- Email: [email protected]
Conductometric graphene sensors decorated with nanoclusters for selective detection of Hg2+ traces in water
Accepted Manuscript Title: Conductometric graphene sensors decorated with nanoclusters for selective detection of Hg2+ traces in water Author: Ahmad I. Ayesh Zainab Karam Falah Awwad Mohammed A. Meetani PII: DOI: Reference:
S0925-4005(15)30001-0 http://dx.doi.org/doi:10.1016/j.snb.2015.06.075 SNB 18652
To appear in:
Sensors and Actuators B
Received date: Revised date: Accepted date:
18-4-2015 10-6-2015 19-6-2015
Please cite this article as: A.I. Ayesh, Z. Karam, F. Awwad, M.A. Meetani, Conductometric graphene sensors decorated with nanoclusters for selective detection of Hg2+ traces in water, Sensors and Actuators B: Chemical (2015), http://dx.doi.org/10.1016/j.snb.2015.06.075 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Conductometric graphene sensors decorated with nanoclusters for selective detection of Hg2+ traces in water Ahmad I. Ayesh1,*, Zainab Karam2, Falah Awwad3, and Mohammed A. Meetani4 Department of Math., Stat. and Physics, Qatar University, Doha, Qatar 2
Department of Electrical Engineering, United Arab Emirates University, Al Ain,
cr
3
Masdar Institute, Abu Dhabi, United Arab Emirates
ip t
1
4
us
United Arab Emirates
Chemistry Department, United Arab Emirates University, Al Ain,
an
United Arab Emirates
* Corresponding author. Email: [email protected], Tel.: +974‐4403‐6592, P. O. Box 2713, Doha,
Abstract:
M
Qatar
d
Mercury ion (Hg2+) sensors based on graphene and gold nanoclusters are presented in this work.
Ac ce pt e
The sensors follow the structure of field‐effect transistor (FET), and their sensing mechanism rely on the change in conductance of graphene and gold nanocluster percolating film. Each sensor was fabricated by deposition of interdigitated metal electrodes on the surface of graphene by thermal evaporation through a shadow mask. Next, gold nanoclusters are generated by magnetron discharge sputtering and inert‐gas condensation inside an ultra‐high compatible system, and they are self‐assembled on the surface of graphene. The sensitivity and selectivity of graphene to Hg2+ are vastly enhanced by incorporating the nanoclusters on its surface. The detection limit of the present sensors is below the safe concentration of Hg2+ in drinking water set by different universal agencies. In addition, those sensors are practical and easy to operate in field for real life applications. 1
Page 1 of 16
Keywords: Graphene, gold nanoclusters, mercury sensor, inert‐gas condensation
ip t
Introduction:
cr
Mercury metal is released into water by different sources including sewage industrial waste, thus, it can enter human food chain. It is considered one of the most harmful pollutant of heavy
us
metals since it is non‐biodegradable, and it can enter to human body by different means including direct consumption (for example, through drinking water), absorption through skin,
an
and by respiratory system. Exposure to mercury cause severe effects on human health such as brain damage, kidney failure, damage in the nervous system, birth defects, chromosome
M
breakage, and paralysis [1, 2]. The United Nation Environment Program (UNEP) assessed the annual released quantity of mercury to 4400 – 7500 tons [3]. In addition, the International
Ac ce pt e
d
World Health Organization regulated the maximum allowed amount of mercury ions (Hg2+) in drinking water to 6 ppb [4]. Nevertheless, previous studies estimated the allowed amount of inorganic Hg2+ in water to 0.5 ppb [5]. Therefore, the development of sensitive, selective, reliable, and cost effective Hg2+ sensors is needed for medical diagnostic, quality control of food industry, as well as water and environment quality monitoring. In the past years different approaches were involved in detection of low concentrations of Hg2+ such as X‐ray absorption spectroscopy [6], spectrophotometry [7], inductively coupled plasma mass spectrometry [8], cold vapor atomic fluorescent spectroscopy [9], and fluorescence assay [10]. However, the previous methods are expensive, and not practical to apply in field due to their delicate instrumentation.
2
Page 2 of 16
Recently our group has been working on developing conductometric sensors based on nanostructures for different stimulate species [11‐13]. These sensors are very sensitive, cheap to manufacture, exhibit low power consumption, and practical to use in outdoor field. The main
ip t
element of those sensors is atomic nanoclusters that are fabricated by sputtering and inert condensation technique inside an ultra‐high compatible system [14]. This technique has many
cr
advantages over other nanocluster syntheses methods where the produced nanoclusters are of
us
high purity since they are prepared inside a vacuum chamber; charged, thus their size can be selected using a suitable mass filter; size tunable by controlling source parameters such as
an
sputtering discharge and inert‐gas condensation; and they can be assembled directly on a substrate without any extra experimental steps [15]. In addition, sputtering is a well‐established
M
technique that is adopted by many industries.
Generally, nanoclusters have unique properties that make them suitable for sensing applications
d
such as the large ratio of surface atoms that are normally unsaturated, thus, they can bind with
Ac ce pt e
other atoms that have suitable chemical activity. A main track of research in this field is to optimize the sensitivity, selectivity, and reproducibility of the fabricated nanoclusters. Among those nanoclusters, gold (Au) was found an attractive material for sensor and catalyst applications [16]. Au nanoclusters were used for different sensor applications [17], including Hg2+ sensor due to their high affinity to mercury [18]. However, the selectivity of Au nanocluster based sensors to Hg2+ ions is still a major challenge that should be addressed to enable commercial production of those sensors [19]. In addition, sensors based on nanoclusters and nanowires are limited because their detection depends on the induced field‐effect [20]. Graphene is a superior nanomaterial that has been experimentally produced in 2004 [21]. It exhibits high electrical conductivity, chemically stable, and it holds unique mechanical and thermal properties that attracted strong scientific interest recently. It consists of carbon atoms
3
Page 3 of 16
that are packed into a lattice of single layer with the structure of honeycomb. Graphene has been used recently for innovative sensor applications to enable the detection of a wide range of stimulant, because of its extraordinary electrical properties, such as glucose [22], heavy metals
ip t
[23], and different gas species [24, 25]. In addition, graphene is predicted to exceed carbon nanotubes in the field of sensor applications because of its large detection area as well as its
cr
biocompatibility. Nevertheless, the selectivity of graphene based sensors needs to be
us
engineered to make it suitable to target particular chemical stimulate.
In this work, we present novel conductometric sensors based on graphene and Au nanoclusters
an
that are highly selective to Hg2+ ions. Electrical electrodes were deposited on the surface of graphene by thermal evaporation. Au nanoclusters were produced by sputtering and inert gas
M
condensation technique inside an ultra‐high vacuum chamber, and they were self‐assembled on the surface of graphene. To the best of our knowledge, the sensors reported here are the first
d
conductometric sensors based on graphene and Au nanoclusters that are utilized for detection
Ac ce pt e
of Hg2+ traces in water. These sensors were exceedingly sensitive to Hg2+ ions, therefore, they have potential to be applied for practical life applications.
Experimental:
Gold nanoclusters were produced by the sputtering and inert gas condensation technique inside an ultra‐high vacuum system as shown in Fig.1 [26]. The system consists of three main chambers (source, mass filter, and deposition chambers) that were pumped initially to a base pressure of 10‐8 mbar using two turbo pumps. A gold target of purity 99.99% (Testbourne ltd, UK) was fixed on a water‐cooled magnetron sputter head. The sputter head was fixed on a motorized linear translator to enable changing the aggregation length (defined as the distance from the surface of the sputtering target to the exit nozzle of the source chamber). Plasma was generated inside
4
Page 4 of 16
the source chamber using argon (Ar) inert gas, and it was used to sputter Au from its target through dc discharge type. The supplied Ar gas was also utilized to condense the sputtered material forming nanoclusters, and to create pressure gradient between the source and
ip t
deposition chambers that enables the formed nanoclusters to travel to the deposition chamber. The inert gas flow rate was controlled within the range 0.0 – 100.0 sccm using a mass flow
cr
controller (MKS Instruments). The size and yield of the produced nanoclusters were controlled
us
by varying the following source parameters: 1) argon inert gas flow rate (fAr), 2) aggregation length (L), and 3) sputtering discharge power (P). The size distribution of the produced
an
nanoclusters was measured using a quadrupole mass filter (QMF) assembled in the path of the produced nanoclusters, see Fig. 1. The mass filter consists of four metal rods, where voltages
M
(combination of dc and ac voltages) of equal magnitudes but opposite polarities were applied to each pair of opposite metal rods. The flux of size‐selected nanoclusters was measured using a
d
Farady cup (located at the exit of QMF) connected with a picoammeter (here, the electrical
Ac ce pt e
current signal represents nanocluster flux).
Each Hg2+ sensor was fabricated using (1 cm X 1cm) commercial graphene layer on SiO2/doped‐ Si substrate (thickness of SiO2 is 285 nm, and Si is p‐type with resistivity of 0.001‐0.005 ohm.cm). Interdigitated parallel Au electrodes (with electrode separation of 100 μm) were fabricated by thermal evaporation using a Torr International evaporator through a shadow mask on the surface of graphene, as shown in Fig. 2(a) [27]. Two batches of sensors were tested in this work: i) sensors based on graphene only, and ii) sensors based on graphene and percolating films of Au nanoclusters, each has a thickness of 5 nm. For nanocluster deposition, each graphene sample was fixed on a cryostat finger. Nanocluster deposition rate was initially measured using a quartz crystal monitor (QCM) facing nanocluster
5
Page 5 of 16
beam and was fixed on a motorized linear translator. Next, the QCM was removed away from the sample, thus, nanoclusters were deposited on graphene surface. Sensitivity measurements were performed using Hg2+ solutions with different concentrations
ip t
(0.05, 0.1, 0.3, 0.6, 3, 6, 20, 40, and 60 ppb). The solutions were prepared as follows: three stock solutions were prepared by dissolving 0.1000 g in 100 mL of 1.0 M HCl (stock A), 0.0100 g HgCl2
cr
in 100 mL of 1.0 M HCl (stock B) and 0.0100 g in 1000 mL of 1.0 M HCl (stock C). The initial
us
concentrations for these three solutions were 1000 ppm, 100 ppm and 100 ppm, respectively. Volumetric flasks, Class A, with a volume capacity of 100 mL, were used in all the preparation
an
steps.
Solution A was used to prepare final concentrations of 60, 40 and 20 ppb by serial dilution. First,
M
6.00 mL of solution A was diluted by 100 mL of 1.0 M HCl to get 60.00 ppm. After that, 1.000 mL of the 60.00 ppm was diluted by 100.0 1.0 M HCl to give 0.60 ppm (600 ppb). Finally, 10.00 mL,
d
6.666 mL, and 3.333 mL volumes of 600 ppb was diluted in three separate volumetric flasks by
Ac ce pt e
100 mL of 1.0 M HCl in each to make the final concentrations of 60, 40, and 20 ppb solutions, respectively.
Solution B, which is 100.0 ppm HgCl2, was used to prepare the final concentrations 6.0 ppb and 3.0 ppb. First, 6.00 mL of solution B was diluted by 100 mL of 1.0 M HCl to get the concentration of 6.0 ppm. After that, 1.000 mL of the 6.0 ppm was diluted by 100 mL 1.0 M HCl to give 0.06 ppm (60.0 ppb). Finally, 10.00 mL, and 5.0 mL volumes of 60.0 ppb was added to 2 volumetric flasks of 100 volumes and diluted to the mark by 1.0 M HCl to make the 6.0 and 3.0 ppb solutions, respectively.
Solution C, which is 100 ppm HgCl2, was used to prepare the last four solutions (0.6, 0.3, 0.1 and 0.05) ppb. First, 6.000 mL of solution C was diluted by 100 mL of 1.0 M HCl to get 6.0 ppm. The 6.00 ppm solution was diluted to give 0.06 ppm (60.0 ppb) by diluting 1.000 mL in 100 mL 1.0 M
6
Page 6 of 16
HCl. Finally, 1.000 mL and 0.500 mL of the 60 ppb solution was diluted by 100mL 1.0 M HCl, using volumetric flask, in order to get 0.6 ppb and 0.5 ppb, respectively. Another 1.000 mL of solution C was diluted in 100 mL of 1.0 M HCl in order to prepare 1.00 ppm
ip t
solution. The 1.00 ppm solution was diluted to give 0.01 ppm (10.0 ppb) by diluting 1.000 mL in 100.0 mL of 1.0 M HCl. Finally, 1.000 mL and 0.500 mL of the 10.0 ppb solution was diluted by
cr
100.0 mL of 1.0 M HCl in order to get 0.10 ppb and 0.05 ppb, respectively.
The selectivity was tested using 0.6 ppb solutions of the following ions: Cr2+, Cd2+, Cu2+, Co2+,
us
Fe2+, Zn2+, and K+. The concentrations of the prepared solutions of Hg2+ and other ions were
an
confirmed using inductively coupled plasma mass spectrometry (ICP‐MS). Electrical measurements of Hg2+ sensors were performed using a Keithley Instruments source
M
measurement unit (KI 236). For sensitivity measurements, a constant voltage of 0.001 V was applied to the sensor, and electrical current signal was monitored as a function of time.
Ac ce pt e
Results and Discussion:
d
Nanoclusters and graphene were imaged using an atomic force microscope (AFM).
The fabricated sensors follow a field‐effect transistor (FET) structure, where graphene layer is the channel material while the p‐type Si substrate is a back gate. The source and drain electrodes of the sensor are shown in Fig.2(a). Figure 2(b) shows an AFM image of graphene that reveals homogeneous and continuous layer. The figure also shows the SiO2 substrate, where a step was used to measure the height of graphene. The height profile of graphene is shown in Fig.2(c), and it reveals a height of 0.8 nm which is consistent with previous publications [28]. The current – voltage (Isd(Vsd)) characteristics as a function of gate voltage (Vg) of the present FET at room temperature are shown in Fig. 3. This figure reveals linear source – drain current within the range of applied Vsd voltage. Increasing the gate voltage causes the current to increase.
7
Page 7 of 16
These characteristics are typical for n‐type channel FET based on graphene [29, 30]. In addition, the characteristics do not reveal current saturation since graphene is a semiconductor with zero‐ gap [31].
ip t
For Hg2+ sensors with Au nanoclusters, the nanoclusters were synthesized using P = 21 W, fAr = 40 sccm, and L = 50 and 60 mm. The size distribution as measured using QMF is shown in Fig.
cr
4(a). The average nanocluster sizes were calculated to 4.67 ± 0.23 nm and 6.82 ± 0.30 nm for
us
aggregation lengths of 50 and 60 mm, respectively. Nanoclusters produced using sputtering and inert gas condensation technique are produced inside the nanocluster source chamber mainly
an
by two mechanisms: three‐body collision, and two‐body collision. Herein, both mechanisms contribute to the formation of nanoclusters. In the former mechanism, nanoclusters are
M
produced due collision between two sputtered atoms and an argon atom to remove excess heat. This process is essential to produce nanocluster seeds. In the later mechanism, two
d
nanoclusters or a nanocluster and an atom collide to produce larger nanoclusters. Therefore,
Ac ce pt e
the two‐body mechanism contribute to further growth in nanocluster size. The increase in nanocluster size with increasing aggregation length is a result of the increase in the aggregation time that increases the number of collisions, thus, nanocluster grow further [14]. In this work, we used Au nanoclusters with an average size of 6.82 ± 0.30 nm. Figure 4(b) demonstrates an AFM image of Au nanoclusters utilized in this work. The image illustrates a percolating film of homogeneous nanoclusters, and their size is consistent with that determined from QMF measurements.
Figure 5 shows the change of conductance of the fabricated Hg2+ sensors when exposed to a 0.5 mL droplet of 0.05 ppb Hg2+ solution. The figure reveals reasonable variation in the electrical current when exposed to Hg2+. The response of the present sensor for Hg2+ stimulant is nearly
8
Page 8 of 16
instantaneous. In addition, the variation in electrical current for sensors with Au nanoclusters is one order of magnitude higher than those based on graphene only. The decrease in the electrical current of the sensor when exposed to Hg2+ can be assigned to the
ip t
adsorption of Hg2+ on surface of the sensor. Herein, the accumulation of positive charges from Hg2+ ions decreases electron concentration in n‐type graphene and increases holes
electrical
cr
concentration which in turn decreases the electrical current. The more significant decrease in
current for sensors with Au nanoclusters can be assigned to additional
us
adsorption of Hg2+ on Au nanoclusters due to their high affinity to Au. The results demonstrate
an
the suitability of using Au nanoclusters to make the graphene sensors selective to Hg2+ ions. Herein, the choice of Au nanoclusters was made because of their high binding energy (Ebind) and
M
adsorption enthalpy (ΔH) to Hg compared with other novel metals as reported previously: Ebind = 56.3 kJ/mol, and ΔH = ‐97.9 kJ/mol [32]. The increase in the sensitivity of the sensor to Hg2+ ions
d
upon decoration with Au nanoclusters can be explained by the increase in electron diffusive
Ac ce pt e
scattering at the surface of nanoclusters because of the presence of Hg2+ ions. Reducing the dimensions of a material below the electron mean free path implies that electrical conduction will be dominated by surface scattering. Hither, the presence of defects or adsorbates normally act as scattering centers which causes inelastic (diffusive) electron scattering, thus, the fraction of electrons that are diffusively scattered increase [33]. Therefore, the adsorption Hg2+ ions on Au nanoclusters leads to a further increase in the electrical resistance. As a result, the deposition of Au nanoclusters on the surface of graphene enhances the adsorption of Hg2+ ions and the sensitivity of the sensor rather than non‐specific adsorption of the ions on graphene [34]. The dependence of the absolute value of the change in electrical current signal (|ΔI|) on Hg2+ concentration is shown in Fig. 6. This figure demonstrates that increasing the concentration of
9
Page 9 of 16
Hg2+ decreases |ΔI|. This behavior has been observed before in [34], and can be explained as result of the increase in the p‐doping effect because of the absorption of Hg2+ on the channel, i.e. nanoclusters/graphene [35]. The value of |ΔI| for sensors with Au nanoclusters is one to two
ip t
orders of magnitudes high than those with graphene only. Furthermore, the detection limit of the present sensors is 0.05 ppb which is below the limit set by the World Health Organization
us
sensors have the potential to be used for practical applications.
cr
(WHO) and that of the United States Environmental Protection Agency (USEPA), thus, our
The selectivity of the sensor was evaluated by exposing the sensor to solutions containing
an
different metal ions that are normally found in natural water resources (Cr2+, Cd2+, Cu2+, Co2+, Fe3+, Zn2+, and K+) with a fixed concentration of 0.6ppb. Figure 7 shows sensor signal for different
M
metal ions. The figure reveals that the fabricated sensors are selective to Hg2+ ions, and the selectivity is evidently enhanced for sensors with Au nanoclusters. It should be noted here that
d
for K+ ions, the signal of the sensor based on graphene and Au nanoclusters is too low thus it
Ac ce pt e
does not appear in the figure. The sensitivity of sensors decorated with Au nanoclusters to Hg2+ ions is higher than that for sensors made of graphene only, which could be assigned to the high binding affinity of Au nanoclusters to Hg2+ ions [33, 36]. This can be explained qualitatively bearing in mind previous studies that investigated the binding energy of the metal ions (under consideration in this work) with either graphene or Au nanoclusters. Upon investigating the binding energy of metal ions with graphene, it was reported that Hg2+ ions have the highest binding energy to graphene compared to other ions (Cr2+, Cd2+, Cu2+, Co2+, Fe3+, Zn2+, and K+), which exhibit slightly lower binding energy to graphene and thus their sensitivity signals are very comparative [36‐39].The sensitivity of Au nanostructures to Hg2+ ions was found to be large due to their high binding energy [33]. In addition, the binding energy of Hg2+ ions to Au
10
Page 10 of 16
nanostructures is evidently higher than that to the other metal ions investigated in this work [40‐43], which makes the graphene‐Au sensor selective to Hg2+ ions.
ip t
Conclusion: We presented in this work conductometric sensors based on graphene and gold nanoclusters
cr
that can be utilized for detection of mercury ions (Hg2+). Herein, the sensor is constructed of graphene on a SiO2/doped‐Si substrate. Interdigitated electrodes were deposited on the surface
us
of graphene followed by self‐assembly of gold nanocluster percolating film. The nanoclusters were fabricated by sputtering and inert‐gas condensation technique inside an ultra‐high vacuum
an
compatible system, which could be utilized by industry. Those nanoclusters are of high purity
M
and narrow size distribution. The sensors were highly sensitive for Hg2+ (0.05 ppb), and extraordinarily selective. The sensing mechanism of the graphene‐based sensors can be
d
summarized as follows: exposure of the graphene sensor to Hg2+ ions decreases electron
Ac ce pt e
concentration in the n‐type graphene, thus, the conductance of the sensor decreases. Decoration of graphene with Au nanoclusters creates electron scattering centers that increase electron diffusive scattering which decreases electrical conductance. Adsorption of Hg2+ ions on Au nanoclusters causes further decrease in the conductance, which implies the increase in the sensitivity and selectivity of the sensor.
The sensitivity of the present sensors is below the minimum allowed limit of Hg2+ in drinking water set by the World Health Organization and that set by the United States Environmental Protection Agency. These sensors are small in size and easy to carry outdoor and have low power requirements, thus, they have a potential to be used for practical field applications.
Acknowledgments: This work was supported by the United Arab Emirates University fund number 31R006. 11
Page 11 of 16
References:
Ac ce pt e
d
M
an
us
cr
ip t
[1] W. Chemnasiri, F.E. Hernandez. Gold nanorod‐based mercury sensor using functionalized glass substrates, Sensors and Actuators B: Chemical 173 (2012) 322‐328. [2] Y. Li, H. Huang, Y. Li, X. Su. Highly sensitive fluorescent sensor for mercury (II) ion based on layer‐by‐layer self‐assembled films fabricated with water‐soluble fluorescent conjugated polymer, Sensors and Actuators B: Chemical 188 (2013) 772‐777. [3] B. Wang, S. Zhuo, L. Chen, Y. Zhang. Fluorescent graphene quantum dot nanoprobes for the sensitive and selective detection of mercury ions, Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 131 (2014) 384‐387. [4] WHO. Guidelines for drinking‐water quality. Fourth edition ed., World Health Organization, 2011. [5] E. U.S. Drinking Water Contaminants. http://water.epa.gov/drink/contaminants/index.cfm#list, 2011. [6] M.K. Kim, K. Shanmuga Sundaram, G. Anantha Iyengar, K.‐P. Lee. A novel chitosan functional gel included with multiwall carbon nanotube and substituted polyaniline as adsorbent for efficient removal of chromium ion, Chemical Engineering Journal 267 (2015) 51‐64. [7] H. Okyere, R.B. Voegborlo, S.E. Agorku. Human exposure to mercury, lead and cadmium through consumption of canned mackerel, tuna, pilchard and sardine, Food Chemistry 179 (2015) 331‐335. [8] M.P. Rodríguez‐Reino, R. Rodríguez‐Fernández, E. Peña‐Vázquez, R. Domínguez‐ González, P. Bermejo‐Barrera, A. Moreda‐Piñeiro. Mercury speciation in seawater by liquid chromatography‐inductively coupled plasma‐mass spectrometry following solid phase extraction pre‐concentration by using an ionic imprinted polymer based on methyl‐mercury– phenobarbital interaction, Journal of Chromatography A. [9] N. Aksuner, B. Basaran, E. Henden, I. Yilmaz, A. Cukurovali. A sensitive and selective fluorescent sensor for the determination of mercury(II) based on a novel triazine‐thione derivative, Dyes and Pigments 88 (2011) 143‐148. [10] W. Chansuvarn, T. Tuntulani, A. Imyim. Colorimetric detection of mercury(II) based on gold nanoparticles, fluorescent gold nanoclusters and other gold‐based nanomaterials, TrAC Trends in Analytical Chemistry 65 (2015) 83‐96. [11] A.I. Ayesh, S.T. Mahmoud, S.J. Ahmad, Y. Haik. Novel hydrogen gas sensor based on Pd and SnO2 nanoclusters, Materials Letters 128 (2014) 354‐357. [12] J.v. Lith, A. Lassesson, S.A. Brown, M. Schulze, J.G. Partridge, A. Ayesh. A hydrogen sensor based on tunnelling between palladium clusters, Appl. Phys. Lett. 91 (2007) 181910. [13] A.I. Ayesh. Gas Sensor Applications using Atomic Nanoclusters. in: Naveen Kumar Navani SS, and J.N. Govil, (Ed.). Nanosensing, vol. 10. Studium Press LLC, Houston, TX 77072, USA, 2012. [14] A.I. Ayesh, N. Qamhieh, H. Ghamlouche, a.M.E.‐S. S. Thaker. Fabrication of size‐selected Pd nanoclusters using a magnetron plasma sputtering source, J. Appl. Phys. 107 (2010) 034317 [15] A.I. Ayesh, S. Thaker, N. Qamhieh, H. Ghamlouche. Size‐controlled Pd nanocluster grown by plasma gas‐condensation method, J. Nanopart. Res. 13 (2011) 1125. [16] M. Penza, R. Rossi, M. Alvisi, G. Cassano, E. Serra. Functional characterization of carbon nanotube networked films functionalized with tuned loading of Au nanoclusters for gas sensing applications, Sensors and Actuators B: Chemical 140 (2009) 176‐184. [17] J.M. Pingarrón, P. Yáñez‐Sedeño, A. González‐Cortés. Gold nanoparticle‐based electrochemical biosensors, Electrochimica Acta 53 (2008) 5848‐5866.
12
Page 12 of 16
Ac ce pt e
d
M
an
us
cr
ip t
[18] T.N. Huan, L.Q. Hung, V.T.T. Ha, N.H. Anh, T. Van Khai, K.B. Shim, H. Chung. Spirally oriented Au microelectrode array sensor for detection of Hg (II), Talanta 94 (2012) 284‐288. [19] D. Martín‐Yerga, M.B. González‐García, A. Costa‐García. Use of nanohybrid materials as electrochemical transducers for mercury sensors, Sensors and Actuators B: Chemical 165 (2012) 143‐150. [20] K.‐I. Chen, B.‐R. Li, Y.‐T. Chen. Silicon nanowire field‐effect transistor‐based biosensors for biomedical diagnosis and cellular recording investigation, Nano Today 6 (2011) 131‐154. [21] K. Novoselov, S. Morozov, D. Jiang, Y. Zhang, S. Dubonos, I. Grigorieva, A. Firsov. Electric field effect in atomically thin carbon films, Science 306 (2004) 666‐669. [22] Y.H. Kwak, D.S. Choi, Y.N. Kim, H. Kim, D.H. Yoon, S.‐S. Ahn, J.‐W. Yang, W.S. Yang, S. Seo. Flexible glucose sensor using CVD‐grown graphene‐based field effect transistor, Biosensors and Bioelectronics 37 (2012) 82‐87. [23] G. Aragay, A. Merkoçi. Nanomaterials application in electrochemical detection of heavy metals, Electrochimica Acta 84 (2012) 49‐61. [24] R. Pearce, T. Iakimov, M. Andersson, L. Hultman, A.L. Spetz, R. Yakimova. Epitaxially grown graphene based gas sensors for ultra sensitive NO2 detection, Sensors and Actuators B: Chemical 155 (2011) 451‐455. [25] D. Bradley. Graphene gas sensor: Carbon, Materials Today 15 (2012) 233. [26] A.I. Ayesh, N. Qamhieh, S.T. Mahmoud, H. Alawadhi. Fabrication of size‐selected bimetallic nanoclusters using magnetron sputtering, Journal of Materials Research 27 (2012). [27] A.I. Ayesh. Electronic transport in Pd nanocluster devices, Applied Physics Letters 98 (2011) 133108. [28] P. Nemes‐Incze, Z. Osváth, K. Kamarás, L.P. Biró. Anomalies in thickness measurements of graphene and few layer graphite crystals by tapping mode atomic force microscopy, Carbon 46 (2008) 1435‐1442. [29] F. Schwierz. Graphene transistors, Nature Nanotechnology 5 (2010). [30] S.J. Park, O.S. Kwon, S.H. Lee, H.S. Song, T.H. Park, J. Jang. Ultrasensitive Flexible Graphene Based Field‐Effect Transistor (FET)‐ Type Bioelectronic Nose, Nano Letters 12 (2012) 5082−5090 [31] Y.‐M. Lin, K.A. Jenkins, A. Valdes‐Garcia, J.P. Small, D.B. Farmer, P. Avouris. Operation of Graphene Transistors at Gigahertz Frequencies, Nano Letters 9 (2009) 422‐426. [32] V. Pershina, T. Bastug, T. Jacob, B. Fricke, S. Varga. Intermetallic compounds of the heaviest elements: the electronic structure and bonding of dimers of element 112 and its homolog Hg, Chemical Physics Letters 365 (2002) 176‐183. [33] Z. Liu, P.C. Searson. Single Nanoporous Gold Nanowire Sensors, The Journal of Physical Chemistry B 110 (2006) 4318‐4322. [34] Z. Yin, Q. He, X. Huang, J. Zhang, S. Wu, P. Chen, G. Lu, P. Chen, Q. Zhang, Q. Yan, H. Zhang. Real‐time DNA detection using Pt nanoparticle‐decorated reduced graphene oxide field‐ effect transistors, Nanoscale 4 (2012). [35] X. Dong, Y. Shi, W. Huang, P. Chen, L.J. Li. Electrical Detection of DNA Hybridization with Single‐Base Specificity Using Transistors Based on CVD‐Grown Graphene Sheets, Advanced Materials 22 (2010) 1649–1653. [36] C. Yu, Y. Guo, H. Liu, N. Yan, Z. Xu, G. Yu, Y. Fang, Y. Liu. Ultrasensitive and selective sensing of heavy metal ions with modified graphene, Chemical Communications 49 (2013) 6492—6494. [37] D. Umadevi, G.N. Sastry. Metal ion binding with carbon nanotubes and graphene: Effect of chirality and curvature, Chemical Physics Letters 549 (2012) 39‐43.
13
Page 13 of 16
Ac ce pt e
d
M
an
us
cr
ip t
[38] W. Su, M. Cho, J.‐D. Nam, W.‐S. Choe, Y. Lee. Highly sensitive electrochemical lead ion sensor harnessing peptide probe molecules on porous gold electrodes, Biosensors and Bioelectronics 48 (2013) 263‐269. [39] D. Gu, J.B. Fein. Adsorption of metals onto graphene oxide: Surface complexation modeling and linear free energy relationships, Colloids and Surfaces A: Physicochemical and Engineering Aspects. [40] S.A. Siddiqui, N. Bouarissa, T. Rasheed, M.S. Al‐Assiri. Quantum chemical study of the interaction of elemental Hg with small neutral, anionic and cationic Aun (n = 1–6) clusters, Materials Research Bulletin 48 (2013) 995‐1002. [41] K.W. Frese Jr. The reorganization energy of the aqueous proton, Journal of Electroanalytical Chemistry and Interfacial Electrochemistry 249 (1988) 15‐24. [42] M.H. Bradbury, B. Baeyens. Modelling the sorption of Mn(II), Co(II), Ni(II), Zn(II), Cd(II), Eu(III), Am(III), Sn(IV), Th(IV), Np(V) and U(VI) on montmorillonite: Linear free energy relationships and estimates of surface binding constants for some selected heavy metals and actinides, Geochimica et Cosmochimica Acta 69 (2005) 875‐892. [43] D. Sareen, P. Kaur, K. Singh. Strategies in detection of metal ions using dyes, Coordination Chemistry Reviews 265 (2014) 125‐154.
14
Page 14 of 16
Figure Captions: Fig. 1: Schematic diagram of the ultra‐high vacuum system and nanocluster deposition.
ip t
Fig. 2: (a) Image of a graphene based sensor and electrode structure. (b) AFM image of graphene
layer. The line indicates the location where the height profile was measured. (c) Height profile of
cr
graphene and SiO2 substrate as measured using the AFM.
us
Fig. 3: Current‐voltage characteristics as a function of gate voltage for graphene sensors. Fig. 4: (a) Size distribution of Au nanoclusters as measured by the QMF. The nanoclusters were
an
produced using P = 21 W and fAr = 40 sccm. (b) AFM image of the produced Au nanoclusters with L = 60mm.
M
Fig. 5: Electrical current response signal of a graphene based sensor for DI water with 0.05 ppb Hg2+. (a) Sensor with graphene only, and (b) sensor with graphene and Au nanoclusters.
d
Fig. 6: Response signal of Hg2+ as a function of concentration in DI water for sensors based on
Ac ce pt e
either graphene only or graphene and Au nanoclusters. Fig. 7: Sensitivity histogram of the graphene based sensors for 0.6 ppb solutions of Cr2+, Cd2+, Cu2+, Co2+, Fe3+, Zn2+, and K+. 15
Page 15 of 16
Ac ce pt e
d
M
an
us
cr
ip t
I) Dr. Ahmad I. Ayesh Associate Professor, Physics of Nanomaterials, Physics Program ‐ Department of Mathematics, Statistics and Physics, Qatar University, Doha, Qatar Dr. Ayesh is the leader of the nanocluster devices group at United Arab Emirates University. He is expert in the nano‐ and micro‐ device fabrication using both the top‐down and bottom‐up approaches as well as the self‐ assembly of the nanostructure within the device. Furthermore, he is expert in nanomaterial fabrication and characterization. II) Miss. Zainab Karam MSC student at Masdar Institute, Abu Dhabi, UAE III) Dr. Falah Awwad Associate Professor, Department of Electrical Engineering, United Arab Emirates University, Al Ain, United Arab Emirates Dr. Falah Awwad received the M.A.Sc. and Ph.D. degrees in Electrical and Computer Engineering from Concordia University (Montreal, QC, Canada) in 2002 and 2006, respectively. He was a Post‐Doctoral Fellow at Ecole Polytechnique de Montréal and Concordia University, Montreal, QC, Canada. Between August 2007 – Feb. 2013, he was an Assistant Professor with the College of Information Technology (CIT) at United Arab Emirates University. Currently, he is an Associate Professor with the Department of Electrical Engineering – College of Engineering (UAE University). He is a member of the editorial Board of Journal of “Nanomaterials & Molecular Nanotechnology” and "Austin Journal of Nanomedicine & Nanotechnology". He is the recipient of the CIT Faculty Outstanding Service Award. The evaluation was done based on his professional and community service during the period 2009 ‐ 2012. His scientific research interests are Low‐Power High‐Speed VLSI Systems including repeater‐insertion methodologies for on‐chip interconnects, Biomedical Imaging Systems, Network on Chip, Wireless Body Sensors/Sensor Networks, and Network Communication algorithms. IV) Dr. Mohammed A. Meetani Associate Professor, Chemistry Department, United Arab Emirates University, Al Ain, United Arab Emirates. Dr. Meetani received his PhD in Applied Analytical Chemistry from Colorado School of Mines in 2003. He is expert in the following fields of research: Biological mass spectrometry, Degradation studies of environmental organic pollutants, Peptides/ protein sequencing, Pyrolysis mass spectrometry, Studying the structural properties of cellulose nanofibrous membranes, development of chromatographic methods for chemical analysis
16
Page 16 of 16
S0925-4005(15)30001-0 http://dx.doi.org/doi:10.1016/j.snb.2015.06.075 SNB 18652
To appear in:
Sensors and Actuators B
Received date: Revised date: Accepted date:
18-4-2015 10-6-2015 19-6-2015
Please cite this article as: A.I. Ayesh, Z. Karam, F. Awwad, M.A. Meetani, Conductometric graphene sensors decorated with nanoclusters for selective detection of Hg2+ traces in water, Sensors and Actuators B: Chemical (2015), http://dx.doi.org/10.1016/j.snb.2015.06.075 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Conductometric graphene sensors decorated with nanoclusters for selective detection of Hg2+ traces in water Ahmad I. Ayesh1,*, Zainab Karam2, Falah Awwad3, and Mohammed A. Meetani4 Department of Math., Stat. and Physics, Qatar University, Doha, Qatar 2
Department of Electrical Engineering, United Arab Emirates University, Al Ain,
cr
3
Masdar Institute, Abu Dhabi, United Arab Emirates
ip t
1
4
us
United Arab Emirates
Chemistry Department, United Arab Emirates University, Al Ain,
an
United Arab Emirates
* Corresponding author. Email: [email protected], Tel.: +974‐4403‐6592, P. O. Box 2713, Doha,
Abstract:
M
Qatar
d
Mercury ion (Hg2+) sensors based on graphene and gold nanoclusters are presented in this work.
Ac ce pt e
The sensors follow the structure of field‐effect transistor (FET), and their sensing mechanism rely on the change in conductance of graphene and gold nanocluster percolating film. Each sensor was fabricated by deposition of interdigitated metal electrodes on the surface of graphene by thermal evaporation through a shadow mask. Next, gold nanoclusters are generated by magnetron discharge sputtering and inert‐gas condensation inside an ultra‐high compatible system, and they are self‐assembled on the surface of graphene. The sensitivity and selectivity of graphene to Hg2+ are vastly enhanced by incorporating the nanoclusters on its surface. The detection limit of the present sensors is below the safe concentration of Hg2+ in drinking water set by different universal agencies. In addition, those sensors are practical and easy to operate in field for real life applications. 1
Page 1 of 16
Keywords: Graphene, gold nanoclusters, mercury sensor, inert‐gas condensation
ip t
Introduction:
cr
Mercury metal is released into water by different sources including sewage industrial waste, thus, it can enter human food chain. It is considered one of the most harmful pollutant of heavy
us
metals since it is non‐biodegradable, and it can enter to human body by different means including direct consumption (for example, through drinking water), absorption through skin,
an
and by respiratory system. Exposure to mercury cause severe effects on human health such as brain damage, kidney failure, damage in the nervous system, birth defects, chromosome
M
breakage, and paralysis [1, 2]. The United Nation Environment Program (UNEP) assessed the annual released quantity of mercury to 4400 – 7500 tons [3]. In addition, the International
Ac ce pt e
d
World Health Organization regulated the maximum allowed amount of mercury ions (Hg2+) in drinking water to 6 ppb [4]. Nevertheless, previous studies estimated the allowed amount of inorganic Hg2+ in water to 0.5 ppb [5]. Therefore, the development of sensitive, selective, reliable, and cost effective Hg2+ sensors is needed for medical diagnostic, quality control of food industry, as well as water and environment quality monitoring. In the past years different approaches were involved in detection of low concentrations of Hg2+ such as X‐ray absorption spectroscopy [6], spectrophotometry [7], inductively coupled plasma mass spectrometry [8], cold vapor atomic fluorescent spectroscopy [9], and fluorescence assay [10]. However, the previous methods are expensive, and not practical to apply in field due to their delicate instrumentation.
2
Page 2 of 16
Recently our group has been working on developing conductometric sensors based on nanostructures for different stimulate species [11‐13]. These sensors are very sensitive, cheap to manufacture, exhibit low power consumption, and practical to use in outdoor field. The main
ip t
element of those sensors is atomic nanoclusters that are fabricated by sputtering and inert condensation technique inside an ultra‐high compatible system [14]. This technique has many
cr
advantages over other nanocluster syntheses methods where the produced nanoclusters are of
us
high purity since they are prepared inside a vacuum chamber; charged, thus their size can be selected using a suitable mass filter; size tunable by controlling source parameters such as
an
sputtering discharge and inert‐gas condensation; and they can be assembled directly on a substrate without any extra experimental steps [15]. In addition, sputtering is a well‐established
M
technique that is adopted by many industries.
Generally, nanoclusters have unique properties that make them suitable for sensing applications
d
such as the large ratio of surface atoms that are normally unsaturated, thus, they can bind with
Ac ce pt e
other atoms that have suitable chemical activity. A main track of research in this field is to optimize the sensitivity, selectivity, and reproducibility of the fabricated nanoclusters. Among those nanoclusters, gold (Au) was found an attractive material for sensor and catalyst applications [16]. Au nanoclusters were used for different sensor applications [17], including Hg2+ sensor due to their high affinity to mercury [18]. However, the selectivity of Au nanocluster based sensors to Hg2+ ions is still a major challenge that should be addressed to enable commercial production of those sensors [19]. In addition, sensors based on nanoclusters and nanowires are limited because their detection depends on the induced field‐effect [20]. Graphene is a superior nanomaterial that has been experimentally produced in 2004 [21]. It exhibits high electrical conductivity, chemically stable, and it holds unique mechanical and thermal properties that attracted strong scientific interest recently. It consists of carbon atoms
3
Page 3 of 16
that are packed into a lattice of single layer with the structure of honeycomb. Graphene has been used recently for innovative sensor applications to enable the detection of a wide range of stimulant, because of its extraordinary electrical properties, such as glucose [22], heavy metals
ip t
[23], and different gas species [24, 25]. In addition, graphene is predicted to exceed carbon nanotubes in the field of sensor applications because of its large detection area as well as its
cr
biocompatibility. Nevertheless, the selectivity of graphene based sensors needs to be
us
engineered to make it suitable to target particular chemical stimulate.
In this work, we present novel conductometric sensors based on graphene and Au nanoclusters
an
that are highly selective to Hg2+ ions. Electrical electrodes were deposited on the surface of graphene by thermal evaporation. Au nanoclusters were produced by sputtering and inert gas
M
condensation technique inside an ultra‐high vacuum chamber, and they were self‐assembled on the surface of graphene. To the best of our knowledge, the sensors reported here are the first
d
conductometric sensors based on graphene and Au nanoclusters that are utilized for detection
Ac ce pt e
of Hg2+ traces in water. These sensors were exceedingly sensitive to Hg2+ ions, therefore, they have potential to be applied for practical life applications.
Experimental:
Gold nanoclusters were produced by the sputtering and inert gas condensation technique inside an ultra‐high vacuum system as shown in Fig.1 [26]. The system consists of three main chambers (source, mass filter, and deposition chambers) that were pumped initially to a base pressure of 10‐8 mbar using two turbo pumps. A gold target of purity 99.99% (Testbourne ltd, UK) was fixed on a water‐cooled magnetron sputter head. The sputter head was fixed on a motorized linear translator to enable changing the aggregation length (defined as the distance from the surface of the sputtering target to the exit nozzle of the source chamber). Plasma was generated inside
4
Page 4 of 16
the source chamber using argon (Ar) inert gas, and it was used to sputter Au from its target through dc discharge type. The supplied Ar gas was also utilized to condense the sputtered material forming nanoclusters, and to create pressure gradient between the source and
ip t
deposition chambers that enables the formed nanoclusters to travel to the deposition chamber. The inert gas flow rate was controlled within the range 0.0 – 100.0 sccm using a mass flow
cr
controller (MKS Instruments). The size and yield of the produced nanoclusters were controlled
us
by varying the following source parameters: 1) argon inert gas flow rate (fAr), 2) aggregation length (L), and 3) sputtering discharge power (P). The size distribution of the produced
an
nanoclusters was measured using a quadrupole mass filter (QMF) assembled in the path of the produced nanoclusters, see Fig. 1. The mass filter consists of four metal rods, where voltages
M
(combination of dc and ac voltages) of equal magnitudes but opposite polarities were applied to each pair of opposite metal rods. The flux of size‐selected nanoclusters was measured using a
d
Farady cup (located at the exit of QMF) connected with a picoammeter (here, the electrical
Ac ce pt e
current signal represents nanocluster flux).
Each Hg2+ sensor was fabricated using (1 cm X 1cm) commercial graphene layer on SiO2/doped‐ Si substrate (thickness of SiO2 is 285 nm, and Si is p‐type with resistivity of 0.001‐0.005 ohm.cm). Interdigitated parallel Au electrodes (with electrode separation of 100 μm) were fabricated by thermal evaporation using a Torr International evaporator through a shadow mask on the surface of graphene, as shown in Fig. 2(a) [27]. Two batches of sensors were tested in this work: i) sensors based on graphene only, and ii) sensors based on graphene and percolating films of Au nanoclusters, each has a thickness of 5 nm. For nanocluster deposition, each graphene sample was fixed on a cryostat finger. Nanocluster deposition rate was initially measured using a quartz crystal monitor (QCM) facing nanocluster
5
Page 5 of 16
beam and was fixed on a motorized linear translator. Next, the QCM was removed away from the sample, thus, nanoclusters were deposited on graphene surface. Sensitivity measurements were performed using Hg2+ solutions with different concentrations
ip t
(0.05, 0.1, 0.3, 0.6, 3, 6, 20, 40, and 60 ppb). The solutions were prepared as follows: three stock solutions were prepared by dissolving 0.1000 g in 100 mL of 1.0 M HCl (stock A), 0.0100 g HgCl2
cr
in 100 mL of 1.0 M HCl (stock B) and 0.0100 g in 1000 mL of 1.0 M HCl (stock C). The initial
us
concentrations for these three solutions were 1000 ppm, 100 ppm and 100 ppm, respectively. Volumetric flasks, Class A, with a volume capacity of 100 mL, were used in all the preparation
an
steps.
Solution A was used to prepare final concentrations of 60, 40 and 20 ppb by serial dilution. First,
M
6.00 mL of solution A was diluted by 100 mL of 1.0 M HCl to get 60.00 ppm. After that, 1.000 mL of the 60.00 ppm was diluted by 100.0 1.0 M HCl to give 0.60 ppm (600 ppb). Finally, 10.00 mL,
d
6.666 mL, and 3.333 mL volumes of 600 ppb was diluted in three separate volumetric flasks by
Ac ce pt e
100 mL of 1.0 M HCl in each to make the final concentrations of 60, 40, and 20 ppb solutions, respectively.
Solution B, which is 100.0 ppm HgCl2, was used to prepare the final concentrations 6.0 ppb and 3.0 ppb. First, 6.00 mL of solution B was diluted by 100 mL of 1.0 M HCl to get the concentration of 6.0 ppm. After that, 1.000 mL of the 6.0 ppm was diluted by 100 mL 1.0 M HCl to give 0.06 ppm (60.0 ppb). Finally, 10.00 mL, and 5.0 mL volumes of 60.0 ppb was added to 2 volumetric flasks of 100 volumes and diluted to the mark by 1.0 M HCl to make the 6.0 and 3.0 ppb solutions, respectively.
Solution C, which is 100 ppm HgCl2, was used to prepare the last four solutions (0.6, 0.3, 0.1 and 0.05) ppb. First, 6.000 mL of solution C was diluted by 100 mL of 1.0 M HCl to get 6.0 ppm. The 6.00 ppm solution was diluted to give 0.06 ppm (60.0 ppb) by diluting 1.000 mL in 100 mL 1.0 M
6
Page 6 of 16
HCl. Finally, 1.000 mL and 0.500 mL of the 60 ppb solution was diluted by 100mL 1.0 M HCl, using volumetric flask, in order to get 0.6 ppb and 0.5 ppb, respectively. Another 1.000 mL of solution C was diluted in 100 mL of 1.0 M HCl in order to prepare 1.00 ppm
ip t
solution. The 1.00 ppm solution was diluted to give 0.01 ppm (10.0 ppb) by diluting 1.000 mL in 100.0 mL of 1.0 M HCl. Finally, 1.000 mL and 0.500 mL of the 10.0 ppb solution was diluted by
cr
100.0 mL of 1.0 M HCl in order to get 0.10 ppb and 0.05 ppb, respectively.
The selectivity was tested using 0.6 ppb solutions of the following ions: Cr2+, Cd2+, Cu2+, Co2+,
us
Fe2+, Zn2+, and K+. The concentrations of the prepared solutions of Hg2+ and other ions were
an
confirmed using inductively coupled plasma mass spectrometry (ICP‐MS). Electrical measurements of Hg2+ sensors were performed using a Keithley Instruments source
M
measurement unit (KI 236). For sensitivity measurements, a constant voltage of 0.001 V was applied to the sensor, and electrical current signal was monitored as a function of time.
Ac ce pt e
Results and Discussion:
d
Nanoclusters and graphene were imaged using an atomic force microscope (AFM).
The fabricated sensors follow a field‐effect transistor (FET) structure, where graphene layer is the channel material while the p‐type Si substrate is a back gate. The source and drain electrodes of the sensor are shown in Fig.2(a). Figure 2(b) shows an AFM image of graphene that reveals homogeneous and continuous layer. The figure also shows the SiO2 substrate, where a step was used to measure the height of graphene. The height profile of graphene is shown in Fig.2(c), and it reveals a height of 0.8 nm which is consistent with previous publications [28]. The current – voltage (Isd(Vsd)) characteristics as a function of gate voltage (Vg) of the present FET at room temperature are shown in Fig. 3. This figure reveals linear source – drain current within the range of applied Vsd voltage. Increasing the gate voltage causes the current to increase.
7
Page 7 of 16
These characteristics are typical for n‐type channel FET based on graphene [29, 30]. In addition, the characteristics do not reveal current saturation since graphene is a semiconductor with zero‐ gap [31].
ip t
For Hg2+ sensors with Au nanoclusters, the nanoclusters were synthesized using P = 21 W, fAr = 40 sccm, and L = 50 and 60 mm. The size distribution as measured using QMF is shown in Fig.
cr
4(a). The average nanocluster sizes were calculated to 4.67 ± 0.23 nm and 6.82 ± 0.30 nm for
us
aggregation lengths of 50 and 60 mm, respectively. Nanoclusters produced using sputtering and inert gas condensation technique are produced inside the nanocluster source chamber mainly
an
by two mechanisms: three‐body collision, and two‐body collision. Herein, both mechanisms contribute to the formation of nanoclusters. In the former mechanism, nanoclusters are
M
produced due collision between two sputtered atoms and an argon atom to remove excess heat. This process is essential to produce nanocluster seeds. In the later mechanism, two
d
nanoclusters or a nanocluster and an atom collide to produce larger nanoclusters. Therefore,
Ac ce pt e
the two‐body mechanism contribute to further growth in nanocluster size. The increase in nanocluster size with increasing aggregation length is a result of the increase in the aggregation time that increases the number of collisions, thus, nanocluster grow further [14]. In this work, we used Au nanoclusters with an average size of 6.82 ± 0.30 nm. Figure 4(b) demonstrates an AFM image of Au nanoclusters utilized in this work. The image illustrates a percolating film of homogeneous nanoclusters, and their size is consistent with that determined from QMF measurements.
Figure 5 shows the change of conductance of the fabricated Hg2+ sensors when exposed to a 0.5 mL droplet of 0.05 ppb Hg2+ solution. The figure reveals reasonable variation in the electrical current when exposed to Hg2+. The response of the present sensor for Hg2+ stimulant is nearly
8
Page 8 of 16
instantaneous. In addition, the variation in electrical current for sensors with Au nanoclusters is one order of magnitude higher than those based on graphene only. The decrease in the electrical current of the sensor when exposed to Hg2+ can be assigned to the
ip t
adsorption of Hg2+ on surface of the sensor. Herein, the accumulation of positive charges from Hg2+ ions decreases electron concentration in n‐type graphene and increases holes
electrical
cr
concentration which in turn decreases the electrical current. The more significant decrease in
current for sensors with Au nanoclusters can be assigned to additional
us
adsorption of Hg2+ on Au nanoclusters due to their high affinity to Au. The results demonstrate
an
the suitability of using Au nanoclusters to make the graphene sensors selective to Hg2+ ions. Herein, the choice of Au nanoclusters was made because of their high binding energy (Ebind) and
M
adsorption enthalpy (ΔH) to Hg compared with other novel metals as reported previously: Ebind = 56.3 kJ/mol, and ΔH = ‐97.9 kJ/mol [32]. The increase in the sensitivity of the sensor to Hg2+ ions
d
upon decoration with Au nanoclusters can be explained by the increase in electron diffusive
Ac ce pt e
scattering at the surface of nanoclusters because of the presence of Hg2+ ions. Reducing the dimensions of a material below the electron mean free path implies that electrical conduction will be dominated by surface scattering. Hither, the presence of defects or adsorbates normally act as scattering centers which causes inelastic (diffusive) electron scattering, thus, the fraction of electrons that are diffusively scattered increase [33]. Therefore, the adsorption Hg2+ ions on Au nanoclusters leads to a further increase in the electrical resistance. As a result, the deposition of Au nanoclusters on the surface of graphene enhances the adsorption of Hg2+ ions and the sensitivity of the sensor rather than non‐specific adsorption of the ions on graphene [34]. The dependence of the absolute value of the change in electrical current signal (|ΔI|) on Hg2+ concentration is shown in Fig. 6. This figure demonstrates that increasing the concentration of
9
Page 9 of 16
Hg2+ decreases |ΔI|. This behavior has been observed before in [34], and can be explained as result of the increase in the p‐doping effect because of the absorption of Hg2+ on the channel, i.e. nanoclusters/graphene [35]. The value of |ΔI| for sensors with Au nanoclusters is one to two
ip t
orders of magnitudes high than those with graphene only. Furthermore, the detection limit of the present sensors is 0.05 ppb which is below the limit set by the World Health Organization
us
sensors have the potential to be used for practical applications.
cr
(WHO) and that of the United States Environmental Protection Agency (USEPA), thus, our
The selectivity of the sensor was evaluated by exposing the sensor to solutions containing
an
different metal ions that are normally found in natural water resources (Cr2+, Cd2+, Cu2+, Co2+, Fe3+, Zn2+, and K+) with a fixed concentration of 0.6ppb. Figure 7 shows sensor signal for different
M
metal ions. The figure reveals that the fabricated sensors are selective to Hg2+ ions, and the selectivity is evidently enhanced for sensors with Au nanoclusters. It should be noted here that
d
for K+ ions, the signal of the sensor based on graphene and Au nanoclusters is too low thus it
Ac ce pt e
does not appear in the figure. The sensitivity of sensors decorated with Au nanoclusters to Hg2+ ions is higher than that for sensors made of graphene only, which could be assigned to the high binding affinity of Au nanoclusters to Hg2+ ions [33, 36]. This can be explained qualitatively bearing in mind previous studies that investigated the binding energy of the metal ions (under consideration in this work) with either graphene or Au nanoclusters. Upon investigating the binding energy of metal ions with graphene, it was reported that Hg2+ ions have the highest binding energy to graphene compared to other ions (Cr2+, Cd2+, Cu2+, Co2+, Fe3+, Zn2+, and K+), which exhibit slightly lower binding energy to graphene and thus their sensitivity signals are very comparative [36‐39].The sensitivity of Au nanostructures to Hg2+ ions was found to be large due to their high binding energy [33]. In addition, the binding energy of Hg2+ ions to Au
10
Page 10 of 16
nanostructures is evidently higher than that to the other metal ions investigated in this work [40‐43], which makes the graphene‐Au sensor selective to Hg2+ ions.
ip t
Conclusion: We presented in this work conductometric sensors based on graphene and gold nanoclusters
cr
that can be utilized for detection of mercury ions (Hg2+). Herein, the sensor is constructed of graphene on a SiO2/doped‐Si substrate. Interdigitated electrodes were deposited on the surface
us
of graphene followed by self‐assembly of gold nanocluster percolating film. The nanoclusters were fabricated by sputtering and inert‐gas condensation technique inside an ultra‐high vacuum
an
compatible system, which could be utilized by industry. Those nanoclusters are of high purity
M
and narrow size distribution. The sensors were highly sensitive for Hg2+ (0.05 ppb), and extraordinarily selective. The sensing mechanism of the graphene‐based sensors can be
d
summarized as follows: exposure of the graphene sensor to Hg2+ ions decreases electron
Ac ce pt e
concentration in the n‐type graphene, thus, the conductance of the sensor decreases. Decoration of graphene with Au nanoclusters creates electron scattering centers that increase electron diffusive scattering which decreases electrical conductance. Adsorption of Hg2+ ions on Au nanoclusters causes further decrease in the conductance, which implies the increase in the sensitivity and selectivity of the sensor.
The sensitivity of the present sensors is below the minimum allowed limit of Hg2+ in drinking water set by the World Health Organization and that set by the United States Environmental Protection Agency. These sensors are small in size and easy to carry outdoor and have low power requirements, thus, they have a potential to be used for practical field applications.
Acknowledgments: This work was supported by the United Arab Emirates University fund number 31R006. 11
Page 11 of 16
References:
Ac ce pt e
d
M
an
us
cr
ip t
[1] W. Chemnasiri, F.E. Hernandez. Gold nanorod‐based mercury sensor using functionalized glass substrates, Sensors and Actuators B: Chemical 173 (2012) 322‐328. [2] Y. Li, H. Huang, Y. Li, X. Su. Highly sensitive fluorescent sensor for mercury (II) ion based on layer‐by‐layer self‐assembled films fabricated with water‐soluble fluorescent conjugated polymer, Sensors and Actuators B: Chemical 188 (2013) 772‐777. [3] B. Wang, S. Zhuo, L. Chen, Y. Zhang. Fluorescent graphene quantum dot nanoprobes for the sensitive and selective detection of mercury ions, Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 131 (2014) 384‐387. [4] WHO. Guidelines for drinking‐water quality. Fourth edition ed., World Health Organization, 2011. [5] E. U.S. Drinking Water Contaminants. http://water.epa.gov/drink/contaminants/index.cfm#list, 2011. [6] M.K. Kim, K. Shanmuga Sundaram, G. Anantha Iyengar, K.‐P. Lee. A novel chitosan functional gel included with multiwall carbon nanotube and substituted polyaniline as adsorbent for efficient removal of chromium ion, Chemical Engineering Journal 267 (2015) 51‐64. [7] H. Okyere, R.B. Voegborlo, S.E. Agorku. Human exposure to mercury, lead and cadmium through consumption of canned mackerel, tuna, pilchard and sardine, Food Chemistry 179 (2015) 331‐335. [8] M.P. Rodríguez‐Reino, R. Rodríguez‐Fernández, E. Peña‐Vázquez, R. Domínguez‐ González, P. Bermejo‐Barrera, A. Moreda‐Piñeiro. Mercury speciation in seawater by liquid chromatography‐inductively coupled plasma‐mass spectrometry following solid phase extraction pre‐concentration by using an ionic imprinted polymer based on methyl‐mercury– phenobarbital interaction, Journal of Chromatography A. [9] N. Aksuner, B. Basaran, E. Henden, I. Yilmaz, A. Cukurovali. A sensitive and selective fluorescent sensor for the determination of mercury(II) based on a novel triazine‐thione derivative, Dyes and Pigments 88 (2011) 143‐148. [10] W. Chansuvarn, T. Tuntulani, A. Imyim. Colorimetric detection of mercury(II) based on gold nanoparticles, fluorescent gold nanoclusters and other gold‐based nanomaterials, TrAC Trends in Analytical Chemistry 65 (2015) 83‐96. [11] A.I. Ayesh, S.T. Mahmoud, S.J. Ahmad, Y. Haik. Novel hydrogen gas sensor based on Pd and SnO2 nanoclusters, Materials Letters 128 (2014) 354‐357. [12] J.v. Lith, A. Lassesson, S.A. Brown, M. Schulze, J.G. Partridge, A. Ayesh. A hydrogen sensor based on tunnelling between palladium clusters, Appl. Phys. Lett. 91 (2007) 181910. [13] A.I. Ayesh. Gas Sensor Applications using Atomic Nanoclusters. in: Naveen Kumar Navani SS, and J.N. Govil, (Ed.). Nanosensing, vol. 10. Studium Press LLC, Houston, TX 77072, USA, 2012. [14] A.I. Ayesh, N. Qamhieh, H. Ghamlouche, a.M.E.‐S. S. Thaker. Fabrication of size‐selected Pd nanoclusters using a magnetron plasma sputtering source, J. Appl. Phys. 107 (2010) 034317 [15] A.I. Ayesh, S. Thaker, N. Qamhieh, H. Ghamlouche. Size‐controlled Pd nanocluster grown by plasma gas‐condensation method, J. Nanopart. Res. 13 (2011) 1125. [16] M. Penza, R. Rossi, M. Alvisi, G. Cassano, E. Serra. Functional characterization of carbon nanotube networked films functionalized with tuned loading of Au nanoclusters for gas sensing applications, Sensors and Actuators B: Chemical 140 (2009) 176‐184. [17] J.M. Pingarrón, P. Yáñez‐Sedeño, A. González‐Cortés. Gold nanoparticle‐based electrochemical biosensors, Electrochimica Acta 53 (2008) 5848‐5866.
12
Page 12 of 16
Ac ce pt e
d
M
an
us
cr
ip t
[18] T.N. Huan, L.Q. Hung, V.T.T. Ha, N.H. Anh, T. Van Khai, K.B. Shim, H. Chung. Spirally oriented Au microelectrode array sensor for detection of Hg (II), Talanta 94 (2012) 284‐288. [19] D. Martín‐Yerga, M.B. González‐García, A. Costa‐García. Use of nanohybrid materials as electrochemical transducers for mercury sensors, Sensors and Actuators B: Chemical 165 (2012) 143‐150. [20] K.‐I. Chen, B.‐R. Li, Y.‐T. Chen. Silicon nanowire field‐effect transistor‐based biosensors for biomedical diagnosis and cellular recording investigation, Nano Today 6 (2011) 131‐154. [21] K. Novoselov, S. Morozov, D. Jiang, Y. Zhang, S. Dubonos, I. Grigorieva, A. Firsov. Electric field effect in atomically thin carbon films, Science 306 (2004) 666‐669. [22] Y.H. Kwak, D.S. Choi, Y.N. Kim, H. Kim, D.H. Yoon, S.‐S. Ahn, J.‐W. Yang, W.S. Yang, S. Seo. Flexible glucose sensor using CVD‐grown graphene‐based field effect transistor, Biosensors and Bioelectronics 37 (2012) 82‐87. [23] G. Aragay, A. Merkoçi. Nanomaterials application in electrochemical detection of heavy metals, Electrochimica Acta 84 (2012) 49‐61. [24] R. Pearce, T. Iakimov, M. Andersson, L. Hultman, A.L. Spetz, R. Yakimova. Epitaxially grown graphene based gas sensors for ultra sensitive NO2 detection, Sensors and Actuators B: Chemical 155 (2011) 451‐455. [25] D. Bradley. Graphene gas sensor: Carbon, Materials Today 15 (2012) 233. [26] A.I. Ayesh, N. Qamhieh, S.T. Mahmoud, H. Alawadhi. Fabrication of size‐selected bimetallic nanoclusters using magnetron sputtering, Journal of Materials Research 27 (2012). [27] A.I. Ayesh. Electronic transport in Pd nanocluster devices, Applied Physics Letters 98 (2011) 133108. [28] P. Nemes‐Incze, Z. Osváth, K. Kamarás, L.P. Biró. Anomalies in thickness measurements of graphene and few layer graphite crystals by tapping mode atomic force microscopy, Carbon 46 (2008) 1435‐1442. [29] F. Schwierz. Graphene transistors, Nature Nanotechnology 5 (2010). [30] S.J. Park, O.S. Kwon, S.H. Lee, H.S. Song, T.H. Park, J. Jang. Ultrasensitive Flexible Graphene Based Field‐Effect Transistor (FET)‐ Type Bioelectronic Nose, Nano Letters 12 (2012) 5082−5090 [31] Y.‐M. Lin, K.A. Jenkins, A. Valdes‐Garcia, J.P. Small, D.B. Farmer, P. Avouris. Operation of Graphene Transistors at Gigahertz Frequencies, Nano Letters 9 (2009) 422‐426. [32] V. Pershina, T. Bastug, T. Jacob, B. Fricke, S. Varga. Intermetallic compounds of the heaviest elements: the electronic structure and bonding of dimers of element 112 and its homolog Hg, Chemical Physics Letters 365 (2002) 176‐183. [33] Z. Liu, P.C. Searson. Single Nanoporous Gold Nanowire Sensors, The Journal of Physical Chemistry B 110 (2006) 4318‐4322. [34] Z. Yin, Q. He, X. Huang, J. Zhang, S. Wu, P. Chen, G. Lu, P. Chen, Q. Zhang, Q. Yan, H. Zhang. Real‐time DNA detection using Pt nanoparticle‐decorated reduced graphene oxide field‐ effect transistors, Nanoscale 4 (2012). [35] X. Dong, Y. Shi, W. Huang, P. Chen, L.J. Li. Electrical Detection of DNA Hybridization with Single‐Base Specificity Using Transistors Based on CVD‐Grown Graphene Sheets, Advanced Materials 22 (2010) 1649–1653. [36] C. Yu, Y. Guo, H. Liu, N. Yan, Z. Xu, G. Yu, Y. Fang, Y. Liu. Ultrasensitive and selective sensing of heavy metal ions with modified graphene, Chemical Communications 49 (2013) 6492—6494. [37] D. Umadevi, G.N. Sastry. Metal ion binding with carbon nanotubes and graphene: Effect of chirality and curvature, Chemical Physics Letters 549 (2012) 39‐43.
13
Page 13 of 16
Ac ce pt e
d
M
an
us
cr
ip t
[38] W. Su, M. Cho, J.‐D. Nam, W.‐S. Choe, Y. Lee. Highly sensitive electrochemical lead ion sensor harnessing peptide probe molecules on porous gold electrodes, Biosensors and Bioelectronics 48 (2013) 263‐269. [39] D. Gu, J.B. Fein. Adsorption of metals onto graphene oxide: Surface complexation modeling and linear free energy relationships, Colloids and Surfaces A: Physicochemical and Engineering Aspects. [40] S.A. Siddiqui, N. Bouarissa, T. Rasheed, M.S. Al‐Assiri. Quantum chemical study of the interaction of elemental Hg with small neutral, anionic and cationic Aun (n = 1–6) clusters, Materials Research Bulletin 48 (2013) 995‐1002. [41] K.W. Frese Jr. The reorganization energy of the aqueous proton, Journal of Electroanalytical Chemistry and Interfacial Electrochemistry 249 (1988) 15‐24. [42] M.H. Bradbury, B. Baeyens. Modelling the sorption of Mn(II), Co(II), Ni(II), Zn(II), Cd(II), Eu(III), Am(III), Sn(IV), Th(IV), Np(V) and U(VI) on montmorillonite: Linear free energy relationships and estimates of surface binding constants for some selected heavy metals and actinides, Geochimica et Cosmochimica Acta 69 (2005) 875‐892. [43] D. Sareen, P. Kaur, K. Singh. Strategies in detection of metal ions using dyes, Coordination Chemistry Reviews 265 (2014) 125‐154.
14
Page 14 of 16
Figure Captions: Fig. 1: Schematic diagram of the ultra‐high vacuum system and nanocluster deposition.
ip t
Fig. 2: (a) Image of a graphene based sensor and electrode structure. (b) AFM image of graphene
layer. The line indicates the location where the height profile was measured. (c) Height profile of
cr
graphene and SiO2 substrate as measured using the AFM.
us
Fig. 3: Current‐voltage characteristics as a function of gate voltage for graphene sensors. Fig. 4: (a) Size distribution of Au nanoclusters as measured by the QMF. The nanoclusters were
an
produced using P = 21 W and fAr = 40 sccm. (b) AFM image of the produced Au nanoclusters with L = 60mm.
M
Fig. 5: Electrical current response signal of a graphene based sensor for DI water with 0.05 ppb Hg2+. (a) Sensor with graphene only, and (b) sensor with graphene and Au nanoclusters.
d
Fig. 6: Response signal of Hg2+ as a function of concentration in DI water for sensors based on
Ac ce pt e
either graphene only or graphene and Au nanoclusters. Fig. 7: Sensitivity histogram of the graphene based sensors for 0.6 ppb solutions of Cr2+, Cd2+, Cu2+, Co2+, Fe3+, Zn2+, and K+. 15
Page 15 of 16
Ac ce pt e
d
M
an
us
cr
ip t
I) Dr. Ahmad I. Ayesh Associate Professor, Physics of Nanomaterials, Physics Program ‐ Department of Mathematics, Statistics and Physics, Qatar University, Doha, Qatar Dr. Ayesh is the leader of the nanocluster devices group at United Arab Emirates University. He is expert in the nano‐ and micro‐ device fabrication using both the top‐down and bottom‐up approaches as well as the self‐ assembly of the nanostructure within the device. Furthermore, he is expert in nanomaterial fabrication and characterization. II) Miss. Zainab Karam MSC student at Masdar Institute, Abu Dhabi, UAE III) Dr. Falah Awwad Associate Professor, Department of Electrical Engineering, United Arab Emirates University, Al Ain, United Arab Emirates Dr. Falah Awwad received the M.A.Sc. and Ph.D. degrees in Electrical and Computer Engineering from Concordia University (Montreal, QC, Canada) in 2002 and 2006, respectively. He was a Post‐Doctoral Fellow at Ecole Polytechnique de Montréal and Concordia University, Montreal, QC, Canada. Between August 2007 – Feb. 2013, he was an Assistant Professor with the College of Information Technology (CIT) at United Arab Emirates University. Currently, he is an Associate Professor with the Department of Electrical Engineering – College of Engineering (UAE University). He is a member of the editorial Board of Journal of “Nanomaterials & Molecular Nanotechnology” and "Austin Journal of Nanomedicine & Nanotechnology". He is the recipient of the CIT Faculty Outstanding Service Award. The evaluation was done based on his professional and community service during the period 2009 ‐ 2012. His scientific research interests are Low‐Power High‐Speed VLSI Systems including repeater‐insertion methodologies for on‐chip interconnects, Biomedical Imaging Systems, Network on Chip, Wireless Body Sensors/Sensor Networks, and Network Communication algorithms. IV) Dr. Mohammed A. Meetani Associate Professor, Chemistry Department, United Arab Emirates University, Al Ain, United Arab Emirates. Dr. Meetani received his PhD in Applied Analytical Chemistry from Colorado School of Mines in 2003. He is expert in the following fields of research: Biological mass spectrometry, Degradation studies of environmental organic pollutants, Peptides/ protein sequencing, Pyrolysis mass spectrometry, Studying the structural properties of cellulose nanofibrous membranes, development of chromatographic methods for chemical analysis
16
Page 16 of 16