Structural Stability and Compressibility Study for ZnO Nanobelts under

ARTICLE pubs.acs.org/JPCC

Structural Stability and Compressibility Study for ZnO Nanobelts under High Pressure Luhong Wang,† Haozhe Liu,*,† Jiang Qian,‡ Wenge Yang,§ and Yusheng Zhao‡,|| †

Natural Science Research Center, Harbin Institute of Technology, Harbin 150080, China Lujan Neutron Scattering Center, Los Alamos National Laboratory, MS-H805, Los Alamos, New Mexico 87545, United States § HPSynC, Carnegie Institution of Washington, Argonne, Illinois 60439, United States High Pressure Science and Engineering Center, University of Nevada, Las Vegas, Nevada 89154, United States

)

‡

ABSTRACT:

The study of nanoscale materials with well-controlled shape could help us to learn more about the morphology effect on phase stability and elastic properties. In situ high-pressure synchrotron angle-dispersive X-ray diffraction for nanobelt and bulk ZnO was measured side-by-side in the same diamond anvil cell at room temperature up to 29 GPa. The pressure-induced wurtzite-type to rocksalt-type structural transition was observed starting at 9.3 GPa for both of these two types of ZnO samples, and no enhanced structural stability for ZnO nanobelts under pressure was found. The relative bigger bulk moduli of wurtzite- and rocksalt-type ZnO nanobelts implicated the morphology effect on its compressibility.

’ INTRODUCTION Functional oxide studies have been an attracting field for both fundamental research and industrial applications. Zinc oxide is one of the most important and widely used materials, such as in optoelectronic devices.1 Nanometer scale sample with well-controlled shapes have been successfully fabricated, and they are potentially great candidates for nanodevices, like field-effect transistors and gas sensors.2 5 The structural stability of such nanomaterials is one of the key parameters for the potential application. Under ambient conditions, ZnO has hexagonal wurtzite-type (space group P63mc, B4-type) structure. ZnO with beltlike morphology, a welldefined geometry, and perfect crystallinity is a perfect model material for the structural stability study.4 Recent studies on the neighborhood wurtzite-type ZnS nanobelts showed an ultrastable behavior under high pressure, and the morphology-tuned stability model has been proposed based on the well delayed pressureinduced phase transition compared with the bulk ZnS sample.6 This naturally promotes a question: whether the morphology effect on structure phase transition works as general scenario. Therefore, the similar wurtzite-type ZnO nanobelts will become an ideal model sample to check its morphology effect on structural stability upon compression. r 2011 American Chemical Society

The pressure-induced phase transition in ZnO was first reported from wurtzite-to-rocksalt (B4-to-B1) structure around 9 GPa in 1962.7 X-ray diffraction (XRD) experimental studies on this phase transition have been performed by many groups at various synchrotron sources.8 13 The mechanism for this B4-to-B1 phase transition was studied by first-principles calculations on the possible path, which would allow the least-cost crossing of the transition enthalpy barrier,14,15 and by high-pressure XRD studies of pressure-dependent pretransition lattice distortion from Rietveld refinement.16 The possibility of the NaCl-type to CsCl-type (B2-type) structural transition under strong compression was experimentally investigated above 200 GPa recently.17,18 When the mean grain size went down to ∼12 nm, the B4-to-B1 phase transition was found to delay to ∼15.1 GPa.19 Once this nanocrystal sample is treated at ∼15 GPa and 550 K, the high-pressure B1 phase Special Issue: Chemistry and Materials Science at High Pressures Symposium Received: May 31, 2011 Revised: October 16, 2011 Published: November 16, 2011 2074

dx.doi.org/10.1021/jp205092e | J. Phys. Chem. C 2012, 116, 2074–2079

The Journal of Physical Chemistry C could be quenched under ambient conditions.20,21 This pressureinduced phase transition was estimated to postpone to ∼12 GPa for ZnO nanowires based on photoluminescence measurement under pressure.22 However, a controversial result for this B4-toB1 phase transition was reported to speed up at ∼4.8 GPa pressure for ZnO nanocrystals with ∼15 nm average grain size23 and showed no difference from bulk ZnO for ZnO nanocrystals with ∼40 nm average grain size.24 ZnO nanotube samples also underwent almost the same transition pressure as bulk upon compression.25 These previous reports indicated the complexity and diversity of the size and shape effect on the stability of wurtzitetype ZnO. In this report, we will present the in situ synchrotron XRD results for the morphology effect on phase stability and

Figure 1. Picture of the nanobelt, bulk ZnO samples, and pressure calibration ruby balls in a diamond anvil cell high pressure sample chamber under 8.7 GPa.

ARTICLE

compressibility of wurtzite-type ZnO nanobelts and bulk ZnO polycrystalline sample under the same pressure environment.

’ EXPERIMENTAL SECTION The ZnO nanobelt sample was synthesized by thermal evaporation method with preferred surface normal direction {2110}.4 Polycrystalline ZnO with a nominal purity of 99.9995% and average grain size ∼0.2 μm was used as bulk samples for comparison. Both samples were loaded in the same sample chamber side-by-side with ∼50 μm apart, which provided almost the same pressure environment for the comparison on the phase transition and compressibility. One image of these two samples under 8.7 GPa pressure conditions is shown in Figure 1, in which a T301 stainless-steel gasket preindented to 45 μm thick with a hole diameter of 100 μm was used in a symmetric diamond anvil cell (DAC). Silicone oil was used as pressure-transmitting medium. The pressure was determined by the ruby luminescence method.26 In situ high-pressure angle-dispersive XRD experiments were carried out at the 16-ID-B station of HPCAT, Advanced Photon Source, Argonne National Laboratory. With a pair of K B mirrors, a focused monochromatic X-ray beam with ∼12 μm diameter and wavelength of 0.4157 Å was used for this investigation. Taking advantage of this fine focused X-ray beam, we were able to obtain diffraction from individual sample at every pressure point. Diffraction patterns were recorded on a 2D MAR345 image plate with a typical exposure time 30 s. The diffraction images were processed with the FIT2D program to get angle-dispersive patterns.27 ’ RESULTS AND DISCUSSION The typical XRD patterns for these two kinds of ZnO samples under various pressure conditions were plotted in Figure 2. The B4-to-B1 phase transition for nanobelt and bulk ZnO, was surprisingly found at almost the same pressure. The very weak diffraction peak (200) of the high-pressure B1 phase was observed for both nanobelt and bulk samples when pressure was increased to 9.3 GPa. This phase transition was sluggish and

Figure 2. Typical XRD patterns of nanobelt and bulk ZnO under different pressure conditions during compression. The B4-to-B1 phase-transition zones are marked by arrows. 2075

dx.doi.org/10.1021/jp205092e |J. Phys. Chem. C 2012, 116, 2074–2079

The Journal of Physical Chemistry C

ARTICLE

Figure 3. Typical XRD patterns of nanobelt and bulk ZnO under different pressure conditions during decompression. The B1-to-B4 phase transition zones are marked by arrows.

completed at 12.4 and 11.7 GPa for nanobelt and bulk ZnO, respectively. To check the reversibility and back transformation sequence of ZnO upon decompression, which may be governed by kinetics of these two kinds of sample, we show the typical XRD patterns during decompression process in Figure 3. Upon decompression, the wurtzite-type structure became the majority below 1.9 GPa and fully recovered to wurtzite-type structure for both samples when pressure was completely released. These large hysteresis behaviors are consistent with the results previously reported for both bulk and nanocrystalline ZnO upon decompression procedure studied separately.16,19 These results showed no clearly enhanced wurtzite-type structure stability for the nanobelt ZnO compared with bulk ZnO under compression. The ZnO nanobelts have two growing directions along [0001] and [010], whereas the ( {2110} surface facets dominate the surface structure of wurtzite-type ZnO with atomic flatness. This is similar as the morphology of ZnS nanobelts, in which the surface structure also is dominated by the low-energy ( {2110} surface facets.6 The high-pressure phase for ZnS is zincblende-type (B3-type) structure. Only a partial atomic rearrangement is required from B4-to-B3 phase transition, and the atomic coordination number is 4 for both cases. As a result, the pressure-induced B4to-B3 phase transformation is explosive style for the ZnS nanobelts due to thin thickness effect, in contrast with its sluggish style phase transformation for bulk counterparts.6 However, the high-pressure phase of ZnO is B1 structure with the atomic coordination number as 6. The B4-to-B1 phase transition involved in complicated symmetry broken and pretransition internal atomic distortion mechanism14 16 and the surface energy difference in the nanoscale beltlike ZnO did not play a significant role in the phase transition pressure. The pressure-induced 4-to-6 coordinated phase transitions have been actively studied for other nanostructured materials. For example, the CdSe nanocrystals, which are a good model for 3D nanostructure, have been investigated for the active volume change and nucleation mechanism during the phase-transition cycles,28 and CdSe nanorods, which are a 2D nanostructure, have been studied for the critical length of the nanorods fracture at the moment of the structural

transformation.29 The typical thickness of the ZnO nanobelts in this study is in the range of 10 to 30 nm with the ratio of width-to-thickness around 5 to 10,4 which is a quasi-1D nanostructure and has different surface-to-volume ratio compared with nanosphere- or nanorod-type samples. The nanoscale thickness alone makes this nanobelt sample have significant different compressibility property for its B4 and B1 phases. So the results from this range of aspect ratio will enrich our knowledge of the morphology dependence of the generic pressureinduced 4-to-6 coordinated phase transitions in semiconductors. The lattice parameters of B4 and B1 phase were obtained from the entire set of XRD whole pattern refinements. The corresponding unit cell volumes via pressure data of the B4 and B1 phases of ZnO were fitted to the second-order Birch equation of state (EoS),30 and results are shown in Figures 4 and 5. This set of EoS data for the wurtzite-type and rocksalt-type nanobelt ZnO will enrich our knowledge for the mechanical strength of various type of nanostructured ZnO because limited compressibility data are available for nanostructured ZnO so far.25,31,32 The experimentally derived bulk moduli under ambient conditions, K0, of the B4 and B1 phase are listed in Table 1. Although normally relatively poor hydrostatic conditions could be generated in sample chamber when silicone oil was used as pressure medium, which is rather common to result in an overestimated trend on the value of K0 compared with that under hydrostatic or better quasi-hydrostatic conditions,16,18,33 it is still meaningful to compare the relative values of K0 between nanobelt and bulk samples because they underwent almost the same compression conditions when they were loaded in the same high-pressure chamber in this study, whereas the hydrostatic conditions were estimated fair at this relative low pressure range in silicone oil medium based on a previous report.34 It is interesting to note that the K0 values of the ZnO nanobelts are 17.5 and 30.4% higher than those of bulk counterparts for B4 and B1 phase, respectively. This is another morphology effect on enhanced bulk moduli of ZnO nanobelts, besides the morphology tuned phase-transition pressure changing in the previous reports for ZnS nanobelts.4 The “tougher” bulk modulus of B4 phase ZnO nanobelts is contributed from its lattice parameters. As shown in Figure 4a,b, 2076

dx.doi.org/10.1021/jp205092e |J. Phys. Chem. C 2012, 116, 2074–2079

The Journal of Physical Chemistry C

ARTICLE

Figure 4. Pressure dependence of (a) lattice parameter a, (b) lattice parameter c, (c) c/a ratio, and (d) unit cell volume for wurtzite-type structure of nanobelt and bulk ZnO.

Table 1. Equation of State Parameters for Nanobelt and Bulk ZnO at Room Temperature, in Which K0 Is the Isothermal Bulk Modulus at Zero Pressure, and Its Pressure Derivative K0 Was Fixed As 4 for the Second-Order Birch EoS Fittinga

nanobelt ZnO bulk ZnO

structural type

K0 (GPa)

V0 (Å3)

B4

164.1 ( 1.3

47.608 (fixed)

B1

253 ( 10

B4 B1

139.6 ( 4.9 194 ( 11

77.9 ( 0.2 47.635 (fixed) 78.7 ( 0.3

V0 is the unit cell volume under ambient conditions, and it was fixed as measured value under ambient conditions for B4 structure during the EoS fitting.

a

Figure 5. Unit cell volume of rocksalt-type structure for nanobelt and bulk ZnO as a function of pressure.

both lattice parameters a and c of nanobelt sample are relatively harder to compress than those of bulk ZnO. In particular, lattice parameter a looks more different from corresponding values of its bulk counterpart, which results in more negative pressure dependence of its c/a ratio, as shown in Figure 4c. It is noticed that the ideal correlation u(c/a) = (3/8)1/2 between c/a ratio and u (cf. the 2077

dx.doi.org/10.1021/jp205092e |J. Phys. Chem. C 2012, 116, 2074–2079

The Journal of Physical Chemistry C fractional atomic coordinates (1/3, 2/3, u) for oxygen in wurtzitetype structure) was experimentally confirmed by the XRD structural refinements of ZnO and ZnS under ambient conditions.35 Therefore, any change of the c/a ratio would counter the variation of u. This help to estimate that the u values increases with pressure faster in nanobelts than that in bulk ZnO, and thus the phase transition pressure should be different.13 However, the almost exactly same pressure-induced phase transition behavior for nanobelt and bulk ZnO observed in this study indicates that the u values estimated from such an ideal correlation loses its validity under compression because of the large distortion of the wurtzite sublattice at pretransition pressure.16 The morphology effect of the nanobelts mainly comes from its nanoscale quasi-1D structure. Because the ZnO nanobelts used in this study have two directions along [0001] and [0110], in which the ({2110} are the major surface facets. These make both a and c axes have the component on the 1D thin layers, which contributes to the relative “tougher” lattice parameters a and c of B4 phase under compression. The B4-to-B1 transition involved in complicated symmetry broken and is accompanied by a large volume collapse (∼17%). Therefore, the beltlike shape may not be perfectly conserved through the phase transition. A very recent high-pressure study for ZnO nanowire demonstrated the evidence of the pressureinduced morphology modifications before and after the compression.32 The fact of the much higher (30.4%) bulk modulus of the highpressure B1 phase in nanobelt-type starting sample than that of bulk sample indicates that the nanobelt sample may roughly remain the beltlike shape after the B4-to-B1 phase transition, which caused the significant harder to compress behavior compared with the bulk counterpart (as shown in Figure 5). For the B1 phase of ZnO nanobelts, the orientation of crystals on beltlike shape is not clear because it is an unquenchable high-pressure phase, and it is very challenging to investigate samples’ preferred orientations from high-pressure XRD data because both latticeand shape-preferred orientations exist in the DAC under high pressure.

’ CONCLUSIONS The major implication from this investigation is that the morphology-enhanced structural stability under high pressure is not a generic feature for wurtzite-type nanobelt materials. The structural stability of nanobelts is also strongly related to their intrinsic crystallographic character of the potential high-pressure phases. The relative higher bulk moduli of wurtzite- and rocksalt-type ZnO nanobelts implicate another morphology effect on the nanobelts’ mechanical property. These results enrich our knowledge on the diversity of the size and morphology influences on the various size- and shape-controlled nanostructure materials, which not only offer great application potentials but also provide perfect models for understanding the morphology effect on the pressureinduced solid to solid-phase transformations. ’ AUTHOR INFORMATION Corresponding Author

*E-mail: [email protected].

’ ACKNOWLEDGMENT This work was performed at HPCAT (sector 16), Advanced Photon Source (APS), Argonne National Laboratory. HPCAT is supported by CIW, CDAC, UNLV, and LLNL through funding

ARTICLE

from DOE-NNSA, DOE-BES, and NSF. APS is supported by DOEBES, under contract no. DE-AC02-06CH11357. This work was partially supported by Natural Science Foundation of China (no. 10975042), and program for Basic Research Excellent Talents and Oversea Collaborative Base Project in Harbin Institute of Technology. We thank Dr. Zhengwei Pan for helpful comments.

’ REFERENCES (1) Ozgur, U.; Alivov, Ya. I.; Liu, C.; Teke, A.; Reshchikov, M. A.; Dogan, S.; Avrutin, V.; Cho, S. J.; Morkoc, H. J. Appl. Phys. 2005, 98, 041301. (2) Wang, Z. L.; Song, J. Science 2006, 312, 242. (3) Kong, X. Y.; Ding, Y.; Yang, R.; Wang, Z. L. Science 2004, 303, 1348. (4) Pan, Z. W.; Dai, Z. R.; Wang, Z. L. Science 2001, 291, 1947. (5) Ahmad, M.; Zhu, J. J. Mater. Chem. 2011, 21, 599. (6) Wang, Z.; Daemen, L. L.; Zhao, Y.; Zha, C. S.; Downs, R. T.; Wang, X.; Wang, Z. L.; Hemley, R. J. Nat. Mater. 2005, 4, 922. (7) Bates, C. H.; White, W. B.; Roy, R. Science 1962, 137, 993. (8) Gerward, L.; Olsen, J. S. J. Synchrotron Radiat. 1995, 2, 233. (9) Desgreniers, S. Phys. Rev. B 1998, 58, 14102. (10) Recio, J. M.; Blanco, M. A.; Lua
na, V.; Pandey, R.; Gerward, L.; Olsen, J. S. Phys. Rev. B 1998, 58, 8949. (11) Decremps, F.; Zhang, J.; Liebermann, R. C. Europhys. Lett. 2000, 51, 268. (12) Decremps, F.; Datchi, F.; Saitta, A. M.; Polian, A.; Pascarelli, S.; Di Cicco, A.; Itie, J. P.; Baudelet, F. Phys. Rev. B 2003, 68, 104101. (13) Solozhenko, V. L.; Kurakevych, O. O.; Sokolov, P. S.; Baranov, A. N. J. Phys. Chem. A 2011, 115, 4354. (14) Saitta, A. M.; Decremps, F. Phys. Rev. B 2004, 70, 035214. (15) Limpijumnong, S.; Jungthawan, S. Phys. Rev. B 2004, 70, 054104. (16) Liu, H.; Ding, Y.; Somayazulu, M.; Qian, J.; Shu, J.; H€ausermann, D.; Mao, H. K. Phys. Rev. B 2005, 71, 212103. (17) Mori, Y.; Niiya, N.; Ukegawa, K.; Mizuno, T.; Takarabe, K.; Ruoff, A. L. Phys. Status Solidi B 2004, 241, 3198. (18) Liu, H.; Tse, J. S.; Mao, H. K. J. Appl. Phys. 2006, 100, 093509. (19) Jiang, J. Z.; Olsen, J. S.; Gerward, L.; Frost, D.; Rubie, D.; Peyronneau, J. Europhys. Lett. 2000, 50, 48. (20) Decremps, F.; Pellicer-Porres, J.; Datchi, F.; Itie, J. P.; Polian, A.; Baudelet, F.; Jiang, J. Z. Appl. Phys. Lett. 2002, 81, 4820. (21) Chen, S. Y.; Shen, P.; Jiang, J. Z. J. Chem. Phys. 2004, 121, 11309. (22) Shan, W.; Walukiewicz, W.; Ager, J. W., III; Yu, K. M.; Zhang, Y.; Mao, S. S.; Kling, R.; Kirchner, C.; Waag, A. Appl. Phys. Lett. 2005, 86, 153117. (23) Wu, Z. Y.; Bao, Z. X.; Zou, X. P.; Tang, D. S.; Liu, C. X.; Dai, J. H.; Xie, S. S.; Li, Q. S.; Shen, Z. X.; Zou, B. S. Mater. Sci. Technol. 2003, 19, 981. (24) Zhuravlev, K. K.; Hlaing, Oo W. M.; McCluskey, M. D.; Huso, J.; Morrison, J. L.; Bergman, L. J. Appl. Phys. 2009, 106, 013511. (25) Hou, D. B.; Ma, Y. Z.; Gao, C. X.; Chaudhuri, J.; Lee, R. G.; Yang, H. B. J. Appl. Phys. 2009, 105, 104317. (26) Mao, H. K.; Xu, J.; Bell, P. M. J. Geophys. Res. 1986, 91, 4673. (27) Hammersley, A. P.; Svensson, S. O.; Hanfland, M.; Fitch, A. N.; H€ausermann, D. High Pressure Res. 1996, 14, 235. (28) Jacobs, K.; Zaziski, D.; Scher, E. C.; Herhold, A. B.; Alivisatos, A. P. Science 2001, 293, 1803. (29) Zaziski, D.; Prilliman, S.; Scher, E. C.; Casula, M.; Wickham, J.; Clark, S. M.; Alivisatos, A. P. Nano Lett. 2004, 4, 943. (30) Birch, F. J. Geophys. Res. 1978, 83, 1257. (31) Kumar, R. S.; Cornelius, A. L.; Nicol, M. F. Curr. Appl. Phys. 2007, 7, 135. (32) Dong, Z.; Zhuravlev, K. K.; Morin, S. A.; Li, L.; Jin, S.; Song, Y. J. Phys. Chem. C 2011DOI: 10.1021/jp205467h. 2078

dx.doi.org/10.1021/jp205092e |J. Phys. Chem. C 2012, 116, 2074–2079

The Journal of Physical Chemistry C

ARTICLE

(33) Liu, H.; Hu, J.; Shu, J.; H€ausermann, D.; Mao, H. K. Appl. Phys. Lett. 2004, 85, 1973. (34) Shen, Y. R.; Kumar, R. S.; Pravica, M. G.; Nicol, M. F. Rev. Sci. Instrum. 2004, 75, 4450. (35) Kisi, E. H.; Elcombe, M. M. Acta Crystallogr. 1989, C45, 1867.

2079

dx.doi.org/10.1021/jp205092e |J. Phys. Chem. C 2012, 116, 2074–2079