Determining the Optimum Exposure and Recovery Periods for Efficient Operation of a QCM Based Elemental Mercury Vapor Sensor

Abstract

In recent years, mass based transducers such as quartz crystal microbalance (QCM) have gained huge interest as potential sensors for online detection of elemental mercury (Hg0) vapor from anthropogenic sources due to their high portability and robust nature enabling them to withstand harsh industrial environments. In this study, we determined the optimal Hg0 exposure and recovery times of a QCM based sensor for ensuring its efficient operation while monitoring low concentrations of Hg0 vapor (<400 [subscript]ppbv[/subscript] ). The developed sensor was based on an AT-cut quartz substrate and utilized two gold (Au) films on either side of the substrate which functions as the electrodes and selective layer simultaneously. Given the temporal response mechanisms associated with mass based mercury sensors, the experiments involved the variation of Hg0 vapor exposure periods while keeping the recovery time constant following each exposure and vice versa. The results indicated that an optimum exposure and recovery periods of 30 and 90 minutes, respectively, can be utilized to acquire the highest response magnitudes and recovery rate towards a certain concentration of Hg0 vapor whilst keeping the time it takes to report an accurate reading by the sensor to a minimum level as required in real-world applications.

K. M. Mohibul Kabir 1 and Samuel J. Ippolito 1, 2 and Glenn I. Matthews 2 and S. Bee Abd Hamid 3 and Ylias M. Sabri 1 and Suresh K. Bhargava 1

Academic Editor:Nick Chaniotakis

1, Centre for Advanced Materials & Industrial Chemistry (CAMIC), School of Applied Sciences, RMIT University, Melbourne, VIC 3001, Australia
2, School of Electrical and Computer Engineering, RMIT University, Melbourne, VIC 3001, Australia
3, Nanotechnology & Catalysis Research Center (NANOCAT), Institute of Postgraduate Studies (IPS), University of Malaya, 3rd Floor, Block A, 50603 Kuala Lumpur, Malaysia

Received 6 May 2015; Revised 16 July 2015; Accepted 21 July 2015; 29 July 2015

This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

1. Introduction

The rapid growth of industrialization in the last century has increased the emission of toxic metal species such as elemental mercury in the atmosphere [1-7]. It is of high importance to control the emission of these metal species from common industrial sources in order to reduce the advert effect they are having on the environment as well as human health. Recently, new and more stringent rules have been introduced by government and environmental bodies worldwide to limit the amount of mercury emitted from industrial processes. For example, the average daily mercury emission from cement kilns in Germany is proposed to be limited to 3.5 [figure omitted; refer to PDF] [8]. In order to comply with these regulations, efficient removal technologies need to be implemented on targeted industry sites. Furthermore, in order to evaluate the efficiency of these removal technologies, highly accurate and sensitive online mercury vapor sensor is required.

In recent years, it has been shown that the mass based transducer such as quartz crystal microbalance (QCM) possesses several major advantages over other commonly used elemental mercury (Hg⁰ ) vapor measurement techniques, which are typically based on atomic absorption spectroscopy (AAS) and atomic fluorescence spectrometry (AFS), and so forth [9-20]. The QCM based Hg⁰ vapor sensors were found to be highly portable and selective and they do not require sample pretreatment which make them highly suitable for online monitoring of Hg⁰ vapor within industrial applications [10]. Moreover, the design and the selective layer of QCM based sensors can be altered to achieve even higher sensitivity and selectivity towards low concentrations of Hg⁰ vapor. We recently reported that a QCM sensor based on gold nanospikes [21] can detect Hg⁰ vapor concentrations down to ~2.5 [figure omitted; refer to PDF] , which is lower than the threshold mercury exposure limit of 5.6 [figure omitted; refer to PDF] set by world health organization [22]. Due to their potential to be used as online or hand-held sensors, it is very important to study QCM based Hg⁰ vapor sensors extensively in order to achieve the best performance for Hg⁰ vapor detection.

The schematic of a typical QCM based gas sensor is shown in Figure 1. The sensor is usually fabricated by depositing two electrodes on both sides of a suitable piezoelectric substrate (i.e., AT cut quartz). A bulk acoustic wave is generated upon the application of an electric potential on one of the electrodes [23-26]. The sensing film can be deposited on the electrode or the electrodes themselves can be made of a material which is selective to target analytes. Crystallographic characteristics and the thickness of the substrate and the electrodes determine the resonant frequency ( [figure omitted; refer to PDF] ) of the sensor. Any perturbation on the sensing film (which typically occurs through mass loading of targeted analytes on the sensing film) results in a shift of the resonant frequency ( [figure omitted; refer to PDF] ) of the sensor. This shift in [figure omitted; refer to PDF] is highly dependent on the amount of analyte that interacts with the sensing film and can be related by the Sauerbrey equation (1) [27]: [figure omitted; refer to PDF] where [figure omitted; refer to PDF] represents the shift in the resonant frequency, [figure omitted; refer to PDF] is the change in the mass of the sensing surface, [figure omitted; refer to PDF] is the active area of the QCM electrodes, and [figure omitted; refer to PDF] and [figure omitted; refer to PDF] are the crystal density and shear modulus of the piezoelectric crystal, respectively. It can be observed from (1) that the shift in [figure omitted; refer to PDF] increases as the mass loading on the sensing film increases, which indicates that the shift in [figure omitted; refer to PDF] is proportional to the concentration of the species being detected.

Figure 1: Schematic diagram of a typical QCM based gas sensor.

[figure omitted; refer to PDF]

In the current study, we focused on the Hg⁰ vapor detection application of a QCM based sensor. Therefore, extensive experimental and analytical approach was taken to determine the optimum exposure and recovery period to be used in order to efficiently monitor low concentration of Hg⁰ vapor using QCM based sensors.

2. Experimental

2.1. Sensor Fabrication and Quality Factor Determination

The QCM based sensor was fabricated on an AT cut quartz substrate having a diameter and thickness of 7.5 mm and 166 μ m, respectively. 100 nm of Au layer on a 10 nm Titanium (Ti) adhesion layer was deposited on both sides of the quartz substrate to function as electrodes as well as the sensing layer simultaneously. The deposition was performed using a Balzers e-beam (BAK 600) evaporator operating at 22°C. Both of the electrodes were circular shaped which were patterned by using a shadow mask. The diameter of the electrodes was 4.5 mm. A photograph of the fabricated sensor is shown in Figure 2(a).

Figure 2: (a) Photograph of the developed QCM based sensor and (b) network analyzer response of the sensor at 2 kHz span around the center frequency.

(a) [figure omitted; refer to PDF]

(b) [figure omitted; refer to PDF]

A network analyzer (Agilent E5100A) was used to measure the frequency response and determine the quality factor of the fabricated sensor. Figure 2(b) shows the frequency response of the sensor at 2 kHz span around the center frequency (9.99909 MHz). The quality factor of the sensor can be calculated using [28] [figure omitted; refer to PDF] where [figure omitted; refer to PDF] . The quality factor of the fabricated sensor was calculated as 6410 (i.e., >2500), indicating the sensor is suitable for gas phase analysis.

2.2. Mercury Testing Setup

The Hg⁰ vapor concentration used in different experiments of this study was 365 [figure omitted; refer to PDF] . This particular concentration of Hg⁰ vapor was generated by setting the temperature of a NIST certified permeation tube (VICI) to 80°C. The Hg⁰ delivery system was also calibrated on site using a potassium permanganate (KMnO₄ ) trapping method. This involved capturing the generated Hg⁰ vapor steam within a train of impingers containing H₂ SO₄ /KMnO₄ and analyzing the solution by inductively coupled mass spectroscopy (ICP-MS) afterwards. This was done in order to ensure the concentration of Hg⁰ vapor was correct. The chamber which housed the QCM sensor had volume of 100 mL and was made of Teflon and stainless steel. The sensor recovery was a one-step process involving the exposure of the sensor toward dry nitrogen (N₂ ). A constant flow rate of 200 sccm and an operating temperature of 30°C were maintained throughout the whole study. The operating temperature of the sensor chamber was kept constant using an active PID controller. Temperature fluctuations within ±0.5°C were observed; however, this did not affect the sensing result due to the high temperature stability of AT-cut quartz. A Maxtek RQCM was used for oscillation as well as to monitor resonant frequency of the sensor.

3. Results and Discussion

3.1. Optimum Hg⁰ Vapor Exposure Time

The performance of the developed Hg⁰ vapor sensor for different exposure times was investigated by exposing the sensor towards 365 [figure omitted; refer to PDF] of Hg⁰ vapor for a range of exposure period ranging from 10 to 120 minutes whilst keeping the recovery period constant at 60 minutes. The structure of the full test is shown in Figure 3(a). It can be seen that after every Hg⁰ vapor exposure period dry N₂ was flushed to the sensor for a period of 60 minutes. This process was performed in order to desorb the Hg⁰ vapor molecules from the Au surface and thus allow the sensor to recover to its baseline frequency. Figure 3(b) shows the dynamic response of the sensor for the entire test. It can be observed that the Hg⁰ exposure resulted in a negative shift in the resonant frequency while the sensor was observed to return it to its baseline frequency during the recovery time.

Figure 3: (a) Structure of the test performed to determine the optimum Hg⁰ exposure time; (b) sensor's dynamic response during the full test. Each exposure/recovery event's cycle was repeated 3 times before the next event started.

(a)

[figure omitted; refer to PDF]

(b) [figure omitted; refer to PDF]

Figure 4(a) shows the sensor's response toward Hg⁰ vapor concentration of 365 [figure omitted; refer to PDF] for an exposure time of 30 and 120 minutes while the recovery period was kept constant at 60 minutes. The pulses shown were extracted from a set of continuous pulses which were measured using a fixed exposure and recovery time for at least 3 cycles. This was done in order to obtain stable performance of the sensors in the final cycle. It can be observed from Figure 4(a) that the sensor response profile between the two exposure periods is different when considering the first 30-minute exposure period. This was mainly due to the different initial state between the two conditions presented (i.e., see Figure 3(b)). This can be further justified from Figures 4(b) and 4(c). That is, it can be observed from Figure 4(b) that the last pulse of 30-minute exposure and 60-minute recovery period cycles has the same adsorption response profile as the first pulse of 40-minute exposure and 60-minute recovery period cycle due to the same initial state of the sensor. This is further confirmed in Figure 4(c) where it can be observed that the last pulse of 90-minute exposure and 60-minute recovery period responds similarly as 120-minute exposure and 60-minute recovery period. Figure 4(d) shows the response magnitudes of the sensor toward the same Hg⁰ vapor concentration (365 [figure omitted; refer to PDF] ) for different exposure periods ranging from 10 to 120 minutes and a constant recovery period of 60 minutes. The response magnitudes of the last pulses for each Hg⁰ vapor exposure and recovery cycle were considered for analysis. The final pulse of each cycle was chosen as it allowed for the sensor to reach stability thereby resulting in desorption of the adsorbed Hg⁰ molecules from the Au surface and thus reducing the impact of a preceding pulse on the characterization of the following Hg⁰ exposure cycle. It can be observed from Figure 4(d) that the sensor's response magnitude increased rapidly when the Hg⁰ vapor exposure time was increased up to 30 minutes. However, it can also be observed that the sensor's response magnitudes did not vary significantly when the Hg⁰ vapor exposure time was increased beyond 30 minutes. This observation indicates that the processes of Hg-Au amalgamation and diffusion of Hg atoms into the Au surface are relatively high up to a period of 30 minutes; however, they are reduced significantly beyond the exposure period of 30 minutes due to the Au surface reaching saturation. Therefore, no significant increase in response magnitudes was observed when the Hg⁰ vapor was exposed for more than 30 minutes. The initial Hg⁰ sorption kinetics (Figure 4(a)) is also observed to have changed (slowed) at the exposure time of 120 minutes due to the higher content of the mercury that was already on the surface from the preceding two cycles of the same pulse (i.e., having 120-minute exposure and 60-minute recovery periods).

Figure 4: (a) Sensor's dynamic response toward Hg⁰ vapor concentration of 365 [figure omitted; refer to PDF] exposed for 30 and 120 minutes; (b) 30 and 40 minutes; and (c) 90 minutes and 120 minutes, all having a constant recovery time of 60 minutes. (d) Sensor's response magnitude towards 365 [figure omitted; refer to PDF] of Hg⁰ exposed for 10 to 120 minutes with a recovery time of 60 minutes.

(a)

[figure omitted; refer to PDF]

(b) [figure omitted; refer to PDF]

(c) [figure omitted; refer to PDF]

(d) [figure omitted; refer to PDF]

3.2. Optimum Hg⁰ Vapor Recovery Time

In order to determine the optimum recovery time of the QCM based Hg⁰ sensor, the sensor was exposed towards Hg⁰ vapor concentration of 365 [figure omitted; refer to PDF] for a period of 30 minutes while the recovery period was varied within a range of 10 to 90 minutes. A period of 30 minutes was chosen as the Hg⁰ vapor exposure time as it was determined to be the optimum time from the previous test (Section 3.1) at which the saturated response magnitude was achieved at a low turnaround time. The test structure for determining optimum recovery period for QCM based Hg⁰ vapor sensor is shown in Figure 5(a).

Figure 5: (a) Structure of the test performed to determine the optimum Hg⁰ recovery time; (b) sensor's dynamic response during the full test. Each exposure/recovery event's cycle was repeated 3 times before the next event started.

(a) [figure omitted; refer to PDF]

(b) [figure omitted; refer to PDF]

The dynamic response of the sensor throughout the test can be observed from Figure 5(b). It can be observed that each exposure and recovery cycle was repeated 3 times before the next cycle started. The sensor's dynamic response for 30 minutes exposure toward Hg⁰ vapor concentration of 365 [figure omitted; refer to PDF] employing three different recovery periods of 70, 90, and 120 minutes can be seen in Figure 6(a). It can be observed that the sensor exhibited a response magnitude of ~17.5 Hz when Hg⁰ exposure and recovery periods were 30 and 70 minutes, respectively. It can also be observed that the sensor's response magnitude rose to ~19 Hz when the recovery period was increased to ~90 minutes while keeping the same Hg⁰ vapor exposure period. Interestingly, the sensor's response magnitudes did not vary significantly when the recovery period was further increased to 120 minutes.

Figure 6: (a) Sensor's dynamic response towards 365 [figure omitted; refer to PDF] of Hg⁰ vapor exposed for 30 minutes with desorption time of 70, 90, and 120 minutes and (b) sensor's adsorption and desorption magnitude towards 30 minutes of Hg⁰ vapor exposure while the desorption period was varied between 10 and 120 minutes. The response magnitudes of the final cycle of each event are plotted.

(a)

[figure omitted; refer to PDF]

(b) [figure omitted; refer to PDF]

A clearer view of sensor's adsorption and desorption magnitudes for the different recovery periods ranging from 10 to 120 minutes can be observed from Figure 6(b). Here, it should be noted that the sensor's desorption magnitudes were taken from the end of the recovery periods. It was interesting to observe that a ~23 Hz response magnitude was observed for a 30-minute exposure and 60-minute recovery periods pulse (Figure 4(d)) while in Figure 6(b) a ~18 Hz response was observed for the same exposure and recovery conditions. However, the difference arises from the preceding pulse cycles that were run for each condition. That is, the reduction in response magnitude for recovery period test was observed because the preceding recovery periods (i.e., 50 minutes) were not efficient enough for the sensor to release the adsorbed Hg⁰ molecules from Au surface as opposed to the longer recovery period (60 minutes) used in all the pulses of the exposure test (Figure 3(a)). It can be seen from Figure 6(b) that the sensor's adsorption and desorption magnitudes increased significantly up to 90 minutes even though the Hg⁰ exposure period was kept constant at 30 minutes. It can also be observed from Figure 6(b) that while the sensor has high recovery efficiency for most of the recovery times tested the 90-minute recovery period showed to have more than 90% recovery efficiency while exhibiting maximum response magnitude towards Hg⁰ vapor. Overall, the results indicate that at least 90 minutes of recovery period is needed to acquire the highest possible response magnitudes as well as the recovery efficiency from the sensor while detecting low concentrations of Hg⁰ vapor without jeopardizing sensor's response time. Getting a higher response magnitude and the recovery efficiency is particularly important as it will enable the sensor to have high sensitivity and thereby result in the efficient detection of low concentrations of Hg⁰ vapor in industrial processes.

4. Conclusion

The Hg⁰ vapor sensing performance of a QCM based sensor was investigated for different Hg⁰ vapor exposure and recovery periods. The developed sensor was based on an AT-cut quartz substrate containing thin Au-film electrodes on both sides. The sensor was tested towards Hg⁰ vapor concentration of 365 [figure omitted; refer to PDF] while the Hg⁰ vapor exposure and recovery periods were varied between 10 and 120 minutes. The overall results indicate that 30-minute Hg⁰ vapor exposure and 90-minute recovery periods can be utilized to achieve the highest response magnitudes and recovery efficiency from the sensor while keeping the turnaround time in a minimum level for real-world applications.

Acknowledgments

The authors acknowledge the Microelectronic and Materials Technology Centre (MMTC) at RMIT University for allowing the use of their facilities. Authors also acknowledge the Australian Research Council (ARC) for supporting this project and Samuel J. Ippolito acknowledges the ARC for APDI fellowship (LP100200859).

Conflict of Interests

The authors declare that there is no conflict of interests regarding the publication of this paper.

References

[1] UNEP Global Mercury Assessment 2013: Sources, Emissions, Releases and Environmental Transport , UNEP Chemicals Branch, Geneva, Switzerland, 2013.

[2] L. B. Chakraborty, A. Qureshi, C. Vadenbo, S. Hellweg, "Anthropogenic mercury flows in india and impacts of emission controls," Environmental Science and Technology , vol. 47, no. 15, pp. 8105-8113, 2013.

[3] L. D. Hylander, M. Meili, "500 years of mercury production: global annual inventory by region until 2000 and associated emissions," Science of the Total Environment , vol. 304, no. 1-3, pp. 13-27, 2003.

[4] D. W. Boening, "Ecological effects, transport, and fate of mercury: a general review," Chemosphere , vol. 40, no. 12, pp. 1335-1351, 2000.

[5] S. Liang, C. Zhang, Y. Wang, M. Xu, W. Liu, "Virtual atmospheric mercury emission network in China," Environmental Science and Technology , vol. 48, no. 5, pp. 2807-2815, 2014.

[6] D. Jampaiah, K. M. Tur, S. J. Ippolito, Y. M. Sabri, J. Tardio, S. K. Bhargava, B. M. Reddy, "Structural characterization and catalytic evaluation of transition and rare earth metal doped ceria-based solid solutions for elemental mercury oxidation," RSC Advances , vol. 3, no. 31, pp. 12963-12974, 2013.

[7] C. T. Driscoll, R. P. Mason, H. M. Chan, D. J. Jacob, N. Pirrone, "Mercury as a global pollutant: sources, pathways, and effects," Environmental Science and Technology , vol. 47, no. 10, pp. 4967-4983, 2013.

[8] K. M. Kabir, Y. M. Sabri, G. I. Matthews, L. A. Jones, S. J. Ippolito, S. K. Bhargava, "Selective detection of elemental mercury vapor using a surface acoustic wave (SAW) sensor," The Analyst , 2015.

[9] K. M. Kabir, Y. M. Sabri, G. I. Matthews, S. J. Ippolito, S. K. Bhargava, "Cross sensitivity effects of volatile organic compounds on a SAW-based elemental mercury vapor sensor," Sensors and Actuators B: Chemical , vol. 212, pp. 235-241, 2015.

[10] Y. M. Sabri, S. J. Ippolito, A. J. Atanacio, V. Bansal, S. K. Bhargava, "Mercury vapor sensor enhancement by nanostructured gold deposited on nickel surfaces using galvanic replacement reactions," Journal of Materials Chemistry , vol. 22, no. 40, pp. 21395-21404, 2012.

[11] P. D. Selid, H. Xu, E. M. Collins, M. S. Face-Collins, J. X. Zhao, "Sensing mercury for biomedical and environmental monitoring," Sensors , vol. 9, no. 7, pp. 5446-5459, 2009.

[12] N. Pirrone, W. Aas, S. Cinnirella, R. Ebinghaus, I. M. Hedgecock, J. Pacyna, F. Sprovieri, E. M. Sunderland, "Toward the next generation of air quality monitoring: mercury," Atmospheric Environment , vol. 80, pp. 599-611, 2013.

[13] D. Martin-Yerga, M. B. Gonzalez-Garcia, A. Costa-Garcia, "Electrochemical determination of mercury: a review," Talanta , vol. 116, pp. 1091-1104, 2013.

[14] S. K. Pandey, K.-H. Kim, R. J. C. Brown, "Measurement techniques for mercury species in ambient air," Trends in Analytical Chemistry , vol. 30, no. 6, pp. 899-917, 2011.

[15] M. Rex, F. E. Hernandez, A. D. Campiglia, "Pushing the limits of mercury sensors with gold nanorods," Analytical Chemistry , vol. 78, no. 2, pp. 445-451, 2006.

[16] B. Mazzolai, V. Mattoli, V. Raffa, G. Tripoli, D. Accoto, A. Menciassi, P. Dario, "A microfabricated physical sensor for atmospheric mercury monitoring," Sensors and Actuators A: Physical , vol. 113, no. 3, pp. 282-287, 2004.

[17] D. P. Ruys, J. F. Andrade, O. M. Guimarães, "Mercury detection in air using a coated piezoelectric sensor," Analytica Chimica Acta , vol. 404, no. 1, pp. 95-100, 2000.

[18] T. Thundat, E. A. Wachter, S. L. Sharp, R. J. Warmack, "Detection of mercury vapor using resonating microcantilevers," Applied Physics Letters , vol. 66, pp. 1695-1697, 1995.

[19] E. P. Scheide, J. K. Taylor, "Piezoelectric sensor for mercury in air," Environmental Science and Technology , vol. 8, no. 13, pp. 1097-1099, 1974.

[20] K. M. Kabir, Y. M. Sabri, A. E. Kandjani, G. I. Matthews, M. Field, L. A. Jones, A. Nafady, S. J. Ippolito, S. K. Bhargava, "Mercury sorption and desorption on gold: a comparative analysis of surface acoustic wave and quartz crystal microbalance-based sensors," Langmuir , 2015.

[21] Y. M. Sabri, S. J. Ippolito, J. Tardio, V. Bansal, A. P. O'Mullane, S. K. Bhargava, "Gold nanospikes based microsensor as a highly accurate mercury emission monitoring system," Scientific Reports , vol. 4, article 6741, 2014.

[22] T. P. McNicholas, K. Zhao, C. Yang, S. C. Hernandez, A. Mulchandani, N. V. Myung, M. A. Deshusses, "Sensitive detection of elemental mercury vapor by gold-nanoparticle-decorated carbon nanotube sensors," Journal of Physical Chemistry C , vol. 115, no. 28, pp. 13927-13931, 2011.

[23] S. K. Vashist, P. Vashist, "Recent advances in quartz crystal microbalance-based sensors," Journal of Sensors , vol. 2011, 2011.

[24] K. Kanazawa, N.-J. Cho, "Quartz crystal microbalance as a sensor to characterize macromolecular assembly dynamics," Journal of Sensors , vol. 2009, 2009.

[25] M. Hoummady, A. Campitelli, W. Wlodarski, "Acoustic wave sensors: design, sensing mechanisms and applications," Smart Materials and Structures , vol. 6, no. 6, pp. 647-657, 1997.

[26] D. Jr. Ballantine, R. M. White, S. J. Martin, A. J. Ricco, E. Zellers, G. Frye, H. Wohltjen Acoustic Wave Sensors: Theory, Design, & Physico-Chemical Applications , Academic Press, 1996.

[27] G. Sauerbrey, "Use of quartz vibration for weighing thin films on a microbalance," Zeitschrift für Physik , vol. 155, pp. 206-212, 1959.

[28] M. D. Ward, D. A. Buttry, "In situ interfacial mass detection with piezoelectric transducers," Science , vol. 249, no. 4972, pp. 1000-1007, 1990.