le vie dello sviluppo attraverso la green economy

3 downloads 29558 Views 1MB Size Report
best indicators of faecal contamination [3] since they originate in the intestine of warm blood- .... characteristics and a laptop PC for data graphing and filing.
07- 10 Novembre 2012

LE VIE DELLO SVILUPPO ATTRAVERSO LA GREEN ECONOMY La Ricerca, gli Strumenti, la Gestione Industriale r Sezione TEMATICHE INTEGRATE r Ecomondo WASTE r Ecomondo ORO BLU r Ecomondo AIR r Ecomondo RECLAIM EXPO r Altri interventi

www.ecomondo.com

La Mascotte di Ecomondo 2012 Il TORDO AMERICANO - American Robin (Turdus Migratorius)

Luciano Morselli - Acrilico su carta latte – 26x18 cm Dedicato a Rachel Carson a 50 anni dalla pubblicazione di Silent Spring

Atti dei seminari a cura di Luciano Morselli

Bacterial concentration detection in water by microfabricated impedance biosensor Marco Grossi [email protected], Bruno Riccò - Department of Electronic Engineering (D.E.I.S.), University of Bologna, Bologna Daniele Gazzola, Manuele Onofri, Giampaolo Zuccheri - Health Sciences and Technologies (HST-ICIR), University of Bologna, Bologna Diego Matteuzzi - Department of Pharmaceutical Sciences, University of Bologna Summary Control of water microbial content is of great importance to guarantee the absence of pathogens. The bacterial concentration is traditionally measured by standard plate count, a technique that is reliable but characterized by long response time and must be performed in microbiology laboratory with the aid of trained personnel. The impedance technique, that measures the bacterial concentration by analyzing the sample electrical characteristics, is competitive with the standard technique since features shorter detection times (3 – 12 hours vs. 24 – 72 hours of plate count) and can be easily realized in automatic form. The present work shows a microfabricated sensor featuring gold electrodes (1mm2 area separated by 100ìm) used to measure the concentration of a wild type coliform strain (isolated in river water). The presented sensor is capable to detect high microbial concentration (106 cfu/ml) in relatively short time (225 minutes) and, compared to other impedance biosensors, has the advantage to properly work at higher frequencies (extending the working frequency range to over 1 MHz) with benefits for measure reliability. Riassunto Il monitoraggio della contenuto microbico delle acque è di grande importanza al fine di garantire l’assenza di microorganismi patogeni. La concentrazione batterica viene determinata tramite conta in piastra, tecnica che risulta affidabile ma richiede tempi lunghi e può solo essere effettuata in laboratori di microbiologia da personale qualificato. La tecnica impedenziometrica, che valuta la concentrazione microbica tramite l’analisi delle caratteristiche elettriche del campione, è una tecnica competitiva con quella tradizionale in quanto garantisce tempi di risposta più brevi (3 – 12 ore rispetto a 24 – 72 ore della conta in piastra) ed è facilmente automatizzabile. Nel presente lavoro viene mostrato un sensore micro fabbricato con elettrodi in oro (di dimensione 1mm2 separati da una distanza di 100ìm) utilizzato per la determinazione della concentrazione di un coliforme isolato nelle acque di fiume. Il sensore si dimostra capace di rilevare alte concentrazioni batteriche (106 cfu/ml) in tempi relativamente brevi (225 minuti) e ha il vantaggio rispetto ad altri biosensori impedenziometrici di poter operare a frequenze più elevate (fino a 1 MHz) con notevoli benefici per quanto riguarda l’affidabilità della misura.

989

1. Introduction Bacterial concentration detection is of great importance in environmental monitoring and existent national and international regulations guarantee water quality and safety [1]. Environmental waters (such as river waters and seawaters) are periodically screened to ensure that microbial concentration is within legal limits and that pathogens that can endanger human health are absent. Usually this is obtained by screening the samples for microorganisms that are related to faecal contamination [2]: in fact, from a statistical standpoint, these present a good correlation with the presence of pathogens. Traditionally, coliforms are considered the best indicators of faecal contamination [3] since they originate in the intestine of warm blooded animals. Water bacterial concentration is usually measured by Standard Plate Count (SPC) technique [4], a reliable method but characterized by long response time (24 – 72 hours depending on the screened microbial strain) and by the need to be performed in microbiology laboratories with the aid of trained personnel. This prevents SPC to be used for fast in-situ detection of bacterial contamination. In the last years many innovative techniques have been proposed for microbial concentration detection based on transduction methods such as amperometry [5], bioluminescence [6], turbidity [7], piezoelectricity [8] and impedance [9]. A set of instruments for coliforms detection in water are produced by IDEXX (Westbrooke, Maine, USA). Colilert, Colilert 18 and Colisure [10][11] are based on the coliform property to produce â-glucuronidase as a result of their methabolism. The IDEXX instruments are however laboratory oriented and the time needed to measure the microbial concentration is only slightly shorter than SPC. The impedance technique based on classic impedance microbiology [12] is highly competitive with SPC because it features singnificantly shorter detection times (3 – 12 hours vs. 24 – 72 hours) and is easily implementable in automatic form with the possibility to be realized as an embedded portable system for in-situ measurements. The impedance technique works as follows: the sample, eventually diluted in a suitable enriched media, is stored at a temperature that favors bacterial growth and its electrical characteristics (the resistive and reactive components of the impedance Z) are measured at time intervals of 5 minutes. Until the sample bacterial concentration is lower than a critical threshold concentration (107 cfu/ml), the electrical parameters are essentially constant (baseline value), while when it exceeds this concentration |Z| begins to decrease (as well as its resistive and reactive components). The time needed to produce a variation of the monitored electrical parameters is called Detect Time (DT) and is known to be linearly related to the logarithm of initial bacterial concentration. Different commercial instruments exist that are based on the impedance technique: Bactometer by Vitek Systems Ltd (Basingstoke, UK), Malthus by Malthus Instruments Ltd (Bury, UK), Bac Trac by Sy-Lab (Purkensdorf, Austria) and RABIT by Don Whitley Scientific (Shipley, UK). Recently, an embedded portable biosensor system based on the impedance technique [13] has been proposed that is particularly suited for in-situ bacterial screening. All the presented instruments feature stainless steel or platinum as the electrodes material, the inter-electrodes distance is in the mm range and a capacity of 3 to 10 ml for the sample under test (SUT) is used. In this work we test a microfabricated sensor, featuring small (1 mm2) gold electrodes separated by 100 ìm, and compare its performance with those of the aforementioned instruments. The results indicate that, even if response time is comparable with that obtained in [13], the microfabricated sensor features a broader working frequency range, thus allowing for more realible measurements.

990

2. Report The proposed sensor has been tested using the experimental setup shown in Fig. 1 (a). The sensor and SUT are stored in a thermal incubator set at 37 °C and the electrical characteristics are measured by an LCR meter Agilent E4980A. A Laptop PC system controls the LCR meter via USB interface and acquires the measured data for graphing and data filing. The sensor, shown in Fig. 1 (b), features 18 gold microelectrodes and, for the measures, 4 electrodes of the top row have been used shorted in couples as shown in Fig. 1 (b). The electrodes consist of 1mm2 gold coated area separated by 100ìm. To measure the electrical characteristics, the sensor has been stimulated with a sinusoidal test voltage of amplitude 10 mVPP and frequency in the range 20 Hz – 2 MHz (logarithmically spaced).

Fig. 1 – (a) Scheme of the experimental setup used in the bacterial concentration measurements. The system features a thermal incubator to store the sample under test, an LCR meter to measure sample electrical characteristics and a laptop PC for data graphing and filing. (b) Microfabricated sensor used to detect bacterial concentration. It features 1mm2 gold electrodes separated by 100ìm. 1: Aluminium Nitride (AlN), 2: gold covered with AlN, 3: gold surface

The enriched medium used to favor bacterial growth is Lauria Bertani (modified to feature low salt content). The medium composition (for 1 liter of distilled water) is as follows: Tryptone 10.0 g, Yeast Extract 5.0 g (pH 7.0). A wild coliform strain (isolated from river water) has been cultured in Lauria Bertani, diluted in different ratios so to obtain different bacterial concentration and inoculated in the medium. The sensor was then immersed in direct contact with the SUT and both stored in the thermal incubator. The equivalent electrical circuit used to model the sensor immersed in the SUT is shown in Fig. 2 (a): Ri represents the interface resistance of the electrodes, Cpar the electrodes capacitance due to the sensor AlN substrate, Rm the medium resistance and ZCPE the impedance of a constant phase element (CPE) modeling the non-ideal capacitive electrode-electrolyte interface. As the

991

bacterial population grows in the SUT, microbial methabolism transforms uncharged compounds in highly charged ones, thus increasing the medium ions concentration and the electrical conductivity (producing Rm decrease). Moreover, since the ions are subjected to different forces if near the electrodes or in the bulk of the electrolyte, a double layer is built at the electrode-electrolyte interface that increases the non-ideal interface capacitance (Q). The parameters of the equivalent circuit that best-fit the measured data have been calculated using Multiple Electrochemical Impedance Spectra Parametrization (MEISP) v. 3.0 by Kumho Chemical Laboratories. In Fig. 2 (a) and (b) the measured data (dash lines) for |Z| and Arg(Z) as well as the simulated curves resulting from the fitted parameters (dot lines) are plotted versus the frequency of the sinusoidal test signal. As can be seen, the simulated curves properly match the measured ones, thus validating the proposed electrical model.

Fig. 2 – (a) Electrical circuit used to fit the measured data. Curves for both measured (dash lines) and simulated (dot lines) data for |Z| (b) and Arg(Z) (c) are plotted vs. the frequency of the sinusoidal test signal

Measures using the experimental setup of Fig. 1 (a) have been carried out with different concentrations of the inoculated coliform strain (from 10 cfu/ml to 107 cfu/ml). Although bacterial methabolism affects both the bulk conductivity and the interfacial capacitance (i.e. Rm and Q), the measured values of Q resulted in poor repeatability and low signal-to-noise ratio, resulting in low realibility in DT calculation. Thus, in the following, only values of Rm are considered. In Fig. 3 (a) the percent decrease of Rm, i.e. [(Rm,baseline – Rm)/Rm,baseline]’”100, is plotted vs. time for two samples characterized by different values of bacterial concentration. As can be seen, the sample featuring lower bacterial concentration (102 cfu/ml) results in higher DT (460 minutes) than the highly contaminated sample (106 cfu/ml) that features a DT of 225 minutes. This shows how the medium resistance Rm can be effectively used to discriminate between different levels of bacterial concentration. However, since the measure of medium resistance Rm requires a multi-frequency approach (i.e. the measure of the electrical parameters on a broad range of frequencies and best-fit of the measured data with the equivalent electrical circuit using a suitable numerical algorithm) we

992

have investigated if a single-frequency measure can reliably estimate the sample bacterial concentration. In Fig. 3 (b) the percent decrease of |Z| is plotted vs. time for the sample characterized by a bacterial concentration of 102 cfu/ml and four different frequencies of the sinusoidal test signal (200 Hz, 2 kHz, 20 kHz, 200 kHz). As can be seen, higher frequencies result in more reliable data, with curves characterized by more stable baseline and thus more accurate determination of DT. Data on higher contaminated samples (106 cfu/ml) confirmed the same results with measured values of DT comparable with those obtained in multi-frequency approach and more reliable results at frequencies higher than 20 kHz. The results show how the proposed microfabricated sensor can broaden the working frequency range. In fact, all the benchtop instruments discussed in the introduction as well as the portable biosensor system in [13] are characterized by a maximum working frequency not higher than 10 kHz, while the proposed sensor broadens this limit to over 1 MHz, with benefits in terms of higher signal-to-noise ratio, more stable baseline and more accurate DT calculation.

Fig. 3 – (a) Percent decrease of Rm plotted vs. time for two samples characterized by different bacterial contamination. Higher contaminated samples feature lower values of DT. (b) Percent decrease of |Z| plotted vs. time for single frequency measurements. Higher frequency measurements result in more stable baseline and more accurate DT calculation

3. Conclusions Bacterial concentration detection is very important in environmental monitoring since the presence of pathogens can endanger human health. Bacterial concentration is usually determined by Standard Plate Count (SPC) technique, a reliable method that is however characterized by slow response (24 – 72 hours depending on the monitored bacteria) and needs a laboratory environment with skilled personnel. The impedance technique for microbial concentration detection is very competitive with SPC since it features shorter response time (3 – 12 hours depending on the sample bacterial contamination) and can be easily automatized and implemented as a portable biosensor system for in-situ measurements. The proposed microfabricated impedance sensor is adequate in bacterial concentration detection since it can measure microbial concentration with response time comparable with that obtained by benchtop commercial systems (225 minutes for a contamination of 106 cfu/ml). Moreover, the small dimensions (1mm2 electrodes separated by 100 ìm) make it possible to test a small quantity of sample and the working frequency range for the sinusoidal test signal is greatly improved (maximum frequency 1 MHz) compared to the other systems (maximum frequency 10 kHz) with benefits in terms of higher signal-to-noise ratio, more stable baseline and more accurate Detect Time calculation.

993

References [1] Corbitt R. A., “Standard handbook of environmental engineering”, McGraw-Hill Publishing Co. (2nd edition), New York, 1999. [2] Meays C. L., Broersma K., Nordin R., Mazumder A., “Source tracking fecal bacteria in water: a critical review of current methods”, Journal of Environmental Management, 2004, Vol. 73, 71-79. [3] Romprè A., Servais P., Baudart J., de Roubin M. R., Laurent P., “Detection and enumeration of coliforms in drinking water : current methods and emerging approaches”, Journal of Microbiological Methods, 2002, Vol. 49, 31-54. [4] Kaspar C. W., Tartera C., “Methods in Microbiology”, Grigorova & J.R. Norris ed., London: Academic Press, 1990, Vol. 22, 497-531. [5] Perez F., Tryland I., Mascini M., Fiksdal L., “Rapid detection of escherichia coli in water by a culture based amperometric method”, Analytica Chimica Acta, 2001, Vol. 427, 149-154. [6] Stanley P.E., “A review of bioluminescent ATP techniques in rapid microbiology”, Journal of Bioluminescence and Chemiluminescence, 2005, Vol. 4 (1), 375-380. [7] Koch A. L., “Turbidity measurements of bacterial cultures in some available commercial instruments”, Analytical Biochemistry, 1970, Vol. 38 (1), 252-259. [8] Kim N., Park I.-S., Kim D.-K., “Characteristics of a label-free piezoelectric immunosensor detecting Pseudomonas aeruginosa”, Sensors and Actuators B, 2004, Vol. 100, 432-438. [9] Yang L., Bashir R., “Electrical/electrochemical impedance for rapid detection of foodborne pathogenic bacteria”, Biotechnology Advances, 2008, Vol. 26, 135-150. [10] Cowbum J. K., Goodall T., Fricker E. J., Walter K. S., Fricker C. R., “Preeliminary study on the use of Colilert for water quality monitoring”, Letters in Applied Microbiology, 1994, Vol. 19 (1), 50-52. [11] Chao K. K., Chao C. C., Chao W. L., “Evaluation of Colilert 18 for detection of coliforms and Escherichia coli in subtropical freshwater”r, Applied and Environmental Microbiology, 2004, Vol. 70 (2), 1242-1244. [12] Firstemberg-Eden R., Eden G., “Impedance microbiology”, New York, Wiley, 1984, Vol. 3, 154-196. [13] Grossi M., Lanzoni M., Pompei A., Lazzarini R., Matteuzzi D., Riccò B., “An embedded portable biosensor system for bacterial concentration detection”, Biosensors and Bioelectronics, 2010, Vol. 26, 983990.

994