honeycomb ceramic composite materials for solar thermal storage applications: Preparation and stability evaluation

Author’s Accepted Manuscript High-temperature alloy/honeycomb ceramic composite materials for solar thermal storage applications: preparation and stability evaluation Xinbin Lao, Xiaohong Xu, Jianfeng Wu, Xiaoyang Xu www.elsevier.com/locate/ceri

PII: DOI: Reference:

S0272-8842(16)32386-0 http://dx.doi.org/10.1016/j.ceramint.2016.12.119 CERI14429

To appear in: Ceramics International Received date: 23 October 2016 Revised date: 11 December 2016 Accepted date: 23 December 2016 Cite this article as: Xinbin Lao, Xiaohong Xu, Jianfeng Wu and Xiaoyang Xu, High-temperature alloy/honeycomb ceramic composite materials for solar thermal storage applications: preparation and stability evaluation, Ceramics International, http://dx.doi.org/10.1016/j.ceramint.2016.12.119 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 galley proof before it is published in its final citable 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.

High-temperature alloy/honeycomb ceramic composite materials for solar thermal storage applications: preparation and stability evaluation Xinbin Lao1*, Xiaohong Xu2, Jianfeng Wu2, Xiaoyang Xu1 1

National Engineering Research Center for Domestic and Building Ceramics, Jingdezhen Ceramic Institute, Jingdezhen 333000, P.R. China; 2

State Key Laboratory of Silicate Materials for Architectures, Wuhan University of Technology, Wuhan 430070, P. R. China

Abstract: SiCw/Al2O3 honeycomb ceramics were engaged as sensible shell materials for encapsulating Al-Si alloys (latent heat materials) in the honeycomb holes to obtain alloy/ceramic composite materials with a high thermal storage capacity for high-temperature solar thermal storage applications. The stability evaluation between the sensible honeycomb ceramics and the latent alloys had been conducted and the failure mechanism for the latent alloys was investigated. Results indicated that the addition of the latent alloys could improve the thermal storage capacity of the sensible honeycomb ceramics significantly by >114% and the thermal storage densities of honeycombs containing Al-12Si and Al-20Si alloys were 1141.3 kJ/kg and 1106 kJ/kg (400-900 °C), respectively. The composite materials exhibited excellent physical and chemical stability. No cracks formed in the honeycomb ceramics and no leakage of alloys was discovered after the composite materials were exposed to 100 thermal cycles in a high-temperature testing environment. The oxidation of Al at >600 °C would lower the latent heat of alloys and the thermal storage densities decreased to 1039.9 kJ/kg and 1013.2 kJ/kg after enduring 100 thermal cycles. This study 1

not only provides a sensible-latent system of thermal storage materials with excellent stability but also gives an insight into the protection of metal containers against the corrosion from Al-based alloys. Keywords: alloy/ceramic composites; solar thermal storage; SiCw/Al2O3 honeycomb ceramics; thermal storage capacity; stability evaluation; failure mechanism;

1. Introduction

Among several types of renewable energy, solar energy is the largest source, whose efficient utilization can relieve greenhouse gas emissions and the shortage of fossil energy [1,2]. In the past decades, solar energy had been comprehensively used in the thermal form in concentrated solar thermal power plants, since the conversion of solar energy into thermal energy became the easiest and the most extensively accepted method [3]. One difficulty of using solar energy is its intermittent nature as it is not available 24 hours per day. For removing the fluctuations caused by the intermittent nature of solar energy, thermal energy storage systems using numerous materials as thermal storage media, which are kept in containers (generally tank made by steel), are demanded to ensure the continuous power supply when the solar irradiation is not favorable [4,5]. During thermal storage process, heat absorbed by a solar collector is transferred to flowing air by convection, which is heated up to 800-1000 °C and then transfers the stored heat to thermal storage systems. There are mainly three types of thermal storage systems, namely sensible storage system, latent storage system and thermo-chemical storage system, which are composed of sensible materials, latent materials (i.e. phase change materials, PCMs) and thermo-chemical storage materials, respectively [6]. Among them, the development level of thermo-chemical 2

storage technology is at an experimental stage, thus only the sensible and the latent ones have been applied in the commercial power plants. In the case of sensible thermal storage systems for high-temperature solar applications (400-900 °C), the amount of energy that is stored using air as a heat transfer medium depends on the temperature change of the sensible solid materials (such as rocks, pebbles, concrete, bricks and ceramics) and can be expressed in equation (1) [7,8] T2

E  m  C p dT

(1)

T1

Here m is the mass and Cp is the specific heat capacity of sensible materials. T1 and T2 represent the lower and higher temperatures, respectively. Sensible thermal storage is the most inexpensive way, which is simpler in the design of thermal storage systems than latent thermal storage as well [9]. However, due to the low specific heat capacity of the sensible materials, sensible thermal storage typically requires a container with larger volume. The other disadvantages associated with the sensible thermal storage systems are the low thermal conductivity and the mechanical degradation during thermal cycling. For instances, the thermal conductivity values of pebbles and rocks are around 1 W·m-1·K-1, which will prolong the period of thermal storage-retrieval process; the compressive strength of concrete decreases about 20% at 400 °C due to the reactions, phase transformations and the thermal expansion mismatch of the raw materials [10]. Consequently, ceramics such as alumina (Al2O3) [11], silicon carbide (SiC) [12,13], zirconia (ZrO2) [14], are found to be highly suitable for sensible thermal storage for the duty considered (400-900 °C) due to the relatively high specific heat capacity, the high refractoriness, the high thermal conductivity as well as the good thermal stability. In general, thermal storage ceramics are used in the form of packed-bed containing pellets or honeycombs which have also wide applications in industrial 3

waste heat recovery and thermal oxidation processes [15]. Unlike pellets, honeycombs do not suffer from the clogging problem caused by the containing impurities due to their particular porous structure. Furthermore, honeycombs not only have the constant thermal storage capacity, but also possess a 4-5 times higher specific surface area, a lower pressure loss and a better heat transfer ability as compared with pellets [16]. In terms of PCM encapsulations, the honeycomb structure is also advantageous to retain PCMs in the holes for fabricating sensible-latent composite materials with a higher thermal storage capacity. However, the crucial aspects about the stability of the honeycomb ceramic shell encapsulated PCMs were largely ignored, although many works concerned with the stability of ceramic pellets encapsulated with PCMs (i.e. PCMs/packed-bed system) had been published. Chen et al. [17] had studied the thermal stability of pellets that microencapsulated with paraffin using silicon dioxide (SiO2) shells by a sol-gel method for low temperature applications. Results indicated that SiO2 shells could improve the thermal stability of composite thermal storage materials. Jalalzadeh-Azar et al. [14] had also evaluated the stability of PCM pellets (composed by 40 wt% Na2SO4 and 60 wt% SiO2) for high-temperature thermal storage applications. The latent core material as dodecanol was encapsulated with SiO2 based on suspension polymerization method by Yin et al. [18] and the thermal stability of the composite pellets was examined. PCMs not only store solar energy by their inherent sensible heat capacity, but also store or release a large amount of latent heat during their phase transitions (melting or solidification), making them more superior than the sensible ceramics in terms of thermal storage capacity [19, 20]. The energy stored by PCMs depends on their mass (m, g), latent heat of fusion (λ, J/g) and sensible heat capacity, which can be separated into the heat capacity of solid (Cp,s, J·g-1·K-1) and liquid (Cp,l, J·g-1·K-1) phases. Thus, 4

E  m  ( 

Tmelt

C

T1

T2

p,s

dT   C p ,l dT )

(2)

Tmelt

Tmelt represents the melting points of PCMs. PCMs have extensive applications in industrial waste heat recovery [21] and solar power plants [22], whilst many types of PCMs with low cost have been developed, including paraffin waxes [23], fatty acid [24] and inorganic molten salts [25], etc. However, paraffin waxes and fatty acid suffer from the poor heat transfer ability as their thermal conductivity values are generally in the range of 0.2-0.5 W/(m·K), thus the heat transfer enhancement technologies for these PCMs have been widely investigated [26-28]. Molten salts for high-temperature applications are characterized by the poor physical and chemical stability and exhibit a great corrosive character for container. The encapsulation of salt melts is also difficult because of the low viscosity, which results in the considerable weight loss during thermal cycling [29]. As stated by Cheng et al [30], Al-based alloys with great potentials, such as Al-Si alloy, Al-Si-Cu alloy, Al-Si-Mg alloy, show a better thermal stability, a higher thermal conductivity (110-116 W·m-1·K-1) and a higher thermal storage capacity than the above mentioned PCMs. Thus Al-based alloys become promising for enhancing heat transfer performances and stability of the latent thermal storage systems. Nevertheless, Al-based alloys are also corrosive for containers made by steel as they have a high chemical reactivity with the surrounding Fe. During working at high temperatures, Al and Fe can interact to form FeAl3 and Fe2Al3 phases, resulting in the corrosion of the containers and the reduction of heat capacity [31]. Thus, there is an urgent need to develop technologies for protecting container from interacting with the latent alloys [32]. In this study, the concept about encapsulating the latent Al-based alloys in the holes of the sensible honeycomb ceramics to fabricate sensible-latent composite thermal storage materials had been put forward, whilst alloys/honeycomb composite materials were prepared using SiC w/Al2O3 5

honeycomb ceramics as the shell materials and Al-Si alloys as PCMs. As the physical and chemical stability of the composite materials was greatly concerned, stability evaluation of the composite materials during high temperature thermal cycling was conducted. This study gives an insight into the preparation of sensible-latent composite thermal storage materials composed of honeycomb ceramics and Al-Si alloys, which can not only increase the thermal storage capacity of sensible honeycomb ceramics significantly, but also provide a new method for the protection against the corrosion of Al-based alloys.

2. Materials and methods 2.1 Honeycomb ceramics and alloys chosen SiCw/Al2O3 honeycomb ceramics with hexagonal holes used as shell materials were a kind of Al2O3-based composite ceramics. They were the laboratory products from the previously published work [33], which were characterized by the high refractoriness (>1400 °C), the high thermal conductivity (8.96~4.06 W·m-1·K-1, room temperature~600 °C), the high specific heat capacity (0.82~1.09 J·g-1·K-1, room temperature~600 °C), the good thermal shock resistance (no cracks occurred after 5 cycles of thermal shock, room temperature~1000 °C), as seen in Table 1. The macro- and micro- structure of the honeycomb ceramics could be seen Fig. 1. Alloys chosen for latent heat storage were commercial Al-12Si and Al-20Si alloys (included 12 wt% and 20 wt% silicon, respectively, supplied by Chengdu Nuclear 857 New Materials Co., LTD), whose comprehensive properties and morphology could be seen in Table 2 and Fig. 2, respectively [34]. It was shown in Table 2 that Al-Si alloys used were featured of the high thermal conductivity and the high fusion heat. Moreover, the melting points of Al-Si alloys were in the temperature range of 400-900 °C, at which the high temperature thermal storage materials were generally used.

6

2.2 Encapsulation processes The precursor of encapsulation materials that used to seal the honeycomb holes for encapsulating the latent alloys was prepared by mixing thoroughly 60 wt% SiCw/Al2O3 ceramic powders (~58μm, obtained by crushing and ball-milling the monolithic SiCw/Al2O3 honeycomb ceramics) and 40 wt% frit (~58μm, supplied by Zibo Guosheng Glaze Co., Ltd., China; the softening temperature was lower than 900 °C and the chemical composition was collected in Table 3) with 5 wt% CMC aqueous solution. This formula of encapsulation materials was designed basing on experiments (trial and measurement). The schematic presentation of encapsulation processes was illustrated in Fig. 3 and the detailed encapsulation procedures were as follows: the precursor that had sealed one side of the partial holes was dried at 100 °C for 24 h in an electric oven, and then the sealed holes were filled with Al-Si alloy powders; after the other side of the holes was sealed, the encapsulation materials were dried and heat-treated at 950 °C at molybdenum disiliciade furnace with a heating rate of 3 °C/min and a holding time of 1 h at the maximum temperature. The unsealed holes were designed for the ventilation of heat transfer medium (usually air) and thereby each sealed hole had four surfaces for heat transfer. The morphology of the fired encapsulation materials bonded with the honeycomb matrix after the heat treatment could be observed in Fig. 4. It was shown in Fig. 4 that the encapsulation materials were bonded with the SiCw/Al2O3 ceramic matrix satisfactorily and the shear strength of the encapsulation materials was tested to be 7.75 MPa, which had significantly surpassed the shear strength requirement of 1 MPa for encapsulation materials as described in Chinese industrial standard (QJ 1643A-1996). The well-bonded encapsulation materials were propitious to inhibit the melted alloys from leakage. As shown in Fig. 5, the photograph of the honeycomb ceramics that encapsulated Al-Si alloys, there was no leakage of alloys observed from the encapsulation 7

materials or the honeycomb ceramic shell after heat-treatment, which showed a good sealing performance. 2.3 Characterization The specific heat capacity (Cp) of the honeycomb shell was measured by TC-7000H laser flash thermal constant analyzer(ULVAC SINKU-RIKO. Inc., Yokohama, Japan). The physical and chemical compatibility between the honeycomb ceramics and Al-Si alloys was studied under a thermal cycle described as follows: the composite samples were placed in a furnace and heated to the preset of temperature (900 °C), soaked for 30 min and then were cooled down (at a rate of 5 °C/min) to 400 °C. The physical compatibility of samples was judged by the cracks or fracture occurred and the leakage of the melted alloys when 100 thermal cycles were completed, while the chemical compatibility was studied by the morphology and the elemental distribution of the combining sites of honeycomb ceramics and alloys. The morphology was observed by SEM (JEOL Ltd., Japan). The elemental distribution was investigated by energy dispersive spectroscopy (EDS), which was taken at 20 kV and a vacuum of 10-4 Pa. The failure mechanism of the latent alloys were studied by the variations in weight, phase compositions, morphology and the fusion heat of alloy materials after enduring 1, 10, 30, 50, 70, 90, 100 thermal cycles. The phase composition was tested by X-ray diffraction (XRD) using a D/MAX-IIIAdiffractometer (Rigaku Corporation, Japan) equipped with Cu Kα (λ=1.54 Å) radiation. The relative content of different phases was calculated by RIR (Reference Intensity Ratio) method semi-quantitatively [35,36]. The fusion heat of the latent alloys was studied by TG-DSC (thermal gravimetry-differential scanning calorimeter) technique by using a simultaneous thermal analyzer (Netzsch Scientific Instruments Trading Ltd., Germany).

8

3. Results and discussion 3.1 Estimation of thermal storage capacity for sensible-latent thermal storage composite materials Thermal storage capacity, i.e. thermal storage density, is defined as the amount of energy accumulated per unit mass or volume, which depends strongly on the application temperature range. For evaluating the thermal storage capacity of alloy/honeycomb ceramic composite thermal storage materials, equations (1) and (2) can be integrated into equation (3)

E  E1  E2

(3)

E is the overall thermal storage energy of the composites. E1 and E2 are the energy contributions from the honeycomb ceramics and Al-Si alloys, respectively. The testing results of specific heat capacity at different temperatures for honeycomb ceramics were gathered in Fig. 6. As known, the specific heat capacity of actual ceramics can be influenced by phase transition, defect formation, thermal vibration, etc. Although the results are the average of the values from five testing points, there is still a little deviation in the testing results [37]. Therefore, as indicated by X. Py et al [7,38], the temperature dependence of the specific heat capacity can be expressed by the following equation (4), which is obtained by the linear fitting of the curve in Fig. 6.

C p  0.8  0.00104T  106  T 2

(4)

According to equations (1) and (4), the thermal storage density of the honeycomb is calculated to be 516.3 kJ/kg (E1, 400-900 °C), whilst basing on equation (2) and the parameters in Table 2, the thermal storage densities of Al-12Si and Al-20Si alloys are 1307.5 kJ/kg and 1233.7 kJ/kg at the constant temperature range, respectively. Due to the large mismatch in the thermal expansion coefficients of honeycombs and Al-Si

9

alloys (see Tables 1 and 2), the honeycomb holes could not be filled up with alloy powders for protecting the honeycomb from the damage caused by the thermal expansion. The volume ratio of the added alloy powders was designed to be 2/3 to the honeycomb holes. Via repeated experiments and measurements, the maximum weight ratio of alloy powders-to-honeycomb was 47.8 %. Thus the energy contributions (E2) from the added Al-12Si and Al-20Si alloys are approximately 625 kJ/kg and 589.7 kJ/kg, both of which have surpassed that of the honeycomb ceramic. Overall, from 400 °C to 900 °C, the thermal storage densities of the composite materials with Al-12Si and Al-20Si alloys are estimated to be 1141.3 kJ/kg and 1106 kJ/kg (400-900 °C), respectively. It is demonstrated that the latent Al-Si alloys show greater energy contributions and the thermal storage capacity of the sensible honeycomb ceramics can be improved by >114%, which makes the honeycomb more suitable for solar thermal storage. 3.2 Stability evaluation of thermal storage composites Apart from the thermal storage capacity, the applicability of thermal storage materials is generally considered in terms of stability and also the stability evaluation is crucial for thermal storage materials composed of PCMs that undergo a melting process. As for the sensible-latent composites, the stability can be characterized by the physical and chemical compatibility between the latent Al-Si alloys and the sensible honeycomb ceramics. Photographs of the alloy/honeycomb ceramic composites after enduring 100 thermal cycles (400-900 °C) were shown in Fig. 7. As seen, these composites have no any signs of physical instability since no cracks form on the surface of honeycomb or encapsulation materials and no leakage of alloys is found, indicating that alloys and ceramics show relatively good physical compatibility. The absence of cracks in the honeycomb shall be attributed to the spare space kept in the holes containing alloys, which can weaken the thermal expansion effect from the melting of 10

alloys. As the undamaged honeycomb ceramics can protect the melted alloys from leakage, no alloys are observed on the surfaces of the composites. Therefore, the corrosion between alloys and the steel container will be inhibited effectively to ensure a long-term operation of thermal storage systems. In an effort to evaluate the chemical compatibility between alloys and honeycomb ceramics, SEM and EDS were applied to study the morphology and elemental distribution of the fractured surfaces of the composites (especially the combining sites), which were shown in Fig. 8 and Fig. 9. It is observed in Figs. 8(a) and 8(b) that the spherical alloy powders melt and form agglomerates after solidification at the combining sites during thermal cycling, whereas the morphology of the honeycomb ceramic remains unchanged as the typical morphology of Al2O3 particles and SiC whiskers can still be observed (see the inset in Fig. 8(b)). By the SEM observations, it is demonstrated that there is no chemical reaction between the honeycomb ceramics and Al-Si alloys even at 900 °C, leading to a good chemical compatibility of the thermal storage composites. Because crystalline phases in the honeycomb ceramics, namely Al2O3 and SiC, are inert to the melting Al-Si alloys as both the possible reactions among Al, Si, Al 2O3 and SiC shown in equations (5) and (6) have a positive Gibbs free energy. 4Al+3SiC=Al4C3+3Si (ΔG>0, 0-900 °C)

(5)

3Si+2Al2O3=4Al+3SiO2 (ΔG>0, 0-900 °C)

(6)

However, when the times of thermal cycle increase up to 100, as seen in Figs. 8(c) and 8(d), the microstructure of honeycomb ceramic beside Al-Si alloy becomes denser with smaller amount of pores. The SEM observations of Al2O3 particles and SiC whiskers are almost lost, whilst the boundary between honeycomb ceramic and alloy becomes ambiguous. Furthermore, at the position nearly 15 μm away from alloys, Al2O3 particles and SiC whiskers are covered by an amorphous 11

phase (see Fig. 8(e)), which is analyzed to be Si according to the EDS mapping images gathered in Fig. 9. As shown in Fig. 9, the honeycomb ceramics that show a denser microstructure are rich in Si element. In these Si-rich areas, Al and O elements with relatively low content can also be detected (see images of Al and O signals in Fig. 9a or the EDS analysis of spot 1 in Fig. 10), indicating that the porous Al2O3-based ceramic matrix is covered by Si which escapes from Al-Si alloys and infiltrates into the honeycomb ceramics during thermal cycling. In contrast to these Si-rich areas, ceramic matrix without the covering Si is porous and abundant in Al and O elements. According to the previous research work about the SiCw/Al2O3 honeycomb ceramic matrix [33], the porosity of honeycomb ceramics could reach up to 51.86% and there were a large amount of the interconnected pores in the microstructure, which provided the pathway that Si infiltrated along. Thus the porous microstructure of the honeycomb ceramics is blamed for the infiltration of Si. However, it shall be emphasized that the contact between Si and the ceramic matrix is still physical rather than chemical as Al2O3 and SiC are inert to Si and the profile of SiC whiskers covered by Si can still be seen in Fig. 8(e). On the contrary, sections belonging to alloy become lack of Si signal due to the loss of Si. As shown in Fig. 9(b), only at the position away from the honeycomb ceramic matrix can the typical segregation of Si in Al-Si alloys be seen. Moreover, apart from Al signal, the strong O signal can be detected in alloys as well, which implies that the alloys beside honeycomb matrix have been oxidized. In Fig. 8(f), the spherical shaped alloy powders become hollow and many scaly crystals form at their surface, which are detected to be Al2O3 (see Fig. 10, the EDS analysis of spots 2 and 3 in Fig. 9(a)), i.e. the products from the oxidized Al. As shown in Fig. 11, the oxidation reaction can also be confirmed by the XRD analysis of Al-Si alloys after enduring thermal cycles of different times. It is observed in Fig. 11 that Al has been oxidized with the detection of Al2O3 and 12

Si remains unoxidized without the formation of SiO2. This is because Al shows a greater reactivity to oxygen than Si in the melts of Al-Si alloys and Al2O3 is the only stable crystalline product from the oxidation reaction of Al [39]. The oxidation reaction can destroy the agglomerate structure of the solidified alloy and produce a great number of conchoidal shaped residuals that cannot contain the unoxidized Si. Then Si escapes from the oxidized alloy and infiltrates into honeycomb ceramic matrix. Herein, the porous microstructure of honeycomb ceramic matrix shall also be blamed for the oxidation of Al-Si alloys as the interconnected pores expose alloys to oxygen and lead to the instability of alloys. In conclusion, although Al-Si alloys and SiCw/Al2O3 honeycomb ceramic show good physical and chemical compatibility, which can enhance the stability of the thermal storage composites, the porous microstructure of the honeycomb ceramic matrix will lead to the oxidation of Al in alloy melts at high temperatures, the infiltration of the remaining Si into the ceramic matrix and thereby the long-term instability of Al-Si alloys. The oxidation of Al will also influence the thermal storage capacity of Al-Si alloys according to the following study about the failure mechanism of alloys. 3.3 Failure mechanism of latent Al-Si alloys during thermal cycling Since the oxidation reaction of Al has changed the phase compositions of Al-Si alloys, which will affect the thermal storage properties of the composites, it is necessary to investigate the failure mechanism of alloys during thermal cycling. Although the alloys beside honeycomb ceramic matrix are oxidized significantly along with the loss of Si during thermal cycling, as shown in Fig. 12, a great number of Si can still be detected in alloys closed to the center of honeycomb unit, which manifests that the interior alloys have a lower oxidation rate than the marginal ones. Thus the agglomerate structure of the solidified alloys can still be maintained to retain Si. The lower oxidation rate shall be attributed to the 13

barriers that form by the infiltrated Si as well as the Al2O3 products resulted from the oxidation of the marginal alloys. As shown in Figs. 8(c) and 8(d), the infiltrated Si blocks up the interconnected pores and the honeycomb ceramic matrix near alloys shows a denser microstructure with a smaller amount of pores, which can limit the rate of oxygen infiltration. As the distribution of the infiltrated Si is inhomogeneous in the honeycomb matrix and the interconnected pores cannot be eliminated completely, the Al2O3 products act as another barrier to prevent the oxygen from infiltration by prolonging and narrowing the aisles of oxygen. According to the semiquantitative XRD analysis results of the crystalline phases of Al-12Si alloy after enduring thermal cycles with different times (see Table 4), the relationship of Al2O3 content and thermal cycles was obtained in Fig. 13. As seen, Al2O3 content in Al-12Si alloy increases significantly at the early stage (≤10 times) and then shows a smaller increasing tendency with the increasing thermal cycles, which indicates that oxidation rate is reduced due to the barrier constructed by the infiltrated Si and the Al2O3 products. However, it shall be noted that as long as the thermal cycling is conducted at temperatures higher than 600 °C, the oxidation will be further enhanced with increasing thermal cycles. As shown in Fig. 14, the TG-DSC analysis results of Al-Si alloys after enduring 100 thermal cycles, the weight gain effects caused by the oxidation of Al become prominent at above 600 °C. Therefore, despite of the decreasing oxidation rate at the late stage, alloys kept at the designed temperature range (400-900 °C) will be oxidized continuously until the complete transformation of Al to Al2O3. Moreover, the latent heat capacity will be degraded with increasing times of thermal cycle, which shall be owing to the oxidation of Al and the segregation of Si. As known, the melting processes of Al-Si alloys begin by breaking the interatomic Al-Al bonding in Al crystals which is weaker than Si-Si bonding, and the energy required for breaking the interatomic bonding is 14

defined as latent heat. Whereas the oxidation reaction decreases the amount of Al crystals and Al-Al bonding with increasing thermal cycles, thus lowering the latent heat of alloys gradually. As for the segregation of Si, it is shown in the EDS images of Si signal (see Fig. 14) that Si phases have segregated to form a large amount of spherical shaped Si aggregations, which lower the interfacial energy of alloys and lead to the phase change with a lower energy [39]. Accordingly, after enduring 100 thermal cycles, the latent heat capacities of Al-12Si and Al-20Si alloys have decreased to 329.3 J/g and 339.3 J/g, respectively (see Fig. 14). As the melting point of pure Si is as high as 1416 °C, which has surpassed the maximum testing temperature in this study, the latent thermal storage mode will lose efficacy after the complete oxidation of Al and only the sensible one remains. To avoid the extensive degradation of the latent heat, it is necessary to limit the maximum application temperature of the thermal storage composites to ≤600 °C or to conduct the surface treatment for the honeycomb shell to eliminate the interconnected pores. 3.4 Estimation of thermal storage capacity for sensible-latent composite thermal storage materials after thermal cycling As mentioned above, the latent heat capacities and the phase compositions of alloys have changed after thermal cycling. Thus the thermal storage capacity of the composites shall be re-estimated and the equation (3) shall turn to equation (7) as follows

E  E1  E2  E3

(7)

E3 is the energy contribution from the Al2O3 products during thermal cycling. It is revealed in Table 4 that with the continuous reduction of Al by oxidation, the weight percentages of Si in the remaining Al-12Si and Al-20Si alloys (exclude the weight ratios of Al2O3) after enduring 100 thermal cycles have increased to 14% and 23.2%. With the increasing Si content in Al-Si alloys, the specific heat capacity will be lowered (the room temperature specific 15

heat capacities of Al and Si are 0.88 J·g-1·K-1 and 0.701 J·g-1·K-1, respectively). Since there are merely 2% and 3.2% variations in the percentage of Si after 100 thermal cycles, the differences in the values of the sensible heat capacities of alloys are rather small (changes of the values are in three decimal places of the dollar) and thus the sensible values of alloys in Table 2 can still be used for evaluating the thermal storage capacity. As shown in Fig.14, the latent heats of Al-12Si and Al-20Si alloys after 100 thermal cycles are reduced to 329.3 J/g and 339.3 J/g, while the melting points are 581.3 °C and 581.5 °C, respectively. Hence the thermal storage densities of the remaining Al-12Si and Al-20Si alloys are calculated to be 1072.4 kJ/kg and 1040.9 kJ/kg (400-900 °C) according to equation (2). In order to determine the weight ratios of the remaining alloys, the weight gains of the composites endured thermal cycles of different times were tested and collected in Table 5. As seen, there are weight gains of 4.32 grams and 3.53 grams for Al-12Si and Al-20Si alloys after thermal cycling tests, respectively, which are resulted from transformation of Al to Al2O3 (88.88% weight gain in this transformation). Thus, it can be confirmed that Al of 4.86 grams and 3.97 grams within Al-12Si and Al-20Si alloys has been oxidized, producing Al2O3 of 9.18 grams and 7.5 grams, respectively. As the additive contents of Al-12Si and Al-20Si alloys are 29.38 grams and 29.7 grams, the weights of the remaining Al-12Si and Al-20Si alloys shall be 24.52 grams and 25.73 grams. The weight ratios of Al2O3 in Al-12Si and Al-20Si alloys are 27.24% and 22.57%, which are approximate to the semiquantitative analysis results shown in Table 4. Overall, the weight ratios of the remaining Al-Si alloys and the produced Al2O3 to the honeycomb ceramic shells are 39.7%, 14.9% for Al-12Si alloy and 40.1%, 12.1% for Al-20Si alloy. Therefore, the energy contributions (E2) from the remaining Al-12Si and Al-20Si alloys are 425.7 kJ/kg and 417.4 kJ/kg. In terms of the specific heat capacity of the produced Al2O3, the theoretical data from [40] in 16

Table 6 can be applied for estimating its thermal storage density, whose calculation method is consistent with that of the sensible honeycomb ceramic. The relationship between the specific heat capacity of Al2O3 and temperature is approximately expressed as equation (8). Via calculation basing on equations (1) and (8), the thermal storage density of the produced Al2O3 shall be 657.3 kJ/kg (400-900 °C) and the sensible energy contributions (E3) from Al2O3 of 14.9% and 12.1% are 97.9 kJ/kg and 79.5 kJ/kg, respectively. By adding the sensible thermal storage density of the honeycomb ceramic (E1, 516.3 kJ/kg), the thermal storage densities of the composite thermal storage materials with Al-12Si and Al-20Si alloys are re-estimated to be 1039.9 kJ/kg and 1013.2 kJ/kg, which have decreased by 8.88% and 8.39%, respectively.

C p  0.7569  0.00154T  106  T 2

(8)

In summary, although the oxidation of alloys has reduced the thermal storage capacity of the composites, the thermal storage densities are still as high as 1039.9 kJ/kg and 1013.2 kJ/kg due to the sensible energy contribution from the Al2O3 products, which makes the composites capable of using as the sensible thermal storage materials even after the complete oxidation of Al in Al-Si alloys.

4. Conclusions In this study, novel high-temperature alloy/honeycomb ceramic composite materials for solar sensible-latent thermal storage applications had been prepared by encapsulating Al-Si alloys in the holes of SiCw/Al2O3 honeycomb ceramics. The thermal storage capacities of the composite materials before and after thermal cycling tests were evaluated. The stability evaluation between Al-Si alloys and honeycomb ceramics was conducted and the failure mechanism for the latent alloys had also been studied. Several key conclusions could be drawn as follows. 1. The addition of Al-Si alloys could improve the thermal storage capacity of the sensible 17

honeycomb ceramics significantly by >114% and the thermal storage densities of the composite materials with Al-12Si and Al-20Si alloys were 1141.3 kJ/kg and 1106 kJ/kg (400-900 °C),, respectively which would decrease to 1039.9 kJ/kg and 1013.2 kJ/kg after enduring 100 thermal cycles. 2. Al-Si alloys and SiCw/Al2O3 honeycomb ceramics exhibited good physical and chemical compatibility, which ensured the high thermal stability up to 900 °C for the composites. The disadvantage of the honeycomb shell was its porous microstructure, which promoted the oxidation of Al in alloys as well as the infiltration of the remaining Si during thermal cycling and resulted in the instability of Al-Si alloys. 3. The latent heat of Al-Si alloys would be lowered with the gradual oxidation of Al and the latent thermal storage of alloys would lose efficacy after the complete transformation of Al to Al2O3. To ensure the long-term stability of Al-Si alloys encapsulated in honeycomb ceramics, further works will focus on the determination of the maximum application temperatures for the latent alloys or the surface treatment for the honeycomb shell.

Acknowledgment Authors are very grateful to the financial support from “973 Program (2010CB227105)”, P. R. China.

References [1] S. Kuravi, J. Trahan, D. Yogi. Goswami, M. M. Rahman, E. K. Stefanakos, Thermal energy storage technologies and systems for concentrating solar power plants, Prog. Energ. Combust. 39 (2013) 285-319. 18

[2] M. Cascetta, G. Cau, P. Puddu, F. Serra, Numerical investigation of a packed bed thermal energy storage system with different heat transfer fluids, Energ. Procedia. 45 (2014) 598-607. [3] B. Xu, P.W. Li, C. Chan, Application of phase change materials for thermal energy storage in concentrated solar thermal power plants: A review to recent developments, Appl. Energ. 160 (2015) 286-307. [4] Y. Tian, C.Y. Zhao, A review of solar collectors and thermal energy storage in solar thermal applications, Appl. Energ. 104 (2013) 538-553. [5] M. Liu, W. Saman, F. Bruno, Review on storage materials and thermal performance enhancement techniques for high temperature phase change thermal storage systems, Renew. Sust. Energ. Rev. 16 (2012) 2118-2132. [6] S. Kuravi, J. Trahan, D. Yogi. Goswami, M. M. Rahman, E. K. Stefanakos, Thermal energy storage technologies and systems for concentrating solar power plants, Prog. Energ. Combust. 39 (2013) 285-319. [7] X. Py, N. Calvet, R. Olives, A. Meffre, P. Echegut, C. Bessada, E. Veron, S. Ory, Recycled material for sensible heat based thermal energy storage to be used in concentrated solar thermal power plants, J. Sol. Energ. Eng. 133 (3) (2011) 1-8. [8] A. Sharma, V.V. Tyagi, C.R. Chen, D. Buddhi, Review on thermal energy storage with phase change materials and applications, Renew. Sust. Energ. Rev. 13(2) (2009) 318-345. [9] G. Zanganeh, A. Pedretti, A. Haselbacher, A. Steinfeld, Design of packed bed thermal energy storage systems for high-temperature industrial process heat, Appl. Eng. 137 (2015) 812-822. [10] A. Gil, M. Medrano, I. Martorell, A. Lárzaro, P. Dolado, B. Zalba, L.F. Cabeza, State of the art on high temperature thermal energy storage for power generation. Part 1-Concepts, materials and modellization, Renew. Sust. Energ. Rev. 14 (1) (2010) 31-55. 19

[11] H. Singh, R.P. Saini, J.S. Saini, A review on packed bed solar energy storage systems, Renew. Sustain. Ener. Rev. 14 (2010) 1059-1069. [12] X.H. Xu, X.B. Lao, J.F. Wu, Y.X. Zhang, X.Y. Xu, K. Li, Synthesis and characterization of Al2O3/SiC composite ceramics via carbothermal reduction of aluminosilicate precursor for solar sensible thermal storage, J. Alloys. Compounds. 662 (2016) 126-137. [13] X.H.Xu, X.B. Lao, J.F. Wu, Z.G. Rao, Y. Zhou, D.Z. He, Y. Liu, Preparation and performance study of Sialon-Si3N4-SiC composite ceramics for concentrated solar power, Int. J. Appl. Ceram. Technol. 12 (2015) 949-956. [14] Ali A. Jalalzadeh-Azar, W. Glenn Steele, George A. Adebiyi, Performance comparison of high-temperature packed bed operation with PCM and sensible-heat pellets, Int. J. Energ. Res. 21 (1997) 1039-1052. [15] J. Song, L.K. Zeng, Y.C. Liu, X.T. Ren, H. Wang, X.S. Cheng, Research and application of ceramic heat storage materials, Chin. Ceram. 43 (6) (2007) 7-10. [16] R.X. Tan, Z.L. Ke, Y.C. Liu, S.A. Ze, H. Wang, P.A. Liu, J. Song, Porous ceramic materials used for heat storage and accumulation, J. Ceram. 2 (27) (2006) 217-226. [17] Z. Chen, L. Cao, G.Y. Fang, F. Shan, Synthesis and characterization of microencapsulated paraffin microcapsules as shape-stabilized thermal energy storage materials, Nanosc. Microsc. Therm. Eng. 17 (2013) 112-123. [18] D. Yin, L. Ma, J. Liu, Q. Zhang, Pickering emulsion: a novel template for microencapsulated phase change materials with polymer-silica hybrid shell. Energ. 64 (2014) 578-581. [19] B. Zalba, J. Marin, L. Cabeza, H. Mehling, Review on thermal energy storage with phase change: materials, heat transfer analysis and applications. Appl. Therm. Eng. 23 (2003) 251-283. 20

[20] S. Khare, M. Dell’Amico, C. Knight, S. McGarry, Selection of materials for high temperature sensible energy storage, Sol. Energ. Mater. Sol. C. 115 (2013) 114-122. [21] F. Pitié, C.Y. Zhao, J. Baeyens, J. Degrève, H.L. Zhang, Circulating fluidized bed heat recovery/storage and its potential to use coated phase-change-material (PCM) particles, 109 (2013) 505-513. [22] Y. Tian, C.Y. Zhao, A review of solar collectors and thermal energy storage in solar thermal applications, Appl. Energ. 104 (2013) 538-553. [23] G. Fang, Z. Chen, H. Li, Synthesis and properties of microencapsulated paraffin composites with SiO2 shell as thermal energy storage materials, Chem. Eng. J. 163 (2010) 154-159. [24] A. San, A. Bicer, Thermal energy storage properties and thermal reliability of some fatty acid esters/building material composites as novel form-stable PCMs, Sol. Energ. Mater. Sol. C. 101 (2012) 114-122. [25] L. Miró, E. Oró, D. Boer, L. F. Cabeza, Embodied energy in thermal energy storage (TES) systems for high temperature applications, Appl. Energ. 137 (2015) 793-799. [26] A. Sari, A. Karaipekli, Preparation, thermal properties and thermal reliability of palmitic acid/expanded graphite composite as form-stable PCM for thermal energy storage, Sol. Energy. Mater. Sol. C. 93 (2009) 571-576. [27] K. Kwak, C. Kim, Viscosity and thermal conductivity of copper oxide nanofluid dispersed in ethylene glycol, Korea-Australia. Rheol. J. 17 (2005) 35-40. [28] Y.J. Zhong, Q.G. Guo, S.Z. Li, J.L. Shi, L. Liu, Heat transfer enhancement of paraffin wax using carbon foam for thermal energy storage, Sol. Energ. Mater. Sol. C. 94 (2010) 1011-1014. [29] A. Jamekhorshid, S.M. Sadrameli, M. Farid, A review of microencapsulation methods of 21

phase change materials (PCMs) as a thermal energy storage (TES) medium, Renew. Sustain. Ener. Rev. 31 (2014) 531-542. [30] X.M. Chen, J. Dong, X.W. Wu, D.Q. Gong, Thermal storage properties of high-temperature phase transformation on Al-Si-Cu-Mg-Zn alloys, Heat. Treatment. Metals. 35 (2010) 13-16. [31] J. Liu, X. Wang, D.B. Zeng, X.T. Guo, Y.P. Zhang, H.F. Di, Y. Jiang, The selectness of high-temperature phase change material Al-Si alloy and experimental research on the container, Acta. Energiae. Solaris. Sinica. 27 (2006) 36-40. [32] J. Xu, S.R. Yu, Research and application progress of Al-based alloy phase change materials using for thermal storage, Mater. Rev. 27 (2013) 93-97. [33] X.H. Xu, X.B. Lao, J.F. Wu, X.Y. Xu, Y.X. Zhang, K. Li, In-situ synthesis of SiCw/Al2O3 composite honeycomb ceramics by aluminium-assisted carbothermal reduction of coal series kaolin, Appl. Clay. Sci. 126 (2016) 122-131. [34] H.T. Cui, P.Y. Peng, J.Z. Jiang, The status and prospect on Al-Si alloy and heat storage unit as phase change material for thermal energy storage, Mater. Rev. 28 (2014) 72-75. [35] L. Eric, X-ray characterization of materials, John Wiley & Sons, Inc., Germany, 1999, pp. 75-77. [36] C.R. Hubbard, R.L. Snyder, RIR-measurement and use in quantitative XRD, Powder. Diffraction. 3 (1988) 74-77. [37] Ceramic society of Japan, Advanced ceramic technologies & products, Springer Japan, Tokyo, Japan, 2012, pp. 47. [38] X. Py, N. Calvet, R. Olives, A. Meffre, P. Echegut, C. Bessada, E. Veron, S. Ory, Recycled material for sensible heat based thermal energy storage to be used in concentrated solar thermal power plants, J. Sol. Energ. Eng. 133 (3) (2011) 1-8. 22

[39] R.Y. Zhang, J.Q. Sun, X.F. Ke, X.X. Zhou, Heat storage properties of Al-Si alloy, Chin. J. Mater. Res. 20 (2006) 156-160. [40] R. Barea, M. Belmonte, M.I. Osendi, P. Miranzo, Thermal conductivity of Al2O3/SiC platelet composites, J. Eur. Ceram. Soc. 23 (2003) 1773-1778.

Fig. 1. (a) Photograph of SiCw/Al2O3 composite honeycomb ceramics; (b)&(c) SEM images. Fig. 2. SEM images of (a) Al-12Si and (b) Al-20Si alloys. Fig. 3. Schematic presentation of encapsulation processes. Fig. 4. The morphology of the fired encapsulation materials bonded with SiCw/Al2O3 honeycomb ceramics matrix. Fig. 5. Photograph of SiCw/Al2O3 composite honeycombs encapsulated with Al-Si alloys. Fig. 6. Specific heat capacity of SiCw/Al2O3 composite honeycombs as a function of temperature. Fig. 7. Photographs of the alloys/honeycomb ceramic composites after enduring 100 thermal cycles (400-900 °C). Fig. 8. SEM images of the fractured surface of thermal storage composites: (a) the combining site of honeycomb and Al-12Si alloy after enduring 50 thermal cycles; (b) magnification of zone 1 in (a); (c) the combining site of honeycomb and Al-12Si alloy after enduring 100 thermal cycles; (d), (e) & (f) magnifications of zones 2, 3 and 4, respectively. Fig. 9. EDS mapping of the combining sites of honeycomb and alloys after enduring 100 thermal cycles: (a) with Al-12Si alloy; (b) with Al-20Si alloy. Fig. 10. EDS (Energy Dispersive Spectroscopy) patterns of three selected spots in Fig. 9a. Fig. 11. XRD patterns of Al-Si alloys after enduring thermal cycles of different times: (a) Al-12Si alloy, (b) Al-20Si alloy. 23

Fig. 12. EDS mapping of alloys away from honeycomb ceramic matrix: (a) Al-12Si alloy; (b) Al-20Si alloy. Fig. 13. The relationship between Al2O3 content in Al-Si alloys and times of thermal cycle. Fig. 14. TG-DSC analysis of Al-Si alloys after enduring 100 thermal cycles: (a) Al-12Si alloy; (b) Al-20Si alloy.

Table 1 Performances of SiCw/Al2O3 composite honeycomb ceramics Performances

Values

Porosity /%

51.86

Compressive strength /MPa

5.44

Softening temperature /°C

>1400

Thermal expansion coefficient /×10-6·°C-1

8.16 No cracks forming after 5 cycles of thermal shock (room temperature~1000 °C, air cooling)

Thermal shock resistance

Table 2 Performances of commercial Al-12Si and Al-20Si alloys Performances

Al-12Si

Al-20Si

Average heat capacity of solid /J·g-1·K-1

1.04

0.97

Average heat capacity of liquid /J·g-1·K-1

1.74

1.65

Melting temperature /°C

575

576

Heat of fusion /J·g-1

560

528.4

Density /g·cm-3

2.67

2.65

Thermal expansion coefficient /×10-6·°C-1

20.5

17.6

Thermal conductivity /W·m-1·K-1

144.6

130

Table 3 Chemical composition of frit Oxides

SiO2

Al2O3

CaO

MgO

K2O

Na2O

BaO

Total 24

Content /wt%

69.0

11.46

6.21

0.54

3.73

7.58

1.48

100

Table 4 Semiquantitative analysis results of crystalline phases in Al-Si alloys after enduring thermal cycles of different times (wt%). Al-12Si

Al-20Si

Thermal cycles Al

Si

Al2O3

Al

Si

Al2O3

1time

81.8

11.6

6.6

77.4

19.9

2.7

5 times

77.9

11.4

10.7

70.1

19.1

10.8

10 times

74.9

11.2

13.9

66.7

18.8

14.5

30 times

70.8

11.0

18.2

64.3

18.5

17.2

50 times

67.0

10.7

22.3

62.6

18.3

19.1

70 times

64.8

10.6

24.6

61.3

18.2

20.5

90 times

62.7

10.4

26.9

60.0

18.0

22.0

100 times

62.4

10.2

27.4

59.1

17.9

23.0

Table 5 Weight gain of samples with Al-12Si and Al-20Si alloys after enduring thermal cycles of different times Weight gain(Δm) /g Thermal cycles Sample with Al-12Si alloy

Sample with Al-20Si alloy

1time

0.94

0.38

5 times

1.56

1.56

10 times

2.05

2.16

30 times

2.74

2.58

50 times

3.38

2.89

70 times

3.85

3.13

90 times

4.26

3.38

100 times

4.32

3.53

Table 6 Values of specific heat capacity of Al2O3 used for the calculation of the sensible energy 25

contribution Testing temperature /°C

25

100

200

300

400

500

600

Specific heat capacity /J·g-1·K-1

0.7719

0.9205

1.0273

1.0906

1.1340

1.1668

1.1933

26