Impedance and dielectric analysis of polycrystalline undoped and Fe doped SnS2

Materials Science in Semiconductor Processing 59 (2017) 18–22

Contents lists available at ScienceDirect

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

crossmark

Impedance and dielectric analysis of polycrystalline undoped and Fe doped SnS2 ⁎

Areej S. Alqarni , B.O. Alsobhi, A.A. Elabbar, O.A. Yassin1 Department of Physics, Faculty of Science, Taibah University, B.O. Box 30002, Prince Naif road, Almadinah Almunawarah, Saudi Arabia

A R T I C L E I N F O

A BS T RAC T

Keywords: Tin disulfide Impedance analysis Dielectric constant Electromagnetic absorption

Polycrystalline Sn1−xFexS2 samples with (x=0, 0.125, 0.250 and 0.375) have been prepared by the molten salt solid state reaction method. The X-ray diffraction (XRD) shows that all the samples crystallize in the hexagonal structure, with P-3m1 space group in preferred orientation of (011). The electrical properties have been studied by complex impedance spectroscopy over the frequency range (20 Hz up to 1 MHz) at room temperature. The Nyquist plot for all samples have been fitted using ZMAN software. The impedance analysis showed that all samples exhibit both bulk and grain boundary contributions and it was found that by increasing the iron content, the resistance increases, but, the dielectric constant and dielectric loss tangent decrease which leads to decrease in conduction. The absorption coefficient (α) has been calculated from the complex dielectric constant. Interestingly, there was a significant correlation between the electromagnetic wave absorption and the reduction in the peak intensity of the XRD patterns indicating that when the iron content increases the sample seems to be a good absorber of electromagnetic waves.

1. Introduction Considerable attention has been paid to the layered semiconductor compound tin disulfide which has many remarkable applications [1–8]. For example, it has been used as high-power lithium ion batteries [1], optoelectronic devices [2], gas sensors and chemical sensors [3,4], photoluminescence material [5], photoconductive [6], pigment [7] as well as anode materials [8]. Moreover, its properties can be tuned by doping. For instance, its photocatalytic activity can be enhanced when zinc or cerium elements are inserted [8,9]. Recently, it has been reported that doping SnS2 with vanadium (V) or tungsten (W) elements makes it a promising parent material candidate for intermediate band solar cells IBSC applications [10]. Fe doped SnS2 was also reported as a good candidate for intermediate band solar cells because iron doping enhanced its optical absorption [11]. On the other hand, many published works have shown that SnS2 can easily be prepared and widely investigated. Useful examples of SnS2 synthesis methods are hydrothermal [12], microwave-assisted [13], spray pyrolysis [10] and molten salt solid state reaction [14]. Studies on the electrical properties of SnS2 are very rare. In spite of that, there is little published data on its electrical properties. In 1996 the dielectric properties of SnS2, single crystals have for the first time

been studied [15] and there is no much work have been reported on its dielectric and impedance properties. Electrochemical impedance spectroscopy (EIS) measures the dielectric properties of a material as a function of frequency. It can be used to elucidate the relaxation mechanism(s) of the transport properties of the materials under study. The response of the material to the applied frequency is usually represented in terms of the most commonly used Bode and Nyquist plots. For a Debye single relaxation process the Nyquist plot is a half circle with its center falls on the x-axis. However, when there is more than one relaxation process the shape in the plot is composed of many joined circles or not well defined half-circle (quasi half circle). The last situation takes place when the relaxation frequencies are close to each other which can be separated by fitting the data to a certain equivalent circuit model. In this work the purpose is to prepare Sn1−xFexS2 (x=0, 0.125, 0.250 and 0.375) powder samples and study their impedance and dielectric properties to inspect the effect of the Fe-doping on relaxation process and the absorption of the electromagnetic radiation.

⁎

Corresponding author. E-mail addresses: [email protected] (A.S. Alqarni), [email protected] (B.O. Alsobhi), [email protected] (A.A. Elabbar), [email protected], [email protected] (O.A. Yassin). 1 Permanent address: Department of Physics, Faculty of Science, Al-Neelain University, Khartoum 11121, Sudan. http://dx.doi.org/10.1016/j.mssp.2016.11.033 Received 30 May 2016; Received in revised form 17 October 2016; Accepted 19 November 2016 1369-8001/ © 2016 Elsevier Ltd. All rights reserved.

Materials Science in Semiconductor Processing 59 (2017) 18–22

A.S. Alqarni et al.

2. Experimental details 2.1. Synthesis of samples The polycrystalline samples of Sn1−xFexS2 (x=0, 0.125, 0.250 and 0.375) were prepared by the molten salt solid state reaction method. All reagents were purchased from Sigma-Aldrich. First, thiourea (N2H4CS), tin(II) chloride dihydrate SnCl2·2H2O and iron(II) chloride hexahydrate FeCl2·6H2O were mixed and ground by an agate mortar. Next, the powders were placed inside a tubular furnace at 280 °C using alumina boats. After heating for 3 h in a flowing N2 gas, they are left to cool naturally. After that, powders were grounded again in an agate mortar to become soft and then washed so many times using distilled water to remove any possible impurity such as (NH4)2SnCl6 which is easily soluble in water. Lastly, they were dried at 90 °C to obtain the samples. 2.2. Characterization The structural properties have been studied using a Shimadzu 6100 X-ray diffractometer. The were collected in the 2θ range from 10◦ to 80◦ with a step size of 0.02. To study the impedance and dielectric properties of the samples, impedance spectroscopy (GW INSTEK LCR-8105 G) was used at room temperature in the frequency range of 20 Hz to 1MHz with the ac-amplitude of 100 mV for a series circuit (X-R). Before the impedance measurement, the powder is pressed into disk pellets of 13 mm diameter and 2 mm thickness using a hydraulic press. It was densified by annealing at 280 °C for 2 h in a flowing of N2 gas. Each pellet sample is loaded between two circular copper electrodes with similar dimensions. The impedance data were acquired using a computer with automated data collection software.

Fig. 1. : The X-ray diffraction patterns of Sn1−xFexS2 (x=0, 0.125, 0.250 and 0.375) with the standard PDF Card - 01–089-2028. The red solid lines are the profile fitting obtained using Fullprof software.

3. Results and discussion 3.1. Structural properties The single phase formation has been confirmed using X-ray diffraction (XRD). Fig. 1 shows the XRD patterns of Sn1−xFexS2 (x=0, 0.125, 0.250 and 0.375). In this figure, the points and the solid line represent the experimental data and the profile fitting created by the FullProf software for a model with a hexagonal lattice and space group P 3m1, respectively. All Bragg's reflections have been indexed and no additional peaks from any possible impurities, like oxides, cubic FeS2 or (NH4)2SnCl6 have been detected. Beside the FullProf result, some matching works [3,16] of the observed peaks and the standard PDF card-01–089-2028 also confirmed that the doping with Fe ions does not change the hexagonal crystal structure of tin disulfide. This result has been previously reported by us [11] in a different preferred orientation of (001) and that may be due to gradually heating during the preparation steps. It has been observed that the intensity of the peaks become weaker when the dopant concentration increase. Fig. 2 shows the variation of the intensity of the detected beam with dopant concentration x. A possible explanation for this is that when the iron content increases the sample seems to absorb a larger amount of the incident X-ray.

Fig. 2. : Shows the variation of the peaks intensity with the dopant concentration x.

3.2. Impedance and dielectric analysis

Fig. 3. : Nyquist plots for Sn1−xFexS2 (x=0, 0.125, 0.250 and 0.375) measured at room temperature, The solid lines are the fits obtained using ZMAN software, the arrow shows the direction of frequency increment. The upper and lower insets show a zoomed view of the high frequency region and equivalent circuit used for modeling the data, respectively.

The electrical properties of semiconductors and insulators may be represented in terms of the complex impedance (Z*), complex dielectric constant (ɛ*), and dielectric loss (tan δ). These quantities are related to each other [17] as follows

Z*

=R+iX =

Rs +i

1 =Rs ωCs

−

1 = iωCs

Z′−

iZ′ ′

ɛ* ɛ′−

(1) 19

=

ɛ′−

i ɛ′ ′=[iωC Z *]−1

i ɛ′ ′=[iωC (Z′ −

−1

iZ′ ′)]

(2) (3)

Materials Science in Semiconductor Processing 59 (2017) 18–22

A.S. Alqarni et al.

Table 1 The values of the bulk resistance R1, grain boundary resistance R2 and the CPE components values obtained from the fitting of impedance data by ZMAN software for Sn1−xFexS2 (x=0, 0.125, 0.250 and 0.375). x

R1

R2

CPE1

CPE2

CPE3

(Ohm cm2)

(Ohm cm2)

Q1 (F cm2)

n1

Q2 (F cm2)

n2

Q3 (F cm2)

n3

0

0.396×106

0.297×106

3.09×10−9

0.91

1.74×10−9

0.72

1.49×10−16

0.82

0.125

1.157×106

1.085×106

2.02×10−9

0.79

9.75×10−10

0.71

1.09×10−7

0.05

0.250

4.762×106

1.513×106

9.06×10−10

0.70

2.07×10−9

0.68

4.56×10−8

0.03

0.375

23.311×106

2.463×106

1.78×10−10

0.70

1.12×10−11

0.92

8.79×10−11

0.80

Fig. 4. : Bode plots for Sn1−xFexS2 (x=0, 0.125, 0.250 and 0.375). the red circles show the phase, the blue squares show the impedance and the dashed lines are to better see the existence of two process.

Fig. 5. : The variation of the dielectric constant with frequency of Sn1−xFexS2 (x=0, 0.125, 0.250 and 0.375) at room temperature.

ɛ′−

i ɛ′ ′=[ωC Z′ ′ + iωC Z′]−1

Fig. 6. : The variation of dielectric loss with frequency of Sn1−xFexS2 (x=0, 0.125, 0.250 and 0.375) at room temperature.

(4)

ɛ′−

20

i ɛ′ ′=

1 ωC Z′ + iωC Z′ ′

×

ωC Z′ ′ − iωC Z′ ωC Z′ ′ − iωC Z′

(5)

Materials Science in Semiconductor Processing 59 (2017) 18–22

A.S. Alqarni et al.

are more than one relaxation process. Each relaxation process exhibits as an RQ element in the model (equivalent circuit). Therefore, the model reflects the probability of the existence of two overlapping semicircles in our samples. This is in agreement with the Bode plots shown in Fig. 4, where only one constant time is identified (|Z| vs. frequency shows straight points) which means overlapping process, however, the Bode phase plots show two time constants (the dashed lines are to better see the existence of two process) [22]. According to brick-layer model, the element R1 and Q1 in our equivalent circuit can be ascribed to the bulk property of the material while R2 and Q2 can be attributed to the grain boundary effect. In some materials, there is also a contribution from capacitive layers at the grain boundary interfaces (parallel to the current and perpendicular to the electrodes) and that appeared in our case as Q3. The electrodes response has not been observed in our samples. The values of the bulk and grain boundary resistance R1 and R2 respectively, reported in Table 1, found to be increasing when the dopant concentration is increased. Figs. 5 and 6 show the variation of the dielectric constant and dielectric loss with frequency of Sn1−xFexS2 (x=0, 0.125, 0.250 and 0.375) at room temperature. It is evident from the figures, that in the measured frequency range, the value of the dielectric constant falls below 2.5 kHz and then becomes slowly varying with frequency. At the lower frequency, higher value of ε' can be attributed to the space-charge polarization. Similar behavior has been reported for SnS2 single crystal [15]. It can be seen that the dielectric loss tanδ decreases with increasing frequency. According to S. A. Masti et al. [23], higher dielectric loss in the low frequency range, may be due to smaller resistivity and the smaller value of dielectric constant at lower frequency range is due to larger value of resistivity. This is explained on the basis of conduction mechanism, wherein the addition of iron content decreases the dielectric constant and increases the resistance which leads to a decrement in conduction.

Fig. 7. : The variation of the electromagnetic absorption coefficient with frequency of Sn1−xFexS2 (x=0, 0.125, 0.250 and 0.375) at room temperature.

ɛ′−

ɛ′−

i ɛ′ ′=

ωC Z′ ′ − iωC Z′ 2

(ωC Z′ ′) +

ωC Z′ ′

i ɛ′ ′=

′ 2

(ωC Z′ ) +

tan δ =

(ωC Z′)2

ɛ′ ′ ɛ′

(ωC Z′)2

(6)

−i

ωC Z′ 2

(ωC Z′ ′) +

(ωC Z′)2

(7)

(8)

where Z' and Z'' are the real and imaginary components of the impedance, respectively, ɛ' and ɛ'' are the real and imaginary components of the dielectric constant, respectively. i= −1 , ω = 2πf the angular frequency, R is resistance, X is reactance of the capacitor, C is A capacitance, C = ɛ d is the geometric capacitance, ɛ˳ is vacuum permittivity, A is the area of the electrode surface, d is the pellet thickness and the subscripts s is for series combinations of the circuit elements. The complex impedance of the present samples has been measured by an LCR meter and analyzed using ZMAN software, which can be used to find the best equivalent circuit model. The results obtained from this model provided a clearer picture of the electrical behavior of these samples. Fig. 3 shows the complex impedance plot between Z'' and Z′ when the frequency varies (the Nyquist plot) for Sn1−xFexS2 (x=0, 0.125, 0.250 and 0.375) measured at room temperature. Based on the polycrystalline structure of our samples and on the brick-layer model, we may expect two semicircles on the complex impedance spectra Z′' vs Z′ due to the contribution from grain (bulk) at high frequency and grain boundary effect at low frequency [18,19]. However, the experimental results showed that all samples had only single semicircular which means that either one-relaxation process in the impedance spectra or two similar relaxation process that caused an overlapping on the semicircles. These semicircles are off-centered (their center is not on real x-axis), and therefore they reflect a non Debye-type relaxation process for all samples [20]. The best fits are obtained from ZMAN software when we use an equivalent circuit that include serial two parallel RQ elements connected in parallel with constant phase element Q3. Q is the constant phase element (CPE) coefficient expressed in Farad unit and given by the following formula:

CPE =

1 Q (iω)n

3.3. The electromagnetic absorption The extinction coefficient (κ ) which is directly related to the absorption coefficient (α) can easily be calculated from the real and imaginary parts of the dielectric constant [24] as follows

κ=

α=

2 1 1 1 [− ε′+(ε′2 + ε′ ′ )2 ] 2 2

4πκ λ

(10) (11)

Fig. 7 shows the variation of the electromagnetic absorption coefficient with frequency of Sn1−xFexS2 (x=0, 0.125, 0.250 and 0.375). From the trend of these curves, we may expect that there is an increase in electromagnetic wave absorption due to increasing amount of the Fe dopant concentration, in agreement of the pervious result from the XRD analysis. It is worth noting that these samples previously reported as candidate for IBSC because of the increment of optical absorption [11]. 4. Conclusions In this work, undoped and iron doped tin disulfides have been successfully synthesized by the molten salt solid state reaction method. The structural and impedance properties have been characterized by XRD and LCR meter, respectively. The results of the structural study indicated that doping with Fe ions does not alter the hexagonal crystal structure of tin disulfide. Moreover, a higher electromagnetic absorption was observed when Fe doping is increased. Studies of impedance properties indicate that the materials exhibit: (a) both bulk and grain boundary contributions, which could not be directly confirmed from Nyquist plots. (b) a decrease in the dielectric constant at low frequency, which is attributed to the space charge polarization [15]. (c) an increase

(9)

where n is the CPE exponent, if n=1, the CPE acts as an ideal capacitor [21]. The CPE components values obtained from the fitting of impedance data by ZMAN software for Sn1−xFexS2 (x=0, 0.125, 0.250 and 0.375) are reported in Table 1. The equivalent circuit formed by (Q1|R1-Q2|R2) |Q3, is shown in the lower inset of Fig. 3. According to brick-layer model it shows that there 21

Materials Science in Semiconductor Processing 59 (2017) 18–22

A.S. Alqarni et al.

[7] W. Anaf, O. Schalm, K. Janssens, K. De Wael, Dye. Pigment. 113 (2015) 409–415. [8] G. Kiruthigaa, C. Manoharan, C. Raju, S. Dhanapandian, V. Thanikachalam, Mater. Sci. Semicond. Process. 26 (2014) 533–539. [9] G. Kiruthigaa, C. Manoharan, M. Bououdina, S. Ramalingam, C. Raju, Solid State Sci. 44 (2015) 32–38. [10] O. a. Yassin, a.a. Abdelaziz, a.Y. Jaber, Mater. Sci. Semicond. Process. 38 (2015) 81–86. [11] A.S. Alqarni, O.A. Yassin, Mater. Sci. Semicond. Process. 42 (2016) 390–395. [12] H. Geng, Y. Su, H. Wei, M. Xu, L. Wei, Z. Yang, Y. Zhang, Mater. Lett. 111 (2013) 204–207. [13] S. Park, J. Park, R. Selvaraj, Y. Kim, J. Ind. Eng. Chem. 31 (2015) 269–275. [14] H. Xiao, Y.C. Zhang, H. Bai, Mater. Lett. 63 (2009) 809–811. [15] S.K.K. Arora, D.H.H. Patel, M.K.K. Agarwal, Mater. Chem. Phys. 45 (1996) 63–65. [16] Y.C. Zhang, Z.N. Du, S.Y. Li, M. Zhang, Appl. Catal. B Environ. 95 (2010) 153–159. [17] E. Barsoukov, J.R. Macdonald, Impedance Spectrosc. Theory Exp. Appl. (2005). [18] N.K. Mohanty, S.K. Satpathy, B. Behera, P. Nayak, R.N.P. Choudhary, J. Adv. Ceram. 1 (2012) 221–226. [19] A. Shukla, R.N.P. Choudhary, A.K. Thakur, J. Mater. Sci. Mater. Electron. 20 (2009) 745–755. [20] A. Abassi, N. Kallel, S. Kallel, K. Khirouni, O. Peña, J. Magn. Magn. Mater. 401 (2016) 853–859. [21] J. Huang, Z. Li, B.Y. Liaw, J. Zhang, J. Power Sources 309 (2016) 82–98. [22] T. Lopes, L. Andrade, H.A. Ribeiro, A. Mendes, Int. J. Hydrog. Energy 35 (2010) 11601–11608. [23] S.A. Masti, A.K. Sharma, P.N. Vasambekar, Adv. Appl. Sci. Res. 4 (2013) 335–339. [24] M.Fox, M.Fox, 2001.

in the iron content led to a reduction to the dielectric constant while increasing the resistivity, which may cause decrement in conduction. (d) higher electromagnetic absorption when the iron concentration is increased. Acknowledgments The authors extremely thank Mr. Awad A. Abdelaziz, Mr. Abdalla Y. Jaber and Ms. Nahlah A. Al-Ahmadi for the technical assistance provided while making the experimental part of this work. References [1] W. Shi, B. Lu, Electrochim. Acta 133 (2014) 247–253. [2] G. Domingo, R.S. Itoga, C.R. Kannewurf, Phys. Rev. 143 (1966) 536–541. [3] W. Shi, L. Huo, H. Wang, H. Zhang, J. Yang, P. Wei, Nanotechnology 17 (2006) 2918. [4] A. Umar, M.S. Akhtar, G.N. Dar, M. Abaker, a. Al-Hajry, S. Baskoutas, Talanta 114 (2013) 183–190. [5] C. Wang, Chem. Phys. Lett. 357 (2002) 371–375. [6] T. Shibata, Y. Muranushi, T. Miura, T. Kishi, J. Phys. Chem. Solids 51 (1990) 1297–1300.

22