A custom-developed, handheld EIS measurement platform

Transcription

1 1 A custom-developed, handheld EIS measurement platform A. Bonyár, I. Bosznai, H. Sántha and G. Harsányi Department of Electronics Technology, Budapest University of Technology and Economics, Budapest, Hungary Abstract In this paper we present a custom-developed, small, handheld Electrochemical Impedance Spectroscopy (EIS) measurement platform designed for various biosensor applications. Our device measures the impedance spectrum in the 2 Hz 100 khz frequency range and evaluates the obtained spectra with built-in circular regression data extraction on a maximum of five electrode pairs (parallel or single measurements). We demonstrate the capabilities of our device with measurements performed on impedance modeling electrical circuits and on real functional biosensors as well. Keywords EIS, biosensor, DNA, platform I. INTRODUCTION Electrochemical Impedance Spectroscopy (EIS) is a widespread measurement method in the field of applied biosensor research and development due to its high sensitivity and simplicity. The combination of this technique with enzyme, antibody/antigen, DNA or living cell based biosensors makes it utilizable in a wide range of application fields such as environmental monitoring, clinical diagnostics or gene analysis. In theory EIS is a method to probe interface properties of biofunctionalized electrode surfaces. O. Panke gives a good overview about the basic principles and applications in the field of biosensing [1]. The complex impedance of an electrochemical system is determined by applying a voltage perturbation on the system and measuring the current response in function of the frequency. For the evaluation of the EIS Nyquist spectra obtained on the system different electrical circuit models are applied. The simplest circuit describing the system is the standard Randles cell model supplemented with the Warburg impedance component as illustrated in Fig. 1. Fig. 1 The Randles cell model supplemented with the Warburg impedance. R s: serial resistance, R ct: charge transfer resistance, C dl: double layer capacitance, W: Warburg impedance Fig. 2 shows a typical EIS Nyquist spectrum with Randles behavior. To extract the impedance data from the measured Nyquist plots the commercially available software utilize circular regression algorithms. Fig. 2 A typical EIS Nyquist spectrum with the representative values and regions

2 2 In order to fit a semicircle on the spectrum with small error it is mandatory to know the spectrum at the 0 range, where 0 is the highest point of the semicircle (see Fig. 2). Thus the theoretical limit of the highest measurable impedance is determined by 0: 0=2π*f 0 =1/(R ct*c dl) (1) In this work we present a custom-developed EIS measurement platform which utilizes a type of two electrode measurement setup without a classical reference electrode, which corresponds to a setup where one electrode is used as a working electrode and the other one as a common counter and reference electrode. We examined and compared this setup to the standard three electrode measurement mode in one of our previous works and found that this simplification causes no decrease in the precision or sensitivity [2]. Because the hardware defined frequency range of our device is 2 Hz 100 khz, the theoretical limit of the highest measurable impedance is determined by (1 : 2Hz 1/(2π*R ct*c dl) (2) For a given bioreceptor layer the impedance of the whole electrochemical system is dependent on the electrolyte concentration. For example in the case of transduction methods utilizing Faradic charge transfer processes (e.g. redox couples such as potassium ferri/ferrocyanide K 3[Fe(CN) 6]/K 4[Fe(CN) 6]) the system impedance can be decreased by increasing the redox couple concentration in the electrolyte. Taking this into account several possible applicable bioreceptor layers can be found in the literature which are compatible with our device specifications. Table 1 collects a few examples utilizing ferri/ferrocyanide redox couples. Table 1 Impedance values of different biofunctionalized gold surfaces Layer type R ct (k cm 2 ) C dl ( Fcm -2 ) f 0 (Hz) c [Fe(CN) 6] -3/-4 (mm) Ref. Bare gold 0, / 0.1M KCl [3] DNA SAM*(48h) 0,0442 7,9 455,8 5 / 0.1M KCl [3] Octadecanethiol 85,9 0,7 2,65 5 / 0.1M KCl [4] Mercaptohexanol 0, / 0.1M Na 2SO 4 [5] Immunosensor** 4,59 0,53 65,42 5 / 10mM PBS [6] *SAM: Self-Assembled Monolayer **Immunosensor for haemoglobin based on mixed SAM After the presentation of the hardware and software features of our device we demonstrate its capabilities with characterization measurements performed on impedance modeling electrical circuits and real biosensors as well. A. Materials II. MATERIALS AND METHODS Custom DNA oligonucleotides, phosphate buffered saline (PBS) ph 7.4, 6-mercapto-1-hexanol (MCH) were purchased from Sigma Aldricht (Germany). FullCure 720 base material and FullCure 705 support material for the 3D printing were purchased from Varinex Inc. (Hungary). B. Electrochemistry For the comparison of different data extraction software in a wide impedance range we used old EIS data measured previously in our laboratory on DNA-MCH SAM based biosensor surfaces. In all the cases the biolayers were deposited on gold thin film working electrodes on glass substrates patterned with laser ablation technology to form interdigitated electrode structures (IDEs). EIS measurements were performed with a Voltalab PGZ 301 potentiostat and the VoltaMaster 4.0 software with 200 mv DC level and 10 mv AC sine wave amplitude, in the 100 khz - 1 Hz frequency range. The measurements were performed in 5 mm ferrocyanide and 5 mm ferricyanide dissolved in 10 mm PBS buffer ph 7.4 in a standard three electrode

3 3 measurement setup with an Ag/AgCl reference electrode. More detailed description about our methods and protocols can be found in our previous papers [2]. A. Hardware setup III. RESULTS AND DISCUSSION Fig. 3 shows a picture of our EIS measurement device. The mechanical parts and the housing of the device were fabricated by 3D rapid prototyping technology utilizing an Objet Eden 250 printer. Fig. 3 Our custom EIS measurement device The disposable biosensor cartridge can be easily introduced and removed with a spring-lock mechanism. The device has built-in fluidic handling with a diaphragm pump from the exchangeable measurement buffer container. Temperature control of the measurement is also possible with a Peltier unit. Fig. 4 presents a block diagram of the main device elements. Fig. 4 Block diagram of the main device elements (PSU: power supply unit, MCU: measurement control unit) The analog circuitry and the mechanical connections are designed for this unique disposable biosensor cartridge which contains five interdigitated electrode pairs (details about the cartridge are about to be published elsewhere). Our reader device can measure the electrode pairs separately and also in a parallel mode (all the five electrode pairs together). With the help of our department s other competences (e.g. printed circuit board manufacturing, 3D rapid prototyping technology) it is possible to re-design and develop the analog circuitry and mechanical parts according to different cartridge designs or other needs.

4 4 B. Software features Our device has two main measurement modes. In standalone mode the device is capable to measure and transfer the extracted data utilizing only the microcontroller s firmware and the Bluetooth module. In laboratory mode (connected to a PC via mini-usb cable) a measurement software provides a graphical user interface (GUI) between the user and the device. The software enables to set advanced measurement options and to visually monitor the obtained results. Fig. 5 shows the main window of the laboratory mode software while on Fig 6 a measured Nyquist plot and the extracted results are presented. Fig. 5 Laboratory mode software main window: 1) electrode pair selection (single/parallel mode), 2) impedance model type selection, 3) frequency range selection, 4) measured Bode plot (amplitude and phase curves) To manage the obtained and uploaded measurement data a secure internet database handled by any web browser is developed. As a user it is possible to upload your measurement data easily with your standalone device from home, while as an administrator (e.g. doctors) it is possible to manage standalone device users and samples from your laboratory (e-health driven data handling system). Fig. 6 Laboratory mode measure window: 1) the measured EIS curve (green) and the fitted semicircle (red), 2) the extracted impedance values

5 5 C. Device calibration To calibrate our EIS measurement device we utilized Randles impedance modeling electrical circuits (discrete resistors and capacitors mounted on printed circuit boards in different values resembling the impedance of biosensor layers). Fig. 7 presents the obtained calibration characteristic of the device. Fig. 7 Calibration characteristic of our EIS device measured on Randles impedance modeling electric circuits The results given on Fig. 7 are an average of at least five measurements. The standard deviation between these consecutive measurements is below 1.5 % in all the cases. The real values of the discrete circuit elements were obtained with a VA 505B type SMD multimeter. The maximal difference between the real and measured values in the case of R ct is 9.94 % (average 2.75 %), in the case of C dl 23 % (average 9.82 %). Fig. 8 Comparison of the extracted R ct and C dl values with the VoltaMaster and ZView software and with our devices built-in circural regression software Fig. 8 presents comparison results between different data extraction software. The evaluated spectra were measured on different biolayers with the VoltaLab PGZ 301potentiostat in a standard three electrode measurement setup. The saved raw data were opened and evaluated with the different software. In the cases of the commercially available VoltaMaster 4.0 and ZView 3.2c software the circular regression fitting is operated and influenced by the user who selects the data points for the

6 6 fitting. In the ZView software the user also has the ability to choose between different circuit models (e.g. to select between a parallel capacitor and constant phase element) which could also affect the extracted values. However speaking about a standalone measurement device one cannot expect any such interaction from the field user (e.g. patients or nurses) so the fully automated circular regression software has to be implemented in the microcontroller level in the device. Fig. 8 shows that inside the 2 Hz theoretical limit the data extraction of our device is in good agreement with the other software, however the difference between the methods increases with higher capacitance and resistance values. The relatively high difference between the two commercially available software (at some values) underlines the influence of the user and the importance of calibration for a well characterized biolayer. IV. CONCLUSIONS In our work we presented a custom-developed, handheld EIS measurement platform. The frequency range of our device is 2 Hz 100 khz which we tested with impedance modeling electrical circuits and which is compatible with a large variety of bioreceptor layers. In standalone software mode the device utilizes the built-in circular regression to evaluate the measured spectra. We demonstrated that the precision of this data extraction is comparable with commercially available software. ACKNOWLEDGMENT The authors would like to thank for the financial support of the following EC funded FP6 project: DVT-IMP (FP IST ). This work is connected to the scientific program of the "Development of quality-oriented and harmonized R+D+I strategy and functional model at BME" project. This project is supported by the New Hungary Development Plan (Project ID: TÁMOP-4.2.1/B-09/1/KMR ). REFERENCES 1. O. Pänke, T. Balkenhohl, Impedance Spectroscopy and Biosensing, Adv Biochem Engin/Biotechnol (2008) 109: A. Bonyár, H. Sántha, Novel interdigitated transducer structures for biosensoric applications Proc of the 32nd International Spring Seminar on Electronics Technology, Brno, Czech Republic, 2009, pp Germarie Sánchez-Pomales, Lenibel Santiago-Rodríguez, Nelson E. Rivera-Vélez, Carlos R. Cabrera, "Control of DNA self-assembled monolayers surface coverage by electrochemical desorption", Journal of Electroanalytical Chemistry, Vol. 611, 2007, pp X.Cui, D.Jiang, et al. Assessing the apparent effective thickness of alkanethiol self-assembled monolayers in different concentrations of Fe(CN) 6 3- /Fe(CN) 6 4- by ac impedance spectroscopy Journal of Electroanalytical Chemistry, Vol. 470, 1999, pp Z. Li, T. Niu, Z. Zhang et al. Studies on the effect of solvents on self-assembly of thioctic acid and mercapto-hexanol on gold, Thin Solid Films (2011), doi: /j.tsf S. Hleli, C. Martelet, et al. An immunosensor for haemoglobin based on impedimetric properties of a new mixed self-assembled monolayer Materials Science and Engineering C Vol. 26, 2006, pp Author: Attila Bonyár Institute: Department of Electronics Technology, Budapest University of Technology and Economics Street: H-1111, Goldmann Gy. Square 3. City: Budapest Country: Hungary bonyar@ett.bme.hu