- Email: [email protected]
Nanoparticles in electrochemical sensors for environmental monitoring
Trends
Trends in Analytical Chemistry, Vol. 30, No. 11, 2011
Nanoparticles in electrochemical sensors for environmental monitoring Liza Rassaei, Mandana Amiri, Ciprian Mihai Cirtiu, Markus Sillanpa¨a¨, Frank Marken, Mika Sillanpa¨a¨ We review the state-of-the-art application of nanoparticles (NPs) in electrochemical analysis of environmental pollutants. We summarize methods for preparing NPs and modifying electrode surfaces with NPs. We describe several examples of applications in environmental electrochemical sensors and performance in terms of sensitivity and selectivity for both metal and metal-oxide NPs. We present recent trends in the beneficial use of NPs in constructing electrochemical sensors for environmental monitoring and discuss future challenges. NPs have promising potential to increase competitiveness of electrochemical sensors in environmental monitoring, though research has focused mainly on development of methodology for fabricating new sensors, and the number of studies for optimizing the performance of sensors and the applicability to real samples is still limited. ª 2011 Elsevier Ltd. All rights reserved. Keywords: Air pollution; Amperometry; Electrochemical sensor; Environmental monitoring; Heavy metal; Herbicide; Modified electrode; Nanoparticle; Pesticide; Voltammetry
1. Introduction Liza Rassaei* MESA+ Institute for Nanotechnology, University of Twente, Enschede, The Netherlands Mandana Amiri Department of Chemistry, University of Mohaghegh Ardabili, Ardabil, Iran Ciprian Mihai Cirtiu Department of Chemistry, McGill University, 801 Sherbrooke Street West, Montreal (PQ) H3A 2K6, Canada Markus Sillanpa¨a¨ Research and Innovation Laboratory, Finnish Environment Institute, P.O. Box 140, FI-00251 Helsinki, Finland Frank Marken Department of Chemistry, University of Bath, Bath BA2 7AY, UK Mika Sillanpa¨a¨ Laboratory of Green Chemistry, LUT Energy, Lappeenranta University of Technology, Patteristonkatu 1, 50101 Mikkeli, Finland
*
Monitoring air, soil and water for hazardous pollutants is important and based on the need to protect the environment and public health from possible distribution of natural and industrial inorganic and organic contaminants. There is a constantly increasing need for online monitoring of contaminants in our environment, driven by new legislation and new technologies. Important examples are (i) control of wastewater quality to avoid release of synthetic drugs residues or detergents and (ii) monitoring of persistent carcinogens or residues of explosives released into the environment. Pesticides (herbicides, fungicides and insecticides) are widely used in agriculture and industry. To limit their toxicity and their accumulation in living organisms, dose adjustment and trace-level monitoring are desirable. Thus, there is an essential need to develop new methods for simple pesticide detection at low concentrations, especially in the field.
Heavy metals occur naturally in the environment, but, due to industrialization, large amounts of heavy metals bound in fossil fuels and mineral materials have been released into the environment and deposited in trace amounts in nearly every part of the planet. Elevated levels of heavy metals in natural water may have a detrimental effect on both human health and the environment [1]. Apart from the direct impact on health or environmental problems, water or soil contamination can cause considerable economic and financial damage, so there is a need for novel instruments capable of real-time, in-situ detection of heavy metals and online monitoring. Air pollution is another serious problem, especially in many heavily populated and industrialized areas, and in newly developing countries. In the majority of the developed world, legislation has already been introduced to an extent that local authorities are requested by law to conduct regular local air-quality monitoring of major urban pollutants (e.g., ozone,
Corresponding author. E-mail: [email protected]
1704
0165-9936/$ - see front matter ª 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.trac.2011.05.009
Trends in Analytical Chemistry, Vol. 30, No. 11, 2011
benzene, SO2, CO, CO2 and NOx produced by industrial activity and transportation) [2]. In order to achieve this goal, pollutant concentrations must be monitored accurately, and ideally in situ, so that sources may be recognized quickly and the atmospheric dynamics of the process are understood. Growing concern worldwide about air pollution and its effect on the environment, industrial safety and hygiene has also made it essential to monitor gaseous species (e.g., SO2, H2S, ozone, H2, NOx, CO, Cl2, NH3, formaldehyde, ethanol, propane or ethylene oxide). Furthermore, such data would lead to real-time environmental decision-making capabilities as a result of hazardous levels being rapidly identified [2,3]. Increasing concern about the distribution and the impact of chemical reagents from analytical methodologies has led to new field methods designed to minimize use of toxic reagents during sample pretreatment and analytical measurement. Electrochemical methods for analysis can be especially sensitive, cheap and portable to provide data even in remote locations. A wide range of electrochemical sensors that fulfill the requirement to be environment friendly are commercially available. For example, sensors for water monitoring have been developed, and several of these are now advanced prototypes. Electrochemical monitoring is especially of interest for in-situ measurements with simple, compact and mobile equipment. Typical electrochemical sensors comprise a sensing electrode (as transduction element), a diffusion barrier, a counter-reference electrode and an electrolyte. The analytical information is obtained from the electrical signal that results from the interaction of the target analyte and the recognition layer at the sensing electrode. Depending on the electrical signal to be measured, electrochemical techniques are divided into potentiometric, conductometric, amperometric or voltammetric methods. Different electrochemical devices have been developed for environmental monitoring, depending on the nature of the analyte, the characteristics of the sample matrix and sensitivity or selectivity requirements. The ideal sensor should possess the following characteristics [4]: (1) specificity for the target species; (2) sensitivity to changes in target-species concentrations; (3) fast response time; (4) extended lifetime of at least several months; and, (5) small size (miniaturization) with the possibility of low-cost manufacture. Nanoparticles (NPs) are attracting attention due to their low cost and unique size-dependent properties. The incorporation of NPs into a variety of matrices to form nanocomposite films is attracting much attention. NPs have been used in many electrochemical, electroanalytical and bioelectrochemical applications. The uniqueness of NPs is due to their mechanical, electrical,
Trends
optical, catalytic and magnetic properties as well as their extremely high surface area per mass. In addition to novel properties, nanomaterials and nanotechnology open up new approaches to manufacture electrodes cost effectively by minimizing the materials needed and waste generation [5]. This is especially relevant to expensive materials (e.g., gold and platinum). For example, inexpensive materials (e.g., carbon coated by NPs) result in a large ratio of surface area to volume for low-cost sensing electrodes. In recent studies, it was demonstrated that NP electrodes could be obtained with high sensitivity and even with individual NPs giving responses [6,7]. The combination of nanotechnology with modern electrochemical techniques allows the introduction of powerful, reliable electrical devices for effective process and pollution control. Although the NPs in general play different roles in different electrochemical sensors, with regard to electroanalysis using a NP-modified electrode has several advantages: (1) effective catalysis; (2) fast mass transport; (3) large effective sensor surface area; and, (4) good control over electrode microenvironment [5]. NPs also present attractive characteristics often notably different from bulk materials in terms of both physical and chemical features. However, while several nanomaterial-based sensor systems have been reported in the literature, their implementation in routine in-situ devices remains a challenge. The monitoring of industrial processes with known sample matrix is usually accurate and reliable; however, in-situ monitoring in environmental analysis often with an unknown or changing matrix provides a bigger challenge in the development of measurement devices [8]. Electrochemical sensors have been employed for several decades for a variety of environmental monitoring applications, including monitoring of water-quality parameters (conductivity, dissolved oxygen or pH) [9], measurement of trace heavy metals [10], and carcinogens and organic pollutants (N-nitroso compounds, aromatic amines and phenols) [11]. There is a myriad of different NP materials, but they are usually divided by their chemical nature into six groups: (1) metals; (2) metal oxides; (3) carbonaceous; (4) polymeric; (5) dendrimeric; and, (6) composites. Most relevant in the design of electrochemical sensors are metal and metal-oxide NPs, so these provide the focus of this review. A further topic developing from the wider use of nanotechnology (not covered in this review) is the environmental impact and potential health effects of NPs, which are still largely unknown. http://www.elsevier.com/locate/trac
1705
Trends
Trends in Analytical Chemistry, Vol. 30, No. 11, 2011
Although several review articles have addressed the application of NPs in electroanalysis [3,4], their applications as sensing elements in electrochemical analysis for environmental monitoring have not previously been the focus, so the aim of this review is critically to assess and to overview the role of NPs used in the design of electrochemical sensors for environmental monitoring. The intention is not to provide a complete, exhaustive review of the available literature, but rather to provide insight into potential applications and benefits of NPs for electrochemical environmental monitoring. First, we introduce different methods of NP preparation, then we address electrode-surface modification and sensor preparation, and finally we describe the application and the performance for several examples of reported NP-based sensors for environmental monitoring.
2. Preparing nanoparticles and modifying electrode surfaces 2.1. Colloidal nanoparticles The development of newer methods for the controlled synthesis of metal NPs with different shapes and sizes has attracted attention during the past decade. One of the advantages of using NPs in sensing design is that their size and their surface characteristics can be tuned during the preparation process to achieve high sensitivity or selectivity. The preparation methods for NPs have been reviewed [12]. Different wet chemical methods have been used for the synthesis of metallic NP dispersions. The most commonly employed techniques involve the use of excess reducing agents (e.g., sodium citrate, tannic acid or NaBH4). NPs can also be synthesized in a two-phase water-organic system, or using an organic reducing agent (e.g., aldehydes) or reverse micelles. The concentrations of the salt and the reducing agent and temperature are important parameters to determine the size of NPs. Finally, NPs can be prepared by unconventional methods (e.g., radiolysis, laser ablation, vacuum evaporation, or microwave techniques). Metal-oxide NPs are prepared based on chemical methods (e.g., hydrolysis) and physical methods, including sputtering, evaporation (thermal and electron beam), pulse-laser deposition and ion implantation. Assemblies of NPs can be prepared by using chemical techniques [e.g., sol-gel, co-precipitation, impregnation, or chemical-vapor synthesis (CVS)] or hydrothermal techniques. There have been reviews on growth of NPs in gas and liquid phases [13]. 2.2. Modification of the electrode surface with nanoparticles Various attachment techniques, including physical and chemical methods, are commonly used for the deposition of colloidal NPs. Particle size and size distribution are the 1706
http://www.elsevier.com/locate/trac
most important characteristics of NPs for their performance in electrochemical-sensor design, because, depending on the size and the distance between adjacent particles, the degree of diffusion-layer overlap (the nature of the voltammetric responses) can be manipulated [14]. Defined and ordered arrangement of NPs is a promising approach for the construction of electrochemical sensors. The electrode surface can be modified with colloidal NPs by simply casting a mixture of NPs and additives (e.g., an enzyme). Methods have been developed with solvent evaporation, growth of an NP-modified sol-gel network on the electrode surface, or electro-aggregation. NPs can be modified with different functional groups and then immobilized on the electrode surface. It is possible to pre-mix the NPs with conductive materials (e.g., different types of nano-carbon to prepare a paste electrode) or to trap them in a conductive polymer. The layer-bylayer method based on electrostatic interaction has been used for charged NPs, which get entrapped in an oppositely-charged poly-electrolyte [15]. In many cases, by applying a reducing potential, NPs are directly electrodeposited from the appropriate salt solutions onto the electrode surface so as to have the desired size. These NPdeposition processes are easy to use, although control of the size of the particles may be greatly depend on deposition time, deposition potential, electrolyte solution, and the salt concentration. Fig. 1 shows typical SEM images illustrating the effect of electrodeposition time on the morphology and the density of Au-NPs on borondoped diamond (BDD)-electrode substrates [16]. It can be seen that the longer the deposition time, the larger the average diameters of deposited Au-NPs. With the deposition time of 10 s, 30 s and 60 s, Au-NP diameters were reported as 79.5 ± 11.3 nm, 152.8 ± 11.4 nm, and 222.8 ± 18.3 nm, respectively. At the deposition time of 600 s, the Au-NPs deposited in a multilayer cluster after covering all BDD surfaces. In general, NPs are preferentially deposited on the substrate defects or at the edges, resulting in a non-uniform distribution [17]. Therefore, using pre-synthesized and functionalized colloidal NPs may solve some of these problems when well-defined nanostructured deposits are needed. When the electrode surface is modified using electrodeposition techniques, the pH or the electrode potential applied may cause unwanted dissolution or formation of surface oxides on the electrode surface. Cu-NPs and NiNPs are oxidized easily so their formation on the electrode surface can also be carried out in situ.
3. Electrochemical characterization of nanoparticle-modified electrodes 3.1. Methods Compared to bulk electrodes, the presence of NPs on the electrode surface enables fast electron-transfer kinetics,
Trends in Analytical Chemistry, Vol. 30, No. 11, 2011
Trends
Figure 1. SEM images of a polycrystalline boron-doped diamond (BDD) electrode surface after direct electrochemical Au-NP deposition: deposition time of (a) 10 s, (b) 30 s (c) 60 s and (d) 600 s [16] (reprinted with permission from Elsevier).
reduces overpotential, increases the electro-active surface area, and causes redox reactions to become kinetically feasible [5]. From electrochemical studies of NP-modified electrodes, it is possible to calculate the number of NPs deposited on the electrode surface, capacitive charging effects or specific underpotential deposition (UPD) effects. Many NPs, including Au-NPs and Pt-NPs, are electrochemically-active materials. They possess distinct electrochemical properties because of their unique electronic structures. Fig. 2A shows typical cyclic voltammograms of an indium tin oxide (ITO) electrode before and after Pt-NPs were grown onto the surface [18]. Due to the simple, well-defined responses, redox models (e.g., the [Fe(CN)6]4/[Fe(CN)6]3 or 3+ 2+ Ru(NH3)6 /Ru(NH3)6 couples) have been widely used to characterize the surface properties of different NPmodified electrodes (see Fig. 2B) [19]. The voltammetry of these two standard redox processes is examined in order to investigate how the NP-surface coverage and electron-transfer kinetics are compared to unmodified electrodes. The electrochemical properties of NP-modified electrodes significantly depend on their surface properties. In cyclic voltammetry experiments, the peakto-peak separation and electron-transfer kinetics of NPmodified electrodes have been shown to change with the amount of NPs. However, there is an optimal coverage for the amount of NPs on the surface. An excess amount of NPs usually gives no further improvement or increases the resistance and the double-layer capacitance of the modified electrode, leading to a decrease in electrochemical sensitivity [20].
Electrochemical impedance spectroscopy (EIS) can also be employed to investigate the changes in impedance of the electrode surface as a function of coverage for NP-modified electrodes. Fig. 2C shows typical electrochemical-impedance spectra of the [Fe(CN)6]3/ [Fe(CN)6]4 redox couples for unmodified and modified electrodes with Pd-NPs. It can be observed that the semicircle diameter of the Nyquist plot for [Fe(CN)6]3/ [Fe(CN)6]4 redox has significantly changed with the loading of Pd-NPs on the substrate [21]. The apparent standard heterogeneous rate constant (assuming [Fe(CN)6]3 = [Fe(CN)6]4) is obtained using [22] k app ¼ RT =ðF 2 Rct AC Þ where Rct, T, F, A and C* represent the gas constant, the absolute temperature, the Faraday constant, the true area and the bulk concentration of the redox couple, respectively [22]. Generally, the kapp values for the [Fe(CN)6]3/[Fe(CN)6]4 redox couple on NP arrays prepared under different growth conditions indicate increasingly facile heterogeneous electron-transfer kinetics, mainly resulting from an increase in active area. For environmental monitoring, NPs are mostly used in two different ways in constructing electrochemical sensors: (1) together with an enzyme (enzymatic sensors) to increase the rate of electron transfer and to improve the sensitivity of the enzyme; or, (2) as individual catalysts (non-enzymatic sensors) to catalyze the reactions and to lower the limits of detection (LODs). As their function in these sensors is different, these two cases are discussed separately.
http://www.elsevier.com/locate/trac
1707
Trends
Trends in Analytical Chemistry, Vol. 30, No. 11, 2011
3.2. Enzymatic sensors A combination of enzymatic reactions with electrochemical methods allows development of selective enzyme-based electrochemical biosensors for determination of environmental pollutants. These sensors, although less robust, exhibit good selectivity, sensitivity, rapid response, and have miniature size. In environmental monitoring, enzyme-based or protein-based NP sensors have mostly been prepared for detection of pesticides (e.g., phenolic compounds) or NOx compounds. A major challenge to develop such sensitive and stable sensors comes from the effective immobilization and ‘‘electrical connection’’ of enzymes to solid electrode surfaces. The most promising areas for the use of enzyme electrodes in environmental monitoring are pesticide biosensors. Enzyme electrodes have been investigated as emerging sensors for faster, simpler detection of pesticides. However, their sensitivity is not yet adequate for application in detecting very low concentrations of pesticides. External supply of substrate is often required to measure the changes in enzyme activity. Compared to free enzyme in solution, immobilized enzyme can be more stable and resistant to various environmental changes. Electrochemical sensors incor-
porating enzymes with nanomaterials combine the recognition and catalytic properties of enzymes with the electronic properties of NPs. The NPs can help enlarge the enzyme loading at the electrode surface. The NPs with smaller size (1–2 nm) have been shown to be more suitable for enzyme immobilization and they increase the biosensor performance because the small size of these NPs results in a significant surface-to-volume ratio and creates more binding sites on the electrode surface for easier contact with enzyme molecules and less denaturation [23]. NPs of small size also allow more freedom in orientating adsorbed proteins, or they introduce conducting channels between the prosthetic groups within the enzyme and the electrode surface. In other words, they maximize the utilization of the bioactive sites of the enzyme and act as electron-transfer pathways [23]. Due to their high surface area and good biocompatibility, surface-functionalized NPs are suitable for many surface-immobilization mechanisms, providing a suitable support to adsorb redox enzymes and proteins. NPs of appropriate design can adsorb redox enzymes and proteins without loss of biological activity. Moreover, since most of the NPs carry charges, they can also electrostatically adsorb biomolecules with opposite
Figure 2. (A) Cyclic voltammograms (scan rate of 50 mV/s) recorded in the solution of H2SO4 for (a) a bare ITO electrode and (b,c) the Pt-NPsITO electrode prepared via (b) 4 and (c) 24 h of growth for a geometric area of 0.031 cm2 [18]. (B) Cyclic voltammograms (scan rate 50 mV/s) of K3[Fe(CN)6] at bare Au electrode (radius of 1.5 mm) and Au-NPs on GCE (radius of 1.5 mm) [19], and (C) Nyquist diagrams (ZÕÕ vs. ZÕ) for the EIS measurements (Amplitude: 5 mV) in 5 mmol/L K3Fe(CN)6/K4Fe(CN)6 in 0.1 mol/L PBS at E1/2 = 0.13 V vs. Ag/AgCl for the bare GCE (h) and Pd-NPs-GCE (o) [21] [(A) reprinted with permission from American Chemical Society, (B) ª Wiley-VCH Verlag GmbH & Co. KGaA, reproduced with permission, and (C) reprinted with permission from Elsevier].
1708
http://www.elsevier.com/locate/trac
Trends in Analytical Chemistry, Vol. 30, No. 11, 2011
charges or selectively bind to active regions on the biomolecule surface. Enzymes usually lack direct electrical communication with the electrode substrate due to their active centers being surrounded by a protein shell. Electron transfer between the electrodes and the active centers is blocked. The conductivity of NPs enhances the rate of direct electron transfer between the active site of the enzymes and electrodes acting as electrical wires. Among different types of NPs, Au-NPs, Pt-NPs, and Ag-NPs have shown biocompatibility with enzymes (e.g., in several recent publications, it was shown that Au-NPs could be used as a conduit for electrons from enzyme redox reactions [24]). Due to their biocompatibility, Au-NPs are considered an excellent option for replacing potentially harmful mediators in the construction of biosensors [25]. The electrode-surface environment and the amount of NPs deposited are also crucial in design of enzyme-based electrochemical sensors. Among different enzymes or proteins commercially available, acetyl cholinesterase (AChE) and tyrosinase (TY) and, in a few cases, hemoglobin protein (Hb) have been used in construction of enzyme-based sensors for environmental monitoring. The application of AChE-nanomaterial-modified electrodes for analysis of pesticides was recently reviewed by Periasamy et al. [26]. In brief, in most AChE-based sensors, organophosphorous (OP)-pesticide detection is mainly based on the irreversible inhibition of AChE activity by OP pesticides. The degree of inhibition is calculated by comparison of the residual activity of the enzyme with the initial activity. TY-NP-based electrodes are usually used for detection of phenolic compounds. The mechanism is based on the oxidation of phenolic compounds (e.g., herbicide atrazine) by tyrosinase to catechol and further to quinone [27]. Hemoglobin (Hb) is known for its catalytic activity towards the reduction of nitrite. However, it is difficult for Hb in solution to exchange electrons directly with bare solid electrodes, as its electroactive center (heme) is buried in its electrochemically-insulated peptide backbone and, furthermore, its adsorption on the electrode surface deteriorates its bioactivity and results in denaturation. NPs may provide a mediator to facilitate electron exchange for Hb [28]. A more uniform film can be obtained when NPs are deposited in a polymer, for example, Au-NPs were codeposited with a biocompatible polymer (chitosan) on the electrode surface to obtain more uniform films [29]. The AChE was then immobilized on this electrode to provide an amperometric sensor to measure different OP pesticides (e.g., malathion, pralidoxime iodide, and monocrotopos). The peak current for oxidation of thiocholine (hydrolysis product of AChE) increased. A negative shift in peak potential compared to bare elec-
Trends
trode or chitosan-modified electrode was related to the presence of Au-NPs and their catalytic behavior. Although enzymatic sensors are more suitable for environmental monitoring, for rapid and simple measurements in the field, their application is still limited, as they need the substrate to activate or to maintain enzyme activity over a lengthy storage time [25,30].
3.3. Non-enzymatic sensors The sensing ability of NPs in non-enzymatic sensors is directly related to their catalytic property. For example, using a seed-mediated growth technique, Cui et al. deposited Au-NPs (mean size 4 nm) on a glassy carbon (GC) electrode [31], which showed excellent catalytic activity towards the oxidation of nitrite in a real wastewater sample without any major interference. However, the catalytic properties of NPs depend on the protecting groups or the presence of additives (e.g., cysteine or tannic acid). Controlling NP size, it is possible to change the catalytic activity of NPs so that it becomes sensitive to different oxidation states of one species. For example, controlling the size of Au-NPs allows Cr(VI) to be measured electrochemically in the presence of excess of Cr(III), offering benefits over the commonly-used diethylenetriamine-pentaacetic-acid (DTPA) method [32]. The stability and activity of this electrode depended on the size and the density of the NPs. Larger NPs were shown to cause not only a negative shift in peak potential and lower peak current but also deterioration in the stability due to the blocking electron transfer between solution species and electrode. This sensor was applied to measure Cr(VI) in tap water, stream water, and sea water [32]. NPs provide a larger surface area for detection of analytes, leading to electrochemical sensors with lower LODs. For example, Au is known to be the most appropriate material to measure arsenic but the analysis of arsenic by the anodic stripping voltammetry (ASV) method is limited by the formation of a monolayer of As0 deposition due to non-conductive property of the As-Au compound. Au-NPs with large surface area are used to overcome this limitation in many studies, but their performance depends on their size, shape and distribution during the nucleation process, which is followed by particle growth [17,33]. Another example is the detection of lead. Lead in acidic media shows UPD on Au electrodes but the maximum response of this process is limited by the electrode area, so the use of Au-NPs with large surface area can solve this problem [34]. Studies have also shown that increasing the surface roughness of a platinum electrode via platinization provides more active sites for catalyzing the oxidation of CO to CO2, leading to increase in sensitivity of this sensor [35]. NPs can also be used in preparing paste electrodes. In this type of electrode, the presence of NPs increases the
http://www.elsevier.com/locate/trac
1709
Trends
Trends in Analytical Chemistry, Vol. 30, No. 11, 2011
surface area of the paste and diminishes the Ohmic resistance of the paste [36]. The small size of NPs and their resulting large surface area can lead to NP-NP aggregation when preparing the sensor. To avoid this, the surface of NPs can be functionalized with thiol groups or other selfassembled monolayers (SAMs) containing amino or carboxylic acid groups. These functional groups introduce new functions in sensor design. In the case of Au-NPs (mean size 5 nm), SAMs stabilize the particles by preventing aggregation. They can also provide an insulating medium between metallic gold cores [18]. In the case of highly monodispersed benzyl mercaptane-stabilized Ag-NP-modified carbon nanotubes (CNTs) on a GC electrode, the phenyl rings of benzyl mercaptane attach to the side wall of CNTs and the free thiol groups at the end of the molecules interact with Ag-NPs to form strong S-Ag bonds. The benzyl mercaptane prevents Ag-NPs from aggregation and promotes interaction with CNTs via p–p interac-
tions to give a uniform dispersion of Ag-NPs on CNTs [37].
4. Sensitivity and selectivity 4.1. Sensors based on metal nanoparticles Metal NPs (e.g., Au, Pt, Pd and Ag) show catalytic activity when they are incorporated into electrochemical sensors. For example, an electrochemical sensor based on Au-NPs (mean diameter 4–6 nm) embedded in 3aminopropyl)triethoxysilane (APS)-derived silicate sol– gel three-dimensional network (APS–Au-NPs) was prepared for simultaneous detection of N2H4, SO32 and NO2 (see Fig. 3). The electro-oxidation peaks appeared at 0.05 V, 0.20 V, and 0.55 V vs. Ag/AgCl for N2H4, SO32 and NO2 in pH 7.2 at the APS–Au-NP-modified GC electrode with a large decrease in the overpotentials to the extent of 750 mV, 600 mV and 250 mV, respectively [38]. Ion-exchange membranes [e.g., Nafion (a
Figure 3. Cyclic voltammograms (scan rate 50 mV/s) at (a) GCE, (b) GCE/APS, (c) GCE/APS-Au-NPs for (A) 250 lM N2H4, (B) SO32 and (C) nitrite (D) (scan rate 20 mV/s) for a mixture containing 250 lM of N2H4, SO32 -and nitrite in 0.1 M PBS pH = 7.2 [38] (reprinted with permission from Elsevier).
1710
http://www.elsevier.com/locate/trac
Trends in Analytical Chemistry, Vol. 30, No. 11, 2011
negatively charged membrane)] can be used beneficially in the sensor design. A screen-printed electrode was modified with a paste of Bi-NPs (mean diameter 50 nm) and Nafion as a binder to help the NPs to adhere to the electrode surface. Well-defined and highly-reproducible electrochemical responses were obtained for detection of Zn(II), Cd(II) and Pb(II) using ASV [39]. The presence of Nafion causes the peak potentials to shift to more negative values due to the cation-exchange behavior of the Nafion coating, which favors an oxidized state of the metal ions. Coating the composite electrode with a Nafion membrane also helped in avoiding anionic interference in the sample (NO3, SO42, CO32). A poly (L -lactide)-stabilized Au-NP-modified (PLA-Au film) electrode has been used for the indirect detection of sulfide by measuring the inhibited oxidation current of As(III) in acidic medium [40]. A substantial decrease in oxidation peak current for As(III) in the presence of sulfide was obtained and it was related to the formation of As2S3 compound. Using this approach with differential pulse stripping voltammetry (DPSV), the amount of sulfide in hot spring water and acid rain was determined with an LOD of 0.04 lM. The following mechanism was suggested for this reaction based on arsenite existing in the form of H3AsO3 in acidic media (see Fig. 4). Na2 S þ 2HCl ! H 2 S þ 2NaClð1Þ 2H 3 AsO3 þ 3H 2 S ! As2 S3 þ 6H 2 Oð2Þ The detection of cyanide and thiocyanide anions in wastewater and industrial samples was reported at Ag-NP-modified electrodes based on the complexation reaction of Ag+ with these anions. The electron-transfer reaction in these processes is controlled by the adsorption of CN or SCN. For example, Taheri et al. developed a cyanide sensor based on mercaptopropyltrimethoxy-silane (MPS)-stabilized Ag-NPs embedded in 3D sol-gel matrix on a gold electrode [41]. MPS acts here as a capping agent. The cathodic current for an Ag-NP-modified electrode decreases in the presence of cyanide ions because of the formation of the Ag+–CN complex. The peak current decreases due to the blockage of the electrode surface by this complex so that less surface area of NPs is available. This electrode was successfully applied for determination of cyanide in industrial samples. In another study, an oleate-stabilized Ag-NP (mean diameter 22 nm)-modified GC electrode was prepared as an amperometric sensor for thiocyanate. In the presence of SCN, a new oxidation peak appeared, due to the formation of the Ag+–SCN complex and the anodic peak current for Ag-NPs decreased due to the electro-reductive activity of Ag-NPs [42]. Ag-NPs with relatively high specific surface area can integrate proteins and help in orientation of adsorbed proteins. Gan et al. modified the pyrolytic graphite-
Trends
electrode surfaces with bioconjugates of Ag-NPs (mean diameter 11 nm) and hemoglobin for determination of nitric oxide [43]. The UV-Vis absorption spectrum confirmed a similar Soret band (405.6 nm) for both Hb with and without Ag-NPs, consistent with Ag-NPs preserving the natural conformation of Hb. The increased rate of electron transfer confirmed the electron-transfer process between protein and electrode was effective. Both electrochemical reactivity of the protein and its detection sensitivity toward NO were enhanced in presence of AgNPs. NPs fixed in an organic matrix have proved to be very useful in construction of electrochemical sensors. The NPs represent the electronic conductivity while the organic matrix has the selective binding sites for selective adsorption of analyte molecules. An interesting part of this approach is the ability to control the sensor properties by molecular design. The charge transport in such NP assemblies sensitively depends on the particle size, the interparticle spacing and the chemical composition (i.e. the electronic structure of the stabilizing or interconnecting ligands). Thus, selective absorption of analytes to the organic matrix among the NPs can be utilized to affect the electrical transport in such hybrid systems (e.g., Wohltjen and Snow developed a colloidal metal-insulator-metal ensemble chemiresistor sensor based on a monolayer of octanethiol-stabilized Au-NPs transducer film for ppm detection of toluene and tetrachloroethylene [44]). Pt-NPs have been used in construction of electrochemical sensors for detecting CO, CO2, NO2, NO, formaldehyde, methanol, naphthols, phenol, mercury and As(III). Pt-NPs can oxidize CO to CO2 in the presence of interferences (e.g., Cl2, SO2, H2S, NO2 and NO) and have been employed for applications in environmental pollution monitoring [45]. The adsorption of CO was shown to be purely chemical and irreversible at Pt-NP (mean diameter 18 nm)-modified electrodes. In this case, CO oxidation occurred through the reaction of adsorbed CO with Pt-OH to form CO2 following the reaction pathway indicated by following equation: CO þ H 2 O ! CO2 þ 2H þ þ 2e
ð3Þ
Photinon et al. showed that the catalytic activity of PtNPs toward oxidation of CO and the sensor response depend on the crystallographic orientation of platinum deposits [46]. It is known that CO2 adsorbs preferentially on the hydrogen-treated Pt surfaces (Pt-H), so detection of Pt-HCOO in acidic media on Pt surfaces is a method to quantify CO2 in a gaseous test environment. The anodic adsorptive stripping of CO2 at high and low concentrations shows the inverse relationship between Pt-H and Pt-HCOO, as depicted in Fig. 5. Electrochemical measurement of As is usually accompanied by interference from Cu. However, Pt-NPmodified electrodes was shown to be less sensitive to Cu,
http://www.elsevier.com/locate/trac
1711
Trends
Trends in Analytical Chemistry, Vol. 30, No. 11, 2011
Figure 4. Indirect detection of sulfide by measuring the inhibited oxidation current of As(III) in acidic medium at poly (L -lactide) stabilized AuNP-modified (PLA-Au film) electrode [40] (with kind permission from Elsevier).
especially for low concentrations of As (less than 100 ppm) [47]. The mechanism suggested for this detection strategy involved formation of PtOH on the electrode surface followed by binding As(OH)3 and electron transfer. The characteristics of mixed-metal NPs, especially their catalytic property, are quite distinct from those of pure-metal NPs. Also, their nano-composite structures are of interest. Very few studies have addressed the codeposition of Pt-NPs and a transition-metal oxide for improving efficient catalysis. For example, co-deposition of Pt-NPs (mean diameter 6 nm) and Fe(III) on a GC electrode resulted in a sensor with good catalytic ability toward the oxidation of nitrite. The Pt-NPs were caged by the Fe(III) network formed by the Fe–O–Fe linkages,
1712
http://www.elsevier.com/locate/trac
and the presence of Fe(III) was shown to be an important factor for the kinetics of NP growth. This sensor showed good response for detection of nitrite in the presence of interferences (e.g., 500-fold Na+, K+, Mg2+, Ni2+, Cu2+, Cl, NO3, SO42, CO32; 200-fold H2O2; 100-fold SO32, S2O32, L -glutamic acid; and, 50-fold dopamine, uric acid, ascorbic acid, glucose, D -fructose, sucrose, oxalic acid, citric acid, D -galactose and malic acid). An LOD of 0.47 lM was obtained for this electrode [48]. In another study, the co-deposition of Pt and Pd-NPs on a Nafion-modified GC electrode was reported by Zhou et al. [49]. This sensor showed remarkable electrocatalytic activity towards oxidation of formaldehyde in aqueous solution of H2SO4 with a broad linear range and
Trends in Analytical Chemistry, Vol. 30, No. 11, 2011
Trends
Figure 5. Sketches for anodic stripping of CO2 at low and high concentrations showing the inverse relationship between Pt-H and Pt-HCOO [46]. E1 is the potential where CO2 adsorbs on the Pt-H surface (0.3 V vs. RHE). E2 is the potential where the non-reacted Pt-H is oxidized from the electrode (0.5 V vs. RHE). E3 is the potential where Pt-HCOO is oxidized (0.75 V vs. RHE).
a low LOD. The Pt–Pd alloy has a synergistic effect to minimize the poisoning of the electrode surface by reaction intermediates, particularly CO. Specific interactions between the analyte species and the organic-shell functional groups on NPs could enhance sensitivity and increase the selectivity of the sensor. For example, a GC electrode modified with Au-NPs capped with 11-mercaptoundecanoic acid units was used for detection of Cu2+ ions and a sensitivity below 1 ppb with improved selectivity in the presence of Fe3+ and Zn2+ ions was reported [50]. Multilayers of conductive NPs gave rise to a large porous surface-area electrode, where the local microenvironment of the metallic NPs was controlled by the cross-linking elements. Jena et al. modified an Au-electrode surface with a thiol-functionalized sol-gel derived from 3D silicate networks via 3-(mercaptopropyl)-trimethoxy-silane (MPTS) and soaked the electrode in citrate stabilized AuNPs (mean size 5–6 nm) solution for self assembly of AuNPs on the thiol groups. A random distribution of Au-NPs through the network with a size distribution of 70–100 nm was obtained. It was shown that the overlap between the diffusion layers at individual Au-NPs produced a diffusion layer linear to the geometrical surface area of the electrode. This electrode showed a good performance for simultaneous detection of As(III), Hg(II) and Cu(II) (5 ppb) in a real sample from an arseniccontaminated area in West Bengal [51].
4.2. Sensors based on metal-oxide nanoparticles Metal oxides provide robust building blocks in nanostructures and offer functionality from electrically conducting to insulating and from highly catalytic to inert. Application of metal oxides in sensors, although less prevalent, is therefore very promising [52]. Zirconia (ZrO2)NPs are selective sorbents in electrochemical sensors for detection of OP pesticides and nerve agents. Zirconia has a strong affinity towards phosphorous groups, so nitroaromatic OPs strongly bind to the ZrO2-NP surfaces [53]. The ZrO2-NP also provide a large surface area and increase the interaction with OP compounds (e.g., methyl parathion, paraoxone and fentrothion). Iron-oxide NPs have been used as the immobilizing matrix for applications in biosensing due to their good biocompatibility, low toxicity, high electron-transfer capability, and high adsorption ability. Iron-oxide NPs provide a favorable environment for immobilization of many biological molecules. For example, Kaushik et al. immobilized single-standard calf-thymus DNA onto chitosan-Fe3O4 hybrid nanobiocomposite film (Fe3O4NPs, mean diameter 22 nm) and deposited it onto an ITO substrate for detection of pyrethroids (insecticides) (e.g., cypermethirn (CM) and permethrin (PM) [54]). The presence of Fe3O4-NPs on the ITO electrode surface increased the electroactive surface area and also promoted electron transfer due to the uniform dispersion throughout the chitosan network on the electrode.
http://www.elsevier.com/locate/trac
1713
Trends
Trends in Analytical Chemistry, Vol. 30, No. 11, 2011
Metal-oxide semiconductors (e.g., WO3, TiO2, ZnO and SnO2) are widely used in gas sensors, having high detection ability and stability. However, the application of NPs as gas-sensing elements is usually limited to chemiresistors, which have been reviewed by Franke et al. [55]. The electrical resistance of chemiresistors changes in the presence of an oxidizing or reducing gas. Many gas sensors operate by detecting changes in the electrical properties of a nanostructured sensing-film material in the presence of a test gas. A review by Toohey addresses the influence of electrode material and NP geometry on sensitivity and selectivity of gas sensors [56]. The gas-sensing NPs (e.g., ZnO, SnO2, TiO2, and WO3) can be deposited on a gold or platinum electrode as a substrate. These sensors operate through transference in the equilibrium of the surface-oxygen reaction by the presence of the target analyte. A reducing gas increases the conductivity of an n-type semiconductor and decreases the conductivity of a p-type semiconductor, and an oxidizing gas behaves vice versa. The use of NP metal oxides increases the sensitivity of the sensor when the surface-oxygen chemisorption reaction determines the sensor response [57]. Mixed metal-oxide-semiconductor NPs enhance the performance of the electrochemical sensors. For example, Yuasa et al. loaded PdO-NPs on SnO2-NPs as a highly sensitive CO gas sensor [58]. They showed that the NP size of PdO on SnO2NPs has a significant effect on gas-sensing properties. By reducing the size of the PdONPs, the number of PdO-NPs in contact with SnO2-NPs increases, so the electrical resistance increases by an order of magnitude. This is because PdO acts as a strong acceptor of electrons and removes electrons from the oxide. However, when PdO is reduced to Pd by the reducing gases, the resistance decreases by back-electron transfer from Pd to SnO2. The difference in the electrical resistance of SnO2 induced by a change in the oxidized and reduced states of Pd is often large, giving rise to a large increase in response to the reducing gases. In another study, WO3-NPs doped with cerium as electrode material was developed for monitoring of environmental gases, such as NOx (NO+NO2) by Luo et al. [59]. They showed that the addition of CeO2 reduces the average grain size of Ce added to WO3, and so enhances the surface area. Impedance spectroscopy analysis confirmed the increase in grain-boundary resistance and the decrease in grain-boundary capacitance with increasing concentration of CeO2. This proved that the Ce ions in the WO3 grain boundaries help to improve the sensing ability of the microstructure. When the volatile organic compound gases (xylene, toluene, benzene and acetone) are in contact with the surface of the samples, a surface reaction decreases the coverage of oxygen ions (O or O2 ) and releases electrons to the conduction band of the material, and that increases the conductivity of the samples. 1714
http://www.elsevier.com/locate/trac
5. Outlook and conclusion NPs have proved to be successful in the preparation of improved electrochemical sensors. However, developing new types of NPs for electrochemical-sensor applications with good stability and selectivity remains a challenge. Compared to NP applications in medicine and bionanotechnology, NP applications in electrochemical environmental monitoring are still underdeveloped. Recent research shows that NPs have considerable potential in improving electrochemical-sensor design for the detection of a wide range of environmental contaminants. Most of the research in this area has been devoted to the development of the new methodologies for fabricating new sensors rather than optimizing their performance (sensitivity or selectivity). Furthermore, sensors are rarely tested with real or industrial samples (e.g., wastewater), which show considerable analytical complexity where NP sensors may offer further advantages. It is important to understand NP chemistry, reactivity and possible mechanisms involved in their interaction with the analyte. More work will be required to exploit the recent availability of new nano-building blocks and better surface science tools for the structural and mechanical analysis of processes at NP surfaces. Finally, a wide range of entirely new sensor materials based on nano-composites is now possible (e.g., mixed metal – metal oxide) and interfacial processes at new types of NP–NP interfaces are accessible. The application of these materials in environmental sensing should enable more advanced electrochemical-sensor systems to be devised in the near future.
Acknowledgements L.R. gratefully acknowledges helpful discussions with Dr. Mohammad Khazaei (Institute for Materials Research, Tohoku University, Japan) and the financial support from EPSRC, European Union and TEKES. References [1] (a) X. Huang, M. Sillanpa¨a¨, E.T. Gjessing, S. Pera¨niemi, R.D. Vogt, River Res. Appl. 27 (2011) 113; (b) X. Huang, M. Sillanpa¨a¨, E.T. Gjessing, R.D. Vogt, Sci. Total Environ. 407 (2009) 6242. [2] M. Richards, M. Ghanem, M. Osmond, Y. Guo, J. Hassard, Ecol. Model. 194 (2006) 274. [3] G.F. Fine, L.M. Cavanagh, A. Afonja, R. Binions, Sensors 10 (2010) 5469. [4] G. Hanrahan, D.G. Patil, J. Wang, J. Environ. Monit. 6 (2004) 657. [5] (a) C.M. Welch, R.G. Compton, Anal. Bioanal. Chem. 384 (2006) 601; (b) F.W. Campbell, R.G. Compton, Anal. Bioanal. Chem. 396 (2010) 241; (c) S.E. Ward Jones, R.G. Compton, Curr. Anal. Chem. 4 (2008) 177. [6] S.R. Belding, E.J.F. Dickinson, R.G. Compton, J. Phys. Chem. C 113 (2009) 11149.
Trends in Analytical Chemistry, Vol. 30, No. 11, 2011 [7] H.J. Zhou, F.R.F. Fan, A.J. Bard, J. Phys. Lett. 1 (2010) 2671. [8] C.M.A. Brett, Pure Appl. Chem. 73 (2001) 1969. [9] A. Jang, Z. Zou, K.K. Lee, C.H. Ahn, P.L. Bishop, Meas. Sci. Technol. 22 (2011) 032001. [10] C. Wei, L. Yi, G. Xiao-Ming, W. Ping, Sensor Lett 9 (2011) 137. [11] Q. Yang, Q. Sun, T. Zhou, G. Shi, L. Jin, J. Agric. Food Chem. 57 (2009) 6558. [12] J. Blackman (Editor), Metallic Nanoparticles, Elsevier, Amsterdam, The Netherlands, 2009. [13] (a)C. Granqvist, L. Kish, W. Marlow (Editors), Gas Phase Nanoparticle Synthesis, Springer, Berlin, Germany, 2004; (b) N.V. Mantzaris, Chem. Eng. Sci. 60 (2005) 4749. [14] Y.-G. Zhou, F.W. Campbell, S.R. Belding, R.G. Compton, Chem. Phys. Lett. 497 (2010) 200. [15] L. Rassaei, M. Sillanpa¨a¨, E.V. Milsom, X. Zhang, F. Marken, J. Solid State Electrochem. 12 (2008) 747. [16] D. Yamada, T.A. Ivandini, M. Komatsu, A. Fujishima, Y. Einaga, J. Electroanal. Chem. 615 (2008) 145. [17] L. Rassaei, M. Sillanpa¨a¨, R.W. French, R.G. Compton, F. Marken, Electroanalysis 20 (2008) 1286. [18] G. Chang, M. Oyama, K. Hirao, J. Phys. Chem. B 110 (2006) 1860. [19] L. Han, X. Zhang, Electroanalysis (NY) 21 (2009) 124. [20] D. Du, J. Ding, J. Cai, A. Zhang, Sens. Actuators, B 127 (2007) 317. [21] S. Thiagarajana, R.-F. Yanga, S.-M. Chen, Bioelectrochemistry 75 (2009) 163. [22] Southampton Electrochemistry Group, Instrumental Methods in Electrochemistry, Horwood Publishing, Chichester, West Sussex, UK, 2006. [23] X. Huang, D. Du, X. Gong, J. Cai, H. Tu, X. Xu, A. Zhang, Electroanalysis (NY) 20 (2008) 402. [24] E. Katz, I. Willner, J. Wang. Electroanalysis (NY) 16 (2004) 19. [25] Y. Xiao, F. Patolsky, E. Katz, J.F. Hainfeld, I. Willner, Science (Washington, DC) 299 (2003) 1877. [26] A.P. Periasamy, Y. Umasankar, S.-M. Chen, Sensors 9 (2009) 4034. [27] W. Song, D.-W. Lia, Y.-T. Lia, Y. Lia, Y.-T. Long, Biosens. Bioelectron. 26 (2011) 3181. [28] W. Yang, Y. Bai, Y. Li, C. Sun, Anal. Bioanal. Chem. 382 (2005) 44. [29] D. Du, J. Ding, J. Cai, A. Zhang, J. Electroanal. Chem. 605 (2007) 53. [30] G.-Y. Kim, M.-S. Kang, J. Shim, S.-H. Moon, Sens. Actuators, B 133 (2008) 1. [31] Y. Cui, C. Yang, W. Zeng, M. Oyama, W. Pu, J. Zhang, Anal. Sci. 23 (2007) 1421.
Trends [32] M.-C. Tsai, P.-Y. Chen, Talanta 76 (2008) 533. [33] X. Dai, R.G. Compton, Anal. Sci. 22 (2006) 567. [34] G. Martı´nez-Paredes, M.B. Gonza´lez-Garcı´a, A. Costa-Garcı´a, Electroanalysis (NY) 21 (2009) 925. [35] B.R. Sathe, M.S. Risbud, S. Patil, K.S. Ajayakumar, R.C. Naik, I.S. Mulla, V.K. Pillai, Sens. Actuators, A 138 (2007) 376. [36] M.H. Mashhadizadeh, K. Eskandari, A. Foroumadi, A. Shafiee, Electroanalysis (NY) 20 (2008) 1891. [37] G.-W. Yang, G.-Y. Gao, C. Wang, C.-L. Xu, H.-L. Li, Carbon 46 (2008) 747. [38] G. Maduraiveeran, R. Ramaraj, Electrochem. Commun. 9 (2007) 2051. [39] G.-J. Lee, H.-M. Lee, C.-K. Rhee, Electrochem. Commun. 9 (2007) 2514. [40] Y.-S. Song, G. Muthuraman, J.-M. Zen, Electrochem. Commun. 8 (29) (2006) 1369. [41] A. Taheri, M. Noroozifar, M. Khorasani-Motlagh, J. Electroanal. Chem. 628 (2009) 48. [42] G.-F. Wang, M.-G. Li, Y.-C. Gao, B. Fang, Sensors 4 (2004) 147. [43] X. Gan, T. Liu, X. Zhu, G. Li, Anal. Sci. 20 (2004) 1271. [44] H. Wohltjen, A.W. Snow, Anal. Chem. 70 (1998) 2856. [45] See, for example: (a) Y. Lee, B.K. Oh, M.E. Meyerhoff, Anal. Chem. 76 (2004) 536; (b) J. Gong, T. Zhou, D. Song, L. Zhang, X. Hu, Anal. Chem. 82 (2010) 567; (c) T. Spa˘tarua, N. Spa˘taru, J. Hazard. Mater. 180 (2010) 777. [46] K. Photinon, S.-H. Wang, C.-C. Liu, Sensors 9 (2009) 7203. [47] S. Hrapovic, Y. Liu, J.H.T. Luong, Anal. Chem. 79 (2007) 500. [48] S. Wang, Y. Yin, X. Lin, Electrochem. Commun. 6 (2004) 259. [49] Z.-L. Zhou, T.-F. Kang, Y. Zhang, S.-Y. Cheng, Microchim. Acta 164 (2009) 133. [50] L.B. Israel, N.N. Kariuki, L. Han, M.M. Maye, J. Luo, C.–J. Zhing, J. Electroanal. Chem. 517 (2001) 69. [51] B.K. Jena, C.R. Raj, Anal. Chem. 80 (2008) 5671. [52] R.L. Vander Wal, G.W. Hunter, J.C. Xu, M.J. Kulis, G.M. Berger, T.M. Ticich, Sens. Actuators, B 138 (2009) 113. [53] G. Liu, Y. Lin, Anal. Chem. 77 (2005) 5894. [54] A. Kaushik, P.R. Solanki, A.A. Ansari, B.D. Malhotra, S. Ahmad, Biochem. Eng. J. 54 (2009) 132. [55] M.E. Franke, T.J. Koplin, U. Simon, Small 2 (2006) 36. [56] M.J. Toohey, Sens. Actuators, B 105 (2005) 232. [57] V.K. Khanna, Defence Sci. J. 58 (2008) 608. [58] M. Yuasa, T. Masaki, T. Kida, K. Shimanoe, N. Yamazoe, Sens. Actuators, B 136 (2009) 99. [59] S. Luo, G. Fu, H. Chen, Z. Liu, Q. Hong, Solid-State Electron. 51 (2007) 913.
http://www.elsevier.com/locate/trac
1715
Trends in Analytical Chemistry, Vol. 30, No. 11, 2011
Nanoparticles in electrochemical sensors for environmental monitoring Liza Rassaei, Mandana Amiri, Ciprian Mihai Cirtiu, Markus Sillanpa¨a¨, Frank Marken, Mika Sillanpa¨a¨ We review the state-of-the-art application of nanoparticles (NPs) in electrochemical analysis of environmental pollutants. We summarize methods for preparing NPs and modifying electrode surfaces with NPs. We describe several examples of applications in environmental electrochemical sensors and performance in terms of sensitivity and selectivity for both metal and metal-oxide NPs. We present recent trends in the beneficial use of NPs in constructing electrochemical sensors for environmental monitoring and discuss future challenges. NPs have promising potential to increase competitiveness of electrochemical sensors in environmental monitoring, though research has focused mainly on development of methodology for fabricating new sensors, and the number of studies for optimizing the performance of sensors and the applicability to real samples is still limited. ª 2011 Elsevier Ltd. All rights reserved. Keywords: Air pollution; Amperometry; Electrochemical sensor; Environmental monitoring; Heavy metal; Herbicide; Modified electrode; Nanoparticle; Pesticide; Voltammetry
1. Introduction Liza Rassaei* MESA+ Institute for Nanotechnology, University of Twente, Enschede, The Netherlands Mandana Amiri Department of Chemistry, University of Mohaghegh Ardabili, Ardabil, Iran Ciprian Mihai Cirtiu Department of Chemistry, McGill University, 801 Sherbrooke Street West, Montreal (PQ) H3A 2K6, Canada Markus Sillanpa¨a¨ Research and Innovation Laboratory, Finnish Environment Institute, P.O. Box 140, FI-00251 Helsinki, Finland Frank Marken Department of Chemistry, University of Bath, Bath BA2 7AY, UK Mika Sillanpa¨a¨ Laboratory of Green Chemistry, LUT Energy, Lappeenranta University of Technology, Patteristonkatu 1, 50101 Mikkeli, Finland
*
Monitoring air, soil and water for hazardous pollutants is important and based on the need to protect the environment and public health from possible distribution of natural and industrial inorganic and organic contaminants. There is a constantly increasing need for online monitoring of contaminants in our environment, driven by new legislation and new technologies. Important examples are (i) control of wastewater quality to avoid release of synthetic drugs residues or detergents and (ii) monitoring of persistent carcinogens or residues of explosives released into the environment. Pesticides (herbicides, fungicides and insecticides) are widely used in agriculture and industry. To limit their toxicity and their accumulation in living organisms, dose adjustment and trace-level monitoring are desirable. Thus, there is an essential need to develop new methods for simple pesticide detection at low concentrations, especially in the field.
Heavy metals occur naturally in the environment, but, due to industrialization, large amounts of heavy metals bound in fossil fuels and mineral materials have been released into the environment and deposited in trace amounts in nearly every part of the planet. Elevated levels of heavy metals in natural water may have a detrimental effect on both human health and the environment [1]. Apart from the direct impact on health or environmental problems, water or soil contamination can cause considerable economic and financial damage, so there is a need for novel instruments capable of real-time, in-situ detection of heavy metals and online monitoring. Air pollution is another serious problem, especially in many heavily populated and industrialized areas, and in newly developing countries. In the majority of the developed world, legislation has already been introduced to an extent that local authorities are requested by law to conduct regular local air-quality monitoring of major urban pollutants (e.g., ozone,
Corresponding author. E-mail: [email protected]
1704
0165-9936/$ - see front matter ª 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.trac.2011.05.009
Trends in Analytical Chemistry, Vol. 30, No. 11, 2011
benzene, SO2, CO, CO2 and NOx produced by industrial activity and transportation) [2]. In order to achieve this goal, pollutant concentrations must be monitored accurately, and ideally in situ, so that sources may be recognized quickly and the atmospheric dynamics of the process are understood. Growing concern worldwide about air pollution and its effect on the environment, industrial safety and hygiene has also made it essential to monitor gaseous species (e.g., SO2, H2S, ozone, H2, NOx, CO, Cl2, NH3, formaldehyde, ethanol, propane or ethylene oxide). Furthermore, such data would lead to real-time environmental decision-making capabilities as a result of hazardous levels being rapidly identified [2,3]. Increasing concern about the distribution and the impact of chemical reagents from analytical methodologies has led to new field methods designed to minimize use of toxic reagents during sample pretreatment and analytical measurement. Electrochemical methods for analysis can be especially sensitive, cheap and portable to provide data even in remote locations. A wide range of electrochemical sensors that fulfill the requirement to be environment friendly are commercially available. For example, sensors for water monitoring have been developed, and several of these are now advanced prototypes. Electrochemical monitoring is especially of interest for in-situ measurements with simple, compact and mobile equipment. Typical electrochemical sensors comprise a sensing electrode (as transduction element), a diffusion barrier, a counter-reference electrode and an electrolyte. The analytical information is obtained from the electrical signal that results from the interaction of the target analyte and the recognition layer at the sensing electrode. Depending on the electrical signal to be measured, electrochemical techniques are divided into potentiometric, conductometric, amperometric or voltammetric methods. Different electrochemical devices have been developed for environmental monitoring, depending on the nature of the analyte, the characteristics of the sample matrix and sensitivity or selectivity requirements. The ideal sensor should possess the following characteristics [4]: (1) specificity for the target species; (2) sensitivity to changes in target-species concentrations; (3) fast response time; (4) extended lifetime of at least several months; and, (5) small size (miniaturization) with the possibility of low-cost manufacture. Nanoparticles (NPs) are attracting attention due to their low cost and unique size-dependent properties. The incorporation of NPs into a variety of matrices to form nanocomposite films is attracting much attention. NPs have been used in many electrochemical, electroanalytical and bioelectrochemical applications. The uniqueness of NPs is due to their mechanical, electrical,
Trends
optical, catalytic and magnetic properties as well as their extremely high surface area per mass. In addition to novel properties, nanomaterials and nanotechnology open up new approaches to manufacture electrodes cost effectively by minimizing the materials needed and waste generation [5]. This is especially relevant to expensive materials (e.g., gold and platinum). For example, inexpensive materials (e.g., carbon coated by NPs) result in a large ratio of surface area to volume for low-cost sensing electrodes. In recent studies, it was demonstrated that NP electrodes could be obtained with high sensitivity and even with individual NPs giving responses [6,7]. The combination of nanotechnology with modern electrochemical techniques allows the introduction of powerful, reliable electrical devices for effective process and pollution control. Although the NPs in general play different roles in different electrochemical sensors, with regard to electroanalysis using a NP-modified electrode has several advantages: (1) effective catalysis; (2) fast mass transport; (3) large effective sensor surface area; and, (4) good control over electrode microenvironment [5]. NPs also present attractive characteristics often notably different from bulk materials in terms of both physical and chemical features. However, while several nanomaterial-based sensor systems have been reported in the literature, their implementation in routine in-situ devices remains a challenge. The monitoring of industrial processes with known sample matrix is usually accurate and reliable; however, in-situ monitoring in environmental analysis often with an unknown or changing matrix provides a bigger challenge in the development of measurement devices [8]. Electrochemical sensors have been employed for several decades for a variety of environmental monitoring applications, including monitoring of water-quality parameters (conductivity, dissolved oxygen or pH) [9], measurement of trace heavy metals [10], and carcinogens and organic pollutants (N-nitroso compounds, aromatic amines and phenols) [11]. There is a myriad of different NP materials, but they are usually divided by their chemical nature into six groups: (1) metals; (2) metal oxides; (3) carbonaceous; (4) polymeric; (5) dendrimeric; and, (6) composites. Most relevant in the design of electrochemical sensors are metal and metal-oxide NPs, so these provide the focus of this review. A further topic developing from the wider use of nanotechnology (not covered in this review) is the environmental impact and potential health effects of NPs, which are still largely unknown. http://www.elsevier.com/locate/trac
1705
Trends
Trends in Analytical Chemistry, Vol. 30, No. 11, 2011
Although several review articles have addressed the application of NPs in electroanalysis [3,4], their applications as sensing elements in electrochemical analysis for environmental monitoring have not previously been the focus, so the aim of this review is critically to assess and to overview the role of NPs used in the design of electrochemical sensors for environmental monitoring. The intention is not to provide a complete, exhaustive review of the available literature, but rather to provide insight into potential applications and benefits of NPs for electrochemical environmental monitoring. First, we introduce different methods of NP preparation, then we address electrode-surface modification and sensor preparation, and finally we describe the application and the performance for several examples of reported NP-based sensors for environmental monitoring.
2. Preparing nanoparticles and modifying electrode surfaces 2.1. Colloidal nanoparticles The development of newer methods for the controlled synthesis of metal NPs with different shapes and sizes has attracted attention during the past decade. One of the advantages of using NPs in sensing design is that their size and their surface characteristics can be tuned during the preparation process to achieve high sensitivity or selectivity. The preparation methods for NPs have been reviewed [12]. Different wet chemical methods have been used for the synthesis of metallic NP dispersions. The most commonly employed techniques involve the use of excess reducing agents (e.g., sodium citrate, tannic acid or NaBH4). NPs can also be synthesized in a two-phase water-organic system, or using an organic reducing agent (e.g., aldehydes) or reverse micelles. The concentrations of the salt and the reducing agent and temperature are important parameters to determine the size of NPs. Finally, NPs can be prepared by unconventional methods (e.g., radiolysis, laser ablation, vacuum evaporation, or microwave techniques). Metal-oxide NPs are prepared based on chemical methods (e.g., hydrolysis) and physical methods, including sputtering, evaporation (thermal and electron beam), pulse-laser deposition and ion implantation. Assemblies of NPs can be prepared by using chemical techniques [e.g., sol-gel, co-precipitation, impregnation, or chemical-vapor synthesis (CVS)] or hydrothermal techniques. There have been reviews on growth of NPs in gas and liquid phases [13]. 2.2. Modification of the electrode surface with nanoparticles Various attachment techniques, including physical and chemical methods, are commonly used for the deposition of colloidal NPs. Particle size and size distribution are the 1706
http://www.elsevier.com/locate/trac
most important characteristics of NPs for their performance in electrochemical-sensor design, because, depending on the size and the distance between adjacent particles, the degree of diffusion-layer overlap (the nature of the voltammetric responses) can be manipulated [14]. Defined and ordered arrangement of NPs is a promising approach for the construction of electrochemical sensors. The electrode surface can be modified with colloidal NPs by simply casting a mixture of NPs and additives (e.g., an enzyme). Methods have been developed with solvent evaporation, growth of an NP-modified sol-gel network on the electrode surface, or electro-aggregation. NPs can be modified with different functional groups and then immobilized on the electrode surface. It is possible to pre-mix the NPs with conductive materials (e.g., different types of nano-carbon to prepare a paste electrode) or to trap them in a conductive polymer. The layer-bylayer method based on electrostatic interaction has been used for charged NPs, which get entrapped in an oppositely-charged poly-electrolyte [15]. In many cases, by applying a reducing potential, NPs are directly electrodeposited from the appropriate salt solutions onto the electrode surface so as to have the desired size. These NPdeposition processes are easy to use, although control of the size of the particles may be greatly depend on deposition time, deposition potential, electrolyte solution, and the salt concentration. Fig. 1 shows typical SEM images illustrating the effect of electrodeposition time on the morphology and the density of Au-NPs on borondoped diamond (BDD)-electrode substrates [16]. It can be seen that the longer the deposition time, the larger the average diameters of deposited Au-NPs. With the deposition time of 10 s, 30 s and 60 s, Au-NP diameters were reported as 79.5 ± 11.3 nm, 152.8 ± 11.4 nm, and 222.8 ± 18.3 nm, respectively. At the deposition time of 600 s, the Au-NPs deposited in a multilayer cluster after covering all BDD surfaces. In general, NPs are preferentially deposited on the substrate defects or at the edges, resulting in a non-uniform distribution [17]. Therefore, using pre-synthesized and functionalized colloidal NPs may solve some of these problems when well-defined nanostructured deposits are needed. When the electrode surface is modified using electrodeposition techniques, the pH or the electrode potential applied may cause unwanted dissolution or formation of surface oxides on the electrode surface. Cu-NPs and NiNPs are oxidized easily so their formation on the electrode surface can also be carried out in situ.
3. Electrochemical characterization of nanoparticle-modified electrodes 3.1. Methods Compared to bulk electrodes, the presence of NPs on the electrode surface enables fast electron-transfer kinetics,
Trends in Analytical Chemistry, Vol. 30, No. 11, 2011
Trends
Figure 1. SEM images of a polycrystalline boron-doped diamond (BDD) electrode surface after direct electrochemical Au-NP deposition: deposition time of (a) 10 s, (b) 30 s (c) 60 s and (d) 600 s [16] (reprinted with permission from Elsevier).
reduces overpotential, increases the electro-active surface area, and causes redox reactions to become kinetically feasible [5]. From electrochemical studies of NP-modified electrodes, it is possible to calculate the number of NPs deposited on the electrode surface, capacitive charging effects or specific underpotential deposition (UPD) effects. Many NPs, including Au-NPs and Pt-NPs, are electrochemically-active materials. They possess distinct electrochemical properties because of their unique electronic structures. Fig. 2A shows typical cyclic voltammograms of an indium tin oxide (ITO) electrode before and after Pt-NPs were grown onto the surface [18]. Due to the simple, well-defined responses, redox models (e.g., the [Fe(CN)6]4/[Fe(CN)6]3 or 3+ 2+ Ru(NH3)6 /Ru(NH3)6 couples) have been widely used to characterize the surface properties of different NPmodified electrodes (see Fig. 2B) [19]. The voltammetry of these two standard redox processes is examined in order to investigate how the NP-surface coverage and electron-transfer kinetics are compared to unmodified electrodes. The electrochemical properties of NP-modified electrodes significantly depend on their surface properties. In cyclic voltammetry experiments, the peakto-peak separation and electron-transfer kinetics of NPmodified electrodes have been shown to change with the amount of NPs. However, there is an optimal coverage for the amount of NPs on the surface. An excess amount of NPs usually gives no further improvement or increases the resistance and the double-layer capacitance of the modified electrode, leading to a decrease in electrochemical sensitivity [20].
Electrochemical impedance spectroscopy (EIS) can also be employed to investigate the changes in impedance of the electrode surface as a function of coverage for NP-modified electrodes. Fig. 2C shows typical electrochemical-impedance spectra of the [Fe(CN)6]3/ [Fe(CN)6]4 redox couples for unmodified and modified electrodes with Pd-NPs. It can be observed that the semicircle diameter of the Nyquist plot for [Fe(CN)6]3/ [Fe(CN)6]4 redox has significantly changed with the loading of Pd-NPs on the substrate [21]. The apparent standard heterogeneous rate constant (assuming [Fe(CN)6]3 = [Fe(CN)6]4) is obtained using [22] k app ¼ RT =ðF 2 Rct AC Þ where Rct, T, F, A and C* represent the gas constant, the absolute temperature, the Faraday constant, the true area and the bulk concentration of the redox couple, respectively [22]. Generally, the kapp values for the [Fe(CN)6]3/[Fe(CN)6]4 redox couple on NP arrays prepared under different growth conditions indicate increasingly facile heterogeneous electron-transfer kinetics, mainly resulting from an increase in active area. For environmental monitoring, NPs are mostly used in two different ways in constructing electrochemical sensors: (1) together with an enzyme (enzymatic sensors) to increase the rate of electron transfer and to improve the sensitivity of the enzyme; or, (2) as individual catalysts (non-enzymatic sensors) to catalyze the reactions and to lower the limits of detection (LODs). As their function in these sensors is different, these two cases are discussed separately.
http://www.elsevier.com/locate/trac
1707
Trends
Trends in Analytical Chemistry, Vol. 30, No. 11, 2011
3.2. Enzymatic sensors A combination of enzymatic reactions with electrochemical methods allows development of selective enzyme-based electrochemical biosensors for determination of environmental pollutants. These sensors, although less robust, exhibit good selectivity, sensitivity, rapid response, and have miniature size. In environmental monitoring, enzyme-based or protein-based NP sensors have mostly been prepared for detection of pesticides (e.g., phenolic compounds) or NOx compounds. A major challenge to develop such sensitive and stable sensors comes from the effective immobilization and ‘‘electrical connection’’ of enzymes to solid electrode surfaces. The most promising areas for the use of enzyme electrodes in environmental monitoring are pesticide biosensors. Enzyme electrodes have been investigated as emerging sensors for faster, simpler detection of pesticides. However, their sensitivity is not yet adequate for application in detecting very low concentrations of pesticides. External supply of substrate is often required to measure the changes in enzyme activity. Compared to free enzyme in solution, immobilized enzyme can be more stable and resistant to various environmental changes. Electrochemical sensors incor-
porating enzymes with nanomaterials combine the recognition and catalytic properties of enzymes with the electronic properties of NPs. The NPs can help enlarge the enzyme loading at the electrode surface. The NPs with smaller size (1–2 nm) have been shown to be more suitable for enzyme immobilization and they increase the biosensor performance because the small size of these NPs results in a significant surface-to-volume ratio and creates more binding sites on the electrode surface for easier contact with enzyme molecules and less denaturation [23]. NPs of small size also allow more freedom in orientating adsorbed proteins, or they introduce conducting channels between the prosthetic groups within the enzyme and the electrode surface. In other words, they maximize the utilization of the bioactive sites of the enzyme and act as electron-transfer pathways [23]. Due to their high surface area and good biocompatibility, surface-functionalized NPs are suitable for many surface-immobilization mechanisms, providing a suitable support to adsorb redox enzymes and proteins. NPs of appropriate design can adsorb redox enzymes and proteins without loss of biological activity. Moreover, since most of the NPs carry charges, they can also electrostatically adsorb biomolecules with opposite
Figure 2. (A) Cyclic voltammograms (scan rate of 50 mV/s) recorded in the solution of H2SO4 for (a) a bare ITO electrode and (b,c) the Pt-NPsITO electrode prepared via (b) 4 and (c) 24 h of growth for a geometric area of 0.031 cm2 [18]. (B) Cyclic voltammograms (scan rate 50 mV/s) of K3[Fe(CN)6] at bare Au electrode (radius of 1.5 mm) and Au-NPs on GCE (radius of 1.5 mm) [19], and (C) Nyquist diagrams (ZÕÕ vs. ZÕ) for the EIS measurements (Amplitude: 5 mV) in 5 mmol/L K3Fe(CN)6/K4Fe(CN)6 in 0.1 mol/L PBS at E1/2 = 0.13 V vs. Ag/AgCl for the bare GCE (h) and Pd-NPs-GCE (o) [21] [(A) reprinted with permission from American Chemical Society, (B) ª Wiley-VCH Verlag GmbH & Co. KGaA, reproduced with permission, and (C) reprinted with permission from Elsevier].
1708
http://www.elsevier.com/locate/trac
Trends in Analytical Chemistry, Vol. 30, No. 11, 2011
charges or selectively bind to active regions on the biomolecule surface. Enzymes usually lack direct electrical communication with the electrode substrate due to their active centers being surrounded by a protein shell. Electron transfer between the electrodes and the active centers is blocked. The conductivity of NPs enhances the rate of direct electron transfer between the active site of the enzymes and electrodes acting as electrical wires. Among different types of NPs, Au-NPs, Pt-NPs, and Ag-NPs have shown biocompatibility with enzymes (e.g., in several recent publications, it was shown that Au-NPs could be used as a conduit for electrons from enzyme redox reactions [24]). Due to their biocompatibility, Au-NPs are considered an excellent option for replacing potentially harmful mediators in the construction of biosensors [25]. The electrode-surface environment and the amount of NPs deposited are also crucial in design of enzyme-based electrochemical sensors. Among different enzymes or proteins commercially available, acetyl cholinesterase (AChE) and tyrosinase (TY) and, in a few cases, hemoglobin protein (Hb) have been used in construction of enzyme-based sensors for environmental monitoring. The application of AChE-nanomaterial-modified electrodes for analysis of pesticides was recently reviewed by Periasamy et al. [26]. In brief, in most AChE-based sensors, organophosphorous (OP)-pesticide detection is mainly based on the irreversible inhibition of AChE activity by OP pesticides. The degree of inhibition is calculated by comparison of the residual activity of the enzyme with the initial activity. TY-NP-based electrodes are usually used for detection of phenolic compounds. The mechanism is based on the oxidation of phenolic compounds (e.g., herbicide atrazine) by tyrosinase to catechol and further to quinone [27]. Hemoglobin (Hb) is known for its catalytic activity towards the reduction of nitrite. However, it is difficult for Hb in solution to exchange electrons directly with bare solid electrodes, as its electroactive center (heme) is buried in its electrochemically-insulated peptide backbone and, furthermore, its adsorption on the electrode surface deteriorates its bioactivity and results in denaturation. NPs may provide a mediator to facilitate electron exchange for Hb [28]. A more uniform film can be obtained when NPs are deposited in a polymer, for example, Au-NPs were codeposited with a biocompatible polymer (chitosan) on the electrode surface to obtain more uniform films [29]. The AChE was then immobilized on this electrode to provide an amperometric sensor to measure different OP pesticides (e.g., malathion, pralidoxime iodide, and monocrotopos). The peak current for oxidation of thiocholine (hydrolysis product of AChE) increased. A negative shift in peak potential compared to bare elec-
Trends
trode or chitosan-modified electrode was related to the presence of Au-NPs and their catalytic behavior. Although enzymatic sensors are more suitable for environmental monitoring, for rapid and simple measurements in the field, their application is still limited, as they need the substrate to activate or to maintain enzyme activity over a lengthy storage time [25,30].
3.3. Non-enzymatic sensors The sensing ability of NPs in non-enzymatic sensors is directly related to their catalytic property. For example, using a seed-mediated growth technique, Cui et al. deposited Au-NPs (mean size 4 nm) on a glassy carbon (GC) electrode [31], which showed excellent catalytic activity towards the oxidation of nitrite in a real wastewater sample without any major interference. However, the catalytic properties of NPs depend on the protecting groups or the presence of additives (e.g., cysteine or tannic acid). Controlling NP size, it is possible to change the catalytic activity of NPs so that it becomes sensitive to different oxidation states of one species. For example, controlling the size of Au-NPs allows Cr(VI) to be measured electrochemically in the presence of excess of Cr(III), offering benefits over the commonly-used diethylenetriamine-pentaacetic-acid (DTPA) method [32]. The stability and activity of this electrode depended on the size and the density of the NPs. Larger NPs were shown to cause not only a negative shift in peak potential and lower peak current but also deterioration in the stability due to the blocking electron transfer between solution species and electrode. This sensor was applied to measure Cr(VI) in tap water, stream water, and sea water [32]. NPs provide a larger surface area for detection of analytes, leading to electrochemical sensors with lower LODs. For example, Au is known to be the most appropriate material to measure arsenic but the analysis of arsenic by the anodic stripping voltammetry (ASV) method is limited by the formation of a monolayer of As0 deposition due to non-conductive property of the As-Au compound. Au-NPs with large surface area are used to overcome this limitation in many studies, but their performance depends on their size, shape and distribution during the nucleation process, which is followed by particle growth [17,33]. Another example is the detection of lead. Lead in acidic media shows UPD on Au electrodes but the maximum response of this process is limited by the electrode area, so the use of Au-NPs with large surface area can solve this problem [34]. Studies have also shown that increasing the surface roughness of a platinum electrode via platinization provides more active sites for catalyzing the oxidation of CO to CO2, leading to increase in sensitivity of this sensor [35]. NPs can also be used in preparing paste electrodes. In this type of electrode, the presence of NPs increases the
http://www.elsevier.com/locate/trac
1709
Trends
Trends in Analytical Chemistry, Vol. 30, No. 11, 2011
surface area of the paste and diminishes the Ohmic resistance of the paste [36]. The small size of NPs and their resulting large surface area can lead to NP-NP aggregation when preparing the sensor. To avoid this, the surface of NPs can be functionalized with thiol groups or other selfassembled monolayers (SAMs) containing amino or carboxylic acid groups. These functional groups introduce new functions in sensor design. In the case of Au-NPs (mean size 5 nm), SAMs stabilize the particles by preventing aggregation. They can also provide an insulating medium between metallic gold cores [18]. In the case of highly monodispersed benzyl mercaptane-stabilized Ag-NP-modified carbon nanotubes (CNTs) on a GC electrode, the phenyl rings of benzyl mercaptane attach to the side wall of CNTs and the free thiol groups at the end of the molecules interact with Ag-NPs to form strong S-Ag bonds. The benzyl mercaptane prevents Ag-NPs from aggregation and promotes interaction with CNTs via p–p interac-
tions to give a uniform dispersion of Ag-NPs on CNTs [37].
4. Sensitivity and selectivity 4.1. Sensors based on metal nanoparticles Metal NPs (e.g., Au, Pt, Pd and Ag) show catalytic activity when they are incorporated into electrochemical sensors. For example, an electrochemical sensor based on Au-NPs (mean diameter 4–6 nm) embedded in 3aminopropyl)triethoxysilane (APS)-derived silicate sol– gel three-dimensional network (APS–Au-NPs) was prepared for simultaneous detection of N2H4, SO32 and NO2 (see Fig. 3). The electro-oxidation peaks appeared at 0.05 V, 0.20 V, and 0.55 V vs. Ag/AgCl for N2H4, SO32 and NO2 in pH 7.2 at the APS–Au-NP-modified GC electrode with a large decrease in the overpotentials to the extent of 750 mV, 600 mV and 250 mV, respectively [38]. Ion-exchange membranes [e.g., Nafion (a
Figure 3. Cyclic voltammograms (scan rate 50 mV/s) at (a) GCE, (b) GCE/APS, (c) GCE/APS-Au-NPs for (A) 250 lM N2H4, (B) SO32 and (C) nitrite (D) (scan rate 20 mV/s) for a mixture containing 250 lM of N2H4, SO32 -and nitrite in 0.1 M PBS pH = 7.2 [38] (reprinted with permission from Elsevier).
1710
http://www.elsevier.com/locate/trac
Trends in Analytical Chemistry, Vol. 30, No. 11, 2011
negatively charged membrane)] can be used beneficially in the sensor design. A screen-printed electrode was modified with a paste of Bi-NPs (mean diameter 50 nm) and Nafion as a binder to help the NPs to adhere to the electrode surface. Well-defined and highly-reproducible electrochemical responses were obtained for detection of Zn(II), Cd(II) and Pb(II) using ASV [39]. The presence of Nafion causes the peak potentials to shift to more negative values due to the cation-exchange behavior of the Nafion coating, which favors an oxidized state of the metal ions. Coating the composite electrode with a Nafion membrane also helped in avoiding anionic interference in the sample (NO3, SO42, CO32). A poly (L -lactide)-stabilized Au-NP-modified (PLA-Au film) electrode has been used for the indirect detection of sulfide by measuring the inhibited oxidation current of As(III) in acidic medium [40]. A substantial decrease in oxidation peak current for As(III) in the presence of sulfide was obtained and it was related to the formation of As2S3 compound. Using this approach with differential pulse stripping voltammetry (DPSV), the amount of sulfide in hot spring water and acid rain was determined with an LOD of 0.04 lM. The following mechanism was suggested for this reaction based on arsenite existing in the form of H3AsO3 in acidic media (see Fig. 4). Na2 S þ 2HCl ! H 2 S þ 2NaClð1Þ 2H 3 AsO3 þ 3H 2 S ! As2 S3 þ 6H 2 Oð2Þ The detection of cyanide and thiocyanide anions in wastewater and industrial samples was reported at Ag-NP-modified electrodes based on the complexation reaction of Ag+ with these anions. The electron-transfer reaction in these processes is controlled by the adsorption of CN or SCN. For example, Taheri et al. developed a cyanide sensor based on mercaptopropyltrimethoxy-silane (MPS)-stabilized Ag-NPs embedded in 3D sol-gel matrix on a gold electrode [41]. MPS acts here as a capping agent. The cathodic current for an Ag-NP-modified electrode decreases in the presence of cyanide ions because of the formation of the Ag+–CN complex. The peak current decreases due to the blockage of the electrode surface by this complex so that less surface area of NPs is available. This electrode was successfully applied for determination of cyanide in industrial samples. In another study, an oleate-stabilized Ag-NP (mean diameter 22 nm)-modified GC electrode was prepared as an amperometric sensor for thiocyanate. In the presence of SCN, a new oxidation peak appeared, due to the formation of the Ag+–SCN complex and the anodic peak current for Ag-NPs decreased due to the electro-reductive activity of Ag-NPs [42]. Ag-NPs with relatively high specific surface area can integrate proteins and help in orientation of adsorbed proteins. Gan et al. modified the pyrolytic graphite-
Trends
electrode surfaces with bioconjugates of Ag-NPs (mean diameter 11 nm) and hemoglobin for determination of nitric oxide [43]. The UV-Vis absorption spectrum confirmed a similar Soret band (405.6 nm) for both Hb with and without Ag-NPs, consistent with Ag-NPs preserving the natural conformation of Hb. The increased rate of electron transfer confirmed the electron-transfer process between protein and electrode was effective. Both electrochemical reactivity of the protein and its detection sensitivity toward NO were enhanced in presence of AgNPs. NPs fixed in an organic matrix have proved to be very useful in construction of electrochemical sensors. The NPs represent the electronic conductivity while the organic matrix has the selective binding sites for selective adsorption of analyte molecules. An interesting part of this approach is the ability to control the sensor properties by molecular design. The charge transport in such NP assemblies sensitively depends on the particle size, the interparticle spacing and the chemical composition (i.e. the electronic structure of the stabilizing or interconnecting ligands). Thus, selective absorption of analytes to the organic matrix among the NPs can be utilized to affect the electrical transport in such hybrid systems (e.g., Wohltjen and Snow developed a colloidal metal-insulator-metal ensemble chemiresistor sensor based on a monolayer of octanethiol-stabilized Au-NPs transducer film for ppm detection of toluene and tetrachloroethylene [44]). Pt-NPs have been used in construction of electrochemical sensors for detecting CO, CO2, NO2, NO, formaldehyde, methanol, naphthols, phenol, mercury and As(III). Pt-NPs can oxidize CO to CO2 in the presence of interferences (e.g., Cl2, SO2, H2S, NO2 and NO) and have been employed for applications in environmental pollution monitoring [45]. The adsorption of CO was shown to be purely chemical and irreversible at Pt-NP (mean diameter 18 nm)-modified electrodes. In this case, CO oxidation occurred through the reaction of adsorbed CO with Pt-OH to form CO2 following the reaction pathway indicated by following equation: CO þ H 2 O ! CO2 þ 2H þ þ 2e
ð3Þ
Photinon et al. showed that the catalytic activity of PtNPs toward oxidation of CO and the sensor response depend on the crystallographic orientation of platinum deposits [46]. It is known that CO2 adsorbs preferentially on the hydrogen-treated Pt surfaces (Pt-H), so detection of Pt-HCOO in acidic media on Pt surfaces is a method to quantify CO2 in a gaseous test environment. The anodic adsorptive stripping of CO2 at high and low concentrations shows the inverse relationship between Pt-H and Pt-HCOO, as depicted in Fig. 5. Electrochemical measurement of As is usually accompanied by interference from Cu. However, Pt-NPmodified electrodes was shown to be less sensitive to Cu,
http://www.elsevier.com/locate/trac
1711
Trends
Trends in Analytical Chemistry, Vol. 30, No. 11, 2011
Figure 4. Indirect detection of sulfide by measuring the inhibited oxidation current of As(III) in acidic medium at poly (L -lactide) stabilized AuNP-modified (PLA-Au film) electrode [40] (with kind permission from Elsevier).
especially for low concentrations of As (less than 100 ppm) [47]. The mechanism suggested for this detection strategy involved formation of PtOH on the electrode surface followed by binding As(OH)3 and electron transfer. The characteristics of mixed-metal NPs, especially their catalytic property, are quite distinct from those of pure-metal NPs. Also, their nano-composite structures are of interest. Very few studies have addressed the codeposition of Pt-NPs and a transition-metal oxide for improving efficient catalysis. For example, co-deposition of Pt-NPs (mean diameter 6 nm) and Fe(III) on a GC electrode resulted in a sensor with good catalytic ability toward the oxidation of nitrite. The Pt-NPs were caged by the Fe(III) network formed by the Fe–O–Fe linkages,
1712
http://www.elsevier.com/locate/trac
and the presence of Fe(III) was shown to be an important factor for the kinetics of NP growth. This sensor showed good response for detection of nitrite in the presence of interferences (e.g., 500-fold Na+, K+, Mg2+, Ni2+, Cu2+, Cl, NO3, SO42, CO32; 200-fold H2O2; 100-fold SO32, S2O32, L -glutamic acid; and, 50-fold dopamine, uric acid, ascorbic acid, glucose, D -fructose, sucrose, oxalic acid, citric acid, D -galactose and malic acid). An LOD of 0.47 lM was obtained for this electrode [48]. In another study, the co-deposition of Pt and Pd-NPs on a Nafion-modified GC electrode was reported by Zhou et al. [49]. This sensor showed remarkable electrocatalytic activity towards oxidation of formaldehyde in aqueous solution of H2SO4 with a broad linear range and
Trends in Analytical Chemistry, Vol. 30, No. 11, 2011
Trends
Figure 5. Sketches for anodic stripping of CO2 at low and high concentrations showing the inverse relationship between Pt-H and Pt-HCOO [46]. E1 is the potential where CO2 adsorbs on the Pt-H surface (0.3 V vs. RHE). E2 is the potential where the non-reacted Pt-H is oxidized from the electrode (0.5 V vs. RHE). E3 is the potential where Pt-HCOO is oxidized (0.75 V vs. RHE).
a low LOD. The Pt–Pd alloy has a synergistic effect to minimize the poisoning of the electrode surface by reaction intermediates, particularly CO. Specific interactions between the analyte species and the organic-shell functional groups on NPs could enhance sensitivity and increase the selectivity of the sensor. For example, a GC electrode modified with Au-NPs capped with 11-mercaptoundecanoic acid units was used for detection of Cu2+ ions and a sensitivity below 1 ppb with improved selectivity in the presence of Fe3+ and Zn2+ ions was reported [50]. Multilayers of conductive NPs gave rise to a large porous surface-area electrode, where the local microenvironment of the metallic NPs was controlled by the cross-linking elements. Jena et al. modified an Au-electrode surface with a thiol-functionalized sol-gel derived from 3D silicate networks via 3-(mercaptopropyl)-trimethoxy-silane (MPTS) and soaked the electrode in citrate stabilized AuNPs (mean size 5–6 nm) solution for self assembly of AuNPs on the thiol groups. A random distribution of Au-NPs through the network with a size distribution of 70–100 nm was obtained. It was shown that the overlap between the diffusion layers at individual Au-NPs produced a diffusion layer linear to the geometrical surface area of the electrode. This electrode showed a good performance for simultaneous detection of As(III), Hg(II) and Cu(II) (5 ppb) in a real sample from an arseniccontaminated area in West Bengal [51].
4.2. Sensors based on metal-oxide nanoparticles Metal oxides provide robust building blocks in nanostructures and offer functionality from electrically conducting to insulating and from highly catalytic to inert. Application of metal oxides in sensors, although less prevalent, is therefore very promising [52]. Zirconia (ZrO2)NPs are selective sorbents in electrochemical sensors for detection of OP pesticides and nerve agents. Zirconia has a strong affinity towards phosphorous groups, so nitroaromatic OPs strongly bind to the ZrO2-NP surfaces [53]. The ZrO2-NP also provide a large surface area and increase the interaction with OP compounds (e.g., methyl parathion, paraoxone and fentrothion). Iron-oxide NPs have been used as the immobilizing matrix for applications in biosensing due to their good biocompatibility, low toxicity, high electron-transfer capability, and high adsorption ability. Iron-oxide NPs provide a favorable environment for immobilization of many biological molecules. For example, Kaushik et al. immobilized single-standard calf-thymus DNA onto chitosan-Fe3O4 hybrid nanobiocomposite film (Fe3O4NPs, mean diameter 22 nm) and deposited it onto an ITO substrate for detection of pyrethroids (insecticides) (e.g., cypermethirn (CM) and permethrin (PM) [54]). The presence of Fe3O4-NPs on the ITO electrode surface increased the electroactive surface area and also promoted electron transfer due to the uniform dispersion throughout the chitosan network on the electrode.
http://www.elsevier.com/locate/trac
1713
Trends
Trends in Analytical Chemistry, Vol. 30, No. 11, 2011
Metal-oxide semiconductors (e.g., WO3, TiO2, ZnO and SnO2) are widely used in gas sensors, having high detection ability and stability. However, the application of NPs as gas-sensing elements is usually limited to chemiresistors, which have been reviewed by Franke et al. [55]. The electrical resistance of chemiresistors changes in the presence of an oxidizing or reducing gas. Many gas sensors operate by detecting changes in the electrical properties of a nanostructured sensing-film material in the presence of a test gas. A review by Toohey addresses the influence of electrode material and NP geometry on sensitivity and selectivity of gas sensors [56]. The gas-sensing NPs (e.g., ZnO, SnO2, TiO2, and WO3) can be deposited on a gold or platinum electrode as a substrate. These sensors operate through transference in the equilibrium of the surface-oxygen reaction by the presence of the target analyte. A reducing gas increases the conductivity of an n-type semiconductor and decreases the conductivity of a p-type semiconductor, and an oxidizing gas behaves vice versa. The use of NP metal oxides increases the sensitivity of the sensor when the surface-oxygen chemisorption reaction determines the sensor response [57]. Mixed metal-oxide-semiconductor NPs enhance the performance of the electrochemical sensors. For example, Yuasa et al. loaded PdO-NPs on SnO2-NPs as a highly sensitive CO gas sensor [58]. They showed that the NP size of PdO on SnO2NPs has a significant effect on gas-sensing properties. By reducing the size of the PdONPs, the number of PdO-NPs in contact with SnO2-NPs increases, so the electrical resistance increases by an order of magnitude. This is because PdO acts as a strong acceptor of electrons and removes electrons from the oxide. However, when PdO is reduced to Pd by the reducing gases, the resistance decreases by back-electron transfer from Pd to SnO2. The difference in the electrical resistance of SnO2 induced by a change in the oxidized and reduced states of Pd is often large, giving rise to a large increase in response to the reducing gases. In another study, WO3-NPs doped with cerium as electrode material was developed for monitoring of environmental gases, such as NOx (NO+NO2) by Luo et al. [59]. They showed that the addition of CeO2 reduces the average grain size of Ce added to WO3, and so enhances the surface area. Impedance spectroscopy analysis confirmed the increase in grain-boundary resistance and the decrease in grain-boundary capacitance with increasing concentration of CeO2. This proved that the Ce ions in the WO3 grain boundaries help to improve the sensing ability of the microstructure. When the volatile organic compound gases (xylene, toluene, benzene and acetone) are in contact with the surface of the samples, a surface reaction decreases the coverage of oxygen ions (O or O2 ) and releases electrons to the conduction band of the material, and that increases the conductivity of the samples. 1714
http://www.elsevier.com/locate/trac
5. Outlook and conclusion NPs have proved to be successful in the preparation of improved electrochemical sensors. However, developing new types of NPs for electrochemical-sensor applications with good stability and selectivity remains a challenge. Compared to NP applications in medicine and bionanotechnology, NP applications in electrochemical environmental monitoring are still underdeveloped. Recent research shows that NPs have considerable potential in improving electrochemical-sensor design for the detection of a wide range of environmental contaminants. Most of the research in this area has been devoted to the development of the new methodologies for fabricating new sensors rather than optimizing their performance (sensitivity or selectivity). Furthermore, sensors are rarely tested with real or industrial samples (e.g., wastewater), which show considerable analytical complexity where NP sensors may offer further advantages. It is important to understand NP chemistry, reactivity and possible mechanisms involved in their interaction with the analyte. More work will be required to exploit the recent availability of new nano-building blocks and better surface science tools for the structural and mechanical analysis of processes at NP surfaces. Finally, a wide range of entirely new sensor materials based on nano-composites is now possible (e.g., mixed metal – metal oxide) and interfacial processes at new types of NP–NP interfaces are accessible. The application of these materials in environmental sensing should enable more advanced electrochemical-sensor systems to be devised in the near future.
Acknowledgements L.R. gratefully acknowledges helpful discussions with Dr. Mohammad Khazaei (Institute for Materials Research, Tohoku University, Japan) and the financial support from EPSRC, European Union and TEKES. References [1] (a) X. Huang, M. Sillanpa¨a¨, E.T. Gjessing, S. Pera¨niemi, R.D. Vogt, River Res. Appl. 27 (2011) 113; (b) X. Huang, M. Sillanpa¨a¨, E.T. Gjessing, R.D. Vogt, Sci. Total Environ. 407 (2009) 6242. [2] M. Richards, M. Ghanem, M. Osmond, Y. Guo, J. Hassard, Ecol. Model. 194 (2006) 274. [3] G.F. Fine, L.M. Cavanagh, A. Afonja, R. Binions, Sensors 10 (2010) 5469. [4] G. Hanrahan, D.G. Patil, J. Wang, J. Environ. Monit. 6 (2004) 657. [5] (a) C.M. Welch, R.G. Compton, Anal. Bioanal. Chem. 384 (2006) 601; (b) F.W. Campbell, R.G. Compton, Anal. Bioanal. Chem. 396 (2010) 241; (c) S.E. Ward Jones, R.G. Compton, Curr. Anal. Chem. 4 (2008) 177. [6] S.R. Belding, E.J.F. Dickinson, R.G. Compton, J. Phys. Chem. C 113 (2009) 11149.
Trends in Analytical Chemistry, Vol. 30, No. 11, 2011 [7] H.J. Zhou, F.R.F. Fan, A.J. Bard, J. Phys. Lett. 1 (2010) 2671. [8] C.M.A. Brett, Pure Appl. Chem. 73 (2001) 1969. [9] A. Jang, Z. Zou, K.K. Lee, C.H. Ahn, P.L. Bishop, Meas. Sci. Technol. 22 (2011) 032001. [10] C. Wei, L. Yi, G. Xiao-Ming, W. Ping, Sensor Lett 9 (2011) 137. [11] Q. Yang, Q. Sun, T. Zhou, G. Shi, L. Jin, J. Agric. Food Chem. 57 (2009) 6558. [12] J. Blackman (Editor), Metallic Nanoparticles, Elsevier, Amsterdam, The Netherlands, 2009. [13] (a)C. Granqvist, L. Kish, W. Marlow (Editors), Gas Phase Nanoparticle Synthesis, Springer, Berlin, Germany, 2004; (b) N.V. Mantzaris, Chem. Eng. Sci. 60 (2005) 4749. [14] Y.-G. Zhou, F.W. Campbell, S.R. Belding, R.G. Compton, Chem. Phys. Lett. 497 (2010) 200. [15] L. Rassaei, M. Sillanpa¨a¨, E.V. Milsom, X. Zhang, F. Marken, J. Solid State Electrochem. 12 (2008) 747. [16] D. Yamada, T.A. Ivandini, M. Komatsu, A. Fujishima, Y. Einaga, J. Electroanal. Chem. 615 (2008) 145. [17] L. Rassaei, M. Sillanpa¨a¨, R.W. French, R.G. Compton, F. Marken, Electroanalysis 20 (2008) 1286. [18] G. Chang, M. Oyama, K. Hirao, J. Phys. Chem. B 110 (2006) 1860. [19] L. Han, X. Zhang, Electroanalysis (NY) 21 (2009) 124. [20] D. Du, J. Ding, J. Cai, A. Zhang, Sens. Actuators, B 127 (2007) 317. [21] S. Thiagarajana, R.-F. Yanga, S.-M. Chen, Bioelectrochemistry 75 (2009) 163. [22] Southampton Electrochemistry Group, Instrumental Methods in Electrochemistry, Horwood Publishing, Chichester, West Sussex, UK, 2006. [23] X. Huang, D. Du, X. Gong, J. Cai, H. Tu, X. Xu, A. Zhang, Electroanalysis (NY) 20 (2008) 402. [24] E. Katz, I. Willner, J. Wang. Electroanalysis (NY) 16 (2004) 19. [25] Y. Xiao, F. Patolsky, E. Katz, J.F. Hainfeld, I. Willner, Science (Washington, DC) 299 (2003) 1877. [26] A.P. Periasamy, Y. Umasankar, S.-M. Chen, Sensors 9 (2009) 4034. [27] W. Song, D.-W. Lia, Y.-T. Lia, Y. Lia, Y.-T. Long, Biosens. Bioelectron. 26 (2011) 3181. [28] W. Yang, Y. Bai, Y. Li, C. Sun, Anal. Bioanal. Chem. 382 (2005) 44. [29] D. Du, J. Ding, J. Cai, A. Zhang, J. Electroanal. Chem. 605 (2007) 53. [30] G.-Y. Kim, M.-S. Kang, J. Shim, S.-H. Moon, Sens. Actuators, B 133 (2008) 1. [31] Y. Cui, C. Yang, W. Zeng, M. Oyama, W. Pu, J. Zhang, Anal. Sci. 23 (2007) 1421.
Trends [32] M.-C. Tsai, P.-Y. Chen, Talanta 76 (2008) 533. [33] X. Dai, R.G. Compton, Anal. Sci. 22 (2006) 567. [34] G. Martı´nez-Paredes, M.B. Gonza´lez-Garcı´a, A. Costa-Garcı´a, Electroanalysis (NY) 21 (2009) 925. [35] B.R. Sathe, M.S. Risbud, S. Patil, K.S. Ajayakumar, R.C. Naik, I.S. Mulla, V.K. Pillai, Sens. Actuators, A 138 (2007) 376. [36] M.H. Mashhadizadeh, K. Eskandari, A. Foroumadi, A. Shafiee, Electroanalysis (NY) 20 (2008) 1891. [37] G.-W. Yang, G.-Y. Gao, C. Wang, C.-L. Xu, H.-L. Li, Carbon 46 (2008) 747. [38] G. Maduraiveeran, R. Ramaraj, Electrochem. Commun. 9 (2007) 2051. [39] G.-J. Lee, H.-M. Lee, C.-K. Rhee, Electrochem. Commun. 9 (2007) 2514. [40] Y.-S. Song, G. Muthuraman, J.-M. Zen, Electrochem. Commun. 8 (29) (2006) 1369. [41] A. Taheri, M. Noroozifar, M. Khorasani-Motlagh, J. Electroanal. Chem. 628 (2009) 48. [42] G.-F. Wang, M.-G. Li, Y.-C. Gao, B. Fang, Sensors 4 (2004) 147. [43] X. Gan, T. Liu, X. Zhu, G. Li, Anal. Sci. 20 (2004) 1271. [44] H. Wohltjen, A.W. Snow, Anal. Chem. 70 (1998) 2856. [45] See, for example: (a) Y. Lee, B.K. Oh, M.E. Meyerhoff, Anal. Chem. 76 (2004) 536; (b) J. Gong, T. Zhou, D. Song, L. Zhang, X. Hu, Anal. Chem. 82 (2010) 567; (c) T. Spa˘tarua, N. Spa˘taru, J. Hazard. Mater. 180 (2010) 777. [46] K. Photinon, S.-H. Wang, C.-C. Liu, Sensors 9 (2009) 7203. [47] S. Hrapovic, Y. Liu, J.H.T. Luong, Anal. Chem. 79 (2007) 500. [48] S. Wang, Y. Yin, X. Lin, Electrochem. Commun. 6 (2004) 259. [49] Z.-L. Zhou, T.-F. Kang, Y. Zhang, S.-Y. Cheng, Microchim. Acta 164 (2009) 133. [50] L.B. Israel, N.N. Kariuki, L. Han, M.M. Maye, J. Luo, C.–J. Zhing, J. Electroanal. Chem. 517 (2001) 69. [51] B.K. Jena, C.R. Raj, Anal. Chem. 80 (2008) 5671. [52] R.L. Vander Wal, G.W. Hunter, J.C. Xu, M.J. Kulis, G.M. Berger, T.M. Ticich, Sens. Actuators, B 138 (2009) 113. [53] G. Liu, Y. Lin, Anal. Chem. 77 (2005) 5894. [54] A. Kaushik, P.R. Solanki, A.A. Ansari, B.D. Malhotra, S. Ahmad, Biochem. Eng. J. 54 (2009) 132. [55] M.E. Franke, T.J. Koplin, U. Simon, Small 2 (2006) 36. [56] M.J. Toohey, Sens. Actuators, B 105 (2005) 232. [57] V.K. Khanna, Defence Sci. J. 58 (2008) 608. [58] M. Yuasa, T. Masaki, T. Kida, K. Shimanoe, N. Yamazoe, Sens. Actuators, B 136 (2009) 99. [59] S. Luo, G. Fu, H. Chen, Z. Liu, Q. Hong, Solid-State Electron. 51 (2007) 913.
http://www.elsevier.com/locate/trac
1715