NOC-17396; No of Pages 5 Journal of Non-Crystalline Solids xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
Journal of Non-Crystalline Solids journal homepage: www.elsevier.com/ locate/ jnoncrysol
Optical properties of As2S3 layers deposited from solutions Vlastimil Matějec a,⁎, Jitka Pedliková a,b, Ivo Barton a, Jiri Zavadil a, Petr Kostka b a b
Institute of Photonics and Electronics, Academy of Sciences of the Czech Republic, Chaberská 57, 182 51 Prague 8, Czech Republic Institute of Rock Structure and Mechanics AS CR, V. Holesovickach 41, 182 09 Prague 8, Czech Republic
a r t i c l e
i n f o
Article history: Received 23 January 2015 Received in revised form 10 April 2015 Accepted 12 April 2015 Available online xxxx Keywords: Optical properties; Layers; Arsenic sulfide; Solution in n-propylamine or ethylenediamine; Dip and spin coating
a b s t r a c t This paper presents results of the preparation and optical characterization of As2S3 layers by using spin or dipcoating technique and As2S3 solutions in ethylenediamine and n-propylamine as solvents. These solutions were prepared from arsenic sulfide powders obtained from laboratory-prepared bulk glasses. Ethylenediamine and n-propylamine solutions with the same As2S3 concentration of 0.7 mol/l were used for the layer preparation. Layers were prepared by both the dip-coating and spin-coating techniques with velocity of 0.2–30 cm/min and 2000 rpm for 90 s, respectively. Substrates used in these experiments were microscopic slides (refractive index of about 1.49). All applied As2S3 layers were dried in vacuum for 60 min and then heat-treated at 180 °C for 60 s in nitrogen atmosphere. Prepared As2S3 layers were characterized by transmission spectroscopy in the wavelength range of 200–100 nm. Thicknesses of applied layers in the range of 200–1500 nm and their refractive indices ranging from 1.7 to 2.3 at 800 nm were estimated from interference bands in measured transmission spectra using Swanepoel's method. Dispersion curves of the refractive index are shown in the paper. Optical band gaps of prepared layers in the range of 2.15–2.35 eV were determined from measured transmission spectra by using Tauc's extrapolation. © 2015 Elsevier B.V. All rights reserved.
1. Introduction In the last ten years great attention has been paid to the development of different types of chalcogenide materials. It is well known that such materials exhibit high transparency in the mid-IR spectral region, high refractive indices (from 2.0 to 3.5) and a high nonlinear refractive index n2 on a level of 3–5.10−18 m/W. Due to such characteristics they are very good candidates for applications in light amplifiers, optical regenerators, or for broadband radiation sources in the mid-IR based on supercontinuum generation [1]. In such applications chalcogenide optical fibers have usually been employed [2]. However, thin films of chalcogenide glasses have also been investigated for the development of optical waveguides, gratings optical memories etc. Physical methods such as thermal evaporation, RF sputtering, or pulsed laser deposition have successfully been used for application of chalcogenide thin films based usually on arsenic sulfide As2S3 [3–5]. Patterning processes such as photolithography and dry or wet etching [6,7], lift-off [8], laser writing [6,9], etc. have been used for fabricating ridge or embedded channel chalcogenide waveguides. However, the performance of such physical methods can be decreased by their limited growth rates, induced internal stresses, delamination of applied films, etc. [10]. In addition to physical methods for the preparation of chalcogenide layers solution-based techniques have also been investigated [11–15].
⁎ Corresponding author.
Such techniques employ solutions of As2S3 in n-propylamine or butylamine. Ethylenediamine has also been added to such solutions in order to improve the As2S3 solubility and annealing of applied films. As2S3 solutions have usually been applied by spin-coating technique. Techniques, such as solution-casting and molding which make possible the fabrication of raised-strip As2S3 waveguides, have been used as well [13,14]. There are some advantages of such solution-based techniques over the physical methods such as low-temperature processing, larger range of layer thicknesses, and more homogenous films. This paper deals with the fabrication of As2S3 layers on glass slides by using arsenic sulfide solutions in n-propylamine and in ethylenediamine. In addition to the spin-coating technique the dip-coating method has also been used. Prepared layers have been characterized by transmission spectroscopy. Refractive indices, thicknesses and optical bandgaps have been determined from measured transmittance spectra. 2. Experimental Layers of As2S3 were applied from their solutions in n-propylamine (PA) or ehylenediamine (ED). At first powders of As2S3 were obtained from bulk glasses prepared in the Institute of Rock Structure and Mechanics by melting technique from high purity elements [16]. For the powder preparation pieces of bulk glass were crushed and finely ground. The powder was sieved by a sieve (100 mesh) and dissolved in the solvent in a sealed glass container under stirring with a magnetic stirrer. The dissolution was carried out in a nitrogen-filled box. As2S3 solutions with the same concentration of 0.7 mol/l in PA and ED
http://dx.doi.org/10.1016/j.jnoncrysol.2015.04.027 0022-3093/© 2015 Elsevier B.V. All rights reserved.
Please cite this article as: V. Matějec, et al., Optical properties of As2S3 layers deposited from solutions, J. Non-Cryst. Solids (2015), http:// dx.doi.org/10.1016/j.jnoncrysol.2015.04.027
2
V. Matějec et al. / Journal of Non-Crystalline Solids xxx (2015) xxx–xxx
respectively were prepared. The solutions were used one day after their preparation. They were filtered with a 0.2 μm filter just before their application on substrates. The solutions of As2S3 were applied onto glass slides (a refractive index of about 148–1.49) by dip-coating method using withdrawing velocities in a range 20 to 300 mm/min and by spin-coating technique at 2000 rpm for 90 s. Applied layers were baked under vacuum at 60 °C for 60 min and then at 180 °C for 60 s in nitrogen. Vacuum heat treatment at 160 °C up to 3 h was also tested. Glass slides used for the layer preparation were cleaned by immersing in a solution of hydrofluoric and nitric acids for few minutes, then rinsed in distilled water and dried in flowing nitrogen. Transmission spectra of glass slides with layers were measured on a spectrometer Perkin Elmer Lambda35 in the wavelength range of 200– 1100 nm with air as a reference. The transmission spectrum of each glass substrate was also measured before the layer application. Each spectrum was registered with a wavelength step of 1 nm. The measurement was repeated three times which allowed us to determine transmittances with a standard deviation better than 6.10− 4 in the wavelength range of 300–1100 nm. The spectrometer is equipped with two built-in light sources, namely with a tungsten lamp for visible and near infrared regions and deuterium flash lamp for the UV region. A controlling software UV WinLab supplied by the producer was used with the spectrometer. Transmission spectra measured by the spectrometer were used without any further corrections for the data treatment (see Section 3).
A photo of a layer of As2S3 applied onto a glass slide and taken by a digital camera is shown in Fig. 1. Examples of measured transmission spectra of prepared layers are shown in Figs. 2–4 with examples of envelope curves TMax, TMin connecting interference maxima and minima. Transmittances were measured with a standard deviation of about 0.06%. In the graphs measured transmittance data were connected by straight lines using the software Origin 7.5. Examples of measured transmission spectra of glass slides used as substrates for layer applications are shown in Fig. 5. Refractive indices of prepared layers were estimated from measured transmission spectra by using Swanepoel's method [17,18] and the following equations rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi Nþ
N2 −n2s
N ¼ 2ns
ns ¼
3. Results and data treatment
n¼
Fig. 2. Transmission spectra of As2S3 layers prepared by dip-coating method from the solution in n-propylamine at different withdrawing velocities; experimental points are connected by straight lines; dashed lines are envelope curves to the spectrum of a layer prepared at v = 200 mm/min.
TMax −TMin n2s þ 1 þ 2 TMax TMin
1 þ Ts
sffiffiffiffiffiffiffiffiffiffiffiffiffiffi 1 −1: T 2s
ð2Þ
ð3Þ
In Eqs. (1)–(3), n and ns are refractive indices of the layer and glass substrate, respectively, Ts is the transmittance measured on the substrate (see Fig. 5), TMax and TMin are the transmittance maximum and corresponding transmittance minimum at a certain wavelength. Such maximum and minimum values were calculated from the envelope curves obtained by fitting a proper function into measured interference maxima and minima. In this paper exponential decay functions of first order (T0 + T1e−λ/λ0) were used for fitting experimental transmittance maxima and minima. Examples of refractive-index dispersion curves for prepared layers and determined from Eqs. (1)–(3) are shown in Figs. 6 and 7. Particular values of the refractive index were calculated with a standard deviation of about 0.01–0.03 refractive index units. Examples of standard deviations are also shown on the plots. The curves were fitted by a dispersion
ð1Þ
Fig. 1. A photo of As2S3 layer (dimensions 25 × 40 mm) prepared from the PA solution by dip-coating technique (v = 200 mm/min).
Fig. 3. Transmission spectra of As2S3 layers prepared by dip-coating method from the solution in ethylenediamine at different withdrawing velocities; experimental points are connected with straight lines; dashed lines are envelope curves for the transmission spectra of a layer prepared with a velocity v = 100 mm/min.
Please cite this article as: V. Matějec, et al., Optical properties of As2S3 layers deposited from solutions, J. Non-Cryst. Solids (2015), http:// dx.doi.org/10.1016/j.jnoncrysol.2015.04.027
V. Matějec et al. / Journal of Non-Crystalline Solids xxx (2015) xxx–xxx
Fig. 4. Transmission spectra of As2S3 layers applied by spin-coating method from the solutions in ethylenediamine and n-propylamine; experimental points are connected with straight lines.
3
Fig. 6. Dispersion of the refractive index of As2S3 layers prepared by dip-coating method; solid lines were obtained by connecting calculated data with straight lines in the software Origin 7.5.
function [17]: B C n¼Aþ 2þ 4: λ λ
ð4Þ
In Eq. (4) A, B and C are characteristic constants for each layer, λ is the wavelength. The software Origin 7.5 and a linear or polynomial function with Y = n and X = 1/λ2 was used in the fitting. Calculated refractive-index dispersion curves were fitted with the dispersion function with correlation coefficients higher than 0.993. Thicknesses of prepared layers were calculated from Eq. (5) [18]. d¼
λ1 λ2 : 2ðλ1 n2 −λ2 n1 Þ
ð5Þ
In Eq. (5) n1 and n2 are refractive indices at two adjacent minima (maxima) λ1 and λ2 respectively. Absorption coefficients of prepared layers were calculated from Eqs. (6)–(8) [18]. α¼−
x¼
ln x d
F−
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi 3 F 2 − n2 −1 n2 −n4s ðn−1Þ3 n−n2s
ð6Þ
F¼
8n2 ns 2TMax TMin T¼ : T TMax þ TMin
ð8Þ
In the region of strong absorption, i.e. without interference bands, TMax = TMin = experimental transmittance value. Optical bandgap values Eg of prepared layers were determined from calculated absorption coefficients α by using Tauc's extrapolation pffiffiffiffiffiffiffiffiffiffi αhν ¼ B hν−Eg :
ð9Þ
In Eq. (9) ν is the frequency, h denotes the Planck' s constant and B is a parameter characteristic for each layer. Examples of plots (αhν)0.5 ~ hν from which bandgap values were determined are in Fig. 8. In these plots the dotted and dashed lines representing Tauc' s extrapolation were obtained by using the software Origin 7.5 and by linear fitting the linear parts of re-calculated experimental data. Values of the refractive index at 800 nm as well as thicknesses and bandgap values of As2S3 layers prepared at different processing conditions are shown in Table 1. Standard deviations of the parameters determined from experimental transmittance spectra were calculated from results obtained at three
ð7Þ
Fig. 5. Examples of transmission spectra of three glass slide substrates used for the application of layers; experimental points are connected with straight lines.
Fig. 7. Dispersion of the refractive index of As2S3 layers prepared by spin-coating method; solid lines were obtained by connecting calculated data with straight lines; dashed curves show the extrapolations of fits of the calculated data with the function in Eq. (4).
Please cite this article as: V. Matějec, et al., Optical properties of As2S3 layers deposited from solutions, J. Non-Cryst. Solids (2015), http:// dx.doi.org/10.1016/j.jnoncrysol.2015.04.027
4
V. Matějec et al. / Journal of Non-Crystalline Solids xxx (2015) xxx–xxx
Fig. 8. Tauc' s extrapolation plots (the dotted and dashed straight lines) used for the bandgap determination on As2S3 films applied from the n-propylamine solution; the solid lines were obtained by connecting re-calculated experimental data with straight lines.
subsequent measurements on the same sample of a particular arsenic sulfide layer.
4. Discussion Arsenic sulfide layers are usually applied from solutions of As2S3 by using the spin-coating technique [11,12]. In this paper the dip-coating method has also been used and its performance compared with that of spin-coating technique. As one can see from Fig. 1 homogeneous layers can be prepared by using this method and the solution of As2S3 in n-propylamine. Similar homogeneous films were also prepared by dip-coating method from the solution of As2S3 in ethylenediamine. Both solutions and the spin coating method have also allowed us to prepare good-homogeneity layers which is in agreement with already published papers [12]. The cleaning of glass substrates has been found to influence, to a large extent, the quality of applied arsenic sulfide layers especially in the case of dip-coated layers. Immersing glass substrate in a solution of hydrofluoric and nitric acids, rinsing in water and drying in a nitrogen stream have been found as the suitable approach for such cleaning. Measured transmission spectra in Figs. 2–4 reflect differences in prepared layers. In order to achieve a good reproducibility at measured spectra, each prepared arsenic sulfide solution was used for three days only and then a fresh one was used. By comparing Figs. 2 and 3 one can see that optical losses of dip-coated layers are higher for those prepared from the ethylenediamine solution. The same conclusion follows from comparing curves in Fig. 4. The curves in Figs. 2–4 exhibit several interference bands and thus they can be used for the determination of the refractive index by the Swanepoel method.
Table 1 Refractive index, thickness and bandgap of As2S3 layers vacuum baked at 60 °C for 1 h. Solvent
Method
n-Propylamine
Dip coating
Velocity n (800 nm)
100 200 300 Spin coating 2000 Ethylenediamine Dip coating 20 50 100 150 Spin coating 2000
1.98 ± 0.02 2.01 ± 0.01 1.98 ± 0.02 2.25 ± 0.01 1.73 ± 0.03 1.81 ± 0.02 1.70 ± 0.03 1.85 ± 0.02 2.01 ± 0.01
d [μm]
Eg [eV]
0.42 ± 0.05 0.45 ± 0.04 0.55 ± 0.04 0.27 ± 0.06 1.0 ± 0.1 1.2 ± 0.1 1.2 ± 0.1 1.5 ± 0.1 0.36 ± 0.05
2.38 ± 0.02 2.36 ± 0.02 2.38 ± 0.02 2.24 ± 0.03 2.34 ± 0.02 2.33 ± 0.02 2.32 ± 0.02 2.29 ± 0.02 2.18 ± 0.03
Note: Velocity units: dip coating [mm/min], spin coating [rpm].
The Swanepoel method needs envelope curves TMax(λ) and TMin(λ) connecting maxima and minima of interference bands. Usually a parabolic function has been used for such purpose [17]. In this paper exponential decay functions of first order were used. From dashed curves in Figs. 2 and 3 one can see that they can be used for connecting extreme points of transmittance curves with reasonable accuracy. Refractive indices have been calculated from the envelope curves and transmittance spectra of glass slides (see Fig. 5). Examples of results are shown in Figs. 6 and 7, and Tables 1 and 2. As such an approach allowed us to determine refractive indices in the range of about 650 to1100 nm experimental data were fitted by Cauchy relation (see Eq. (4) and dashed curves in Fig. 7) which were used for calculations of refractive indices at shorter wavelengths. From Figs. 6 and 7 and Table 1 one can see that refractive indices of the layer at 800 nm are between 1.7 and 2.3 at 800 nm. Higher values were determined for spin-coated layers (compare Figs. 6 and 7) and for layers applied from the solution in n-propylamine These values are usually lower than values from 2.1 to 2.5 reported elsewhere for spincoated layers [12]. In that paper a solution of As2S3 in n-propylamine with a concentration of 0.8 mol/l was used as well as different drying temperature and times. However, a value of 2.25 for the spin-coated layer from PA solution (see Table 1) is comparable with that reported elsewhere [12]. From Figs. 6 and 7 and Table 1 it is evident that lower refractive indices were determined for layers prepared from ED solution which can be related to lower vapor pressures of ethylenediamine and to slower evaporation. Thus, higher amounts of ethylenediamine can be kept in layers prepared from ED solutions than in those applied from PA solutions. The refractive index of As2S3 layers can be correlated with the layer thickness (see Table 1). A higher thickness is related to a lower refractive index provided that other processing parameters are the same. A set of experiments has been carried out to test the effects of higher drying temperature and longer drying times. Some of the results are in Table 2. They show that by increasing the drying temperature to 160 °C at the vacuum treatment a refractive-index value is increased to 2.07. However, it has been found that longer drying times decreased the refractive index which doesn't agree with the opposite trend reported elsewhere [12]. On the other hand it has been found that longer drying times of solution-based As2S3 layers increased their porosity [15]. And the porosity as well as the rest of the solvents decrease the refractive index of such layers. Thus, drying conditions for arsenic sulfide layers coated from solutions have to be determined for each solvent, the concentration and coating procedure. Values of photonic bandgap shown in Table 1 are comparable with those reported elsewhere [12]. One can conclude that they depend on the preparation technique (compare results for dip- and spin-coating techniques in Table 1) and a little on the solvent used. 5. Conclusions Homogeneous arsenic sulfide layers can be prepared on glass slides from solutions of arsenic sulfide in n-propylamine and ethylenediamine by using spin-coating and dip-coating techniques. Refractive indices of such layers determined from their measured transmittance spectra by using the Swanepoel method range from 1.7 to 2.3 at 800 nm. Higher Table 2 Effect of thermal treatment on the refractive index and thickness of As2S3 layers. Vacuum treatment Temperature [°C]
Duration [h]
60 160 160 150
1 1 2 3
n (800 nm)
d [μm]
2.01 ± 0.01 2.07 ± 0.01 1.90 ± 0.02 1.88 ± 0.02
0.42 ± 0.05 0.33 ± 0.05 0.30 ± 0.05 0.25 ± 0.06
Please cite this article as: V. Matějec, et al., Optical properties of As2S3 layers deposited from solutions, J. Non-Cryst. Solids (2015), http:// dx.doi.org/10.1016/j.jnoncrysol.2015.04.027
V. Matějec et al. / Journal of Non-Crystalline Solids xxx (2015) xxx–xxx
refractive indices were determined for layers prepared from the npropylamine solutions and for layers with smaller thicknesses on the level of 200–400 nm. Optical bandgaps in the range of 2.18–2.38 determined from measured transmission spectra are dependent on the solvent and coating technique used. Future experiments will deal with layers based on other chalcogenide materials. Acknowledgment This work was supported by the Czech Science Foundation (contract no. P106/12/2384). References [1] F. Smektala, C. Quemard, L. Leneindre, J. Lucas, A. Barhlemy, C. De Angelis, Chalcogenide glasses with large non-linear refractive indices, J. Non-Cryst. Solids 239 (1998) 139–142. [2] S. Cui, R. Chahal, C. Boussard-Pledel, V. Nayabal, J.-L. Doualan, J. Troles, J. Lucas, B. Bureau, From selenium- to tellurium-based glass optical fibers for infrared spectroscopies, Molecules 18 (2013) 5373–5388. [3] A.B. Seddon, Chalcogenide glasses: a review of their preparation, properties and applications, J. Non-Cryst. Solids 184 (1995) 33–50. [4] Benjamin J. Eggleton, Barry Luther-Davies, Kathleen Richardson, Chalcogenide photonics, Nat. Photonics 5 (2011) 141–148. [5] V. Balan, C. Vigreux, A. Pradel, Chalcogenide thin films deposited by radiofrequency sputtering, J. Optoelectron. Adv. Mater. 6 (2004) 875–882. [6] J.-F. Viens, C. Meneghini, A. Villeneuve, T.V. Galstian, É.J. Knystautas, M.A. Duguay, K.A. Richardson, T. Cardinal, Fabrication and characterization of integrated optical waveguides in sulfide chalcogenide glasses, J. Lightwave Technol. 17 (1999) 1184–1191.
5
[7] S.J. Madden, D.-Y. Choi, D.A. Bulla, A.V. Rode, B. Luther-Davies, V.G. Ta'eed, M.D. Pelusi, B.J. Eggleton, Long, low loss etched As2S3 chalcogenide waveguides for alloptical signal regeneration, Opt. Express 15 (2007) 14414–14421. [8] J. Hu, V. Tarasov, N. Carlie, N.-N. Feng, L. Petit, A. Agarwal, K. Richardson, L. Kimerling, Si-CMOS compatible lift-off fabrication of low-loss planar chalcogenide waveguides, Opt. Express 15 (2007) 11798–11807. [9] N. Hô, M.C. Phillips, H. Qiao, P.J. Allen, K. Krishnaswami, B.J. Riley, T.L. Myers, N.C. Anheier Jr., Single-mode low-loss chalcogenide glass waveguides for the midinfrared, Opt. Lett. 31 (2006) 1860–1862. [10] A. Zakery, Y. Ruan, A.V. Rode, M. Samoc, B.L. Davies, Low-loss waveguides in ultrafast laserdeposited As2S3 chalcogenide films, J. Opt. Soc. Am. B 20 (2003) 1844–1852. [11] G.C. Chern, I. Lauks, Spin-coated amorphous chalcogenide films, J. Appl. Phys. 53 (1982) 6979–6982. [12] S. Song, J. Dua, C.B. Arnold, Influence of annealing conditions on the optical and structural properties of spin-coated As2S3 chalcogenide glass thin films, Opt. Express 18 (2010) 5472–5480. [13] C. Tsay, Y. Zha, C.B. Arnold, Solution-processed chalcogenide glass for integrated single-mode mid-infrared waveguides, Opt. Express 18 (2010) 26744–26753. [14] C. Tsay, E. Mujagic, C.K. Madsen, C.F. Gmachl, C.B. Arnold, Mid-infrared characterization of solution-processed As2S3 chalcogenide glass waveguides, Opt. Express 18 (2010) 15523–15530. [15] Y. Zha, S. Fingerman, S.-J. Cantrell, C.-B. Arnold, Pore formation and removal in solution-processed amorphous arsenic sulfide films, J. Non-Cryst. Solids 369 (2013) 11–16. [16] D. Lezal, J. Pedlıkova, J. Zavadil, P. Kostka, M. Poulain, Preparation and characterization of sulfide, selenide and telluride glasses, J. Non-Cryst. Solids 326&327 (2003) 47–52. [17] D. Poelman, P.-F. Smet, Methods for the determination of the optical constants of thin films from single transmission measurements: a critical review, J. Phys. D. Appl. Phys. 36 (2003) 1850–1857. [18] I. Ay, H. Tolunay, Optical transmission measurements on glow-discharge amorphous silicon nitride films, Turk. J. Phys. 25 (2001) 215–222.
Please cite this article as: V. Matějec, et al., Optical properties of As2S3 layers deposited from solutions, J. Non-Cryst. Solids (2015), http:// dx.doi.org/10.1016/j.jnoncrysol.2015.04.027