Effect of Ablation Rate on the Microstructure and ... - ACS Publications

Subscriber access provided by University of Massachusetts Amherst Libraries

Article

Effect of ablation rate on the microstructure and electrochromic properties of pulsed laser deposited Molybdenum oxide thin films S. Santhosh, M. Mathankumar, Selvaraj Selva Chandrasekaran, Amal Kaitheri Nanda Kumar, Palanichamy Murugan, and Balasubramanian Subramanian Langmuir, Just Accepted Manuscript • DOI: 10.1021/acs.langmuir.6b02940 • Publication Date (Web): 05 Dec 2016 Downloaded from http://pubs.acs.org on December 7, 2016

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Langmuir is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 47

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

Effect of ablation rate on the microstructure and electrochromic properties of pulsed laser deposited Molybdenum oxide thin films S. Santhosha,b, M. Mathankumara,b, S. Selva Chandrasekarana,b, A. K. Nanda Kumara, P. Murugana,b, B. Subramanian*a,b a

CSIR-Central Electrochemical Research Institute, Karaikudi- 630 003,India

b

Academy of Scientific and Innovative Research (AcSIR), Central Electrochemical Research Institute, Karaikudi 630 003, India *Corresponding author Tel: +91 4565 241538, Fax: +914565 227713 E-mail: [email protected], [email protected]

Abstract: Molybdenum trioxide (MoO3) is a well-known electrochromic material. In the present work, n-type α-MoO3 thin films with both direct and indirect band gaps were fabricated by varying the laser repetition (ablation) rate in a pulsed laser deposition (PLD) system at a constant reactive O2 pressure. The electrochromic properties of the films are compared and correlated to microstructure and molecular level coordination. Mixed amorphous and textured crystallites evolve at the microstructural level. At the molecular level, using NMR and EPR, we show that the change in the repetition rate results in a variation of the molybdenum coordination with oxygen: at low repetition rates (2 Hz), larger the octahedral coordination and greater the texture whereas at 10 Hz, tetrahedral coordination is significant. The anion vacancies also introduces a large density of defect states in the band gap, evidenced by XPS studies of the valence band and supported by DFT calculations. The electrochromic contrast improved remarkably by almost 100% at higher repetition rates whereas the switching speed decreased by almost six times. While the electrochromic contrast and coloration efficiency was better at higher repetition rates, the switching speed, reversibility and stability was better at low repetition rates. This difference in the electrochromic properties of the two MoO3 films are attributed to the variation in the defect and molecular coordination states of the Mo cation. Keywords:

MoOx,

Pulsed

Laser

Deposition,

repetition

Electrochromism.

ACS Paragon Plus Environment

rate,

oxygen

vacancy,

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

1. Introduction The present day display technologies based on Liquid Crystal Displays (LCDs) and Light Emitting Diodes (LEDs) have slowly replaced the earlier generation of cathode ray tube (CRT) displays. While the CRT did indeed produce clear images with fast switching and good coloration, the energy requisite to maintain the high vacuum inside the tube added to the high potential needed to accelerate the electron beam led to its gradual phase-out (1). In the LCDs, the images are sharp and clear, but fabrication of large area display turns out to be very complicated because the stringent conditions of exactly parallel electrode plates to ensure field-uniformity are hard to meet in large area displays and also owing to the complexity in addressing a single pixel (2). The LED displays show good coloration and wide angle view, but it requires a constant current source, cooling technique etc. The colors emitted by old LEDs appear faded when compared to new LEDs of the same material. In the current market, a combination of LED-LCDs is also used to enhance the display over large areas and angles. The new technology which competes with the CRT and LCD are electrochromic display devices, which can be fabricated using electrochromic materials (1). Electrochromic devices utilize low dc voltages to change the optical state and it does not require any energy to maintain its state, thus lowering the energy consumption (3). Electrochromic materials can also be used as switchable mirrors, anti-glare glasses, optical shutters and infra-red (IR) radiation absorbers (4,5). This technology is now replacing the cooling system in our habitats (6,7). One square meter of an electrochromic window can reduce the energy usage by nearly 170 KWh annually (8). Electrochromism is a reversible phenomenon by which a material is able to change its optical properties under the application of modest electric fields (9-12). Electrochromic materials can be classified into three types depending on the type of the electrochrome (electroactive species). Type I electrochrome exists in a solution phase (methyl viologen in water). The solution (electrolyte) becomes colored during the redox process. In the case of type II electrochrome, initially the electroactive species is in a liquid phase, but becomes colored during precipitation (for instance, heptyl viologen). Type III electrochrome exists completely in the solid phase and examples are molybdenum oxide, tungsten oxide, Prussian


Page 2 of 47

Page 3 of 47

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

blue etc (1). In a device-level electrochromic system, three parameters are often used to describe the chromism of the active material: fast switching (typically between a few seconds or lesser), electrochromic contrast (color contrast between the colored and bleached state at the wavelength of interest, typically observed from its optical transmission) and stability (the number of charges that can be inserted and withdrawn without loss) (13-15). While both organic materials like viologen, pyrazoline, some conductive organic polymers and inorganic transition metal oxides like MoO3, WO3, IrO2, NiOx, and V2O5 (16,17) have been identified as electrochromic materials, devices fabricated using the organic electrochrome require a good sealing system to prevent leakage since most of the organic electrochromic materials are in a liquid phase. The diffuse coloration and lack of stability associated with organic electrochromes has made inorganic materials ubiquitous for electrochromic devices that require fast switching, contrast and stability. Moreover, both anodic/cathodic (or sometimes both, as in V2O5) coloration can be achieved in various metal oxides, thereby lending a certain degree of flexibility in the choice of the oxide material. Nanostructured Materials reportedly show enhancement in the electrochromic performance (18,19). Among these transition metal oxides, molybdenum trioxide (MoO3) is superior to other electroactive oxides like IrO2, NiOx etc. because its high visible light absorption centered at around 500, 625 and 770 nm results in good coloration efficiency (20, 21). It is a type III electrochromic material with cathodic coloration. It is used in applications ranging from catalysis, electrodes in rechargeable lithium ion batteries, gas sensors and in thermochromic and photochromic applications due to its various oxidation states, coordination number and stoichiometry (22-24). Stoichiometric MoO3 is a wide band gap ptype semiconductor but can be used as a n-type semiconductor by creating oxygen vacancies (25-27). It exists in three different phases: the thermodynamically stable orthorhombic (αMoO3), metastable monoclinic (β-MoO3), and hexagonal (h-MoO3) structures (28,29). Among these, orthorhombic α-MoO3 has several applications due to its layered structure. The unit cell of α-MoO3 comprises a set of four Mo ions coordinated in an octahedral arrangement to 6 oxygen anions (see Figure 1c). It crystallizes in the space group, Pnma (Pbnm in the non-standard notation, obtained by a cyclic permutation of the a,b,c axes from the standard Pnma notation).It exhibits an intrinsically layered arrangement with corner and edge-shared MoO6 octahedra. The unit cell contains three kinds of oxygen ions for Mo−O bonding, repeating their ordering along the long b axis, providing enough free volume for small cation intercalation during redox and Faradaic reactions. Weak electrostatic and van der


Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 47

Waals contributions are responsible for the interaction between bilayer and the adjacent bilayers along b since it is non-covalent (30,31). In a real time device, the electrochromic metal oxide is coated as a thin film over a suitable (often transparent) substrate. The deposition method strongly affects the adhesion, occurrence, and variations of the photo/electro-optic effect in the films (32, 33). They also alter the morphology leading to a change in impedance, diffusion of the intercalating species, electrode reaction kinetics etc. Different film growth techniques are currently used for molybdenum oxide thin film deposition such as electron beam evaporation, sputtering, chemical vapor deposition (CVD)and electrodeposition. (34-37) In comparison to the above, the pulsed laser deposition (PLD) is an excellent technique to maintain the stoichiometry of complex oxide substances like La-Ba-Cu-O, YBaCuO7 etc (38). High quality thin films ranging from epitaxial single crystal to fully amorphous

microstructures

of

complex

oxides,

cuprates,

nitrides,

ferroelectric,

ferromagnetic, and dielectric oxides can be grown using this technique. The morphology, oxygen stoichiometry, crystal structure and lattice strain are altered as a consequence of varying the deposition parameters like energy of the laser pulse, substrate temperature, partial pressure of (O2) gas, frequency of the laser shots etc. Highly adhesive films are also obtained using PLD owing to the bombardment of high energetic species on the substrate (39). High rates of film growth are achieved between each laser pulse. Growth occurs between two consecutive pulses if an interval sufficient for surface diffusion of the adatoms can be achieved. When arrested, for instance by increasing the pulse frequency (repetition rate), the nuclei size can be controlled allowing controlled microstructures to be developed, ranging from epitaxial to fully amorphous (40). In this work, thin films of Molybdenum oxide were grown on glass and FTO substrates by PLD technique, and at five different laser pulse repetition rates viz 2, 4, 6, 8 and 10 Hz (designated MO-2Hz, MO-4Hz, MO-6Hz, MO-8Hz and MO-10Hz respectively). The duration of the deposition was varied between 2to 10 mins so as to maintain a total of 1200 shots for each batch of samples and to maintain nearly equal thickness of all specimens. In the following sections we report the observed microstructure and electrochromic properties of the films. The intention of this work is to observe general trends in the electrochromic properties as the repetition rate is varied. For the sake of conciseness, only results pertaining


Page 5 of 47

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

to the two extreme cases (2 and 10 Hz) are presented here. Where remaining data for the intermediate cases are also given either in main or in supplementary information.

2. Experiments and methods 2.1 Thin films fabrication Sintered pellets made from commercially purchased MoO3 powder (Merck, 99% pure) were used as the targets for laser ablation. Well cleaned glass and FTO (Fluorine doped tin oxide) coated conducting glass plates (1 cm2) were used as substrates. The target was ablated using a KrF excimer laser (COMPex Pro, Coherent, Germany) operating at a wavelength of λ= 248 nm and pulse width 9 ns and the laser repetition rate was fixed from 2 Hz to 10 Hz (corresponding to a deposition rate of about 0.25 nm/shot). The ablation energy density for all depositions was fixed at 150 mJ/cm2. The chamber was initially pumped using a combination of rotary (HINDHIVAC direct drive pump, model ED-21 with the free air displacement capacity of 350 litre/minute) and turbo pumps (PFEIFFER Germany, model HiPace 300 DN 100 ISO-K, (PMP03900) with the pumping speed of 260 liter/second for N2) to 5 x 10-6 mbar, after which O2 gas was leaked to the pressure of 5 x 10-2 mbar in the chamber. To maintain approximately equal thickness for all sets of samples, the total number of laser shots was kept constant (1200 shots), ensuring a reasonable thickness of ≈300 nm. All the films were fabricated at room temperature. The cross section FESEM images of MO2Hz and MO-10Hz are shown in Figure SS1.

2.2 Characterizations The phase and structure of the thin films coated on glass substrates were investigated using X-Ray diffraction (D8 Advance, Bruker) with Cu Kα radiation (λ = 1.5418 Å) at a scanning rate of 0.02 steps s-1. Samples for transmission electron microscopy studies were observed using both conventional TEM (Tecnai 20 G2, FEI make, the Netherlands) and high resolution (HR) TEM (Tecnai 20, G2,FEI) operating at 200 kV and capable of an information limit of 0.14 Å. The oxidation states of Mo and O in the MoO3 thin films were examined using X-ray Photoelectron Spectrometer (Thermo scientific model: MULTILAB 2000), operating with an Mg source with hυ = 1253.6 eV. The cross section images of the films was observed with a field emission SEM (Zeiss Ultra). The optical absorption and transmittance studies were carried out using UV-visible spectrophotometer (UNICO-model: 4802). The


Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

photoluminescence spectra were recorded using a florescence spectrophotometer (Make: VARIAN, Model: Cary Eclipse, Source: Xenon pulse lamp) with an external assembled high temperature setup that can increases the top to 600 K. The excitation wavelength of 300 nm is used for PL studies. Nuclear magnetic resonance (NMR) studies were carried out using a Bruker Advance 400 MHz spectrometer. The MO-2Hz and MO-10Hz samples for liquid mode NMR studies were prepared by dispersing the flakes peeled out from relatively thick films in heavy water D2O solvent. The magnetic field strength was 9.3950 T and the resonant frequency was 26.059 MHz. NMR studies were also performed on two other Mo-containing systems: bulk α-MoO3 powder and Na2MoO4 compound. The Mo-O coordination is completely octahedral in bulk MoO3 powders and tetrahedral in Na2MoO4 powder, thereby providing a reference level to identify six-fold and four-fold co-ordinations. Electron paramagnetic resonance (EPR) spectroscopy was conducted in a Bruker EMX-Plus instrument operating in the Q-band (υ=9.859 GHz) at room temperature.

2.3 Electrochromic experiments: The electrochemical studies were carried out in a conventional three-electrode cell, in which the MoO3 film coated on FTO substrate was a working electrode, a platinum wire acted as the counter electrode and Ag/AgCl was the reference electrode. A 1 M LiClO4 in Propylene Carbonate was used as an electrolyte solution. The cyclic voltammetry (CV) and chronoamperometry (CA) experiments were carried out using PARSTAT 2273-Advanced electrochemical system. 2.4 Computation methods: The DFT method which has proven to be one of the most accurate methods for the computation of the electronic structure (41). First principles density-functional calculations were performed to study the electronic structure in both pristine and O-deficient MoO3 compounds using the Vienna ab initio simulation package (VASP) codes (42, 43). In the calculations, the wave functions of valance electrons in the Mo (4p, 5s, 4d) and O (2s, 2p) atoms are considered as augmented plane waves and the generalized gradient approximations were used to estimate exchange-correlation energetics (44, 45). To optimize the lattice constants of MoO3, an energy cut-off of 350 eV was used and the Brillouin zone of the 2×2×1 supercell was sampled by a 5×5×2 k-mesh. Each electronic and ionic step was alternatively and self-consistently performed until the energy difference between two


Page 6 of 47

Page 7 of 47

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

consecutive steps reached 0.001 meV. Finally, to calculate the band structure and density of states, the optimized Brillouin zone was sampled with 360 k-points along high symmetry directions of both pure MoO3 and O-vacancy defected MoO3. The calculated lattice parameters of orthorhombic α-MoO3 (space group of Pbnm) are a = 3.901 Å, b = 13.904 Å, and c = 3.70 Å which are in close agreement with that of experimental values (46).

3. Results 3.1 Structural and microstructural characterization Figure 1(a) compares the X-ray diffraction pattern obtained from bulk α-MoO3 powder with those thin film. The reflections from the films matched with the thermodynamically stable layered orthorhombic α-MoO3 phase with a= 3.9630 Å, b =13.8560 Å and c = 3.6966 Å (ICDD Card No. 005-0508, space group #62). Linear behavior of intensity with respect to repetition rate were observed which is shown in Figure 1(b). A relatively higher intensity of the peaks was observed for the MO-2Hz compared to the MO10Hz, indicating a higher degree of crystallinity at lower repetition rates. We attribute this to the slow arrival rate of the ablated species on the substrate at low repetition rates, allowing for equilibration (crystallization) of the depositing atoms and ions by surface diffusion and desorption. Consequently, the MO-2Hz film also crystallizes much earlier than the MO-10Hz films (after roughly 15 nm, the former revealed widespread incipient nuclei whereas the latter showed no evidence of crystallization), which was confirmed by HRTEM (discussed in the following section). Interestingly, both the films showed strong texturing predominantly along (060) and with a relatively lesser intensity along (200), implying that most of the crystals grew with the long b axis oriented normal to the plane of the substrate. The texture coefficient (Thkl) of all (hkl) planes was measured from XRD pattern using the formula(47) =

( ) ( ) ( ) ∑ ( )

(1)

where, I(hikili) is the intensity of (hikili) diffraction peak of the sample and Io(hikili) is the intensity of (hikili) plane given in the ICDD card, N is the number of diffractions obtained. The texture coefficients of all diffraction peaks are given in Table 1. The peak widths (FWHM) of the strong (060) increased by around 4% for the MO-10Hz (0.0904) in comparison with the MO-2Hz (0.0872) sample. The texture coefficient of the most intense


Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

peak (060) decreased with increasing repetition rate, implying that the material tends to become polycrystalline with increasing repetition rate. To observe the microstructure of two extreme films in detail, TEM analysis was carried out. The microstructure at the initial stages of growth was studied by depositing the films (for ~30 s) directly over formvar-supported Cu grids. The final microstructure was also studied by scratching a thin layer of the deposited films (microtomy) and drop casting them over Ar-etched Cu grids.

The HRTEM images of MO-2Hz are shown in Figure2a-c. The initial stages of growth are defined by a predominantly amorphous matrix in which ultrafine nuclei (~ 5 nm) could be distinguished by the observation of lattice fringes. The TEM micrographs of the film after deposition are shown in Figure2d-f. Compared to the microstructure at the initial stages of growth, this sample showed three types of diffraction contrast: amorphous contrast from some areas, ring patterns indicating small crystallites and strong reflection arcs indicating a tendency for texturing of the particles. Dark field images showed clustering of fine grains (the average grain size after a log normal fit was 10.35±0.81 nm). However, these clusters were only occasionally observed while a large part of the microstructure showed textured diffraction from larger grains. Interestingly, fringes with oscillating contrast were also visible, indicating a layered morphology. In the initial stages of growth the MO-10Hz appeared to be completely amorphous and devoid of any diffraction contrast (Figure 3a-c). In comparison with MO-2Hz, the crystallization kinetics during the initial stages of deposition appeared to be slower for MO10Hz. After deposition, the MO-10Hz specimen showed fine isotropic nanocrystallites and larger crystals (appearing with a dark contrast in the bright field images shown in Figure3df). The diffraction pattern revealed both rings (from the ultrafine particles) and also spots from some of the larger grey contrast areas. Dark field imaging from the rings established the presence of nanocrystallites dispersed uniformly in a grayish background, similar to the MO2Hz specimen. This was used to calculate the particle size distribution (23.3±0.73 nm). However, the distribution had a long tail at the higher end, indicating a non-uniform grain growth and also clustering within coalesced grains. The large areas of gray contrast, after appropriate tilting showed spot pattern which were subsequently indexed to the [020] zone axis of MoO3. Interestingly, the dark field images from the reflections revealed the regions to be a composite mixture of many small crystallites, all of which were


Page 8 of 47

Page 9 of 47

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

approximately oriented along the [020] axis. Probably, the crystal growth mechanism appears to be one of small crystallites coalescing to form a larger crystal in a bid to lower their surface free energy. The tendency for coalescence seems to be higher for MO-10Hz than MO-2Hz. 3.2 Core level and valance band XPS studies. XPS high resolution spectra corresponding to Mo and O are shown in Figure 4. The elemental composition was calculated using CasaXPS software by fitting the peaks of individual species. The results are shown in Table 2. All the samples showed slight oxygen deficiency. It was observed that the oxygen deficiency increases with increasing repetition rate. The Mo-3d on deconvolution gives additional peaks attributed to Mo5+ oxidation state. Similarly, O 1sgives two deconvoluted peaks attributed to lattice oxygen at lower energy and adsorbed oxygen at higher energy. The positions of the peaks are given in Table 2.The binding energy difference (∆BE) between the O-1s and Mo6+ 3d5/2 were calculated to be around 297 eV which is in good agreement with previous reports of MoO3(48, 49). The percentage of Mo5+ is increased with increasing repetition rate. The valance band (VB) spectra of the films acquired at a resolution of 0.02 eV are shown in Figure 5. The asymmetric peak of valance band confirms the presence of oxygen vacancies. Using a linear fit to the raising edge of the VB and noting its intersection to the base line, the VB maximum was found to lie ~3.7 eV below the Fermi level (EF). This value in comparison with the observed band gap of 2.8 – 3.7 eV suggests that the EF is pushed into the conduction band (CB) indicating n-type conductivity, by the presence of oxygen vacancies. Interestingly, in the case of MO-6Hz, EF lies below the conduction band. This observation is agreement with independent dc conductivity results (discussed in section 3.3). A high density of (mid gap) defect states are also observed near the EF as seen from the tapering background level in the VB spectra. This observation of defect states populating the band gap is also reflected in the loss peaks emerging from the O 1s core level peak shown in FigureSS2. The photoelectrons from the O 1s core level undergo inelastic energy losses due to their interaction with the VB and CB plasmons and band-to-band excitations. These losses lead to the occurrence of secondary satellites~5 eV higher than the core level peak. Following the method of Nohiraet. al., (50) the binding energy (BE) separation between the core and the loss peaks were equated to the band gap and determined to be 2.8 and 3.01 eV for MO-2Hz


Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 47

and MO-10Hz respectively. This value is in good agreement with the Eg calculated from UVvis spectrophotometry (discussed in section 3.3). However, the loss peaks are also rather broad and almost continuous to the core level peak in the MO-10Hz specimen indicating a dispersion in the plasmon-core electron interaction, hinting at the occurrence of a substantial population of electrons in the mid gap states also. In comparison to the sample deposited at low repetition rates, it then appears that at high deposition rates not only the microstructure, but the electronic structure is also altered significantly. 3.3 Optical studies: The optical band gaps were calculated from UV absorption data using the Tauc equation: (51) αhυ = A × (hυ− Eg)m

(2)

where, α is the absorption coefficient, h is the Plank’s constant, υ is the frequency of the photon, A is a absorption edge width parameter, Eg is the band gap, and m is the exponent which predicts the type of transition (m can take values ½,3/2, 2 and 3 respectively for direct allowed, direct forbidden, indirect allowed and indirect forbidden transitions). Generally, MoO3 is an indirect band gap semiconductor (31). Figure 6 shows the variation of band gap with increasing repetition rate. The band gap values of all films lie between 2.8 to 3.7 eV which is in good agreement with previous reports (17, 24, 25, 27, 32). The plot of band gap with respect to repetition rate (Figure 6) shows a bowing effect as observed in alloys. DC conductivity measurements for the samples are shown in figure 6, which also shows a bowing effect. The independent measurements of resistivity and band gap both indicating a bowing effect suggest that there is a gradual transition in the properties from 2 Hz to 10 Hz. MO-10Hz and MO-8Hz samples shows the favorable microstructure for electric conduction even though these films has larger optical band gap than MO-2Hz. The Tauc plot of the five samples are shown in FigureSS3. MO-10Hz and MO-8Hz gave a satisfactory linear fit with m=2 agreeing with the general notion of MoO3 being an indirect band gap semiconductor. But, interestingly in the case of MO-2Hz, MO-4Hz and MO-6Hza satisfactory linear fit is obtained only for m=1/2 indicating a direct band gap. Therefore, the repetition rate alters the electronic transition from indirect to direct.


Page 11 of 47

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

To further elucidate the occurrence and evaluation of defect states as a function of these repetition rate, the ln absorbance was plotted against energy of the incident photon and are shown in figure SS4. The MO-10Hz has a tail region whereas MO-2Hz does not show any tail below the absorption edge. The oxygen deficient MO-10Hz naturally has a higher intrinsic inhomogeneity. According to Wang and Pan’s Model (52), this results in a fluctuating potential leading to indirect transition. To confirm the observation, high temperature photoluminescence (PL) studies were carried out and the spectra are shown in Figure SS5. The peak broadening of MO-10Hz and MO-8Hz are asymmetric in nature which reveals the presence of multiple peaks. The fitted data of photoluminescence spectrum are tabulated in the Table SS1. Area of the deconvoluted peak at ~526 nm for MO-10Hz and MO-8Hz tends to increase whereas the area of the peak at ~530 nm tends to decrease with increasing temperature. In the case of MO-6Hz, the minor peak (~531nm) is observed along with a major peak (~526nm) at 300 K and 400 K, but this minor peak disappeared and an enhancement of the major peak was observed with further increasing the temperature. This shows at room temperature MO-6Hz film shows indirect band gap, at higher temperature it shows direct band gap. Tauc plot of MO-6Hz film gives good linear fit with m=2 and m=1/2, which shows it has both direct and indirect component. The best fit is m=1/2 which is direct band gap. In the case of MO-2Hz and MO-10Hz there is absence of extra peaks. The spectrum is easily fitted using single peak with perfect fit. Also the area of the peaks are higher than the remaining three films (53). The defect states and trapping centers formed due to oxygen vacancies may responsible for the direct to indirect transition. These oxygen vacancies are also a cause strain in the film. So the strain of MO-2Hz and MO10Hz are calculated using Williamson hall plot, and it was shown in Figure SS6. This confirms that the compressive strain arises due to oxygen vacancies. So, DFT calculations was used to confirm whether it showed any transition from direct to indirect band gap on compressive loading. The DFT results (Figure SS7) showed that both unstrained MoO3 and 2% elongation strain along z-axis have direct transition. There is no indication of indirect band gap during elongation, so 2% compressive strain was supplied along z-axis, which slightly shows the indication of indirect transition. Further compression (4% along z-axis) confirmed the presence of indirect band gap. So this confirms that the repetition rate indirectly produces/alter strain in the film which alters the electronic band structure of the films resulting in direct to indirect band gap transition.


Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 47

3.4 Electrochemical studies of MoO3 thin film: 3.4.1 Cyclic voltammetry Cyclic voltammograms of MoO3 films are shown in Figure 7a. The potential was swept from +1.0 to −1.0 V (versus Ag/AgCl) at a potential scan rate (v) of 50 mV s−1 in 1 M (LiClO4 + PC) electrolyte. During the potential scan the current obtained due to Li+ ion intercalation and deintercalation was recorded and the variation is shown (Figure7a). Coloration was obtained during the cathodic potential sweep and bleaching was observed during the anodic scan. When compared the two extreme condition films, first interesting aspect was the observation of a higher current density of about 8.21mA cm-2 for the MO10Hz film compared to 5.34 mA cm-2 for the MO-2Hz.Moreover, the coloration threshold potential of MO-10Hz was lower than MO-2Hz. The diffusion coefficient of Li+ in MoO3 thin films was calculated using the relation (54) ip = 2.69 x 105⋅n3/2⋅D1/2⋅ C0⋅A ⋅v1/2

(3)

where ip is the peak current in mA, n is the number of electrons involved in the redox reaction, D is the diffusion coefficient (cm2 s-1) of Li in MoO3, C0 is Li+ concentration in the electrolyte (1 mole), A is the area of an electrode in cm2, and v is the potential scan rate (V s1

). The diffusion coefficient of MO-2Hz thus calculated was 5.793 x 10-15 cm2s-1 whereas for

MO-10Hz it was 16.45 x 10-15 cm2 s-1. Higher diffusivity of Li+ ion was observed for MO10Hz. 3.4.2 Chronoamperometry In order to determine the kinetics of the reduction and oxidation switching mechanism, chronoamperometric cycling experiments were carried out with the potential step of +1 and −1 V versus Ag/AgCl for 60 seconds each and the results are shown in Figure 7b and c (zoomed image). Both coloration current (ic) and bleaching current (ib) decreased continuously with time, which was used to measure the speed of electrochromic response of the films. The response time in both films was shorter for coloration than bleaching. Interestingly, the MO-10Hz film required a much longer duration for the standard 70% coloration (15 s) as well as bleaching (40 s) than MO-2Hz (which colored in 2 s, while bleaching required 10s). The MO-2Hz showed faster switching between the two states


Page 13 of 47

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

(colored/bleached) (Table3). The area covered under the coloration (cathodic) and bleaching (anodic polarization) curves correspond to the amount of charge intercalated and deintercalated. Reversibility of the electrochromic material is also an important parameter to check the stability of the phenomenon. This is the ratio between the de-intercalated charge Qdi during bleaching and intercalated charge Qi during coloration. In an ideal case, the reversibility is 100%, implying that the material ejects all intercalated charges during the bleaching process. The reversibility of MO-2Hz and MO-10Hz were calculated to be 86% and 64% respectively. In this aspect, the MO-2Hz shows a higher reversibility than MO10Hz.

3.4.3 Optical modulation The transmittance value at 600 nm of MoO3 thin films before and after Li+ intercalation as a function of repetition rate are shown in Figure 7d and the transmission spectrum of all samples are shown in figure SS8. The highest electrochromic contrast(∆T) at λ =600 nm, was found to be 36.1 % and 45.5 % for MO-2Hz and MO-10Hz respectively. A larger modulation of visible light is observed in MO-10Hz compared to MO-2Hz. The amount of charge required to produce the optical change is determined by the coloration efficiency (η) of the material. The coloration efficiency (optical change per unit charge of injection) of the MO-2Hz and MO-10Hz was calculated using the relation

η=

} !

{

#

cm2/C

(4)

where the term, ln(Tb/Tc) is called the optical density and Tb and Tc are the transmittance of bleached and colored films respectively. Q is the amount of charge intercalated or deintercalated. Table 3 shows the calculated values of optical density and coloration efficiency. The electrochromic coloration efficiency of MO-10Hz (of 65.25 cm2/C) was slightly better than the MO-2Hz (50.04 cm2/C). 3.4.4 Electrochemical stability of the films:


Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 47

The stability of the material was studied under an applied potential swept from 1 V to -1V vs Ag/AgCl and liquid electrolyte (1M LiClO4 in PC) at a scan rate of 50 mV/s. The films were subjected to 100 cycles with a delay of one second per cycle. The performances of the films are shown in Figure8. MO-2Hz retains the same shape of curve and peak current after 100 cycles, but in the case of MO-10Hz the peak current decreases substantially and the curve also shrinks from the initial cycle indicating that its stability is lower than MO-2Hz. 4.Discussion: The data reported in the previous sections and summarized in Tables suggest that the films prepared at the different laser repetition rates are not just trivially different in their degree of crystallization, but have a rather basic disparity in their electronic structure and consequently in their functionality. Fundamentally, there is clear evidence of a change from direct to indirect band gap by decreasing the repetition rate accompanied by a corresponding change (decrease) in the fraction of Mo5+ and improved the crystalline nature. On the other hand, the electrochemical studies categorize the MO-2Hz to have an almost similar diffusion coefficient of Li+, but lower peak current, and lower coloration efficiency than MO-10Hz. But they score favorably in the aspects of reversibility, fast switching time and stability (upto the tested 100 cycles) compared with MO-10Hz. It is worth pointing out that the diffusion coefficient alone does not translate into coloration efficiency, since the coloration occurs only by charge transfer between Mo5+ and Mo6+ions (9). The efficiency is limited by the performance of the entire cell (diffusion, intercalation and charge transfer) and its components. In the case of Li+ intercalation, the equilibrium reaction that is to be established is: MoO3 + x(M+ + e-) →[Mx+Mo(1-x)6+Mo(x)5+]O3

(5)

At low values of x (Mo5+/Mo6+

Effect of Ablation Rate on the Microstructure and ... - ACS Publications

Recommend Documents