- Email: [email protected]
Initial pits for electrochemical etching in hydrofluoric acid
Sensors and Actuators 85 Ž2000. 390–394 www.elsevier.nlrlocatersna
Initial pits for electrochemical etching in hydrofluoric acid H. Ohji a,) , P.J. French b, S. Izuo a , K. Tsutsumi a a b
Mitsubishi Electric Corporation, AdÕanced Technology R & D Center, 8-1-1, Tsukaguchi-Honmachi, Amagasaki, Hyogo 661-8661, Japan Delft UniÕersity of Technology, ITS r Et, Laboratory for Electronic Instrumentationr DIMES, Mekelweg 4, Delft 2628 CD, The Netherlands Received 4 October 1999; received in revised form 15 February 2000; accepted 21 February 2000
Abstract This paper reports on a characterization of structures fabricated by electrochemical etching in hydrofluoric acid ŽHF. using different types of initial pits. An initial pit has been thought to require a sharp tip in order to collect electronic holes generated by illumination. Therefore, the initial pits have been formed by etching in a KOH solution, which suffers from crystal orientation dependence. We successfully demonstrate the structures fabricated by the electrochemical etching with the initial pits, which have flat or round base. These pits are formed by isotropic wet etching or reactive ion etching ŽRIE., which are free from crystal orientation of silicon substrate. Furthermore, regular hole patterns can be achieved without initial pits. This makes the electrochemical etching process more simple. The electric field in the silicon is calculated and the calculated results support the experimental results. q 2000 Elsevier Science S.A. All rights reserved. Keywords: Electrochemical etching; Initial pit; Wet etching
1. Introduction The electrochemical etching in hydrofluoric acid ŽHF. acid has been known as a technique for porous silicon formation w1x. There are three types of porous silicon. One is macroporous silicon, another is mesoporous silicon and the other is microporous silicon. These three types of porous silicon are defined by the pore diameter. For the micro- and mesoporous silicon, the pores are formed isotropically and highly interconnected and thus, the pore direction and structure cannot be controlled. On the other hand, a regular macropore pattern has been demonstrated using initial pits as a starting point w2x. This makes it possible to control the location of the macropores. If v-shaped grooves are used as an initial pit, the trench structures can be achieved by the electrochemical etching in Ž100. silicon substrate w3x. During the electrochemical etching, the etched width can be changed with the current density controlled by the light intensity w4x. Using this effect, the current density can be increased after obtaining the trenches to connect the trenches under the structures w3x. Thus, the free-standing beams can be obtained in one step. Furthermore, an accelerometer structure has been ) Corresponding author. Tel.: q81-6-6497-7511; fax: q81-6-64977295. E-mail address: [email protected] ŽH. Ohji..
demonstrated using the single-step electrochemical etching for microstructures ŽSEEMS. w5x. However, the structures suffer from crystal orientation of silicon substrate due to the anisotropic KOH etching for initial pits. The initial pits have been thought to require a sharp tip to collect the electronic holes, which work as a trigger for chemical reaction. The influence of the initial pit shapes on the etched structure has not previously been studied. In this study, the initial pits are formed by isotropic wet etching or reactive ion etching ŽRIE. followed by the electrochemical etching. These two methods for initial pit formation do not suffer from crystal orientation of silicon substrate. Thus, the etched structures fabricated by the electrochemical etching are also free from crystal orientation. The effects of the initial pit shape on the etched structure are given in this paper.
2. Experimental A sketch of the electrochemical etching set-up is given in Fig. 1. A sample is fixed on a copper sample holder, which is screwed in the etch bath containing 5% HF w5x. The sample holder has a hole to illuminate the back side of the sample with a white light of which the intensity can be varied. A positive voltage is applied to the sample with reference to a platinum counter electrode. Current is ad-
0924-4247r00r$ - see front matter q 2000 Elsevier Science S.A. All rights reserved. PII: S 0 9 2 4 - 4 2 4 7 Ž 0 0 . 0 0 4 3 0 - 1
H. Ohji et al.r Sensors and Actuators 85 (2000) 390–394
391
Fig. 3. Initial pit formed by an isotropic wet etching.
Fig. 1. Electrochemical etching set-up.
justed by the light intensity and monitored by an ammeter. The process flow for electrochemical etching is given in Fig. 2. Starting material is n-type Ž100. silicon with the resistivity of 1–10 V cm. A 0.3 mm silicon nitride layer was formed and patterned to define the location of the initial pits. The initial pits were formed by an isotropic wet etching or RIE followed by the electrochemical etching. In the case of isotropic wet etching, HF, HNO 3 and acetic acid were used as an etchant. The electrochemical etching was done using the set-up in Fig. 1. After the electrochemical etching, the sample was cut to observe a cross-section by scanning electron microscopy ŽSEM..
given in Fig. 3. A side etching can be observed under the mask layer and the shape of the initial pit is relatively rounded with the depth of 1 mm. This formation technique for the initial pits is free from substrate crystal orientation. The SEM micrograph of the holes, which were fabricated by the electrochemical etching following the initial pits, is given in Fig. 4. Four holes can be seen under each of the initial pits with the pitch of 8 mm. The electrochemical etching starts at the four corners of each initial pit due to higher current density at corners of an initial pit w5x. On the other hand, the electrochemical etching always starts at the center of initial pits formed by KOH etching, because the initial pit has inverse pyramid shape in which the tip collects electronic holes. When the higher current density was used in the electrochemical etching, the hole width enlarged, holes at each corner of the initial pit were
3. Results 3.1. Initial pit formed by isotropic wet etching The SEM micrograph of the initial pit formed by the isotropic wet etching Žusing HF, HNO 3 and acetic acid. is
Fig. 4. Holes fabricated by the electrochemical etching following the initial pits formed by an isotropic wet etching.
Fig. 2. Process flow for electrochemical etching.
Fig. 5. Holes fabricated by higher current density.
H. Ohji et al.r Sensors and Actuators 85 (2000) 390–394
392
Fig. 6. Holes fabricated by small pitch of initial pits.
connected and became one hole with the width of 7 mm, as shown in Fig. 5. The same initial pits layout of Fig. 4 was used in this experiment. If different current density is used, width and pitch of etched hole can be varied. A regular hole pattern with width and pitch of 3 and 4 mm, respectively, is shown in Fig. 6. In this electrochemical etching, the current density was the same as for Fig. 4. However, the resulting width and pitch of the initial pit were different. In this case, the initial pit was too small to keep the four etched holes independently at the four corners of the initial pit. Thus, one etched hole can be achieved under each initial pit in Fig. 6. In this way, many types of etched hole configuration can be obtained using a certain light intensity and a different initial pit layout. It has been confirmed that it is not always necessary for the initial pit with sharp tip to control the location of the etched holes.
Fig. 8. Holes fabricated by the different depths of initial pits.
hole can be observed under each initial pit. This makes the etched structure more precise in dimensions and gives
3.2. Initial pit formed by RIE The SEM micrograph of the initial pit formed by RIE is given in Fig. 7. The bottom of the initial pit is flat and the wall is slightly wavy. Fig. 8 shows the etched holes formed by the electrochemical etching using the different depths of initial pits made by RIE and the same current density was used in these experiments. When the depth of initial pit was shallower than 1.2 mm, in Ža. and Žb., four etched holes were formed under the each initial pit. This configuration is as same as in Fig. 4. On the other hand, when the initial pits were deeper than 3 mm, one etched
Fig. 7. Initial pit formed by RIE.
Fig. 9. Etched holes fabricated without initial pits.
H. Ohji et al.r Sensors and Actuators 85 (2000) 390–394
393
Fig. 10. Model for the calculation.
considerable freedom for a design of devices due to the independence of the substrate crystal orientation for the initial pits and resulting etched structure. It is difficult to make a deep initial pit using the isotropic wet etching due to the side etching effect and thus, the etched shape of the structure has some limitations. However, the wet etching has an advantage in which the process flow for the electrochemical etching is much simpler. Therefore, it depends on the structures whose method for initial pit should be used.
holes was defined by patterning for the SiN masking layer. When the pitch of open square in SiN mask is large, 8 mm, a set of four etched holes can be observed in Fig. 9a. In the case of the small pitch of the open square, 6 mm, a regular etched hole pattern can be achieved in Fig. 9b. Therefore, with careful design of the silicon nitride masking layout, it is not always necessary to form a initial pit. This makes the electrochemical etching process more simple.
3.3. Without initial pit
From the theoretical point of view, the electronic holes work as a trigger for the chemical reaction in the electrochemical etching. In order to clarify the etching mechanism, the electric field in the sample was calculated by the MIDSIP-T ŽMitsubishi Device Simulation Program Trian-
The SEM micrographs of the etched hole pattern, which were fabricated without initial pits, are given in Fig. 9a and b. In these experiments, the location of the etched
3.4. Calculation
Fig. 11. Calculated electric field at the edge of the mask Ža. and the initial pit Žb..
394
H. Ohji et al.r Sensors and Actuators 85 (2000) 390–394
gular mesh.. The model for the calculations is given in Fig. 10. The sample was n-type silicon with the doping density 5 = 10 15 rcm3, and n q layer with the doping density 1 = 10 19 rcm3 was formed to make an ohmic contact. The masking layer was silicon dioxide with the thickness 0.5 mm and the open area was 8 = 8 mm2 for the electrochemical etching. One volt was applied to the n q layer. The difference of the Fermi level between the silicon and the etchant was assumed to be 0.5 eV. In the n-type silicon, the electronic hole is a minority carrier, so the electronic holes are generated by the illumination from the back side of the sample, as shown in Fig. 1. Then the electronic holes diffuse to the Sirelectrolyte interface due to the difference of the electronic hole density. In this calculation, the electron hole pairs were supplied from the surface of the n q layer with the density of 1 = 10 20 pairs cmy3 sy1 as the photogenerated holes. Fig. 11 shows the electric field for two types. One is the initial pit with the depth of 1 mm, and the other has no initial pit on the silicon surface. These two results, Fig. 11a and b, are quite similar. Higher electric field can be seen at the edge of the mask and the initial pit. While inside the silicon, the electric field is very weak. It means that the electronic holes diffuse from the back to the front and come close to the edge of initial pit. The electronic holes are accelerated by the strong electric field near the edge of initial pit and reach the edge of initial pit. Thus, the electrochemical etching starts at the corners of initial pit and the mask edge. These calculated results support the experimental results in ŽFigs. 4, 8a,b and 9a..
4. Conclusions In this paper, the etched holes made by the electrochemical etching have been demonstrated. When the certain mask layout and current density are chosen, the initial pits formed by an isotropic wet etching or RIE can be used for the electrochemical etching. It has been confirmed that it is not always necessary for the initial pit with sharp tip to control the location of the etched holes. Some of the methods for the initial pits are free from substrate crystal orientation. This gives the considerable freedom for the device design using the electrochemical etching. Furthermore, a regular etched hole pattern can be obtained without an initial pit although the shape of the etched structure has some limitations. This makes the electrochemical etching process even more simple. However, the location of the etched hole cannot be controlled precisely in this case. The electric field in the silicon substrate has been calculated. The calculated results support the experimental results.
Acknowledgements The authors would like to thank the staff and processing crew of DIMES for assistance in preparing the samples. In addition, we wish to thank T. Ozeki of Mitsubishi Electric for his encouragement.
References w1x D.R. Turner, Electropolishing silicon in HF acid solutions, J. Electrochem. Soc. 105 Ž1958. 402–408. w2x V. Lehmann, H. Foll, Formation mechanism and properties of electrochemically etched trenches in n-type silicon, J. Electrochem. Soc. 137 Ž1990. 653–659. w3x H. Ohji, P.J. Trimp, P.J. French, Fabrication of free standing structure using single-step electrochemical etching in hydrofluoric acid, Sensors and Actuators, A 73 Ž1999. 95–100. w4x H. Ohji, S. Lahteenmaki, P.J. French, Macro porous silicon formation for micromachining, Proceedings SPIE, Micromachining and Microfabrication Process Technology III, Austin, TX, USA, September Ž1997. 189–197. w5x H. Ohji, P.T.J. Gennissen, P.J. French, K. Tsutsumi, Fabrication of accelerometer using single-step electrochemical etching for microstructures ŽSEMMS.,IEEE MEMS Conference ’99, Orlando, USA, 1999, pp. 61–65.
Biographies Hiroshi Ohji received his BSc and MSc in Precision Engineering from Osaka University, Japan in 1986 and 1988, respectively. In the same year, he joined Mitsubishi Electric, Japan. He moved to the Netherlands in 1996. For 2 years, he was a member of the Laboratory for Electronic Instrumentation, Delft University of Technology as a visiting scientist in the field of microfabrication technologies and their applications. He returned to Mitsubishi Electric in July 1998 and has been working on automotive sensors and wet etching technologies. Paddy French received his BSc in Mathematics and MSc in Electronics from Southampton University, UK, in 1981 and 1982, respectively. In 1986, he obtained his PhD, also from Southampton University, which was a study of the piezoresistive effect in polysilicon. After 18 months as a post-doctoral candidate at Delft University, the Netherlands, he moved to Japan in 1988. For 3 years, he worked on sensors for automotives at the Central Engineering Laboratories of Nissan Motor. He returned to Delft University in May 1991 and is now a staff member of the Laboratory for Electronic Instrumentation with interests in micromachining and process optimisation related to sensors. Shin-ichi Izuo received the BSc and MSc in Material Engineering from Kyoto University, Japan in 1996 and 1998, respectively. In the same year, he joined Mitsubishi Electric, Japan. He has been working on micromachining and wet etching technologies. Kazuhiko Tsutsumi received the BS Ž1976. and MS Ž1978. degrees in Material Engineering and PhD degree in Electrical Engineering Ž1986. from Osaka University, Japan. Since 1982, he has worked as Research Scientist at Mitsubishi Electric, Japan and he is now a group manager in Advanced Technology R&D Center. Research interests in his group include device physics, microsensors, micromachining and process optimisation related to sensors.
Initial pits for electrochemical etching in hydrofluoric acid H. Ohji a,) , P.J. French b, S. Izuo a , K. Tsutsumi a a b
Mitsubishi Electric Corporation, AdÕanced Technology R & D Center, 8-1-1, Tsukaguchi-Honmachi, Amagasaki, Hyogo 661-8661, Japan Delft UniÕersity of Technology, ITS r Et, Laboratory for Electronic Instrumentationr DIMES, Mekelweg 4, Delft 2628 CD, The Netherlands Received 4 October 1999; received in revised form 15 February 2000; accepted 21 February 2000
Abstract This paper reports on a characterization of structures fabricated by electrochemical etching in hydrofluoric acid ŽHF. using different types of initial pits. An initial pit has been thought to require a sharp tip in order to collect electronic holes generated by illumination. Therefore, the initial pits have been formed by etching in a KOH solution, which suffers from crystal orientation dependence. We successfully demonstrate the structures fabricated by the electrochemical etching with the initial pits, which have flat or round base. These pits are formed by isotropic wet etching or reactive ion etching ŽRIE., which are free from crystal orientation of silicon substrate. Furthermore, regular hole patterns can be achieved without initial pits. This makes the electrochemical etching process more simple. The electric field in the silicon is calculated and the calculated results support the experimental results. q 2000 Elsevier Science S.A. All rights reserved. Keywords: Electrochemical etching; Initial pit; Wet etching
1. Introduction The electrochemical etching in hydrofluoric acid ŽHF. acid has been known as a technique for porous silicon formation w1x. There are three types of porous silicon. One is macroporous silicon, another is mesoporous silicon and the other is microporous silicon. These three types of porous silicon are defined by the pore diameter. For the micro- and mesoporous silicon, the pores are formed isotropically and highly interconnected and thus, the pore direction and structure cannot be controlled. On the other hand, a regular macropore pattern has been demonstrated using initial pits as a starting point w2x. This makes it possible to control the location of the macropores. If v-shaped grooves are used as an initial pit, the trench structures can be achieved by the electrochemical etching in Ž100. silicon substrate w3x. During the electrochemical etching, the etched width can be changed with the current density controlled by the light intensity w4x. Using this effect, the current density can be increased after obtaining the trenches to connect the trenches under the structures w3x. Thus, the free-standing beams can be obtained in one step. Furthermore, an accelerometer structure has been ) Corresponding author. Tel.: q81-6-6497-7511; fax: q81-6-64977295. E-mail address: [email protected] ŽH. Ohji..
demonstrated using the single-step electrochemical etching for microstructures ŽSEEMS. w5x. However, the structures suffer from crystal orientation of silicon substrate due to the anisotropic KOH etching for initial pits. The initial pits have been thought to require a sharp tip to collect the electronic holes, which work as a trigger for chemical reaction. The influence of the initial pit shapes on the etched structure has not previously been studied. In this study, the initial pits are formed by isotropic wet etching or reactive ion etching ŽRIE. followed by the electrochemical etching. These two methods for initial pit formation do not suffer from crystal orientation of silicon substrate. Thus, the etched structures fabricated by the electrochemical etching are also free from crystal orientation. The effects of the initial pit shape on the etched structure are given in this paper.
2. Experimental A sketch of the electrochemical etching set-up is given in Fig. 1. A sample is fixed on a copper sample holder, which is screwed in the etch bath containing 5% HF w5x. The sample holder has a hole to illuminate the back side of the sample with a white light of which the intensity can be varied. A positive voltage is applied to the sample with reference to a platinum counter electrode. Current is ad-
0924-4247r00r$ - see front matter q 2000 Elsevier Science S.A. All rights reserved. PII: S 0 9 2 4 - 4 2 4 7 Ž 0 0 . 0 0 4 3 0 - 1
H. Ohji et al.r Sensors and Actuators 85 (2000) 390–394
391
Fig. 3. Initial pit formed by an isotropic wet etching.
Fig. 1. Electrochemical etching set-up.
justed by the light intensity and monitored by an ammeter. The process flow for electrochemical etching is given in Fig. 2. Starting material is n-type Ž100. silicon with the resistivity of 1–10 V cm. A 0.3 mm silicon nitride layer was formed and patterned to define the location of the initial pits. The initial pits were formed by an isotropic wet etching or RIE followed by the electrochemical etching. In the case of isotropic wet etching, HF, HNO 3 and acetic acid were used as an etchant. The electrochemical etching was done using the set-up in Fig. 1. After the electrochemical etching, the sample was cut to observe a cross-section by scanning electron microscopy ŽSEM..
given in Fig. 3. A side etching can be observed under the mask layer and the shape of the initial pit is relatively rounded with the depth of 1 mm. This formation technique for the initial pits is free from substrate crystal orientation. The SEM micrograph of the holes, which were fabricated by the electrochemical etching following the initial pits, is given in Fig. 4. Four holes can be seen under each of the initial pits with the pitch of 8 mm. The electrochemical etching starts at the four corners of each initial pit due to higher current density at corners of an initial pit w5x. On the other hand, the electrochemical etching always starts at the center of initial pits formed by KOH etching, because the initial pit has inverse pyramid shape in which the tip collects electronic holes. When the higher current density was used in the electrochemical etching, the hole width enlarged, holes at each corner of the initial pit were
3. Results 3.1. Initial pit formed by isotropic wet etching The SEM micrograph of the initial pit formed by the isotropic wet etching Žusing HF, HNO 3 and acetic acid. is
Fig. 4. Holes fabricated by the electrochemical etching following the initial pits formed by an isotropic wet etching.
Fig. 2. Process flow for electrochemical etching.
Fig. 5. Holes fabricated by higher current density.
H. Ohji et al.r Sensors and Actuators 85 (2000) 390–394
392
Fig. 6. Holes fabricated by small pitch of initial pits.
connected and became one hole with the width of 7 mm, as shown in Fig. 5. The same initial pits layout of Fig. 4 was used in this experiment. If different current density is used, width and pitch of etched hole can be varied. A regular hole pattern with width and pitch of 3 and 4 mm, respectively, is shown in Fig. 6. In this electrochemical etching, the current density was the same as for Fig. 4. However, the resulting width and pitch of the initial pit were different. In this case, the initial pit was too small to keep the four etched holes independently at the four corners of the initial pit. Thus, one etched hole can be achieved under each initial pit in Fig. 6. In this way, many types of etched hole configuration can be obtained using a certain light intensity and a different initial pit layout. It has been confirmed that it is not always necessary for the initial pit with sharp tip to control the location of the etched holes.
Fig. 8. Holes fabricated by the different depths of initial pits.
hole can be observed under each initial pit. This makes the etched structure more precise in dimensions and gives
3.2. Initial pit formed by RIE The SEM micrograph of the initial pit formed by RIE is given in Fig. 7. The bottom of the initial pit is flat and the wall is slightly wavy. Fig. 8 shows the etched holes formed by the electrochemical etching using the different depths of initial pits made by RIE and the same current density was used in these experiments. When the depth of initial pit was shallower than 1.2 mm, in Ža. and Žb., four etched holes were formed under the each initial pit. This configuration is as same as in Fig. 4. On the other hand, when the initial pits were deeper than 3 mm, one etched
Fig. 7. Initial pit formed by RIE.
Fig. 9. Etched holes fabricated without initial pits.
H. Ohji et al.r Sensors and Actuators 85 (2000) 390–394
393
Fig. 10. Model for the calculation.
considerable freedom for a design of devices due to the independence of the substrate crystal orientation for the initial pits and resulting etched structure. It is difficult to make a deep initial pit using the isotropic wet etching due to the side etching effect and thus, the etched shape of the structure has some limitations. However, the wet etching has an advantage in which the process flow for the electrochemical etching is much simpler. Therefore, it depends on the structures whose method for initial pit should be used.
holes was defined by patterning for the SiN masking layer. When the pitch of open square in SiN mask is large, 8 mm, a set of four etched holes can be observed in Fig. 9a. In the case of the small pitch of the open square, 6 mm, a regular etched hole pattern can be achieved in Fig. 9b. Therefore, with careful design of the silicon nitride masking layout, it is not always necessary to form a initial pit. This makes the electrochemical etching process more simple.
3.3. Without initial pit
From the theoretical point of view, the electronic holes work as a trigger for the chemical reaction in the electrochemical etching. In order to clarify the etching mechanism, the electric field in the sample was calculated by the MIDSIP-T ŽMitsubishi Device Simulation Program Trian-
The SEM micrographs of the etched hole pattern, which were fabricated without initial pits, are given in Fig. 9a and b. In these experiments, the location of the etched
3.4. Calculation
Fig. 11. Calculated electric field at the edge of the mask Ža. and the initial pit Žb..
394
H. Ohji et al.r Sensors and Actuators 85 (2000) 390–394
gular mesh.. The model for the calculations is given in Fig. 10. The sample was n-type silicon with the doping density 5 = 10 15 rcm3, and n q layer with the doping density 1 = 10 19 rcm3 was formed to make an ohmic contact. The masking layer was silicon dioxide with the thickness 0.5 mm and the open area was 8 = 8 mm2 for the electrochemical etching. One volt was applied to the n q layer. The difference of the Fermi level between the silicon and the etchant was assumed to be 0.5 eV. In the n-type silicon, the electronic hole is a minority carrier, so the electronic holes are generated by the illumination from the back side of the sample, as shown in Fig. 1. Then the electronic holes diffuse to the Sirelectrolyte interface due to the difference of the electronic hole density. In this calculation, the electron hole pairs were supplied from the surface of the n q layer with the density of 1 = 10 20 pairs cmy3 sy1 as the photogenerated holes. Fig. 11 shows the electric field for two types. One is the initial pit with the depth of 1 mm, and the other has no initial pit on the silicon surface. These two results, Fig. 11a and b, are quite similar. Higher electric field can be seen at the edge of the mask and the initial pit. While inside the silicon, the electric field is very weak. It means that the electronic holes diffuse from the back to the front and come close to the edge of initial pit. The electronic holes are accelerated by the strong electric field near the edge of initial pit and reach the edge of initial pit. Thus, the electrochemical etching starts at the corners of initial pit and the mask edge. These calculated results support the experimental results in ŽFigs. 4, 8a,b and 9a..
4. Conclusions In this paper, the etched holes made by the electrochemical etching have been demonstrated. When the certain mask layout and current density are chosen, the initial pits formed by an isotropic wet etching or RIE can be used for the electrochemical etching. It has been confirmed that it is not always necessary for the initial pit with sharp tip to control the location of the etched holes. Some of the methods for the initial pits are free from substrate crystal orientation. This gives the considerable freedom for the device design using the electrochemical etching. Furthermore, a regular etched hole pattern can be obtained without an initial pit although the shape of the etched structure has some limitations. This makes the electrochemical etching process even more simple. However, the location of the etched hole cannot be controlled precisely in this case. The electric field in the silicon substrate has been calculated. The calculated results support the experimental results.
Acknowledgements The authors would like to thank the staff and processing crew of DIMES for assistance in preparing the samples. In addition, we wish to thank T. Ozeki of Mitsubishi Electric for his encouragement.
References w1x D.R. Turner, Electropolishing silicon in HF acid solutions, J. Electrochem. Soc. 105 Ž1958. 402–408. w2x V. Lehmann, H. Foll, Formation mechanism and properties of electrochemically etched trenches in n-type silicon, J. Electrochem. Soc. 137 Ž1990. 653–659. w3x H. Ohji, P.J. Trimp, P.J. French, Fabrication of free standing structure using single-step electrochemical etching in hydrofluoric acid, Sensors and Actuators, A 73 Ž1999. 95–100. w4x H. Ohji, S. Lahteenmaki, P.J. French, Macro porous silicon formation for micromachining, Proceedings SPIE, Micromachining and Microfabrication Process Technology III, Austin, TX, USA, September Ž1997. 189–197. w5x H. Ohji, P.T.J. Gennissen, P.J. French, K. Tsutsumi, Fabrication of accelerometer using single-step electrochemical etching for microstructures ŽSEMMS.,IEEE MEMS Conference ’99, Orlando, USA, 1999, pp. 61–65.
Biographies Hiroshi Ohji received his BSc and MSc in Precision Engineering from Osaka University, Japan in 1986 and 1988, respectively. In the same year, he joined Mitsubishi Electric, Japan. He moved to the Netherlands in 1996. For 2 years, he was a member of the Laboratory for Electronic Instrumentation, Delft University of Technology as a visiting scientist in the field of microfabrication technologies and their applications. He returned to Mitsubishi Electric in July 1998 and has been working on automotive sensors and wet etching technologies. Paddy French received his BSc in Mathematics and MSc in Electronics from Southampton University, UK, in 1981 and 1982, respectively. In 1986, he obtained his PhD, also from Southampton University, which was a study of the piezoresistive effect in polysilicon. After 18 months as a post-doctoral candidate at Delft University, the Netherlands, he moved to Japan in 1988. For 3 years, he worked on sensors for automotives at the Central Engineering Laboratories of Nissan Motor. He returned to Delft University in May 1991 and is now a staff member of the Laboratory for Electronic Instrumentation with interests in micromachining and process optimisation related to sensors. Shin-ichi Izuo received the BSc and MSc in Material Engineering from Kyoto University, Japan in 1996 and 1998, respectively. In the same year, he joined Mitsubishi Electric, Japan. He has been working on micromachining and wet etching technologies. Kazuhiko Tsutsumi received the BS Ž1976. and MS Ž1978. degrees in Material Engineering and PhD degree in Electrical Engineering Ž1986. from Osaka University, Japan. Since 1982, he has worked as Research Scientist at Mitsubishi Electric, Japan and he is now a group manager in Advanced Technology R&D Center. Research interests in his group include device physics, microsensors, micromachining and process optimisation related to sensors.