Unique chemical activity in porous YbB2C2 ceramics with high porosity and high compressive strength

High purity layered YbB2C2 powder is synthesized by a boro/carbothermic reduction method using YbBO3, B4C and graphite powders as raw materials. Its X-ray diffraction data are presented, and the space group P4/mbm (No. 127) is confirmed. The lattice parameters are a = b = 5.3389 Å and c = 3.5683 Å, and the atom positions are Yb (0.0000, 0.0000, 0.0000), B (0.3621, 0.8621, 0.5000), and C (0.1606, 0.6606, 0.5000). Porous YbB2C2 ceramics have a high porosity in the range of 69.89–58.11% and a high compressive strength in the range of 19.49–63.44 MPa. Furthermore, the as-produced porous YbB2C2 ceramics show unique chemical activity. Porous YbB2C2 ceramic with a porosity of 69.89% emits so much heat that it can burn a piece of paper when this ceramic is wetted by water. The rate of reaction between the porous YbB2C2 ceramic and water can be simply controlled by adjusting the porosity. The solid reaction products are YbB6, C and an unknown amorphous phase.


Scientific Reports
| (2020) 10:20227 | https://doi.org/10.1038/s41598-020-77267-9 www.nature.com/scientificreports/ Experimental YbB 2 C 2 powder and porous YbB 2 C 2 ceramics were synthesized by the boro/carbothermic reduction method with a mixture of YbBO 3 , B 4 C (99%, 200 mesh, Jingangzhuan, China), and graphite (99%, 200 mesh, Tianyuan, China) powders. The routes for fabricating YbB 2 C 2 powder can be described as follows. First, raw powders with a 4.00:1.28:18.16 molar ratio were milled in a polypropylene jar for 6 h in alcohol to obtain a homogeneous mixture. The powder mixture was then dried in an oven at 60 °C for 24 h. After that, the powder mixture was placed in a graphite crucible heated at a rate of 10 °C/min to 1950 °C and held for 1.0 h under flowing Ar. The route for synthesizing porous YbB 2 C 2 ceramics can be generalized as follows: at first, the homogeneous mixture powders of YbBO 3 , B 4 C, and graphite (as above) were uniaxially pressed into columnar compacts by applying different pre-pressures (10,20,30,50,100,200 and 300 MPa) with a dwell time of 5 min. Then, these green bodies were placed in a graphite crucible and heated at a rate of 8 °C/min to 1980 °C and held for 1.5 h under flowing Ar. The phase composition was identified using an X-ray diffractometer with Cu Kα radiation (Rigaku D/ max-2400, Tokyo, Japan). Rietveld refinement was performed on the XRD results using the GSAS suite with EXPGUI 28,29 . The morphology was investigated by a SUPRA 35 scanning electron microscope (SEM) (LEO, Oberkochen, Germany) equipped with an energy dispersive spectroscopy (EDS) system. Selected area electron diffraction (SAED) was performed using a Tecnai G2 F20 (FEI, Eindhoven, the Netherlands) instrument equipped with a field emission gun. During the process, a perforated carbon/copper net served as a support for the as-prepared powder.
The density of the porous YbB 2 C 2 ceramics was calculated geometrically, by measuring the volume and weight of five identical samples and averaging the data to ensure accuracy. The porosity (η) of porous YbB 2 C 2 ceramics was determined by Eq. (1), where ρ and ρ 0 refer to the sintered density and the theoretical density of YbB 2 C 2 ceramics, respectively. The compressive strength was measured using a microcomputer control electron universal testing machine (CMT4204, Shenzhen SANS testing machine Co., Ltd, Shenzhen, China). The tested samples were rectangular bars with dimensions of 5 mm × 5 mm × 10 mm. The compressive strengths of the as-prepared porous YbB 2 C 2 ceramics were tested in the directions parallel (σ // ) and perpendicular (σ ⊥ ) to the forming direction. For each group, five samples with the same density were used to obtain the average value. The crosshead speed was 0.5 mm/min Results and discussion XRD analysis and Rietveld refinement of YbB 2 C 2 . To date, the crystal structure of REB 2 C 2 is still controversial. Bauer 2,30 and Sakai et al. 15 believed that its space group is P-24c (No. 112), but an increasing number of researchers' studies 4,8,10,31,32 confirmed that its space group is P4/mbm (No. 127). In this work, the P4/mbm space group is used as a starting model for refining the YbB 2 C 2 sample, and the initial values of the cell parameters and atom coordinates come from the isostructural compound YB 2 C 2 26 . The experimental XRD spectrum (red) and the calculated pattern obtained by Rietveld analysis (black) from 2θ = 20° to 80° are shown in Fig. 1a. Only the YbB 2 C 2 phase is detected, which confirms that high purity YbB 2 C 2 powder has been successfully synthesized. Considering the difficulty of synthesizing the YbB 2 C 2 compound 10,14 , the as-fabricated YbB 2 C 2 powder is remarkable, and the boro/carbothermic reduction method shows more advantages than other methods [8][9][10][11][12][13][14][15][16][17] . It should be noted that the difference between the fabricated molar ratio (4.00:1.28:18.16) and theoretical molar ratio (4.00:1.00:19.00) of the raw powders (YbBO 3 , B 4 C and graphite) may come from their different vapor losses under high temperature.
It is well known that the goodness of fit (GOF) for the peak shape and position, structure and background is measured in terms of profile R (reliability) factors, and that relatively lower values of R wp (the weighted spectrum R factor) and R p (the spectrum R factor) are considered to indicate good profile refinement 33 . R wp and R p can be written as Eqs. (2) and (3), where W i , Y oi and Y ci refer to the weight factor based on statistics, diffraction intensity from observation and diffraction intensity from calculation, respectively 34 . Obviously, the observed and calculated spectra are in good agreement, and low values of R wp (10.37%) and R p (7.52%) are obtained www.nature.com/scientificreports/ The lattice parameters of YbB 2 C 2 calculated from the refinement are a = b = 5.3389 Å and c = 3.5683 Å, respectively, and the atom positions are Yb (0.0000, 0.0000, 0.0000), B (0.3621, 0.8621, 0.5000), and C (0.1606, 0.6606, 0.5000). The crystal structure of YbB 2 C 2 can be seen in Fig. 1b,c. The results are in accordance with the cell parameters and (or) atom positions of the known REB 2 C 2 compound 4,11,26,32 . The detailed XRD results of the structural refinements are grouped in Table 2. In addition, the theoretical density of the YbB 2 C 2 ceramic calculated from the refinement is 7.05 g/cm 3 .
Microstructure of the YbB 2 C 2 powder. The morphology of the YbB 2 C 2 powder observed by SEM is shown in Fig. 2a,b. The YbB 2 C 2 powder consists of typical laminated plates with a size of approximately several microns in thickness and tens of microns in length and width. The laminated plates of YbB 2 C 2 reflect the alternate arrangement of the B-C networks and Yb sheets, as shown in Fig. 1c. To clearly observe the morphology of the YbB 2 C 2 powder, TEM was used as well. Figure 2c shows the bright field image (BFI) of YbB 2 C 2 . The as-prepared YbB 2 C 2 possesses a typical laminar microstructure. Figure 2d,e show the diffraction results along the [001] and [010] zone axes. All maxima can be indexed on the basis of a tetragonal cell, with parameters a = 5.2835 Å and c = 3.4920 Å; both values are in good agreement with those measured results from the XRD spectra. In addition, Fig. 2d,e and Table 2 show that the reciprocal space spots were not extinct as long as they met "0kl: k = 2n, h00: h = 2n or hkl: h + k = 2n". According to the International Tables of Crystallography 35 , the reflection conditions are compatible with the space group P4/mbm (No. 127), which confirms the previous choice for Rietveld refinement.
Porous YbB 2 C 2 ceramics. Advanced porous ceramics (also called ceramic foams) usually have low density, high porosity, large specific surface area, thermal shock resistance, corrosion and wear resistance, and high chemical stability [36][37][38] . SEM images of the as-prepared porous YbB 2 C 2 samples with porosities of 69.89%, 61.58% and 58.11% are shown in Fig. 3. The microstructure can be described as an incompact stacking of uniform lamellae with a size of approximately one or two microns in thickness and tens of microns in length and width. It should be noted that the microstructure of porous YbB 2 C 2 is slightly different from that of porous YB 2 C 2 .
(2) www.nature.com/scientificreports/  www.nature.com/scientificreports/ This difference in microstructure comes from the different kinds of raw materials employed. For porous YB 2 C 2 ceramics, Y 2 O 3 , BN and C were used as raw materials, and small grains and large lamellar grains were obtained due to the in situ reaction process and inheritance from graphite, respectively 27 . In the porous YbB 2 C 2 ceramics, the lamellas are oriented to a certain degree, more or less, in the plane perpendicular to the forming direction; this may be due to the raw material of graphite, which has a typical lamellar structure. The anisotropic microstructure of the porous YbB 2 C 2 becomes more obvious with decreasing porosity. For porous ceramics, the compressive strength is an important parameter. To confirm the anisotropic mechanical properties of porous YbB 2 C 2 samples, a compression test was carried out in two directions: σ // (the loading direction parallel to the forming direction) and σ ⊥ (the loading direction perpendicular to the forming direction), as shown in Fig. 4a and Table 3. The σ // strengths of the as-prepared porous YbB 2 C 2 samples with porosities of 69.89-58.11% are in the range of 19.49-63.44 MPa. It is obvious that the σ // strength decreases with increasing porosity. Different from the trend of the σ // strength, the maximum σ ⊥ strength occurs at a porosity of 61.58% (62.40 MPa). The σ ⊥ strengths are higher than the corresponding σ // strengths for samples with porosities of 69.89-61.58%, while the former are lower than the latter for samples with porosities of 58.98-58.11%. Two factors determine the σ ⊥ strengths, namely, the porosity and anisotropic microstructure. With increasing porosity, the σ ⊥ strength decreases as well, which is similar to the trend of the σ // strength. The anisotropic microstructure, i.e., the YbB 2 C 2 lamellas are oriented in the plane perpendicular to the forming direction has an adverse effect on the σ ⊥ strength. This may be due to the weak adhesion between the YbB 2 C 2 laminates. The influence of the anisotropic microstructure of the porous YbB 2 C 2 samples on their compressive strengths can be understood by a model, as shown in Fig. 4b,c. On one hand, when lamellas of porous YbB 2 C 2 samples are oriented to a small www.nature.com/scientificreports/ degree in the plane perpendicular to the forming direction (Fig. 4b), some lamellas act as diagonal bracings that enhance the σ ⊥ strength (samples with porosity of 69.89-61.58%, σ // < σ ⊥ ). On the other hand, when the lamellas of the porous YbB 2 C 2 samples are oriented ideally in the plane perpendicular to the forming direction (Fig. 4c), the oriented lamellas easily disintegrate because of the weak adhesion between them, which weakens the σ ⊥ strength (samples with porosities of 58.98-58.11%, σ // > σ ⊥ ). For the porous YbB 2 C 2 sample with a porosity of 61.58%, the porosity is not high enough and the lamellas are oriented to an appropriate degree. These two factors work together and ultimately enhance the σ ⊥ strength (a maximum of 62.40 MPa). According to Chen et al. 27    www.nature.com/scientificreports/ is different from that in porous YB 2 C 2 ceramics. For the latter, the mechanism can be described as follows: the failure of σ // is caused by accumulated damage 27,39 , and the failure of σ ⊥ is caused by permanent damage 27,40 .
In addition, the as-prepared porous YbB 2 C 2 ceramics were found to produce a large amount of heat when they were wetted by water. One typical example is the reaction between a porous YbB 2 C 2 sample (Φ10 × 3 mm) with a porosity of 69.89% and water, as shown in Fig. 5a-d. The sample was first immersed in deionized water for 1 min (Fig. 5a), then the soaked sample was wrapped by a piece of A4 paper (Fig. 5b), and the paper was completely burned up after about approximately 3 h (Fig. 5c,d). It is clear that a large amount of heat was emitted during the reaction process. This result implies that porous YbB 2 C 2 ceramics possess individual chemical activity in wet environments, which is different from the chemical inertness of other porous ceramics 37,38 . It was also found that the reaction between porous YbB 2 C 2 ceramic and water became slow and non-obvious with decreasing porosity. For example, when the porous YbB 2 C 2 ceramic's porosity was reduced to 58.11%, the paper could not be burned up by reaction heat. Obviously, the chemical activity of the porous YbB 2 C 2 ceramics can be controlled by adjusting their porosity, which can be understood by a sponge model (Fig. 6). During the reaction process, the porous YbB 2 C 2 ceramic acts as a sponge that can easily soak up water at the beginning. The soaked water can keep the reaction going, while a large amount of heat begins to accumulate. Finally, the wrapped paper was burned up by the accumulated heat. Figure 5e,f show the microstructures of the as-prepared porous YbB 2 C 2 sample and its solid reaction products with water for 3 h. The laminated plates of the porous YbB 2 C 2 sample are smooth (Fig. 5e), while the plates of its products with water are rough and look like floccules (Fig. 5f). The XRD spectrum of the solid reaction products of the as-prepared porous YbB 2 C 2 sample with water for 3 h is shown in Fig. 7a. Obviously, apart from the residual YbB 2 C 2 , the solid products consist of YbB 6 , C and an unknown amorphous phase. The amorphous phase Figure 5. Photographs of the as-prepared porous YbB 2 C 2 sample (η = 69.89%) and its reaction with water (a-d). SEM micrograph of the as-prepared porous YbB 2 C 2 sample (η = 69.89%) (e). SEM micrograph of the solid reaction products of the YbB 2 C 2 sample (η = 69.89%) with water for 3 h (f).