CONTROLLABLE GROWTH OF WELL-ALIGNED ZnO NANORODS ON QCM BY A NOVEL WET CHEMICAL ROUTE FOR HIGH PERFORMANCE NH3 GAS SENSOR

Vu Anh Minh, Truong An Hien, Bui Van Sang, Do Van Lam, Nguyen Van Toan, Vu Ngoc

Hung, Nguyen Van Hieu, and Nguyen Van Quy International Training Institute for Materials Science, Hanoi University of Technology

No. 1, Dai Co Viet Road, Hanoi, Vietnam; Email: quy@itims.edu.vn

Abstract. A wet chemical route method has been developed for synthesizing vertically well-aligned ZnO nanorods at a low temperature of 90 ºC. The uniformly vertical ZnO nanorods with different height were synthesized on a gold electrode of quartz crystal microbalance (QCM) for a high performance NH3 gas sensor. The fabricated sensors indicated well sensitivity and selectivity detection with NH3 gas. Effect of the height of ZnO nanorods on the change in frequency of the sensors was investigated. The flow rate of injection gas was at 15 sccm and 25 sccm while the response time of the sensor is 300 s and 80 s, respectively. At higher flow rate, the response time of sensor is faster. The flow rate of injection gas was not affected on the sensitivity of the sensor. The selectivity of the sensor was tested with several gases including of carbon monoxide, nitrous oxide and LPG (which consists of hydrocarbons like CH4, C3H8, C4H10). Keywords: Gas sensor, Zinc oxide, QCM, Resonance frequency.

INTRODUCTION

Rerently, ZnO has become an attractive semiconductor metal-oxide material due to its unique

properties, such as a direct band gap (3.37 eV), large exciton binding energy (60 meV), high thermal and chemical stability, transparence, biocompatibility, wide electrical conductivity range, and high melting temperature [1-3]. Moreover, one-dimensional (1D) ZnO nanostructures have attracted much attention because of their large aspect ratio, which makes them a good candidate for gas sensing applications [4, 5]. Most gas sensors using semiconductor metal-oxide materials are based on the change in electrical conductivity with the composition of the surrounding gas atmosphere. Major disadvantages in conductivity-based gas sensors are the high operating temperature, poor gas selectivity, and unstableness. These sensors are based on the changes in electrical resistance of the materials upon gas adsorption [4-8]. Otherwise, in the conductivity-based sensors using 1D nanostructured materials, the sensing layer is required to be continuous to ensure that conductivity exists in two electrodes of the sensor. Therefore, the sensing layers usually consist of the 1D nanostructures in parallel orientation with the substrate. Hence, the exposing area of the sensing layer is reduced significantly.

In this work, we fabricated a highly sensitive and reproducible ammonia gas sensor with a combination of ZnO nanorods and quartz crystal microbalance (QCM). Change in the resonant frequency can be related to change in the mass according to the Sauerbrey equation [9]. As such, QCM technique has been applied in biochemistry, analytical science, and other fields [10-12]. QCM coated with a sensing layer was also applied to the sensor due to its high reproducibility and stability. As the QCM is an extremely sensitive mass device that can detect the change in mass of a molecule, sensors based on the QCM have high sensitivity and accuracy. The operating principle of the QCM coated with ZnO nanorods sensor is based on the change in mass of the adsorbed gas in the sensing layer. Thus, mass-based gas sensors measure directly the adsorption process and do not require the charge carriers of the semiconductor materials to be the sensing layer to overcome the activation energy barrier. The gas sensor based on the QCM coated with ZnO can usually be operated at room temperature. In order to synthesize ZnO nanorods, the wet chemical route was chosen due to its obvious advantages of low synthesizing temperature, simple equipment, and easy operation. Moreover, this technique can synthesize on any substrate with a large-scale area, precise position control, and easy control of synthesis conditions. In comparison with the other synthesis methods, the wet chemical route has many advantages such as low cost, low temperature operation, high preferred orientation, and environmental friendliness. An important advantage of the method is its ability to synthesize the vertically well-aligned ZnO nanorods with high quality and uniform length and distribution. In the present paper, the fabrication processes of mass-based gas sensor using QCM coated with a sensing layer of vertically well-aligned

The 5th International Workshop on Advanced Materials Science and Nanotechnology (IWAMSN2010) - Hanoi, Vietnam - November 09-12, 2010

ZnO nanorods were reported. The effect of the height of ZnO nanorods on the sensitivity of as-fabricated sensors was examined. The characteristics, including sensitivity, reproducibility, response, and recovery times of the sensors, were also investigated with different ammonia concentrations.

EXPERIMENTAL

The QCM was cleaned in the acetone, ethanol, and de-ionized (DI) water to remove adsorbed dust and surface termination. Vertically aligned ZnO nanorods were then directly grown on the Au electrode of the QCM device by a wet chemical route. First, zinc acetate [Zn(COOCH3)2•2H2O] diluted in butanol was coated on the Au electrode by the spin-coating technique and was followed by heat-treatment at 300 ºC in air for 30 min to form a seed layer of ZnO nanocrystals. Subsequently, the QCM coated with the seed layer was vertically floated upside down on the aqueous solution surface of equal molar zinc nitrate [Zn(NO3)2•6H2O] and hexamethylenetetramine (HMTA) (C6H12N4). The hydrothermal process was conducted at 90 ºC. Growth time was maintained for 2 h and 3 h to examine effect of the height of ZnO nanorods on sensitivity of as-fabricated gas sensors. After reactions, the substrates were removed from the solution, rinsed with DI water, and dried with dry air blow. The morphology of the as-grown ZnO nanorods was analyzed by field emission scanning electron microscopy (FE-SEM).

Fig.1. Diagram of the testing gas system.

Gas sensing measurement of the fabricated sensors was conducted using a gas sensing system, as shown in figure 1. A flow-through technique with a constant flow rate was employed for the gas sensing test. The gas concentration was controlled by changing the mixing ratio of the parent gases on MFC1 and synthetic air on MFC2 in a mixing chamber. The testing process was conducted by two steps as follows. At the beginning, air flow (MFC4) was flowed through the chamber to obtain the baseline of the frequency. Balance NH3 gas flow (MFC3), that was similar to air flow, was then flowed through the chamber to replace the air flow. The frequency shifts of the sensors were monitored by a frequency counter, QCM200, which was connected to a computer system via the SRSQCM200 software program and stored in a PC.

RESULTS AND DISCUSSION

Figure 2 shows SEM images of the as-grown ZnO nanorods on the Au electrode of the QCM

synthesized by wet chemical route method with various growth times of 2 h (Fig. 2a) and 3 h (Fig. 2b). It can be clearly seen that the morphology and aspect ratio of as-grown ZnO nanorod array are closely relative to the growth times. Namely, the height of ZnO nanorods synthesized for 2 and 3 h was 1.7 μm and 3 μm, respectively (Fig. 2 (bottom)). Though the height of ZnO nanorods was very different, the average diameter of ZnO nanorods were around 100 nm and 80 nm correspond to height of 1.7 and 3 μm, respectively (Fig. 2 (upper)). The morphology of the ZnO nanorods with a hexagonal structure was vertically well-aligned and uniformly distributed on the Au electrode of the QCM. This shows that the exposed area of the sensing layer was remarkably enhanced compared with the sensing layer of the ZnO nanowires.

Fig. 2. SEM images of top-view (upper) and side view (bottom) of as-grown ZnO nanorod synthesized for 2 h

and 3 h. Figure 3 shows the response transients of the as-grown QCM sensor with various ZnO nanorod

heights to switching-on and off of the NH3 gas-flow with different concentrations (50, 100, 200, 400 and 800 ppm) at room temperature (25 ºC). In the first stage, the sensor flushed a reference air gas flow of 15 sccm to obtain a baseline. The sensor was then exposed to a NH3 gas flow of 15 sccm with a certain concentration, which leads to frequency response until a steady stage was reached, indicating maximum adsorption of NH3 gas onto the QCM sensor. The NH3 gas flow was finally replaced by the air gas flow and the sensor returned back to its baseline. In both cases, the flow rate of the diluted ammonia gas and dry air was fixed at 15 sccm. Hence, in the gas sensing chamber, the flow and pressure were ensured to be constant. Both cases also expose that when the NH3 gas flow was replaced by air flow, the resonant frequency returned to its original value. This indicates that the absorbed NH3 molecules were completely removed. The vertically aligned ZnO nanorods grown on the QCM electrodes might have enhanced this fact. However, the shift frequency at each NH3 concentration is resonant with the case of ZnO nanorod height of 3 μm. Namely at 50 ppm of NH3 the change in frequency of sensor with ZnO nanorod height of 1.7 and 3 μm is 7 and 9 Hz, respective. It can be explained that the surface area of ZnO nanorod height of 3 μm is larger than that of ZnO nanorod height of 1.7 μm. Therefore, more NH3 molecules are adsorbed on the ZnO nanorods with larger surface area.

Fig. 3. Variation in the resonant frequency with time during adsorption of different NH3

concentrations of as-fabricated sensor with (a) 1.7 μm and (b) 3 μm of ZnO nanorod height. In order to examine the effect of flow rate of injection gas on sensitivity of sensors, various flow

rate of injection gas of 25 and 15 sccm was tested. Figure 4a shows the variation in the resonant frequency with time during adsorption of different flow rate of injection gas. The flow rate of injection gas was not affected on the sensitivity of the sensor. The flow rate of injection gas was at 15 sccm and 25

sccm while the response time of the sensor is 300 s and 80 s, respectively. Thus, at higher flow rate, the response time of sensor is faster. One of the most important characteristics of the gas sensor is its selectivity. To examine the selectivity of the QCM coated with ZnO nanorods gas sensor, we tested the sensitivity of the sensor with carbon dioxide and LPG, which consists of hydrocarbons like CH4, C3H8, C4H10. The conductivity-based gas sensor using ZnO thin film indicates significant sensitivity to LPG significantly [13]. For the gas sensing properties, the fabricated sensors were not responsive at a concentration of about 1% LPG. The change in frequency of the sensors was realized when the concentration of LPG was about 2%. Figure 4b shows the comparison of the resonant frequency of the device exposed to LPG and carbon dioxide with that exposed to ammonia. Although the concentration of LPG was 2%, the frequency shift of the sensor only changed by 0.8 Hz. In comparison with 50, 100, or 200 ppm NH3, the response of the sensor with LPG was insignificant. When the LPG concentration was doubled (4%), the frequency shift of the sensor was only 1.8 Hz, as shown in Fig. 4b. In comparison with conductivity-based sensors using ZnO material [14], as-fabricated sensor had almost no sensitivity with LPG. This is the difference between the conductivity-based sensor and mass-based sensor using QCM coated with ZnO nanorods. The selectivity of the sensors was also indicated when exposed to carbon dioxide. The frequency shift of the sensors with the carbon dioxide was even less than that with the LPG. The figure 4b shows clearly the frequency shift of the sensor when exposed to carbon dioxide. At 2% and 4% carbon dioxide concentrations, the change in frequency was only 0.6 and 1.3 Hz, respectively. Therefore, the sensor based on QCM coated with ZnO nanorods had high selectivity with each particular gas.

Fig. 4 (a) Effect of flow rate of injection gas on response time and (b) selective NH3 gas of QCM

coated with ZnO nanorods sensor.

CONCLUSIONS

Vertical well-aligned ZnO nanorods were successfully grown on gold-electrode of QCM by wet chemical route method. The ZnO nanorods were uniformly distributed on the substrate with height to be 1.7 and 3 μm correspond to synthesis time of 2 and 3 h, respectively. The sensor based on QCM coated with ZnO nanorods shows a good interaction with the ammonia and could detect low concentration of several tens part per million. The results indicated the reproducible and reversible performance of the sensor. Moreover, the sensor shows the selective characteristics via less sensitivity with the LPG gas and carbon dioxide.

Acknowledgments. This work is supported by Ministry of Education and Training under research project code: B2010-01-387 (2010-2011). References

1. Y.Z. Wang, B.L. Chu, Superlattices Microstruct. 44 (2008) 54–61. 2. J.C. Ronfard-Haret, J. Lumin. 104 (2003) 103–114. 3. Lu, Y. Wang, X. Lin, Mater. Lett. 63 (2009) 2321–2323.

4. M.W. Ahn, K.S. Park, J.H. Heo, D.W. Kim, K.J. Choi, J.G. Park, Sens. Actuators B 138 (2009) 168–173

5. M.C. Carotta, A. Cervi, V. di Natale, S. Gherardi, A. Giberti, V. Guidi, D. Puzzovio, B. Vendemiati, G. Martinelli, M. Sacerdoti, D. Calestani, A. Zappettini, M. Zha, L. Zanotti, Sens. Actuators B 137 (2009) 164–169.

6. X. Du, S.M. George, Sens. Actuators B 135 (2008) 152-160. 7. N.V. Hieu, Sens. Actuators B 144 (2010) 425-431. 8. Ibrahim A. Al-Homoudi, J.S. Thakur, R. Naik, G.W. Auner, G. Newaz, Appl. Surf. Sci. 253 (2007)

8607-8614. 9. G.W. Sauerbrey, Z. Phys. 155 (1959) 206-222. 10. A. Rudrake, K. Karan, J.H. Horton, J. colloid interface Sci. 332 (2009) 22-31. 11. T. Abe, M. Essahi, Sens. Actuators A 82 (2000) 139-143. 12. P.A. Lieberzeit, A. Rehman, B. Najafi, F.L. Dickert, Anal. Bioanal. Chem. 391 (2008) 2897–2903. 13. D. Mishra, A. Srivastava, A. Srivastava, P.K. Shukla, Appl. Surf. Sci. 255 (2008) 2947. 14. X.H. Wang, J. Zhang, Z.Q. Zhu, Ammonia sensing characteristics of ZnO nanowires studied by

quartz crystal microbalance, Appl. Surf. Sci. 252 (2006) 2404-2411.