects process time. Simulation runs were conducted for A46 melting process. In the first model only conduction was considered as the heat transfer mechanism whence at the second one, convection was allowed to occur simultaneously with conduction. Fig. 10 illustrates contours of liquid mass fraction as time progresses in both cases (on the left, convection occurs along with conduction and on the right, conduction is only considered). For the initial steps of the phenomenon (up to 250 s), liquid mass fraction is almost the same. As time progresses and liquid increases within the

4.2. Computational results and discussion Computational approach is used to verify the effect of convection

Fig. 9. Computational grid of the simulated heat exchanger. 95


M.K. Koukou et al.

Fig. 10. Computational contours of liquid mass fraction for 100–300–500–700 s during melting (left: conduction and convection – right: conduction).

96


M.K. Koukou et al.

Fig. 11. Time difference in liquid fraction evolution during melting (with and without convection).

Acknowledgements

tank, buoyancy currents result in convection that speeds up melting time. Fig. 11 illustrates the difference in melting time assuming pure conduction and conduction along with convection. When both heat transfer mechanisms are inserted to the simulation, melting time is reduced by 17% (1000 s–1200 s) indicating how the phenomenon speeds up.

This work is supported by TESSe2b project which has received funding from the European Union’s Horizon 2020 research and innovation programme under Grant agreement No. 680555. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at https://doi.org/10.1016/j.tsep.2018.05.004.

5. Conclusions

References

In this work, the performance of a developed LHTES unit using a staggered HE was studied using four different organic PCMs all suitable according to product information for a low temperature solar application. However, all the organic PCMs used showed a super cooling effect directly affecting the performance of LHTES mainly in terms of heat discharge. The LHTES unit using a staggered HE was an effective way for thermal energy storage successfully applied for the recovery and reuse of wasted heat although not optimized according to the properties of the thermal storage materials used. Furthermore as PCMs are substances of very low thermal conductivity the thermal transfer mechanisms are crucial, controlling the heat exchange process and eventually the heat storage performance of the LHTES. The larger the HTF flow rate of the HE the fastest both processes (melting and solidification) are, but as PCM liquefies, convection (via buoyancy currents) speeds up melting process. During solidification when a thin layer of PCM is formed on the HE fins and tubes, convection is limited and conduction is the main heat transfer mechanism. This indicates the need for special HE design approaches or materials solutions that should be followed in order to facilitate heat and mass transfer especially on the discharging mode. The TES unit studied can be considered as an effective solution with short response times up on charging. Discharging performance is strongly affected by the thermal transfer limitations as well as the super cooling of the PCMs. Such issues using PCM may be addressed either by optimised TES operation control or HE design.

[1] A. Abhat, Low temperature latent heat thermal energy storage: heat storage materials, Sol. Energy 30 (4) (1983) 313–332. [2] A.A. Robaidi, Development of novel polymer phase change material for heat storage application, Int. J. Mater. Sci. Appl. 2 (6) (2013) 168–172. [3] L. Liu, D. Su, Y. Tang, G. Fang, Thermal conductivity enhancement of phase change materials for thermal energy storage: a review, Renewable Sustainable Energy Rev. 62 (2016) 305–317. [4] V. Chalkia, N. Tachos, P.K. Pandis, A. Giannakas, M.K. Koukou, M.Gr. Vrachopoulos, L. Coelho, A. Coelho, V.N. Stathopoulos, Influence of organic phase change materials on the physical and mechanical properties of HDPE and PP polymers, RSC Adv. (2018) submitted. [5] H. Jouhara, A. Chauhan, T. Nannou, S. Almahmoud, B. Delpech, L.C. Wrobel, Heat pipe based systems – advances and applications, Energy 128 (2017) 729–754. [6] H. Mroue, J.B. Ramos, L.C. Wrobel, H. Jouhara, Experimental and numerical investigation of an air-to-water heat pipe-based heat exchanger, Appl. Therm. Eng. 78 (2015) 339–350. [7] I. Iliopoulos, A. Karampekios, P.K. Pandis, N. Vourdas, H. Jouhara, S. Tassou, V.N. Stathopoulos, Evaluation of organic coatings for corrosion protection of condensing economizers, Procedia Struct. Integrity (2018) submitted. [8] Z. Khan, Z. Khan, A. Ghafoor, A review of performance enhancement of PCM based latent heat storage system within the context of materials, thermal stability and compatibility, Energy Convers. Manage. 115 (2016) 132–158. [9] H.P. Garg, S.C. Mullick, A.K. Bhargava, Solar Thermal Energy Storage, Springer, Netherlands, 1985. [10] D.N. Nkwetta, F. Haghighat, Thermal energy storage with phase change materials. A state-of-the art review, Sustain Cities Soc 10 (2014) 87–100. [11] S. Seddegh, X. Wang, A.D. Henderson, Z. Xing, Solar domestic hot water systems using latent heat energy storage medium: a review, Renewable Sustainable Energy Rev. 49 (2015) 517–533. [12] E. Oro, A. de Gracia, A. Castell, M.M. Farid, L.F. Cabeza, Review on phase change materials (PCMs) for cold thermal energy storage applications, Appl. Energy 99

97


M.K. Koukou et al.

[28] S. Paria, S. Baradaran, Ahmad Amiri, A.A.D. Sarhan, S.N. Kazi, Performance evaluation of latent heat energy storage in horizontal shell-and-finned tube for solar application, J. Therm. Anal. Calorim. 123 (2016) 1371–1381. [29] Y. Zhang, A. Faghri, Heat transfer enhancement in latent heat thermal energy storage system by using the internally finned tube, Int. J. Heat Mass Transfer 39 (1996) 3165–3173, http://dx.doi.org/10.1016/0017-9310(95)00402-5. [30] I. Jamal, M. Baccar, Numerical study of PCM solidification in a finned tube thermal storage including natural convection, Appl. Therm. Eng. 84 (2015) 320–330, http://dx.doi.org/10.1016/j.applthermaleng.2015.03.065. [31] A. Amini, J. Miller, H. Jouhara, An investigation into the use of the heat pipe technology in thermal energy storage heat exchangers, Energy 136 (2017) 163–172. [32] P. Promoppatum, S. Yao, T. Hultz, D. Age, Experimental and numerical investigation of the cross-flow PCM heat exchanger for the energy saving of building HVAC, Energy Build. 138 (2017) 468–478. [33] A. Frazzica, V. Palomba, D.L. Rosa, V. Brancao, Experimental comparison of two heat exchanger concepts for latent heat storage applications, Energy Procedia 135 (2017) 183–192. [34] K. Merlin, D. Delaunay, J. Soto, L. Traonvouez, Heat transfer enhancement in latent heat thermal storage systems: comparative study of different solutions and thermal contact investigation between the exchanger and the PCM, Appl. Energy 166 (2016) 107–116. [35] A. Khalifa, L. Tan, A. Date, A. Akbarzadeh, A numerical and experimental study of solidification around axially finned heat pipes for high temperature latent heat thermal energy storage units, Appl. Therm. Eng. 70 (2014) 609–619, http://dx.doi. org/10.1016/j.applthermaleng.2014.05.080. [36] J. Gasia, J. Diriken, M. Bourke, J. Van Bael, L.F. Cabeza, Comparative study of the thermal performance of four different shell-and-tube heat exchangers used as latent heat thermal energy storage systems, Renewable Energy 114 (2017) 934–944. [37] http://www.pcmproducts.net. [38] T. Hatakeyama, Z. Liu, Handbook of Thermal Analysis, John Wiley and Sons. [39] V.N. Stathopoulos, A. Papandreou, D. Kanellopoulou, C.J. Stournaras, Structural ceramics containing electric arc furnace dust, J. Hazard. Mater. 262 (2013) 91–99. [40] M. Delgado, A. Lázaro, J. Mazo, B. Zalba, Review on phase change material emulsions and microencapsulated phase change material slurries: materials, heat transfer studies and applications, Renewable Sustainable Energy Rev. 16 (2012) 253–273. [41] ANSYS® Academic Research Mechanical, Release 18.1. [42] A.D. Brent, V.R. Voller, K.J. Reid, Enthalpy-porosity technique for modeling convection-diffusion phase change: application to the melting of a pure metal, Num. Heat Transfer 13 (3) (1988) 297–318, http://dx.doi.org/10.1080/ 1040778880891361.

(2012) 513–533. [13] D. Zhou, C.Y. Zhao, Y. Tian, Review on thermal energy storage with phase change materials (PCMs) in building applications, Appl. Energy 92 (2012) 593–605. [14] N. Shamsundar, R. Srinivasan, Analysis of energy storage by phase change with an array of cylindrical tubes, in: ASME Winter annu. Meet., USA: San Francisco; 1978, pp. 35–40. [15] P. Brousseau, M. Lacroix, Study of the thermal performance of a multi-layer PCM storage unit, Energy Convers. Manage. 37 (1996) 599–609. [16] J. Banaszek, R. Domanski, M. Rebow, F. El-Sagier, Experimental study of solidliquid phase change in a spiral thermal energy storage unit, Appl. Therm. Eng. 19 (1999) 1253–1277, http://dx.doi.org/10.1016/S1359-4311(98)00120-3. [17] J. Banaszek, R. Domanski, M. Rebow, F. El-Sagier, Numerical analysis of the paraffin wax-air spiral thermal energy storage unit, Appl. Therm. Eng. 20 (2000) 323–354. [18] K.A.R. Ismail, J.R. Henrıquez, Numerical and experimental study of spherical capsules packed bed latent heat storage system, Appl. Therm. Eng. 22 (2002) 1705–1716. [19] M. Lacroix, Study of the heat transfer behavior of a latent heat thermal energy storage unit with a finned tube, Int. J. Heat Mass Transfer 36 (8) (1993) 2083–2092. [20] F. Agyenim, P. Eames, M. Smyth, A comparison of heat transfer enhancement in a medium temperature thermal energy storage heat exchanger using fins, Sol. Energy 83 (9) (2009) 1509–1520. [21] A. Castell, M. Belusko, F. Bruno, L.F. Cabeza, Maximisation of heat transfer in a coil in tank PCM cold storage system, Appl. Energy 88 (11) (2011) 4120–4127. [22] M. Medrano, M.O. Yilmaz, M. Nogués, I. Martorell, J. Roca, L.F. Cabeza, Experimental evaluation of commercial heat exchangers for use as PCM thermal storage systems, Appl. Energy 86 (10) (2009) 2047–2055. [23] N.H.S. Tay, M. Belusko, A. Castell, L.F. Cabeza, F. Bruno, An effectiveness-NTU technique for characterising a finned tubes PCM system using a CFD model, Appl. Energy 131 (2014) 377–385. [24] N.I. Ibrahim, F.A. Al-Sulaimana, S. Rahmana, B.S. Yilbas, A.Z. Sahin, Heat transfer enhancement of phase change materials for thermal energy storage applications: a critical review, Renewable Sustainable Energy Rev. 74 (2017) 26–50. [25] F. Agyenim, P. Eames, M. Smyth, Heat transfer enhancement in medium temperature thermal energy storage system using a multitube heat transfer array, Renewable Energy 35 (2010) 198–207. [26] E. Oró, A. Gil, L. Miró, G. Peiró, Thermal energy storage implementation using phase change materials for solar cooling and refrigeration applications, Energy Procedia 30 (2012) 947–956. [27] M.K. Rathod, J. Banerjee, Thermal performance enhancement of shell and tube latent heat storage unit using longitudinal fins, Appl. Therm. Eng. 75 (2015) 1084–1092.

98