Characterization and optical properties of bismuth chalcogenide films prepared by pulsed laser deposition technique

Characterization and optical properties of bismuth chalcogenide films prepared by pulsed laser deposition technique

Materials Science in Semiconductor Processing 57 (2017) 210–219 Contents lists available at ScienceDirect Materials Science in Semiconductor Process...

2MB Sizes 0 Downloads 74 Views

Materials Science in Semiconductor Processing 57 (2017) 210–219

Contents lists available at ScienceDirect

Materials Science in Semiconductor Processing journal homepage: www.elsevier.com/locate/mssp

Characterization and optical properties of bismuth chalcogenide films prepared by pulsed laser deposition technique

crossmark



A.M. Adama,b,c, , E. Lilova, V. Lilovaa, P. Petkova a b c

Physics Department, University of Chemical Technology and Metallurgy, Sofia, Bulgaria Physics Department, Faculty of Science, Sohag University, Egypt College of novel materials and nanotechnology, National University of Science and Technology MISiS, 119049, Moscow, Leninsky Prospekt, 4, Russia

A R T I C L E I N F O

A BS T RAC T

Keywords: Chalcogenides Thin films Optical characteristics

Thin films of Bi-based chalcogenides were prepared by pulsed laser deposition (PLD) technique according to the stoichiometric formula: Bi2(Se1−xTex)3. Their optical properties were studied aiming to find the suitable area of application and the optimum composition amongst the samples under study. X-ray diffraction analysis proved the crystallinity of the deposited samples; in addition, surface roughness and films homogeneity were studied by atomic force microscopy (AFM) confirming the suitability of PLD technique to prepare homogenous and smooth films of the concerned alloys. Absorption coefficient calculations showed higher absorption values of 5×105 and 6×105 cm−1 for Te contents of 90% and 100% in the Bi2(Se1−xTex)3 system respectively. Optical band gap of the concerned films were calculated and found to be in the range of 0.76–1.11 eV, exhibiting comparable values with the previously reported by other authors. Optical studies conformed direct and allowed transitions in all films. Refractive index (n) and dielectric constants (Ɛr) and (Ɛi) were calculated and studied as a function of the wavelength. Values and behavior of (n), (Ɛr) and (Ɛi) indicated strong dependence on the composition and the wavelength range.

1. Introduction

The chalcogenide system Bi2Se3−xTex has long been the subject of a number of investigations in the present time because of the possible applications of these alloys, in particular, in thermoelectric devices, since defects can be obtained by Te doping in Bi2Se3. Using pulsed laser deposition (PLD) to synthesize thin films such as thermoelectric Bi-Chalcogenide materials received only little attention for the preparation of thermoelectric materials. However, the technique allows growing high quality films at lower deposition temperature than any other technique which in turn enables to restore quite easily the stoichiometry of the target [7]. As reported in [8], the thermoelectric performances can be improved through a suitable modification of electron and phonon transport mechanisms which is strongly predicted for low dimensional or nanostructured materials, however, this requires a control of the material structure down to the nanoscale. In [8] they show that pulsed laser deposition provides a good control on the film composition, phase and structure, necessary for a comprehension of the relationship between structure and thermoelectric properties. In other words, the engineering of tailored Bi2Te3 thin films with improved thermoelectric properties can be achieved. The present work aims to prepare thermoelectric thin Bi-chalco-

Bismuth Selenide (Bi2Se3) system is one of the best known topological insulators because it has a gapless single Dirac cone and a bulk band-gap that is larger than other comparable materials. Selenium (Se) vacancies or antisites defects are usually formed when this material is grown, such kind of defects serve as donors and further shift the Fermi energy considerably above the band gap [1]. The compound (Bi2Se3) is quite often used in thermoelectric generators, it is an exceptionally good electrical conductor–as good as gold and is transparent to infrared light as well. On the other hand, Bi2Te3 is one of the most state-of-the art efficient materials working near room temperature; in addition, Bi-Te based alloys are just suited for many applications involving small temperature difference, such as harvesters operating between room and human skin temperatures. Accordingly, lots of applications were benefited from such small power, for instance, pacemakers and blood pressure regulators [2]. Furthermore, Bi2Te3 based alloys find applications in microelectronics, optoelectronics and electromechanical devices, such as photoconductive targets in TV cameras, IR spectroscopy, IR detectors sensors and memory devices [3–6].



Corresponding author at: College of novel materials and nanotechnology, National University of Science and Technology MISiS, 119049, Moscow, Leninsky Prospekt, 4, Russia. E-mail address: [email protected] (A.M. Adam).

http://dx.doi.org/10.1016/j.mssp.2016.10.043 Received 5 July 2016; Received in revised form 18 October 2016; Accepted 21 October 2016 Available online 29 October 2016 1369-8001/ © 2016 Elsevier Ltd. All rights reserved.

Materials Science in Semiconductor Processing 57 (2017) 210–219

A.M. Adam et al.

respectively. Furthermore, it is noteworthy mentioning that the films are not purely single phased of Bi2Se3 or Bi2Te3, since other phases such as BiTe and Bi3Te4 were detected. As in [1], Bi2Te3 thin films were deposited by pulsed laser deposition technique, a polyhedral (PH) structure that was composed of 3D triangular and polygonal crystals was observed, and the PH exhibited a diminished density because of the presence of microvoids between crystals. the PH film possessed Bi4Te5 phase (JCPDS 22-0115), which was associated with a composition of approximately 51.5 at% Te. However, the degree of purity of the material obtained is relatively high. Thus, the recorded XRD patterns do not show peaks due to the various forms of Bi2O3 or bismuth selenium/telluride oxide (Bi2O5Se/Te) that could be also formed during the deposition process. It is worth mentioning here that the structure is significantly dependent on the Te content in the Bi2(Se1−xTex)3 host alloy, furthermore, the binary Bi2Se3 sample shows a completely different crystal structure which may result in different optical properties of Bi2Se3 from the other samples. Regarding the sample containing 50% Bi2Se3-50% Bi2Se3, we cannot observe evidence of any phase transformations in this system as all available evidences refer to the existence of a continuous solid solution range between Bi2Se3 and Bi2Te3 [11]. In other words Bi2(Se1−xTex)3 is a pseudo-binary system, as the distribution coefficient of bismuth is unity [12]. Generally, the XRD peaks are shifted to lower two-theta values indicating that the Te ions are inserted into the Bi2Se3 hosting lattice. As shown in Fig. 2, a notable left shift in the position of the (015) peak (a common peak in the three samples) can be noticed as a result of Te ions incorporation into the Bi2Se3 lattice, indicates the compositional dependence of the crystal structure of the samples under the study. Accordingly, impact of Te/Se ratio's variation in the Bi2(Se1−xTex)3 system on the concerned optical properties is expected to be strong. From X-ray diffraction pattern for maximum intensity peak, the crystallite size (D) of the PLD prepared thin films was calculated using Scherrer's equation:

genide films by PLD technique and investigate their optical parameters at different contents of the chalcogenide elements: Se and Te. Bi2Se3 and Bi2Te3 binary system were studied as well, since tellurium is substituted for selenium in the molecular formula Bi2(Se1−xTex)3, until Bi2Se3is transformed into Bi2Te3. 2. Experimental technique Thin films the Bi2(Se1−xTex)3 system were synthesized using pulsed laser deposition technique (PLD). Target materials are previously prepared bulk crystalline alloys, prepared by melting method for pure elements, 5N pure (99.999%). PLD was performed with a KrF* excimer laser source (λ=248 nm, τFWHM=25 ns) operating at a repetition rate of 2 Hz. The incident laser fluence was 3.3 J/cm2. The targets were prepared from previously synthesized bulk materials. The distance target – substrate was 3 cm. The working pressure was maintained at 4×10−4 Pa. A total number of 1000 pulses were applied on the target for the deposition of each layer. According to studies [9], significant improvements in terms of the properties of the deposited films based on Bi2Te3 were resulted from using PLD as a preparation technique. Interestingly enough; the first attempt to produced Bi2Te3 thin films via pulsed laser deposition was reported in 1996. In this work, it was found that, the deposited film is Te-deficient at positions close to the incoming laser, as the laser interacts with the plasma plume coming off the target [10]. The thickness of the deposited films was around 2500 Å. Thin films have been characterized by the X-ray diffraction technique, energy dispersive analysis and atomic force microscopy; the transmission spectra of the thin films in the spectral range 400–2700 nm were obtained using a double beam ultraviolet–visible–near infrared spectrophotometer V-670. 3. Results and discussion 3.1. X-ray diffraction (XRD) analysis

D=

X-ray diffraction analysis was carried out to study the crystal structure of Bi2(Se1−xTex)3 films at x=0.00, 0.50 and 1.00. Fig. 1 declares the XRD diffractograms of the PLD synthesized films; the three samples of the concerned films are polycrystalline with different positions for the main peak, which reflects different direction for the preferable orientation of atoms constituting each film which in turn refers to the compositional dependence of the structure on the film's constituents. (01,11), (10,10) and (220) were the preferable direction of atoms orientation for the samples X=0.00, X=0.50 and X=1.00

Kλ β cos θ

Since, D is the grain size, λ is the X-ray wavelength, k is a dimensionless parameter known as shape factor, β is the line broadening at half maximum of intensity and θ is the Bragg angle. Very small sizes of grains were induced from Scherrer's equation and listed in Table 1, referring to smooth surface and nano-structure nature of the concerned films.

Fig. 1. XRD patterns of Bi2(Se1−xTex)3 thin films prepared by PLD.

Fig. 2. Compositional dependence of the peak (015) position.

211

Materials Science in Semiconductor Processing 57 (2017) 210–219

A.M. Adam et al.

granular in nature with hill shapes on the upper surface. The height and diameters of the grains of each sample are dependent on the composition itself, therefore, the surface morphology at x=0.50 and 1.00 is highly different from x=0.00. While the two binary compounds X=0.00 and x=1.00 show very similar 2D images, x=0.50 reflects the mixed structure which seems highly denser and completely different from the fishing net structure that exhibited by the two other films. It can be observed also that the surface roughness of the film decreases with addition of more Te which may be attributed to the better suitability of Bi2Te3 for deposition by PLD than Bi2Se3 which is because of the good adhesion of Te with substrate. Surface roughness approximations indicate smooth and uniform films. Aluminum coated silicon cantilevers TAP150-Al-G (Budget Sensors Innovative Solutions Bulgaria Ltd.) with resonant frequency of 150 kHz and spring constant of 5 N/m were used for PLD prepared thin films. The radius of the cantilever's tip is smaller than 10 nm. These cantilevers have a nominal resonance frequency of ca 65 kHz and a typical force constant of 0.5 N/m. The tip nominal radius is less than 8 nm. Roughness analysis was performed by means of Nanoscope 7.30 program. The surface morphology of the thin films was investigated by AFM measurements that were carried out using a Multimode V (Bruker, Santa Barbara, CA). For all measurements, taken in 5 µm scale, scan rate was 1 Hz and the image resolution was 512 lines per scan direction. The observed roughness (confirmed also by metallographic microscope) of the films surface was found to vary between 12 and 30 nm/μm3. In terms of comparison, Bi2Te3 Single Layers were prepared by PLD and tested for thermoelectric applications [13]. The AFM and profilometer films roughness Ra for various energy density E (for TS=360 °C) and various substrate temperatures TS (for E=3 J cm−2) were measured. The smoothest layers (0.4–1.4 nm) were prepared at 200 °C, 3 J cm−2, in argon atmosphere of 13 Pa, for 40 mm targetsubstrate distance. There was the excess of Bi (Bi/Te–0.76) compared to target (Bi/Te–0.67). Layers prepared from Bi2Te3 target were found

Table 1 Grains sizes of Bi2(Se1−xTex)3 thin films at x=0.00, 0.50 and 1.00 calculated from Scherrer's equation. Sample



β

D (nm)

X=0.00 X=0.50 X=1.00

42.799 38.839 49.799

0.02093 0.01605 0.00977

7.4 9.6 16.3

3.2. Energy dispersive X-rays analysis (EDX) A quantitative elemental analysis of the as prepared films was carried out at room temperature with the aid of EDX analysis. EDX scan confirmed the presence of oxygen and carbon in some samples prepared by PLD. It was anticipated that the stoichiometric formula of each compound is nearly ideal, however the presence of Carbon and Oxygen in some samples is the reason behind the deviation from ideality as can be seen in Fig. 3. 3.3. Atomic force microscopy (AFM) Surface characteristics with extremely high magnifications, up to 1,000,000× and with very accurate resolution ranging from 100 µm to less than 1 µm can be achieved using the AFM. A quantitative method to examine the surface morphology and structure is obtained by analyzing the surface roughness using AFM. Fig. 4 Shows 2D and 3D images of Bi2(Se1−xTex)3 films at x=0.00, 0.50 and x=1.00 with image area size (5 µm×5 µm). The deposited films show crystalline nature with small grains. 2D images indicate well covered surface of each film which is could be due to agglomeration of particles, the surface of Bi2Se3 (x=0.00) can be seen as a fishing net structure. In the shown 3D images, intensity strip indicates the height of the surface grains along z-axis. The AFM study reveals that the particles are

Fig. 3. EDX analysis of Bi2(Se1−xTex)3 thin films samples at x=0.00, 0.20, 0.50 and 1.00.

212

Materials Science in Semiconductor Processing 57 (2017) 210–219

A.M. Adam et al.

Fig. 4. a) 2D images by AFM of Bi2(Se1−xTex)3 films prepared by PLD at x=0.00, 0.50 and x=1.00. b) 3D images by AFM of Bi2(Se1−xTex)3 films prepared by PLD at x=0.00, 0.50 and x=1.00.

to be N-type semiconductor. The studied layers shoved semi-metallic behavior i.e. flat or slightly increasing electrical resistivity with increasing temperature. The in-plane electrical resistivity about 1 mΩ cm and the Seebeck coefficient of about −61 μV K−1 were measured at room temperature. Low room temperature ZT about 0.13 was determined by Harman method. From the ZT we estimated thermal conductivity approximately 0.86 W K−1 m−1.

Additionally the complete profile of the three samples is shown below: The 3D images of Bi2(Se1−xTex)3 films at x=0.00, 0.50 and x=1.00 with image area size (5 µm×5µm) are illustrates as shown below:

213

Materials Science in Semiconductor Processing 57 (2017) 210–219

A.M. Adam et al.

Fig. 4. (continued)

3.4. Optical study 3.4.1. Transmission data Transmittance and reflectance spectra of the as-deposited thin films were measured at room temperature as a function of wavelength in a wavelength range between 400 nm and 2700 nm using double beam Jasco spectrophotometer UV–vis–NIR Model V-670. All optical transmission data are normalized to the transmission of bare glass substrate. Variation of transmittance with the wavelength is shown in Fig. 5. All samples exhibited the same behavior of transmittance against wavelength, since the transmissivity increased as wavelength increases for all samples. Generally, the PLD synthesized films showed very low transmittance over the whole wavelength range. However, Bi2Se3 film showed a relatively high transmittance compared with the other films. All samples showed an absorption edge depends only on the type of material. The maximum transmittance was obtained at a wavelength of about 2500 nm, in the IR region i.e. at low energies since there are no appropriate electronic transitions possible that is why transmission is very high in this range. On the other hand, the minimum transmittance was observed at the beginning of the visible light region for all samples. It is notable that for lower wavelengths (≤1000 nm) which is corresponding to high energies, the transmittance is relatively low because most of the light is absorbed in this region. All films are almost opaque, for UV–vis range in particular. In the actual experimental part 1000 and 3000 laser pulses were applied, however, the transmittance of the films prepared at 3000 pulses showed very low values over the whole taken range of wavelengths. Interestingly and worth mentioning that, there is a technological difficulty to develop large size sample of thermoelectric materials such as the concerned Bi2(Se1−xTex)3 system with optimal properties by using PLD. Nevertheless, applicable sizes were obtained in our work as the substrates with the dimensions of 12×12 mm were covered homogenously with Bi2(Se1−xTex)3 layers. AFM observations confirmed that.

Fig. 5. Transmission spectra of Bi2(Se1−xTex)3 films prepared by PLD at x=0.00, 0.10, 0.20, 0.50, 0.80, 0.90 and x=1.00.

214

Materials Science in Semiconductor Processing 57 (2017) 210–219

A.M. Adam et al.

Fig. 8. Compositional dependence of the optical reflectivity at the middle of visible light region.

Fig. 6. Compositional dependence of transmittance at the mid-range of the concerned wavelength range.

observed in the high energy range (low wavelength region). Noteworthy, the observed interference patterns are due to the interference between the wave fronts generated at the air and substrate interface, which defines the sinusoidal behavior of the curves. Furthermore, a similar behavior of reflectance was presented by [16]. In the middle of visible light range the effect of Te content on the amount of the reflected light was tested. Fig. 8 presents the reflectivity variation against Te content. The variation of R alongside with that of T (as in Fig. 6) for the concerned system indicates that both T and R are strongly dependent on the film composition. Nevertheless, a general increase in the value of R as Te content increased can be concluded.

At a selected point of 1650 nm (in the middle of the selected wavelength), the effect of Te addition on the absolute value of the transmissivity was studied. As can be seen in Fig. 6, more Te content leads to more opacity of the prepared materials because of the formation Bi2Te3 with higher opacity than that of the hosting Bi2Se3 crystal. On the other hand, the notable decrease in the transmittance as Te content increased could be attributed to the enhanced surface scattering due to decreasing of surface roughness [14,15].

3.4.2. Reflectance spectra On the other hand, reflectance as a function of wavelength of the PLD deposited films was examined. In contrast to transmittance spectra, reflectance spectra showed maximum values at lower wavelength range as can be noticed in Fig. 7. As a complementary part, reflectance spectra go in the opposite direction of transmittance against wavelength variation. It is obvious that all samples demonstrate reflection maximal values, and also clear that an increase in the Te content enhances the value of the reflection maxima. The ambiguous behavior of reflectivity against wavelength can be attributed to the film's structure. The films under study seem to possess multi-layers structure, as light waves go through the film; they are reflected many times on the film layers which a matter results in appearing maximal and minimal values of reflectivity. In addition, the reflection of the light photons on the film layers was can be clearly

3.4.3. Absorption coefficient and optical band gap To determine how strongly our films absorb light at a given wavelength and how tellurium addition affects the thin films absorption ability, the variation of absorption coefficient (α ) and extinction coefficient (k) was studied as a function of photon energy in addition to the study of their behavior against tellurium content. The absorption coefficient (α) was calculated based on the measured Transmittance (T), reflectance spectra (R) and the film thickness (d), using the following formula [17]:

1 1−R α= Ln ( ) d T The same values of (α) were exactly obtained using the equation:

α=

−Ln(x) d

where,

x=

−1 + 2R−R2 + T2 +

(1 − 2R+R2 − T2)2 +4T2 2T

The dependence of absorption coefficient (α) on the incident photon energy (hν) for the samples under the study is illustrated in Fig. 9. As shown in Fig. 9, (α) exhibits a long tail at low energies. Such a feature, typical for layered type materials was studied and attributed to internal photon scattering on the different layers of material [18]. All samples have a considerable absorption at all wavelength ranges, furthermore, all films have the same spectral pattern, The high absorbance in the UV region is due to the high energy of the UV radiation, since the recorded behavior is consistent and common in all films. High values of (α) support the allowed direct band transition in our films, as well as suggest potential applications of the films under study in the field of optical recording devices manufacturing [19–22]. In the middle point of the concerned wavelength range, the variation of α as a function of Te content was studied. As clear in

Fig. 7. Reflectance spectra of Bi2(Se1−xTex)3 films prepared by PLD at x=0.00, 0.10, 0.20, 0.50, 0.80, 0.90 and x=1.00.

215

Materials Science in Semiconductor Processing 57 (2017) 210–219

A.M. Adam et al.

Fig. 9. Photon energy dependence of the absorption coefficient of PLD prepared Bi2(Se1−xTex)3 thin films at x=0.00, 0.10, 0.20, 0.50, 0.80, 0.90 and x=1.00.

Fig. 11. Tauc plots of PLD prepared Bi2(Se1−xTex)3 thin films with x=0.00, 0.10, 0.20, 0.50, 0.80, 0.90 and x=1.00.

Fig. 10. Absorption coefficient corresponding to Te content.

Fig. 10, α shows a strong compositional dependence. Optical band gaps (Egopt) of the as deposited films were calculated using the “direct” band gap Tauc plots [23], which is obtained from the relation of (αhν)2 vs. hν, where α is the absorption coefficient and hν is the energy of the incident light. Obeying the well known Tauc's equation, the optical band gap has been determined by the extrapolation of the linear regions on energy axis, from the plots of (αhν)2 versus hν. From (Egopt) calculations one can notice that, direct and allowed transition can take place from the valence to the conduction band, confirming that the prepared materials are free from any imperfections. Table 2 indicates the calculated band gap for the prepared samples as interpreted from Fig. 11 which shows the Tauc plots of the PLD prepared Bi2(Se1−xTex)3 thin films. An illustration of the compositional dependence of the calculated Fig. 12. The composition dependence of optical band gap of the PLD synthesized thin films.

Table 2 The obtained values of (Egopt) of the PLD prepared Bi2(Se1−xTex)3 thin films. Film sample

Egopt (eV)

0.00 0.10 0.20 0.50 0.80 0.90 1.00

0.59 0.55 0.54 0.57 0.60 0.58 0.59

optical band gap for PLD prepared Bi2(Se1−xTex)3 thin films can be seen in Fig. 12. The obtained values of (Egopt) in our work are comparable [24] and also in some cases are a little bite higher than the calculated value of the concerned materials as reported in many previous publications. The difference between our results and the results from other workers could be due the polycrystallinity nature of our films, it is known that the optical band gap energy changes with film thickness and crystallinity [25–27]. In our case, we have not single crystal structure of the 216

Materials Science in Semiconductor Processing 57 (2017) 210–219

A.M. Adam et al.

prepared materials as well as a relatively high thickness. Regarding of the increase in Egopt as Te content increases, MossBurstein effect should be taken into consideration to understand the detected behavior. Therefore, an increase in the apparent band gap was observed as a result of Moss-Burstein shift. That is why; in our measurements, the calculated band gap is equal to the actual band gap+Moss-Burstein shift. On the other hand, the thickness of the film is considered as an effective parameter in determination of Egopt values. As in [28], when the thickness of the film increased, the Egopt values increased as well. 3.4.4. Refractive index The complex refractive index of a substance is defined as N=n +ik=Ɛ1/2 , where n is the real refractive index and k, the extinction coefficient. The optical constants n and k are real positive numbers and can be determined by optical measurements. 3.4.5. Extinction coefficient (k) The parameter (k) was calculated with the aid of absorption coefficient (α) values according to the following equation:

k=

Fig. 14. Extinction coefficient as a function of Te amounts in the Bi2(Se1−xTex)3 structure.

coefficient (k) as follows:

αλ 4π

R=

As can be seen in Fig. 13, k follows two opposite trends against wavelength, the first trend is a sharp decrease, whilst, the second is a slight increase as wavelength increased. However, the behavior is typically observed for the factor (α). Due to the direct dependent of k on α, the observed variation of k against wavelength is absolutely logical. The effect of Te/Se ratio on the parameter k in the Bi2(Se1−xTex)3 system is demonstrated in Fig. 14. The figure presents the variation of k as a function of the Te amounts in the Bi2(Se1−xTex)3 matrix at a certain wavelength value. It is clear that k is strongly dependent on the film composition. No general trend can be decided, since there is a strong increase in k as Te added to the Bi2Se3 system, however, the strong increase was followed by a strong decrease and then a slight increase happened thereafter. In addition, the calculated k values in our work lay in the range of previously determined values by many research workers [29]. Mostly, all samples show a similar behavior, the behavior itself can be attributed to the behavior of absorption coefficient (α).

(n − 1)2 ++K2 (n + 1)2 ++K2

Roots of the aforementioned equation for the values of n can be determined from the following relation:

n=

⎡ R+1 2 ⎤1/2 1+R ± ⎢( ) − (1 + K2) ⎥ ⎣ R−1 ⎦ 1−R

According to the above equation, we can use the minus or the negative sign before the square root to get two different values for (n). In our study, more reasonable calculations were obtained using the minus root in the previous equation. The refractive index of thin films was studied by many researchers and have shown variation with different parameters such as deposition techniques, substrate temperature, annealing temperature, degree of oxidation, mixing ratio, density, doping, hydrogen content, films thickness, non-stoichiometry, inhomogeneity, and anisotropy of the deposited films. The refractive index (n) of Bi2(Se1−xTex)3 thin films prepared by PLD is presented in Fig. 15. It can be seen that (n) changes with films composition. It is also observed that n for all films decreases with the increase in wavelength except for Bi2Se3 in which the refractive index increases up to a certain value and then decreases with the increase in wavelength. The decrease in refractive index with wavelength shows the normal dispersion behavior of the Bi2Se3 material. Generally, the exhibited behavior is due to the increase in transmittance and reduction in absorption coefficient with wavelength.

3.4.6. Real refractive index (n) Real refractive index (n) calculations were derived using an equation, linking refractive index (n), reflectance (R) and extinction

Fig. 13. Wavelength dependence of the extinction coefficient (k) of Bi2(Se1−xTex)3 thin films.

Fig. 15. Refractive index of Bi2(Se1−xTex)3 thin films as a function of wavelength.

217

Materials Science in Semiconductor Processing 57 (2017) 210–219

A.M. Adam et al.

Fig. 16. Refractive index of Bi2(Se1−xTex)3 thin films as a function of wavelength. Fig. 18. Variation the real part (εr) of the dielectric function against wavelength for the system Bi2(Se1−xTex)3.

The decrease in refractive index with wavelength shows the normal dispersion behavior of the material. As is clear, Bi2Se3 films has the highest (n) amongst all films over the whole wavelength range, as well as a completely different behavior. Substantial reduction of the refractive index as wavelength increased indicates the decrease of light absorption. Furthermore, refractive index calculations using the positive sign in equation above are shown in Fig. 16. Noteworthy, as presented in Fig. 15 (n) possessed values less than one in some cases showing that the deposited layers in some samples act as dispersive medium, in dispersive medium the wave speed depends on the frequency of the wave, consequently, when the light wave is traveling with a frequency close to the resonance frequency (n) gets smaller than 1. The compositional dependence of n is given in Fig. 17. It is clearly obvious that the Te containing films have significantly small n that the B2Se3 film. The more Te content added in the sample, the small n value gets. This behavior is mainly due to the change in the composition's density as Te content varies [30]. The optical properties of Bi2(Se1−xTex)3 films can be described in terms of the dielectric function ε=εr+iεI, Real part of dielectric constant is calculated using the relation, εr=n2−k2, whilst the imaginary part is calculated using εi=2nk, Worth mentioning that, the real part (εr) and imaginary part (εi) of the dielectric function gives the transparency and absorption, respectively, in the concerned material [31]. In Fig. 18, the wavelength

dependence of (εr) is depicted. Far away from the un-doped sample (Bi2Se3), all samples exhibited the same behavior which is very similar to that of transparency (shown in Fig. 5). As can be seen in, Fig. 18, for all films, εr increase steadily with an increase in the wavelength, attain a peak and subsequently decrease or saturate, except in the Bi2Se3 thin film sample. On the other hand, imaginary part (εi) of the dielectric function is shown in Fig. 19. As a complementary part, (εi) exhibits an opposite behavior to that of (εr). As a general view, εr is higher than εi indicating the low transmittance of the films in the range of wavelengths under study. The variation of both εr and εi with wavelength follows the same trend as that of refractive index and extinction coefficient. 4. Conclusion High quality films of thermoelectric materials based on the chalcogenide Bi2(Se1−xTex)3 system were successfully prepared by the PLD technique; the obtained films were found to be crystalline with homogenous and continuous surface. The morphology of the prepared films, as well as roughness, was measured by AFM. The observed roughness of the films surface was found to vary between 12 and 30 nm/μm3. The effect of Te addition on the structural and optical properties of the concerned system was studied, it was found to be strong. The

Fig. 17. Refractive index variation as a function of Te amount in the Bi2(Se1−xTex)3 thin films system.

Fig. 19. The imaginary part (εi) of the dielectric function as a function of wavelength for Bi2(Se1−xTex)3 thin film samples.

218

Materials Science in Semiconductor Processing 57 (2017) 210–219

A.M. Adam et al.

References

incorporation of Te atoms into the system enhanced the crystallinity of the concerned system significantly, as proven by XRD analysis. The size of grains was seen to be increase as Te added a matter suggesting enhancement of electrical properties due to the decrease of grain boundaries. Optical characteristics of PLD prepared films were studied in this work. All films showed high reflectance and low transmittance in the visible to near UV range of the concerned wavelengths which is further confirmed by the extinction coefficient and dielectric constant analysis. High absorption coefficient was observed for the concerned films (in the range of 107 m−1), which is proper for application in the field of optical recording devices manufacturing. Additionally, all samples exhibited maximum absorption coefficient in the region of visible light. Furthermore, Te addition influenced the absorption coefficient notably. Optical band gap was derived with aid of Tauc’ plots extrapolations. The estimated values were in the range of 0.54–0.60 eV depending on the Te amount inside the Bi2(Se1−xTex)3 system. The optical gap calculations indicated that direct and allowed transition can take place from the valence to the conduction band, confirming that the prepared materials are free from any imperfections, which in turn confirms the suitability of PLD technique to prepare crystalline chalcogenides for different applications, especially for thermoelectric power applications. Imaginary and real part of the materials refractive index were calculated and studied as a function of wavelength, normal dispersion was observed. Remarkably, Te addition increasing led to refractive index's decreasing. The real and imaginary parts of the dielectric function were studied against wavelength. The real part of the dielectric function showed values of as twice as that of the imaginary part. Moreover, the imaginary part was seen as a complimentary for the real part over the concerned wavelength. We believe that a precise control of the structural and morphological properties and a better understanding of the growth mechanisms of thermoelectric films are necessary for a detailed investigation of the relationship between structural and transport properties of polycrystalline materials. Therefore, we can conclude that the pulsed laser deposition technique is suitable for fast and cheap study of wide scale of parameters having influence on final properties of the product material.

[1] Phuoc Huu Le, Kaung Hsiung Wu, Chih Wei Luo, Jihperng Leu, Thin Solid Films 534 (2013) 659–665. [2] A.Kamal, H.Abu Bakr, Z.Wang, H.El Samman, P.Fiorini, S.Sedky, in: Proceedings of the Multifunctional Nanocomposites & Nanomaterials International Conference & Exhibition, MN2008-47020, Sharm El-Sheikh, Egypt, January 11– 13, 2008, p. 33. [3] D. Arivuoli, F.D. Gnanam, P. Ramasamy, Growth and microhardness studies of chalcogneides of arsenic, antimony and bismuth, J. Mater. Sci. Lett. 7 (1988) 711–713. [4] N. Sakai, T. Kajiwara, K. Takemura, S. Minomura, Y. Fujii, Pressure-induced phase transition in Sb2Te3, Solid State Commun. 40 (1981) 1045–1047. [5] M. Stölzer, M. Stordeur, H. Sobotta, V. Riede, IR transmission investigations of (Bi1−xSbx)2Te3 single crystals, Phys. Status Solidi (b) 138 (1986) 259–266. [6] L. Jansa, P. Lošťák, J. Šrámková, J. Horák, The change of the electric conductivity type in crystals of Bi2−xInxTe3 solid solutions, J. Mater. Sci. 27 (1992) 6062–6066. [7] D.B. Chrisey, G.K. Kubler, Pulsed Laser Deposition of Thin Films, Wiley, New York, 1994. [8] A. Bailini, F. Donati, M. Zamboni, V. Russo, M. Passoni, C.S. Casari, A. Li Bassi, C.E. Bottani, Appl. Surf. Sci. 254 (2007) 1249–1254. [9] Jane E. Cornett, Oded Rabin, Solid-State Electron. 101 (2014) 106–115. [10] A. Dauscher, A. Thomy, H. Scherrer, Pulsed laser deposition of Bi2Te3 thin films, Thin Solid Films 280 (1996) 61–66. [11] N. Fuschillo, J.N. Blerly, F.J. Donahoe, J. Phys. Chem. Solids 8 (1959) 430–433. [12] J.P.Mchugh, W.A.Tiller, Westinghouse Scientific Paper No. 431FD271-Pl, 1958. [13] M. Jelinek, R. Zeipl, J. Vanis, T. Kocourek, J. Remsa, J. Navratil, J. Lorinčík, J. Mater. Sci. Chem. Eng. 4 (2016) 52–64. [14] M.I. Abd-Elrahman, Rasha M. Khafagy, Shiamaa A. Zaki, M.M. Hafiz, J. Alloy. Compd. 571 (2013) 118–122. [15] A. Kathalingam, Mi-Ra Kim, Yeon-Sik Chae, Jin-Koo Rhee, S. Thanikaikarasan, T. Mahalingam, J. Alloy. Compd. 505 (2010) 758. [16] Austine A. Mulama, Julius M. Mwabora, Andrew O. Oduor, Cosmas Muiva, Afr. Rev. Phys. (2014) (9-0006). [17] Sodky H. Mohamed, André Anders, Surf. Coat. Technol. 201 (2006) 2977–2983. [18] G. Domingo, R.S. Itoga, C.R. Kannewurf, Phys. Rev. 143 (1966) 536. [19] T.E. Manjulavalli, T. Balasubramanian, D. Nataraj, Chalcogenide Lett. 5 (11) (2005) 297–302. [20] Saji Augustine, S. Ampili, Jeung Ku Kang, Elizabeth Matha, Mater. Res. (2005) 1314–1325. [21] C. Ghosh, B.P. Varma, Thin Solid Films 60 (1979) 61. [22] B.B. Nayak, H.N. Acharya, T.K. Choudhari, G.B. Mitra, Thin Solid Films 92 (1982) 309. [23] J. Tauc, R. Grigorovici, A. Vancu, Optical properties and electronic structure of amorphous germanium, Phys. Status Solidi 15 (2) (1966) 627–637. [24] Saji Augustine, S. Ampili, Jeung Ku Kang, Elizabeth Mathai, Mater. Res. Bull. 40 (2005) 1314–1325. [25] C.B. Satterhwaite, R.W. Ure, Electrical and thermal properties of Bi2Te3, J. Phys. Rev. 108 (1957) 1164–1170. [26] S. Shigetomi, S. Mori, Electrical properties of Bi2Te3, J. Phys. Soc. Jpn. 11 (1956) 915. [27] A. Goswami, S.S. Koli, Semiconducting properties of Bi2Te3 and Bi2Se3 films, Indian. J. Pure Appl. Phys. 7 (1969) 166. [28] M.M. Wakkad, E.Kh Shokr, S.H. Mohamed, J. Non-Cryst. Solids 265 (2000) 157–166. [29] Bhakti Jariwala, Dimple shah, N.M. Ravindra, Thin Solid Films 589 (2015) 396–402. [30] A.M. Adam, E. Lilov, P. Petkov, Mater. Sci. Semicond. Process. 52 (2016) 1–7. [31] N.M. Shah, C.J. Panchal, V.A. Kheraj, J.R. Ray, M.S. Desai, Growth, structural and optical properties of copper indium diselenide thin films deposited by thermal evaporation method, Sol. Energy 83 (2009) 753–760.

Acknowledgements The authors gratefully acknowledge the financial support of the Ministry of Education and Science of the Russian Federation in the framework of Increase Competitiveness Program of NUST «MISiS» (№ К4-2016-2018), implemented by a governmental decree dated 16th of March 2013, N 211. Authors are also grateful to the Bulgarian academy of science for cooperation.

219