Preparation and characterization of NiO based nano-ceramic composites as alternative anode materials for solid oxide fuel cells (SOFCs)

Document Type : Research Paper

Authors

Department of Chemistry, Karunya Institute of Technology and Sciences, Karunya Nagar, India

10.22052/JNS.2020.03.004

Abstract

Solid oxide fuel cell (SOFC) is being developed all over the world at present as a future energy conversion device. The alternative anode materials are also being studied in order to develop efficient low temperature SOFCs (LTSOFCs) operating below 800o C. Ni‐based cermets are still the most promising anode materials and some targeted modifications are needed to improve the coking resistance and to enhance their electro-catalytic activity relatively at low temperature (~600 – 700 oC) range. Present work is aimed to develop alternate anode materials such as, NiO-Ce0.9Gd0.1O2-δ-Ce0.9Y0.1O2-δ, NiO-Ce0.8Gd0.2O2-δ-Ce0.8Y0.2O2-δ, NiO-Ce0.9Gd0.1O2-δ-Ce0.9Sm0.1O2-δ, NiO-Ce0.8Gd0.2O2-δ-Ce0.8Sm0.2O2-δ, NiO-Ce0.9Gd0.1O2-δ and NiO-Ce0.8Gd0.2O2-δ for by simple chemical precipitation route and characterize them towards application in SOFC systems. The precursor materials used in this synthesis were cerium nitrate hexahydrate, gadolinium nitrate, yttrium nitrate, samarium nitrate [as basic materials] and sodium hydroxide [as precipitator material]. C-TAB (cetyl trimethylammonium bromide) was used as a surfactant in order to avoid the agglomeration of the nanoparticles. The influence of Gd, Y and Sm doping on the phase structure development of ceria was investigated. The prepared anode materials were characterized by TGA, XRD, FTIR, particle size analysis, SEM and EDAX. The electrical characteristics of the materials were studied by electrochemical impedance spectroscopy. The results obtained were discussed in order to use the materials as alternate anode materials for SOFCs.

Keywords


 

INTRODUCTION

            Solid oxide fuel cell (SOFC) is an electrochemical device that converts the energy of a chemical reaction directly into electrical energy. Recently, alternative anode materials have been proposed notably for low temperature SOFCs (LTSOFCs) operating below 800 °C. The role of an anode in SOFC is electro-oxidation of fuel by catalyzing the reaction, and facilitating fuel access and product removal. Therefore, the anode materials should meet certain requirements reported [1,2]:

  • Stability in reducing environment.
  • Sufficient electronic and ionic conductivity.
  • Porous structure.
  • Thermal expansion coefficient (TEC) matching with electrolyte materials.
  • High catalytic activity.

            In SOFCs, anode is the place for the fuel oxidation with oxygen ions to generate H2O or CO2. Of these, the most common SOFC anode is the Ni‐based cermet, which exhibits a good compatibility with electrolytes, an excellent activity for hydrogen electro‐oxidation and a sufficient electronic conductivity for electron transfer. However, when fed with hydrocarbons, quick carbon deposition over the Ni‐based anode may be experienced due to the low coking resistance, and then, a quick cell performance degradation was observed. As a result, besides the development of new fuels for SOFCs, the rational design of the coking‐resistant anodes is also important for the commercialization of hydrocarbon‐fueled SOFCs. Recently, numerous efforts have been devoted to improve the coking resistance of hydrocarbon‐fueled SOFCs, e.g., adopting perovskite‐based anode materials, applying non‐nickel‐based anodes and modifying/decorating the Ni‐based cermet anodes [3]. The vast majority of SOFCs have a nickel anode because of its low cost compared to precious metals. The most frequently used anode materials are cermet composed of nickel and solid electrolyte, such as Ni-YSZ, Ni-SDC, targeting at maintaining porosity of anode by preventing sintering of the Ni particles and offering the anode with a TEC comparable to that of the solid electrolyte relatively lower temperature (600 oC).

Yttria-doped ceria (YDC) and samaria-doped ceria (SDC) were proposed as potential anodes for intermediate temperature SOFCs [4,5]. The common feature of these two candidate materials is, in reducing atmosphere typical of anode environment, they exhibit characteristic of mixed ionic-electronic conductivity. Similar to Ni/YSZ cermet anode, Ni/doped ceria cermet has also been formulated and tested for use as potential anode in intermediate temperature SOFCs [6]. Nanocrystalline materials such as ZnO/NiO were proposed as alternate anode materials for LTSOFC operating below 500 °C [7].It was found that nanocrystalline based anodes can exhibit good catalytic activity, high electronic and ionic conductivity in turn which may help to achieve high power output at reduced temperatures.

            It was found that nanocrystalline based anodes can exhibit good catalytic activity, high electronic and ionic conductivity in turn which may help to achieve high power output. Yang Nai-Tao et al. have proposed new Ni-Ce0.8Gd0.2O1.9 (CGO) cermet anodes for IT-SOFC application [8,9]. They found that NiO in the anodes can be reduced to Ni completely by the mixture of 20%H2 ~ 80%He at 700 °C without the phase transformation of CGO. There are various processing routes for the synthesis of ceramic composite materials. Those routes include solid state route, sol-gel route, plasma spraying, laser synthesis, hydrothermal synthesis and co-precipitation route, green synthesis,  combustion, microwave synthesis, thermal decomposition route, etc. [10-27]. In this research work, we report the preparation and characterization of a set of nanoceramic composite materials by a chemical precipitation route for application as alternate anode materials for SOFC systems.

MATERIALS AND METHODS

Chemicals

            In this experiment, various chemicals have been used to prepare the nanocrystalline anode materials. The chemicals used and their details are presented in the Table 1. The aqueous solution containing a known amount of nickel nitrate, cerium nitrate, Gd2O3, Sm2O3/Y2O3 (as basic materials, dissolved in nitric acid) and sodium hydroxide (as precipitator material) was prepared in distilled water. Gadolinium nitrate, yttrium nitrate and samarium nitrate were prepared by dissolving the required quantity of Gd2O3, Y2O3 and Sm2O3 in HNO3 respectively.  The chemicals were taken for synthesis based on the stoichiometric calucation. The amount of precursor materials used for the preparation of different nanocrystalline materials are indicated in Table 2.

Materials synthesis

Initially, the precipitating solution (sodium hydroxide) was mixed with 2 ml of 10% CTAB (as surfactant material). In order to avoid agglomeration this surfactant added to solution. To this mixture, Ni(NO3)3, Ce(NO3)3, Gd(NO3)3 and Y(NO3)3/Sm(NO3)3 solutions were subsequently added one by one drop wise. They were mixed thoroughly by a magnetic stirring apparatus (1,000 rpm) at room temperature for 2-3 hours under controlled alkaline pH range. The pH of the solution was adjusted in range of 11 – 14 by the addition of sodium hydroxide pellets[28,29]. The resultant precipitate ((Ni(OH)2 + Ce(OH)4 + Gd(OH)3+ Y(OH)3) or (Ni(OH)2 + Ce(OH)4 + Gd(OH)3+ Sm(OH)3 or (Ni(OH)2 + Ce(OH)4 + Gd(OH)3)  with CTAB was filtered and then washed with deionized water and ethanol in the ratio of 9:1 (v/v) five to ten times. The product was dried at 50 °C to 100 °C for 24 h. The resultant material was calcined at 300, 450, 600  and 750 °C for 2 hours each in air. During calcination, the surfactant was removed, and phase-pure nanocrystalline material was formed. The schematic representation of the synthesis of nanocrystalline materials by chemical precipitation method is shown in the Fig. 1.

Reaction mechanisms involved        

            The main reactions involved in the preparation of nanocrystalline ceramic composite oxide materials during the experimental procedure can be written briefly as follows:          

Reaction mechanisms involved        

            The main reactions involved in the preparation of nanocrystalline ceramic composite oxide materials during the experimental procedure can be written briefly as follows:          

Reaction mechanism involved in the preparation of NiO-Ce0.9Gd0.1O2-δ - Ce0.9Y0.1O2-δ

 

 

 

 

Reaction mechanism involved in the preparation of  NiO-Ce0.8Gd0.2O2-δ - Ce0.8Y0.2O2-δ

 

 

 

Reaction mechanism involved in the preparation of NiO-Ce0.9Gd0.1O2-δ - Ce0.9Sm0.1O2-δ

 

 

 

Reaction mechanism involved in the preparation of NiO-Ce0.8Gd0.2O2-δ - Ce0.8Sm0.2O2-δ

 

 

 

Reaction mechanism involved in the preparation of NiO-Ce0.9Gd0.1O2-δ

 

 

Reaction mechanism involved in the preparation of NiO-Ce0.8Gd0.2O2-δ

 

 

Characterizations

         The precursor samples were subjected to thermo gravimetric analysis experiments with Perkin Elmer TGA 7 instrument to know about the temperature of formation of phase pure materials. The heat treated powder was characterized by Shimadzu XRD6000 X-ray diffractometer using CuKα radiation. The lattice parameters were calculated by least square fitting method using DOS computer programming. The theoretical density of the powders was calculated with the obtained XRD data. The crystallite sizes of the powder were calculated by Scherrer’s formula. JASCO FTIR spectrometer was employed to record the FTIR spectra of materials in the range of 4000 – 400 cm-1. The surface morphology of the particles was studied by means of JEOL Model JSM-6360 scanning electron microscope. EDAX analysis was also performed with JEOL Model JSM-6360 to find out the percentage of elements present in the samples. The particle size of the powder was measured using Malvern Particle Size Analyzer using triple distilled water as medium.

            The circular pellets (8 mm diameter x 1.5 mm thickness) were prepared from the materials by applying a pressure of 1.2 ton with a mini hydraulic pelletizer. They were sintered at 700o C for 3 hours and finally subjected to high temperature electrochemical impedance spectroscopy studies. The complex impedance spectroscopy measurements have carried out using a Solatron 1260 frequency response analyzer (FRA) combined with Solatron 1296 electrochemical interface (ECI).

RESULTS AND DISCUSSION

Thermogravimetric analysis

            The dried precursor precipitate samples [(Ni(OH)2 + Ce(OH)4 + Gd(OH)3 + Y(OH)3, Ni(OH)2 + Ce(OH)4 + Gd(OH)3 + Sm(OH)3 and Ni(OH)2 + Ce(OH)4 + Gd(OH)3) with C-TAB,] with an initial mass 6-12 mg was placed in an open platinum crucible. The TGA patterns obtained with the precursor precipitate materials are indicated in Fig. 2 (a, b, c, d, e & f).

            From the figures, it was understood the total weight loss was found to be in the range of various percentage levels from the temperature of 25 to 700 °C. From the curve, it was understood that the weight loss begins to appear from  the initial stage itself. From the curves the thermal decomposition of the molecule can be divided into four separate regions as explained in the literature [30].The thermo gravimetric analysis data obtained on the precursor materials is presented in Table 3.

From the curves, it was found that, the weight loss of about 2% is found at around 100 °C, which may be due to the removal of water molecule from the sample. Then, the total weight loss of 5-6% is found at around 250 °C in all the samples, which may be attributed to the phase formation of NiO in the sample [30]. The further weight loss present in the sample until 700 °C is due to the decomposition of remaining carbon/nitrogen-based compounds from the sample. At around 700 °C, the weight loss is stable, which indicates the formation of phase-pure ceramic oxide materials. The total weight loss of about 11-22% is found in all the samples.

Structural analysis by X-ray Diffraction studies

            The XRD patterns of the nanocrystalline ceramic composite oxide materials prepared by the chemical precipitation method with CTAB as surfactant are shown in           Fig. 3. The XRD patterns of the calcined NiO based nanocrystalline ceramic composite oxide powders reveal the formation of well-crystallined single-phase materials.

The XRD patterns obtained on these materials were compared with the standard data for NiO (JCPDS card No. 75-0197) [31] and ceria (JCPDS card No. 81-0792) [32]. The lattice parameters, the theoretical density (DX) and Crystallite size of the samples were calculated as shown in Table 4.

The x-ray broadening provides information about crystallite size by Debye- Scherrer equation. Crystallite size of the samples was calculated from XRD line broadening method using the following Scherrer relationship (Equation 1).

 

 

            Where ‘Dp’ is the crystallite size in nm, ‘k’ is a numerical constant (~0.9), ‘λ’ is the wavelength of X-rays (for Cu Kα radiation, λ = 1.5418 Å), ‘β’ is the effective broadening taken as a full width at half maximum (FWHM) (in radians), ‘θ’ is the diffraction angle for the peak.        

            The lattice parameters are calculated from 2θ values in the X-ray diffraction patterns by using DOS computer programming.   The theoretical density (DX) for the samples was calculated according to the formula (Equation 2).

 

 

            Where, Z = number of chemical species in the unit cell, M = molecular mass of the sample (g/mol), N = Avogadro’s number (6.022 x 1023) and V is the volume of the crystalline unit cell as determined by x-ray diffraction.

   The XRD patterns of the heat-treated NiO based ceria based ceramic composite oxide powders reveal the formation of well-crystallined materials with cubic fluorite structure geometry [33-35]. No impurity peaks were observed in the XRD patterns of all the nanocrystalline materials. The crystallographic planes observed at (111), (200), (220), (311), (222), (400), (331), (420) and (422) as per JCPDS No: 81-0792 are indicated in CeO2 phase [32].The crystallographic planes observed at (111), (200) and (220) as per JCPDS No: 75-0197 are indicated in NiO phase [31].

FTIR Studies

            Fig. 4 (a, b, c, d, e & f) shows the FTIR spectra obtained on NiO based ceria nanocrystalline ceramic composite oxide materials prepared by the chemical precipitation method. FTIR measurements were done using KBr method at room temperature (RT).

            As seen from the spectra, only a prominent peak ~400 cm-1 is found. This peak represents to Ni-O stretching vibrations reported earlier [36].It is reported that pure CeO2 has shown a broad band at 1383 cm-1 [37].In our study, all the samples have shown the characteristic peak of CeO2 exactly at 1383.6 cm-1. The samples showed peaks at around 2360 cm-1, which may be due to the presence of dissolved or atmospheric CO2 in the sample [37]. The peak appeared at 1600 cm-1 is attributable to H-O-H bending mode and is indicative of the presence of molecular water in the samples [37]. The wide absorption bands that appeared in the spectra nearly at 3400 cm-1 are attributed to the stretching vibration of water H-O bond (moisture) [37]. The absorption of atmospheric CO2 and water molecule is common in CeO2 based materials [38].The bands in the region 1,000 to 650 cm-1 have been assigned to the stretching modes, and the region 650 to 450 cm-1contains bridging stretching modes in the samples [39]. The important peaks observed from FTIR spectra assigned for characteristic peaks have shown in the Table 5.

Particle size measurements

            The particle size distribution curves of NiO-based nanocrystalline ceramic composite oxide materials are shown in Fig. 5. For all the measurements, the sample is sonicated with triple distilled water for about 10 minutes and after that the sample is subjected for particle size analysis. Particle characteristics data (based on volume) obtained with NiO-based nanocrystalline ceramic composite oxide powders prepared by chemical precipitation method is indicated in Table 6.

            From the Figures and the particle characteristics data (Table 6) it is observed that the samples size present between the diameters of 166-230 nm range. And also, it was noticed that both nano and micro particles present in all the samples. The presence of larger particles with micron size may be due to the high temperature treatment. It was reported that the particles may agglomerate when they subject to high temperature calcination process [40].

Morphological studies by Scanning Electron Microscope (SEM)

            The SEM pictures obtained on NiO-based nanocrystalline ceramic composite oxide powders (NiO-Ce0.9Gd0.1O2-δ-Ce0.9Y0.1O2-δ, NiO-Ce0.8Gd0.2O2-δ-Ce0.8Y0.2O2-δ, NiO-Ce0.9Gd0.1O2-δ-Ce0.9Sm0.1O2-δ, NiO-Ce0.8Gd0.2O2-δ-Ce0.8Sm0.2O2-δ,  NiO-Ce0.9Gd0.1O2-δ and  NiO-Ce0.8Gd0.2O2-δ) calcined at 750 °C are shown in Figs. 6-11 (a & b) respectively. For all the nanocomposite ceramic oxide powders, SEM photographs have taken at two resolutions, such as, 10,000 and 30,000.

            The morphology of the NiO-based nanocrystalline ceramic composite oxide powders revealed that the particles exhibit spherical shape and the size of these particles shown around 100 nm. From the pictures it is observed that, there are some micro particles also presented. These particles are formed from nano particles due to agglomeration and look like micro particles [40].The addition of surfactant (CTAB) prevented the possibility of high agglomeration and helped to get fine nanocrystalline powders.

EDAX Analysis

            The EDAX spectra obtained with nanocrystalline ceramic composite oxide powders are reported in Fig. 12.

            The chemical composition data observed for the samples from the EDAX analysis is indicated in Table 7. From the data, it was found that the elements were present as per the necessity.                    

Conductivity studies

            Doped ceria has emerged as an alternate material for SOFC. Ceria exhibits mixed ionic and electronic conductivity in reducing atmospheres. Doping with rare earth metals, like gadolinium, increases its stability and conductivity. The relative high values for ionic conduction were observed with Sm, Gd and Y as acceptor dopants [41]. Considering the abundance of these rare-earth elements, rare earth doped ceria used as a potentially low-cost electrolyte for IT-SOFC. Recently, mixed ionic electronic conductivity (MIEC) anode materials such as NiO-GDC, NiO-SDC, etc. are proposed for SOFC application [40-41]. Therefore our proposed nanocrystalline ceramic composite oxide materials were subjected to electrical conductivity measurements as per the details mentioned below.

Sample preparation

            NiO-based nanocrystalline ceramic composite oxide powders were finely ground with the addition of 2-3 drops of binding agent poly ethylene glycol (PEG). They allowed to dry at 50-100 °C for 10-15 minutes. This mixture was subjected to made the circular compacts (with 2 mm thickness and 10 mm diameter) by applying a pressure of 1.2 ton with hydraulic pressure pelletizer. The pellets were sintered at 750 °C for 3 hours in air. The final pellets were obtained with gray color after sintering as shown in Fig. 13. After sintering, the porosity of the pellets became highly dense. These pellets were subjected to high temperature impedance spectroscopy studies in air.

Electrochemical impedance measurements

            The impedance measurements were carried out different temperatures, such as, room temperature, 300, 400, 500 and 600 °C for all the specimens. The conditions maintained for the analysis of the sample are: 1.3 volts; frequency range 42 Hz to 5 KHz. The impedance curves obtained on the ceria doped nanocrystalline pellets are shown in the Figs. 15-20. Fitting of the measurement data was performed with the software ZSimpwin of version 3.20. The impedance data of the NiO based nanocrystalline sintered pellets was fitted with the equivalent circuit R(C(R(CR)) as shown in Fig. 14. The impedance spectra was fitted to the conventional equivalent electronic circuit containing three Resistance- Constant Phase Element (R-CPE) sub circuits in series, which generates three semicircles on the Nyquist plots. Equivalent circuit modeling has been accepted as the means of interpreting electrochemical impedance results [42], as this offers a convenient way of analyzing and investigating changes in cell behavior. The semicircle corresponding to the bulk conductivity is lost from the spectrum above 350 °C. At higher temperatures, around 500 °C, the grain boundary semicircle is also lost. This is caused by the effect on the spectra of inductances generated within the experimental apparatus.

The electrical conduction of ceria based materials results from impurity and intrinsic factors. At low temperatures, its conduction is dominated by the dissociated electron concentration from the energy gap of the impurity, whose activation energy of electrical conduction is much lower than that of the intrinsic conduction. At high temperatures, the conductivity increase is predominantly due to the intrinsic factor, while electrons from the energy gap of the impurity are all dissociated and activated. The impedance curves obtained on sintered circular compacts of NiO-based nanocrystalline ceramic composite oxide pellets are indicated in Figs. 15-20. The bulk resistance (Rb) can be determined from the intercept of the low-frequency part of the arc with real Z’-axis. As the temperature increases, the Rb value shifts towards a lower impedance value. The bulk conductivity can be obtained from the following equation. (Equation 3).

 

 

Where ‘σ’ is represented as total conductivity, ‘t’ and ‘A’ are the thickness and cross sectional area of the pellet and ‘R’ is a resistance. Conductivity values calculated for sintered nanocrystalline ceramic composite oxide pellets at different temperatures as shown in Table 8.

From the data, it is observed that NiO-Ce0.9Gd0.1O2-δ exhibited the highest value of conductivity at 600 oC (8.0544 × 10-04 Scm_1). Petrovsky et al. prepared a novel Y- doped ZrO2 for application in ITSOFC and its electrical conductivity at 700 °C was 0.1 Scm-1 in air and >100 Scm-1 in reducing atmosphere [43].

The activation energies of all materials have been calculated by using Arrhenius linear fit relationship equation. Activation energies, correspond to the conductivity in high temperature range, were determined from the linear fit of the Arrhenius curves. Activation energy of all samples was calculated by using Arrhenius relationship equation. (Equation 4)

 

 

 ‘σ’ is represented as direct current conductivity

T = temperature

σo = Pre-exponentional factor

Ea = activation energy and

KB= Boltzmann constant

The activation energy values calculated for our samples are shown in Table 9. From the data, it was found that with the increase in the conductivity, the activation energy values were also increased.

            Arrhenius-type plots of total conductivity derived from these impedance spectra are presented in Fig. 20 for all anode nanoceramic composite materials. In general, the conductivity was higher for the material sintered at high temperature. Also, there is a decrease in conductivity in all samples as a function of time at 400 and 500 °C.  The conductivity values obtained in the samples are found to be less than the reported data because the samples were sintered at temperatures less than the reported data.  Hence, it is inferred that by increasing the sintering temperature of the circular anode compacts, the conductivity values may further enhance. The conductivity of anode materials for SOFC can be improved well by sintering them at high temperature for prolonged duration. This will improve the microstructural characteristics of anodes which may result in high electronic conductivity [44].

CONCLUSIONS

In this research work, a set of nanocrystalline anode materials (NiO-Ce0.9Gd0.1O2-δ-Ce0.9Y0.1O2-δ, NiO-Ce0.8Gd0.2O2-δ-Ce0.8Y0.2O2-δ, NiO-Ce0.9Gd0.1O2-δ-Ce0.9Sm0.1O2-δ, NiO-Ce0.8Gd0.2O2-δ-Ce0.8Sm0.2O2-δ,  NiO-Ce0.9Gd0.1O2-δ and  NiO-Ce0.8Gd0.2O2-δ) for SOFC have been successfully developed using a cost-effective facile synthesis, i.e., chemical precipitation method. TGA patterns obtained on the precursor samples revealed the methodology to get phase pure materials. The powder XRD data obtained for all the samples is in agreement with the standard reported JCPDS data. From the FTIR spectra, it was observed that characteristic peak of metal-oxygen bond is present in all the samples. The presence of nano particles was confirmed by the particle size analysis and SEM. Elemental composition of all the samples was studied by EDAX analysis, which is in accordance with the theoretical data. The conductivity measurements obtained on all the samples revealed that the samples proposed in this research work may be considered as SOFC anodes after sintering them at high temperatures for prolonged duration.

ACKNOWLEDGMENTS

ASN would like to thank Central Power Research Institute (Ministry of Power, Govt. of India) (Grant No. CPRI/ R&D/ TC/ GDEC/ 2019, dated 06-02-2019) for financial support. DR and ASN thank Karunya Deemed University for providing necessary facilities in the Department of Chemistry to carry-out this research work.

CONFLICT OF INTEREST

            The authors declare that there are no conflicts of interest regarding the publication of this manuscript.