2015 XVIII AISEM Annual Conference
Thermal Characterization of Thin Film Heater for Lab-On-Chip Application Giulia Petrucci, Domenico Caputo, Augusto Nascetti, Nicola Lovecchio, Emanuele Parisi Samia Alameddine and Giampiero de Cesare DIET & DIAEE, “Sapienza” University of Rome Rome, Italy Email:
[email protected] Abstract— This paper presents the design, fabrication and characterization of a thin film heater functional for thermal treatments in lab-on-chip system. The spatial temperature distribution determined by different heater geometries has been studied through electro-thermal simulations by using COMSOL Multiphysics. The heater showing the more uniform temperature distribution has been subsequently fabricated and characterized. A very good agreement between modeled and measured data has been attained. Results show a spatial temperature distribution of about ±1ºC over an area comparable to the heater area and a directly USB powered heater, demonstrating the suitability of the proposed device for lab-on-chip thermal applications. Keywords— Lab-on-chip; Thin film heater on Temperature distribution; Electro-thermal simulations.
I.
glass;
INTRODUCTION
Lab-on-chip technology is receiving a lot of attention thanks to the different possible applications in biology and in genomics. Lab-on-chips are a compact and miniaturized platform able to achieve complete biological and chemical analyses in shorter time and with lower reagents consumption than a standard laboratory [1]. The development of lab-onchip requires the integration of multiple functions, in order to implement the different analytical protocols. The regulation of temperature is a critical issue in some of these protocols as occurs for example in the Polymerase Chain Reaction (PCR) technique for DNA amplification [2]. In order to implement a precise thermal control, the vast majority of studies exploits external approaches, such as external metal blocks (Peltier elements), which are, however, bulky and high power consumption [3]. Other technologies, embedding a microheater within the system, have been developed to reduce the lab-on-chip dimensions. A serpentine geometry was studied for example by Lao et al. [4]. They developed, on a silicon-based micromachined fluidic chamber, platinum heaters and sensors, integrating a digitally temperature control. Different heater geometries have been studied by Hsieh et al. [5], [6], by comparing their temperature response. They studied two blocks of heaters, two blocks with additional side heaters, an array of heaters with additional side heaters and self-compensated array of heaters. Selva et al. [7],
978-1-4799-8591-3/15/$31.00 ©2015 IEEE
Andleeb Zahra Centre for Life Nano Science @ Sapienza Istituto Italiano di Tecnologia Rome, Italy
[8] showed also that it is possible to optimize heater shape in order to generate different temperature profiles by making a finite element study of the thermal response of the heater. By patterning the substrate with an optimized resistor, it is possible to generate a uniform temperature distribution on a selected area with great accuracy and with short response time. The aim of this work is to design and characterize a thin film heater deposited on a glass substrate, featuring an heater area comparable to the active area of the thermal treatment, a temperature uniformity of ±1°C suitable for isothemal PCR and a low power consumption. Different heater geometries have been designed using finite element simulations (COMSOL Multiphysics), coupling the electrostatic and the heat transfer problems and obtaining the temperature profiles due to a potential difference applied across the resistor. Material selection and layer thicknesses have been also designed aiming to a USB powered device (at maximum 5 V and 500 mA). After the design process, the device showing the more uniform temperature distribution has been fabricated using thin film technologies. To validate the modeled results, an infrared thermo-camera measured the thermal distribution over the heater in steady state condition. The organization of the paper is as follows. In Section II the thin film heater design and electro-thermal simulations are described. Section III shows the fabrication of the thin film heaters. Section IV discusses the paper results by comparing the simulations and experiments. Finally, Section V draws the main conclusions. II.
SIMULATIONS
The investigated heater geometries, deposited on a 50x50 mm2 glass substrate, are reported together with their dimensions in Fig. 1. Each heater includes a central part devoted to the heating (represented by the dense vertical lines in Fig. 1a and 1b and by the circular geometry in Fig. 1c) and two vertical side lines for the electrode contacts (represented in the figures by the more wide vertical black pads). Fig. 1c also presents two additional pads for the 4-wire measurements, to determine precisely the voltage applied across the heater. The central region of each geometry is the effective area of the device, useful to biological protocols, over which the control of the thermal profile is required.
2015 XVIII AISEM Annual Conference
Fig. 1. Studied geometries: meander (a), chirp (b) and circular (c) with their main dimensions. Red squares represent the active area, i.e. the area required for the thermal treatment.
This active area is defined in the figure by the red squares. To develop a multiple function lab-on-chip, the dimensions of the active area are designed as close as possible to that of the central part of the heaters. When a current (or a voltage) is applied to the pad, the electrical power is dissipated as heat, inducing an increase of the temperature on the structure. The spatial temperature distribution induced on the glass substrate by thin film heaters have been investigated by using COMSOL Multiphysics software. This tool couples the electrical and the thermal problems via the Joule effect. The thermal module determines the heat transfer in the structure due to conduction, convection and radiation processes. In our
simulations, the heat exchange by radiation has not be considered, while the contribution by convection has been evaluated considering an external natural convection whose heat transfer coefficient (h) depends on the surface, on the atmospheric pressure and on the external temperature [h = hair (A, pa, Text)]. Heater geometries, materials and thicknesses have been designed considering a USB power supply, having a current limit of 500 mA and a supply voltage less than 5 V, and the size of the active area on which the uniform temperature distribution is required (about 8.5 mm x 8 mm). In order to compare the performances of the different geometries, the temperature distribution has been evaluated by setting the electrical power dissipated by each heater to induce a temperature of about 60°C (required for an isothermal PCR) inside the active area. We monitored the temperature distribution along the x- and y-axis, at three different coordinates for each axis. Results are summarized in Fig. 2. In the first column a sketch of the active area with the cut lines taken at three different coordinates are reported, while the following columns show the temperature trends for the three geometries. In particular, the meander geometry has a ∆Tx along each x line of about 5°C and a ∆Ty, always along each y line, of about 7°C. The maximum temperature difference is about 10°C, showing that this geometry provides a not satisfying temperature distribution. In particular, the temperature in the central part is higher than at the heater edges.
Fig. 2. Comparison between temperature trends of the three geometries along x- and y-axis. Cut lines have been taken in the active area at three different coordinates: red lines refer to 3500 μm from the center in upper and left direction for x- and y- axis respectively, blue lines have been taken in the center, while green lines refer to 3500 μm from the center in lower and right direction for x- and y- axis respectively.
978-1-4799-8591-3/15/$31.00 ©2015 IEEE
2015 XVIII AISEM Annual Conference For this reason, the geometry has been changed increasing the distance between the central arms and decreasing that of the side arms and at the same time increasing and decreasing their respectively width. For this new geometry (called chirp structure) the temperature uniformity along x-axis has been improved (∆Tx= 3°C). We can observe a plateau in the central part of the active area, along which there is a constant temperature. However, the temperature difference along the yaxis is still quite high (∆Ty=7.5°C). The maximum temperature difference is still of 10°C. The best results have been obtained with the circular geometry. As we can see from the last column of Fig. 2, the temperature profiles determined by this geometry along both x and y-axis are quite constant, demonstrating a very uniform temperature distribution. Indeed the maximum temperature difference on the whole active area is 2.5°C. In this simulation, the voltage was set to 4 V with a current of 177 mA, reaching in the steady-state condition the temperature desired for PCR application.
III.
FABRICATION
All heaters were deposited on a 50x50x1 mm3 glass substrate. One of the key points in the design of a lab-on-chip with integrated thermal management is related to the substrate material on which the device is fabricated, which, in turn, is often imposed by the requirements of the final application. Glass is the material of choice for conventional analytical applications and therefore glass substrates are of particular interest for lab-on-chip devices [9]-[11]. In our system, the electrical pads are made by 30nm/150nm/30nm-thick Cr/Al/Cr stacked layer, while the central part is a 1100nm-thick Indium Tin Oxide (ITO) film. ITO was chosen because it is a transparent material and allows a possible monitoring of biological reactions performed on the same lab-on-chip. On the other hand, the metal pads allow to keep the total heater resistance as low as possible. The fabrication of the heater has been implemented with the following processes: •
vacuum evaporation of three metal Cr/Al/Cr stack;
•
definition of the contact part of thin film heater by conventional photolithography and lift-off technique to prevent that chemical etching solution could damage other structures on the glass (see Fig. 3a);
•
sputtering of ITO;
•
definition of the central part of thin film heater by conventional photolithography and lift-off technique (see Fig. 3b).
The thicknesses of the metal stack and of the ITO have been chosen according to the simulation results. A picture of the fabricated heater is reported in Fig. 3c.
978-1-4799-8591-3/15/$31.00 ©2015 IEEE
Fig. 3. Masks used for Cr/Al/Cr (a) and ITO (b) patterning. In Figure 3c, an example of realized heaters.
IV.
RESULTS AND DISCUSSION
At first, a 4-wire measurement of the fabricated resistance has been performed. A value of 23.1 Ω for both ITO resistors has been found. This value is very close to the one utilized in the simulations. Indeed, when a voltage of 4 V has been imposed on the heater, the resulting current is 123 mA. To validate the results obtained through simulations, we have utilized an infrared thermo-camera (FLIR A325), to view and record the thermal distribution on the structure. A picture of the spatial temperature profile measured with the thermocamera in the steady-state condition is reported in Fig. 4a. For comparison, in Fig. 4b is reported the corresponding simulation. In order to have information about the uniformity of the temperature distribution, the average temperature and the standard deviation have been calculated in the black points of Fig. 4a and 4b, which are at the same coordinates. Results are reported in Table 1. An excellent agreement between simulations and measurements is achieved. Indeed, almost the same average temperature and same standard deviation is found in both simulations (60.90±2.31°C) and measurements (60.92°C±2.13°C). TABLE 1. AVERAGE TEMPERATURE AND STANDARD DEVIATION IN SIMULATION AND IN MEASUREMENT. Simulation
Measurement
Taverage (°C)
60.90
60.92
3*σ (°C)
2.31
2.13
2015 XVIII AISEM Annual Conference ACKNOWLEDGMENT Authors would like to thank the Center for Life Nano Science @ Sapienza, Istituto Italiano di Tecnologia (Rome, Italy) and the Italian Ministry of Education, University and Research (MIUR) through PRIN 2010-2011 Project ARTEMIDE (ref. 20108ZSRTR) and through University Research Project 2013 (prot. C26A13HKFB) for their financial support. REFERENCES [1]
Fig. 4. Simulations (a) and measurements (b) results.
V.
CONCLUSIONS
The presented work shows the thermal characterization of a thin film heater on a glass substrate for isothermal PCR. Different geometries have been studied with electro-thermal simulations with COMSOL Multiphysics, in order to achieve the more uniform temperature distribution. The geometry showing the best temperature profiles has been fabricated and characterized with a thermo-camera. A very good agreement between simulations and experimental results has been achieved, showing that a careful design of the heater geometry leads to a spatial temperature distribution of about ±1ºC (as the one required by the isothermal PCR) and a directly USB powered heater, whose area is very close to the thermal active zone.
978-1-4799-8591-3/15/$31.00 ©2015 IEEE
P. Abgrall, A.M. Gué, “Lab-on-chip technologies: making a microfluidic network and coupling it into a complete microsystem- a review”, Journal of Micromechanics and Microengineering, 17, 2007, R15-R49J. [2] M. Parida, S. Sammarangaiah, P. Kumar Dash, P. V. L. Rao and K. Morita, “Loop mediated isothermal amplification (LAMP): a new generation of innovative gene amplification technique; perspective in clinical diagnosis of infectious deseases”, Rev. Med. Virol. 200818:407-421, 2008. [3] V. Miralles, A. Huerre, F. Malloggiì i, M.C. Jullien, “A review oh heating and temperature control in microfluidic systems: techniques and applications” Diagnostic 2013, 3(1), 33-67. [4] A.I.K. Lao, T.M.H. Lee, I.-M. Hsing, N.Y. Ip, “Precise temperature control of microfluidic chamber for gas and liquid phase reactions”, Sens. Actuator. A 2000, 84, 11–17. [5] T.-M. Hsieh, C.-H. Luo, F.-C. Huang, J.-H. Wang, L.-J. Chien, G.-B. Lee, “Enhancement of thermal uniformity for a microthermal cycler and its application for polymerase chain reaction”, Sens. Actuator. B 2008, 130, 848–856. [6] T.-M. Hsieh, C.-H. Luo, J.-H. Wang, J.-L. Lin, K.-Y. Lien, G.-B. Lee, “Enhancement of thermal uniformity for a microthermal cycler and its application for polymerase chain reaction”, Microfluid. Nanofluid. 2009, 6, 797–809. [7] B. Selva, J. Marchalot, M.-C. Jullien, “An optimized resistor pattern for temperature gradient control in microfluidics”, J. Micromech. Microeng. 2009, 19, 065002:1–065002:10. [8] B. Selva, P. Mary, M.-C. Jullien, “Integration of a uniform and rapid heating source into microfluidic systems”, Microfluid. Nanofluid. 2010, 8, 755–765. [9] A. Scorzoni, M. Tavernelli, P. Placidi, P. Valigi, D. Caputo, G. de Cesare, G. Petrucci, A. Nascetti, “Thermal Characterization of a Thin Film Heater on Glass Substrate for Lab-on-Chip Applications”, in Instrumentation and Measurement Technology Conference (I2MTC) Proceedings, 2014 IEEE International (pp. 1089-1094). IEEE. [10] A. Scorzoni, M. Tavernelli, P. Placidi, P. Valigi, S. Zampolli, D. Caputo, G. Petrucci, A. Nascetti, “Improvement of Thermal Resistance of Thin Film Heaters on Glass Substrate for Lab-On-Chip Applications”, Procedia Engineering 87 ( 2014 ) 959 – 962. [11] A. Scorzoni, D. Caputo, G. Petrucci, P. Placidi, S. Zampolli, M. Tavernelli, G. de Cesare, A. Nascetti, “Design and Experimental Characterization of Thin Film Heaters on Glass Substrate for Lab-onChip Applications”, Sens and Actuator. A (in press).
