HIGHLY SENSITIVE PRESSURE SENSOR USING A GOLD-COATED ELASTIC PYRAMID ARRAY PRESSING ON A RESISTOR N. Thanh-Vinh, K. Matsumoto, and I. Shimoyama The University of Tokyo, Tokyo, JAPAN ABSTRACT We propose a highly sensitive pressure sensor using an elastic micro pyramid array with a gold-coated surface pressing on a resistive silicon beam. When a pressure was applied on the pyramid array, the deformation of each pyramid induced the larger contact area between each pyramid and the resistor. Since the pyramid array was coated by a gold layer, which had much lower resistivity than that of the silicon beam, the resistance of each contact area drastically decreased. Therefore the resistance between two electrodes of the silicon beam decreased. Our demonstrated sensor showed the non-linear response to the applied pressure, which was caused by the increase in elasticity of each pyramid as its deformation became larger. This non-linear property can be used as an advantage to enlarge the sensing range of the sensor. The sensitivity of our sensor for the pressure from 0 to 1 kPa was 0.61 kPa-1. The resolution derived from the largest error, was 0.16 kPa.
Figure 1 The schematic sketch of our pressure sensor. The sensor is consisted of a gold-coated elastic micro pyramid array pressing on doped silicon beam (as a resistor).
INTRODUCTION Recently, various types of tactile sensors have been intensively reported with many promising applications such as dexterous manipulation in robotics or surgery systems, force measurement in sports, etc [1-2]. In order for a tactile sensor to be used in practical applications, the sensitivity and durability are critically required. Among many types of tactile sensor, piezoresistive and capacitive tactile sensors are most well-developed. Piezoresistive tactile sensor has the advantages such as high sensitivity, low noise and simple electronics [3-5]. However, since the output of a piezoresistive tactile sensor is proportional to the deformation of the piezoresistive elements, improving the sensitivity usually requires more fragile structures. Capacitive tactile sensor is also superior in sensitivity, however, the noise from parasitic capacitance, and complex readout circuit are major drawbacks [6, 7]. In this study, we propose a pressure sensor with high sensitivity and simple readout circuit. The concept of our sensor is shown in Figure 1. The sensor consists of a polydimethylsiloxane (PDMS) pad with god-coated micro pyramid array, which is pressed on a resistive silicon beam. The silicon beam was doped to reduce the resistance of the layer close to the surface. The sensing scheme of our sensor is to utilize the ohmic contacts between the elastic pyramid array with a high electrical conductivity surface and the rigid silicon beam. These contact areas expand under pressure, causing the relatively large fractional resistance change of the beam compared to the fractional resistance change obtained by the piezoresistive effect in conventional piezoresistive type tactile sensor. Actually, the fractional resistance change of
Figure 2 Pressure sensing principle of our sensor. When the pressure is applied, the ohmic contact areas between gold layer and the silicon beam increase, causing the resistance of the beam to decrease. The cracks prevent the electrical short between two adjacent pyramids. our sensor was large enough so that no amplifier was required. This advantage brings about the great simplicity for the measurement circuit. Moreover, in our sensor, there is no fragile silicon element and thus we can eliminate the failure caused by the fracture of micro silicon structure which is the main drawback of conventional piezoresisitve tactile sensors [1, 4].
PRESSURE SENSING PRINCIPLE Figure 2 shows the sensing principle of our pressure sensor. When a pressure is applied, the pyramid array
deforms, causing the contact areas between the array and the beam to increase. Since the pyramid array is coated with gold which has much lower electrical resistivity than doped silicon, the resistances of the contact areas are drastically reduced and thus, the resistance between the two electrodes of the beam decreases. We created cracks around the base of all the pyramids simply by bending the PDMS pad. When a pressure is applied, these cracks are widened, electrically isolating any two adjacent pyramids. In the other words, the silicon beam is partly electrically shorted in isolated contact areas. By measuring the resistance of the silicon beam, the applied pressure can be detected. It is worth noticing that in the sensing scheme of our sensor, the silicon beam does not deform, which is fundamentally different from the sensing method of conventional piezoresisitve tactile sensor.
FABRICATION Fabrication of the PDMS pyramid array The fabrication process flow of the PDMS pyramid array is shown in Figure 3(a). First, a glass layer was formed on the surface of a silicon wafer by thermal oxidization. This glass layer then was patterned and etched by buffered hydrogen fluoride. In the next step, the silicon wafer was anisotropically etched using a 20% Tetramethylammoniumhydroxid solution at 80 degree Celsius. The etching rate of this step was 0.5µm per minute. After that, we coated a thin layer of C4F8 onto the silicon wafer as an anti-adhesive layer before pouring PDMS in. After baking the PDMS at 70 degree Celsius for one hour, the PDMS pad with pyramid array was peeled off from the silicon mould. Since the PDMS itself is porous, it is difficult to directly deposit the gold layer on the PDMS surface. In our fabrication process, we deposited a 1 µm-thick parylene layer and then a 30 nm-thick gold layer onto the PDMS pyramid. The gold layer was easily deposited on the smooth parylene surface. Fabrication of the silicon beam The fabrication of the silicon beam is shown in Figure 3(b). First, the device silicon layer of a silicon on insulator (SOI 50µm/2µm/300µm) wafer was doped by rapid thermal diffusion process [8]. Then a gold layer was deposited on to the device silicon layer. After patterning the gold layer, the beam was created by etching silicon using ICP-RIE. The electrodes of the beam were formed by etching the unnecessary gold area. The doped silicon beam was chosen to be the resistor in our sensor because it can be easily fabricated by conventional MEMS process. Moreover, the electrical conductivity of a silicon beam can be increased by doping process, resulting in a relatively low resistance which can be easily measured by a simple bridge circuit. Sensor device Before attaching the PDMS pyramid array onto the silicon beam, we created the cracks around each pyramid by simply bending the PDMS pad (Figure 3(c)). Many methods could be used to obtain the electrical isolation between two
Figure 3 Fabrication process of the PDMS micro pyramid array (a) and the resistive Si beam (b). (c) The introduction of cracks and sensor device assembly. adjacent pyramids, such as patterning the gold layer or depositing the gold layer with a shadow mask. However these methods are complicated and difficult due to the height of the micro pyramids. Fabrication results The fabricated PDMS pad with micro pyramid array is shown is shown in Figure 4. The size of the PDMS pad was 10 mm × 10 mm × 1 mm. The base size and pitch of the pyramid array were 10 µm × 10 µm and 12.5 µm, respectively. An SEM image in Figure 4 (b) shows the smooth surface of the pyramid array before the PDMS pad was bent. However, as shown in Figure 4 (c) and (d), after the bending the PDMS pad, cracks appeared around each pyramid enabling the electrical isolation between any two
Figure 4 (a) Fabricated gold-coated PDMS pyramid array. (b) Smooth surface of the pyramids before bending. (c), (d) Cracks appeared around the pyramids after bending the PDMS pad. The size of the pyramid base was 10 µm × 10 µm. The pitch of the pyramid array was 12.5 µm.
Figure 5 (a) Photograph of the fabricated doped silicon beam as a resistor. (b) Photograph of the sSensor device with the PDMS pyramid array attached on the beam. adjacent pyramids. The fabricated silicon beam and the completed sensor device are shown in Figure 5. The dimensions and initial resistance of the silicon beam were 10 mm × 3 mm × 50 µm and 410 Ω, respectively.
EXPERIMENTS AND RESULTS Response of the sensor to pressure The experiment set up to evaluate the response of our sensor to pressure is shown in Figure 6a. We applied the pressure on the sensor by using a linear stage with the speed of 3mm/min. The applied pressure was measured by a digital force gauge (IMADA ZP-50N). A sponge was sandwiched
Figure 6 (a) The experiment set up to evaluate the response of our sensor to applied pressure. (b) The relationship between the applied pressure and the resistance change of the silicon beam. As the pressure increased, the sensor showed the non-linear response with the high sensitivity for the low pressure range. the increase of pressure as the force gauge was moved by the linear stage. Figure 6b shows the resistance change of the silicon beam responding to pressure. The result shows the non-linearly response of our sensor to the applied pressure. The resistance decreased rapidly as the applied pressure increased for the low pressure range (less than 1 kPa). When the pressure became larger (more than 5 kPa), the decreasing slope of the resistance became smaller. This characteristic in the response of our sensor was directly caused by the non-linear deformation of each pyramid under the pressure. As the pressure increased, the contact area between each pyramid and the silicon beam increased which also resulted in the larger elastic resistance of each pyramid. Therefore the contact areas between the pyramid array and the silicon beam first increased rapidly in the small pressure range but slowed down as the pressure increased. This non-linear property in deformation of a micro pyramid array was also reported in previous study [9]. The non-linear property is actually a benefit to enlarge the sensing range of the sensor. For pressure below 1 kPa, the sensitivity of our sensor was 0.61 kPa-1 which is more than 10000-fold higher than the sensitivity of the pressure sensor using a 300nm-thick piezoresistive beam embedded inside PDMS [4]. From the maximum error indicated in the graph, the resolution of our
ACKNOWLEDGEMENT The photolithography masks were made using the University of Tokyo VLSI Design and Education Center (VDEC)’s 8 inch EB writer F5112 + VD01 donated by ADVANTEST Corporation.
REFERENCES
Figure 7 The response of our sensor when placing and removing a weight of 2 gram, corresponding to a pressure of about 0.2 kPa between the force gauge and the sensor in order to slow down sensor was calculated to be 0.16 kPa. From the result, one may find that the resolution of our sensor in low pressure range was actually larger than that in high pressure range. The large error in low pressure range is thought to be caused by the unstable contact between the pyramid array and the silicon beam when the pressure was small. As the pressure increased, the contact between the pyramid array and the beam became more stable, making the error in the change of the resistance become smaller. We believe that by improving the packaging process, the more stable contact between the pyramid array and the silicon beam and thus higher sensing resolution can be obtained. Real time response to a weight of 2 gram To confirm the reversibility of the deformation of the pyramid array, we measured the resistance of the silicon beam when placing then removing a 2-gram weight corresponding to a pressure of 0.2 kPa. As shown in Figure 7, the resistance of the silicon beam immediately recovered after removing the weight. This result indicates that the pyramid array did not stick to the silicon beam and was able to recover to the initial shape as the load was removed.
CONCLUSIONS In conclusions, we proposed a highly sensitivity pressure which is consisted of a gold-coated PDMS micro pyramid array pressing on a resistive silicon beam. The pressure sensing scheme is to utilize the ohmic contact between the pyramid and the beam, which increases as the pressure is applied on the PDMS pyramid array. The sensor exhibited the non-linear response to applying pressure, which was caused by the nature in deformation of the pyramids. For pressure below 1 kPa, the sensitivity of our sensor was 0.61 kPa-1, which was more than 10000 times higher than that of conventional piezo-resistive tactile sensor. Our sensor achieved the sensing resolution of 0.16 kPa.
[1] Hanna Yousef, Mehdi Boukalle, Kaspar Althoefer, “Tactile sensing for dexterous in-hand manipulation in robotics - A review”, Sens. Actuator A-Phys., 167, pp. 171-187, 2011. [2] M.E.H. Eltaib, J.R. Hewit, “Tactile sensing technology for minimal access surgery - A review”, Mechatronics, 13, 1163-1177, 2003. [3] Hidetoshi Takahashi, Akihito Nakai, Kiyoshi Matsumoto, Isao Shimoyama, “A shear force detector using piezo-resistive beams with side-wall doping”, Proceedings of 25th IEEE MEMS Conference (MEMS2012), Paris, Jan. 29- Feb. 2, pp. 599 – 602. [4] Kentaro Noda, Y. Hashimoto, Y. Tanaka, I. Shimoyama, “MEMS on robot applications”, in Digest Tech. Papers Transducers‘09 Conference, Denver, June 21-25, pp. 2176–2181. [5] M. Sohgawa, T. Mima, H. Onishi, T. Kanashima, M. Okuyama, K. Yamashita, M. Noda, M. Higuchi, and H. Noma, “Tactile array sensor with inclined chromium/silicon piezoresistive cantilevers embedded in elastomer”, in Digest Tech. Papers Transducers‘09 Conference, Denver, June 21-25, pp. 284–287. [6] H. Lee, J. Chung, S. Chang, E. Yoon, “Normal and shear force measurement using a flexible polymer tactile sensor with embedded multiple capacitors”, J. Microelectromech. Syst., vol. 17 pp.934–942, 2008. [7] M. Y. Cheng, C. L. Lin, Y. T. Lai, Y. J. Yang, “A Polymer-Based Capacitive Sensing Array for Normal and Shear Force Measurement”, Sens. Actuator A-Phys, p.009, 2010. [8] Murat Gel and Isao Shimoyama, “Force sensing submicrometer thick cantilevers with ultra-thin piezoresistors by rapid thermal diffusion”, J. Micromech. Microeng., vol. 14, pp. 423-428, 2003. [9] Stefan C. B. Mannsfeld, Benjamin C-K. Tee, Randall M. Stoltenberg, Christopher V. H-H. Chen, Soumendra Barman, Beinn V. O. Muir, Anatoliy N. Sokolov, Colin Reese and Zhenan Bao, “Highly sensitive flexible pressure sensors with microstructured rubber dielectric layers”, Nat. Mater., vol. 9, pp. 859-864, 2010.
CONTACT *Nguyen Thanh-Vinh The University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo, 113-8656, Japan Tel: +81-3-5841-6318, Fax: +81-3-3818-0835 E-mail:
[email protected]
