Cerium nitride (CeN), as a rare earth mononitride, is the only semi-metallic conductor material among the rare earth-element-based compounds. CeN has excellent electro-magnetic properties and special electronic valence characteristics: its effective valence state is between +3 and +4[1–3]. CeN can easily react with the oxygen, resulting in difficulties to prepare, store, and transport it, which greatly restricts its investigation and applications[4–5].
Conventionally, nitride powders are prepared by the carbothermal reduction nitridation reactions between solid reactants (metal oxides and solid carbon) and nitrogen source gas. Obviously, the homogeneous mixing of solid reactants is difficult, and the long-term reaction at high temperatures requires excellent heat resistance of the experimental equipment. It is reported that CeN can be obtained by the thin film preparation, which mainly uses radio-frequency ion plating and metal atom sputtering methods[6–9]. In these processes, the target poisoning easily occurs, which hinders or even terminates the sputtering process. Thus, CeN films can hardly form. Additionally, among the CeN powder preparation methods, the carbothermal reduction method takes precedence over nitridation synthesis[10–12] and novel reactive milling synthesis[13] methods. However, the obtained target products are multi-phase mixtures, and the content and purity of CeN should be further improved[14].
The sol-gel method is an effective low-temperature syn-thesis method with wide application in the synthesis of homo-geneous metal oxides[15–18]. Different from the solid phase method, the sol-gel method is a typical wet chemical process, which can be used to prepare the homogeneous precursors with homogeneity at the atomic or molecular level[19–21]. The addition of chelating agent to the metal salt solution can form a stable and homogeneous metal chelate, and the target powder can be obtained after specific heat treatments[22–25].
In this research, the cerium nitrate hexahydrate Ce(NO3)3·6H2O and citric acid (CA) were used as the cerium source and organic carbon sources, respectively. The Ce3+-CA precursors were prepared by sol-gel method, then treated by in-situ carbonization process to obtain CeO2/C powder, and finally combined with carbothermal reduction nitridation reaction to prepare CeN powder.
Ce(NO3)3·6H2O (AR) and CA (AR) were purchased from Sichuan Cologne Chemical Reagent Co., Ltd. Deionized water was used as the solvent in the experiments.
Firstly, Ce(NO3)3·6H2O of fixed molar amount was added into the deionized water to obtain the solution with metal salt concentration of 1 mol/L. The solution was magnetically stirred to obtain a homogenous solution. Afterwards, CA was added to the obtained solution according to the molar ratio of Ce(NO3)3·6H2O to CA as 1:1, 1:2, and 1:3. In this case, the molar ratio of Ce(NO3)3·6H2O to CA is equal to the molar ratio of Ce3+ to CA. The mixed solution was magnetically stirred in water bath at 80 °C for 6 h to obtain uniform and stable Ce3+-CA chelate. The Ce3+-CA chelate was dried in a drying oven at 120 °C for 4 h. Due to the solvent evaporation and the decomposition of organic matter, a brownish webbed precursor material was obtained. The precursor was ground into powder, then placed in a small tube furnace, and treated at 600 °C for 2 h under the flowing high-purity Ar conditions to achieve the in-situ carbonization. Then, the obtained CeO2/C powder was placed in a high-temperature tube furnace and treated at 1500 °C under high-purity N2 atmosphere for 4 h, i.e., the CeO2/C powder was treated through the carbothermal reduction nitridation reaction. Finally, the target product CeN powder was obtained after cooling to room temperature and it was stored in the glove box. The experiment process is shown in Fig.1.
Fig.1 Experiment process of CeN preparation through carbothermal reduction nitridation reaction assisted by sol-gel method
The phase composition of the products was identified by X-ray diffractometer (XRD, DX-2700). All specimens were characterized by XRD with Mylar thin-film as a protective film. The morphology and component distribution of the products were investigated through the scanning electron microscope (SEM, JEOL JSM-7900F) equipped with energy dispersive spectroscope (EDS). The mechanism of organic precursor formation in the aqueous phase of sol-gels was investigated by Fourier transform infrared spectroscope (FTIR, INVENIO R). The in-situ carbonization process was analyzed by thermogravimetry (TG) and differential scanning calorimetry (DSC, PerkinElmer-STA8000). Raman spectra were obtained by laser Raman spectrometer (Raman, LabRAM HR).
2.1 Preparation of precursor powder by sol-gel method
In this research, the carbon source originates from CA, and the molar ratio of carbon greatly influences the phase composition of the product. The effect of molar ratio of Ce3+ to CA on the phase composition of the reaction products was investigated to optimize the experiment design and to improve the purity of CeN powder.
Fig.2 shows XRD patterns of the Ce3+-CA chelate precursors with different molar ratios of Ce3+ to CA. It can be seen that no significant diffraction peaks exist in the Ce3+-CA chelate precursors, indicating that the chelate precursors are mainly amorphous structures. Fig.3 shows SEM morphologies of different precursor powders. When the molar ratio of Ce3+ to CA is 1:1, some spongy porous structures appear. With increasing the CA concentration, the porous structure of the inner linkage gradually disappears and is transformed nto the block structure, as shown in Fig.3b–3c.
Fig.2 XRD patterns of Ce3+-CA chelate precursors with different molar ratios of Ce3+ to CA
Fig.3 SEM morphologies of Ce3+-CA chelate precursors with different molar ratios of Ce3+ to CA: (a) 1:1, (b) 1:2, and (c) 1:3
Fig.4 shows FTIR spectra of CA and Ce3+-CA chelate pre-cursors with different molar ratios of Ce3+ to CA. It can be seen
Fig.4 FTIR spectra of CA and Ce3+-CA chelate precursors with different molar ratios of Ce3+ to CA
from the CA spectrum that the peaks at 3492 and 3283 cm-1 are related to the -OH vibration and the vibration of O-H among water molecules, respectively. The peaks at 1743 and 1425 cm-1 correspond to the antisymmetric and symmetric vibrational peaks of C=O in the carboxyl group of CA, respectively. The peak at 1692 cm-1 corresponds to the stretching vibration peak of -COOH in CA. The characteristic peak of -CH2 group appears in the range of 1430–1350 cm-1. FTIR spectra of Ce3+-CA chelate precursors are significantly different from that of pure CA. The vibrational absorption peak near 1610 cm-1 corresponds to the symmetric stretching vibrational peak of C=O bond in the ionized carboxyl group (-COO). The vibrational absorption peak of NO3- and the symmetric peak of C=O can be observed in the range of 1390–1370 cm-1. The peaks in the range of 1090–1030 cm-1 are related to the vibrational absorption peaks of C-O-C-. The peak intensity of NO3- and C-O-C- is gradually weakened with increasing the CA concentration in the preparation.
As shown in Fig.4, the disappearance of -OH, O-H, and carboxyl C=O stretching vibration peaks of CA, the weakening of the antisymmetric stretching peak of C=O, and the appearance of NO3- and C-O-C- vibration absorption peaks all indicate the occurrence of polyester reaction, i.e., Ce3+ interacts with the hydroxyl and carboxyl groups of CA, forming the Ce3+-CA chelate. The formation of the chelate structure promotes the immobilization of Ce3+ in the three-dimensional network structure of the organic precursor. Thus, the homogeneous mixing at molecular scale is achieved and the required diffusion distance between the atoms of Ce and C sources is reduced. Briefly, the in-situ carbonization process forms CeO2/C powder with increasing the contact area of CeO2 and C, therefore reducing the migration distance between Ce and C atoms in the carbothermal reduction nitridation reaction, which promotes the CeN preparation at 1500 °C during carbothermal reduction nitridation process.
2.2 In-situ carbonization of precursor powder
TG-DSC curves of the representative precursor powder are displayed in Fig.5. With increasing the temperature from room temperature to 600 °C, a series of exothermic peaks appear in DSC curve (curve of heat flow) and significant mass loss can be observed in TG curve (curve of remained mass).
Fig.5 TG-DSC curves of representative precursor powder
As shown in Fig.5, there are four stages of mass loss with increasing the temperature. The first stage is that partial crystallization water molecules leave the Ce(NO3)3·6H2O com-pound at temperatures from room temperature to 70 °C, there-fore forming Ce(NO3)3·4H2O. The second stage occurs at 70–142 °C, where the adsorbed and crystallized water of the mate-rial continues to evaporate, thereby forming Ce(NO3)3·H2O. The third stage is at 142–353 °C, where the last crystallization water is removed, thus forming Ce(NO3)3. Subsequently, the in-situ carbonization gradually occurs to form CeO2. Mean-while, the thermal decomposition of CA and the oxidative decomposition of CA-related composites occur. The fourth stage occurs at 353–634 °C, which mainly consists of the decomposition of nitrate and organic matter as well as the in-situ carbonization of precursors. It is clear that the thermal treatment of sol-gel precursors consists of two main parts: one is the water evaporation, including the adsorbed and crystallized water; the other is the decomposition of organic matter and in-situ carbonization of precursor.
Fig.6 shows XRD patterns of the in-situ carbonization products of precursor with different molar ratios of Ce3+ to CA. A distinctive diffraction peak of CeO2 can be observed in the in-situ carbonization product with molar ratio of Ce3+:CA=1:1. When the molar ratio of Ce3+ to CA is 1:2 and 1:3, some weaker and wider diffraction peaks can be observed, and they may be related to the CeO2 diffraction peaks. This phenomenon indicates the formation of CeO2 with poor crystalline phase when the molar ratio of Ce3+ to CA is 1:2 and 1:3. In other words, CeO2 exists in the in-situ carbonization products. No diffraction peaks of carbon can be observed in Fig.6, inferring that the carbon exists in the amorphous form at this stage. With increasing the molar ratio of Ce3+ to CA, the intensity of the diffraction peaks is gradually reduced. This is probably because with increasing the CA concentration, the thickness of the amorphous carbon layer on the product surface is gradually increased, which leads to the decrease in the intensity of CeO2 diffraction peaks. SEM morphologies of the in-situ carbonization products of precursor with different molar ratios are shown in Fig.7, which basically maintain the characteristics of the precursor powders (Fig.3). The more clear porous structure of the products in Fig.7a may be caused by the continuous release of gas during the in-situ carbonization.
Fig.6 XRD patterns of in-situ carbonization products of precursors with different molar ratios of Ce3+ to CA
Fig.7 SEM morphologies of in-situ carbonization products of precursors with different molar ratios of Ce3+ to CA: (a) 1:1, (b) 1:2, and (c) 1:3
Fig.8 shows the Raman spectra of pure CeO2 powder and in-situ carbonization products with different molar ratios of Ce3+ to CA. No peak of carbon can be observed in the Raman spectrum of pure CeO2. However, two distinctive Raman
Fig.8 Raman spectra of pure CeO2 and in-situ carbonization products with different molar ratios of Ce3+ to CA
peaks of disordered amorphous carbon, D-peak and G-peak, exist in the in-situ carbonization products with different molar ratios of Ce3+ to CA. The D-peak near 1350 cm-1 indicates the boundary vibration mode of the hexagonal Brillouin zone induced by disorder, which is mainly caused by the material defects. The larger the D-peak intensity, the more the defects in the material[26]. It can be seen that the D-peak intensity is gradually decreased with changing the molar ratio of Ce3+ to CA from 1:1 to 1:3. Therefore, it can be inferred that the defects in the in-situ carbonization products are gradually reduced with increasing the CA concentration. The G-peak near 1580 cm-1 suggests the stretching vibration mode of the bonds within the carbon atom (sp2) plane, which is related to the graphitization degree. The presence of D-peak and G-peak indicates the existence of carbon in the in-situ carbonization specimens and the formation of a strong bond between CeO2 and carbon.
To further demonstrate the existence of carbon in the in-situ carbonization products and to investigate the distribution of the product elements, EDS analysis was conducted based on the in-situ carbonization product with molar ratio of Ce3+:CA=1:3 and the results are shown in Fig.9. The in-situ carbonization specimen only contains Ce, O, and C elements. It can be seen that Ce, O, and C elements are uniformly distributed in the specimen. Therefore, it can be inferred that the in-situ carbonization product CeO2 is uniformly distributed in the cleaved C matrix. The distributions of Ce, O, and C elements in the in-situ carbonization products with molar ratio of Ce3+ to CA as 1:1 and 1:2 are similar to those in Fig.9.
Fig.9 SEM morphology (a) and EDS mapping results of Ce (b), O (c), C (d) elements of in-situ carbonization product with molar ratio of Ce3+:CA=1:3
2.3 Carbothermal reduction nitridation reaction
XRD patterns of the products after carbothermal reduction nitridation reaction are shown in Fig.10. When the molar ratio of Ce3+ to CA is 1:1, the carbothermal reduction nitridation reaction product is a mixture of Ce2O3, Ce2ON2, and CeN. When the molar ratio of Ce3+ to CA is 1:2, the diffraction peaks of Ce2O3 and Ce2ON2 phases disappear and only the diffraction peaks of CeN phase exist. The C content further increases when the molar ratio of Ce3+ to CA is 1:3, and the product is still the CeN phase. As shown in Fig.10, with increasing the C content, the CeN peak intensity is decreased, indicating that the crystallinity of CeN decreases. Additionally, the excess increase in initial CA content leads to the excess carbon in the carbothermal reduction nitridation products. Therefore, the molar ratio of Ce3+ to CA of 1:2 is the optimal parameter for the preparation of CeN powder through the carbothermal reduction nitridation reaction assisted by sol-gel method.
Fig.10 XRD patterns of carbothermal reduction nitridation reaction products with different molar ratios of Ce3+ to CA
SEM morphologies and EDS element mappings of the carbothermal reduction nitridation reaction products with molar ratio of Ce3+ to CA as 1:2 are shown in Fig.11. As shown in Fig.11a–11b, the micron-sized powder with irregular and massive morphology can be observed. In addition, the agglomerative sintering between the powder particles occurs. Mainly Ce and N elements can be detected, as shown in Fig.11c–11d. It is found that both Ce and N elements are homogeneously distributed.
Fig.11 SEM morphologies of carbothermal reduction nitrida- tion reaction products with molar ratio of Ce3+ to CA as 1:2 (a–b); EDS mapping results of Ce (c) and N (d) elements corresponding to Fig.11b
1) CeN powder can be prepared by the carbothermal reduction nitridation reaction assisted by sol-gel method, using Ce(NO3)3·6H2O as the cerium source and citric acid (CA) as the chelating agent. This method uses organic carbon as the carbon source instead of conventional solid carbon.
2) The preparation process can be divided into two main stages. One is the aqueous phase Ce3+-CA chelation stage, which achieves the homogeneous mixing of the Ce source with C source at the molecular level. The other is the heat treatment stage, which includes the in-situ carbonization to obtain CeO2/C powder and carbothermal reduction nitridation reaction to obtain CeN powder.
3) The in-situ carbonization process forms CeO2/C powder through increasing the contact area of CeO2 and C, therefore reducing the migration distance between Ce and C atoms in the carbothermal reduction nitridation reaction, which promotes the CeN preparation at 1500 °C during carbothermal reduction nitridation process.
References
2 Patthey F, Cattarinussi S, Schneider W D et al. Europhysics Letters[J], 1986, 2(11): 883 [Baidu Scholar]
3 Aerts C M, Strange P, Horne M et al. Physical Review B[J], 2004, 69(4): 5115 [Baidu Scholar]
4 Norman M R, Koelling D D, Freeman A J et al. Physical Review Letters[J], 1984, 53(17): 1673 [Baidu Scholar]
5 Norman M R, Wimmer E, Freeman A J. Physical Review B[J], 1985, 32(12): 7830 [Baidu Scholar]
7 Xiao S Q, Takai O. Thin Solid Films[J], 1998, 317(1–2): 137 [Baidu Scholar]
9 Xiao W, Guo Q, Wang E G. Journal of Physical Chemistry B[J], 2005, 109(11): 4953 [Baidu Scholar]
10 Nakagawa T, Matsuoka H, Sawa M et al. Journal of Nuclear Materials[J], 1997, 247: 127 [Baidu Scholar]
11 Nakagawa T, Matsuoka H, Sawa M et al. Journal of Nuclear Materials[J], 1997, 247: 147 [Baidu Scholar]
12 Hirota M, Katsura M, Miyake M. Journal of Alloys and Compounds[J], 1994, 207–208: 409 [Baidu Scholar]
13 Callahan P G, Jaques B J, Marx B M. Journal of Nuclear Materials[J], 2009, 392(1): 121 [Baidu Scholar]
14 Butt D P, Jaques B. Synthesis and Optimization of the Sintering Kinetics of Actinide Nitrides[R]. Boise: Boise State University, 2009 [Baidu Scholar]
15 Fernandes J D G, Melo D M A, Zinner L B et al. Materials Letters[J], 2002, 53(1–2): 122 [Baidu Scholar]
16 Kwon H T, Bukhovko M P, Mahamulkar S et al. AIChE Jour- nal[J], 2018, 64(11): 4019 [Baidu Scholar]
18 Alanazi H M, AlHaddad M, Shawky A et al. Catalysis Communications[J], 2023, 177: 106 646 [Baidu Scholar]
19 Cushing B L, Kolesnichenko V L, O'Connor C J. Chemical Reviews[J], 2004, 104(9): 3893 [Baidu Scholar]
21 Zhao Shijiao, Ma Jingtao, Zhao Xingyu et al. Rare Metal Materials and Engineering[J], 2020, 49(2): 732 (in Chinese) [Baidu Scholar]
23 Guo H X, Wang J R, Bai J et al. Journal of the American Ceramic Society[J], 2018, 101(7): 2786 [Baidu Scholar]
24 Wang Weichao, Liu Gu, Wang Liuying et al. Rare Metal Materials and Engineering[J], 2022, 51(3): 1003 (in Chinese) [Baidu Scholar]
25 Ouyang L F, Yang X J, Li X L et al. Rare Metal Materials and Engineering[J], 2022, 51(6): 2039 [Baidu Scholar]
26 Ferrari A C, Meyer J C, Scardaci V et al. Physical Review Letters[J], 2006, 97(18): 187 401 [Baidu Scholar]