• 沒有找到結果。

Effect of Oxygen on the Microstructural Growth and Physical Properties of Transparent Conducting Fluorine-Doped Tin Oxide Thin Films Fabricated by the Spray Pyrolysis Method

N/A
N/A
Protected

Academic year: 2021

Share "Effect of Oxygen on the Microstructural Growth and Physical Properties of Transparent Conducting Fluorine-Doped Tin Oxide Thin Films Fabricated by the Spray Pyrolysis Method"

Copied!
5
0
0

加載中.... (立即查看全文)

全文

(1)

Effect of Oxygen on the Microstructural Growth and Physical

Properties of Transparent Conducting Fluorine-Doped Tin Oxide

Thin Films Fabricated by the Spray Pyrolysis Method

N. Chantarat,aYu-Wei Chen,aShu-Han Hsu,bChin-Ching Lin,cMei-Ching Chiang,c and San-Yuan Chena,z

aDepartment of Materials Science and Engineering, National Chiao Tung University, Hsinchu, Taiwan bNational Nano Device Laboratories, Science-based Industrial Park, Hsinchu, Taiwan

cMaterials and Chemical Research Laboratories, Industrial Technology Research Institute, Chutung, Hsinchu, Taiwan

We reported the fabrication of an FTO conducting thin film via a spray deposition method, which was used to investigate the effect of oxygen content in the carrier gas on deposited film morphology and properties. Using a carrier gas containing various O2/N2concentrations (0%, 20%, 50%, 80%, and 100%) led to significant changes in the thickness, size, and shape of grain growth.

The deposited films with 0–50% oxygen content yielded a low resistivity of∼10−4-cm and a transmittance in the range of 76–96% at 550 nm. Furthermore, by changing the carrier gas concentration, the FTO films displayed different charge transport, recombination, and collection properties due to the surface and interfacial effects. These films with modified properties can be applied to dye-sensitized solar cells (DSSCs). Overall, the conversion efficiency of a solar cell based on a 0% O2sample was increased

by approximately 2% from that of a 100% O2sample. The higher efficiency is mainly the result of the lower O2content, which

minimized the grain boundaries (spacing) and improved the electron transport on the FTO film surface. © 2013 The Electrochemical Society. [DOI:10.1149/2.005309jss] All rights reserved.

Manuscript submitted February 12, 2013; revised manuscript received May 14, 2013. Published June 11, 2013.

Transparent conductive oxides (TCOs) have many applications in modern electronics. Because of their high optical transparency and metal-like conductivity, they have been widely used for electrode applications in devices such as thin-film solar cells,1 optoelectronic devices,2gas sensors,3frost-resistant surfaces,4e-windows,5etc. Up to now, most well-known TCO materials such as tin-doped indium oxide (ITO) exhibit n-type conductivity and oxygen deficiency due to oxygen vacancies and possible incorporated oxygen. However, ITO has two obvious disadvantages: (1) Providing a stable supply of ITO for the current expanding market is difficult because of the high price and scarcity of indium. (2) The electrical properties are significantly degraded after an essential annealing process in an oxidizing atmo-sphere with a temperature over 300◦C. Therefore, a substitution for ITO is needed. Fluorine-doped tin oxide (FTO) has been found to be one of most comparable materials to ITO. FTO has attracted much at-tention as a promising alternative because of its wide energy gap (Eg

= 3.67 eV), low cost of production, thermal stability, chemical in-ertness, and high transparency, although lower conductivities com-pared with the∼10−4 S/cm of ITO hinder its commercial applica-tions. In addition, FTO thin films have been extensively prepared by various methods, such as chemical vapor deposition (CVD), metallor-ganic deposition, rf sputtering, sol–gel, and spray pyrolysis deposition (SPD).6–8

Considerable effort has gone into investigating how to develop a textured surface on FTO conducting electrodes to improve the perfor-mance of solar cell devices by reducing light reflection or making light scattering more efficient. An anisotropic post-etching method using lithographic patterning and etching steps is typically used to achieve the desired morphology.9,10However, this might be deleterious to the film properties and difficult to control and could increase processing costs. Therefore, this becomes an important issue for developing high surface roughness and quality stable films during growth. For exam-ple, fabricating self-textured films via several methods, the electrical resistivity of PLD-grown 210–1000 nm thick FTO films was reported to be in the range of 2-3 m-cm,11and an MOCVD-grown FTO film with a similar thickness possessed a resistivity value in the range of 1.2-2.1-cm.12

Among the various techniques, spray pyrolysis deposition (SPD) has been economically conducted to produce self-textured FTO films with a simple scalable deposition and easy doping process that de-pends on manipulation of the substrate temperature, calcinations, gas pressure, and flow rate. On the basis of the proposed principle, the

zE-mail:[email protected]

variation in the growth and nucleation of deposited FTO films is strongly determined by the incorporated oxygen concentration, which has an effect on the surface mobility and the oxidation rate of chemical precursors.

In the present work, we report the effect of the oxygen content in the carrier gas on the FTO film morphology and optoelectronic properties during the spray pyrolysis deposition. The deposition of 400–550 nm thick films on a glass substrate led to significant changes in the thickness, size, and shape of grain growth, which is indica-tive of variable correlation between optical and electrical properties. Here, our deposited films reach a low resistivity of∼10−4-cm and a transmittance of 76%∼96% at 550 nm. Finally, the photovoltaic im-plementation of dye-sensitized solar cells (DSSCs) and electrochemi-cal impedance spectroscopy (EIS) also revealed particular behaviors, such as the charge transport, recombination, and collection properties of the surface and interfacial effects. Consequently, the higher growth rate for spray pyrolysis deposition compared with other methods pro-vides an improved understanding of the grain growth mechanisms of FTO film because various surface morphologies could be obtained by using high deposition temperatures in air. Moreover, other param-eters, including deposited temperature, duration, gas flow rate, and spray gun settings, could be further optimized to produce better elec-trical and optical film properties, leading to a significant enhancement of efficient solar cell devices.

Experimental

Fluorine-doped tin oxide films (SnO2:F) were deposited over

Corn-ing glass substrates by ultrasonic spray pyrolysis. The precursor solu-tion containing a mixture of SnCl· 4H2O (1 mM) and NH4F (0.5 mM)

was dissolved in 500 mL of DI water. To improve the solubility of the solution, 10 mL of HCl was added to the solution and stirred vigorously (∼30 min) until the solution turned transparent. The prod-uct solution was placed in an ultrasonic nebulizer reactor, which can produce an aerosol with a controlled droplet size. The experimen-tal apparatuses consist of a specific homemade ultrasonic atomizer, spray gun, and graphite hotplate. The deposition parameters, such as substrate temperature, spray rate, and distance from the spray gun to the substrate, were kept constant at the optimized value: 400◦C, 20 mL/min, and 30 cm, respectively. In addition, different morpholo-gies of textured FTO films can be obtained by applying a carrier gas containing O2 and N2during the deposition process (O2: 0%, 20%,

(2)

Figure 1. (a–e) AFM topographic 5× 5 μm scans and (f–j) cross-sectional

TEM images for oxygen-incorporated FTO films deposited at various oxygen concentrations: 0%, 20%, 50%, 80%, and 100%.

A morphology analysis (i.e., roughness, grain size, and other cross-sectional observations) of the samples was performed by a field-emission scanning electron microscope (SEM, JEOL JSM-6700F), a high-resolution transmission electron microscope (TEM, JEOL 2100) operated at 200 keV, and an atomic force microscope (AFM). The crys-tal structures were determined using an XRD diffractometer equipped with a Cu Kα radiation source (λ = 0.154 nm), and the diffraction data were collected for 2θ = 10∼70◦. For the optical analysis, a UV-VIS double beam spectrophotometer (Model: Evolution 300 BB) was em-ployed to record the optical transmission spectra of the deposited films over the wavelength range of 300–700 nm. Hall effect measurements were carried out at room temperature with a van der Pauw sample con-figuration (magnetic field: 4000 Gauss, electric current: 0.1 mA), using indium balls as the ohmic electrode, to demonstrate electrical prop-erties such as carrier concentration (n), resistivity (ρ), and mobility (μ). For device performance, dye-sensitized solar cells (DSSCs) were fabricated using N3 dye, TiO2sol, liquid electrolyte, and a Pt counter

electrode. The photovoltaic properties, including short-circuit current density (Jsc), open-circuit voltage (Voc), and conversion efficiency (η),

were measured under AM 1.5 sunlight illumination (Model YSS-80, Yamashita Denso, Japan; intensity: 100 mW/cm2) from an illuminated

area of 0.5 cm× 0.5 cm. The electrochemical impedances of the cells were also measured using the photocurrent-potential-impedance ana-lyzer by applying an open-circuit voltage as the bias in the frequency range of 0.1 Hz - 1 MHz.

Results and Discussion

Based on our spray pyrolysis method, different O2/N2

concen-trations (0%-100%) serve as the carrier gas for depositing the FTO films with different surface morphologies. The grain size, estimated from the top AFM images (Figure1a–1e), is apparently smaller for films with a higher O2 content. The optimized deposition

temper-ature (constant at 400◦C) and duration (varied by 5–10 min) were used for all film types to form films with a similar thickness in the range of 400–450 nm, as indicated in the cross-sectional TEM images (Figure1f–1j). As a result, the surface roughness (RMS) varies in a small range of 22–27, and the thickness was slightly increased. This behavior was more pronounced for higher O2 concentrations. Two

types of growth for FTO films can be classified in two cases: island-like structures at low O2 content (0%-20%) and pyramid-like

struc-tures at high O2 content (50%-100%). The side-view TEM images

reveal the grain growth mechanism, as indicated in the figures. The first type of film may exhibit strong lateral grain development rather than grain boundary scattering; therefore, dense and thick films were grown in geometrical directions with circular coverage. In contrast, in the second type of film, it is possible that a higher oxygen concentra-tion will suppress the lateral growth and individually form cone seeds with apexes. The nucleation site density becomes much higher than that of the first type due to the effect of the low surface energy at higher oxygen concentrations. Moreover, the low-magnification SEM images in Figure2a–2cillustrate the top views of a portion of the large-area FTO films (O2: 0%, 20%, and 100%, respectively). An examination of

the surface morphology shows how the high-density microstructural

(3)

Figure 3. X-ray diffraction (XRD) patterns of FTO films deposited on glass

substrates with various concentrations of incorporated O.

grains are homogeneously fabricated in an array over a large area with a certain size. The films are very compact, but the textural boundaries are clearly visible, and the result shows a correlation with the AFM and TEM images.

According to nucleation theory, the nucleation density (N) can be expressed by the following equation:

N= Aexp 

−G RT



whereG is the activation energy of nucleation, which is composed of the volume free energy (GV) and the surface free energy (GS).

R, T, and A are the gas constant, growth temperature, and a constant, respectively. Earlier research explained that the change in the volume free energy decreases as the content of O2in the ambient atmosphere

of the deposition process increases. Thus, the activation energy of nucleation becomes lower under O-rich conditions, which enhances the surface mobility for the formation of the initial layer of the FTO film, resulting in smaller grains.13

The XRD pattern in Figure3reveals the evolution of FTO thin films with varied contents of O2/N2carrier gas. All the FTO films are

polycrystalline, indicating a SnO2rutile (tetragonal) structure without

additional peaks for SnO or Sn phases. As-deposited FTO film (O2:

0%) shows the preferential orientation of (110), (200), and (211) at 26.5, 38, and 51.5 degrees, respectively. When the oxygen content was increased, i.e., 80% and 100%, the (200) peak monotonically dominated the patterns, and the (110) peak decreased. The intensity ratio between (110) and (200) decreases from 1.3 to 0.7, corresponding to a change in the carrier gas from pure nitrogen (O2: 0%) to pure

oxygen (O2: 100%). In principle, this particular orientation can be

influenced by the grain evolutionary selection that follows the concept of periodic bond chain theory:1,4the thermodynamic equilibrium form of the SnO2structure typically shows a dominant (110) plane of polar

faces composed of both tin and oxygen atoms. However, Korotkov et al. reported that the addition of a high concentration of oxygen may enhance the presence of the polar halogen-rich region due to the formation of by-product HCl from the spray pyrolysis process,5 leading to quenching of the [110] growth direction and growth rate saturation. In contrast, for a low surface energy at a high oxygen concentration, it was found that (200) deposition could be induced by the formation of a seed layer, which allowed the enhanced nucleation density to change the orientation of the above layer, forming a highly texture surface useful for TCO applications.

In Figure4, the overall transmission of the films was characterized by UV-Vis measurements. The 100% O2 sample shows the highest

percentage in transmission (96% at the wavelength of 550 nm). Fur-thermore, the samples with the lower O2content of 0%, 20%, 50%,

and 80% have lower transmission values at 76%, 82%, 86%, and 89%,

Figure 4. Transmission (T%) spectra of FTO films with varying oxygen

con-centrations, deposited on Corning glass.

respectively. The evaluation results can be explained by a correlation of light scattering and reflection, depending on the size and shape of the grain structures. As shown above, the first result is attributed to an enhanced scattering at the surface and the internal layer and diffraction at structures with sizes close to the wavelength of a photon (Figure1). This behavior is due to the smaller grain growth with a high density in a definite area compared with what occurs for lower O2concentrations. Note that the theoretical consideration of this

be-havior also refers to an increase in waveguide sensitivity. At 80% and 100%, the fluctuation of frequent waves can be clearly observed due to a high surface texture, e.g., the pyramid-like nanostructure together with a large refractive index (n). In contrast, a dome-like structure with a flat grain surface may give rise to the possibility of light reflec-tion, which results in the loss of the surface-normal incidence based on the Fresnel effect.14,15The desired sideways waveguiding behavior is somewhat reduced because the dome with a large incident angle acts as an individual converging lens, only redirecting the light in the forward and backward direction, with less internal diffusion.

Figure5shows the electrical properties of FTO film at room tem-perature as a function of the oxygen content of the carrier gas (0%, 20%, 50%, 80%, and 100%), including the carrier concentration (n), resistivity (ρ), and Hall mobility (μ). The carrier concentration in-creased to 16.12× 1020cm3 at 50% and then leveled off for further

increases in the oxygen content. In a consistent manner, the values of resistivity increased to 18-cm and 70 -cm at 80% and 100%, respectively. As indicated by the above results, increasing the oxygen

Figure 5. FTO film electronic transport properties: carrier concentration (n),

resistivity (ρ), and Hall mobility (μ) as a function of the oxygen concentration (0-100%).

(4)

Figure 6. J-V characteristics of FTO film-based DSSCs. The illumination

intensity of 100 mW cm−2with AM 1.5 and an active area of 0.25 cm2were applied.

content up to 50% is expected to improve the internal oxidization of metallic Sn, which forms well-dispersed SnO2. Instead, at oxygen

concentrations above 50%, the excessive oxygen ions (O2−) generated from O2 molecules at high working temperatures (> 300◦C) might

predominantly disrupt the stoichiometric FTO film by reducing the amount of intrinsic donor defects (and Sni4+). Thus, the O2−ion will

serve as a compensating acceptor defect to trap free electrons and raise the height of the potential barrier along grain boundaries, re-sulting in a lower carrier concentration and a higher resistivity in the matrix. Additionally, the mobilities were as high as 14.45 cm2/V-s at

0% and 16.25 cm2/V-s at 20% because the formation of large grains

allows a smoother surface to develop compared with other films with higher oxygen contents. Thus, the mobility significantly decreased to 1.17 cm2/V-s at 100% because the small grain size and numerous

boundaries between the grains hindered the intergrain carrier mobility. Due to the increased grain boundary and ionized impurity scattering of the carriers, the electrical properties tend to degrade in proportion to the oxygen concentration.

The photovoltaic properties of the textured FTO film-based DSSCs with different amounts of incorporated O2(0%, 20%, and 100%) in the

films are summarized in Figure6and TableI. At 100%, the conversion efficiency (η) obtained is as low as 3.22%; it then increased to 4.98% and 5.25% at O2:20% and O2:0%, respectively. This is mainly due

to the increased short-circuit current density (Jsc) from 6.20 to 9.65

and 10.03 mA/cm2, respectively, although the open-circuit voltage

(Voc) was not changed much (it only varied in the range of 0.750–

0.761 V). Because Vocstrongly depends on recombination or

back-electron transport, the high values with only a small variation could be associated with suppressing those effects, resulting in high electron throughput. In contrast, a large change in Jscis indicated by differences

in grain boundaries (spacings) in each case, which can affect electron transport on the FTO film surface (as described by Hall measurements) and the interfacial pathway (contact area) between the TiO2 layer

and the FTO film. Moreover, due to abundant O2 incorporation, the

low electrical characteristics (ρ and μ) of the 100% O2sample were

assumed to dominate the performance of DSSCs, although the film showed the highest transmission in the UV-Vis analysis (Figure4).

Table I. Summary of FTO film properties used in DSSCs.

O2-content Voc(V) Jsc(mA/cm2) FF (%) H (%)

0% 0.756 10.03 0.692 5.25

20% 0.750 9.65 0.688 4.98

100% 0.761 6.20 0.682 3.22

Figure 7. Representative Nyquist plot of the electrical impedance spectra of

the FTO films measured in the frequency range of 0.1 Hz – 1 MHz under AM 1.5.

Due to the grain boundary effect mentioned above, we further ex-amined the correlation between the cell performance and the internal resistance of the DSSCs via electrochemical impedance spectroscopy (EIS) measured at Vocunder 1.5 AM (Figure7). Regarding the

signif-icant difference in surface morphology (grain growth), the 0%, 20%, and 100% O2samples were used to elucidate the behavior of the

elec-tron interception and the elecelec-tron diffusion to the collecting anode. The Nyquist plots of the samples exhibited four impedance compo-nents of a DSSC, including the resistance element (Rh) of the FTO

layer in the high-frequency region (> 106 Hz) and the other three

impedance elements Z1(ω1: 103-105Hz), Z2(ω2: 1-103Hz), and Z3

(ω2: 0.1-1 Hz) associated with charge transfer processes at the counter

electrode, the interfaces between the FTO/TiO2/dye/electrolyte, and

the Nernstian diffusion within the electrolyte, respectively. A specific feature of the Z2curve for the 100% O2sample increased by 11.5

in width compared with those of the 0% and 20% samples, which is attributed to the electrochemical interfacial resistance between the FTO and the TiO2working electrode. Simultaneously, the Rhof the

100% O2sample also increases by 6, showing good agreement with

the electrical properties of the Hall measurement in Figure4, whereas the other values (Z1and Z3) remain almost identical.

Conclusions

In summary, fluorine-doped tin oxide (FTO) electrodes have been investigated as an alternative anode to indium tin oxide (ITO) in dye-sensitized solar cells (DSSCs). FTO films were grown by ultrasonic spray pyrolysis deposition on Corning glass substrates at 400◦C us-ing O2/N2deposition concentrations of 0, 20, 50, 80, and 100%. The

structural, electrical, and optical properties of the FTO texture films were studied as a function of surface morphology. A significant re-duction in grain size was found to increase the surface roughness, which depended strongly on the increased oxygen concentration. In contrast, the formation of a flattened film is a crucial point that reduces the film resistivity (ρ) to 10−14-cm and enhances the transmittance

(T%) up to 96% at 550 nm. Moreover, electrochemical impedance spectroscopy (EIS) indicates that a flat surface can minimize the dis-continuity of TiO2 particles on an FTO textured film. The key

im-provements in the electrical and optical properties reduced the Auger recombination, so the photovoltaic performance of FTO film-based DSSCs with 0% O2showed a higher conversion efficiency (5.25%)

compared with DSSCs with 100% O2 (3.22%); the numerous grain

boundaries and pyramid-shaped structures made a significant contri-bution to the degradation of the electron transport (diffusion length).

(5)

Acknowledgments

This work was financially supported by the National Science Coun-cil of the Republic of China, Taiwan, under Contract NSC100-3113-E-009-002. We also thank the Industrial Technology Research Institute of Taiwan for technical and financial support.

References

1. I. Volintiru, M. Creatore, B. J. Kniknie, C. I. M. A. Spee, and M. C. M. Van De Sanden,Journal of Applied Physics, 102, 043709 (2007). 2. S. Lee, J. H. Noh, H. S. Han, D. K. Yim, D. H. Kim, J.-K. Lee, J. Y. Kim, H. S. Jung,

and K. S. Hong,The Journal of Physical Chemistry C, 113, 6878 (2009). 3. J. H. Lee, C. Y. Chou, Z. X. Bi, C. F. Tsai, and H. Wang,Nanotechnology, 20, 395704

(2009).

4. C. Y. Kim and D. H. Riu,Thin Solid Films, 519, 3081 (2011).

5. R. Y. Korotkov, P. Ricou, and A. J. E. Farran,Thin Solid Films, 502, 79 (2006). 6. J. Kane and H. P. Schweizer,Journal of the Electrochemical Society, 123, 270 (1976). 7. T. N. Blanton and M. Lelental,Materials Research Bulletin, 29, 537 (1994). 8. K. Y. Rajpure, M. N. Kusumade, M. N. Neumann-Spallart, and C. H. Bhosale,

Ma-terials Chemistry and Physics, 64, 184 (2000).

9. C. Xiang, S.-C. Kung, D. K. Taggart, F. Yang, M. A. Thompson, A. G. G¨uell, Y. Yang, and R. M. Penner,ACS Nano, 2, 1939 (2008).

10. D.-G. Choi, H. K. Yu, S. G. Jang, and S.-M. Yang,Journal of the American Chemical Society, 126, 7019 (2004).

11. H. Kim, G. Kushto, R. Auyeung, and A. Piqu´e,Applied Physics A: Materials Science & Processing, 93, 521 (2008).

12. J. W. Bae, S. W. Lee, and G. Y. Yeom,Journal of the Electrochemical Society, 154, D34 (2007).

13. K. Omura, P. Veluchamy, M. Tsuji, T. Nishio, and M. Murozono,Journal of the Electrochemical Society, 146, 2113 (1999).

14. M. Dashtdar and M. T. Tavassoly,J. Opt. Soc. Am. A, 25, 2509 (2008).

15. H. Fouckhardt, I. Steingoetter, M. Brinkmann, M. Hagemann, H. Zarschizky, and L. Zschiedrich,Advances in OptoElectronics, 2007, 1 (2007).

數據

Figure 1. (a–e) AFM topographic 5 × 5 μm scans and (f–j) cross-sectional
Figure 4. Transmission (T%) spectra of FTO films with varying oxygen con-
Table I. Summary of FTO film properties used in DSSCs.

參考文獻

相關文件

Then, we tested the influence of θ for the rate of convergence of Algorithm 4.1, by using this algorithm with α = 15 and four different θ to solve a test ex- ample generated as

Particularly, combining the numerical results of the two papers, we may obtain such a conclusion that the merit function method based on ϕ p has a better a global convergence and

Then, it is easy to see that there are 9 problems for which the iterative numbers of the algorithm using ψ α,θ,p in the case of θ = 1 and p = 3 are less than the one of the

volume suppressed mass: (TeV) 2 /M P ∼ 10 −4 eV → mm range can be experimentally tested for any number of extra dimensions - Light U(1) gauge bosons: no derivative couplings. =>

Define instead the imaginary.. potential, magnetic field, lattice…) Dirac-BdG Hamiltonian:. with small, and matrix

incapable to extract any quantities from QCD, nor to tackle the most interesting physics, namely, the spontaneously chiral symmetry breaking and the color confinement.. 

• Formation of massive primordial stars as origin of objects in the early universe. • Supernova explosions might be visible to the most

We investigate some properties related to the generalized Newton method for the Fischer-Burmeister (FB) function over second-order cones, which allows us to reformulate the