This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.
The design and construction of high selective voltammetric sensor for 2,4,6-trinitrotoluene (TNT) by using molecularly imprinted polymer (MIP) as a recognition element was introduced. TNT selective MIP and non imprinted polymer (NIP) were synthesized and then were used for carbon paste (CP) electrode preparation. The MIP-CP electrode showed very high recognition ability in comparison to NIP-CP. It was shown that electrode washing after TNT extraction, led to enhance the selectivity. Some parameters affecting sensor response, was optimized and then calibration curve was plotted. Dynamic linear range of 5-1000 nM was obtained. The detection limit of sensor was calculated equal to 1.5 nM. This sensor was used successfully for TNT determination in the different water and soil samples.
Keywords: 2,4,6-trinitrotoluene; Molecularly imprinted polymer; Voltammetric sensor; Carbon paste
The importance of analysis of explosives can be considered in two different fields (Hill et al., 2000). One is the threat of an illegal use of these compounds against the security of the nation. The other threat is growing concern about health risks associated with the release of explosives into the environment from military sites and former ammunition plants. An important characteristic of nitroaromatic compounds is their ability to rapidly penetrate the skin. They can cause the formation of methemoglobin on acute exposures and anemia on chronic exposures. 2,4,6-Trinitrotoluene (TNT) explosive can readily enter groundwater supplies and has been classified as toxic at concentrations above 2 ng.ml−1 by the Environmental Protection Agency (Environmental Protection Agency, 1989) as it presents harmful effects to all life forms (Talmage et al.,1999). It causes liver damage and aplastic anemia. Deaths from aplastic anemia and toxic hepatitis were reported in TNT workers prior to 1950s (McConnell et al., 1946). Other occasional effects include leukocytosis or leukopenia, peripheral neuritis, muscular pains, cardiac irregularities, renal irritation and bladder tumors (Djerassia, 1998; Djerassia and Vitany, 1975). These compounds are generally recalcitrant to biological treatment and remain in the biosphere, where they constitute a source of pollution due to both toxic and mutagenic effects on humans, fish, algae and microorganisms. The vapor or dust can cause irritation of mucous membranes resulting in sneezing, cough and sore throat. However, relatively few microorganisms have been described as being able to use nitroaromatic compounds as nitrogen and/or carbon as energy source (Duque et al., 1993; French et al., 1998; Vorbeck et al., 1994).
Due to health, ecological and security risks caused by long- and short-term exposure to explosive compounds, there is considerable interest in their measurements in environmental samples. Extensive efforts have been devoted to the development of innovative and effective sensors, capable of monitoring explosives rapidly in the low concentration and complex matrixes. Electrochemical sensors are among the mostly used devices for TNT determination because of inherent sensitivity of these methods (Saravanan et al., 2006; Zimmermann and Broekaert, 2005; Agu et al., 2005; Wang and Thongngamdee, 2003). In order to enhance the sensitivity and also selectivity, the use of chemically modified electrodes is common option (Wang et al., 2004; Hrapovic et al., 2006; Wang and Pumera, 2006; Shi et al., 2007). However the moderate selectivity of these sensors can be considered as a main problem in this case. The use of electrochemical biosensors is the proper way to overcome the moderate selectivity problem of the electrochemical sensors based on the chemically modified electrodes (Naal et al., 2002).
Although biological receptors have specific molecular affinity and have been widely used in diagnostic bioassays and chemo/biosensors, they are often produced via complex protocols with a high cost and require specific handling conditions because of their poor stability, and the natural receptors for many detected analytes don't exist (Whitcombe et al., 2000; Wulff, 2002; Haupt and Mosbach, 2000; Ye and Haupt, 2004). Thus, there has been a strong driving force in synthesizing artificial recognition receptors. Molecular imprinting is one of the most efficient strategies that offer a synthetic route to artificial recognition systems by a template polymerization technique (Mosbach, 2006; Urraca et al., 2006; Tao et al., 2006; Sun et al., 2004; Piletskaya et al., 2005, Huang et al., 2004; Hall et al., 2006).
In molecular imprinting technology, the specific recognition sites can be spatially organized by imprinting template molecules in a polymeric organic/inorganic matrix through the template-monomer complexes such as covalent and non-covalent interactions. After crosslinking copolymerization, the removal of templates from the crosslinked matrix generates the recognition cavities complementary to the shape, size and functionality of templates, leaving the steric and chemical information of the imprinted molecules. Therefore, the target species can selectively rebind into the molecularly imprinted polymers (MIPs) through the specific interaction with these imprinted sites. The synthesis technique is simple and cheap, and the resultant MIP materials exhibit high selectivity, excellent mechanical strength, durability to heat, acid and base conditions and better engineering possibility than biological counterparts (Ye and Haupt, 2004). Moreover, the introduction of synthetic design into molecular imprinting strategy can even make a host element suitable for the analyte for which the natural receptor does not exist. These characteristics allow MIP materials as recognition elements to be used in a wide range of fields (Andersson et al., 1990; Spivak, 2005).
Although the main applications continue to be in selective separation, MIP-based sensors for the detection of active molecules, pharmaceuticals and environmental pollutants are perhaps the most challenging, and have attracted considerable interest in recent years (Piletsky and Turner, 2002; Holthoff and Bright, 2007; Holthoff et al., 2007; Stephenson and Shimizu, 2007). When the combination of MIP materials with a transducer in a suitable format, the sensors with MIP recognition can identify and quantify a target species by converting the analyte-MIP binding event into a physically readable signal,
In this work a new and simple method was applied for design and preparation the high selective and sensitive electrochemical sensor for TNT determination. Molecularly imprinted polymer having recognition sites for TNT was used as recognition element, in the carbon paste electrode. This biomimetic modifier functioned as selectivity increasing and pre-concentrator agent for TNT determination. The prepared electrode was used for TNT determination by three steps procedure including analyte extraction in the electrode, electrode washing and electrochemical measurement of TNT. It was found that the washing step had main effect on the selectivity improvement of the sensor by removing the weakly absorbed interferences from the electrode without considerable effect on the sensitivity of sensor. The optimized sensor was used successfully for TNT determination in water and soil samples.
2.1. Instruments and reagents
Electrochemical data were obtained with a three-electrode system using a potentioastat/galvanostat model PGSTAT302, Metrohm. The differently prepared MIP or NIP involved sensors were used as a working electrode. A platinum wire and an Ag/AgCl electrode were used as the counter and reference electrodes respectively. Methacrylic acid (MAA), obtained from Sigma-Aldrich (Munich, Germany), was purified by passing them through a short column of neutral alumina, followed by distillation under reduced pressure. Ethylene glycol dimethacrylate (EDMA), obtained from Fluka (Buchs, Switzerland), distilled under reduced pressure in the presence of hydroquinone inhibitor and stored at 4C° until use. 2,4,6-trinitrotoluene, para-nitrophenol, aniline, phenol, nitrobenzene, n-eicosane and 2, 2'-azobisisobutyronitrile (AIBN) were supplied by Sigma-Aldrich (Munich, Germany), and used as received. Graphite powder was purchased from Fluka (Buchs, Switzerland). Other chemicals were of analytical grade and were purchased from Merck (Darmstadt, Germany)
2.2. Molecularly imprinted Polymers preparation
Molecularly imprinted polymer having TNT recognition sites was designed and prepared according to precipitation procedure. In order to molecularly imprinted polymer synthesis template molecule (1 mmol), methacrylic acid (4 mmmol) and 50 ml of dry chlorofom were placed into a 100 ml round-bottomed flask and the mixture was left in contact for 10 min. Subsequently, EDMA (24 mmol) and AIBN (0.2 mmol) were added. The flask was sealed and the mixture was purged with nitrogen for 15 min. Polymerization took place in a water bath at 60 -C for 24 h. the final polymer was simply powdered and the template was removed by Soxhlet extraction with methanol for 48 h. the complete removing of template from the polymer was traced by square wave voltammetry method. Non imprinted polymer (NIP) was prepared similar to MIP except that the template was not present in the polymerization media.
In order to obtain finer and smaller MIP particles the obtained powder was sequentially immersed three times in the acetonitrile and the supernatant portions were collected for final using.
2.3. Preparation of the sensors
In order to construction the sensor (MIP-CP or NIP-CP) 0.05 gr graphite was homogenized in a mortar with 0.01 gr of powdered para-nitrophenol MIP or NIP for 10 min. Subsequently, n-eicosane, 0.03 gr was melted in a dish in a water bath heated at 45 -50 C°. The graphite/MIP blend was then added to the melted n-eicosane and mixed with a stainless steel spatula. The final paste was used to fill a hole (2.00 mm in diameter, 3 mm in depth) at the end of an electrode body previously heated at 45 C°. After cooling at room temperature, the excess of solidified material was removed with the aid of sand paper.
2.4. Electrochemical measurements
The electrochemical measurement of TNT was carried out according to the following sequentially procedure:
Extraction step: the prepared electrode was inserted into the solutions containing determined or undetermined concentrations of TNT in which the pH was fixed equal to 4.5 by acetate buffer. All solutions in the extraction period were stirred at fixed stirring rate and for a determined time.
Washing step: the electrode was removed from the first solution and then inserted into the washing solution composed of water/acetonitrile (97:3) remaining in this solution for 15 s.
Analyzing step: the electrode was placed in the electrochemical cell containing 10 ml of HCl (0.07M). For voltammetry experiment at firs the potential of -1.0 V was applied to the electrode for 20 s and then the potential was scanned in the aimed range.
2.5. TNT measurement in real samples
In order to TNT analysis, the imprinted sensor was incubated in the spiked solution, and then inserted into water/acetonitrile (97:3) solution for 15 s followed by transferring it into an electrochemical cell containing 10 ml acetate buffer with pH= 4.5. Square wave voltamograms in the range of 0.0 to 1.0V were recorded and the current peak was used for final determination.
3. Results and discussion
3.1. Cyclic voltammetry behavior of TNT
Electrochemical behavior of TNT in different mediums and various electrodes has been studied and reported (Schmelling et al., 1996; Plambeck, 1982; You et al., 1997; Wang et al., 1998). However the cyclic voltammetry of TNT was investigated by using pure carbon paste electrode having no MIP or NIP. The obtained voltammogram is shown in Fig. 1. As can be seen three distinct reductions were observed during the decreasing potential sweep, likely corresponding to the formation of hydroxylamine from the sequential reduction of the three NO2 groups. However, the increasing sweep indicated only one oxidation peak at positive potential. The oxidation peak may correspond to further transformation of the moieties formed during the decreasing potential sweep. Thus it is related to the oxidation of hydroxylamine (Hilmi et al., 1999). In the further investigation in the case of both cyclic voltammetry and differential pulse voltammetry at first the proper negative pre-potential was applied to the MIP-CP or NIP-CP, converting the -NO2 groups of TNT to the -NHOH groups, and then the potential sweep was carried out in the positive potential range.
3.2. MIP-CP and NIP-CP electrodes responses to TNT and washing effect
In order to study the TNT recognition ability of MIP, the MIP-CP, NIP-CP and CP electrodes were prepared and inserted into the TNT containing solutions. After 7 min the electrodes were removed from the TNT solution and cyclic voltammetry was carried out. The obtained results are shown as voltamograms, related to the oxidation peak of reduced product of TNT which was obtained after applying the pre-potential of -1.0 V for 10 s to each electrode (Fig. 2). As can be seen the CV signal of MIP-CP electrode (voltamogram a) is higher than that for NIP-CP electrodes (voltamogram c). This indicates that the MIP in the carbon paste electrode can intensively uptake TNT from the aqueous solution in comparison to NIP-CP. In order to evaluate the TNT keeping power by MIP after its extraction, we performed other experiments in which the previous experiments were repeated again, but in this case the electrodes were inserted into the washing solution for a short time (15 s) just after electrode removing from the TNT solution. The obtained results are shown as the voltamograms (b) and (d) in Fig. 2. As it is clear, the electrodes washing after TNT extraction in the electrodes, do not affect noticeably the TNT signal in the MIP-CP whereas at the same time the response of NIP-CP electrode is decreased considerably indicating more affinity of MIP-CP electrode for TNT. This observed behavior of MIP-CP can be used for selectivity enhancement of MIP-CP by simple washing step. The washing process can remove the weakly and nonspecifically absorbed TNT molecules from the electrode surface, the state which is dominant in the case of NIP-CP. However TNT molecules which are incorporated in the selective sites of MIP are not removed such easily by washing process.
The constructed MIP particles, during non covalent approach, usually contain selective sites with various affinities for template. Some of them are cavities with the sizes matchable with template molecule. These are template recognition sites, constructed with regular and perfect shape in the polymerization period and thus have more affinity for TNT. TNT molecules which are presented in such cavities are tightly absorbed to the MIP and thus the washing of electrode, modified with MIP, don't noticeably disrupt the corresponding interactions, (Caroa et al., 2002).
A main drawback of non-covalent systems is the unavoidable heterogeneity of the binding sites arising from the multitude of complexes formed between the template and the functional monomers which are apparently preserved to some extent during the polymerization. Usually an excess of functional monomer relative to the template is required to favor template-functional monomer complex formation and to maintain it's integrity during the polymerization. As a result, a fraction of the functional monomers are randomly incorporated in the polymer matrix resulting in the formation of nonselective binding sites (Joshi et al., 1998; Whitcombe and Volfson, 1995; Whitcombe et al., 1997; Takeuchi and Matsui, 1996). The cavities with incomplete or irregular shape and also the non selective binding sites cannot absorb TNT molecules so tightly. The portion of TNT molecules absorbed by such mentioned binding sites can be removed from MIP-CP electrode by washing process.
In the case of NIP-CP electrode washing step removed the TNT from the electrode easily because the binding site in the NIP is non selective and act almost similar to non selective sites of MIP.
Molecularly imprinting polymer can act more selective than other artificial recognition elements. However according to what was explained above, because of presence of non selective and poorly selective sites in the MIP and also non selective absorption property of carbon particles in the MIP-CP, importance of washing step in our proposed method is explicit.
3.3. Electrochemical method selection
In order to achieve a high sensitive sensor the selection of proper electrochemical technique is of most important. Thus we tested different voltammetric methods such as differential pulse voltammetry and square wave voltammetry, known as high sensitive methods, in the same conditions of extraction and washing. The obtained results of this experiments are shown in Fig. 3. As it is evident, the response of square wave voltammetry for TNT is higher than that of differential pulse voltammetry. Thus this method was selected as a main electrochemical method for TNT determination by prepared sensor.
3.4. Optimization of parameters affecting TNT determination
The optimization process for designed sensor was divided into three sections including optimization of carbon paste composition, extraction parameters and electrochemical determination conditions.
3.4.1. MIP-CP composition optimization
In order to find the best composition for MIP-CP electrode the amount of different ingredients of the electrode including MIP, carbon and n-eicosane was changed in the fixed conditions of extraction and voltammetric determination and the obtained responses were used for conclusion. The MIP-CP electrode was prepared with fixed amount of carbon and n-eicosane and different amount of MIP. The resulted electrode at each case was used for TNT extraction and determination. The obtained results are presented in Fig. 4 (I). It is clear that the maximum response for prepared sensor appear in the MIP amount of 0.025gr. Higher amount of MIP in the MIP-CP electrode can increase the sensor response because of providing more recognition sites on the electrode surface as it is evident in the corresponding curve. However enhancement the MIP amount, more than a threshold amount, leads to decrease the prepared sensor response probably because of electrode surface conductivity decreasing. Similar experiments were also carried out in order to investigate the effect of carbon and n-eicosane amount on the prepared electrode response for TNT by variation of these parameters amounts in the MIP-CP electrode followed by recording the obtained results. These results are shown in the Fig. 4 (II) and (III). From the corresponding curves the optimum amount of carbon and n-eicosane were found to be 0.035 and 0.025 gr for carbon and n-eicosane amounts, respectively. Increasing the carbon content of MIP-CP electrode leads to increase the corresponding electrode response because of electron transferring capability enhancement of the electrode. Increasing electrode sensitivity with carbon amount increasing continues to limited point and afterwards the electrode response decreases with carbon content enhancement because of carbon amount of electrode surface increases in expense of MIP content decreasing.
The optimum amount of n-eicosane is required for MIP-CP electrode preparation. Presence of higher amount of binder (n-eicosane) in the MIP-CP electrode leads to decrease in electrode response because of electrode surface conductivity decreasing.
3.4.2. TNT extraction conditions optimization
The pH of TNT solution as a commonly considering parameter was noticed and its effect on the TNT extraction in the electrode was studied. For this purpose the pH of TNT solutions were fixed at different pH and at each case, the prepared electrode was inserted into the TNT solution for 7 min at constant stirring state. After the mentioned time the electrode was removed from the solution and then was immersed into the solution of electrochemical cell. The results of this experiment are shown in Fig. 5 (I). As it can be seen, in the pH range of 2-5 the TNT related electrochemical signal and thus the TNT extraction amount, is relatively high and it seems that no considerable variation in extraction of TNT is resulted by pH changing in this range. In the pH higher than 6 the extraction amount tends to decrease. According to these results the pH of 4.5, fixed with acetate buffer, was chosen as an optimum pH for TNT extraction in the electrode.
Since the electrode contacting area with the TNT containing solution was partially small, therefore the analyte extraction in the electrode was carried out at stirring state. In order to optimize the stirring rate in the extraction period, TNT was extracted in the prepared MIP-CP electrodes at various stirring rates whereas the other extraction parameters such as time, pH and TNT entration were the same and constant. The obtained results, howing the TNT related square wave voltametry signal variation against stirring rate, re presented in Fig. 5 (II). As can be seen the more is the stirring rate, the greater is the electrode response for TNT, indicating the high effect of stirring rate on the TNT extraction in the MIP-CP electrode. The growth in TNT voltammetric response with stirring rate increasing continues noticeably till 400 r.p.m but after that it seems that the parathion extraction enhancement is not so much and the variation of extraction with stirring rate changing is small. Thus we selected the value of 500 r.p.m as optimum for this optimization purpose.
Extraction time was another main parameter which was examined. For this aim the prepared electrodes were inserted into the TNT solutions of the fixed concentration and stirring rate for various times and afterwards the electrodes were removed from the solutions followed by square wave voltammetry. The obtained results are shown in Fig. 5 (III). According to this figure the increasing of extraction time leads to intensive increasing in the TNT extraction amount in the electrode till about 10 min and afterwards the response increasing rate with time enhancement is not so considerable. In order to decrease the TNT analyzing time, as much as possible, the time of 10 min was selected for the extraction time.
3.4.3. TNT electrochemical determination conditions optimization
In the case of square wave voltammetry measurement, the main important parameters which could be optimized were; pH of solution of electrochemical cell, applied pre-potential amount, time of exerted pre-potential and also frequency of used potential in square wave voltammetry. Fig. 6 (I) shows that the anodic peak current is dependent on the pH. The anodic peak current decreased with decreasing pH. Thus in order to obtain high electrochemical response of the sensor the acidic pH of about 1.5 was fixed by using HCl (0.07 M) solution.
The effect of pre-potential magnitude and its exertion time was investigated. It was found that the pre-potential of -1.0 V is appropriate and higher value pre-potential applying to the electrode did not increase further the response amount.
The optimization of pre-potential applying time is demonstrated in Fig. 6 (II), where the peak current increases with increase in the pre-potential applying time till about 30 s where it attains plateau Thus the optimum amount of 30 s was considered for this parameter. The effect of square wave frequency on the final response of sensor was examined. The obtained results are shown in Fig. 6 (III). As can be deduced the frequency of 150 Hz is the best option in this case.
3.5. Analytical characterization
After the optimization and establishment of determination method for prepared MIP-CP sensor, various ions and molecules were examined with respect to their interference with the determination of TNT. It is worth noticing that the values of current response used for the calibration curve are actually the absolute values of the oxidative peak current, observed after electrode incubating in different concentration of TNT solution and cathodic pre-potential of -1.0 V applying to the electrode. For 70 nM of TNT the results showed that over 500- fold excess concentration of K+, Ca2+, Cl-, SO42-, Co2+, Ni2+ , Fe2+, Zn2+, Cu2+, Pb2+, Hg2+ and Hg2+ did not interfere the TNT response. Phenol, aniline, para-nitrophenol and benzoic acid showed no interference till 30 fold excess over TNT. The interference for para-nitrophenol and nitrobenzene was appeared in their concentration 60 fold higher than that for TNT. Phenol and aniline in the concentrations 100 fold excess than TNT showed their interference. The interference level was considered the error of 5% in the TNT determination by aimed interfere compound.
Fig. 7 shows the prepared and optimized MIP-CP sensor response against TNT concentration variation. Precise inspection in plotted calibration curve proved that a linear relationship can be found over TNT concentration in the range of 5 -10−9 to 5.0-10−6 M with a detection limit of 1.5-10−9 M (S/N = 3).
Determination of TNT in water and soil samples
The analytical usefulness of the prepared electrochemical sensor for determination of TNT was demonstrated by applying it to the determination of TNT in water and soil samples. The absence of TNT was first verified in the non-spiked samples. Water samples of tap water and ground water were spiked with TNT. For analysis of soil samples, approximately 1 g soil, previously spiked with TNT at the 200 µg.g-1 levels, was accurately weighed and introduced into a 30 mL centrifuge tube. Next, 2 mL acetonitrile was added and the mixture was shaken for 30 min. After centrifugation at 3500 rpm for 20 min the liquid was filtered through a Nylon syringe filter. A 10 µL aliquot of this extract was diluted to 10 mL with acetate buffer solution of pH 4.5. The electrode was immersed into the resulted solution for 10min followed by inserting the electrode in the washing solution for 15 s. finally the electrode was emplaced in the electrochemical cell containing HCl (0.07M) and the SW voltammetry response was recorded for determination purpose according to the calibration curve, previously described. Table 1 summarizes the results obtained for the three samples analysed. As can be seen, the recoveries achieved are acceptable for all the samples tested, the confidence intervals being calculated for a significance level of 0.05.
Very high selective square wave voltammetry sensor for TNT determination at low concentration was proposed. The MIP functioned as both pre-concentrator and high selective recognition element in the carbon paste structure. Washing of MIP-CP electrode after TNT extraction led to enhance the selectivity without considerable loss in sensitivity and detection limit of the sensor. The proposed sensor was used successfully for TNT determination in the real samples.