The Effects of the Acid Treatment of ZrB₂ Particles on Their Purity and Aqueous Dispersibility

Abstract

Oxide impurities such as boria (B₂O₃) and zirconia (ZrO₂) on the surfaces of zirconium diboride (ZrB₂) particles are known to limit their sinterability. Among the impurities, B₂O₃ on the surface of ZrB₂ particles could be easily removed by methanol or hydrofluoric acid. However, the remaining ZrO₂ still gave negative influences on the sinterability. In this study, ZrB₂ particles were treated with various acids to remove oxide impurities on their surfaces. The acid treatments were found to vary in efficacy, according to acid type, and affect the crystallinity and morphology of ZrB₂ particles to varying degrees, in some cases forming additional impurities. In particular, the change in the oxygen content of the ZrB₂ particles induced by acid treatment was found to depend on the type of acid. The results of the acid treatments were compared which revealed that HNO₃ treatment optimizes the purity of ZrB₂ particles. In addition, the effects of acid treatment on the surface properties of ZrB₂ particles were considered. In particular, the correlation between the surface properties of the acid-treated ZrB₂ particles and their dispersibility in aqueous solution was investigated.

1. Introduction

A critical factor affecting the development of aerospace technologies such as hypersonic flight vehicles is the requirement for materials with high-temperature tolerance, excellent mechanical strength, and oxidation resistance [1,2,3]. A specific class of materials developed for use under extreme environmental conditions is ultra-high-temperature ceramics (UHTCs) and research interest in this class of materials is increasing [1,2,3,4].

Among UHTCs, zirconium diboride (ZrB₂) is especially suitable for application in extreme environments owing to properties such as a high melting point (3246 °C); high-temperature stability and strength; high electrical and thermal conductivities; excellent corrosion resistance and hardness; a high elastic modulus; and a chemically stable crystal structure [5,6,7,8,9,10,11]. These physicochemical properties result from its strong covalent bonds and low self-diffusion coefficient [5,6,7,12].

However, these properties of ZrB₂ that inform its classification as a UHTC also contribute to its limited sinterability [13,14,15]. Moreover, an incidental reduction is caused by the unavoidable oxide impurities in non-oxide inorganic materials [14,15,16]. Oxide impurities such as boron trioxide (B₂O₃) and zirconium dioxide (ZrO₂) were formed easily on ZrB₂ particles during milling processes or storage due to moisture in the atmosphere [17,18,19,20]. Amorphous B₂O₃ impurities enable rapid diffusion to accelerate coarsening. During grain coarsening, the surface area of the particles decreases as the driving force for densification is consumed, thus having a negative influence on the sinterability of ZrB₂ [17,18,20,21,22]. A previous study showed that most of the B₂O₃ content of ZrB₂ powder can be removed by washing the powder with methanol [14,20]. In contrast, ZrO₂, another significant oxide impurity, is chemically stable, regardless of its crystalline phase, and is difficult to remove.

This study investigated the efficacy of acid treatment in enhancing the purity of commercial ZrB₂ powder, i.e., removing Zr–O impurities which are one of the main factors limiting the sinterability of ZrB₂. The effects of acid treatment (using various acids) on the surface properties, microstructure, and aqueous dispersibility of ZrB₂ particles were studied.

2. Materials and Methods

2.1. Pretreatment of ZrB₂ Particles with Methanol

Commercial ZrB₂ powder (Japan New Metals Co., Osaka, Japan), with a mean particle size of 2.61 μm, was used as a starting material. To eliminate oxides like B₂O₃, the ZrB₂ powder was subjected to repeated washing processes, and each time it was ultrasonicated during centrifugation in 40 mL methanol (99.9%, Daejung Chemicals and Metals Co., Siheung, Korea) for 10 min [14]. The washed particles were dried in a vacuum drying oven under a pressure of −0.1 MPa at 40 °C for 12 h.

2.2. Treatment of ZrB₂ Particles with Various Acids

The surfaces of methanol-washed, commercial ZrB₂ particles were modified through treatment with various acids, namely hydrochloric acid (HCl, 36.46%, Daejung Chemicals and Metals Co., Siheung, Korea), nitric acid (HNO₃, 61%, Junsei Chemical Co., Tokyo, Japan), sulfuric acid (H₂SO₄, >98%, Daejung Chemicals and Metals Co., Korea), and hydrofluoric acid (HF, 50.5%, Duksan Pure Chemicals, Ansan, Korea). The acquired acids were diluted with deionized (DI) water to a concentration of 1 mol% to enable the comparison of their effects on ZrB₂ particles. Methanol-washed ZrB₂ powder (5 g) and diluted acid (250 mL) were mixed at 300 rpm for 1 h at 80 °C in a 500 mL three-necked round-bottom flask and then cooled to room temperature. Each modified ZrB₂ powder was subjected to repeated methanol washing processes to remove residual acid; each time it was ultrasonicated during centrifugation. The washed particles were dried in a vacuum drying oven under a pressure of −0.1 MPa at 40 °C for 12 h.

2.3. Characterization

The crystallinity of the ZrB₂ particles was analyzed by X-ray diffraction (XRD; Ultima IV, Rigaku, Japan) using CuKα radiation (wavelength (λ) = 1.54178Å). The microstructures of the particles were analyzed by field emission scanning electron microscopy (FE-SEM; JSM-6330F, JEOL, Tokyo, Japan). To determine the oxide impurities, the oxygen content of the particles was investigated using a carrier gas hot extraction method (TC-600, LECO Co., St. Joseph, MI, USA). The surface properties of the particles were determined by Fourier transform infrared spectroscopy (FT-IR; IRAffinity-1 S, Shimadzu, Kyoto, Japan). The zeta potential and the particle size distribution (PSD) was determined using the dynamic light scattering (Zetasizer Nano ZSP, Malvern, PA, USA) method.

3. Results and Discussion

ZrO₂, an oxide impurity that acts as a sintering inhibitor of ZrB₂, exists on the surface rather than within ZrB₂ particles, considering that it is formed mainly as a result of water in the atmosphere [17]. Therefore, it was accepted that the acids in this study react with ZrO₂ on the surface of ZrB₂ particles according to the following chemical reactions [19,23]:

ZrO₂ + 6HCl → [ZrCl₆]²⁻ + 2H₃O⁺

(1)

ZrO₂ + 6HNO₃ → [Zr(NO₃)₆]²⁻ + 2H₃O⁺

(2)

ZrO₂ + 2H₂SO₄ → [Zr(SO₄)₂]²⁻ + 2H₃O⁺

(3)

ZrO₂ + 6HF → [ZrF₆]²⁻ + 2H₃O⁺

(4)

Figure 1 shows the XRD patterns of as-received ZrB₂ particles and acid-treated ZrB₂ particles. In the XRD patterns of the as-received ZrB₂ particles, diffraction peaks are observed at 25.2°, 32.5°, 41.5°, and 51.5°; these peaks correspond to the (011), (100), (101), and (002) planes, respectively, of hexagonal ZrB₂ (JCPDS no. 89-3930). The XRD patterns of the HCl- and HNO₃-treated ZrB₂ particles do not differ significantly from that of the as-received ZrB₂ particles; suggesting that HCl and HNO₃ treatments do not significantly affect the crystallinity and structure of ZrB₂ particles. On the other hand, the XRD patterns of the H₂SO₄- and HF-treated ZrB₂ particles differ significantly from that of the as-received ZrB₂ particles; the peaks observed in the XRD of the as-received ZrB₂ particles are barely discernable in the XRD of the H₂SO₄- and HF-treated ZrB₂ particles and a number of new low-intensity peaks are observed. The XRD peaks observed at 30.2° and 50.2° correspond to the (101), (110), and (112) planes, respectively, of tetragonal ZrO₂ (JCPDS no. 81-1544), while the peaks at 28.1°, 31.4°, 34.9°, and 50.1° correspond to the (-111), (111), (020), and (220) planes, respectively, of monoclinic ZrO₂ (JCPDS no. 81-1314). The XRD peaks at 22.0°, 23.4°, 34.9°, and 37.7° correspond to the (003), (012), (104), and (021) planes, respectively, of B₄C (JCPDS no. 86-1129). In particular, the XRD of H₂SO₄-treated ZrB₂ particles shows peaks at 30.1°, 35.4°, and 50.4°, corresponding to the (101), (110), and (112) planes, respectively, of sulfated zirconia (SZ) (JCPDS no. 80-2155); as well as peaks at 28.1°, 32.5°, 41.5°, and 50.1°, corresponding to the (100), (101), (102), and (110) planes of ZrS₂ (JCPDS no. 11-0679) [24,25]. Oxidation produced both tetragonal and monoclinic ZrO₂ while H₂SO₄ and HF dissolved the ZrB₂ particles. SZ was formed because SO₄²⁻ from H₂SO₄ attached to the ionized surface of ZrO₂ formed by oxidation. The unexpected peaks associated with B₄C is likely owed to the presence of small quantities of B₄C within the commercial ZrB₂ particles that are released by the acid treatment, i.e., the dissolution of the ZrB₂ particles.

Figure 2 shows the microstructures of as-received ZrB₂ particles, methanol-washed ZrB₂ particles, and HCl-, HNO₃-, H₂SO₄-, and HF-treated ZrB₂ particles. The as-received ZrB₂ particles have roundish shapes, however, the roundish shapes of the HCl- and HNO₃-treated ZrB₂ particles are more distinct. The morphologies of the HCl- and HNO₃-treated ZrB₂ particles are similar, exhibiting fairly smooth surfaces. Unlike the HCl- and HNO₃-treated ZrB₂ particles, the H₂SO₄-treated ZrB₂ particles (which also produced distinct XRD patterns), appear aggregated and merged. In particular, the morphology of HF-treated ZrB₂ particles is unique, as shown in Figure 2f, exhibiting a rhapis shape. During acid treatment, H₂SO₄ and HF dissolve ZrB₂ particles and transform them from ZrB₂ to oxide phases, monoclinic ZrO₂ and tetragonal ZrO₂, which contain significant amounts of oxygen, and SZ in the case of H₂SO₄; as revealed by the XRD results (Figure 1). The various shapes of the acid-treated ZrB₂ particles suggest that their properties are dissimilar. To elucidate the differences in the properties of the various ZrB₂ particles, they were quantitatively analyzed.

Figure 3 shows the oxygen contents of ZrB₂ particles subjected to 0 to 4 washing processes in methanol. The major oxide impurities on the surfaces of ZrB₂ particles are ZrO₂ and B₂O₃ [26]. A previous study revealed that B₂O₃ on the surfaces of ZrB₂ particles can be efficiently dissolved in methanol [6]. Accordingly, commercial ZrB₂ particles were subject to repeated washing processes, each time the particles were ultrasonicated during centrifugation in 40 mL methanol for 10 min. The initial oxygen content of the commercial ZrB₂ particles was 3.52 ± 0.19 wt%. After three and four washing processes in methanol, the oxygen contents of the ZrB₂ particles are 1.28 ± 0.13 wt% and 1.23 ± 0.09 wt%, respectively. It was found that the washing processes in methanol are unable to reduce the oxygen content of the commercial ZrB₂ particles to less than approximately 1.2 wt%. The oxygen contents of ZrB₂ particles subjected to three and four washing processes are not significantly different; however, the standard deviation in the oxygen content of ZrB₂ particles subjected to four washing processes is noticeably lower than the standard deviation in the oxygen content of ZrB₂ particles subject to three washing processes. Despite the efficacy of the washing process, the oxygen content of the ZrB₂ particles still exceeds 1 wt% owing to the presence of ZrO₂.

It was found that treating the methanol-washed ZrB₂ particles with HCl, reduced their oxygen content from 1.23 ± 0.09 wt% to 0.56 ± 0.06 wt% as shown in

Table 1. The oxygen contents of as-prepared of ZrB₂ and acid-treated ZrB₂.

Materials	Oxygen Contents (%)	Standard Deviation (%)
Methanol-treated ZrB2	1.23	0.09
HCl-treated ZrB2	0.56	0.06
HNO3-treated ZrB2	0.54	0.04
H2SO4-treated ZrB2	15.04	1.46
HF-treated ZrB2	12.66	1.80

. Similarly, the oxygen content of HNO₃-treated ZrB₂ particles was relatively low at 0.54 ± 0.04 wt%. The oxygen content and the standard deviation in the oxygen content of HNO₃-treated ZrB₂ particles was lower than those of HCl-treated ZrB₂ particles. The removal of oxide impurities on the surface of the ZrB₂ particles could be deduced based on the reduction in oxygen content. The treatment of methanol-washed ZrB₂ particles with H₂SO₄ and HF increased their oxygen content to 15.04 ± 1.46 wt% and 12.66 ± 1.80 wt%, respectively. The oxygen contents of the HCl- and HNO₃-treated ZrB₂ particles were lower than those of the H₂SO₄- and HF-treated ZrB₂ particles owing to the high reactivity of H₂SO₄ and HF towards ZrB₂. The increase in oxygen contents of H₂SO₄- and HF-treated ZrB₂ particles was consistent with the change in the crystallinity of ZrB₂ particles induced by H₂SO₄ and HF treatments, as shown in the XRD results (Figure 1). These results revealed that H₂SO₄ and HF treatments were ineffective at reducing the oxygen content of ZrB₂ particles. Consequently, only HCl and HNO₃ treatments were investigated further.

Figure 4 shows the zeta potentials of as-received ZrB₂ particles and HCl- and HNO₃-treated ZrB₂ particles. The zeta potential of the as-received ZrB₂ particles is −11.3 ± 0.98 mV. In contrast, the zeta potentials of the HCl- are HNO₃-treated ZrB₂ particles are −33.4 ± 0.47 mV and −40.2 ± 0.46 mV, respectively. The zeta potentials and standard deviation in the zeta potentials of the acid-treated ZrB₂ particles are lower than those of the as-received ZrB₂ particles; moreover, the zeta potential and standard deviation in the zeta potential of the HNO₃-treated ZrB₂ particles are lower than those of the HCl-treated ZrB₂ particles. Evidently, HCl and HNO₃ treatments increase the absolute zeta potential of ZrB₂ particles.

Figure 5 shows the FT-IR spectra of as-received ZrB₂ particles and HCl- and HNO₃-treated ZrB₂ particles. The spectrum of the as-received ZrB₂ particles shows peaks at 3221 cm⁻¹ and 1477 cm⁻¹ corresponding to O–H stretching vibrations and C–H bending vibrations, respectively. The spectra of the HCl- and HNO₃-treated ZrB₂ particles are similar to that of the as-received ZrB₂ particles; however, the intensities of the O–H peaks in the spectra of the acid-treated ZrB₂ particles are higher than that of the O–H peak in the spectrum of the as-received ZrB₂ particles. Additionally, in the spectrum of the HNO₃-treated ZrB₂ particles, a low-intensity peak is observed at 1525 cm⁻¹ corresponding to N–O stretching vibrations. The hydroxyl groups on the surfaces of the acid-treated ZrB₂ particles enhance the electrostatic repulsion of the particles, accounting for the relatively low zeta potentials of the acid-treated ZrB₂ particles (Figure 4). The observed N–O peak in the spectrum of the HNO₃-treated ZrB₂ particles is likely caused by the reaction of ZrB₂ with the nitrogen in HNO₃.

Figure 6 shows the PSD of as-received ZrB₂ particles and HCl- and HNO₃-treated ZrB₂ particles in DI water. The as-received ZrB₂ particles show a wide and bimodal PSD with peaks at 697 nm and 1977 nm. In contrast, the HCl- and HNO₃-treated ZrB₂ particles show monomodal PSDs with peaks at 691 nm and 583 nm, respectively. The mean size of the HNO₃-treated ZrB₂ particles (631 nm) is less than that of the HCl-treated ZrB₂ particles (778 nm). The differences in the PSD curves are related to the electrostatic repulsion of the particles. Particles with higher absolute zeta potential are more dispersed (Figure 4). The bimodal PSD of the as-received ZrB₂ particles is due to the aggregation of the particles resulting from insufficient electrostatic repulsion. The HCl and HNO₃ treatment of ZrB₂ particles enhance their dispersibility; accounting for the monomodal PSDs of the HCl- and HNO₃-treated ZrB₂ particles. Evidently, acid treatment enhances the dispersion of ZrB₂ particles; however, HNO₃ treatment is more effective than HCl treatment in enhancing the dispersion of ZrB₂ particles as it induces a higher absolute zeta potential in ZrB₂ particles than HCl treatment.

4. Conclusions

Oxide impurities are formed on ZrB₂ particles during milling processes and storage because of moisture in the atmosphere; this phenomenon is exacerbated by the large surface area and high surface energy of ZrB₂ particles. Oxide impurities, such as ZrO₂ and B₂O₃, limit the sinterability of ZrB₂ particles. In this study, washing processes with methanol and acid treatments were performed to remove the oxygen impurities on commercial ZrB₂ particles. The oxygen content of ZrB₂ particles is reduced by washing the particles in methanol, owing to the removal of B₂O₃; however, oxygen impurities such as ZrO₂ remain. To remove ZrO₂ impurities, the methanol-washed ZrB₂ particles were treated with various acids. HCl and HNO₃ treatments were found to preserve the crystallinity of ZrB₂ particles while noticeably reducing their oxygen content. In contrast, H₂SO₄ and HF treatments were found to significantly change the crystallinity of ZrB₂ particles, forming impurities such as ZrO₂, which contains significant amounts of oxygen, as well as SZ. Furthermore, HF-treated ZrB₂ particles assume a rhapis shape. These results clearly demonstrate that H₂SO₄ and HF treatments do not enhance the purity of ZrB₂ particles.

The dispersibility of ZrB₂ particles were also found to be affected by acid treatment. The absolute zeta potentials of HCl- and HNO₃-treated particles exceed that of as-received commercial ZrB₂ particles. The FT-IR spectra of HCl- and HNO₃-treated particles are similar to those of the as-received ZrB₂ particles; however, their peaks corresponding to the stretching vibrations of O–H groups are more intense. Moreover, the O–H peak in the spectrum of the HCl-treated ZrB₂ particles is less intense than the corresponding peak in the spectrum of the HNO₃-treated ZrB₂ particles. These results suggest that HCl and HNO₃ treatments improve the electrostatic repulsion of ZrB₂ particles by forming hydroxyl groups on their surfaces. The as-received ZrB₂ particles show bimodal PSD while the acid-treated particles show monomodal PSD. This is because the aggregation of particles is prevented by the enhanced electrostatic repulsion between the acid-treated ZrB₂ particles. This study revealed that the oxygen content and aqueous dispersibility of ZrB₂ particles treated with HNO₃ are lower and higher, respectively, than those of ZrB₂ particles treated with HCl, H₂SO₄, and HF.