Phase and structure development of spontaneously ambient-grown ZnO·xH₂O and TiO₂·xH₂O nanostructures towards oxide single crystals

Abstract

Stress-induced ZnO·xH₂O and TiO₂·xH₂O nanocrystals were spontaneously grown on ZnO and TiO₂ films in an ambient atmosphere based on a bond breaking–hydrolysis–reconstruction mechanism, without the use of any other precursors. The development of their phase and structure towards ZnO and TiO₂ was in situ and ex situ studied. The formation of unstable near-amorphous belt-like orthorhombic ZnO·1.5H₂O nanowires and partially crystalline column-like (pyramid top) monoclinic TiO₂·2.5H₂O nanorods was initiated in a high relative humidity of 98%, whereas more stable polycrystalline faceted orthorhombic ZnO·H₂O nanowires and bamboo leaf-shaped monoclinic TiO₂·0.4H₂O nanoflakes were formed in 70% humidity. Upon receiving energy through in situ exposure to an electron beam or ex situ thermal annealing, the ZnO·xH₂O and TiO₂·xH₂O transformed into hexagonal (wurtzite) ZnO and orthorhombic (brookite) TiO₂, respectively. Exposure of non-polar TiO₂ to an electron beam or annealing generated a polycrystalline structure. Exposure of polar ZnO to a low-energy electron beam caused the formation of aligned subgrains, while high-energy annealing yielded a single-crystalline structure, both with a longitudinal [101̄0] orientation, via a self-assembly process that involved nanocrystallite agglomeration, subgrain tilting and boundary elimination.

---

Introduction

Low-dimensional nanostructures of metal oxide semiconductors, such as cheap, non-toxic ZnO and TiO₂ with good piezoelectric and photonic properties as well as remarkable photocatalytic activities, have been considered for use in optoelectronic devices, photocatalysis (pollutant degradations), H₂O splitters and even CO₂ converters, among other devices.^1–11 To synthesize low-dimensional ZnO and TiO₂ nanostructures, expensive, high-temperature vaporous processes^3,12,13 and, in particular, low-cost, low-temperature aqueous (hydrothermal) methods^3,4,14–21 have been developed. During the reaction of the precursors in the aqueous processes,^20–25 ZnO and TiO₂ hydrate templates, i.e. ZnO·xH₂O (typically Zn(OH)₂) and TiO₂·xH₂O (Ti(OH)₄), are initially formed and ultimately transform into ZnO and TiO₂ nanocrystals by a series of hydrolysis–dehydration processes (dissolution–reprecipitation, crystallization and solid-phase transformation).²³ In all aqueous synthetic routes, precursors and catalysts, additives or surfactants are required, and the surface chemistry (H₂O adsorption/hydrolysis) of the reactants^26–28 as well as the structural development and phase transformations of the reaction products^20–25 play important roles in the aqueous growth of oxide nanostructures.

Based on aqueous synthesis and “mechanochemistry” (the breaking of the surface bonds of oxides to form oxy-products),²⁹ we previously developed a bond breaking–hydrolysis–reconstruction (BHR) mechanism³⁰ through which ZnO and TiO₂ nanocrystals spontaneously grew from films in an ambient atmosphere at room temperature (RT) without the need of precursors or catalysts. The covalent oxide nanocrystals were formed in regions of highly concentrated tensile stress around the crest edges of scratch tracks,^30,31 whereas the extrusion-induced growth of metal whiskers (such as Sn and Bi with weak metallic bonds and high mobility) occurs in regions of compressive stress.^32,33 The formation of oxide nanocrystals by simple solid-phase transformations in the BHR process without the use of any precursors is of greater interest than an aqueous synthesis. A series of reactions, including the ambient adsorption of water or hydroxide (OH⁻; hydrolysis) onto stress-generated surface defects (bond breaking), the subsequent hydration of the oxide surface (large-volume hydrolysis) and the final dehydration (reconstruction) may be important in the ambient growth of oxide nanocrystals. To comprehensively elucidate the BHR mechanism and to systematically clarify the evolution of the BHR products into single-crystalline oxide nanostructures for potential applications, in this study, stress-induced ZnO·xH₂O and TiO₂·xH₂O template nanostructures were spontaneously grown in an ambient atmosphere with different humidities. Their structural development and phase transformations to crystalline or even single-crystalline ZnO and TiO₂ are studied using transmission electron microscopy (TEM), both in situ upon exposure to an electron beam and ex situ after thermal annealing.

Experimental section

Thin ZnO and TiO₂ films with a thickness of several hundred nanometers were deposited on Si or glass substrates by a radio-frequency magnetron sputtering using ZnO and TiO₂ targets, respectively. The films were then scratched using a UMIS nanoindenter (Based Model, CSIRO) with a Berkovich diamond tip and a nanoscratch module. The scratch load was ramped up from 0 to 400 mN over a distance of about 1 mm. The scratched samples were then simply stored in a normal ambient atmosphere with a relative humidity (RH) of 40%, 70% or 98% saturated vapor pressure at room temperature for various durations to allow the spontaneous growth of ZnO·xH₂O and TiO₂·xH₂O nanocrystals. To investigate the phase transformations and structural evolution of the grown nanocrystals, in situ exposure to an electron beam (during TEM observations) or ex situ thermal annealing at 300–500 °C for one hour was conducted on the samples that had been stored in the ambient atmosphere. The surface morphologies of the as-grown and the annealed nanocrystals were observed using a scanning electron microscope (SEM, JEOL JSM-6500F), and the microstructures and lattice structures were examined using high-resolution transmission electron microscopes (JEOL JEM-2100F and FEI Tecnai F20 (low accelerate voltage of 120 kV)) that were equipped with an energy dispersive spectrometer (EDS). Thin foil specimens with nanocrystal/film interfaces (around the nanocrystal roots) and short TiO₂·xH₂O nanocrystals (covered by a top Pt protective coating) were prepared using a focus ion beam (FIB, FEI Nova-200) at an ultralow current of 1 nA down to 30 pA to avoid any damage that might be caused by ion bombardment. Specimens of long ZnO·xH₂O nanowires were attached to a Cu grid (with a graphite film) using an optical microscope that was equipped with three-axis micromanipulation probes (Olympus, BX51). These specimens were then observed using TEM.

Results and discussion

Stress-induced spontaneous ambient growth

The SEM images in Fig. 1a clearly show the growth of TiO₂·xH₂O nanocrystals around the scratch track on a TiO₂ film, after the film was scratched and stored in a RH of 98% at RT for 50 days. Along the scratch track, the applied stress was critical to the growth of the nanocrystals by the BHR mechanism:³⁰ (1) at the track front, under a small stress, almost no nanocrystals were formed, while at the end of the track, under a large stress, many nanocrystals appeared; (2) a threshold stress was required to break Ti–O bonds and initiate the growth of nanocrystals; (3) few nanocrystals were formed in the indented region of the scratch track (under a compressive stress) but many were formed at the crest edges (under a tensile stress³¹), unlike for the compressive stress-induced growth of Sn whiskers.³² As shown in Fig. 1b, ZnO·xH₂O nanowires also grew around the side edges of the tracks at the end of the tracks on scratched ZnO films, under the same conditions for a shorter storage period of 20 days. Moreover, longer ZnO·xH₂O nanowires than TiO₂·xH₂O nanocrystals were grown because the Zn–O bonds were weaker than Ti–O and more easily broken for defect initiation and water adsorption. Long, tangled belt-like nanowires (non-faceted) with a diameter of about 200 nm and a length of about 10 μm grew in a RH of 98%, while shorter faceted nanowires (three-fold symmetric) with a diameter of 100–200 nm and a length of several μm were formed in a RH of 70%.

Fig. 1 (a) SEM images of TiO₂·xH₂O nanocrystals around a scratch track on a TiO₂ film, formed in a RH of 98% at RT for 50 days, at the track (a1) front, (a2) center and (a3) end. (b) ZnO·xH₂O nanocrystals around scratch tracks on ZnO films, formed in a RH of (b1) 98% and (b2) 70% at RT for 20 days.

Fig. 2 systematically presents the features, amounts and sizes of the TiO₂·xH₂O nanocrystals that were formed in different humidities, at various temperatures, after various periods of storage. In a RH of 98% at RT, more and larger nanocrystals grew as the period of storage increased: at 10 days, a few bean-like nanodots with a size of less than 100 nm had just nucleated; at 30 days, more column-like nanorods with diameters of 100–200 nm and lengths of about 400 nm appeared; at 50 days, many more larger column-like nanorods had grown, clearly exhibiting a faceted feature and a pyramid-like top, similar to those of typical brookite or rutile TiO₂ nanocrystals.³ In a RH of 70%, differently structured nanocrystals were formed: long, flat, bamboo leaf-like nanoflakes, similar to those of anatase TiO₂, were formed by surface hydration,³⁴ and had a width of about 200 nm and a length of about 1 μm, possibly because the phase/crystal structure differed from that formed in a RH of 98%. In a dry atmosphere, with a RH of 40%, the lack of moisture for the hydration resulted in the suppressed growth of small dots with a size below 100 nm. At the roots of all of the nanocrystals, including ZnO·xH₂O and TiO₂·xH₂O, a film-like hydrate base was observed, clearly revealing the prior adsorption of water and hydration of the oxide film surfaces before the formation of the nanocrystals at local stress-caused defects.³⁰ The sample stored in a RH of 98% at 60 °C, however, did not exhibit the accelerated longitudinal growth of TiO₂·xH₂O nanorods, but large stacked nanosheets were formed. At a higher temperature, and consequently a higher saturated vapor pressure, more moisture and a higher reaction rate induced more and faster hydration, from which the film-like hydrate base on the oxide film surfaces directly and rapidly developed into nanosheets.

Fig. 2 SEM images of TiO₂·xH₂O nanocrystals around scratch tracks on TiO₂ films, formed under different humidities, temperatures and durations (scale bar: 2 μm for low-magnification and 400 nm for high-magnification (arrowed and dash line-framed) images).

Phase and structure initiation in different humidities

Detailed TEM analysis, shown in Fig. 3 and 4, revealed the initiation of different structures and phases of ZnO·xH₂O and TiO₂·xH₂O nanocrystals in different humidities. Notably, during TEM observations, the as-grown hydrate nanocrystals that contained a certain amount of water molecules decomposed (dehydrated) upon exposure to an electron beam, releasing water molecules and causing the gradual formation of bubble-like pores in the nanocrystals or blisters. Therefore, only the structures before decomposition, at a very early stage of observation (less than 10 s), were examined. Fig. 3a–c present the short column-like, faceted TiO₂·xH₂O nanorods (with a pyramid top) that were formed in a high relative humidity of 98%. The high-magnification and lattice images in Fig. 3c and d revealed a partially crystalline structure that comprised a disordered matrix and some short-range ordered clusters with sizes of several nanometers and with interplanar spacings of (803̄) 0.243 nm and (602) 0.231 nm (an angle of 60°), consistent with the structure of monoclinic Ti₄H₂O₉·9H₂O (i.e. TiO₂·2.5H₂O, JCPDS #39-0040). Similarly, as presented in the TEM image and diffused nano-beam SAD pattern in Fig. 3e and f, the ZnO·xH₂O, in the form of long belt-like (non-faceted) nanowires, also had an amorphous structure with a Zn : O ratio of 1 : 2.5 (EDS analysis), suggesting the possible formation of orthorhombic Zn(OH)₂·0.5H₂O (ZnO·1.5H₂O, JCPDS #20-1436). When supplied with abundant moisture for continuous hydration, rather than simultaneous dehydration, the ZnO and TiO₂ film surfaces in which bonds were broken by stress, particularly around some high-energy column boundaries, were steadily hydrolyzed to form a large amount of hydrates, as revealed by the roughened (down-penetrating) crystal/film interfaces and the dissolved column boundaries in Fig. 3b and e. With the hydrates as nuclei and supplements, the protruding ZnO·1.5H₂O and TiO₂·2.5H₂O nanostructures were formed with a high water content and near-amorphous (partially crystalline, with short-range order) structures.

Fig. 3 TEM analysis of TiO₂·xH₂O nanocrystals around a scratch track on a TiO₂ film, formed in a RH of 98% at RT for 50 days: (a) specimen preparation by FIB cutting, (b) low-magnification image of nanocrystal 1 (NC1, in a2), (c) high-magnification image of nanocrystal 2 (NC2, in a2), (d) lattice image of region 1 (R1, in NC2 in c). TEM analysis of ZnO·xH₂O nanocrystals around scratch tracks on ZnO films, formed at RT for 20 days: in a RH of 98%, (e) low-magnification image of nanocrystal 3 (NC3) and (f) high-magnification image of region 2 (R2, in NC3 in e); in a RH of 70%, (g) low-magnification image of nanocrystal 4 (NC4) and (h) high-magnification image of region 3 (R3, in NC3 in g; insets in f and h: nano-beam SAD patterns).

Fig. 4 TEM analysis of TiO₂·xH₂O nanocrystals around a scratch track on a TiO₂ film, formed in a RH of 70% at RT for 50 days: (a) specimen preparation by FIB cutting, (b1) low- and (b2) high-magnification (circle I in b1) images of nanocrystal 1 (NC1, in a2; in an early observation stage), (c) lattice image of region 1 (R1, in circle I in b2); (d) low-magnification image of NC1 (in a2; in a late observation stage), (e) high-magnification image of NC1 (circle II in d; in a late observation stage), (f) lattice image of region 2 (R2, in circle II in e), (g) high-magnification image of NC1 (circle III in d; in a late observation stage), (h) lattice image of region 3 (R3, in circle III in g).

In lower humidity, a RH of 70%, less water was incorporated into the formed ZnO·xH₂O and TiO₂·xH₂O nanocrystals. On the ZnO film surface, as shown in the TEM images in Fig. 3g and h, polycrystalline ZnO·xH₂O nanowires (faceted, three-fold symmetric, as shown in Fig. 1b) grew from a film-like hydrate base, and nanograins with sizes of several nanometers (clusters with sizes of several times the lattice parameter) were dispersed in the nanowires. A Zn : O ratio of about 1 : 2 in some regions and 1 : 1 in others, the lattice structure with an interplanar spacing of 0.28 nm (ZnO (101̄0)), and the nano-beam SAD spots (ZnO (0002) and (101̄0), zone axis [112̄0]^5,34,35) in Fig. 3h all revealed the formation of crystalline orthorhombic (wulfingite) ε-Zn(OH)₂ (ZnO·H₂O, JCPDS #38-0385) and partly hexagonal (wurtzite) ZnO (JCPDS #36-1451) after concurrent hydration/dehydration in an unsaturated moist atmosphere. Additionally, the TEM images in Fig. 4a–c indicated the growth of TiO₂·xH₂O nanocrystals with a lower water content in a RH of 70%: the formed nanoflakes were more stable and so did not decompose until after a long period of exposure to an electron beam (5 min). In the early stage of observation, as presented in Fig. 4c, the clear lattice structure with a (202) interplanar spacing of 0.473 nm suggested the formation of partly dehydrated, polycrystalline monoclinic H₂Ti₅O₁₁·H₂O (TiO₂·0.4H₂O, JCPDS #44-0130) in the lower humidity.

The surface chemistry of the ZnO and TiO₂ films, including the adsorption of water or hydroxide (OH⁻) and dehydration, is believed to dominate the above initiation of different phases of ZnO·xH₂O and TiO₂·xH₂O in different humidities. A range of water adsorption behaviors are proposed based on dynamic density functional calculations and scanning tunneling microscopic analysis of the hydrolysis/dehydration in aqueous solutions.^34–45 In this study, tensile stresses facilitate the breakage of Zn–O and Ti–O bonds to form oxygen vacancies, greatly increasing the surface reactivity (or wettability) for water adsorption.^36,37 The first-layer chemical adsorption of OH⁻ and H⁺ (dissociated from H₂O) onto oxygen vacancies (adsorption energy ∼0.2 eV, which is much smaller than the adsorption energy of H₂O molecules, ∼0.7 eV (ref. 1 and 38–41)) and onto oxygen (to form OH pairs), respectively, forms a monolayer⁴² and facilitates the second-layer physical adsorption of H₂O molecules.^1,38–41 In ZnO, as an example, Zn²⁺ at the surface (similar to Zn²⁺_(aq) in aqueous solutions) spontaneously adsorbs OH⁻, following reactions (1) and (2):⁴³

Zn²⁺_(aq) + OH⁻ → Zn(OH)⁺_(aq) (ΔG = −37.58 kJ mol⁻¹) (1)

Also, the unstable TiO²⁺ on TiO₂ adsorbs water, possibly as described in reaction (3):^44,45

TiO²⁺_(s) + O²⁻ + 2H₂O → Ti(OH)_4(s) (ΔH = −675.54 kJ mol⁻¹) (3)

Similar to the transformation of LiF crystals to LiOH in a humid atmosphere via surface OH⁻ adsorption and penetrating (inward) hydrolysis,⁴⁶ ZnO and TiO₂ film surfaces are expected to hydrolyze, forming a hydrate base and protruding hydrate nanostructures. Consistent with the growth of TiO₂ nanocrystals with different phases and morphologies (such as brookite nanorods and anatase nanoflowers) in aqueous synthetic solutions with different OH⁻ concentrations,⁴⁷ orthorhombic ZnO·1.5H₂O and monoclinic TiO₂·2.5H₂O, containing a large amount of water, grew in a humid atmosphere, whereas orthorhombic ZnO·H₂O or hexagonal (wurtzite) ZnO and monoclinic TiO₂·0.4H₂O, containing less water were formed at lower humidity.

In situ and ex situ evolution of phase and structure

In the late stage of observation after long-term exposure (5 min) to an electron beam, as presented in Fig. 4d, some of the bamboo leaf-like TiO₂·0.4H₂O nanocrystals decomposed, and some pores formed in the nanocrystals and the hydrate base. The lattice images of the non-decomposed regions in Fig. 4e–h present several nanocrystallites with interplanar spacings of (2̄16) 0.274 nm and (022) 0.234 nm (angle of 80.5°), (2̄16) 0.274 nm and (1̄2̄7) 0.240 nm (angle of 55°; not presented herein), and (022) 0.235 nm (angle of 41.9°), all corresponding to triclinic Ti₉O₁₇ (JCPDS #71-0631). Fig. 5a–d reveal that the TiO₂·0.4H₂O nanocrystals were more stable under thermal annealing at 300 °C, but they partially decomposed at 500 °C (which is approximately the decomposition temperature of TiO₂ hydrates⁴⁸). In the TEM images of the TiO₂·0.4H₂O nanocrystals that were decomposed at 500 °C in Fig. 5e–h, most regions were found to have been completely transformed (dehydrated) into polycrystalline TiO₂ with interplanar spacings of (321) 0.190 nm and (202) 0.225 nm (angle of 51°), corresponding to typical orthorhombic (brookite) TiO₂ (JCPDS #76-1934), although very little monoclinic TiO₂·0.4H₂O with interplanar spacings of (006) 0.224 nm and (1̄2̄04) 0.195 nm (angle of 86°) remained in Fig. 5h (JCPDS #44-0130). The above observations resulted in the conclusion that partially crystalline (short-range-ordered) monoclinic TiO₂·2.5H₂O, that had formed in high humidity, would transform into polycrystalline monoclinic TiO₂·0.4H₂O and then further into polycrystalline triclinic Ti₉O₁₇ or orthorhombic (brookite) TiO₂ upon exposure to an electron beam or thermal annealing, similar to reaction (4):^44,45

Ti(OH)_4(s) → TiO_2(s) + 2H₂O (ΔG = −44.9 kJ mol⁻¹) (4)

Fig. 5 Nanoscopic structure analysis of annealed TiO₂·xH₂O nanocrystals (formed in a RH of 70% at RT for 180 days). SEM images: as grown, (a) low- and (b) high-magnification images; annealed at (c) 300 and (d) 500 °C. TEM analysis, 500 °C annealed: (e) specimen preparation by FIB cutting, (f1) low- and (f2) high-magnification (circle I in f1) images of nanocrystal 1 (NC1, in e2; in a late observation stage); lattice images of (g) region 1 (R1) and (h) region 2 (R2, in circle I in f2).

The in situ nanoscopic structural analysis, based on the TEM images in Fig. 6 captured following exposure to an electron beam, and ex situ observations after annealing at 450 °C, shown in Fig. 7, suggested the same transformation sequence occurred from near-amorphous ZnO·1.5H₂O to crystalline or even single-crystalline ZnO nanowires, when the nanowires received energy. Upon exposure to an electron beam for a very short time (less than 10 s), as verified by the high-resolution TEM images, the nano-beam SAD and fast Fourier transform (FFT) patterns in Fig. 6a and d, the ZnO·1.5H₂O nanowire had a disordered structure in most regions, yielding diffused hollow diffraction rings, with a few nanoscale atom clusters (quasi-crystals) with a size of several times the lattice parameter dispersed in the amorphous matrix. As the exposure time was increased (to 1 min), as shown in Fig. 6b and e, the nanowire received more energy and began to dehydrate at its surface where the nanocrystallites nucleated and agglomerated forming a clear lattice structure with a size of about 2–3 nm, possibly following reaction (5),⁴³ in a manner similar to the recovery of LiOH to LiF in a dry atmosphere:⁴⁶

Fig. 6 In situ nanoscopic structure analysis of ZnO·xH₂O nanowires (formed in a RH of 98% at RT for 20 days) under electron beam exposure. High-resolution TEM images: exposed for (a) less than 10 s, (b) 1 min, (c) 5 min (insets: nano-beam SAD patterns); lattice images: exposed for (a) less than 10 s, (b) 1 min, (c) 5 min (ZnO wurtzite structure, SAB: small-angle subgrain boundary; insets: FFT patterns along a [12̄10] axis).

Fig. 7 Nanoscopic structure analysis of 450 °C annealed ZnO·xH₂O nanowires (formed in a RH of 98% at RT for 20 days). TEM analysis of single-crystalline ZnO nanowire: (a1) low- and (a2) high-magnification images (background: graphite film on a Cu grid), (b) high-resolution image (insets: FFT pattern along a [12̄10] axis and lattice image). TEM analysis of a single-crystalline ZnO nanowire with a polycrystalline defect: (c) high-magnification image, (d) high-resolution images and nano-beam SAD patterns of region 1 (R1) and region 2 (R2, in c); lattice images of (e) R1 and (f) R2 (in d; SAB: small-angle subgrain boundary, HAB: high-angle subgrain boundary; dashed lines: subgrain boundaries; solid lines: lattice orientations of small-angle subgrains NG1 and NG2 in e and high-angle subgrains NG3, NG4 and NG5 in f).

As verified by the lattice images and the nano-beam SAD and FFT patterns in Fig. 6c and f, polycrystalline hexagonal (wurtzite) ZnO with a (0001) interplanar spacing of 0.26 nm was ultimately formed, and crystallization and grain growth to 5–10 nm via nanocrystallite agglomeration were observed after long-term exposure (5 min) to an electron beam. The nanocrystallites (or subgrains) tilted with a longitudinal [101̄0] preferred orientation, leaving the {0001} surfaces and the small-angle subgrain boundaries (SABs) among the aligned nanocrystallites. When the exposure time was extended further (10 min), the subgrain structure remained due to the inefficiency of a low-energy electron beam in eliminating SABs.

The high energy provided by thermal annealing activated almost complete dehydration and structural evolution of ZnO·1.5H₂O to ZnO. As observed in Fig. 7a and b, annealing at 450 °C generated a perfect lattice structure with a (0001) interplanar spacing of 0.26 nm and a (101̄0) spacing of 0.28 nm, as well as simple FFT diffraction spots and a stoichiometric Zn : O ratio of 1 : 1, indicating further subgrain tilting and boundary elimination, which yielded the longitudinal [101̄0] orientation and (0001) surface of single-crystalline hexagonal (wurtzite) ZnO nanowires. However, probably due to non-uniform dehydration, a few bump-like or blister defects with a polycrystalline structure were also observed on the surfaces of some single-crystalline nanowires, as presented in Fig. 7c and d. The lattice images that were obtained around a defect, shown in Fig. 7e and f, in which different lattice orientations of several small- and high-angle subgrains are marked, clearly reveal the presence of SABs in the near-single-crystalline region and high-angle boundaries (HABs) in the polycrystalline region. Fig. 8 schematically depicts the structural development of amorphous ZnO·1.5H₂O to single-crystalline ZnO nanowires (without or with a polycrystalline defect). From Fig. 8a, soon after an amorphous ZnO·1.5H₂O nanowire receives energy, dehydration firstly occurs on both surfaces where ZnO clusters begin to precipitate. More nanocrystallites then form and grow preferentially along the [0001] direction (inwards) into columnar subgrains that have SABs and {0001} surfaces. A self-assembly process that includes nanocrystallite agglomeration, subgrain tilting and boundary elimination ultimately yields a single-crystalline ZnO nanowire. Three main growth directions ([0001], [101̄0] and [12̄10]) of a wurtzite ZnO structure, dependent on the experimental parameters and morphology, have been reported.⁴⁹ In this study, the observed longitudinal [101̄0] growth of the nanowires, with {0001} top/bottom surfaces and (12̄10) side surfaces, is attributable to polar surface-induced growth to reduce electrostatic energy, as proposed by Korgel and Kong:^50,51 when ZnO nanocrystallites form on surfaces, the Zn-terminated (0001) surface is positively charged and the O-terminated (0001̄) surface is negatively charged; agglomerated nanocrystallites with surface polar charges are then automatically rearranged in parallel to reduce the total electrostatic energy, causing small-angle alignment (with the (0001)/(0001̄) surfaces perpendicular to the longitudinal [101̄0]). However, as shown in Fig. 8b, non-uniform dehydration, such as that caused by early heating of the ZnO·1.5H₂O nanowire interior, causes the aggregation and expansion of released water molecules below surface, forming a blister, which interferes with the parallel alignment and packing of the polarized ZnO nanocrystallites. High-angle boundaries therefore remain around the blister, generating polycrystalline defects on the surfaces of some single-crystalline nanowires. In contrast, for non-polar TiO₂ without electrostatic energy, a randomly polycrystalline structure is formed either upon exposure to a low-energy electron beam or by high-energy thermal annealing. Similar to the polycrystalline ZnO structure with blister defects and HABs, agglomerated non-polar TiO₂ nanocrystallites of different orientations, also with HABs, barely tilt for boundary elimination even when they receive high energy over a long period, so a polycrystalline TiO₂ structure remains.

Fig. 8 Schematic illustrations of the structural development of spontaneously ambient-grown amorphous ZnO·xH₂O to crystalline ZnO nanowires: (a) a single-crystalline structure, (b) a single-crystalline structure with a polycrystalline defect.

The fact that the stress-induced spontaneous growth of ZnO·xH₂O and TiO₂·xH₂O nanostructures of a certain length (of the order of micrometers) in an ambient atmosphere takes as long as several to tens of days is unpromising. However, this work promotes the development of a new route that facilitates surface hydrolysis for the subsequent formation of hydrate/oxide nanocrystals. Additionally, the in situ and ex situ investigations herein are believed to contribute towards our understanding of the underlying mechanisms of the phase and structural transformations of hydrates towards oxides, although thermal annealing that accelerates dehydration may change the final phases of the ZnO and TiO₂ from those that are slowly formed at room temperature.

Conclusions

The phase and structural transformations of stress-induced ZnO·xH₂O and TiO₂·xH₂O nanocrystals, that spontaneously grew in ambient atmospheres of different humidities without the use of any precursors, were studied in situ upon exposure to an electron beam and ex situ after thermal annealing to elucidate their development into ZnO and TiO₂. As confirmed by nanoscopic analysis, water adsorption/dehydration-dependent surface chemistry dominated the initiation of ZnO·xH₂O and TiO₂·xH₂O. In a moisture saturated atmosphere, near-amorphous belt-like orthorhombic ZnO·1.5H₂O nanowires and partially crystalline column-like monoclinic TiO₂·2.5H₂O nanorods were formed; in lower humidity, polycrystalline faceted orthorhombic ZnO·H₂O nanowires and bamboo leaf-shaped monoclinic TiO₂·0.4H₂O nanoflakes developed. When the ZnO·xH₂O and TiO₂·xH₂O received thermal energy, hexagonal wurtzite ZnO and orthorhombic brookite TiO₂ nanocrystallites nucleated. Polarity induced the formation of single-crystalline ZnO nanowires with a [101̄0] orientation and {0001} surfaces via self-assembly, which included nanocrystallite agglomeration, subgrain tilting and boundary elimination.

Acknowledgements

The authors gratefully acknowledge the financial support provided for this research by the National Science Council, Taiwan, under Grant no. NSC-102-2221-E-005-026-MY3, and in part by the Ministry of Education, Taiwan, under the ATU plan. Ted Knoy is appreciated for his editorial assistance.

References

1. U. Diebold, Surf. Sci. Rep., 2003, 48, 53–229.
2. O. Carp, C. L. Huisman and A. Reller, Prog. Solid State Chem., 2004, 32, 33–177.
3. X. Chen and S. S. Mao, Chem. Rev., 2007, 107, 2891–2959.
4. Z. L. Wang, Mater. Sci. Eng., R, 2009, 64, 33–71.
5. Z. L. Wang, R. Yang, J. Zhou, Y. Qin, C. Xu, Y. Hu and S. Xu, Mater. Sci. Eng., R, 2010, 70, 320–329.
6. S. Barth, F. H. Ramirez, J. D. Holmes and A. R. Rodriguez, Prog. Mater. Sci., 2010, 55, 563–627.
7. S. Xu, Y. Qin, C. Xu, Y. Wei, R. Yang and Z. L. Wang, Nat. Nanotechnol., 2010, 5, 366–373.
8. K. J. Choi and H. W. Jang, Sensors, 2010, 10, 4083–4099.
9. Y. C. Huang, S. Y. Chang, C. F. Lin and W. J. Tseng, J. Mater. Chem., 2011, 21, 14056–14061.
10. J. Song, Y. Zhang, C. Xu, W. Wu and Z. L. Wang, Nano Lett., 2011, 11, 2829–2834.
11. S. Chu, G. Wang, W. Zhou, Y. Lin, L. Chernyak, J. Zhao, J. Kong, L. Li, J. Ren and J. Liu, Nat. Nanotechnol., 2011, 6, 506–510.
12. J. J. Wu and C. C. Yu, J. Phys. Chem. B, 2004, 108, 3377–3379.
13. C. A. Chen, Y. M. Chen, Y. S. Huang, D. S. Tsai, K. K. Tiong and P. C. Liao, CrystEngComm, 2009, 11, 2313–2318.
14. Q. Zhang and L. Gao, Langmuir, 2003, 19, 967–971.
15. Y. Yu and D. Xu, Appl. Catal., B, 2007, 73, 166–171.
16. X. Y. Kong and Z. L. Wang, Nano Lett., 2003, 3, 1625–1631.
17. J. S. Na, B. Gong, G. Scarel and G. N. Parsons, ACS Nano, 2009, 3, 3191–3199.
18. Y. Sun, L. Wang, X. Yu and K. Chen, CrystEngComm, 2012, 14, 3199–3204.
19. R. Shi, P. Yang, J. Wang, A. Zhang, Y. Zhu, Y. Cao and Q. Ma, CrystEngComm, 2012, 14, 5996–6003.
20. N. T. Khoa, S. W. Kim, D. V. Thuan, D. H. Yoo, E. J. Kim and S. H. Hahn, CrystEngComm, 2014, 16, 1344–1350.
21. N. T. Khoa, S. W. Kim, D. H. Yoo, S. Cho, E. J. Kim and S. H. Hahn, ACS Appl. Mater. Interfaces, 2015, 7, 3524–3531.
22. T. Ichikawa and S. Shiratori, Inorg. Chem., 2011, 50, 999–1004.
23. S. Xu and Z. L. Wang, Nano Res., 2011, 4, 1013–1098.
24. M. Wang, Y. Zhou, Y. Zhang, S. H. Hahn and E. J. Kim, CrystEngComm, 2011, 13, 6024–6026.
25. N. J. Nicholas, G. V. Franks and W. A. Ducker, CrystEngComm, 2012, 14, 1232–1240.
26. Y. Ding and Z. L. Wang, Surf. Sci., 2007, 601, 425–433.
27. A. Önsten, D. Stoltz, P. Palmgren, S. Yu, M. Göthelid and U. O. Karlsson, J. Phys. Chem. C, 2010, 114, 11157–11161.
28. J. T. Newberg, D. E. Starr, S. Yamamoto, S. Kaya, T. Kendelewicz, E. R. Mysak, S. Porsgaard, M. B. Salmeron, G. E. Brown, J. A. Nilsson and H. Bluhm, J. Phys. Chem. C, 2011, 115, 12864–12872.
29. S. L. James, C. J. Adams, C. Bolm, D. Braga, P. Collier, T. Friščić, F. Grepioni, K. D. M. Harris, G. Hyett, W. Jones, A. Krebs, J. Mack, L. Maini, A. G. Orpen, I. P. Parkin, W. C. Shearouse, J. W. Steedk and D. C. Waddelli, Chem. Soc. Rev., 2012, 41, 413–447.
30. S. Y. Chang, N. H. Yang, Y. C. Huang, S. J. Lin, T. Z. Kattamis and C. Y. Liu, J. Mater. Chem., 2011, 21, 4264–4271.
31. T. Shibutani, Q. Yu, T. Yamashita and M. Shiratori, IEEE Trans. Electron. Packag. Manuf., 2006, 29, 259–264.
32. W. Shim, J. Ham, K. Lee, W. Y. Jeung, M. Johnson and W. Lee, Nano Lett., 2009, 9, 18–22.
33. M. D. Vaudin, Y. B. Gerbig, S. J. Stranick and R. F. Cook, Appl. Phys. Lett., 2008, 93, 193116.
34. C. Wöll, Prog. Surf. Sci., 2007, 82, 55–120.
35. D. Raymand, A. C. T. van Duin, W. A. Goddard III, K. Hermansson and D. Spångberg, J. Phys. Chem. C, 2011, 115, 8573–8579.
36. A. Calzolari and A. Catellani, J. Phys. Chem. C, 2009, 113, 2896–2902.
37. H. Hu, H. F. Ji and Y. Sun, Phys. Chem. Chem. Phys., 2013, 15, 16557–16565.
38. H. Zhou, H. Alves, D. M. Hofmann, W. Kriegseis, B. K. Meyer, G. Kaczmarczyk and A. Hoffmann, Appl. Phys. Lett., 2002, 80, 210–212.
39. Y. He, W. K. Li, X. Q. Gong, O. Dulub, A. Selloni and U. Diebold, J. Phys. Chem. C, 2009, 113, 10329–10332.
40. X. Cui, Z. Wang, S. Tan, B. Wang, J. Yang and J. G. Hou, J. Phys. Chem. C, 2009, 113, 13204–13208.
41. C. Sun, L. M. Liu, A. Selloni, G. Q. Lu and S. C. Smith, J. Mater. Chem., 2010, 20, 10319–10334.
42. O. Dulub, B. Meyer and U. Diebold, Phys. Rev. Lett., 2005, 95, 136101.
43. A. Moezzi, M. Cortie and A. McDonagh, Dalton Trans., 2011, 40, 4871–4878.
44. K. G. Knauss, M. J. Dibley, W. L. Bourcier and H. F. Shaw, Appl. Geochem., 2001, 16, 1115–1128.
45. V. N. Misra, S. C. Das and T. Subbaiah, Emerging Trends in Mineral Processing and Extractive Metallurgy, Allied Publishers, Delhi, India, 2005, p. 497.
46. S. Sakthivel, M. V. Shankar, M. Palanichamy, B. Arabindoo, D. W. Bahnemann and V. Murugesan, Water Res., 2004, 38, 3001–3008.
47. Y. H. Jin, S. H. Lee, H. W. Shim, K. H. Ko and D. W. Kim, Electrochim. Acta, 2010, 55, 7315–7321.
48. S. Bela, A. S. W. Wong and G. W. Ho, J. Phys. D: Appl. Phys., 2010, 43, 035401.
49. Z. L. Wang, J. Phys.: Condens. Matter, 2004, 16, R829–R858.
50. B. A. Korgel, Science, 2004, 303, 1308–1309.
51. X. Y. Kong, Y. Ding, R. Yang and Z. L. Wang, Science, 2004, 303, 1348–1351.

---

Footnote

† These authors contributed equally.

---