Capacitive Pressure Sensor by Silicon Micro-Machined

BILKENT UNIVERSITY MECHANICAL ENGINEERING DEPARTMENT

ME 657: Nano/Micro Manufacturing PROJECT FINAL REPORT Capacitive Pressure Sensor by Silicon Micro-Machined

Group Members: Ergün Ümitcan GÜVENIR

Submission Date:

21000154

25/05/2015

Süleyman Fatih KELEŞ

21002112

Abstract

In this report, capacitive micro pressure sensor’s design and fabrication methods are examined. For design step SolidWork are used and geometrical constraints are specified by literature source. The Sensor is working based on capacitance phenomena. Depend on this phenomena sensor is designed. Fabrication steps are separated by two part one of them is glass part and second part is SOI Wafer. Steps are briefly explain in fabrication part. In the discussion part challenging steps are defined.

1. Introduction Capacitive Pressure Sensor by Silicon Micro-Machined is selected as a project. MEMS is a new technology that gives opportunity designing and production of micro electro-mechanical devices. MEMS devices are very useful for this reason their usage area is increasing such as mechanical, optical, chemical sensors; projection displays, fiber switches, DNA amplification, medical diagnostics, material testing, lab-on-a-chip, micro robots, and many others. Compered to macro systems, MEMS have some advantages: They are small and their weight is low. In micro linear dimension, transient time is shorter. Therefore MEMS are more efficient systems. Because of the MEMS size they are less effected the environmental affect like temperature and humidity. Also, MEMS works on small size, expensive materials can be used for this small size. So, their performance can be improved. Although, the fabrication equipment are expensive compere to macro fabrication machines, fabrication costs are cheaper than macro fabrication. Pressure sensor is one of the most popular areas of MEMS technology. Firstly, Micro pressure sensors are used especially in automotive industry for airbag systems. Micro pressure sensors have extensive usage from inkjet cartridges to blood pressure testers. In pressure sensors, mechanical energy transformed from the pressure to a form of electrical signal, such as (I)current, (V)voltage and (C)capacitance. Methods of transduction that are based on basic physical laws, such as piezo-resistive, capacitive or resonant phenomena. Capacitive phenomena is chosen because they are more controllable characteristic, larger output range, more sensitive to pressure than the piezo-resistive ones, less temperature dependence. For fabrication, Bulk and surface micro machining are proposed. Bulk micro machining are used for mostly substrate machining. The substrate can be fabricated by etching, spark machining, mechanical, ultrasonic milling. Surface machining can be defined as additive process on the MEMs. Depositing and patterning are used. Structural and sacrificial materials

are used to fabricate MEMs. Proposed fabrication techniques which is bulk and surface micro machining can be used to fabricate micro pressure sensors. 2 Design and Analyze

A capacitive pressure is basically a capacitor which have deformable electrodes parallel to each other. When pressure effects on the capacitive pressure sensor, the electrodes bend and the capacitance of sensor change. Measurement of pressure change bases on the change in capacitance of the sensor. The capacitance of parallel plate capacitor is given by the following equation:

∁=

𝜖𝐴 𝑑

(1)

Where: 

ε is permittivity of the dielectric material between the electrodes.



A is area of the parallel plates of the capacitor.



d is the gap distance between parallel plates. The capacitance of the capacitive pressure sensor is proportional to gap distance

between plates. As mentioned above, the capacitive pressure sensor working principle bases on the change in the capacitance in other words change in gap distance between plates. In order to design a capacitive pressure sensor, a basic capacitor should be design. However there are differences between capacitive pressure sensor and parallel plate capacitor. The main difference is deformability. As mentioned, the sensor works by change in gap distance, therefore the electrodes of the sensor should be deformable. For this reason, the sensor should has diaphragm plate which bend to other plate when pressure has impact on it.

Before to proceed on the design, literature research about capacitive pressure sensor is made. During the research, studies about the capacitive pressure sensor are examined. Moreover, in order to decide a primary design of the sensor, design of other capacitive pressure sensor design on market are analyzed. After research, primary design of the capacitive pressure sensor is decided. Basically, it has two plates, one bottom substrate two wall substrate and support beam. Bottom substrate places base of the sensor as the name suggests. Two wall substrate are placed right and left side of the bottom substrate. The wall substrate are parallel to each other and their purpose is to hold the plate which is at top. At the midpoint of gap between two wall substrate, the support beam is placed in order to provide support surface to plate which is at middle. After all parts of the sensor, primary design is formed and in Figure 1 the primary design of the capacitive pressure sensor is shown.

c

b

b a

d

Figure 1 : Primary Design of the Sensor e

In the primary design of the capacitive pressure sensor: 

a is bottom substrate.



b is wall substrate.



c is diaphragm plate.



d is middle plate.



e is support beam.

After the primary design, the second step of the designing process is making first design based on the primary design. For designing process Solidworks® program is used. However, the main difficulty of the designing process is to decide the dimension of the design and the most important dimension is the gap distance between plates because all project is based on this distance. Moreover the diaphragm plate thickness is also important because sensitivity of the sensor is based on the diaphragm plate, its flexibility and its bending ability. There are some parameters about the dimension of the sensor. The parameters are reached during literature researching. Throughout designing process and deciding dimensions of the sensor’s parts the parameters are considered because the project is a conceptual design and the test of the sensor is not possible. As mentioned above, the important dimensions are the distance of gap between plate and the thickness of diaphragm plate therefore the dimension of them are decide according to constrains which is reached by literature research. The diaphragm’s thickness should be selected according to SOI wafer properties. The lower limit of the thickness is 3 micrometer and the upper limit of it is 20 micro meter. The planed limit of the designed sensor is using under maximum 5 MPa, for this reason the diaphragm of the sensor should resist 5 MPa pressure. According to this parameters the thickness of the diaphragm is selected as 15 micrometer. The other important dimension is gap distance. It also should be selected according to SOI wafer properties. With respect to constraint, the lower limit of the gap is 2 µm and the upper limit of the gap is 10 µm. As mentioned above, the planned maximum pressure is 5 MPa, for this reason, the distance of the gap between plates is selected as 8 µm. Then; according to these important dimensions, other parts of the sensor are drafted and dimensions are selected. Finally, the design of the capacitive pressure sensor is reached. Following Figure 2 show technical drawing of the sensor.

Figure 2: Technical Drawing of the Primary Design



Unit of all dimensions in the technical drawing is micrometer.

Next process after the design is structural analyzing of the sensor. Firstly, intended method for analyzing is using ANSYS Workbench 15 program. However, the computer which is used for analyzing cannot work properly for analyzing because CPU of the computer failed during analyzing. For this reason, more basic analyzing method should be selected, Solidworks

program has its own analyzing tools, although Solidworks Analyzing Tools is not recommended for structural analyzing, there is no another option for it.

Figure 3: Primary Design

Firstly, first design is tested under 3 MPa pressure. Pressure is arranged to perpendicular to diaphragm plate of the sensor. Furthermore, in order to fasten bottom and wall substrates, fixed support is arranged to surface of these substrates. The following Figure 4 shows stress distribution by structural analyze under 3 MPa pressure.

Figure 4: Von- Mises Stress Distribution under 3 MPa

The maximum Von-Mises stress is not higher than the yield stress of the SOI therefore, the design can work under 3 MPa in terms of physical strength. However, displacement of the design should be considered to decide whether it works or not. The following Figure 4 shows displacement of the design under 3 MPa. As considered most displacement occurs at middle of the diaphragm. However, under 3 MPa pressure, gap distance between plates is too small and almost plates contact each other and it is not acceptable because when 5 MPa is applied, they will be contact each other and it means that the sensor is failed. For this reason, in order to reach 5 MPa as maximum acceptable pressure for the sensor, the distance of the gap between. According to constraints which reached by literature research,

maximum distance for the gap is 10 micrometer. For the final design of the sensor the gap dimension is selected as 10 µm.

Figure 5: Displacement of the Design under 3 MPa

During the final design process, as mentioned above the gap dimension is changed and selected as far as possible, in other words 10 µm. On the other hand, when results of analyze of the first design are examined, it is decided that area of the plates can be increased because, as mentioned above the capacitance limit is depended with the area of the plate. If the area increase, the capacitance of the sensor also increase ant it means that sensitivity of the capacitive pressure sensor increase because by the large area, small changes in pressure of the circumstances can be detected. Then, the final design is reached. The following Figure 6 shows technical drawing of the final design.

Figure 6: Technical Drawing of the Final Design



Unit of all dimensions in the technical drawing is micrometer.

Like for the primary design, next step is structural analyzing. For the final design, in Solidworks Analyzing Tools, all parameters such as pressure, fixed support arranged same as first analyzing. The followings figures show the result of analyze of the final design under 3 MPa.

Figure 7: Final Design

Figure 8: Von- Mises Stress Distribution under 3 MPa for Final Design

Figure 10: Displacement of the Design under 3 MPa of Final Design

As it is expected, the final design works under 3 MPa. The displacement of the diaphragm is in the acceptable limit. Furthermore, the desired maximum pressure is 5 MPa, therefore the next step is that to apply 5 MPa to diaphragm perpendicularly. The following figures show the Von-Mises stress distribution and displacement of the final design under 5 MPa. When the result of analyze under 5 MPa, the maximum Von-Mises stress is not higher than yield stress of SOI. However, maximum displacement of the diaphragm plate is not in the acceptable limit. The maximum value of the displacement is almost 10 µm and it is gap distance of design and if the plate bends 10 µm the two plate contact each other and it means that sensor fail. Therefore the limit of the designed capacitive sensor is 3 MPa.

Figure 11: Von- Mises Stress Distribution under 5 MPa for Final Design

2. Fabrication After the glass wafer is cleaned in a wet bench, a photoresist for large pattern lithography is spin-coated and patterned. Sensor is planned to produce by two part. One of two is glass fabrication second one is SOI wafer. Fabrication steps drawn by using AutoCAD. Figure 12: Displacement of the Design under 5 MPa of Final Design

3. Fabrication

After the glass wafer is cleaned in a wet bench, a photoresist for large pattern lithography is spin-coated and patterned. Sensor is planned to produce by two part. One of two is glass fabrication second one is SOI wafer. Fabrication steps drawn by using AutoCAD.

Spin coat and pattern PR

Glass etching

Photoresist removing

Spin coat and pattern Photoresist

Glass etching

Photoresist removing

Pattern Photoresist

Sputter Cr/Au

Photoresist lift off

Fabrication of SOI Wafer

LPCVD silicon Nitride

Spin coat and pattern Photoresist

Metallization

PECVD oxide

Photoresist patterning

oxide RIE etching

PECVD polysilicon

PR patterning

polysilicon plasma etching

sacrificial etching By anodic bonding two part of sensor are bonded.

3. Conclusion and Discussion

First problematic part is metallization, we have not enough information about this process. However, we need to metallization because of the making electrode. As learned from the lectures, we tried to design and propose the fabrication step. All these things are conceptual, actually we do not know efficiency of our design sensor. We did some simulations on the SolidWorks but these simulations involve only mechanical properties. Electrical performance of the proposed sensor are not calculated or tested. For sputtering, cooling is required. Also sputtering can damage the substrate. PECVD require high cost equipment, and can damage the substrate. LPCVD causes high temperature. In spin coating edge bead can be happen. But it can be solved by remover solvents. During analyzes two assumptions are made in order to reached the result. First one is about design. Actually sensor should be design in Solidworks part by part and by assembly tools of Solidworks, the parts matched each other. However, during design process the sensor is drafted

like monolithic because in the structural analyze, if assembly is used, result cannot be reached because of the computer fail. Second assumption is the determined the forces, between two plates there should be electrostatic force but as first assumption when electrostatic force is defined between two plate, analyze fail because of CPU fail. Instead of the electrostatic force, two fixed force perpendicular to plates in opposite direction are defined. Magnitudes of the force is selected as 100 N and it is an assumption.