Silicon-based miniature sensor for electrical tomography

https://doi.org/10.1016/j.sna.2003.08.012Get rights and content

Abstract

The paper describes the fabrication of a novel miniature sensor for electrical tomography. The sensor comprises a number of copper electrodes that are fabricated around a small hole that is etched through a silicon wafer. Copper electrodes are electroplated to fill channels that are formed in thick photo-resist on top of the silicon wafer. Electrodes with a thickness of 60 μm, surrounding a hole of diameter 300 μm, have been realised. Initial measurements have been made using a commercial LCR meter applied to an eight-electrode sensor and images of a 80 μm diameter wire have been obtained. Future work will consider the integration of measurement circuitry alongside the electrodes in order to reduce parasitic capacitances.

Introduction

Over the last 15 years, electrical tomography has emerged as a valuable design, monitoring and diagnosis tool for industry [1]. The basic aim of tomography is to determine the distribution of materials in some region of interest by obtaining a set of measurements using sensors that are distributed around the periphery. The measurements are non-intrusive, perhaps penetrating the “wall” of the vessel but not entering into the medium, and also, ideally, non-invasive such that the sensors are located on the outside of the “wall”. Each measurement is affected, to a greater or lesser degree, by the location of materials in the region of interest. Typically, a source of energy is imposed on the vessel from one orientation and a number of measurements are taken by distributed sensors to create a projection of data. The source location is then moved to provide another projection and so on around the vessel until a frame of data is accumulated. In medicine, X-ray are commonly the source of energy. The X-ray beam is attenuated as it passes through the vessel in a way that depends on the density distribution of materials. Usually, the frame of data is translated, in software, into a cross-sectional image representing the distribution of density. This can then be interpreted to yield parameters of interest, for instance, the identification of tumours in medical applications. X-ray tomography was the first “modality” to be used in medicine, providing density contrast, but it has been augmented subsequently by techniques such as magnetic resonance imaging (water contrast) and positron emission tomography (dopant contrast).

Fuelled by developments in personal computing, sensor design and electronics, research into applications of tomography in industrial processes has gained in popularity, using electronic scanning techniques and no moving parts. Techniques have been influenced by the widespread success in medicine, however, in many cases, the demands of industrial applications are significantly different. It is not uncommon to require many cross-sectional images per second, at a low cost, using “mobile” equipment. For these reasons, nucleonic techniques are often inappropriate and alternatives have emerged. For instance, the literature includes descriptions of instruments that are based on acoustic, optical, infra-red, microwave and electrical sources of energy [2], [3]. Recent work has seen the migration of the technique onto the factory floor and instruments are now being established on production lines to yield benefit in terms of quality, environmental impact, process efficiency and manufacturing agility [4].

Typical applications of tomography to industry consider process vessels that range from centimetres to metres in diameter and instrumentation has been developed to deliver rapid, accurate measurements. New applications are beginning to emerge that may benefit from miniaturised electrical tomography sensors and the present work describes early efforts to explore one possible approach to realise such sensors. Although some miniature electrical tomography sensors have been described previously [5], [6], this paper introduces the first such sensor that is integrated on a silicon wafer. The work comprises two distinct aspects namely, creation of a hole through the silicon wafer by etching, and fabrication of thick electrodes, accurately located around the hole, using electroplating. It is desirable to realise the measurement circuitry as close as possible to the electrodes to reduce stray signals and, in this respect, silicon offers unique advantages. In addition, for capacitance measurements it is desirable to have relatively large electrodes in order to increase the measured signal. Unless the number of electrodes surrounding the hole is reduced, the only way in which this can be achieved is by increasing the thickness of the electrode material around the hole. A significant part of the present work has explored fabrication techniques to realise these aims. Results from finite element simulations predict the voltages that are to be expected on the electrodes when a current is injected into the region. The sensor has been tested using a commercial LCR meter, with switching matrix, to image a metal wire that is 80 μm in diameter. Although this is not ideal instrumentation for the purpose, being slow and bulky, it provides the necessary evidence to support further work on the sensor. The following sections describe, fabrication of the hole and electrodes; results of finite element simulations to predict voltages on the electrodes and preliminary results.

Section snippets

Fabrication process

The following sections describe the approach that has been adopted to realise a miniature sensor for electrical tomography comprising “thick” electrodes that are distributed around a small hole that is etched through a silicon chip. The work has been undertaken using the clean room facilities at UMIST.

Test results

An eight-electrode sensor, fabricated according to the above procedures has been tested using a tomography system that is based on a commercial Hewlett-Packard 4284A LCR meter. A block diagram of this system is shown in Fig. 3. The LCR meter is able to make four-point impedance measurements up to an excitation frequency of 1 MHz. To realise tomographic images, it is necessary to measure capacitance between all pairs of electrodes, i.e. 1–2, 1–3, 1–4, …, 2–3, 2–4, … For eight electrodes, this

Conclusions

The paper describes a fabrication process for creating “thick” copper electrodes around a hole that is etched through a silicon chip. The motivation for the work is to explore the possibility of producing miniature sensors for electrical tomography. The process includes two novel steps. Firstly, the hole has to be fabricated and this is achieved by etching. Secondly, the thick electrodes have to be created and accurately located around the hole. The electrodes are fabricated using thick

Trevor York obtained his PhD in 1982 in “electron spectroscopy of reactive molecules” at the University of Manchester. In 1985, he was appointed as a lecturer in the Department of Electrical Engineering and Electronics at UMIST, where he is now a reader. His research interests focus on aspects of process tomography notably exploitation of microelectronics, hardware acceleration of data processing and industrial applications. He has published over 90 papers.

References (9)

  • T.A York et al.

    Particle detection using an integrated capacitance sensor

    Sens. Actuators A

    (2001)
  • T.A York

    Status of electrical tomography in industrial applications

    J. Electron. Imag.

    (2001)
  • T.A. York (Ed.), Proceedings of the 1st World Congress on Industrial Process Tomography, Buxton, UK, April...
  • R. Mann (Ed.), Proceedings of the 2nd World Congress on Industrial Process Tomography, Hannover, Germany, August...
There are more references available in the full text version of this article.

Cited by (25)

  • Fabrication of microchannel with 60 electrodes and resistance measurement

    2010, Flow Measurement and Instrumentation
    Citation Excerpt :

    Griffiths et al. reported a circular array of 16 electrodes that was developed for a miniature sensor in the electrical computed tomography [11]. Also, York et al. [12,13] showed the 16 electrodes of one cross section in a miniature sensor by silicon etching, and reconstructed the concentration of a gas–solid flow in the one cross section by ECT. However, the miniature sensors are available for the only one cross section in the two phase flow.

View all citing articles on Scopus

Trevor York obtained his PhD in 1982 in “electron spectroscopy of reactive molecules” at the University of Manchester. In 1985, he was appointed as a lecturer in the Department of Electrical Engineering and Electronics at UMIST, where he is now a reader. His research interests focus on aspects of process tomography notably exploitation of microelectronics, hardware acceleration of data processing and industrial applications. He has published over 90 papers.

Sun Liling graduated in 1984 with a BSc degree in semiconductor physics and devices from the North West University in Xi’an province. She is a senior engineer of the Xi’an 771 Microelectronics Institute in China and has undertaken projects on, for instance, photo-resistant and plasma etching technology, multi-layered thick film substrates and high density assemblies. She spent 12 months, from April 2000, as an academic visitor in the Department of Electrical Engineering and Electronics at UMIST.

Chris Gregory received his BSc degree in microbiology from the University of Sheffield in 1994. He then received MSc in biosensors from the University of Manchester in 1995. His PhD degree was awarded in 1999 for work on an anti-fouling system for sensors working in harsh fluid environments. He is currently working as a research associate in the Department of Electrical Engineering and Electronics at UMIST in the field of medical ultrasound transducer development.

John Hatfield received his BSc degree in physics from the University of Leeds, Leeds, UK, in 1973, and the MSc degree from UMIST, Manchester, UK, in 1984. He received his PhD degree from the same university in 1988 for his research into position-sensitive particle detectors. Currently, he is a reader in the Department of Electrical Engineering and Electronics at UMIST. His research interests are in the field of integrated sensors and transducers and he has published more than 90 papers in this area.

View full text