- Email: [email protected]
Agro-waste ash and mineral oxides derived glass-ceramics and their interconnect study with Crofer 22 APU for SOFC application
Ceramics International xxx (xxxx) xxx–xxx
Contents lists available at ScienceDirect
Ceramics International journal homepage: www.elsevier.com/locate/ceramint
Agro-waste ash and mineral oxides derived glass-ceramics and their interconnect study with Crofer 22 APU for SOFC application Gaurav Sharma, K. Singh* School of Physics and Materials Science, Thapar Institute of Engineering and Technology, Patiala, 147004, India
A R T I C LE I N FO
A B S T R A C T
Keywords: Glass-ceramic Interaction Sealant Solid oxide fuel cells X-ray diffraction
In this study, silicate glass-ceramic is synthesized from wheat straw ash (WSA) using melt-quench technique. Xray diffraction is done to confirm the nature of the as-quenched samples. Inductive coupled plasma-mass spectroscopy (ICP-MS) and energy dispersive spectroscopy (EDS) are performed to evaluate the exact chemical composition of the as-prepared samples. Similar composition is also melted using conventional mineral oxides for comparison with WSA sample. The WSA sample shows good thermal stability in temperature range of 300–900 °C. The coefficient of thermal expansion (CTE) of WSA sample is more or less in the required range for solid oxide fuel cells (SOFC) application. The powder of WSA derived glass-ceramic is used to form the diffusion couple with the interconnect (Crofer 22 APU) for interfacial study. The excellent adhesion with the smooth and porosity-free interface is formed between glass-ceramic and Crofer 22 APU, when WSA is used as a sealant as compared to the mineral based glass-ceramics. Semi-coherent interface does not show any delimitation or separation even after five thermal cycles. The micro hardness (384 HV) of the interface is indicated good bonding between interconnect and glass-ceramic.
1. Introduction
may be initiated the thermal stress at the interface, and it reduces the overall efficiency of SOFC. Therefore, minimum reaction should be taken place between glass-sealants and other components of SOFC [11,12]. All the required components, interconnect is a very crucial and vital factor to retain the performance of SOFC. Generally, chromiumrich steel (Crofer 22 APU) is being used as an inter-connector in SOFC. Thus, WSA glass-ceramic and Crofer 22 APU have been selected to check the interaction study with numerous thermal cycles. Additionally, similar chemical composition is also melted in similar condition for comparison the results with WSA sample. On the other hand, the existing waste materials draw the major interest of the scientific community to find out the effective way to utilization for suitable applications in the engineering fields, which also reduces waste management-related problems [13–15]. In general, the ashes of agricultural waste are being used in the civil and construction due to their low-density and small particle size with higher content of silica with some other trace elements [16–18]. It is well-reported in the literature that the extracted silica with other oxides from agricultural waste is exploited to form glasses and glass-ceramic [19–21]. Mostly sealing glasses have 40–60% SiO2 as network-former along with alkaline-earth metal oxides 10–30 (mol%) [10]. Additionally, some intermediate oxides like Al2O3, Y2O3, TiO2 and ZrO2, etc., are also being used to control viscosity, CTE and
Sealing material is very important and essential part of the planar solid oxide fuel cell (SOFC) [1]. Generally, sealant is applied at the edges of the flat plates [2,3]. SOFC device is required high operating temperature, so, the conventional sealing materials like polymer, alloys and organic adhesive cannot sustain at the working temperature of SOFC (800–1000 °C) [4–6]. That why, various types of glasses and glass-ceramics are typically used as the sealant materials to improve the lifespan of the SOFC after many thermal cycles at the higher temperature. The quality of sealing materials must be high, even miner leakage can affect the cell potential, and it can reduce the overall performance of the device [7]. However, silicate glasses and glass-ceramics are always preferable as compare to other formers based glasses and glassceramics like B2O3 and P2O5 for this purpose due to better thermal stability, particularly, in variable oxygen partial pressure and presence of water vapor [5,8,9]. Basically, glass-sealant made an interface with all the components in SOFC's [10]. Furthermore, formed interface usually delaminates during the thermal cycles or long run of SOFC due to the thermal expansion mismatch between different components of SOFC. In addition to this, due to high working temperature of SOFC, some crystalline phases are also formed due to inter-diffusion and intradiffusion among SOFC components. The observed detrimental phases *
Corresponding author. E-mail address: [email protected] (K. Singh).
https://doi.org/10.1016/j.ceramint.2019.07.029 Received 28 May 2019; Accepted 2 July 2019 0272-8842/ © 2019 Elsevier Ltd and Techna Group S.r.l. All rights reserved.
Please cite this article as: Gaurav Sharma and K. Singh, Ceramics International, https://doi.org/10.1016/j.ceramint.2019.07.029
Ceramics International xxx (xxxx) xxx–xxx
G. Sharma and K. Singh
Table 1 The chemical composition of WSA (wt%) analyzed by EDS and similar composition of mineral derived sample. Label
SiO2
CaO
Na2O
K2O
MgO
WSA MOD-sample
85 85
8 8
0.82 0.82
4.85 4.85
4.85 4.85
characteristic temperatures of the glasses [7,22–24]. However, advanced and better performing seal materials are still required, which can sustain for longer time without having any detrimental reaction with the other components of SOFC. The interfacial behavior between components and glass-sealants are needed to know the thermal stability, interface formation its growth and interfacial behavior with respect to the thermal cycles before using sealant in real SOFC conditions. Usually, it is considered that alkali oxide like K2O and Na2O, etc. increase the CTE of glasses. Contrary to this, they are also diffuse very fast, which can cause drastic drop in electrical resistance of glass-sealants at the higher temperature. Alkali metals oxides also enhanced volatility of the chromium and poisoning of cathode [10]. In the present study, the attempt has been made to form the glass-ceramic from wheat straw ash (WSA) and conventional mineral oxides derived glass-ceramic i-e., MOD-sample. The WSA sample is unique, and new in the sense that the agricultural waste derived sample probable the first time investigated as the sealing material. However, exact mineral based glassceramic is not any kind of suitability for this application. In the present study, agricultural waste derived glass-ceramic is revealed better adhesion with interconnect than mineral oxides synthesized glass-ceramics. This study is performed in order to understand WSA based glassceramic and MOD glass-ceramic sealing properties in SOFC's.
Fig. 2. Thermal gravimetric (TGA) curve of the as-quenched WSA sample.
crucible at 1550 °C in the programmable electric-muffle furnace using a heating rate of 5 °C/min. The molten substance is quenched in the air from 1550 °C to room temperature. The thermal gravimetric measurement (TGA) of the WSA sample was performed with NETZSCH STA449 F3 Jupiter in the air atmosphere at the heating rate of 10 °C/min from 50-950 °C. During the experiment, the partial pressure of the air was 5 kg/cm2. The temperature and weight loss detection limit of the instrument are ± 1 °C and 0.001 mg, respectively. The coefficient of thermal expansion (CTE) of the as-quenched WSA sample was determined by using Netzsch DIL402 PC in the temperature range 30–850 °C in the air atmosphere. The amorphous and crystalline nature of the samples was checked by X-ray diffraction. The XRD pattern was obtained by PANalytical's X'Pert Pro X-ray diffractrometer. During experiment, the scan speed was 0.001°/minutes. Fourier transforms infrared (FTIR) spectra of the as-quenched WSA, MOD-sample and powder of glass-ceramic of WSA sample from interface was obtained on the Perkin Elmer-Spectrum-RX-IFTIR spectrometer from 4000400 cm−1 range. For FTIR measurement, the pellets were made of the
2. Experimental details Two samples (WSA and MOD) were synthesized using wheat straw ash and exact same chemical composition via minerals, respectively as given in Table 1. Initially, wheat straw ash is processed at 1000 °C for 1 h to remove any volatile substance. Further, both the samples were made in similar experimental conditions as follows. The powder was ground in an agate-mortar pestle and melted in re-crystallized alumina
Fig. 1. Schematic design of diffusion couple of glass-ceramic sample with Crofer 22 APU. 2
Ceramics International xxx (xxxx) xxx–xxx
G. Sharma and K. Singh
100–800 °C to know the insulating nature of the heat-treated sample. A circular pellet is made from glass-ceramic powder and heat-treated at 900 °C for 10 h. The heat-treated pellet is gold sputtered to make the proper electrical contact. The powder of the as-prepared sample was mixed with polyvinyl alcohol (PVA) and put as a sandwich between two ultrasonically cleaned Crofer 22 APU sheets of size 1.0 × 1.0 × 0.2 cm as shown in Fig. 1 to make diffusion couple. This diffusion couple was kept at 900 °C for 1, 2, 10, 100 and 500 h. SEM measurement was performed by JOEL-JSM 6460 LV (Tokyo, JAPAN) microscope at the acceleration voltage of 15 kV. Prior to the observations, surface was coat with a gold layer of about 100 Å thickness to avoid the charging under electron beam. The energy dispersive spectroscopy analysis was carried out using OXFORD instrument INCA-X act. The micro hardness test was performed on Mitutoyo micro hardness tester (MVK-HO, Japan) on the polished and ultrasonically cleaned diffusion couple. The indentations were made at 49 and 98 N load with 10 s dwell time. The micro hardness of various places of diffusion couple was measured by using the following equation [25,26].
Fig. 3. Coefficient of thermal expansion (CTE) of wheat straw ash (WSA) sample.
HV = 1.854(F/d2) Where, F and d are applied load and area of indentation, respectively. 3. Results and discussion 3.1. EDS and ICP analysis The chemical composition of the processed wheat straw ash was estimated by energy dispersive spectroscopy (EDS). To confirm the exact and precise chemical composition of the as-quenched WSA sample was calculated using inductive coupled plasma (ICP-MS) technique. The elemental percentage of the present sample is more or less similar of minerals derived glass-ceramics. However, both samples have the higher amount of SiO2 with alkali and alkaline earth-metal, but some very miner trace elements are presented in the WSA composition inherently as compared to MOD sample. 3.2. Thermogravimetric (TG) and CTE analysis The thermal stability of the given WSA sample is checked using TGA experiment as shown in Fig. 2. Initially, sample exhibits some weight loss i.e., 1.5% up to 350 °C. This weight loss is related to water molecules absorbed by the sample. After that, TGA curve do not show any trend with respect to temperature i-e., 350–950 °C, it indicates the good thermal stability of WSA sample. Additionally, the coefficient of thermal expansion (CTE) is very essential parameter for sealing applications; normally, it should be in the range of 9–13 × 10−6/°C for SOFC applications [1,7,27]. The CTE of as-prepared sample is 10 × 10−6/°C as shown in Fig. 3. Usually, silica-rich glasses have low CTE, this amount of silica is deciding factor for CTE in silicate glasses [28]. The presence of the modifying elements in the samples can increase the CTE due to the modifying overall structure of the glass-network means the silicate-network. However, the small amount i-e., only ~4 wt% of the modifiers is found in the WSA sample. XRD of the melt-quench sample (discussed in the next section) shows the presence of tridymite crystalline phase. The tridymite phase has higher CTE as reported by Nurur et al. [29]. Thus, CTE of the present sample may be arisen due to the presence of the crystalline phase as well as some trace elements and their bonding with glass-matrix.
Fig. 4. XRD pattern of (a) as-prepared sample, (b) after chemical interaction at 900 °C for 500 h and (c) MS-sample derived using mineral oxides. Table 2 Weight percentage (%) of the present oxides in the as-quenched glass-ceramic of WSA and MOD sample calculated using ICP and EDS. Compounds
SiO2 CaO Na2O MgO K2O Al2O3
Elemental analysis (wt %) WSA (ICP)
WSA (EDS)
84.94 8.02 0.62 0.82 4.85 0.72
84.53 4.85 1.46 1.83 4.85 2.40
MOD-sample (EDS)
85.40 9.50 1.25 0.55 2.25 1.05
3.3. X-ray diffraction analysis The nature of the as-quenched samples derived from WSA and MOD sample are checked using X-ray diffraction. Additionally, the powder of both the samples is used to make the diffusion couples (as shown in experimental section) for interaction study. After exposing for different
sample's powder with KBr in the 1:1 wt ratio. The impedance measurement performed using SOLATRON (1260S) impedance spectroscopy using frequency from 102-106 Hz range for temperature range of 3
Ceramics International xxx (xxxx) xxx–xxx
G. Sharma and K. Singh
exposed WSA sample (at 900 °C for 500 h) is shown in Fig. 4 (b). The XRD pattern is indexed with the tridymite crystalline SiO2 phases. Interestingly, the Al2O3 crystalline phase, presented in the initial glassceramic, is completely either suppressed or dissolved in the matrix as indicated by XRD of 500 h exposed sample. Many researchers have reported that sometime meta-stable crystalline phase may dissolve to form new crystalline phase in prolonged heat-treatment of glasses and glass-ceramics [5,30]. As heat-treatment evolved, either it dissolves or converts into more stable crystalline phase. The SiO2 crystobalite phase is usually detrimental in SOFCs application point of view since large volume change occurs during thermal cycles, which led the thermal stresses. However, in the present case, the crystobalite phase is not formed. Moreover, In WSA sample, no other mixed crystalline phases are formed even after 500 h exposed duration. Usually, chemical interaction for longer duration leads to the formation of many crystalline phases in the multi-component glasses/glass-ceramics with other components of SOFCs when glasses are made of mineral oxides. Conclusively, minimum reaction is taken place without forming any detrimental crystalline phase at the interface between sample and Crofer 22 APU. In this respect, the wheat straw ash derived glass-ceramic seems to be better alternate sealing materials than conventional glass-sealant formed using mineral oxides.
Fig. 5. FTIR spectra of (a) as-prepared WSA (b) exposed glass-ceramic and (c) MOD as-quenched sample.
thermal cycles, the powders were taken from diffusion couple and investigated by XRD. X-ray diffraction patterns are given in Fig. 4(a) and (b). The X-ray diffraction pattern of the as-prepared sample exhibits the broad halo along with some weak embedded crystalline peaks. These peaks are indexed with crystalline rhombohedral phases of Al2O3 (ICDD card no.00-042-1468). The volume fraction of rhombohedral Al2O3 crystalline phase is less than 2% as calculated from fullprof XRD software. It is also supported by ICP analysis as given in Table 2. On the other hand, MOD-sample exhibits crystobalite crystalline phase with ICDD card no-(01-075-0923), which is given in Fig. 4(c). This crystalline phase is very detrimental for sealing point of view, since during thermal cycle, this phase showed large volume change led to create thermal stress at the interface of Crofer 22 APU and glass seal. After exposed at 900 °C for 500 h, the interface powder is taken from the diffusion couple for further experimentation. The XRD pattern of
3.4. FTIR analysis FTIR spectra are taken in the range of 4000–400 cm−1 of the asprepared and exposed (900 °C/500 hh) samples as shown in Fig. 5(a)(c). FTIR spectra of as-prepared as well as exposed samples at 900 °C for 500 h exhibit some striking differences in the bands. For comparison, the FTIR spectra of MOD sample is also given in Fig. 5 (c). In case of asquenched and exposed samples, the bands are diffused particularly in as-quenched sample. Exposed and MOD (as-quenched) samples exhibit sharp bands, particularly in MOD sample. Obviously, MOD sample is exhibits fully crystobalite crystalline phase. In general, the broadness of bands is related to the presence of different structure units of glassformer (SiO2) in the sample. However, in both the case, these bands are weak at the higher wave number. Most of the strong bands are present
Fig. 6. SEM images of diffusion couple of WSA glass-ceramic with Crofer 22 APU: heat-treated for: (a) 10 h, (b) 100 h and (c) 500 h. 4
Ceramics International xxx (xxxx) xxx–xxx
G. Sharma and K. Singh
Fig. 7. Dot profile of different elements across the diffusion couple of WSA and Crofer 22 APU after heat-treatment at 900 °C for 500 h.
in the fingerprint region. The majority of bands belong to the SiO4 polyhedra along with the hydroxyl group related bands in both the spectra. The band at 466-472 cm−1 belongs to Si–O–Si bending vibration [13]. The band at 790-782 cm−1 belongs to O–Si–O stretching vibration [31]. The strong band at 1062-1094 cm−1 is an asymmetric stretching vibration band of Si–O–Si [32,33]. The band at about, 16201629 cm−1 is the flexural vibration band of absorbed water H–OH [34]. The bands at 2997 cm−1 indicate the bending of Si–OH groups. The band at 3482-3414 cm−1 is arisen due to water absorbed by the sample. In comparison to the as-prepared sample with the exposed glassceramic sample exhibits two other weak bands at 685 and 621 cm−1. The band at 685 cm−1 belongs to inner vibration of Si–O–Si [35]. On the other hand, 621 cm−1 bands are associated with chromium oxide. Additionally, all the bands shift towards the higher wavenumber in an exposed sample. It is associated with the diffusion of Cr and Fe from
Crofer 22 APU to glass sealant. Since bonds between diffused species (Cr) and oxygen are stronger than bonds between silicon and oxygen, which will shift bands towards the higher wave number. The presence of the hydroxyl group, as observed in the FTIR spectra, it is responsible to promote the bonding between glass-sealant and interconnect. Initially, it may provide wettability between sealant and interconnect.
3.5. Resistivity analysis Glass-sealant must be an insulator at the working temperature of SOFC. Therefore, insulating nature of present WSA glass-ceramic is investigated at the working temperature of SOFC. The total resistance of the sample was estimated with the help of Nyquist plots. The total resistivity of the sample is observed ~106 Ω-cm at the 700 °C. The activation energy is also calculated from Arrhenius plot, i.e. ~0.35 eV, 5
Ceramics International xxx (xxxx) xxx–xxx
G. Sharma and K. Singh
Fig. 8. Line profile of the different elements across the diffusion couple heat-treated at 900 °C after 500 h.
3.6. SEM/EDS analysis The preparation of diffusion couple is shown in schematic diagram in Fig. 1. A detailed investigation on the interface between glassceramic and Crofer 22 APU has been carried out after exposing diffusion couple at 900 °C for different time periods. Interestingly, MOD sample did not form any interface with interconnect material, So, SEM study couldn't conducted on this diffusion couple. Fig. 6 shows the SEM micro graphs of the interface of glass-ceramic with Crofer 22 APU heattreated at 900 °C for 10, 100 and 500 h exposed time duration. Initially, diffuse dendrite growth is clearly visible to the glass-ceramic side. As the time duration increase from 10 h to 100 h, it becomes clear and long in length as marked in Fig. 6 (a) and (b). Finally, in 500 h of exposure time, it becomes longer and clearer as shown in Fig. 6 (c). The dendrite types light gray rods correspond to tridymite crystalline phase. It is also supported by XRD. The interface is very smooth and free from porosity even after 500 h exposure with the five different thermal cycles. It is unusual since SiO2 rich glasses and glass-ceramic usually form crystoballite and quartz crystalline phases after exposing. The formation of these phases is very detrimental to SOFCs application point of view due to the big change in their volume fraction, particularly, crystoballite phase during thermal cycles. The big change in their volume creates thermal stress at the interface leads delamination of the interface. In case of present study, a minimum chemical reaction between interconnect and glass-sealant has taken place. It is showing good compatibility between both the components. Basically, once bonds are formed between SOFC components and glass-sealant then vigorous reaction may form some crystalline phases, they change the CTE, which may delaminate the interface. The dot mapping of the elemental distribution has been carried out and shown in Fig. 7. In case of chromium distribution, very interesting feature is observed, after uniform and equal distribution up to 160 μm followed by less dense and equally distributed chromium in the glass sealant. Similarly, calcium diffused from 200 μm glass side to interconnect side, whereas, alkali metals like K+ and Na+ are equally distributed throughout the diffusion couple. Silicon and oxygen clearly show the interface boundaries in dot mapping of these elements. Furthermore, to obtained clear picture. The line profile of different elements across the diffusion couple is also
Fig. 9. Hardness of glass-ceramic, interface and Crofer 22 APU and glassceramic at different point.
Table 3 Hardness data of Crofer 22 APU, interface and WSA glass-ceramic after heattreatment 500 h at 900 °C different points in Fig. 8. Hardness (HV) of the Crofer 22 APU, interface and glass-ceramic Crofer 22 APU
Interface
Glass-ceramic
156 190 130
237 384 343
439 490 475
which indicates the ion conducting nature of the present sample. The resistivity is in the insulating range, which is required for the sealing proposes in SOFC.
6
Ceramics International xxx (xxxx) xxx–xxx
G. Sharma and K. Singh
performed as shown in Fig. 8. It verifies from the dot mapping results. Interestingly, the present WSA glass-ceramic exhibits good bonding and smooth interface even after 500 h. It seems that the presence of some trace elements may play a crucial role in formation of interface. However, the role of trace elements in agricultural waste derived glass and glass-ceramic is not well-known yet.
[4] A.A. Reddy, D.U. Tulyaganov, S. Kapoor, A. Goel, M.J. Pascual, V. V Kharton, J.M.F. Ferreira, Study of melilite based glasses and glass-ceramics nucleated by Bi2O3 for functional applications, RSC Adv. 2 (2012) 10955–10967, https://doi. org/10.1039/c2ra22001f. [5] V. Kumar, S. Sharma, O.P. Pandey, K. Singh, Thermal and physical properties of 30SrO-40SiO2-20B2O3-10A2O3 (A= La,Y,Al) glasses and their chemical reaction with bismuth vanadate for SOFC, Solid State Ionics 181 (2010) 79–85, https://doi. org/10.1016/j.ssi.2009.12.005. [6] Y. Guo, R. Ran, Z. Shao, A novel way to improve performance of proton-conducting solid-oxide fuel cells through enhanced chemical interaction of anode components, Int. J. Hydrogen Energy 36 (2011) 1683–1691, https://doi.org/10.1016/j.ijhydene. 2010.10.081. [7] J.W. Fergus, Sealants for solid oxide fuel cells, J. Power Sources 147 (2005) 46–57, https://doi.org/10.1016/j.jpowsour.2005.05.002. [8] M. Reben, H. Li, Thermal stability and crystallization kinetics of MgO-Al2O3-B2O3SiO2 glasses, Int. J. Appl. Glass Sci. 2 (2011) 96–107, https://doi.org/10.1111/j. 2041-1294.2011.00039.x. [9] G. Sharma, K. Singh, Recycling and utilization of agro-food waste ashes: syntheses of the glasses for wide-band gap semiconductor applications, J. Mater. Cycles Waste Manag. 10 (2019) 1–9, https://doi.org/10.1007/s10163-019-00839-z. [10] G. Kaur, O.P. Pandey, K. Singh, Chemical interaction study between lanthanum based different alkaline earth glass sealants with Crofer 22 APU for solid oxide fuel cell applications, Int. J. Hydrogen Energy 37 (2011) 3883–3889, https://doi.org/ 10.1016/j.ijhydene.2011.04.104. [11] A.K. Yadav, P. Singh, A review of structure of oxide glasses by Raman spectroscopy, RSC Adv. 5 (2015) 67583–67609, https://doi.org/10.1039/c0xx00000x. [12] J.B. Goodenough, Y.H. Huang, Alternative anode materials for solid oxide fuel cells, J. Power Sources 173 (2007) 1–10, https://doi.org/10.1016/j.jpowsour.2007.08. 011. [13] S.S. Danewalia, G. Sharma, S. Thakur, K. Singh, Agricultural wastes as a resource of raw materials for developing low-dielectric glass-ceramics, Sci. Rep. 6 (2016) 24617, https://doi.org/10.1038/srep24617. [14] I.A. Cornejo, S. Ramalingam, J.S. Fish, I.E. Reimanis, Hidden treasures : turning food waste into glass, Am. Ceram. Soc. Bull. 93 (2014) 24–27. [15] B.S. Todkar, O.A. Deorukhkar, S.M. Deshmukh, Extraction of silica from rice husk ash, Int. J. Eng. Res. Dev. 12 (2016) 69–74. [16] S.R. Kiran, V. Gupta, Experimental study on concrete when cement is partially replace by fly ash, wheat straw ash, rice husk ash, saw dust ash, Int. J. Sci. Res. Dev. 3 (2015) 311–313. [17] A. Shrivas, D. Jain, R. Joshi, Utilisation of waste plastics as a partial replacement of coarse aggregate in concrete blocks, Int. J. Sci. Technol. Eng. 2 (2015) 89–107, https://doi.org/10.17485/ijst/2015/v8i12/54462. [18] C. Cobreros, J.L. Reyes-Araiza, A. Manzano-ramírez, R. Nava, M. Rodríguez, M. Mondragón-Figueroa, L.M. Apátiga, E.M. Rivera-Muñoz, Barley straw ash: pozzolanic activity and comparison with other natural and artificial pozzolans from Mexico, BioResources 10 (2015) 3757–3774, https://doi.org/10.15376/biores.10. 2.3737-3774. [19] F. Naghizadeh, M.R. Abdul Kadir, A. Doostmohammadi, F. Roozbahani, N. Iqbal, M.M. Taheri, S.V. Naveen, T. Kamarul, Rice husk derived bioactive glass-ceramic as a functional bioceramic: synthesis, characterization and biological testing, J. NonCryst. Solids 427 (2015) 54–61, https://doi.org/10.1016/j.jnoncrysol.2015.07.017. [20] R. Yuvakkumar, V. Elango, V. Rajendran, N. Kannan, High-purity nano silica powder from rice husk using a simple chemical method, J. Exp. Nanosci. 9 (2012) 272–281, https://doi.org/10.1080/17458080.2012.656709. [21] G. Sharma, S.K. Arya, K. Singh, Optical and thermal properties of glasses and glassceramics derived from agricultural wastes, Ceram. Int. 44 (2018) 947–952, https:// doi.org/10.1016/j.ceramint.2017.10.027. [22] P. Sharma, K.L. Singh, A.P. Singh, C. Sharma, S. Mago, The influence of various dopants and sintering techniques on the properties of the yttria-ceria-zirconia system as an electrolyte for solid oxide fuel cells, J. Electron. Mater. 48 (2019) 3527–3536, https://doi.org/10.1007/s11664-019-07094-w. [23] D. Kuščer, M. Hrovat, J. Holc, S. Bernik, D. Kolar, Some characteristics of Al2O3and CaO-modified LaFeO3-based cathode materials for solid oxide fuel cells, J. Power Sources 61 (1996) 161–165, https://doi.org/10.1016/S0378-7753(96) 02364-6. [24] M. Garai, N. Sasmal, B. Karmakar, Effects of M2+ ( M = Ca, Sr, and Ba) addition on crystallization and microstructure of SiO2 -MgO-Al2O3-B2O3-K2O-F glass, Indian J. Mater. Sci. 2015 (2015) 1–7. [25] M. Tiegel, R. Hosseinabadi, S. Kuhn, A. Herrmann, C. Rüssel, Young's modulus, Vickers hardness and indentation fracture toughness of alumino silicate glasses, Ceram. Int. 41 (2015) 7267–7275, https://doi.org/10.1016/j.ceramint.2015.01. 144. [26] P. Jha, S.S. Danewalia, G. Sharma, K. Singh, Antimicrobial and bioactive phosphatefree glass–ceramics for bone tissue engineering applications, Mater. Sci. Eng. C 86 (2018) 9–17. [27] N. Lahl, K. Singh, L. Singheiser, K. Hilpert, Crystallisation kinetics in AO-Al2O3SiO2-B2O3 glasses (A=Ba,Ca,Mg), J. Mater. Sci. 35 (2000) 3089–3096. [28] M. Elisa, B.A. Sava, A. Diaconu, L. Boroica, D. Ursu, I. Stamatin, F. Nastase, C. Nastase, C. Logofatu, Thermal properties of ecological phosphate and silicate glasses, Glas, Phys. Chem. 35 (2010) 596–601, https://doi.org/10.1134/ s108765960906008x. [29] A. Nurur, H. Nakazawa, Thermal changes in monoclinic tridymite, Am. Mineral. 63 (1978) 1252–1259 doi:10003-004x/7 8/1112-1252502. [30] S.H. Javed, A.S.A. Umair, S. Tajwar, Precipitated silica from wheat husk, J. Pak. Inst. Chem. Eng. 39 (2011) 51–54. [31] M.I.M. Zamratul, A.W. Zaidan, A.M. Khamirul, M. Nurzilla, S.A. Halim, Formation,
3.7. Hardness of interface The micro hardness is measured across the interface at the different point as shown in Fig. 9. The micro hardness of the Crofer 22 APU side varies from 130 to 190 HV. It is similar to earlier reports for the Crofer 22 APU. On the other hand, at the interface of glass-ceramic and Crofer 22 APU, the micro hardness increases and become 237 to 384 HV shown in Table 3. At the boundary of glass-ceramic and Crofer 22 APU, micro hardness is less than glass-ceramic and higher than the Crofer 22 APU. This clearly indicates that the bonding between glass-ceramic and Crofer 22 APU is semi-coherent in nature [36]. Towards the glassceramic side, micro hardness is higher than the interface and Crofer 22 APU, i.e. 440–490 HV. Silicate glasses usually have the micro hardness in above-said range, whereas borate glasses have higher micro hardness. In other words, the micro hardness test clearly shows continuous and superior bonding between Crofer 22 APU and glass sealant even after 500 h exposed of the diffusion couple. Conclusively, the glassceramic derived from agricultural waste can be used as a sealant in SOFCs. 4. Conclusions The comparative study of the wheat straw ash and mineral oxides based glass-ceramic have been done in the light of structural, mechanical and interaction behavior with Crofer 22 APU for SOFC application. WSA exhibits good thermal stability with insulating nature even at 700 °C. The CTE of present WSA glass-ceramic is in required range for SOFC application ~10 × 10−6/°C. The major band of SiO2 polyhedra with the presence of the hydroxyl group, which supports the wetting and adhesion between interconnect and WSA. While mineral oxides derived glass-ceramic of similar composition as WSA sample could not form the interface with interconnect. Semi-coherent interface is strong, smooth and well adhere to Crofer 22 APU even after five thermal cycles for the different time duration. The formation of thin chromium layer (160 μm) at the interface as well as the presence of some trace elements might be responsible for strong bonding between Crofer 22 APU and glass-ceramic sealant derived from WSA. The present study demonstrates that agricultural derived glass-ceramic over the mineral oxides based glass-ceramic can be promising sealant materials for SOFC's applications. The maximum micro hardness at the interface is 384 HV. Acknowledgement Authors are gratefully thankful to Dr. Devender Kumar (Mechanical department, TIET) and for measuring the hardness and Neetu Bansal for discussed the results. One of the authors (Gaurav Sharma) is thankful to SAI Labs, Thapar Institute of Engineering and Technology, Patiala, for providing the characterization techniques. References [1] M.K. Mahapatra, K. Lu, Seal glass for solid oxide fuel cells, J. Power Sources 195 (2010) 7129–7139, https://doi.org/10.1016/j.jpowsour.2010.06.003. [2] J.M. Haag, D.M. Bierschenk, S.A. Barnett, K.R. Poeppelmeier, Structural, chemical, and electrochemical characteristics of LaSr2Fe2CrO9-δ-based solid oxide fuel cell anodes, Solid State Ionics 212 (2012) 1–5, https://doi.org/10.1016/j.ssi.2012.01. 037. [3] N. Lahl, D. Bahadur, K. Singh, L. Singheiser, K. Hilpert, Chemical interactions between aluminosilicate base sealants and the components on the anode side of solid oxide fuel cells, J. Electrochem. Soc. 149 (2002) 607–614, https://doi.org/10. 1149/1.1467945.
7
Ceramics International xxx (xxxx) xxx–xxx
G. Sharma and K. Singh
[34] V.G. Plotnichenko, V.O. Sokolov, E.M. Dianov, Hydroxyl groups in high-purity silica glass, J. Non-Cryst. Solids 261 (2000) 186–194, https://doi.org/10.1016/S00223093(99)00654-7. [35] P. Thuadaij, A. Nuntiya, Preparation and characterization of faujasite using fly ash and amorphous silica from rice husk ash, Procedia Eng. 32 (2012) 1026–1032, https://doi.org/10.1016/j.proeng.2012.02.049. [36] K.K. Chawla, Interface, Composite Materials, Science and Engineering vol. 50, Springer, 2013, pp. 105–133.
structural and optical characterization of neodymium doped-zinc soda lime silica based glass, Results Phys. 6 (2016) 295–298, https://doi.org/10.1016/j.rinp.2016. 05.014. [32] A.G. Kalampounias, IR and Raman spectroscopic studies of sol-gel derived alkalineearth, Bull. Mater. Sci. 34 (2011) 299–303. [33] P. Jha, K. Singh, Effect of MgO on bioactivity, hardness, structural and optical properties of SiO2-K2O-CaO-MgO glasses, Ceram. Int. 42 (2015) 436–444, https:// doi.org/10.1016/j.ceramint.2015.08.128.
8
Contents lists available at ScienceDirect
Ceramics International journal homepage: www.elsevier.com/locate/ceramint
Agro-waste ash and mineral oxides derived glass-ceramics and their interconnect study with Crofer 22 APU for SOFC application Gaurav Sharma, K. Singh* School of Physics and Materials Science, Thapar Institute of Engineering and Technology, Patiala, 147004, India
A R T I C LE I N FO
A B S T R A C T
Keywords: Glass-ceramic Interaction Sealant Solid oxide fuel cells X-ray diffraction
In this study, silicate glass-ceramic is synthesized from wheat straw ash (WSA) using melt-quench technique. Xray diffraction is done to confirm the nature of the as-quenched samples. Inductive coupled plasma-mass spectroscopy (ICP-MS) and energy dispersive spectroscopy (EDS) are performed to evaluate the exact chemical composition of the as-prepared samples. Similar composition is also melted using conventional mineral oxides for comparison with WSA sample. The WSA sample shows good thermal stability in temperature range of 300–900 °C. The coefficient of thermal expansion (CTE) of WSA sample is more or less in the required range for solid oxide fuel cells (SOFC) application. The powder of WSA derived glass-ceramic is used to form the diffusion couple with the interconnect (Crofer 22 APU) for interfacial study. The excellent adhesion with the smooth and porosity-free interface is formed between glass-ceramic and Crofer 22 APU, when WSA is used as a sealant as compared to the mineral based glass-ceramics. Semi-coherent interface does not show any delimitation or separation even after five thermal cycles. The micro hardness (384 HV) of the interface is indicated good bonding between interconnect and glass-ceramic.
1. Introduction
may be initiated the thermal stress at the interface, and it reduces the overall efficiency of SOFC. Therefore, minimum reaction should be taken place between glass-sealants and other components of SOFC [11,12]. All the required components, interconnect is a very crucial and vital factor to retain the performance of SOFC. Generally, chromiumrich steel (Crofer 22 APU) is being used as an inter-connector in SOFC. Thus, WSA glass-ceramic and Crofer 22 APU have been selected to check the interaction study with numerous thermal cycles. Additionally, similar chemical composition is also melted in similar condition for comparison the results with WSA sample. On the other hand, the existing waste materials draw the major interest of the scientific community to find out the effective way to utilization for suitable applications in the engineering fields, which also reduces waste management-related problems [13–15]. In general, the ashes of agricultural waste are being used in the civil and construction due to their low-density and small particle size with higher content of silica with some other trace elements [16–18]. It is well-reported in the literature that the extracted silica with other oxides from agricultural waste is exploited to form glasses and glass-ceramic [19–21]. Mostly sealing glasses have 40–60% SiO2 as network-former along with alkaline-earth metal oxides 10–30 (mol%) [10]. Additionally, some intermediate oxides like Al2O3, Y2O3, TiO2 and ZrO2, etc., are also being used to control viscosity, CTE and
Sealing material is very important and essential part of the planar solid oxide fuel cell (SOFC) [1]. Generally, sealant is applied at the edges of the flat plates [2,3]. SOFC device is required high operating temperature, so, the conventional sealing materials like polymer, alloys and organic adhesive cannot sustain at the working temperature of SOFC (800–1000 °C) [4–6]. That why, various types of glasses and glass-ceramics are typically used as the sealant materials to improve the lifespan of the SOFC after many thermal cycles at the higher temperature. The quality of sealing materials must be high, even miner leakage can affect the cell potential, and it can reduce the overall performance of the device [7]. However, silicate glasses and glass-ceramics are always preferable as compare to other formers based glasses and glassceramics like B2O3 and P2O5 for this purpose due to better thermal stability, particularly, in variable oxygen partial pressure and presence of water vapor [5,8,9]. Basically, glass-sealant made an interface with all the components in SOFC's [10]. Furthermore, formed interface usually delaminates during the thermal cycles or long run of SOFC due to the thermal expansion mismatch between different components of SOFC. In addition to this, due to high working temperature of SOFC, some crystalline phases are also formed due to inter-diffusion and intradiffusion among SOFC components. The observed detrimental phases *
Corresponding author. E-mail address: [email protected] (K. Singh).
https://doi.org/10.1016/j.ceramint.2019.07.029 Received 28 May 2019; Accepted 2 July 2019 0272-8842/ © 2019 Elsevier Ltd and Techna Group S.r.l. All rights reserved.
Please cite this article as: Gaurav Sharma and K. Singh, Ceramics International, https://doi.org/10.1016/j.ceramint.2019.07.029
Ceramics International xxx (xxxx) xxx–xxx
G. Sharma and K. Singh
Table 1 The chemical composition of WSA (wt%) analyzed by EDS and similar composition of mineral derived sample. Label
SiO2
CaO
Na2O
K2O
MgO
WSA MOD-sample
85 85
8 8
0.82 0.82
4.85 4.85
4.85 4.85
characteristic temperatures of the glasses [7,22–24]. However, advanced and better performing seal materials are still required, which can sustain for longer time without having any detrimental reaction with the other components of SOFC. The interfacial behavior between components and glass-sealants are needed to know the thermal stability, interface formation its growth and interfacial behavior with respect to the thermal cycles before using sealant in real SOFC conditions. Usually, it is considered that alkali oxide like K2O and Na2O, etc. increase the CTE of glasses. Contrary to this, they are also diffuse very fast, which can cause drastic drop in electrical resistance of glass-sealants at the higher temperature. Alkali metals oxides also enhanced volatility of the chromium and poisoning of cathode [10]. In the present study, the attempt has been made to form the glass-ceramic from wheat straw ash (WSA) and conventional mineral oxides derived glass-ceramic i-e., MOD-sample. The WSA sample is unique, and new in the sense that the agricultural waste derived sample probable the first time investigated as the sealing material. However, exact mineral based glassceramic is not any kind of suitability for this application. In the present study, agricultural waste derived glass-ceramic is revealed better adhesion with interconnect than mineral oxides synthesized glass-ceramics. This study is performed in order to understand WSA based glassceramic and MOD glass-ceramic sealing properties in SOFC's.
Fig. 2. Thermal gravimetric (TGA) curve of the as-quenched WSA sample.
crucible at 1550 °C in the programmable electric-muffle furnace using a heating rate of 5 °C/min. The molten substance is quenched in the air from 1550 °C to room temperature. The thermal gravimetric measurement (TGA) of the WSA sample was performed with NETZSCH STA449 F3 Jupiter in the air atmosphere at the heating rate of 10 °C/min from 50-950 °C. During the experiment, the partial pressure of the air was 5 kg/cm2. The temperature and weight loss detection limit of the instrument are ± 1 °C and 0.001 mg, respectively. The coefficient of thermal expansion (CTE) of the as-quenched WSA sample was determined by using Netzsch DIL402 PC in the temperature range 30–850 °C in the air atmosphere. The amorphous and crystalline nature of the samples was checked by X-ray diffraction. The XRD pattern was obtained by PANalytical's X'Pert Pro X-ray diffractrometer. During experiment, the scan speed was 0.001°/minutes. Fourier transforms infrared (FTIR) spectra of the as-quenched WSA, MOD-sample and powder of glass-ceramic of WSA sample from interface was obtained on the Perkin Elmer-Spectrum-RX-IFTIR spectrometer from 4000400 cm−1 range. For FTIR measurement, the pellets were made of the
2. Experimental details Two samples (WSA and MOD) were synthesized using wheat straw ash and exact same chemical composition via minerals, respectively as given in Table 1. Initially, wheat straw ash is processed at 1000 °C for 1 h to remove any volatile substance. Further, both the samples were made in similar experimental conditions as follows. The powder was ground in an agate-mortar pestle and melted in re-crystallized alumina
Fig. 1. Schematic design of diffusion couple of glass-ceramic sample with Crofer 22 APU. 2
Ceramics International xxx (xxxx) xxx–xxx
G. Sharma and K. Singh
100–800 °C to know the insulating nature of the heat-treated sample. A circular pellet is made from glass-ceramic powder and heat-treated at 900 °C for 10 h. The heat-treated pellet is gold sputtered to make the proper electrical contact. The powder of the as-prepared sample was mixed with polyvinyl alcohol (PVA) and put as a sandwich between two ultrasonically cleaned Crofer 22 APU sheets of size 1.0 × 1.0 × 0.2 cm as shown in Fig. 1 to make diffusion couple. This diffusion couple was kept at 900 °C for 1, 2, 10, 100 and 500 h. SEM measurement was performed by JOEL-JSM 6460 LV (Tokyo, JAPAN) microscope at the acceleration voltage of 15 kV. Prior to the observations, surface was coat with a gold layer of about 100 Å thickness to avoid the charging under electron beam. The energy dispersive spectroscopy analysis was carried out using OXFORD instrument INCA-X act. The micro hardness test was performed on Mitutoyo micro hardness tester (MVK-HO, Japan) on the polished and ultrasonically cleaned diffusion couple. The indentations were made at 49 and 98 N load with 10 s dwell time. The micro hardness of various places of diffusion couple was measured by using the following equation [25,26].
Fig. 3. Coefficient of thermal expansion (CTE) of wheat straw ash (WSA) sample.
HV = 1.854(F/d2) Where, F and d are applied load and area of indentation, respectively. 3. Results and discussion 3.1. EDS and ICP analysis The chemical composition of the processed wheat straw ash was estimated by energy dispersive spectroscopy (EDS). To confirm the exact and precise chemical composition of the as-quenched WSA sample was calculated using inductive coupled plasma (ICP-MS) technique. The elemental percentage of the present sample is more or less similar of minerals derived glass-ceramics. However, both samples have the higher amount of SiO2 with alkali and alkaline earth-metal, but some very miner trace elements are presented in the WSA composition inherently as compared to MOD sample. 3.2. Thermogravimetric (TG) and CTE analysis The thermal stability of the given WSA sample is checked using TGA experiment as shown in Fig. 2. Initially, sample exhibits some weight loss i.e., 1.5% up to 350 °C. This weight loss is related to water molecules absorbed by the sample. After that, TGA curve do not show any trend with respect to temperature i-e., 350–950 °C, it indicates the good thermal stability of WSA sample. Additionally, the coefficient of thermal expansion (CTE) is very essential parameter for sealing applications; normally, it should be in the range of 9–13 × 10−6/°C for SOFC applications [1,7,27]. The CTE of as-prepared sample is 10 × 10−6/°C as shown in Fig. 3. Usually, silica-rich glasses have low CTE, this amount of silica is deciding factor for CTE in silicate glasses [28]. The presence of the modifying elements in the samples can increase the CTE due to the modifying overall structure of the glass-network means the silicate-network. However, the small amount i-e., only ~4 wt% of the modifiers is found in the WSA sample. XRD of the melt-quench sample (discussed in the next section) shows the presence of tridymite crystalline phase. The tridymite phase has higher CTE as reported by Nurur et al. [29]. Thus, CTE of the present sample may be arisen due to the presence of the crystalline phase as well as some trace elements and their bonding with glass-matrix.
Fig. 4. XRD pattern of (a) as-prepared sample, (b) after chemical interaction at 900 °C for 500 h and (c) MS-sample derived using mineral oxides. Table 2 Weight percentage (%) of the present oxides in the as-quenched glass-ceramic of WSA and MOD sample calculated using ICP and EDS. Compounds
SiO2 CaO Na2O MgO K2O Al2O3
Elemental analysis (wt %) WSA (ICP)
WSA (EDS)
84.94 8.02 0.62 0.82 4.85 0.72
84.53 4.85 1.46 1.83 4.85 2.40
MOD-sample (EDS)
85.40 9.50 1.25 0.55 2.25 1.05
3.3. X-ray diffraction analysis The nature of the as-quenched samples derived from WSA and MOD sample are checked using X-ray diffraction. Additionally, the powder of both the samples is used to make the diffusion couples (as shown in experimental section) for interaction study. After exposing for different
sample's powder with KBr in the 1:1 wt ratio. The impedance measurement performed using SOLATRON (1260S) impedance spectroscopy using frequency from 102-106 Hz range for temperature range of 3
Ceramics International xxx (xxxx) xxx–xxx
G. Sharma and K. Singh
exposed WSA sample (at 900 °C for 500 h) is shown in Fig. 4 (b). The XRD pattern is indexed with the tridymite crystalline SiO2 phases. Interestingly, the Al2O3 crystalline phase, presented in the initial glassceramic, is completely either suppressed or dissolved in the matrix as indicated by XRD of 500 h exposed sample. Many researchers have reported that sometime meta-stable crystalline phase may dissolve to form new crystalline phase in prolonged heat-treatment of glasses and glass-ceramics [5,30]. As heat-treatment evolved, either it dissolves or converts into more stable crystalline phase. The SiO2 crystobalite phase is usually detrimental in SOFCs application point of view since large volume change occurs during thermal cycles, which led the thermal stresses. However, in the present case, the crystobalite phase is not formed. Moreover, In WSA sample, no other mixed crystalline phases are formed even after 500 h exposed duration. Usually, chemical interaction for longer duration leads to the formation of many crystalline phases in the multi-component glasses/glass-ceramics with other components of SOFCs when glasses are made of mineral oxides. Conclusively, minimum reaction is taken place without forming any detrimental crystalline phase at the interface between sample and Crofer 22 APU. In this respect, the wheat straw ash derived glass-ceramic seems to be better alternate sealing materials than conventional glass-sealant formed using mineral oxides.
Fig. 5. FTIR spectra of (a) as-prepared WSA (b) exposed glass-ceramic and (c) MOD as-quenched sample.
thermal cycles, the powders were taken from diffusion couple and investigated by XRD. X-ray diffraction patterns are given in Fig. 4(a) and (b). The X-ray diffraction pattern of the as-prepared sample exhibits the broad halo along with some weak embedded crystalline peaks. These peaks are indexed with crystalline rhombohedral phases of Al2O3 (ICDD card no.00-042-1468). The volume fraction of rhombohedral Al2O3 crystalline phase is less than 2% as calculated from fullprof XRD software. It is also supported by ICP analysis as given in Table 2. On the other hand, MOD-sample exhibits crystobalite crystalline phase with ICDD card no-(01-075-0923), which is given in Fig. 4(c). This crystalline phase is very detrimental for sealing point of view, since during thermal cycle, this phase showed large volume change led to create thermal stress at the interface of Crofer 22 APU and glass seal. After exposed at 900 °C for 500 h, the interface powder is taken from the diffusion couple for further experimentation. The XRD pattern of
3.4. FTIR analysis FTIR spectra are taken in the range of 4000–400 cm−1 of the asprepared and exposed (900 °C/500 hh) samples as shown in Fig. 5(a)(c). FTIR spectra of as-prepared as well as exposed samples at 900 °C for 500 h exhibit some striking differences in the bands. For comparison, the FTIR spectra of MOD sample is also given in Fig. 5 (c). In case of asquenched and exposed samples, the bands are diffused particularly in as-quenched sample. Exposed and MOD (as-quenched) samples exhibit sharp bands, particularly in MOD sample. Obviously, MOD sample is exhibits fully crystobalite crystalline phase. In general, the broadness of bands is related to the presence of different structure units of glassformer (SiO2) in the sample. However, in both the case, these bands are weak at the higher wave number. Most of the strong bands are present
Fig. 6. SEM images of diffusion couple of WSA glass-ceramic with Crofer 22 APU: heat-treated for: (a) 10 h, (b) 100 h and (c) 500 h. 4
Ceramics International xxx (xxxx) xxx–xxx
G. Sharma and K. Singh
Fig. 7. Dot profile of different elements across the diffusion couple of WSA and Crofer 22 APU after heat-treatment at 900 °C for 500 h.
in the fingerprint region. The majority of bands belong to the SiO4 polyhedra along with the hydroxyl group related bands in both the spectra. The band at 466-472 cm−1 belongs to Si–O–Si bending vibration [13]. The band at 790-782 cm−1 belongs to O–Si–O stretching vibration [31]. The strong band at 1062-1094 cm−1 is an asymmetric stretching vibration band of Si–O–Si [32,33]. The band at about, 16201629 cm−1 is the flexural vibration band of absorbed water H–OH [34]. The bands at 2997 cm−1 indicate the bending of Si–OH groups. The band at 3482-3414 cm−1 is arisen due to water absorbed by the sample. In comparison to the as-prepared sample with the exposed glassceramic sample exhibits two other weak bands at 685 and 621 cm−1. The band at 685 cm−1 belongs to inner vibration of Si–O–Si [35]. On the other hand, 621 cm−1 bands are associated with chromium oxide. Additionally, all the bands shift towards the higher wavenumber in an exposed sample. It is associated with the diffusion of Cr and Fe from
Crofer 22 APU to glass sealant. Since bonds between diffused species (Cr) and oxygen are stronger than bonds between silicon and oxygen, which will shift bands towards the higher wave number. The presence of the hydroxyl group, as observed in the FTIR spectra, it is responsible to promote the bonding between glass-sealant and interconnect. Initially, it may provide wettability between sealant and interconnect.
3.5. Resistivity analysis Glass-sealant must be an insulator at the working temperature of SOFC. Therefore, insulating nature of present WSA glass-ceramic is investigated at the working temperature of SOFC. The total resistance of the sample was estimated with the help of Nyquist plots. The total resistivity of the sample is observed ~106 Ω-cm at the 700 °C. The activation energy is also calculated from Arrhenius plot, i.e. ~0.35 eV, 5
Ceramics International xxx (xxxx) xxx–xxx
G. Sharma and K. Singh
Fig. 8. Line profile of the different elements across the diffusion couple heat-treated at 900 °C after 500 h.
3.6. SEM/EDS analysis The preparation of diffusion couple is shown in schematic diagram in Fig. 1. A detailed investigation on the interface between glassceramic and Crofer 22 APU has been carried out after exposing diffusion couple at 900 °C for different time periods. Interestingly, MOD sample did not form any interface with interconnect material, So, SEM study couldn't conducted on this diffusion couple. Fig. 6 shows the SEM micro graphs of the interface of glass-ceramic with Crofer 22 APU heattreated at 900 °C for 10, 100 and 500 h exposed time duration. Initially, diffuse dendrite growth is clearly visible to the glass-ceramic side. As the time duration increase from 10 h to 100 h, it becomes clear and long in length as marked in Fig. 6 (a) and (b). Finally, in 500 h of exposure time, it becomes longer and clearer as shown in Fig. 6 (c). The dendrite types light gray rods correspond to tridymite crystalline phase. It is also supported by XRD. The interface is very smooth and free from porosity even after 500 h exposure with the five different thermal cycles. It is unusual since SiO2 rich glasses and glass-ceramic usually form crystoballite and quartz crystalline phases after exposing. The formation of these phases is very detrimental to SOFCs application point of view due to the big change in their volume fraction, particularly, crystoballite phase during thermal cycles. The big change in their volume creates thermal stress at the interface leads delamination of the interface. In case of present study, a minimum chemical reaction between interconnect and glass-sealant has taken place. It is showing good compatibility between both the components. Basically, once bonds are formed between SOFC components and glass-sealant then vigorous reaction may form some crystalline phases, they change the CTE, which may delaminate the interface. The dot mapping of the elemental distribution has been carried out and shown in Fig. 7. In case of chromium distribution, very interesting feature is observed, after uniform and equal distribution up to 160 μm followed by less dense and equally distributed chromium in the glass sealant. Similarly, calcium diffused from 200 μm glass side to interconnect side, whereas, alkali metals like K+ and Na+ are equally distributed throughout the diffusion couple. Silicon and oxygen clearly show the interface boundaries in dot mapping of these elements. Furthermore, to obtained clear picture. The line profile of different elements across the diffusion couple is also
Fig. 9. Hardness of glass-ceramic, interface and Crofer 22 APU and glassceramic at different point.
Table 3 Hardness data of Crofer 22 APU, interface and WSA glass-ceramic after heattreatment 500 h at 900 °C different points in Fig. 8. Hardness (HV) of the Crofer 22 APU, interface and glass-ceramic Crofer 22 APU
Interface
Glass-ceramic
156 190 130
237 384 343
439 490 475
which indicates the ion conducting nature of the present sample. The resistivity is in the insulating range, which is required for the sealing proposes in SOFC.
6
Ceramics International xxx (xxxx) xxx–xxx
G. Sharma and K. Singh
performed as shown in Fig. 8. It verifies from the dot mapping results. Interestingly, the present WSA glass-ceramic exhibits good bonding and smooth interface even after 500 h. It seems that the presence of some trace elements may play a crucial role in formation of interface. However, the role of trace elements in agricultural waste derived glass and glass-ceramic is not well-known yet.
[4] A.A. Reddy, D.U. Tulyaganov, S. Kapoor, A. Goel, M.J. Pascual, V. V Kharton, J.M.F. Ferreira, Study of melilite based glasses and glass-ceramics nucleated by Bi2O3 for functional applications, RSC Adv. 2 (2012) 10955–10967, https://doi. org/10.1039/c2ra22001f. [5] V. Kumar, S. Sharma, O.P. Pandey, K. Singh, Thermal and physical properties of 30SrO-40SiO2-20B2O3-10A2O3 (A= La,Y,Al) glasses and their chemical reaction with bismuth vanadate for SOFC, Solid State Ionics 181 (2010) 79–85, https://doi. org/10.1016/j.ssi.2009.12.005. [6] Y. Guo, R. Ran, Z. Shao, A novel way to improve performance of proton-conducting solid-oxide fuel cells through enhanced chemical interaction of anode components, Int. J. Hydrogen Energy 36 (2011) 1683–1691, https://doi.org/10.1016/j.ijhydene. 2010.10.081. [7] J.W. Fergus, Sealants for solid oxide fuel cells, J. Power Sources 147 (2005) 46–57, https://doi.org/10.1016/j.jpowsour.2005.05.002. [8] M. Reben, H. Li, Thermal stability and crystallization kinetics of MgO-Al2O3-B2O3SiO2 glasses, Int. J. Appl. Glass Sci. 2 (2011) 96–107, https://doi.org/10.1111/j. 2041-1294.2011.00039.x. [9] G. Sharma, K. Singh, Recycling and utilization of agro-food waste ashes: syntheses of the glasses for wide-band gap semiconductor applications, J. Mater. Cycles Waste Manag. 10 (2019) 1–9, https://doi.org/10.1007/s10163-019-00839-z. [10] G. Kaur, O.P. Pandey, K. Singh, Chemical interaction study between lanthanum based different alkaline earth glass sealants with Crofer 22 APU for solid oxide fuel cell applications, Int. J. Hydrogen Energy 37 (2011) 3883–3889, https://doi.org/ 10.1016/j.ijhydene.2011.04.104. [11] A.K. Yadav, P. Singh, A review of structure of oxide glasses by Raman spectroscopy, RSC Adv. 5 (2015) 67583–67609, https://doi.org/10.1039/c0xx00000x. [12] J.B. Goodenough, Y.H. Huang, Alternative anode materials for solid oxide fuel cells, J. Power Sources 173 (2007) 1–10, https://doi.org/10.1016/j.jpowsour.2007.08. 011. [13] S.S. Danewalia, G. Sharma, S. Thakur, K. Singh, Agricultural wastes as a resource of raw materials for developing low-dielectric glass-ceramics, Sci. Rep. 6 (2016) 24617, https://doi.org/10.1038/srep24617. [14] I.A. Cornejo, S. Ramalingam, J.S. Fish, I.E. Reimanis, Hidden treasures : turning food waste into glass, Am. Ceram. Soc. Bull. 93 (2014) 24–27. [15] B.S. Todkar, O.A. Deorukhkar, S.M. Deshmukh, Extraction of silica from rice husk ash, Int. J. Eng. Res. Dev. 12 (2016) 69–74. [16] S.R. Kiran, V. Gupta, Experimental study on concrete when cement is partially replace by fly ash, wheat straw ash, rice husk ash, saw dust ash, Int. J. Sci. Res. Dev. 3 (2015) 311–313. [17] A. Shrivas, D. Jain, R. Joshi, Utilisation of waste plastics as a partial replacement of coarse aggregate in concrete blocks, Int. J. Sci. Technol. Eng. 2 (2015) 89–107, https://doi.org/10.17485/ijst/2015/v8i12/54462. [18] C. Cobreros, J.L. Reyes-Araiza, A. Manzano-ramírez, R. Nava, M. Rodríguez, M. Mondragón-Figueroa, L.M. Apátiga, E.M. Rivera-Muñoz, Barley straw ash: pozzolanic activity and comparison with other natural and artificial pozzolans from Mexico, BioResources 10 (2015) 3757–3774, https://doi.org/10.15376/biores.10. 2.3737-3774. [19] F. Naghizadeh, M.R. Abdul Kadir, A. Doostmohammadi, F. Roozbahani, N. Iqbal, M.M. Taheri, S.V. Naveen, T. Kamarul, Rice husk derived bioactive glass-ceramic as a functional bioceramic: synthesis, characterization and biological testing, J. NonCryst. Solids 427 (2015) 54–61, https://doi.org/10.1016/j.jnoncrysol.2015.07.017. [20] R. Yuvakkumar, V. Elango, V. Rajendran, N. Kannan, High-purity nano silica powder from rice husk using a simple chemical method, J. Exp. Nanosci. 9 (2012) 272–281, https://doi.org/10.1080/17458080.2012.656709. [21] G. Sharma, S.K. Arya, K. Singh, Optical and thermal properties of glasses and glassceramics derived from agricultural wastes, Ceram. Int. 44 (2018) 947–952, https:// doi.org/10.1016/j.ceramint.2017.10.027. [22] P. Sharma, K.L. Singh, A.P. Singh, C. Sharma, S. Mago, The influence of various dopants and sintering techniques on the properties of the yttria-ceria-zirconia system as an electrolyte for solid oxide fuel cells, J. Electron. Mater. 48 (2019) 3527–3536, https://doi.org/10.1007/s11664-019-07094-w. [23] D. Kuščer, M. Hrovat, J. Holc, S. Bernik, D. Kolar, Some characteristics of Al2O3and CaO-modified LaFeO3-based cathode materials for solid oxide fuel cells, J. Power Sources 61 (1996) 161–165, https://doi.org/10.1016/S0378-7753(96) 02364-6. [24] M. Garai, N. Sasmal, B. Karmakar, Effects of M2+ ( M = Ca, Sr, and Ba) addition on crystallization and microstructure of SiO2 -MgO-Al2O3-B2O3-K2O-F glass, Indian J. Mater. Sci. 2015 (2015) 1–7. [25] M. Tiegel, R. Hosseinabadi, S. Kuhn, A. Herrmann, C. Rüssel, Young's modulus, Vickers hardness and indentation fracture toughness of alumino silicate glasses, Ceram. Int. 41 (2015) 7267–7275, https://doi.org/10.1016/j.ceramint.2015.01. 144. [26] P. Jha, S.S. Danewalia, G. Sharma, K. Singh, Antimicrobial and bioactive phosphatefree glass–ceramics for bone tissue engineering applications, Mater. Sci. Eng. C 86 (2018) 9–17. [27] N. Lahl, K. Singh, L. Singheiser, K. Hilpert, Crystallisation kinetics in AO-Al2O3SiO2-B2O3 glasses (A=Ba,Ca,Mg), J. Mater. Sci. 35 (2000) 3089–3096. [28] M. Elisa, B.A. Sava, A. Diaconu, L. Boroica, D. Ursu, I. Stamatin, F. Nastase, C. Nastase, C. Logofatu, Thermal properties of ecological phosphate and silicate glasses, Glas, Phys. Chem. 35 (2010) 596–601, https://doi.org/10.1134/ s108765960906008x. [29] A. Nurur, H. Nakazawa, Thermal changes in monoclinic tridymite, Am. Mineral. 63 (1978) 1252–1259 doi:10003-004x/7 8/1112-1252502. [30] S.H. Javed, A.S.A. Umair, S. Tajwar, Precipitated silica from wheat husk, J. Pak. Inst. Chem. Eng. 39 (2011) 51–54. [31] M.I.M. Zamratul, A.W. Zaidan, A.M. Khamirul, M. Nurzilla, S.A. Halim, Formation,
3.7. Hardness of interface The micro hardness is measured across the interface at the different point as shown in Fig. 9. The micro hardness of the Crofer 22 APU side varies from 130 to 190 HV. It is similar to earlier reports for the Crofer 22 APU. On the other hand, at the interface of glass-ceramic and Crofer 22 APU, the micro hardness increases and become 237 to 384 HV shown in Table 3. At the boundary of glass-ceramic and Crofer 22 APU, micro hardness is less than glass-ceramic and higher than the Crofer 22 APU. This clearly indicates that the bonding between glass-ceramic and Crofer 22 APU is semi-coherent in nature [36]. Towards the glassceramic side, micro hardness is higher than the interface and Crofer 22 APU, i.e. 440–490 HV. Silicate glasses usually have the micro hardness in above-said range, whereas borate glasses have higher micro hardness. In other words, the micro hardness test clearly shows continuous and superior bonding between Crofer 22 APU and glass sealant even after 500 h exposed of the diffusion couple. Conclusively, the glassceramic derived from agricultural waste can be used as a sealant in SOFCs. 4. Conclusions The comparative study of the wheat straw ash and mineral oxides based glass-ceramic have been done in the light of structural, mechanical and interaction behavior with Crofer 22 APU for SOFC application. WSA exhibits good thermal stability with insulating nature even at 700 °C. The CTE of present WSA glass-ceramic is in required range for SOFC application ~10 × 10−6/°C. The major band of SiO2 polyhedra with the presence of the hydroxyl group, which supports the wetting and adhesion between interconnect and WSA. While mineral oxides derived glass-ceramic of similar composition as WSA sample could not form the interface with interconnect. Semi-coherent interface is strong, smooth and well adhere to Crofer 22 APU even after five thermal cycles for the different time duration. The formation of thin chromium layer (160 μm) at the interface as well as the presence of some trace elements might be responsible for strong bonding between Crofer 22 APU and glass-ceramic sealant derived from WSA. The present study demonstrates that agricultural derived glass-ceramic over the mineral oxides based glass-ceramic can be promising sealant materials for SOFC's applications. The maximum micro hardness at the interface is 384 HV. Acknowledgement Authors are gratefully thankful to Dr. Devender Kumar (Mechanical department, TIET) and for measuring the hardness and Neetu Bansal for discussed the results. One of the authors (Gaurav Sharma) is thankful to SAI Labs, Thapar Institute of Engineering and Technology, Patiala, for providing the characterization techniques. References [1] M.K. Mahapatra, K. Lu, Seal glass for solid oxide fuel cells, J. Power Sources 195 (2010) 7129–7139, https://doi.org/10.1016/j.jpowsour.2010.06.003. [2] J.M. Haag, D.M. Bierschenk, S.A. Barnett, K.R. Poeppelmeier, Structural, chemical, and electrochemical characteristics of LaSr2Fe2CrO9-δ-based solid oxide fuel cell anodes, Solid State Ionics 212 (2012) 1–5, https://doi.org/10.1016/j.ssi.2012.01. 037. [3] N. Lahl, D. Bahadur, K. Singh, L. Singheiser, K. Hilpert, Chemical interactions between aluminosilicate base sealants and the components on the anode side of solid oxide fuel cells, J. Electrochem. Soc. 149 (2002) 607–614, https://doi.org/10. 1149/1.1467945.
7
Ceramics International xxx (xxxx) xxx–xxx
G. Sharma and K. Singh
[34] V.G. Plotnichenko, V.O. Sokolov, E.M. Dianov, Hydroxyl groups in high-purity silica glass, J. Non-Cryst. Solids 261 (2000) 186–194, https://doi.org/10.1016/S00223093(99)00654-7. [35] P. Thuadaij, A. Nuntiya, Preparation and characterization of faujasite using fly ash and amorphous silica from rice husk ash, Procedia Eng. 32 (2012) 1026–1032, https://doi.org/10.1016/j.proeng.2012.02.049. [36] K.K. Chawla, Interface, Composite Materials, Science and Engineering vol. 50, Springer, 2013, pp. 105–133.
structural and optical characterization of neodymium doped-zinc soda lime silica based glass, Results Phys. 6 (2016) 295–298, https://doi.org/10.1016/j.rinp.2016. 05.014. [32] A.G. Kalampounias, IR and Raman spectroscopic studies of sol-gel derived alkalineearth, Bull. Mater. Sci. 34 (2011) 299–303. [33] P. Jha, K. Singh, Effect of MgO on bioactivity, hardness, structural and optical properties of SiO2-K2O-CaO-MgO glasses, Ceram. Int. 42 (2015) 436–444, https:// doi.org/10.1016/j.ceramint.2015.08.128.
8