It can be concluded that most high-temperature macro encapsulations are fabricated using a metallic shell mostly made of steel, while micro encapsulations are mainly fabricated using plastic shells due to easy in production, which can be operated at low temperature. However, in a corrosive environment, the metallic encapsulation cannot be used for a long duration of operation of the latent heat thermal storage system. Also, the more flexible encapsulation material is required to accommodate the volume change and generated vapor pressure of PCM (phase change material) during solid to liquid phase change, which is linked to the stability of the capsules. In this context, a polymer can be used as the encapsulation material, although these materials are not widely available and well-characterized at high temperature (>200 C).
This will also help in reducing the weight of the storage tank, which would result in capital cost. Hence, in this study, the feasibility of using polymeric materials at high temperature for solar applications is investigated. High-performance polymers PTFE, PEEK and PEKK are chosen as encapsulation materials for medium temperature (~200 C) solar thermal energy storage. In order to verify the use of PTFE, PEEK and PEKK, these polymers are taken through accelerated thermal cycles and then tested for thermo-mechanical properties to understand the degradation in the encapsulation material with thermal cycle. The findings from the present study will provide useful information and direction on the further development of the polymers as encapsulation materials for storing PCM for medium temperature solar applications.
Three different types of high-temperature polymers are used and tested for various parameters to understand their use as encapsulation materials for medium temperature solar thermal energy storage (between 200 and 300 C). Polytetrafluoroethylene (PTFE, trade name Teflon), Polyetheretherketone (PEEK) and Polyetherketoneketone (PEKK) of Ghardha Chemicals Pvt. Ltd. make are chosen as the encapsulation materials in this study. Appropriate sealants are used to leak-proof the capsule while testing. Table 2 shows the physical properties of polymers chosen for the study [43]. It can be noted that the density of PTFE is higher than that of PEEK and PEKK. PEKK shows higher glass transition temperature due to the presence of an extra-stiff ketone group in its structure, while PEEK has a tendency to rotate about the ether bond. All the three polymers are thermoplastic; therefore, operation at very high temperature might lead to a change in its shape. A commercially available, organic material, A164 (make: Phase Change Material Products Limited, UK) is used as the phase change material (PCM). The melting temperature of the PCM is found as 164 C.
Fig. 1 displays the front and side view of a hollow spherical capsule fabricated from PTFE, PEEK and PEKK rods using CNC machining. It shows the positive-negative halves of the capsule. The region projecting out in the positive half fits perfectly in the slot available in the negative half, restricting the leakage of liquid PCM. This positive-negative interlocking is one of the crucial steps to stop the leakage of liquid PCM from the capsule at elevated temperatures. The two halves are joined together with stainless steel (SS) screws after filling with the PCM. The volume of PCM filled is equal to the 90% of the capsule volume to allow expansion and to accommodate vapor pressure during solid to liquid phase change.
The thickness of spherical capsules is calculated using equation (1) as, s F ¼ Pr 2tcp where P is the design pressure applied on the outer surface of the capsule, F is the factor of safety, s is the compressive yield stress, and r and tf depict the outer radius and thickness of the capsule, respectively. In the calculation of the thickness of spherical capsules, the following assumptions are made, (i) pressure in the storage tank is chosen as 200 psi, which corresponds to the pressure due to flowing water for optimal operations in the turbine operated low-temperature thermal power plant, (ii) compressive yield stress of PTFE at 250 C is taken as 3.4 MPa [43], (iii) vapor pressure of PCM and joint efficiency factor are neglected and (iv) factor of safety is considered as 2 due to the possibility of localized heating near the sealing material. Thus, the thickness of the PTFE spherical capsule of r ¼ 15 mm is calculated as 6.08 mm, which is taken as 6 mm in the fabrication process. It can be noted that the thickness of the spherical capsule should be calculated based on the operating conditions of the location where the thermal energy storage system will be installed in a solar thermal power plant.
In order to understand the degree of degradation of polymer with the thermal cycles, as well as, at elevated temperatures for a long duration of operation, TGA is used. Dried samples, typically 5e25 mg of polymer is loaded into platinum pans and then heated up based on the required thermal cycle. The TGA/DTA analysis is conducted on a Diamond Thermogravimetric/Differential Thermal Analyzer by PERKIN ELMER, USA. The samples, collected after every 10 thermal cycles, are heated from room temperature to 750 C. All the three materials are also heated for 10e12 h at 320 and 340 C to gauge the degradation of material with time.
Tensile/compression testing is used to understand the change in mechanical properties of the polymeric materials after being exposed to a thermal cycle. The tensile/compression testing is performed on Tinius Olsen H25KS. The strain rate in case of tensile test is 5 mm/min, and the sample is allowed to reach the breakpoint, while for compression testing, the strain rate is 0.5 mm/ min and the sample is compressed 30e35% of its original length. For these testing, dog bone tensile sample and cylindrical compression samples are machined from 10 mm diameter rods of PTFE, PEEK and PEKK. The samples are manufactured using rotating Lathe machine.
T-type thermocouples (make: T.C. Ltd., temperature range: 0e350 C ± 0.2 C) are inserted at various locations of the two identical capsules of the same material to understand the change in temperature at the selected points. Fig. 2 illustrates the location of thermocouples with the nomenclature. The thermocouples are calibrated with a constant temperature bath before the experiments. A M2700 Keithley data logger is used for recording the temperature data at a time interval of 2 s. A resistive furnace with a temperature range of 30e1500 C is used for thermal profiling of the capsules. For uniform heating and continuous maintenance of temperature, the furnace is set at 20 C more than the required temperature to compensate the high loss of thermal energy during removal and placement of the capsules inside the furnace. The furnace is heated to the required temperature at a rate of 20 C/min until the required temperature is achieved and is held isothermally.
Two identical capsules of all the three materials are undergone through thermal cycles of 10 min of heating and 20 min of cooling. The heating temperature is decided according to the degradation temperature of the material and is listed later, while the capsules are cooled in the air using natural convection. The time for cooling is taken more as the thermal conductivity of PCM is low, which causes slow dissipation of heat. As shown in Fig. 2, a small hole of 1 mm is drilled in capsule 2 for inserting the thermocouple at the center and half-center of the capsule, which is then sealed with the help of two different industrial grade sealants for PEEK and PEKK. TC 3, TC 4 and TC 5 in Fig. 2 are named as Open1, Open2 and Open3, respectively.
Tensile dog-bone samples and cylindrical compression samples are also taken through a similar thermal cycle and are then tested. The capsules are placed at the center of the furnace for uniform heating. The time required for moving the capsules in and out of the furnace is minimized to reduce the loss of heat from the furnace. Thermocouples are placed at various locations on the capsules to understand the difference in heating, while also recording the maximum surface temperature of the capsule. The temperature is monitored continuously to ensure that it does not cross the degradation temperature of the polymer and lead to failure of the capsule.
In this study, accelerated thermal cycling of the three materials PTFE, PEEK, PEKK are conducted to optimize the material selection and design improvements. Samples are collected after every 10 cycles and are characterized for various parameters, such as weight, dimensional uniformity. The samples after every 10 cycles are also characterized using TGA, DTA, tensile and compression testing to characterize their mechanical properties under the thermal cycle. The behavior of encapsulated materials in solar thermal energy storage plant can be understood by accelerated thermal cycle, in which each material is tested for thirty thermal cycles of 10 min of heating, followed by 20 min of cooling in a furnace. The operating temperatures of the three materials, as shown in Table 3, is chosen near the degradation temperature of the three materials to understand their behavior in the extreme conditions.
Fig. 3 shows the temperature at different locations of the encapsulated PCMs with thermal cycles for PTFE at 245 C in the furnace. The temperature of the PCM inside the PTFE capsule stabilizes at 164 C for both the thermocouples (center, half-center) around 13000 s, indicating the initiation of melting of PCM in the encapsulation, however no complete melting of PCM is observed, which is due to the low thermal conductivity of PTFE (~0.39 W/m.K) and high thermal resistance to heat transfer from the capsule outer surface to the inner surface because of encapsulation thickness of 6 mm. Hence, it can be concluded that the thickness of the PTFE capsule should be chosen carefully to obtain the effect of latent heat of fusion of PCM on the stabilization of temperature for a given thermal cycle. The maximum surface temperature reaches 215 C after 160 min. In Fig. 3(d), the heating and cooling periods of a thermal cycle are increased to 20 min and 40 min respectively. It can be noted that the PCM reaches its melting temperature at 15 min, while in the cooling period, the time to reach the phase change temperature is 5 min. In this case, identifiable nonuniformity in shape and dimensions is not observed. Fig. 4 shows the capsules made of PTFE before and after every 10 thermal cycles. The initial diameter of the PTFE capsule is measured as 29.55 ± 0.1 mm (Fig. 4(a)) with encapsulation thickness 6 ± 0.15 mm.
It can be observed that the PTFE capsules retain their shape and structure, even after being subjected to a furnace temperature of 245 C and the maximum surface temperature of 220 C, showing that it can be used as an encapsulation material for solar thermal energy storage up to a temperature range of 215 C shows the PCM and capsules after 30 thermal cycles. No visual sign of non-uniformity in its shape and structure is evident in the PTFE capsule, which is confirmed by the measurements of the dimensions of the capsules. The absence of any leakage of the PCM also confirms dimensional and structural retention of the PTFE capsule during the thermal cycling. The positive-negative interlock that is carved in the spherical capsule (Fig. 1) ensures leak proof. This can be attributed to the glassy to rubbery state transition of PTFE at elevated temperatures due to its lower glass transition temperature, thereby sealing the PCM inside the capsule.
Fig. 5 shows the temperature profile at different locations of the encapsulated PEEK capsule for a furnace temperature of 265 C. The PCM reaches its melting temperature of 164 C at ~100 min (not shown in the figure). Although complete melting of PCM is not observed, it is apparent from the figure that the melt fraction of PCM (i.e. stabilization time of temperature during solidification of PCM) is higher than that in case of the PTFE capsule due to higher surface temperature. Initially, extensive leakage is observed in the PEEK capsule, which can be attributed to its rigid crystalline structure than PTFE, and the leakage is reduced using commercially available sealant along the positive-negative interlock. No abnormal behavior of the thermocouple is observed for 1e20 thermal cycles with PEEK. In Fig. 5, a rapid change in temperature of the thermocouple at half-centers is observed after 810 min suggesting that the thermocouple lost contact with the PCM and started showing the temperature of the air inside the encapsulation capsule.
it shows the condition of PEEK capsule at 0, 10, 20 and 30 thermal cycles. Slight degradation of sealant on the outer surface of the PEEK capsule can be observed. Extensive black spot at point 1 can be observed because of the localized heating due to screws, as the screws closer to the heating rods of the furnace absorb heat faster and temperature rises more than the surface temperature of capsule exposed to air suggesting that the use of screws for higher temperature application should be avoided and alternative locking mechanism can be adopted. The effect of sealant can be seen. The amount of degradation in region 2 is much lower than that in region 1 because of the insulating effect of the sealant. it shows the inner section of the PEEK capsule, which also turned black in the regions, where sealant was applied. The difference in the amount of black substance in regions 1 and 2 of the figure indicates that the black residue originates from the degradation of the sealant rather than the PEEK. Better and improved sealing techniques like ultrasonic welding can be used so that the effect/degradation of sealant can be reduced.
Region 3 in Fig. 6(e) shows that a small portion PCM adhering to the capsule wall is also not able to be to withstand the high temperature and turned black depicting its degradation (maximum operating temperature of PCM ¼ 280 C). Fig. 7 shows the thermal cycle of PEKK capsule of thickness 2 mm at the furnace temperature of 290 C. PEKK also being a rigid polymer shows leakage without the use of sealant. Therefore, a sealant is used to provide a leak-proof system. The temperature measured by the thermocouples inside the encapsulated capsule (Centre, Half-Centre) in Fig. 7(aec) depicts a slow rise in PCM temperature in the first thermal cycle, as PCM reaches to its melting temperature. In this case, the stabilization period is high compared to that of PTFE and PEEK, indicating high melt fraction of PCM. It can be noted in Fig. 7 that the PCM temperature inside the PEKK capsule during the cooling period decreases below its melting temperature (~155 C) before stabilizes at the melting temperature (164 C) of the PCM, which is known as sub-cooling of PCM depending on the cooling rates.
Fig. 8 shows the PEKK capsule before and after every 10 thermal cycles. Fig. 8(aed) shows no visual change in the shape and structure of PEKK capsule and further investigation suggest that there is no change in thermo-mechanical properties PEKK as well (vide infra). Fig. 8(e) shows the inner cross section of the PEKK capsule. The blackening of PCM near the surface of the capsule suggests that the PCM has reached its degradation temperature, which is due to the two reasons, viz. (i) high furnace temperature, therefore leading to high surface temperature, (ii) less shell thickness leading faster and more heat transfer from air to PCM. Hence, the operating temperature of the capsule is required to be selected properly to prevent degradation of PCM during operation.
Fig. 9 shows the thermogravimetric analysis of PTFE, PEEK, PEKK kept for 12 h at 320 and 340 C. The figure shows that all the materials do not undergo substantial weight loss even after being exposed to the high temperature for a long duration. The maximum weight loss is found to be 0.6% for PTFE at 340 C after 1000 min while 1.4% in case of PEEK. This factor of stability at elevated temperature proves that these materials can be used for the applications at these temperatures; however the change in mechanical properties is to be studied to understand if the encapsulation can withstand the vapor pressure generated inside the capsule during the melting of PCM and also the weight of capsules above it in the container, which will be discussed later.
Fig. 10 shows the TGA graphs for three materials tested after every 10 cycles from room temperature (~30 C) to 750 C. The TGA graphs for PTFE, as shown in Fig. 10(a), depicts a very small deviation indicating that there is a very less change in the polymer morphology. While in the case of PEEK and PEKK, the onset temperature of degradation increases after the first 10 thermal cycles and remains constant thereafter. This is due to the release of residual stresses from the material during the thermal cycle. Both PEEK and PEKK rods, which are used to fabricate the spherical capsule are extruded under very high pressure leading to the development of residual stresses in the material, which will be validated with the results later. It can be seen from the TGA graphs (Fig. 10(b) and (c)) that the degradation of PEEK and PEKK is a twostep process.
This is because of the presence of two different functional groups that is ether and ketone in PEEK and PEKK. The first step of thermal decomposition is attributed to the random chain scission of the ether and ketone bonds, while the second decomposition is due to the cracking and dehydrogenation of the cross-linked residue produced in the first stage of decomposition, which results in thermally stable carbonaceous char. The amount of residue depends on the type of atmosphere used during the experiment. The degradation temperature of PTFE is always lower than that of PEEK and PEKK. The presence of residue in case of PEKK confirms the presence of carbon fiber in the polymer, which is added to increase its strength.
3.3. Differential thermal analysis (DTA) Fig. 11 shows the DTA profile of PTFE, PEEK and PEKK at zeroth, and after 10th thermal cycles. A slight drop in the DTA curve suggests the melting of the polymer. Therefore, it can be seen that PTFE melts at 369 C (Fig. 11(a)), while PEEK and PEKK start melting at 378 C (Fig. 11(b)) and 395 C (Fig. 11(c)), respectively. Two separate peaks are observed during the degradation of PEEK and PEKK, which suggests that the degradation of PEEK and PEKK is a two-step process, which is explained earlier. 3.4. Compression testing the compression tests of PTFE, as shown in Fig. 12, show a continuous decrease in the amount of stress required for 30% strain, suggesting that PTFE (Fig. 12(a)) undergoes loss of mechanical strength with the accelerated thermal cycle. While in case of PEEK and PEKK (Fig. 12 (b, c)), an increase in the compressive strength can be seen due to the removal of residual stresses generated in the material during the manufacturing of capsules. Graphs for PEEK and PEKK show a minimal change with progressive thermal cycles, therefore reaching a stable value, which can be used to optimize the capsule thickness. Hence, it is essential to remove the residual stress from the PEEK and PEKK rods by the application of high temperature and then normalizing to obtain constant and improved properties. The compressive strength of PEEK and PEKK is three times more than that of PTFE; therefore, the thickness of the capsule can be reduced for a given pressure. This leads to a reduction in the quantity of required material, resulting in a reduction in the cost. The use of high strength material will help in reducing the volume of the thermal storage system, as well as the weight of the capsules. Due to the presence of carbon fiber in PEKK, the compressive strength is always more than that of PEEK, though the difference is small.
3.5. Tensile testing Fig. 13 shows the tensile testing of three materials, PTFE, PEEK and PEKK. Tensile testing of PTFE (Fig. 13(a)) shows a similar curve as the compression testing. A continuous decrease in the tensile strength of PTFE with the thermal cycle can be observed. In the case of PEEK (Fig. 13(b)), the maximum strain at the point of fracture reduces, which happens due to the increase in cross-linking within the PEEK polymers, therefore making the material more rigid and brittle. For PEKK (Fig. 13(c)), the tensile strength improves with thermal cycle. From the stress-strain relation, Young’s modulus is calculated. It can be noted that the variation in the strain stress of polymer materials is obtained after reaching the encapsulated PCM at the room temperature. It is found that after 0, 10, 20 and 30 cycles for PTFE, Young’s modulus is 0.1833, 0.1696, 0.1662, 0.1600 GPa, respectively. Similarly, for PEEK and PEKK, Young’s modulus is determined as 1.658, 1.714, 1.869, 1.916 GPa and 1.506, 1.757, 2.025, 1.657 GPa, respectively after 0, 10, 20, 30 thermal cycles. As discussed earlier in the case of PEEK and PEKK, the strength of the material increases because of the release of the residual stress during fabrication.
3.6.1. Compression testing on the different samples of the same material Fig. 14(aec) shows the compression testing of polymers on different samples of the same material at different temperatures. It can be observed that the elasticity behavior of PTFE, PEEK and PEKK decreases with increasing temperature. The elastic behavior for PEEK at 140, 200, 250 C can be observed up to the strain of 4.18%, 3.97%, 2.84% respectively, followed by a plateau which indicates the plastic behavior for the same material. The PEKK also shows a similar trend of decreasing strain with increasing temperature. For temperatures of 165, 225 and 275 C, the strains up to which elastic behavior is found to be 6.52%, 5.45%, 3.56%, respectively. In the case of PTFE, the elastic behavior exists at percentage strain of 3.71%, 2.98%, 1.97% for 120, 180 and 280 C, respectively. Fig. 14(d) shows the variation of Young’s modulus with temperature for PTFE, PEEK, PEKK. The Young’s modulus decreases with increasing temperature for three polymers, which indicates that as the temperature increases, the mechanical properties of the polymer starts to deteriorate. The rate of decrease in Young’s modulus for PTFE is higher as compared to than for PEEK and PEKK. The PEEK and PEKK have similar values of Young’s modulus at 140, and 160 C as the rate of decrease in Young’s modulus for PEEK is faster than that for PEKK with increasing temperature due to stronger two ketone bonds in PEKK compared to two ether bonds in PEEK. Therefore, it can be concluded that PEKK is thermally and mechanically stable as compared to PEEK and PTFE at a relatively higher temperature. The PEEK can be used in the range of room temperature to 250 C while in the case of PEKK this range is wider that is up to 275 C. This set of data can be useful in the application of thermal diffusion where there is a constant temperature value or a small range of temperature.
3.6.2. Compression testing on the same sample of the same material In order to complement and substantiate the understanding developed during the tensile testing, compression testing is performed on the same samples. As the glass transition temperature of PTFE is 115 C, three experiments are carried out at 120, 180 and 250 C. Similar experiments are performed for PEEK and PEKK having glass transition temperature of 143 and 175 C, respectively. Fig. 15(aec) shows the variation of stress with strain for PTFE, PEEK, and PEKK at different temperatures on the same sample. From Fig. 15(a), the elastic behavior is observed at temperatures of 120, 180 and 250 C up to the strain of 0.12%, 0.25% and 0.45% followed by the plastic behavior. This indicates that the elasticity of PTFE increases with temperature. Similarly, the elasticity behavior of PEKK is observed for the strain of 5%, 2.4% and 2.2% for 165, 225 and 275 C, whereas, in case of PEEK, the strain is 4.7%, 2.3% and 0.25% respectively, for 140, 200 and 250 C. Due to the higher glass transition temperature, the PEKK shows the elastic behavior at a higher temperature compared to PEEK and PTFE. The percentage strain range for all three temperatures is almost comparable for PEKK, while in case of PEEK, the strain range is small (~22.5%) at 250 C, indicating that the polymer starts to deteriorate from 250 C. The compression testing on the same sample of PTFE, PEEK and PEKK reveals that the plastic and elastic behavior can not be distinguished at a temperature of 250 C. Fig. 15(d) depicts the variation of Young’s modulus with temperature. It can be observed that with increasing temperature, the strength of the material decreases for all the polymers. Young’s modulus for PTFE at 120 C is 0.276 GPa, which becomes 0.05 at 250 C. Whereas, Young’s modulus is found as 2 and 0.1 GPa at 140 and 250 C, respectively for PEEK and 1.89 and 0.4 GPa at 165 and 275 C for PEKK. In comparison with PEKK, PEEK has a higher value of Young’s modulus indicating that PEEK can remain stable at the higher temperature and can be used for higher temperature applications. As the temperature increases, all the polymeric materials show more plastic behavior, which indicates that the material becomes brittle. Therefore, it can be concluded that the temperature range that is safe for the continuous use of applications is 0e120 C for PTFE. For PEEK and PEKK, these can sustain up to the temperature of 250 and 275 C, respectively
3.7. Dimensional analysis the dimensional analysis of the flange diameter and the outer diameter of the capsule, which shows a slight change in their shapes. The diameter slightly increases with the thermal cycle, which is because of the generation of the vapor pressure of PCM inside the capsule. Capsule 1 of PEEK shows the least increase (~0.2%) in its diameter while capsule 1 of PTFE shows the maximum increase (~0.32%). The percentage increase in diameter of PEKK is almost comparable to that of PEEK, which suggests that PEEK and PEKK are more suited for use as an encapsulation material for medium temperature solar thermal energy storage than PTFE.
4. Conclusions In this work, the suitability of PTFE, PEEK and PEKK for thermal energy storage is studied based on the several parameters. The dimensional and weight analyses of PTFE, PEEK and PEKK shows no sign of degradation. It is found that PTFE shows a decline in mechanical properties while there is an improvement in mechanical properties after first 10 cycles for PEEK and PEKK due to the removal of residual stresses developed during extrusion and machining of the parts. Therefore, PEEK and PEKK should be preheated before the use for best results from encapsulation. TGA curves for all the material reveals similar results proving that there is an insignificant change in the structure and arrangement of the molecules in the polymer. All three encapsulation materials can withstand 30 accelerated thermal cycles near their degradation temperatures with no leakage. Thermal analysis at 320 and 340 C for 10e12 h shows a weight loss of up to 3 wt%, indicating that polytetrafluoroethylene (PTFE), polyether ether ketone (PEEK) and polyether ketone ketone (PEKK) can withstand elevated temperature for a long duration which is essential for medium temperature solar applications. The compression testing shows that PEEK and PEKK can be used encapsulation of phase change materials with melting temperature in the range of 200e250 C as these polymers show Young’s modulus of 2 and 0.1 GPa at 140 and 250 C respectively for PEEK and 1.89 and 0.4 GPa at 165 and 275 C for PEKK. Hence, it can be concluded that the use of high-temperature polymeric material as microencapsulation for PCM in medium temperature solar thermal energy storage is a viable option.