2023 Volume 64 Issue 7 Pages 1585-1590
Zinc oxide (ZnO) with a rocksalt crystal structure is attractive because of the bandgap which lies in the range of visible light absorption (1.2–2.6 eV). However, the rocksalt structure is not stable at ambient pressure and temperature according to an equilibrium phase diagram. Nevertheless, this study demonstrates, for the first time, that it is possible to realize a 100% fraction of the rocksalt structure at ambient pressure and temperature. ZnO powder is initially processed by severe plastic deformation under high pressure through a technique of high-pressure torsion (HPT). The HPT-processed ZnO is then examined using a high-pressure application system available at BL04B1 of SPring-8 and in situ X-ray diffraction (XRD) analysis is conducted under high pressures at elevated temperatures. It is shown that the initial presence of the rocksalt structure produced by the HPT process is effective to attain a 100% fraction of the rocksalt structure.
Zinc oxide (ZnO) is stable with a wurtzite crystal structure at ambient pressure and temperature according to a pressure-temperature phase diagram.1,2) It exhibits an allotropic transformation to a rocksalt crystal structure as the pressure increases to more than ∼6 GPa. The ZnO with the rocksalt structure is attractive because its band gap is in the range of 1.2–2.6 eV3–5) so that it can absorb visible light. This is in contrast with the wurtzite structure because its band gap energy is as high as 3.1–3.4 eV6,7) so that the absorption is limited only to UV light. Thus, the rocksalt structure is more promising than the wurtzite structure for effective use as a photocatalyst. However, because the rocksalt structure is not stable at ambient pressure and temperature, it is a challenging task to stabilize it even at such atmospheric conditions.
Recently, Razavie-Khosroshahi et al.8) demonstrated that the rocksalt structure can remain present at ambient pressure and temperature when ZnO is processed by high-pressure torsion (HPT) and exhibit a photocatalytic activity under visible light. However, the fraction of the rocksalt phase is not sufficient and thus it is an important subject to increase it for more efficient use of ZnO. Here, HPT is a well-known processing technique to introduce severe plastic strain under high pressure as schematically illustrated in Fig. 1(a).9) As extensively documented in many review papers,10–16) the HPT process can promote functionalities in wide ranges of materials such as metals and alloys,17–24) and ceramics8,25–28) and semiconductors.29–38) In particular, application of the HPT process to ZnO including mixtures with TiO2 and GaN was reported8,39–41) where photocatalytic activity is improved through control of vacancy concentrations as well as of the fraction of high-pressure phases.
(a) Principle of high-pressure torsion (HPT) process and (b) schematic illustration of high-pressure application system for in situ X-ray diffraction analysis at SPring-8.
In this study, high pressure is applied to HPT-processed ZnO and the change in the fraction of the rocksalt structure is examined using in situ synchrotron high-energy X-ray diffraction (XRD) analysis in SPring-8. A high-pressure application system developed recently42) is adopted for this XRD experiment, of which principle was introduced earlier43) and modification was made to the in situ experiment as illustrated in Fig. 1(b).42) This study thus shows that a 100% fraction of the rocksalt structure is successfully attained through the in situ analysis.
This study used ZnO powder with a wurtzite structure with an average particle size of ∼20 nm and a purity of 98.3%. The powder of 0.66 g was first compressed in a 10 mm diameter mold under a pressure of ∼50 MPa at room temperature. The disk was then subjected to HPT processing at room temperature under a pressure of 2 GPa or 6 GPa for a total number of 3 turns with a rotation speed of 1 rpm. The final dimensions of the HPT-processed disk were 0.7 mm in thickness and 10 mm in diameter.
2.2 Analysis using in situ XRDIn situ X-ray diffraction (XRD) analysis was carried out at BL04B1 of SPring-8 in JASRI using the high-pressure processing unit described in detail earlier.42) Monochromatic X-rays with an energy of 60 keV or white X-rays in the energy range of 20–150 keV were used for the analysis.
For the monochromatic X-ray analysis, the HPT-processed disk was mounted on a supporter, and selected positions with areas of 0.2 × 0.2 mm2 were illuminated from the center to the edge along the radius of the disk. X-ray diffraction was recorded for a duration of 1 min. by a CCD (Charge-Coupled Device) camera. The diffraction images were then processed such that the diffracted intensities were integrated around the incident beam direction at an equal distance (i.e. at an equal angle) from the center of the image. The intensity profile was then depicted as a function of the diffraction angle. The camera length was determined by acquiring X-rays using ceria (CeO2) particles.
For the analysis with white X-rays, a strip sample with dimensions of 1 mm width, 7 mm length and 0.5 mm thickness was placed between the anvils and only the X-rays diffracted at the mid-width of the strip sample was collected using a collimator and detected by an energy dispersive Ge-type solid state detector (Ge-SSD). The collimator and the detector were located along the direction at an angle of 6 degrees from the incident beam direction. X-ray acquisition was continued for 3 min. and X-ray signals for those fulfilled the diffraction conditions were recorded as a function of X-ray energy.
Figure 2 shows XRD profiles after HPT processing under (a) 2 GPa and (b) 6 GPa for 3 turns (N = 3) including (c) under 6 GPa without rotation (N = 0). Here, each disk after the processing was placed with the disk plane perpendicular to the beam direction and was illuminated by X-rays at four different distances (r = 1, 1.5, 3 and 4.5 mm) from the disk center. The formation of the rocksalt structure is confirmed after processing under 6 GPa for N = 3, but no rocksalt peak was visible after processing under 2 GPa for N = 3 and under 6 GPa without rotation. This indicates that, for the presence of the rocksalt structure at ambient pressure, the application of the higher pressures as 6 GPa as well as the introduction of strain is important. These results are consistent with those reported earlier by Razavie-Khosroshahi et al.8) and with reports that high-pressure phases are more likely to form with the assistance of straining.43,44)
X-ray profiles after HPT processing under (a) 2 GPa and (b) 6 GPa for 3 turns (N = 3), and (c) under 6 GPa without rotation (N = 0). X-ray acquisition was made at four different distances (r = 1, 1.5, 3 and 4.5 mm) from disk center.
Figure 3 shows X-ray profiles after HPT processing under 6 GPa through N = 3 but now the sample was heated to 180°C. X-ray acquisition was made by every 20°C during elevating the temperature except for the first increase by 35°C from room temperature (25°C) and by holding for 5 min after reaching the designated temperature before X-ray acquisition.
X-ray profiles after HPT processing under 6 GPa for 3 turns. X-ray acquisition was made before heating and during heating to 180°C.
The peaks for the rocksalt structure diminish as the temperature increases and become completely invisible when the temperature reaches 160°C. This is far below the melting temperature of ZnO which is 1975°C and thus equivalent to the homologous temperature of 0.19Tm where Tm is the melting temperature. This in situ heating experiment in Fig. 2 suggests that the reverse transformation from the rocksalt structure to the wurtzite structure occurs as a thermally activated process although the homologous temperature is very low as 0.19Tm. It should be noted that the occurrence of this reverse transformation is consistent with the equilibrium phase diagram where the wurtzite structure is stable at ambient pressure and temperature.
3.3 In situ XRD under high pressures and at elevated temperaturesThe in situ X-ray measurement during elevating temperatures was further conducted, but the measurement was now made under the application of high pressures using the high-pressure application unit. The results are shown in Fig. 4, where the X-ray profiles are displayed first for the states during increasing the load by every 10 kN to a maximum of 60 kN at room temperature (25°C) and then, while this load was kept applied, the temperature was elevated to 180°C by every 20° as for the measurement in Fig. 3. Following the X-ray measurement at 180°C, the sample was cooled by circulating chilled water through the upper and lower anvils. Cooling was made to room temperature within ∼5 minutes and X-ray measurement was performed before and after releasing the applied load. Here, the maximum load of 60 kN corresponds to a pressure of 8.6 GPa by dividing the applied load by the area of the strip sample, which is 7 mm2, the real pressure on the region of the measurement will be evaluated later and described together with the transformation behavior to the rocksalt structure in detail in the Discussion section.
X-ray profiles after HPT processing under 6 GPa for 3 turns. X-ray acquisition was made under loading to 60 kN (nominal pressure of 8.6 GPa), during heating to 180°C and after cooling to room temperature. X-ray profiles before loading and after loading are also included.
Inspection of Fig. 4 reveals that the rocksalt peaks increase relative to the wurtzite peaks and this occurs as the applied load increases at room temperature. Peak shifts also occur due to the application of high pressures. The increase of the rocksalt peaks becomes more prominent as the temperature rises. In particular, this is clearly found from the (200) rocksalt peak which does not overlap with any wurtzite peaks. It should be noted that a peak visible at a slightly higher energy side of the (200) rocksalt peak is due to the anvils which contain diamond particles. It is important to emphasize that the rocksalt peaks tend to dominate the sample even after releasing the pressure. This experiment thus indicates that raising the temperature under the application of pressure is an effective way to promote the formation of the rocksalt structure and results in a considerable increase in the fraction of the rocksalt structure.
3.4 In-situ XRD for powderThe same X-ray measurement was undertaken for ZnO powder which was not processed by HPT. The results are shown in Fig. 5 for X-ray profiles during increasing the applied load at room temperature and during elevating the temperature under a maximum load of 50 kN for this measurement. The temperature was raised to 220°C and the sample was cooled by circulating the chilled water as for Fig. 4. X-ray measurement was then undertaken before and after unloading. No rocksalt peaks are visible but only the wurtzite peaks are detected in the initial powder state and at the states after loading to 50 kN. Furthermore, no rocksalt peaks appear until the temperature reaches 140°C under the loading but become appreciable above 160°C and become prominent as the temperature further increases, as evident particularly from the (200) rocksalt peak. The rocksalt peaks are detected when the temperature decreased to room temperature and this is also the case after releasing the pressure. This indicates that the temperature increase under the application of pressure is important for the formation of the rocksalt structure.
X-ray profiles for powder without HPT processing. X-ray acquisition was made under loading to 50 kN (nominal pressure of 7.1 GPa), during heating to 220°C and after cooling to room temperature. X-ray profiles before loading and after loading are also included.
As recognized from Figs. 4 and 5, peak shifts occurred upon application of loads. The real pressure at the region of measurement may be estimated from the shift of the peak energy through the following Birch-Murnaghan equation.45,46)
| \begin{equation} P_{c}(\rho) = \frac{3}{2}\beta_{0}(\eta^{\frac{7}{3}} - \eta^{\frac{5}{3}}) \left[ 1 + \frac{3}{4} (\eta^{\frac{2}{3}} - 1) (\beta'{}_{0} - 4) \right] \end{equation} | (1) |
The pressures so estimated are plotted in Fig. 6 with respect to the temperatures at the time of X-ray acquisition. The symbols in black and blue are for the XRD profiles obtained from the HPT-processed sample shown in Fig. 4 and from the powder shown in Fig. 5, respectively. The number labeled on each symbol represents the order of the XRD analysis from which the (200) rocksalt peak begins to appear. The symbols are distinguished such that they are filled more as the fraction of the rocksalt structure increases. Therefore, if the symbol is all closed, the rocksalt structure occupies the whole region of the analysis. By contrast, the powders without HPT processing never completed the transformation to the rocksalt structure within the conditions covered in this study. It is obvious that straining by HPT is effective to promote the formation of the rocksalt structure and, as mentioned in association with Fig. 4, to cover entirely the analyzed area around 80°C under a pressure of 9 GPa. It should be noted that this pressure is in good accordance with the applied pressure 8.6 GPa which is derived from the applied load (60 kN) divided by the area of the strip sample (7 mm2). It is considered that the increase in the pressure with increasing the temperature may be attributed to a thermal expansion of metallic jigs which support the ceramic anvils on the strip sample. It is also important to note that almost 100% fraction of the rocksalt structure remains present even after reducing to ambient pressure and temperature which is not in an equilibrium condition according to the phase diagram.1,2)
Plots of pressure against temperature calculated from peak shift. Black and blue symbols are for HPT-processed sample and powder sample, respectively. Numbers labeled on symbols represent order of the XRD analysis from which (200) rocksalt peak begins to appear. Symbols are filled more as fraction of rocksalt structure increases, and fully closed symbols indicate rocksalt structure occupies whole analyzed regions.
Based on such in situ XRD experiments, it is anticipated that it may be possible to produce a higher fraction of the rocksalt structure using the HPT process. Thus, ZnO powder with the wurtzite structure was processed by HPT under 6 GPa at room temperature for 3 turns. Thereafter, the temperature was increased to 180°C and kept for 1 hour before the pressure was released to ambient pressure. Figure 7 shows XRD profiles acquired at 5 different positions from the disk center after removing from the HPT machine. The rocksalt peaks are visible at any position, and the fraction increases as the position is away from the center (i.e., as the imparted strain increases). The results are consistent with the in situ experiment where the strain promotes the formation of the rocksalt structure. However, the fraction of the rocksalt structure is insufficient to reach 100% in the whole area. This may be attributed to the reason that the pressure of 6 GPa adopted for this check is not high enough during the subsequent holding in the HPT machine.
X-ray profiles for sample processed by HPT under 6 GPa at room temperature for 3 turns and subsequently heated to hold at 180°C for 1 hour. X-ray acquisition was made at 5 different positions (r = 2, 3, 4, 4.5 and 5 mm) from disk center.
Figure 8 shows a pressure-temperature phase diagram reproduced from a report by Bayarjargal and Winkler,1) where a second harmonic generation technique was used for the phase identification. Open symbols and closed symbols represent phase regions corresponding to the wurtzite structure and the rocksalt structure, respectively. Figure 8 also includes the results by Decremps et al.,2) where the transformation point from the rocksalt to wurtzite structure is marked by open blue squares and that from the wurtzite to rocksalt structure by closed blue squares. The present results are also plotted as marked by star symbols in red. Here, the results are taken from Fig. 6 only for the cases where the rocksalt structure occupies 100% of the analyzed region.
Temperature-pressure phase diagram of ZnO. Open and closed circles are reproduced from Bayarjargal and Winkler1) for wurtzite and rocksalt structures, respectively. Open and close blue squares are from Decremps et al.,2) representing transformation points from rocksalt to wurtzite structure and from wurtzite to rocksalt structure, respectively. Star symbols are replotted from Fig. 6 only for when rocksalt structure occupies 100%.
Comparison reveals that the presence of the rocksalt structure under high pressures is well consistent with the reports by Bayarjargal and Winkler1) and Decremps et al.2) However, this is not the case for the presence of the rocksalt structure under ambient pressure and temperature. This exceptional result arose most probably because the strain introduced in the sample by the HPT process generates internal pressure due to the mutual interaction of dislocations as argued in earlier papers.42,44)
This study examined if the rocksalt structure of ZnO can increase the fraction to 100%. In situ high-energy XRD experiments were conducted to check the transformation behavior from the wurtzite structure to the rocksalt structure, where the former is stable at ambient pressure and temperature while the latter is formed at high pressures and temperatures. The following conclusions were reached in this study.
This work was supported by a Grant-in-Aid for Scientific Research (A) from the MEXT, Japan (JP19H00830). The synchrotron radiation experiments were performed at the BL04B1 of SPring-8 with the approval of the Japan Synchrotron Radiation Research Institute (JASRI) (Proposal No. 2019B1496 and 2021A1390).
