Pulsed laser assisted growth of aligned nanowires of WO₃: role of interface with substrate

Abstract

We report here a systematic study of the growth of aligned WO₃ (002)-oriented nanowires (NWs) on a (111)-oriented platinised silicon substrate using a pulsed laser deposition (PLD) method. Transmission electron microscopy (TEM) analysis has shown that the wires are single crystalline and grow along the [001]_t or [100]_t directions. X-ray diffraction (XRD) measurements confirm phase and structural analysis. We investigated the effect of ablated particle flux on nanowire growth, in particular, the role of the nucleating centre at the interface as it gets modified by the ablated particle flux. We observe a critical value of the laser energy (that determines a critical flux and energy of ablated moieties) at which a compact nanograin film gets converted to an aligned nanowire film. We attribute the existence of such a threshold to the desorption process from the catalyst droplet. By cross-sectional TEM and compositional mapping accompanied by simulation, we confirm that the interfacial layer between the substrate and NW is modified by the ablated particle flux and energy. Aligned NWs above the threshold energy can be attributed to the formation of favorable nucleation sites for a preferred orientation.

---

1. Introduction

In search of new materials with novel properties, many interesting materials have been synthesized in the past two decades that find application in low-dimensional devices and systems. Tungsten oxide (WO₃) belongs to one such class of materials. It has a wide variety of applications in optoelectronics, microelectronics, selective catalysis, and environmental engineering.^1–11 In nanowire form, WO₃ is even more attractive for novel device concepts and for miniaturization. Hence for past two decades, there had been a surge of interest, which still continues, in the growth of WO₃ nanowires with the desired structure for smart device applications.¹² The physical properties of these materials mainly depend on the crystal structure, chemical composition, surface morphology and phase stability of the resulting structures.¹³ WO₃ films can be grown by various techniques that include Radio Frequency (RF) sputtering,¹⁴ chemical vapor deposition,¹⁵ and sol–gel process.¹⁶ Pulsed laser deposition (PLD) is largely a thin film deposition technique used for the deposition of complex, multi-component materials in thin film form (ceramics, super-conductors, metals, etc.). PLD utilizes the output of a pulsed UV laser focused onto a solid target which rapidly heats, vaporizes and creates nearly atomic (partially ionized) vapor that is deposited upon a heated substrate, which is typically comprised of a ceramic, metal or glass. Pulsed laser deposition (PLD) technique is also a suitable technique used for the preparation of WO₃ film because of its repeatability and controlling ability of the stoichiometry and crystal structure of the film.^17,18

The first report of the WO₃ films prepared using the PLD method was by Rougier et al. in 1999.¹⁹ They investigated the effects of the processing temperature, pressure of atmosphere and post annealing on the structural and optical properties of the WO₃ films. There exists reports which elucidates that in PLD the ablated particle flux plays an important role on the synthesis of well-defined oxides including the crystallinity, the stoichiometry, and so on.^20–22 The vapor–liquid–solid (VLS) or vapor–solid (VS) growth mechanism using these techniques allows to control the aspect ratio and positioning of the synthesized NWs on metal supported substrates.¹⁸ The unidirectional growth rate must be much higher than growth rates of other surfaces and interfaces in cases of NW growth. Wacaser et al. proposed a mechanism which is based on preferential nucleation at the interface between the deposited material and a crystalline solid (substrate).²³ This growth mechanism required modification of substrates with the help of laser fluence or laser parameters. Hence the growth can be termed as pulse laser assisted growth where the laser energy/fluence assist to control the nature of the film/substrate interface. In terms of the growth mechanism the substrate plays an important role in controlling the surface morphology as well as the crystal structure of the material. Several reports are there on the growth of metal supported NWs but Pt(111)/SiO₂/Si supported aligned nanowires have not been reported. As the growth of WO₃ nanowires by PLD is a bottom-up technique, the nature of substrates has a vital role for the determination of nanowire dimension and alignment. In many applications of WO₃ (e.g., photochromic property, photoconductive property etc.) there is need for one electrode that is conducting. By using Pt(111)/SiO₂/Si as the substrate for growth, one of the electrode is made conducting. Pt has the advantage that it does not attach with WO₃ and also has a very little lattice mismatch of ∼3%, which may help in reduction of surface energy of WO₃ along a particular direction.

In WO₃ nanostructures the physical properties can be tailored by accurate control of their morphology and crystallographic ordering, exploiting the variety of phases that depends on temperature.²⁴ In a recent report, the photoresponse of WO₃ nanostructures had been greatly modified by tuning surface morphology.¹⁷ Compared to single nanowires, aligned nanowire arrays have a higher surface to volume ratio and thus can be successfully used to fabricate large area, highly sensitive gas sensors for practical applications.²⁵ However, studies on synthesis of tungsten oxide nanowire arrays are comparatively rare due to the high melting point of tungsten metal and tungsten compounds. Several groups have performed studies on nanowire array growth using different techniques.^26–29 Generally, among all the tungsten oxide nanostructures synthesized for gas sensors, W₁₈O₄₉ are mostly obtained and studied.^30,31 One of the interesting works was demonstrated by Huang et al.²⁹ in which they successfully synthesized W₁₈O₄₉ nanowire arrays directly on ITO (indium tin oxide) coated glass substrate using tungsten trioxide powders by a thermal evaporation method at a relatively low temperature (1100 °C). According to Phillip M. Wu³² the difficulty in understanding the physical properties of WO₃ lies in the diverse structures that WO₃ can form when prepared under different growth conditions. Therefore slight distortions of the ideal cubic perovskite-like structure, results in WO₆ octahedra forming a lattice with either corner-sharing or edge-sharing configurations giving rise to various phases. Hence stabilization of a particular phase of higher symmetry is difficult to obtain in case of fully stoichiometric WO₃ material. Dongyun Ma et al. suggested that these materials if are randomly oriented then the active surface area were not sufficiently utilized, due to the compact stacking which reduces the overall electrochemical performance of the device.³³ Thus according to the Dongyun Ma et al., designing a nanostructure with unique surface morphology along with a proper crystal structure is the key to the success of device performance. Consequently a wide knowledge on the morphological and structural study as well as the interfacial layer that acts as a nucleation site for this oxides are now a days gaining importance.

According to the previous reports detailed studies exist on the growth morphology of WO₃ thin films, nanoparticles, nanosheets and nanowires.³⁴ To the best of the present authors' knowledge, no detail work exists regarding the tuning of the growth morphology of WO₃ nanowires on platinised-silicon substrates by controlling the growth parameters (mainly laser energy) especially in PLD. The physical properties of WO₃ are significantly influenced by the shape, size, and crystallinity³⁵ of WO₃ nanostructures. Therefore, many efforts have been focused on the improvement of synthetic routes of 1D WO₃ nanostructures such as nanowires and exploration of the 1D direction crystal growth mechanism. However, except several report on the synthesis researches, there is still no deep understanding of the growth mechanism of 1D orthorhombic-WO₃ nanostructures that can be widely accepted. According to several reports it has been a challenge to stabilize the higher symmetric phases of WO₃ such as orthorhombic (o), tetragonal (t), and cubic (c) at room temperature, since WO₃ reverts to monoclinic-WO₃, due to a second-order Jahn–Teller distortion, irrespective of thermal treatment³⁶ In this paper it has been shown that by controlling the deposition parameters (that in turn control the nucleation pathways) it is possible to grow and stabilize compact array of nearly highly oriented single crystalline orthorhombic (higher symmetry phase) WO₃ nanowires especially by PLD. We also investigate the cause for this attractive growth morphology and establish the role of the interface at the nucleating site and its modifications by the ablated species that in turn is controlled by the laser fluence. It is well known fact that atomically sharp interface between the substrate and the material is well deserved in terms of controlling the device performance as well as the material growth.³⁷ In that regards our study provides an in-depth knowledge regarding the material substrate interface that ultimately controls the material physical properties. This type of study according to the present authors' knowledge was not reported earlier. A precise electron microscopy analysis were performed which provides knowledge on the growth pathway and structure of aligned WO₃ nanowires. The interface that acts as a favorable nucleation site for the aligned NWs growth have been investigated by preparing cross-sectional TEM (X-TEM) samples with the help of Focussed Ion Beam (FIB) based lithography and a proper explanation have been provided correlating the interface between the substrate/aligned NW with the growth mechanism.

2. Experimental

2.1. Material synthesis

Commercially available WO₃ powder of 99.9% purity from Sigma Aldrich was used to make pressed target for PLD. The pellet of WO₃ was sintered at 1100 °C for 10 h to make it compact and defect free target for PLD. In this work KrF excimer laser with laser wavelength 248 nm has been used. Thin film and nanowires of WO₃ were grown on Pt(111) (60 nm)/SiO₂ (300 nm)/Si substrate by controlling the growth parameter of pulsed laser deposition, mainly the laser fluence. Systematic increase of laser fluence changes film like morphology to vertically aligned nanowires. Processing conditions for WO₃ nanocrystalline film and nanostructures (NWs) are summarized in Table 1.

Table 1 Growth parameters of pulsed laser deposition

PLD parameters	Film	NW	Aligned NW
Laser energy	110 mJ	160 mJ	200 mJ
Laser fluence	2 J cm^−2	3 J cm^−2	4 J cm^−2
Gas used	High purity oxygen (99.7%)	High purity oxygen (99.7%)	High purity oxygen (99.7%)
Oxygen partial pressure	30 pascal	30 pascal	30 pascal
Substrate temperature	600 °C	600 °C	600 °C
Post annealing	600 °C for 30 min	600 °C for 30 min	600 °C for 30 min
Laser shots	20 min	20 min	20 min

---

2.2. Cross-sectional specimen preparation for TEM by using FIB

In order to have a detail study of the effect of interfacial layer on NW growth, investigation of NW substrate interface using cross-sectional transmission electron microscopy is necessary. We have prepared samples for X-TEM using Focused Ion Beam (Helios 600 Dual beam FESEM, FEI Make) and Omniprobe. Initially a protective Pt layer is deposited on the film using Gas Injection System (GIS) to avoid sputtering of the film during sample preparation. Initially a column of about (18 μm × 5 μm × 2 μm) having Pt/WO₃/substrate is prepared by milling the sample at 30 kV and 2.4 nA beam current. Then the column was transferred from the sample to TEM grid using Omniprobe and final thinning of the sample down to <100 nm is done at 16 kV and 0.24 nA beam current.

2.3. Characterization

The information about the phases and purity of samples was obtained by X-ray powder diffraction. X-ray diffraction (XRD) studies were carried out with a X'pert pro diffractometer (PANalytical) using a nickel filtered CuKα₁ radiation operated at 45 kV and 40 mA. For phase identification and texture determination of WO₃ films/wires XRD patterns were collected in the 2θ range of 20–80° in standard Bragg–Brentano geometry. The morphologies of as-prepare samples were analyzed by using scanning electron microscope (SEM, Quanta FEG 250) and transmission electron microscopy 200 kV (Tecnai G², TF-20, FEI Make). TEM was employed to study the structure and phase locally. The structural analysis on the basis of the TEM micrographs on the film/nanowires was done by multifunctional simulation software named Java Electron Microscopy Simulation (JEMS) software.³⁸ JEMS simulates HRTEM images by the multislice and Bloch-wave methods. It can construct crystallographic models of various materials in both direct and reciprocal spaces, stereographic projections, and three-dimensional images. By setting the necessary parameters (microscopic and crystallographic), it is possible to observe the changes in HRTEM images in real time.³⁹ We have used JEMS for simulating the diffraction pattern and obtaining the change in the coordination of the atoms. We have kept the initial defocus value (226 nm) and the microscopic parameters (200 kV, envelope illumination) similar for all the cases.

3. Results and discussion

Nanocrystalline thin film like structure of WO₃ was grown on platinised-silicon [Pt(111) (60 nm)/SiO₂ (300 nm)/Si] substrate by using pulsed laser deposition at laser energy 110 mJ and with the increase in the laser energy, the structure as well as morphology changes. At laser energy of 160 mJ, the film like morphology transformed into rod like structure and finally at 200 mJ, the NWs become vertically aligned with the substrate.

3.1. X-ray analysis

The corresponding XRD pattern of all the three samples is shown in Fig. 1a–c. Phase characterization using XRD indicates that the deposited WO₃ film/NWs are well crystallised. The degree of orientation i.e. f_hkl is calculated by eqn (1):⁴⁰here, I_hkl is the relative intensity of a particular crystallographic plane in WO₃. The film like morphology; basically nanocrystalline film, which is obtained at a lower laser energy (110 mJ) (as shown in SEM image of Fig. 2) is monoclinic^22,41 in structure with unit cell parameters as a = 0.7297 nm, b = 0.7539 nm, c = 0.7688 nm (JCPDS 43-1035) (ICSD-80056; space group P2₁/n) and a degree of orientation of 43% along (002) plane (Fig. 1a). But with the increase in the laser energy (160 mJ) the film morphology changes to NWs (Fig. 2) and the structure becomes orthorhombic⁴¹ with unit cell parameters as a = 0.7361 nm, b = 0.7573 nm, c = 0.7762 nm (ICSD-50730; space group Pbcn) and an increasing degree of orientation of about 64% along (002) (Fig. 1b). The NWs becomes much more aligned as the laser energy increased further to an optimal value of 200 mJ (Fig. 2). The structure of the aligned NWs are orthorhombic (ICSD-50730; space group Pbcn) as similar to the NWs grown at 160 mJ laser energy but degree of orientation along (002) has been increased to 82% (Fig. 1c), the increased intensity of X-ray diffraction pattern clearly indicates that. It is well reported that with the change in the deposition/post-annealing temperature phase change in WO₃ occurs. Several reports are available regarding the change in crystallographic ordering or symmetry of WO₃ depending on the substrate temperature during growth as well as the post-annealing temperature.^22,31 In our case as the temperature during deposition as well as post-annealing remains same, it is clear that the laser energy also has an immense effect on the phase as well as degree of orientation. The monoclinic (Fig. 1d) and orthorhombic (Fig. 1e) phases can be identified by splitting of the XRD-peak positions at around 23° (2θ).⁴¹ Later in the microscopy Section (3.3) the SAED patterns (Fig. 5a–c) shows the change in the structure from nanocrystalline monoclinic phase (film) to orthorhombic phase (nanowires) with the increase in the laser energy.

Fig. 1 (a–c) XRD results showing phase change (monoclinic to orthorhombic) and increased degree of orientation along (002) with the increase in the laser energy/deposition flux (d) enlarged view of (a) in the 2θ range of 22–25° showing splitting of the XRD peak of monoclinic structure at around 23° (2θ) (e) enlarged view of (b) in the 2θ range of 22–25° showing the splitting of the XRD peak of orthorhombic structure at around 23° (2θ).

Fig. 2 FESEM micrograph showing changes in the morphology (nanocrystalline film to NW) of WO₃ with the increase in the laser energy.

3.2. FESEM analysis

As illustrated in Fig. 1a–c the different growth possibilities with the increase in the laser energy is well supported by the FESEM microstructure as well as the FIB cross-sections of the WO₃ film/NWs (Fig. 2 & 3 respectively). From Fig. 2, it is quite evident that laser flux modifies the surface morphology from nanocrystalline film to NWs and finally oriented growth of NWs was possible at highest laser flux used in experiment. Since we don't have a quantification of the ablated material and plume characteristics, we will use the laser fluence as a “marker” for the same. This is a valid marking because increase of the fluence leads to enhancement of the energy as well as the flux of the ablated material.

Fig. 3 Cross-sectional samples prepared by FIB of the WO₃ films and NWs showing evolution in morphology (film to NW) with the increase in the deposition flux/laser energy. FIB prepared nanocrystalline film (110 mJ) or non-aligned NW (160 mJ) shows sharp interface while the aligned NW shows a modified interface of about ∼70 nm.

The cross-sections of the WO₃ samples prepared by FIB also support the pictorial representation of the growth with the change in the laser energy. Fig. 3 shows the cross-sectional view of the samples. It is seen that with the increase in the laser energy there is a change in the thickness of the samples though the other growth parameters are constant. It is seen at 160 mJ laser energy the NWs formed are randomly oriented (non-aligned) which grows as a stacking as well as tripods and tetrapods in different directions (Fig. 2). On the other hand the NWs grown at laser energy of 200 mJ are much more aligned, controlled and compact (Fig. 2). This may be because with the increase of the laser pulse energy, the kinetic energy and density of the laser-ablated plasma increases and the materials are deposited faster, which would lead to a compactness in the overall system. The growth of nanowires will depend on the process of nucleation and in particular on the interface that the NW forms with the Pt surface that covers the Si. To investigate this aspect we studied the cross-sectional microscopy. In Fig. 3 the cross-sectional views are shown. It can be clearly seen that the non-aligned NWs grows directly from the Pt layer while the aligned NWs grows from an interfacial layer of thickness ∼70 nm. In order to have an insight about the interface of the aligned nanowires (200 mJ) cross-sectional FESEM-EDS analysis (Fig. 4) was done on the sample. From Fig. 4, the FESEM-EDS analysis shows that in case of growth of aligned nanowires the Pt layer gets modified with the presence of W in that active interface layer. We show later from the TEM analysis (Fig. 10 and 11), the composition of the active layer. In both the cases the other growth parameters being constant, only the laser energy driven characteristic of the ablated material difference leads to change in the interfacial layer. The higher laser energy (200 mJ) helps in impinging of W/WO_x adatoms in Pt layer.

Fig. 4 EDS analysis done in FESEM images showing the presence of Pt and W at the active layer of ∼70 nm at the Pt/NW interface of the aligned NWs deposited at 200 mJ laser energy.

3.3. Microstructural analysis

In order to further confirm and comprehend the growth mechanism of the WO₃ film/NWs, TEM, HRTEM, scanning TEM-EDS line scan and mapping were carried out to observe the morphology, composition, and the structure. Fig. 5a–c shows the change in the SAED patterns from the film (Fig. 5a) to single nanowires (Fig. 5b and c) with the change in the laser energy. From Fig. 5a–c the change in the structure from monoclinic to orthorhombic are clearly visible.⁴¹ Details of the WO₃ films grown at laser energy of 110 mJ were further clarified by TEM (Fig. 6a and b). In Fig. 2 it is already seen that the morphology of the nanocrystalline film is granular with grain size ∼30 nm. The cross-section of the sample shows amorphisation of some portion of the film due to sample preparation in FIB. It is well known that sample preparation with FIB leads to amorphisation and Ga implantation in many cases. This can be optimized but cannot be fully reduced. During the sample preparation in our case we have taken care of amorphisation and Ga implantation (∼4%) to some extent in the three samples by using a low energy ion beam (16 kV) and low beam current (0.24 nA) from the beginning to the end of the lamella preparation. Higher beam energy/current leads to huge implantation of Ga. Fig. 6a shows that only a thin layer of WO₃ film (∼40 nm) under the protective layer of Pt remains crystalline. The fringes corresponds to spacing of about 0.32 nm (Fig. 6b), which corresponds to the (220) plane of the monoclinic cell having unit cell parameters a = 0.7297 nm, b = 0.7539 nm, c = 0.7688 nm.⁴¹ The Fast Fourier Transformation (FFT) of HRTEM image (Fig. 6d) of a single nanowire (Fig. 6c inset) exhibits that the nanowire is single-crystalline for non-aligned NWs (160 mJ laser energy). The fringes here are consistent with the orthorhombic symmetry, identifying normal distances of 0.3828 nm corresponding to the diffraction plane (002). Similar observation was made in case of the aligned nanowires (200 mJ laser energy) (Fig. 6e and f). Here also the FFT and the fringes are consistent with the orthorhombic structure.⁴¹ Stoichiometry of both the nanowires has been confirmed by STEM-EDS mapping (Fig. 7a–f) which shows uniform distribution of the W and O within a single nanowire. The only difference between the two wires as observed from the SAED pattern (Fig. 8b and d) is the zone axis. Hence to find out the growth direction of the NWs as well as the coordination of the atoms in the octahedra JEMS® simulation software is employed. The HREM maps generated using multislice option simulation in JEMS® (Fig. 8a and c)³⁸ shows that though the crystal structure for both the single nanowires are orthorhombic but there is a distortion in the crystal symmetry of the wires. According to JEMS® analysis of the SAED patterns the growth direction of the wires are found to be different. The growth direction is [001] for a single NW grown at 160 mJ and [100] for a single NW grown at 200 mJ (Fig. 8b and d the simulated SAED patterns are shown by red spots on top of the experimental SAED pattern). In the review work⁴¹ we have a clear idea about the relation between the different crystal symmetry of WO₃ and the related growth directions. In the case of monoclinic (m), orthorhombic (o) and tetragonal (t) the lattice parameters are related by the relation a_m ≈ a_o ≈ (a_t − b_t); b_m ≈ c_o ≈ (a_t + b_t); c_m ≈ b_o ≈ c_t.⁴¹ In experiments using the gas condensation technique,¹⁵ the selective preparation of WO₃ [001]_t preferential growth direction could be obtained by varying the O₂ partial pressure. From the simulated HREM map a schematic (Fig. 9a–c) was drawn to emphasize the distortions between the two wires and their crystal symmetry. The superimposed schematic (Fig. 9a tilted 18° w.r.t zone axis and superimposed on Fig. 9b and yield, Fig. 9c) clearly shows that there is a distortion in the co-ordination of the atoms in the two wires though the crystal structure being orthorhombic in both the cases. This is due to the difference in the growth directions of the wires which is due to the effect of the laser energy that attributes to different growth dynamics.

Fig. 5 Selected area electron diffraction pattern showing the structural change from nanocrystalline monoclinic structure to single crystalline orthorhombic wires. (a) SAED pattern from the film (inset) grown at 110 mJ showing monoclinic structure (b) SAED pattern from the single nanowire (inset) grown at 160 mJ showing orthorhombic structure and (c) SAED pattern from the single nanowire (inset) grown at 200 mJ showing orthorhombic structure.

Fig. 6 HRTEM micrographs showing (a) nanocrystalline film at 110 mJ (b) fringes corresponds to (220) orientation of monoclinic structure from the marked yellow portion of the film; HRTEM micrographs showing (c) single nanowire at 160 mJ (thickness ∼ 200 nm) with FFT at the inset showing single crystallinity (d) fringes corresponds to (002) orientation of orthorhombic structure (e) single nanowire at 200 mJ (thickness ∼ 70 nm) with FFT at the inset showing single crystallinity (f) fringes corresponds to (002) orientation of orthorhombic structure.

Fig. 7 The STEM-EDS area mapping shows uniform distribution of W and O for single NWs (a) STEM image of a single NW deposited at 160 mJ, (b & c) O and W mapping of the same wire (d) STEM image of a single NW deposited at 200 mJ and (e & f) O and W mapping of the same wire.

Fig. 8 (a & c) Simulated HREM map generated on the basis of the TEM analysis from the SAD pattern for the single NWs deposited at a laser energy 160 mJ and 200 mJ respectively. (b & d) Simulated SAD patterns (red spots) showing difference in the zone axis between the wires deposited at 160 mJ and 200 mJ laser energy.

Fig. 9 Schematic showing orthorhombic crystal symmetry in a single NW deposited at (a) 160 mJ (b) 200 mJ and (c) both the structure superimposed [(a) tilted 18° w.r.t zone axis and superimposed on (b) yield (c)] to show a distortion in the coordination of the atoms and the crystal symmetry.

3.3.1. Nature of interface. To understand the growth mechanism and the role of NW substrate interface, cross-sectional TEM analysis was done on the samples. In the case of the film grown at lower laser energy; no modification had been observed at the interface. In the case of the non-aligned NWs grown at middle laser energy (160 mJ) slight diffusion of W is noticed but the extent of diffusion is very less (Fig. 10) with no drastic modification of the substrate and the interface. But for the aligned NWs grown with higher laser energy (200 mJ) a complete modification of the NW substrate interface is noticed (Fig. 11). It has been noticed in case of aligned NWs that one interfacial layer has been formed at the interface (∼70 nm), as seen from cross sectional image (Fig. 3). In order to find out the elemental composition of the interfacial layer EDS mapping was done at the aligned NW substrate interface for 200 mJ samples shown in Fig. 4, which is already identified to be a mixture of Pt and W. For further confirmation, we probed our FIB prepared cross-section of specimen of NWs with STEM-EDS line scan and area mapping on both the samples grown at 160 mJ and 200 mJ (Fig. 10 and 11 respectively). From Fig. 10 it is clearly visible that there is a little diffusion of W in the Pt but no such modification of the substrate or formation of an interfacial layer is noticed. This is also confirmed by the line scan spectrum of W and O. In this case due to the non-alignment of the NWs, Pt from the protective layer diffuses throughout the sample and gets accumulated near the interface during FIB sample preparation which can also be clearly visible from the line scan profile of Pt. From the line scan spectrum (Fig. 11) presence of W and O is clearly visible within the Pt layer at the interface in the case of aligned nanowires grown at 200 mJ. From the STEM-EDS area mapping also the presence of non-stoichiometric WO₃i.e. WO_x is ensured. This may happen due to the implantation of WO_x into the Pt layer due to the high laser energy induced diffusion as already seen in cross-sectional EDS analysis in FESEM. This modified Pt layer during the growth has an immense effect on the growth morphology of the aligned nanowires. It is well reported that favorable nucleation sites reduce the overall free energy contribution of the system and thus ensure the growth of a particular phase. In the present case, the modified Pt layer probably reduces the nucleation barrier represented by interface energy and thus serves as favorable nucleation site for the growth of aligned nanowires.

Fig. 10 STEM-EDS line scan and area mapping of the non-aligned nanowires grown at 160 mJ showing the overall uniformity in the chemical composition of the wire and extent of diffusion at the wire/Pt interface. The yellow box shows the mapping area and the red line shows the line scan length along the upward direction from the interface to the top.

Fig. 11 STEM-EDS line scan and area mapping of the aligned nanowires grown at 200 mJ showing the overall uniformity in the chemical composition of the wire along with modified Pt layer due to diffusion of W and O. The yellow box shows the mapping area and the red line shows the line scan length along the upward direction from the interface to the top.

4. Growth mechanism

In our case the growth is PLD assisted. In PLD the growth mechanism depends upon different factors: (i) laser radiation interaction with target, (ii) dynamics of the ablation materials, (iii) deposition of the ablation materials on the substrate and (iv) nucleation and growth of film/NW on substrate surface. As the instantaneous ablation rate depends highly on the fluence of the laser, the morphology, stoichiometry and the crystallographic structure is largely governed by the fluence. The emitted materials move towards the substrate according to the laws of gas-dynamics. As the energy of the ejected materials depend on the laser fluence, at higher laser energy the ejected high energy species impinge onto the substrate surface. As this laser induced diffusion occurs, it serves as a source of condensation of the particles supplied. When the condensation rate is higher than the rate of the particles supplied, thermal equilibrium condition (supersaturation or lowering of surface energy) reached and film/NW grows on the substrate surface at the expenses of the direct flow of ablation particles. In case of PLD assisted growth nucleation rate and growth of the film/NW depends on the two thermodynamic parameters one is the substrate temperature (T_s) and the other is the supersaturation (D_s) given by eqn (2). Here K is the Boltzmann constant, R is the actual deposition rate that is fluence dependant and R_e is the equilibrium value at T_s.

D_s = KT_s ln(R/R_e) (2)

It is well known from the literatures that the nanowires grown by physical vapor deposition normally follow two main growth modes one is vapor solid (VS) and the other vapor–liquid–solid (VLS).^42–45 For the both growth modes, catalyst plays the role of promoting the formation of the crystal nucleus. In the VS growth mode,^42,43 the substrate temperature usually is not much high, and the catalyst grains are in the solid form on the substrates. The vapor from the ablated target gets deposited preferentially on the solid catalysts forming an unidirectional growth. In the VLS mode,^44,45 the substrate temperature usually is higher reaching the eutectic point leading to formation of liquid form of catalyst and vapors diffuse through the liquid catalyst and saturate on the substrates to form catalyst leading NWs. Because of the instability of catalyst pellets, the nanowires were usually crooked. Another growth mode is Preferential Interface Nucleation (PIN).²³ In PIN induced growth the formation of a nanostructure is considered independent of the phase diagram or the eutectic temperatures of the involved materials and proceeds only by nucleation at one of the interfaces between the catalyst and the growing nanostructure, i.e. the catalyst–nanostructure interface.²³

Fig. 12 shows the schematic of the different growths that occurs in our case with the increase in the deposition flux. At lower laser energy (110 mJ), the formation of film like morphology is justified; it has already been reported earlier in cases of other binary oxides that below a certain laser energy no nanowire formation occurs.⁴⁶ Hence similar kind of observation is also applicable in our case where a laser energy less than 160 mJ promotes only film like morphology. This is because of the fact that when the laser energy is less then the R/R_e value is also less which lowers the value of D_s and hence large island like growth occurs which coalesce due to the presence of T_s and forms film like morphology. With the increase in the laser fluence (160 mJ) the deposition rate (R) increases that increase D_s and hence promotes wire like morphology over a critical value of laser fluence. The wires are non-aligned in this case follows VS growth mode. As discussed earlier both VS and VLS are catalyst induced growth mode and here Pt act as catalyst layer making an interface between substrate and NW layer.

Fig. 12 Schematic diagram showing the growth mechanism of WO₃ film/NWs on the platinised silicon substrate with the increase in the deposition flux (laser energy from 110 mJ to 200 mJ). This schematic also shows the change in the morphology of the deposited WO₃ with the increase in the laser energy.

If we consider the case of aligned NWs, as our deposition temperature is not reasonably high enough to melt Pt (∼600 °C), it can not support VLS growth mode. However, higher laser energy (200 mJ) helps to increase R and consequently D_s. It is known that in case of pulsed laser assisted growth at higher laser energy, the laser induced diffusion of ablated materials may occur. This has already been observed and confirmed by FESEM and STEM-EDS analysis, that diffusion of W and O adatoms within Pt catalyst forms a mixed layer of ∼70 nm (Fig. 4 & 11) at the interface. Hence the interfacial region which is formed by the modification of the Pt layer acts as favorable nucleation site for the growth of aligned WO₃ NWs by reducing the overall contribution of the energies provided by surface energies of the substrate and the growing crystal as well as the interface energy between the two. Thus in our case of higher laser energy, growth mode may be assigned to PIN.

5. Conclusion and perspective

Aligned WO₃ NWs were grown (111)-oriented crystalline Pt/SiO₂/Si by PLD. The influence of laser energy on the growth morphology is studied in details. The laser energy used for the ablation, controls the plume and the energy and flux of the ablated material that then reaches the substrate on which growth occurs. This ablation parameter in PLD can be controlled which in turn leads to control of growth morphology to grow dense well aligned single crystalline nanowires. XRD confirmed the monoclinic to orthorhombic change in the structure of the WO₃ depending on the laser energy. The STEM-EDS line scan and area mapping shows a modification of the Pt layer. This modified layer reduces the surface energy of (002) phase and hence promotes the growth of aligned NWs. Jems simulation related to TEM shows the growth direction changes in a single NW with the change in the laser energy. At higher laser energy PIN induced growth of aligned WO₃ NWs was observed. The incorporation of WO_x into the Pt layer was found to promote the growth of NWs providing favorable nucleation sites by reducing the surface energy that acts as interface energy in aligned NWs. This work presents a simple and a general strategy to improve the growth of aligned WO₃ NWs on Pt/SiO₂/Si substrate, which in turn provides an opportunity to integrate them in devices. Further investigations are needed to understand the mechanism of the critical Pt catalyst size effect and the deposition temperature on the growth of NWs. These will help to have a control over NWs size and their distribution density, which is one of the critical problems for their effective integration in practical devices.

Acknowledgements

Authors, AG, SRM and BG want to thank DST, Govt. of India under UNANST Phase II (Ref. No. SR/NM/NS-53/2010) for financial support. Ankita Ghatak (one of the authors) thankfully acknowledges SNBNCBS and DST for providing her fellowship as PDRA-I. SRM would like to thank ICON Analytical Equipment Pvt. Ltd. for their constant support and financial help. The work has been done in DST supported Unit for Nanoscience. The authors wish to sincerely thank Distinguished Prof. A. K. Raychaudhuri for his helpful suggestions during this work. They also thank Mr Joy Bandhopadhyay for his untiring help during FIB sample preparation.

References

1. C. G. Granqvist, Handbook of Inorganic Electrochromic Materials, Elsevier, NewYork, 1995.
2. C. Santato, M. Odziemkowski, M. Ulmann and J. Augustynski, J. Am. Chem. Soc., 2001, 123, 10639–10649.
3. S. Djerad, L. Tifouti, M. Crocoll and W. Weisweiler, J. Mol. Catal. A: Chem., 2004, 208, 257–265.
4. F. Huguenin, E. R. Gonzalez and O. N. Oliveira Jr, J. Phys. Chem. B, 2005, 109, 12837–12844.
5. S. H. Lee, H. M. Choeng, J. G. Zhang, A. Mascarenhas, D. K. Benson and S. K. Deb, Appl. Phys. Lett., 1999, 74, 242–244.
6. C. B. Trasferetti, F. P. Rouxinol, R. V. Gelamo, M. A. Bica de Moraes, C. U. Davanzo and D. L. de Faria, J. Phys. Chem. B, 2004, 108, 12333–12338.
7. M. Satnkova, X. Vilanova, E. Llobet, J. Calderer, C. Bittencourt, J. J. Pireaux and X. Correig, Sens. Actuators, B, 2005, 105, 271–277.
8. A. I. Gavrilyuk, Electrochim. Acta, 1999, 44, 3027–3037.
9. M. Li, W. Gao, A. Posadas, C. H. Ahn and E. I. Altman, J. Phys. Chem. B, 2004, 108, 15259–15265.
10. G. Soto, W. D. L. Cruz, J. A. Diaz, F. F. Castillon and M. H. Farias, Appl. Surf. Sci., 2003, 218, 282–290.
11. S. Ma and B. G. Frederick, J. Phys. Chem. B, 2003, 107, 11960–11969.
12. B. Cao, J. Chen, X. Tang and W. Zhou, J. Mater. Chem., 2009, 19, 2323–2327.
13. Y. Djaoued, S. Balaji and R. Brüning, J. Nanomater., 2012, 2012, 674168,  DOI:10.1155/2012/674168.
14. M. Penza, M. A. Tagliente, L. Mirenghi, C. Gerardi, C. Martucci and G. Cassano, Sens. Actuators, B, 1998, 50, 9–18.
15. P. Tagtstrom and U. Jansson, Thin Solid Films, 1999, 352, 107–113.
16. M. Tong, G. Dai and D. Gao, Mater. Chem. Phys., 2001, 69, 176–179.
17. S. Roy Moulik, S. Samanta and B. Ghosh, Appl. Phys. Lett., 2014, 104, 232107.
18. R. Nechache, M. Nicklaus, N. Diffalah, A. Ruediger and F. Rosei, Appl. Surf. Sci., 2014, 313, 48–52.
19. A. Rougier, F. Portemer, A. Quede and M. El Marssi, Appl. Surf. Sci., 1999, 153, 1–9.
20. T. Yanagida, T. Kanki, B. Vilquin, H. Tanaka and T. Kawai, Phys. Rev. B: Condens. Matter Mater. Phys., 2004, 70, 184437.
21. T. Yanagida, T. Kanki, B. Vilquin, H. Tanaka and T. Kawai, J. Appl. Phys., 2005, 97, 033905.
22. T. Yanagida, T. Kanki, B. Vilquin, H. Tanaka and T. Kawai, J. Appl. Phys., 2006, 99, 053908.
23. B. A. Wacaser, K. A. Dick, J. Johansson, M. T. Borgstrom, K. Deppert and L. Samuelson, Adv. Mater., 2008, 21, 153–165.
24. H. Zheng, J. Z. Ou, M. S. Strano, R. B. Kaner, A. Mitchell and K. Kalantar-zadeh, Adv. Funct. Mater., 2011, 21, 2175–2196.
25. A. Ponzoni, E. Comini, G. Sberveglieri, J. Zhou, S. Z. Deng, N. S. Xu, Y. Ding and Z. L. Wang, Appl. Phys. Lett., 2006, 88, 203101.
26. J. Thangala, S. Vaddiraju, R. Bogale, R. Thurman, T. Powers, B. Deb and M. K. Sunkara, Small, 2007, 3, 890–896.
27. F. Xu, S. D. Tse, J. F. Al-Sharab and B. H. Kear, Appl. Phys. Lett., 2006, 88, 243115.
28. J. Zhou, L. Gong, S. Z. Deng, J. Chen, J. C. She, N. S. Xu, R. S. Yang and Z. L. Wang, Appl. Phys. Lett., 2005, 87, 223108.
29. K. Huang, Q. T. Pan, F. Yang, S. B. Ni and D. Y. He, Mater. Res. Bull., 2008, 43, 919–925.
30. R. M. Ko, S. J. Wang, Z. F. Wen, J. K. Lin, G. H. Fan, W. I. Shu and B. W. Liou, Jpn. J. Appl. Phys., 2008, 47, 3272–3276.
31. S. B. Sun, Y. M. Zhao, Y. D. Xia, Z. D. Zou, G. H. Min and Y. Q. Zhu, Nanotechnology, 2008, 19, 305709.
32. P. M. Wu, S. Ishii, K. Tanabe, K. Munakata, R. H. Hammond, K. Tokiwa, T. H. Geballe and M. R. Beasley, Appl. Phys. Lett., 2015, 106, 042602.
33. D. Ma, G. Shi, H. Wang, Q. Zhang and Y. Li, J. Mater. Chem. A, 2013, 1, 684–691.
34. J. R. G. Navarro, A. Mayence, J. Andrade, F. Lerouge, F. Chaput, P. Oleynikov, L. Bergström, S. Parola and A. Pawlicka, Langmuir, 2014, 30, 10487–10492.
35. Z. Gu, H. Li, T. Zhai, W. Yang, Y. Xia and Y. Mal, J. Solid State Chem., 2007, 180, 98–105.
36. M. F. Daniel, B. Desbat, J. C. Lassegues, B. Gerand and M. Figlarz, J. Solid State Chem., 1987, 67, 235–247.
37. A. Bose and M. Sreemany, Appl. Surf. Sci., 2014, 289, 551–559.
38. E. I. Suvorova, P. A. Stadelmann and P. A. Buffat, Crystallogr. Rep., 2004, 49, 343–352.
39. P. A. Stadelmann, Utramicroscopy, 1987, 21, 131–146.
40. A. Bose, T. Maity, S. Bysakh, A. Seal and S. Sen, Appl. Surf. Sci., 2010, 256, 6205–6212.
41. S. Pokhrel, J. Birkenstock, A. Dianat, J. Zimmermann, M. Schowalter, A. Rosenauer, L. C. Ciacchi and L. Mädler, CrystEngComm, 2015, 17, 6985–6998.
42. L. Chen, X. Fu, J. Lai, J. Sun, Z. Ying, J. Wu and N. Xu, J. Electron. Mater., 2012, 41, 1941–1947.
43. J. Lai, L. Chen, X. Fu, J. Sun, Z. Ying, J. Wu and N. Xu, Appl. Phys. A: Mater. Sci. Process., 2011, 102, 477–483.
44. R. S. Wagner and W. C. Ellis, Appl. Phys. Lett., 1964, 4, 89.
45. T. I. Kamins, R. S. Williams, D. P. Basile, T. Hesjedal and J. S. Harris, J. Appl. Phys., 2001, 89, 1008–1016.
46. H. Li, L. Chen, Y. Zhao, X. Liu, L. Guan, J. Sun, J. Wu and N. Xu, Nanoscale Res. Lett., 2014, 9, 91.

---