Occurrence of non-equilibrium orthorhombic SnO₂ phase and its effect in preferentially grown SnO₂ nanowires for CO detection

Abstract

Tin oxide nanowires (NWs) have been successfully deposited on silicon substrates by thermal evaporation of tin powder and used for gas sensing applications. This method has good potential for large-scale fabrication of one dimensional (1D) metal oxide materials. Deposited NWs have been characterized by XRD and FESEM. XRD analysis revealed that SnO₂ NWs exhibit a mixture of orthorhombic and tetragonal phases or single phase tetragonal structure depending on source to substrate distance, and it also plays a vital role on the size and morphology of NWs. The effect of multiphase SnO₂ NWs on CO sensing is studied using an indigenously built sensing setup. The FESEM micrographs reveal that tin oxide NWs of about 50 to 350 nm in diameter can be grown using suitable process parameters like partial O₂ pressure, distance between the substrate and source. CO sensitivity measurements exhibit enhancement in the case of mixed phase SnO₂ NWs as compared to single phase NWs. Texture coefficient is found to have a profound effect on the sensing properties of SnO₂-NWs.

---

1. Introduction

Among semiconductors SnO₂ is the most attractive multifunctional material being used in transparent conducting glass,¹ lithium ion rechargeable batteries,² energy storages,³ piezoelectricity,⁴ field emission displays,⁵ heterogeneous oxidation catalyst,⁶ photoluminescence⁷ and gas sensors⁸ etc. It is also a biodegradable and biocompatible material, suitable for medical and biological applications.⁹ As compared to other II–VI group compound semiconductors, SnO₂ has direct and wide-band gap (E_g = 3.6 eV at room temperature) and high exciton binding energy of 130 meV, which makes the exciton state stable even at room temperature. Band gap tailoring can be done using nanostructures of SnO₂ by a variety synthesis techniques namely R-F magnetron sputtering,¹⁰ metal-organic chemical vapor deposition (MOCVD),¹¹ spray pyrolysis,¹² sol gel,¹³ hydrothermal process,¹⁴ template growth process¹⁵ etc. Bulk SnO₂ sensors have a tetragonal rutile (R-) phase, however, Zn doping¹⁶ is found to stabilize a high temperature and high pressure orthorhombic (O-) phase of SnO₂ along with a tetragonal rutile SnO₂ when the molar ratio of Zn/Sn is around 0.133; these mixed phase thin films exhibited an enhanced sensor response to CO detection.¹⁶ However, whether Zn doping alone or the orthorhombic structure itself is responsible for the enhancement in sensitivity is not known. Although it is difficult to stabilize the O-phase of SnO₂ at normal atmospheric conditions, SnO₂ nanowires (NWs) synthesized on a silicon substrate by thermal evaporation of tin powder in a controlled atmosphere exhibited mixed phases (R + T) or single rutile (R) phase depending on the distance between the substrate and the evaporation source. This method has the potential for large-scale fabrication of one dimensional (1D) metal oxide materials. In this paper, we present XRD, FESEM and sensor measurement results on these NWs in order to clarify the effect of the orthorhombic phase of SnO₂ on CO detection.

2. Experimental

2.1 Synthesis

Fig. 1 shows the experimental set up for the deposition of SnO₂ NWs. It consists of a horizontal tubular furnace and a mass flow controller. Tin (Sn) powder from Sigma Aldrich (10 μm, 99% trace metals basis) was used as source material for the synthesis of SnO₂ NWs. After that Sn powder was loaded in an alumina boat and placed inside the center of the heating zone of the furnace.

Fig. 1 Schematic sketch of tubular furnace.

A silicon wafer with thickness 380 ± 25 μm, resistivity 1–10 Ω cm, orientation 〈100〉 ± 0.50 (WRS materials USA) was chosen as the substrate. These silicon substrates were placed on the downstream side at a distance of 5 cm (sample S1) and 10 cm (sample S2) from the source materials, and then the furnace was evacuated to 10⁻² Torr by using rotary vacuum pump and purged with argon gas at 200 sccm. Then the furnace temperature was raised to growth temperature, i.e. 850 °C. Once the growth temperature was attained high purity oxygen gas with flow rate 50 sccm and the growth process was maintained for 180 minutes. Then the process flow of carrier gases were stopped and the furnace was cooled down to room temperature. These process parameters were tuned such that samples S1 and S2 exhibited a single phase tetragonal microstructure and mixed-phase (R + O) microstructure, respectively. It is found that the morphology of the deposited structure depends on deposition time, substrate temperature, furnace temperature and gas flow rate, however, the existence of O-phase mainly depends on the distance between the source and the substrate position. For sensing measurements, gold pattern electrodes were deposited on the sensing layer and the sensing area, 0.51 cm², was kept constant for both the sensors S1 and S2. The operating temperature was varied from 50 °C to 350 °C in order to find out the optimum working temperature with varying concentrations of CO gas passed to the sensing chamber using a mass flow controller. Change in resistance of the sensor films was measured as a function of time through a computer interfaced EXTECH MultiMaster™ 560A True RMS digital multimeter. Detailed description of the indigenous sensing setup has already been reported elsewhere.¹⁷

2.2 Characterization tools

All the samples were characterized by X-ray diffraction (XRD) using a Bruker AXS, Germany (Model D8 Advanced) diffractometer in the scanning range of 20–80° (2θ) using CuKα radiation (λ = 1.5418 Å). Surface morphology of these deposited nanowires was studied using a Carl Zeiss Field Emission Scanning Electron Microscope (Carl Zeiss AG, Germany). Sensor testing was carried out using the indigenous sensing set up.

3. Results and discussions

3.1 XRD analysis

Fig. 2 displays XRD patterns of S1 and S2 samples. All the Bragg diffraction peaks of S1 can be indexed to the cassiterite (R-rutile) structure of SnO₂ with tetragonal lattice parameters a = b = 4.738(1) Å and c = 3.1855(9) Å consistent with ICDD 85-0423, whereas S2 exhibits a few but distinct peaks of orthorhombic (O-) phase (ICDD 78-1063) along with majority R-phase. The volume fraction of O-phase in S2, is estimated by calculating the ratio (I_O/I_R) of the intensity of the strongest orthorhombic reflection (111) to the intensity of the strongest tetragonal reflection (110),¹⁸ revealed that S2 contains 8% of O-phase and 92% of R-phase. Generally the orthorhombic structured SnO₂ is formed under high pressures at this temperature.¹⁹ Orthorhombic phase formation at moderate temp and low pressure in oxygen deficient atmosphere has also been reported in the literature.^19–21

Fig. 2 XRD pattern of SnO₂ NWs deposited at 850 °C (a) 5 cm away (S1) (b) 10 cm away (S2) from the source material.

Detailed texture studies based on Harris’s analysis were performed in order to calculate the texture coefficient of R-phase using the relationship:²²

where P(hkl) is the texture coefficient of the plane specified by Miller indices (hkl); I(hkl) and I₀(hkl) are the specimen and standard integrated intensities, respectively, for a given peak and n is the number of diffraction peaks. Comparison of the texture coefficients of different planes (Table 1) reveals that for S1, nanowires are highly textured along the (110) plane, which has highest packing fraction of planar density in the rutile structure, useful for sensing, as this plane is effective in electron transfer via oxygen adions, (mostly O⁻).²³ In S2, growth of the (110) plane in R-phase SnO₂ may be restricted due to the presence of O-phase as evidenced by a reduced texture coefficient of the (110) plane in S2 and instead a strong increase in texture along the (200) plane is noticed.¹⁷

Table 1 Texture coefficient of samples S1 and S2

Sample name	Planes (hkl)	Texture coefficient
S1	110	1.926628
	101	1.33498
	200	0.829961
	211	0.474329
	112	0.434102
S2	110	1.007898
	101	1.112282
	200	1.489113
	211	0.667335
	112	0.723372

---

3.2 FESEM analysis

Fig. 3 shows FESEM micrographs of samples S1 and S2. NWs are clearly seen with a large aspect ratio in the range of 40–50.

Fig. 3 FESEM micrograph of S1 and S2 sensors.

The diameters of the NWs are found to be in the range of 50 to 400 nm with some branchings. The thinnest NWs of 50 nm diameter were grown in sample S2 where as the highest diameter NWs in S1 sample was in the range of 300 to 400 nm. These observations suggest that the thickness of deposited NWs decreases with the distance between the substrate and the source. The NWs in both S1 and S2 showed a very good adhesion with the substrate.

3.3 TEM analysis

TEM imaging of S2 which has mixed phases (O + T) is shown in Fig. 4(a). NWs are of ∼50 nm diameter is observed. The SAD pattern of S2 shown in Fig. 4(b) reveals that NWs have a tetragonal structure with ring patterns observed for the (110) and (211) planes. However, an orthorhombic phase could not be detected owing to its low volume fraction (∼8%) determined by XRD analysis. As TEM is a highly localized probe, as compared to XRD, it was not possible to determine the existence of an orthorhombic SnO₂ phase from TEM analysis.

Fig. 4 (a) TEM image of S2 NWs sensors, (b) electron diffraction pattern of the nanowires shown in (a).

3.4 Sensing measurements

Fig. 5(a) and (b) shows the CO sensing response of samples S1 and S2 at optimum operating temperature under the exposure of different gas concentrations. The gas sensor response is defined as:where, R_g is the electrical resistance in the presence of CO in air and R_a is the electrical resistance in air. The sensors made of NWs from S1 & S2 were tested at a temperature range from 50 °C to 300 °C.

Fig. 5 Sensor response of (a) S1 for different concentration of CO at 225 °C (b) S2 for different concentration of CO at 175 °C as a function of time.

S1 shows a maximum response at 225 °C and S2 shows a maximum response at 175 °C for different concentration of CO. Sensing response of the NWs depends upon the exchange rate of gas, density of charge carriers and hence the temperature. It is observed that for S1 sensor the sensing response is increased with temperature up to 225 °C and then decreased beyond 225 °C for all gas concentrations as shown in Fig. 6(c) and for S2 sensor the sensing response is increased with temperature up to 175 °C and decreased beyond 175 °C for all gas concentrations as shown in Fig. 6(d). From Fig. 5(a) and (b) the measured response- (recovery-) times for S1 are 1–2 s (20–55 s) and for S2 are 1–2 s (15–40 s), respectively.

Fig. 6 Sensor response of S1 and S2 as a function of concentration (a and b) & temperature (c and d).

The sensors were capable of detecting CO in the range of 50–150 ppm at temperatures 50 °C to 300 °C and recover quickly once the CO is removed, as shown in Fig. 6 (a) and (b). S1 and S2 NWs exhibit a maximum sensitivity of 102 and 150 for 150 ppm of CO, respectively. These results show that mixed phase NWs with (R + O) structure exhibit higher sensitivity than single (R-) phase NWs. Our results also indicate that a large surface area available due to reduction in diameter of NWs also plays a very important role in the sensitivity of the sensors, which is also supported by literature.^24,25 These results are also consistent with texture analysis. Texture coefficients calculated for different Miller planes are tabulated in Table 1. It reveals that as compared to samples S1, the texture coefficient of the (200) plane increased noticeably in S2, at the cost of a decrease in planar density along the (110) plane. Such a change in texture coefficients was reported earlier with dopants like Fe, In and W.^17,26 Reduced texture coefficient of a dense (110) plane can act as sink of vacancies and thereby improve sensing response in S2.

Cross sensitivity behavior of these sensors S1 and S2 was tested with 150 ppm of CO, C₂H₅OH and NH₃ gas, as shown in Fig. 7. It is confirmed from the Fig. 7 that S2 sensor is highly sensitive to CO gas at its working temperature 175 °C. Several literature reports revealed that pure SnO₂ is selective for CO at a particular operating temperature.^27–29

Fig. 7 Sensors response of S1 and S2 sensors for different target gases at 150 ppm.

The higher sensor response of S2 sample for CO, as seen in Fig. 7, in the present work, might be due to NWs of smaller diameters, mixed phase and huge reduction in texture coefficient along the most densely packed (110) plane.

4. Conclusions

Preferentially grown tin oxide NWs along the (110) and (200) planes are deposited on a silicon substrate at a deposition temperature of 850 °C. The NWs have an average diameter of 50–350 nm. XRD studies reveal that tetragonal or mixed phase (orthorhombic and tetragonal) SnO₂ NWs can be stabilized selectively by changing the distance between source and substrate. TEM studies reveal that 50 nm tin oxide NWs have tetragonal structure. It is found that the texture coefficient, diameter and mixed phase structure of NWs play very important role for sensitivity of the sensors. Sensor response increases with increase in texture coefficient of (200) plane with concurrent decrease in texture coefficient of (110) plane in mixed phase SnO₂ NWs. The mixed phase (orthorhombic and rutile) SnO₂ NWs with lower diameters (50–200 nm) thus, exhibit enhanced sensitivity for CO detection as compared to single phase.

Acknowledgements

Authors are thankful to Vice Chancellor, Defence Institute of Advanced Technology, Girinagar, Pune-411025 (India) for his support. ACA and MK acknowledges the ER-IPR grant no. ERIP/ER/1003883/M/01/908/2012/D (R&D)/1416 through DRDO-DIAT program on nanomaterials.

References

1. N. Kikuchi, E. Kusano, E. Kishio and A. Kinbara, Vacuum, 2002, 6, 365–371.
2. Y. Zhao, Q. Zhou, L. Liu, J. Xu, M. Yan and Z. Jiang, Electrochim. Acta, 2006, 51, 2639.
3. J. Seop Lee and J. Jang, J. Ind. Eng. Chem., 2014, 20, 363–371.
4. S. Su, R. Zuo, X. Wang and L. Li, Mater. Res. Bull., 2010, 45, 124–128.
5. Z. Zhu, Appl. Surf. Sci., 2006, 253, 792–796.
6. S. Pilkenton and D. Raftery, Solid State Nucl. Magn. Reson., 2003, 24, 236–253.
7. L.-W. Chang, M.-W. Huang, C.-T. Li and H. C. Shih, Appl. Surf. Sci., 2013, 279, 167–170.
8. M. Batzill and U. Diebold, Prog. Surf. Sci., 2005, 79, 47–154.
9. L. Wang, Y. Chen, J. Ma, L. Chen, Z. Xu and T. Wang, Sci. Rep., 2013, 3, 3500,  DOI:10.1038/srep03500.
10. W. Yang, S. Yu, Y. Zhang and W. Zhang, Thin Solid Films, 2013, 542, 285–288.
11. R. Y. Korotkov, P. Ricou and A. J. E. Farran, Thin Solid Films, 2006, 502, 79–87.
12. G. Korotcenkov, I. Blinov, M. Ivanov and J. R. Stetter, Sens. Actuators, B, 2007, 120, 679–686.
13. X. Zhong, B. Yang, X. Zhang, J. Jia and G. Yi, Particuology, 2012, 10, 365–370.
14. Y. Guan, D. Wang, X. Zhou, P. Sun, H. Wang, J. Ma and G. Lu, Sens. Actuators, B, 2014, 191, 45–52.
15. C. Zheng, X. Zheng, Z. Hong, X. Ding and M. Wei, Mater. Lett., 2011, 65, 1645–1647.
16. S. Tian, Y. Gao, D. Zeng and C. Xie, J. Am. Ceram. Soc., 2012, 95, 436–442.
17. M. Kumar, A. Kumar and A. C. Abhyankar, Ceram. Int., 2014, 40, 8411–8418.
18. R. A. Spurr and H. Myers, Anal. Chem., 1957, 29, 760.
19. F. J. Lamelas and S. A. Reid, Phys. Rev. B: Condens. Matter Mater. Phys., 1999, 60, 9347–9352.
20. Z. R. Dai, Z. Wei Pan and Z. L. Wang, Adv. Funct. Mater., 2003, 13, 9–24.
21. Z. R. Dai, J. L. Gole, J. D. Stout and Z. L. Wang, J. Phys. Chem. B, 2002, 106, 1274–1279.
22. G. B. Harris, Philos. Mag., 1952, 43, 113.
23. M. A. Maki-Jaskari and T. T. Rantala, Phys. Rev. B: Condens. Matter Mater. Phys., 2002, 65, 1–6.
24. D. Meng, N. M. Shaalan, T. Yamazaki and T. Kikuta, Sens. Actuators, B, 2012, 169, 113–120.
25. N. M. Shaalan, T. Yamazaki and T. Kikuta, Sens. Actuators, B, 2011, 153, 11–16.
26. M. Kumar, A. Kumar and A. C. Abhyankar, ACS Appl. Mater. Interfaces, 2015, 7, 3571–3580.
27. G. Tulzer, S. Baumgartner, E. Brunet, G. C. Mutinati, S. Steinhauer, A. Kock and C. Heitzinger, Procedia Eng., 2012, 47, 809–812.
28. A. M. Gaskov and M. N. Rumyantseva, Inorg. Mater., 2000, 36, 293–301.
29. P. G. Harrison and M. J. Willet, Nature, 1998, 332, 337–339.

---