Effect of the thickness of the MoO3 layers on optical properties of MoO3/Ag/MoO3 multilayer structures

(1)

Effect of the thickness of the MoO3 layers on optical properties of MoO3/Ag/MoO3 multilayer structures

D.-T. Nguyen, S. Vedraine, L. Cattin, P. Torchio, M. Morsli et al.

Citation: J. Appl. Phys. 112, 063505 (2012); doi: 10.1063/1.4751334 View online: http://dx.doi.org/10.1063/1.4751334

View Table of Contents: http://jap.aip.org/resource/1/JAPIAU/v112/i6 Published by the American Institute of Physics.

Additional information on J. Appl. Phys.

Journal Homepage: http://jap.aip.org/

Journal Information: http://jap.aip.org/about/about_the_journal Top downloads: http://jap.aip.org/features/most_downloaded Information for Authors: http://jap.aip.org/authors

(2)

Effect of the thickness of the MoO

₃

layers on optical properties of MoO

₃

/Ag/MoO

₃

multilayer structures

D.-T. Nguyen,¹S. Vedraine,²L. Cattin,¹P. Torchio,²M. Morsli,³F. Flory,²and J. C. Berne`de⁴

1Universite de Nantes, Institut des Materiaux Jean Rouxel (IMN), CNRS-UMR 6502, 2 rue de la Houssinie`re, BP 92208, Nantes F 44000, France

2Aix-Marseille Universite, Institut Materiaux Microelectronique Nanosciences de Provence (IM2NP), CNRS-UMR 7334, Domaine Universitaire de Saint-Jer^ome, Marseille F 13397, France

3Universite de Nantes Faculte des Sciences et des Techniques, 2 rue de la Houssinie`re, BP 92208, Nantes F 44000, France

4Universite de Nantes, Moltech Anjou, CNRS-UMR 6200, 2 rue de la Houssinie`re, BP 92208, Nantes F 44000, France

(Received 10 May 2012; accepted 9 August 2012; published online 18 September 2012)

The electrical and optical properties of MoO3/Ag/MoO3 multilayer structures have been studied using the Ag deposition rate and layer thicknesses as parameters. When the silver film is deposited at 0.20 nm/s rate, the silver layer thickness necessary to achieve the percolation threshold of the resistivityqtowards conductive structures is 10 nm. Below 10 nm, the films are semiconductor and above the films are conductors. In the present work, the variation of the thicknesses of top and bottom MoO₃layers is shown to strongly modify the optical properties of the multilayer structures.

By using a Ag thickness of 10 nm, we demonstrate an increasing of the transmittance of the MoO3/ Ag/MoO₃structures by optimizing the MoO₃layers thicknesses. When the MoO₃bottom layer is 20 nm thick, and the MoO₃top layer is 35 nm, the maximum transmission is 86% at the wavelength of 465 nm, while the averaged transmission in the visible range (350 nm–800 nm) is 70%. The best measured conductivity,r¼1.110⁵(Xcm)¹, corresponds also to this MoO₃(20 nm)/Ag (10 nm)/

MoO₃(35 nm) structure. A good qualitative agreement between the theoretical calculations of the variation of the optical transmittance and reflectance of the MoO3/Ag/MoO3 structures is also highlighted.V^C 2012 American Institute of Physics. [http://dx.doi.org/10.1063/1.4751334]

I. INTRODUCTION

The need of transparent conductive coatings is increasing continually due to numerous applications of these coatings in many optoelectronic devices such as electrochromic, electroluminescent, and photovoltaic solar cells. Indium tin oxide (ITO) is the current commercial choice. However, indium being scarce makes ITO expensive,¹ especially since the development of solar cells based on CIGS (Cu(In, Ga)Se₂) increases the demand. Moreover, an increasing in- terest concerns the use of flexible substrates for optoelectronic devices and ITO films are well known to perform quite poorly under repeated bending.² As a consequence, there is a need for alternative transparent conductive electrode. For organic optoelectronic devices, some additional specific requirements have to be taken into account. Most of the organic layers are very brittle and temperature sensitive.

Therefore, the deposition of transparent conductive oxide (TCO) onto organic layers by sputtering of doped metallic oxide (such as ZnO:Al) is difficult because plasma causes damage to organic layers. Hence, the choice of transparent electrode material and processing conditions are very re- stricted. Thin metal films allow soft deposition conditions and exhibit high conductivity and ductility. However, even by considering the silver which presents the highest conductivity at room temperature (6.1410⁵S cm¹), when thick enough to be conductive exhibits only a limited transmittance in the visible part of the spectrum. On the contrary, TCOs showing a high transmittance have relatively poor

electrical properties when deposited with soft deposition at room temperature. In order to achieve good conductivity and transmittance, a very thin silver film sandwiched between two TCO films offers a reasonable solution.^3–11 Neverthe- less, the low work function of classical TCO results in imperfect work function alignment with hole extraction or injection layers in organic photovoltaic cells (OPVs) or organic light emitting diodes (OLEDs).¹²Therefore, thin films of transition metal oxide (TMO) such as WO₃, MoO₃, V₂O₅, which can be deposited by gentle thermal sublimation were employed and have demonstrated to be very efficient buffer layer between the anode and the hole organic transporting layer. The combination of the high transmission of TMO and the high conductivity of metal by sandwiching a thin silver layer between two TMO layers appears to be a very promis- ing approach.^13–17In this work, the effect of the thicknesses of the MoO₃and Ag layers on optical and electrical properties of MoO₃/Ag/MoO₃multilayer structures is investigated.

These results are confronted to those deduced from modelling by using software based on a finite-difference time-domain (FDTD) method. The influence of the deposition rate of Ag onto the structures properties is also studied.

II. EXPERIMENTAL TECHNIQUES

MoO₃/Ag/MoO₃ multilayer structures were deposited onto glass substrates, under vacuum, in a simple Joule effect evaporation plant. The various films were successively deposited without vacuum break, using two tungsten crucibles:

0021-8979/2012/112(6)/063505/8/$30.00 112, 063505-1 VC2012 American Institute of Physics

(3)

one loaded with MoO₃powder, the other with Ag wires. The substrate temperature during operation was room temperature, while deposition rate and film thickness were measured in situby quartz monitor. The deposition rate of the MoO₃ films was 0.05 nm s¹. The thicknesses of the MoO₃films varied from 1 nm to 50 nm for the bottom layer and from 5 nm to 35 nm for the top layer. When deposited by Joule heating at the deposition rate of 0.2 nm s¹, the optimum Ag layer thickness has been demonstrated to be 10 nm;¹³in the present work, the effect of Ag deposition rate (0.15 and 0.2 nm/s) on the multilayer structure properties is studied.

The physic-chemical, optical and electrical properties of the MoO₃/Ag/MoO₃structures have been studied using both MoO₃ film thicknesses as parameters. X-ray photoelectron spectroscopy (XPS) measurements were performed in order to investigate the surface and depth profiles of the structures.

XPS analyses were performed with a magnesium x-ray source (1253.6 eV) operating at 10 kV and 10 mA. During these measurements, the vacuum was 10⁷Pa and the pass energy for high resolution spectra was 50 eV. In the used apparatus, the C1s peak of the C-C bonds had a well defined position at 284.6 eV which was used as reference to estimate the electrical charge effect. The quantitative studies were based on the determination of each peak after subtraction of s-shaped background and on the sensitivity factors, s, given by the manufacturer: Mo3d, s¼2.5, Ag3d_5/2, s¼2.9, and O1s, s¼0.6. The samples were grounded with a conductive paste to decrease the charge effect. The depth profile of the structures was studied by recording successive XPS spectra obtained after argon ion etching for short periods. Sputtering was carried out at pressures of less than 510⁴ Pa, a 10 mA emission current, and a 3 kV beam energy using an ion gun. With these experimental conditions, all the surface of the sample was sputtered. The fine morphology of the different structures used as anode was observed through scanning electron microscopy (SEM) with a JEOL 7600F. The optical measurements were carried out at room temperature using a Carry spectrophotometer. The optical density was measured at wavelengths from 0.3lm to 1.5lm. The majority carrier type was checked by the hot probe technique. An n-type con- stantan wire was used as the reference sample. The electrical resistivity was determined by measurements in a Van der Pauw configuration.

Numerical results were obtained via the commercial software FDTD SOLUTION^#. The optical constants of MoO₃ material were obtained via the spectroscopic ellipsometer

"SOPRA GES-5."

III. RESULTS AND DISCUSSION A. Experimental results

According to results previously obtained, a 10 nm-thick Ag film and a 35 nm-thick MoO₃top layer^13,18 were used.

First, optical properties of manufactured Glass/MoO₃/Ag/

MoO₃structures were investigated by varying the MoO₃bottom layer thickness between 1 and 50 nm. Spectral transmittance curves of such structures are reported in Figure1. It can be noticed that the maximum transmittance, 86% at k¼465 nm, is achieved for MoO₃ bottom layer thickness

smaller than 30 nm. Moreover, it can be seen a red shift of the transmittance area when the MoO₃bottom layer thickness increases from 1 nm to 30 nm. Such red shift effect has already been put in evidence.¹⁵Figure 2shows (curve with black stars) the averaged transmittance of the Glass/MoO₃ (X nm)/Ag (10 nm)/MoO₃ (35 nm) electrode in the visible spectral domain (350 nm up to 800 nm), when X varies between 1 nm and 50 nm. The optimum transmittance is achieved for a MoO₃bottom layer thickness of 20 nm. For other thicknesses, the transmittance is lower. Consequently, a thickness of 20 nm for the MoO₃bottom layer was retained in order to secondly study the influence of the MoO₃ top layer thickness. The Figure3presents the spectral transmittance measured from Glass/MoO₃ (20 nm)/Ag (10 nm)/

MoO₃ (X nm) electrodes with X varying between 5 and 35 nm. A red shift of the transmittance maximum is observed when the thickness of the MoO₃top layer increases. Figure2 displays (squares) the dependence of the averaged transmittance of such structures according to the variation of the MoO₃top layer thickness. It can be seen that the transmittance of the multilayer significantly increases with the thickness of the MoO₃top layer. A thickness of 35 nm was finally selected for the MoO₃ top layer to obtain the optimal

FIG. 1. Transmittance spectra of Glass/MoO3 (X nm)/Ag (10 nm)/MoO3

(35 nm) structures with X ranging between 1 nm and 50 nm.

FIG. 2. Averaged transmittance (350 nm to 800 nm) as a function of the thickness of the bottom ($) and top (䊏) MoO3layers.

063505-2 Nguyenet al. J. Appl. Phys.112, 063505 (2012)

(4)

electrode. It is worthy to note that the best measured conductivity,r¼1.110⁵(Xcm)¹, corresponds also to the optimal structure MoO₃(20 nm)/Ag (10 nm)/MoO₃(35 nm).

Following optimization of the MoO₃films thickness, the Ag layer was deepen studied by investigating the influence of its deposition rate and thickness on the whole structure conductivity. Therefore, the MoO₃ (20 nm)/Ag (X nm)/

MoO₃(35 nm) structures, when X varies between 5 nm and 16 nm, were carried out for two deposition rates: 0.20 nm/s and 0.15 nm/s. When the silver film is deposited at 0.20 nm/s rate, there is a threshold of the resistivityq from 110⁶to 110⁴ Xcm at 10 nm. Below this 10 nm value for silver thickness, the multilayer films are semiconductors; above this value, the films are conductors with an n-type conductivity. At this same deposition rate of 0.2 nm/s, an optimum transmittance is obtained for Ag film thicknesses varying between 10 nm to 13 nm. Actually, such transmittance values are 20% higher than those obtained from structures with Ag films deposited at 0.15 nm/s rate. Moreover, when the Ag film thickness is 10 nm, and for a deposition rate of 0.15 nm/

s, the resistivity is very sensitive and varies from one sample to another (q¼110³to 10²Xcm). The visualization of the MoO₃/Ag bilayer structure surface by scanning electron microscopy (Figure4) allows to understand the dependence of the conductivity of the MoO₃(20 nm)/Ag (X nm)/MoO₃ (35 nm) structures with the silver deposition rate. For a silver thickness of 10 nm deposited at 0.2 nm/s rate, connected aggregated Ag islands create continuous paths along the silver films (Figure 4(b)). When the same silver thickness is more slowly deposited, i.e., at 0.15 nm/s rate, the silver films are not continuous and present broad intergrains paths empty of material (Figure4(a)). Therefore, the thickness of the silver film which allows achieving the percolation threshold, i.e., the formation of a closed and continuous film, depends on the deposition rate of the silver film. The change of conductivity of the trilayer electrodes is consequently caused by such transition of the silver layer morphology from uncon- nected Ag islands to a continuous Ag film. Therefore, the threshold thickness value corresponds to the percolation of

the metal nanostructures. Below this thickness the films are discontinuous; above this thickness, they are continuous and the structures are conductive. So, if we compare the sheet resistance of the multilayer structures to that of the corresponding silver thin film alone, the values are nearly the same.¹⁹ This shows that the multilayer structures resistivity depends faintly of the oxide resistance. The schematic diagrams of energy band structures of Ag and MoO₃ before and after contact are shown in Figure 5. The Ag work function is 4.4 eV, while it has been recently shown that MoO₃ has a work function of 6.8 eV.²⁰ Therefore, when Ag and MoO₃ are in contact, the Fermi level alignment induces transfer of electrons from Ag to MoO₃. It results an accumulation type contact between Ag and MoO₃layers due to band bending at the contact. So electrons are easily injected from Ag to MoO₃.

In order to check that the transversal conductivity does not constitute a limit of the structure, we have realized sandwich structures Al/MoO₃/Ag/MoO₃/Al (and Ag/MoO₃/Ag/

MoO₃/Ag), the Al (Ag) thin films being used as electrodes to measure the resistance of the sandwich structures. The measured resistances of these structures have been compared to those of sandwiches using similar electrodes, length, and thickness, but without the MoO₃/Ag/MoO₃structure, i.e., Al/

Al (Ag/Ag) structures. As a mater of fact, in the precision

FIG. 3. Transmittance spectra of Glass/MoO3 (20 nm)/Ag (10 nm)/MoO3

(X nm) structures with X ranging between 5 nm and 35 nm.

FIG. 4. Scanning electron microscopy images of the surface morphology of Glass/MoO3(20 nm)/Ag (X nm) structures, with Ag deposited at different deposition rates: (a) 0.15 nm/s, (b) 0.20 nm/s.

(5)

range of the apparatus used, similar values have been achieved. This testifies that the limiting value of these structures is the resistance of the electrodes themselves. The conductivity of our MoO₃/Ag/MoO₃ structures is of the same order of magnitude than that of corresponding metal films.

That result has already been highlighted by Guillen and Her- rero.¹⁹ The thickness of the structure being 6510⁶mm, its resistance is far smaller than that of the electrodes which are at least 1 mm long. This result allows checking that the transversal conductivity of our MoO₃/Ag/MoO₃structures is of the same order of magnitude than the measured planar conductivity. This is in agreement with the schematic diagrams of energy levels of Figure5.

The transmittance spectra of Glass/MoO₃ (20 nm)/Ag (X nm)/MoO₃ (35 nm) structures with X ranging between 5.5 nm and 17.5 nm, when Ag is deposited at 0.2 nm/s rate, are reported in Figure6. It can be seen that the transmittance of the structures increases with the Ag film thickness up to 10 nm, and decreases beyond this thickness value, while the conductivity increases only slightly. A red shift effect of the transmission spectrum, when the silver thickness increases from 5.5 nm up to 10 nm, is also observed. It should be noted that after 7 months storage in room air, the transmittance of the Glass/MoO₃ (20 nm)/Ag (X nm)/MoO₃ (35 nm) structures increases slightly (Figure7), while there is no significant variation of the conductivity.

The effect of the presence, or not, of the different MoO₃ layers were experimentally checked. As shown in Figure8,

the reflectance and transmittance of the following structures were measured: Glass/Ag (10 nm), Glass/Ag (10 nm)/MoO₃ (35 nm), Glass/MoO₃(20 nm)/Ag (10 nm), and Glass/MoO₃ (20 nm)/Ag (10 nm)/MoO₃(35 nm). More precisely, it can be seen in Figure 8(c) that the transmittance of the structures glass/Ag/MoO₃ becomes weak with the increase of wavelength whatever the side of irradiation is. On the other hand, when we introduce a MoO₃layer between the glass substrate and the Ag layer, the transmittance is changed dramatically.

The transmission peak is far broader than that obtained in the absence of this MoO₃film. It is due to the fact that MoO₃ films enhance the light coupling. Furthermore, the outer MoO₃ layer allows managing the surface work function value of the electrode, which is very important in the field of organic device.

Such 3-layer designed structures present an enhanced transmission and a lower reflection as compared to bilayer structures or to single silver layer deposited on glass.

The XPS depth profile of a Glass/MoO₃ (20 nm)/Ag (10 nm)/MoO₃ (35 nm) structure is shown in Figure 9. Ag

FIG. 5. Schematic diagrams of energy levels of Ag and MoO3before (a) and after (b) contact.

FIG. 6. Transmittance spectra of Glass/MoO3 (20 nm)/Ag (X nm)/MoO3

(35 nm) structures with X ranging between 5.5 nm and 17.5 nm.

FIG. 7. Averaged transmittance (350 nm to 800 nm) of Glass/MoO3(20 nm)/

Ag (X nm)/MoO3 (35 nm) structures, with X ranging between 10 nm and 17 nm, as a function of the storage duration in room air (䊏) just after deposition, after one month (䊉) and after 7 months ($).

(6)

depth profile is not symmetric, which is essentially due to the fact that the silver is deposited onto the bottom layer of MoO₃(but a possible contribution of the XPS etching pro- cess cannot be completely excluded). A tail is clearly visible towards the bottom layer (right side of the profile) ended by

around 1 at. % silver in the bulk of the MoO₃bottom layer, by comparing of about 0.5 at. % silver in the MoO₃top layer (left side of the profile). Concerning the possible silver oxide presence, with formation of interfacial Ag-O phase caused by oxygen outdiffusion from MoO₃layers, such reaction has been shown to be difficult²¹by considering the oxide formation enthalpy, which is confirmed here by the XPS analysis.

The bonding energy of the Ag3d_5/2peak of the silver present in the MoO3film is 368.5 eV, which corresponds to metallic silver. The constant position, after etching, of the Ag3d_5/2 peak indicates that no reaction occurs with Ag₂O formation.

At the surface of the structure, from the EFFermi level, position relatively to the band conduction (E_V), it can be confirmed that the structures behaves as n-type structures because EVEF¼3 eV.

Following the optical, electrical, and SEM study, when the silver layer thickness is lower than 10 nm (discontinuous layer), or for low silver deposition rate, the transmittance decreasing can be due to scattering losses. The transmittance maximum is achieved for silver film thickness of 10 nm.

When the Ag layer thickness increases, the transmittance decreases due to higher light reflectance and absorbance. As previously discussed,^13,18,22it can be concluded that the optical properties of the dielectric/metal/dielectric structures can be controlled by different optical effects such as interferen- tial, scattering, and plasmon effects.^3,7,23 Experimental results are hereafter confronted to theoretical study.

B. Numerical results

In order to investigate the theoretical behaviour of the optical properties of our manufactured multilayer electrodes, 3D calculations using a FDTD method were performed. This method is able to rigorously solve the Maxwell’s equations and makes it possible to obtain the electromagnetic field versus time and position. Our simulation zone (Fig.10) presents the schematics of the Glass/MoO₃/Ag/MoO₃design used for calculations by considering infinite size in the x and y direction. Along the z direction, the boundary conditions are per- fectly matched layers (PML) which absorb all waves moving towards the exterior of the simulation zone (back and front

FIG. 8. Experimental reflectance (a) and transmittance (b) versus wavelength for the four following structures: Glass/Ag (10 nm), Glass/Ag (10 nm)/MoO3 (35 nm), Glass/MoO3 (20 nm)/Ag (10 nm), Glass/MoO3

(20 nm)/Ag (10 nm)/MoO3(35 nm), (c) transmittance of Glass/Ag (10 nm)/

MoO3(35 nm), and Glass/MoO3(20 nm)/Ag (10 nm)/MoO3(35 nm), when illuminated from glass and top layer sides.

FIG. 9. XPS depth profile of a glass/MoO3 (20 nm)/Ag (10 nm)/MoO3

(35 nm) structure.

(7)

directions from the illumination source) without reintroduc- ing reflection. The source is a polychromatic plane wave polarised along the x-axis. The selected space-mesh size was 0.4 nm inside the silver layer and up to 11 nm far away from its interfaces. The time-mesh size was 1.310¹⁸ s. The values of the silver optical indices are taken from Palik.²⁴ The values of the (n, k) optical constants of MoO3 were measured by spectroscopic ellipsometry and are shown versus wavelength in Figure11. Each layer is considered to own a plane surface, i.e., without any roughness.

Simulations of direct transmittance data for multilayer structures were performed withFDTD software, starting from the optogeometrical design and refractive indices. Similar modelling was successfully performed for calculating optical properties in thin film coatings with integrated silver nano- particles.^25,26 The transmittance spectrum of the structure Glass/MoO₃(20 nm)/Ag (10 nm)/MoO₃ (35 nm) is first calculated and compared with experimental measurements.

Results are shown in Figure 12. It can be seen a relatively good agreement in the behaviour of the transmittance between experimental and numerical results. Differences in amplitude can be due to the silver film morphology which is experimentally non uniform but constituted of coalescent Ag aggregates. Moreover, we can notice that maximum of trans-

mittance obtained about the wavelength of 450 nm corresponds to the minimum ofkextinction coefficient shown in Figure11.

The role of both MoO₃layers was studied. The air side oxide layer thickness is first considered, by setting the thicknesses of the other oxide layer (glass side) to 20 nm and of the silver layer to 10 nm (as found to be experimentally opti- mized). The calculated transmittance spectra of the structure Glass/MoO3(20 nm)/Ag (10 nm)/MoO3(X nm) are obtained by varying X from 0 to 35 nm. The frequency shift role of the MoO₃top layer (air side) is illustrated in Figure13. We clearly observe a red-shift of the transmittance curves according to the increase of the thickness of this oxide layer.

These numerical results calculated in the 360–800 nm spectral range are well correlated with the experimental results presented in Figure3. If the absolute transmittance levels are different (mainly due to the non-uniformity of the experimental silver thickness and to the probable uncertainty in the theoretical refractive indices used for calculation), they are rather close in relative. It is by example easy to notice that curves move according to the same sequence towards the red spectral range when the thickness of MoO₃ increases. The influence of the other (glass side) MoO₃layer’s thickness is also discussed. The thickness of silver layer is set to 10 nm and that of the oxide layer (air side) to 35 nm, as previously proposed in Sec.III. We consider the structure of type Glass/

FIG. 11. MoO3optical constants dispersion with real part "n" and imaginary part "k" measured by spectroscopic ellipsometry.

FIG. 12. Comparison of measured and calculated transmittance spectra of Glass/MoO3(20 nm)/Ag (10 nm)/MoO3(35 nm) structures.

FIG. 13. Calculated transmittance spectra of Glass/MoO3 (20 nm)/Ag (10 nm)/MoO3(X nm) structures with X ranging between 0 nm and 35 nm.

FIG. 10. Schematics of the Glass/MoO3/Ag/MoO3design used for FDTD calculations.

(8)

MoO₃(X nm)/Ag (10 nm)/MoO₃(35 nm) by varying X from 10 to 50 nm. The reflectance spectra obtained are plotted in Figure14(a), where the effect of this oxide layer acting as an anti-reflection layer is clearly demonstrated for this thickness range. Indeed, the reflectance decreases when the MoO₃ layer thickness (glass side) increases. However, the calculation of the absorbance (Figure14(b)) indicates that the most antireflective layer is also the highest absorbing layer. This dielectric layer has therefore a significant influence on the whole absorption of the multilayer structure. A compromise between the beneficial anti-reflectance effect and the harmful absorptance needs to be found for being applied as efficient transparent electrode. The experimental study has proved that the optimum value of this oxide layer is around 20 nm.

IV. CONCLUSION

The optical properties of MoO₃/Ag/MoO₃ multilayer structures have been studied using the Ag deposition rate and layer thicknesses as parameters. When the silver film is deposited at 0.20 nm/s rate, the Ag film thickness necessary to achieve conductive structures is 10 nm. The passage from discontinuous to continuous film corresponds to the threshold percolation of path formation, which explains the sudden increase of the conductivity values. Above this percolation thickness value, the conductivity increases slightly with the silver layer thickness. However, further increase of Ag thickness also results in a decrease in transmittance. The trade off

between sheet resistance and transmittance of the MoO₃/Ag/

MoO₃structure shows that the optimum structure is obtained with a silver thickness of 10 nm. When the MoO₃ bottom layer is 20 nm thick, and the MoO₃top layer is 35 nm, the maximum transmittance is 86% at the wavelength of 465 nm, while the averaged transmittance in the visible range (350 nm–800 nm) is 70%. A significant improvement of the optical properties of the described multilayer structures are achieved, by comparison to some previous results where the structures from type MoO₃(40 nm)/Ag(10 nm)/

MoO₃(37.5 nm) presented a maximum transmittance of 80%

at the wavelength of 525 nm and an averaged transmittance of 70% in a narrower spectral domain (410 nm–710 nm). The 10 nm-thick silver layer has been shown to be structured.

This kind of metallic structure can lead to plasmon resonance in the MoO₃/Ag/MoO₃structure. Since it was demonstrated that plasmon resonance effect enhances the photocurrent in organic solar cells^27,28such structure can be used as anode in organic solar cells. Numerically, a good agreement between the theoretical calculations of the variation of the optical transmittance of MoO₃/Ag/MoO₃structures and experimental tendencies is proved. Modelling is an efficient numerical tool to help understanding the effect of each layer and can be useful for further optimization of complex structures. In this work, the antireflective effect of the glass side oxide layer and the frequency-shifting effect induced by the air side oxide layer are successfully highlighted. This will allow us to further optimize numerically the MoO₃/Ag/MoO₃structures for use as alternative ITO-free electrodes in thin film organic photovoltaic cells. The red-shift of the transmittance spectrum with the oxide layers thicknesses can also allow to bet- ter match the MoO₃/Ag/MoO₃ transmittance window towards the absorptance spectral range of the active layer in organic solar cells.

ACKNOWLEDGMENTS

OTC-2012-2013 project supported by the Nanorgasol Network of Mission Ressources et Competences Technologi- ques du CNRS FRANCE.

1T. Minami,Thin Solid Films516, 1314 (2008).

2C. D. Willams, R. O. Robles, M. Zhang, S. Li, R. H. Baughman, and A. A.

Zakhidov,Appl. Phys. Lett.93, 183506 (2008).

3K.-H. Choi, H.-J. Nam, J.-A. Jeong, S.-W. Cho, H.-K. Kim, J.-W. Kang, D.-G. Kim, and W.-J. Cho,Appl. Phys. Lett.92, 223302 (2008).

4D. R. Sahu and J.-L. Huang,Thin Solid Films516, 208 (2007).

5D. R. Sahu, S.-Y. Lin, and J.-L. Huang, Thin Solid Films 516, 4728 (2008).

6J. Mayer, S. Hamwi, T. B€ulow, H.-H. Johannes, T. Riedl, and W. Kowal- sky,Appl. Phys. Lett.91, 113506 (2007).

7S. Y. Ryu, J. H. Noh, B. H. Hwang, C. S. Kim, S. J. Jo, J. T. Kim, H. S.

Jeong, C. H. Lee, S. Y. Song, S. H. Choi, and S. Y. Park,Appl. Phys. Lett.

92, 02306 (2008).

8M. Fahland, T. Vogt, W. Schoenberger, and N. Schiller,Thin Solid Films 516, 5777 (2008).

9D. Zhang, P. Wang, R. Murakami, and X. Sang,Appl. Phys. Lett. 96, 233114 (2010).

10M. Philipp, M. Knupfer, B. B€uchner, and H. Gerardin,J. Appl. Phys.109, 063710 (2011).

11T. Winkler, H. Schmidt, H. Fl€ugge, F. Nikolayzik, I. Baumann, S.

Schmale, T. Weimann, P. Hinze, H.-H. Johannes, T. Rabe, S. Hamwi, T.

Riedl, and W. Kowalsky,Org. Electron.12, 1612–1618 (2011).

FIG. 14. Calculated reflectance spectra (a) and absorption spectra (b) of Glass/MoO3(X nm)/Ag (10 nm)/MoO3 (35 nm) structures with X ranging between 10 nm and 50 nm.

(9)

12A. Godoy, L. Cattin, L. Toumi, F. R. Diaz, M. A. del Valle, G. M. Soto, B.

Kouskoussa, M. Morsli, K. Benchouk, A. Khelil, and J. C. Berne`de,Sol.

Energy Mater. Sol. Cells94, 648–654 (2010).

13L. Cattin, M. Morsli, F. Dahou, S. Y. Abe, A. Khelil, and J. C. Berne`de, Thin Solid Films518, 4560–4563 (2010).

14J. Leng, Z. Yu, T. Zhang, Y. Jiang, J. Zhang, and D. Zhang,J. Appl. Phys.

108, 073109 (2010).

15F. Li, S. Ruan, Y. Xu, F. Meng, J. Wang, W. Chen, and L. Shen, Sol.

Energy Mater. Sol. Cells95, 877–880 (2011).

16D. Oh, Y. S. No, S. Y. Kim, W. J. Cho, K. D. Kwack, and T. W. Kim, J. Appl. Phys.509, 2176–2179 (2011).

17L. Shen, Y. Xu, F. Meng, F. Li, S. Ruan, and W. Chen,Org. Electron.12, 1223–1226 (2011).

18L. Cattin, M. Morsli, and J. C. Berne`de, J. Polym. Eng.31, 329–332 (2011).

19C. Guillen and J. Herrero,Thin Solid Films520, 1–17 (2011).

20M. Kr€oger, S. Hamwi, J. Meyer, T. Riedl, W. Kowalsky, and A. Kahn, Appl. Phys. Lett.95, 123301 (2009).

21J.-A. Jeong, Y.-S. Park, and H.-K. Kim,J. Appl. Phys.107, 023111 (2010).

22J.-A. Jeong and H.-K. Kim,Sol. Energy Mater. Sol. Cells93, 1801–1809 (2009).

23L. X. Cai, J. Qiao, J. Mao, and N. Jiang,Thin Solid Films441, 200 (2003).

24E. D. Palik,Handbook of Optical Constants of Solids(Academic, New York, 1985).

25D. Duche, P. Torchio, L. Escoubas, F. Monestier, J. J. Simon, F. Flory, and G. Mathian,Sol. Energy Mater. Sol. Cells93, 1377 (2009).

26S. Vedraine, P. Torchio, D. Duche, F. Flory, J. J. Simon, J. Le Rouzo, and L. Escoubas,Sol. Energy Mater. Sol. Cells95, S57 (2011).

27B. P. Devi, K.-C. Wu, and Z. Pei,Sol. Energy Mater. Sol. Cells95, 2102–

2106 (2011).

28X. Chen, C. Zhao, L. Rothberg, and M.-K. Ng, Appl. Phys. Lett.93, 123302 (2008).

Effect of the thickness of the MoO3 layers on optical properties of MoO3/Ag/MoO3 multilayer structures

Effect of the thickness of the MoO3 layers on optical properties of MoO3/Ag/MoO3 multilayer structures

Related Articles

Additional information on J. Appl. Phys.

Effect of the thickness of the MoO

layers on optical properties of MoO

/Ag/MoO

multilayer structures