MEMS pressure sensor using area-change capacitive technique

This invention relates a pressure sensor, and more particular to a micro-electro-mechanical system (MEMS) pressure sensor.

U.S. Pat. No. 7,187,185 discloses force sensors that use area-change capacitive sensing techniques to determine the magnitude of the force. These sensors are made using conventional machining processes.

In one device, a first vertical electrode is mounted on a top layer and a second vertical electrode is mounted on a bottom layer. The top and the bottom layers are separated by an inert gas. When a force is asserted, it changes the overlap area and therefore the capacitance between the vertical electrodes. The change in the capacitance can be correlated to the magnitude of the force applied to the device.

In another device, an inner conductive surface in the form of a metal tube is inserted into a middle conductive surface in the form of a hollow tube. A spacer may hold the inner conductive surface inside the middle conductive surface so they are separated by a gap. Like the earlier device, when a force is asserted, it changes the overlap area and therefore the capacitance between the conductive surfaces. The change in the capacitance can be correlated to the magnitude of the force applied to the device.

In one embodiment of the invention, a micro-electro-mechanical system (MEMS) pressure sensor includes a silicon spacer, a silicon membrane layer mounted above the spacer, and a silicon sensor layer mounted above the silicon membrane layer. The silicon spacer defines an opening to the silicon membrane layer. A first portion of the silicon membrane layer opposite of the spacer opening forms a diaphragm while a second portion of the silicon membrane layer opposite of the spacer forms a stationary perimeter around the diaphragm.

The silicon sensor layer includes a movable electrode and a stationary electrode. The electrodes are separated by a conventional semiconductor manufacturing technique (e.g., etching) to create a substantially constant gap between the electrodes. The movable electrode is located above the diaphragm while the stationary electrode is located above the stationary perimeter.

The movable electrode and the diaphragm move in response to a pressure applied to the diaphragm. The vertical displacement of the movable electrode changes the overlap area between sidewall surfaces of the electrodes but not the gap between the electrodes. This in turn changes the capacitance between the electrodes by an amount linearly proportional to the change in the pressure.

In the above structure, the movable and the stationary electrodes are formed on the same membrane layer using conventional semiconductor manufacturing processes. This allows great precision to be achieved in the alignment of the electrodes, especially in controlling the gap between the electrodes, over conventional machining processes. The result is a highly accurate pressure sensor created in an efficient and cost-effective manner.

FIG. 1 illustrates a side view of a MEMS pressure sensor in one embodiment of the invention.

FIG. 2 illustrates a top view of the MEMS pressure sensor of FIG. 1 in one embodiment of the invention.

FIG. 3 illustrates a side view of the MEMS pressure sensor of FIG. 1 under pressure in one embodiment of the invention.

FIGS. 4A , 4 B, 4 C, and 4 D illustrate a process for making the MEMS pressure sensor of FIG. 1 in one embodiment of the invention.

FIG. 5 illustrates a top view of the MEME pressure sensor of FIG. 1 in another embodiment of the invention.

FIG. 6 illustrates a side view of a MEMS pressure sensor in one embodiment of the invention.

Use of the same reference numbers in different figures indicates similar or identical elements.

FIGS. 1 and 2 illustrate a MEMS pressure sensor 100 in one embodiment of the invention. Sensor 100 is formed using conventional semiconductor manufacturing processes.

Sensor 100 includes a silicon spacer 102 that defines an opening 104 . A silicon membrane layer 406 is mounted above spacer 102 . A portion of membrane layer 406 exposed by opening 104 forms a diaphragm 108 and a portion mounted above spacer 102 forms a stationary perimeter 110 around the diaphragm. In one embodiment, opening 104 is a round opening and diaphragm 108 is therefore round.

A first oxide layer 404 has a portion 114 between spacer 102 and stationary perimeter 110 . As will be describe later, oxide layer 404 acts as an etch stop in the process for making opening 104 in spacer 102 .

A silicon sensor layer 410 is mounted above membrane layer 406 . Sensor layer 410 includes a movable electrode 118 and a stationary electrode 120 around the movable electrode. Movable electrode 118 is located substantially above diaphragm 108 while stationary electrode 120 is located substantially above stationary perimeter 110 . Movable electrode 118 and stationary electrode 120 are separated by a gap 122 .

A second oxide layer 408 has a portion 126 between movable electrode 118 and diaphragm 108 , and a portion 128 between stationary electrode 120 and stationary perimeter 110 . Oxide layer 408 acts as an insulating layer to prevent electrodes 118 and 120 from being shorted.

FIG. 2 illustrates one embodiment where movable electrode 118 is a disk and stationary electrode 120 has a round inner perimeter that matches the shape of the movable electrode. As a result of oxide etch and release of movable electrode 118 and stationary electrode 120 from oxide layer 408 , holes 202 (only two are labeled for clarity) are formed on the outer circumference of movable electrode 118 and the inner circumference of stationary electrode 120 . Note that the phantom lines shown indicate the perimeter of oxide portions 126 and 128 from oxide layer 408 ( FIG. 1 ).

Referring back to FIG. 1 , the capacitance between the electrodes can be sensed by a capacitance sensing circuit 132 when a voltage is supplied to electrodes 118 and 120 . The value of the capacitance depends on the overlap area and the gap between the sidewalls of electrodes 118 and 120 as follows:

C=ε 0 A/d,

where C is the capacitance, ε 0 is the permittivity of free space, A is the overlap area between the electrodes, and d is the gap distance.

Referring to FIG. 3 , a pressure difference between the opposing sides of diaphragm 108 (simply illustrated as pressure 130 asserted on the diaphragm) causes the diaphragm and movable electrode 118 to translate. As the pressure is evenly distributed across diaphragm 108 , the translation of the diaphragm and movable electrode 118 are substantially vertical so gap 122 between the movable electrode and stationary electrode 120 remain substantially constant.

This vertically translation of movable electrode 118 changes the overlap area (symbolically illustrated by a distance 302 ) between sidewalls of movable electrode 118 and stationary electrode 120 . The change in the overlap area changes the capacitance between electrodes 118 and 120 . The change in the capacitance is linearly proportional to the change in pressure 130 . The exact relationship can be mathematically or empirically determined to find the change in pressure 130 .

FIGS. 4A , 4 B, 4 C, and 4 D illustrate a method for making sensor 100 using convention semiconductor manufacturing processes in one embodiment of the invention. In the first step illustrated in FIG. 4A , a double silicon-on-insulator substrate 400 is provided. Substrate 400 includes a first silicon layer 402 , a first oxide layer 404 above silicon layer 402 , a second silicon layer 406 above oxide layer 404 , a second oxide layer 408 above silicon layer 406 , and a third silicon 410 above oxide layer 408 . As used here, a silicon layer refers to any material with silicon, including single crystal silicon, polysilicon, and silicon carbide. Silicon layer 406 can be selected to be stiff or soft depending on the sensitivity to pressure desired for sensor 100 .

In the second step illustrated in FIG. 4B , silicon layer 402 is etched down to oxide layer 404 to form spacer 102 with opening 104 . In this step, oxide layer 404 acts as an etch stop.

In the third step illustrated in FIG. 4C , silicon layer 410 is etched to form electrodes 118 and 120 . Also in this step, holes 202 ( FIG. 2 ) are etched in the outer perimeter of electrode 118 and the inner perimeter of 120 in preparation for the oxide etch and release in the next step.

In the fourth step illustrated in FIG. 4D , an oxide etch is performed to remove portions of oxide layer 404 and 408 exposed by silicon layers 402 and 410 . Specifically, this releases electrode 118 and diaphragm 108 from the remainder of the structure so they can translate vertically under a change in pressure.

FIG. 5 illustrates a variation of MEMS pressure sensor 100 (hereafter “sensor 500 ”) in one embodiment of the invention. Sensor 500 is essentially the same as sensor 100 except that a movable electrode 118 A now has spokes 501 with teeth 502 that are interdigitated with teeth 504 (only one of each is labeled for clarity) from a stationary electrode 120 B. Teeth 502 and 504 increase the overlap area between electrodes 118 A and 120 A at any vertical displacement in order to make sensor 500 more sensitive to pressure changes. In addition, movable electrode 118 has spokes 505 that are coupled by springs 506 to stationary pads 510 (only one of each is labeled for clarity) insulated from stationary electrode 120 A. This allows wire bonds to be formed two or more stationary pads 510 for coupling to voltage sources. While interdigitated teeth are shown, other complementary features may be used to increase the overlap area between electrodes 118 A and 120 A.

FIG. 6 illustrates a variation of MEMS pressure sensor 100 (hereafter “sensor 600 ”) in one embodiment of the invention. Sensor 600 is essentially the same as sensor 100 except a channel 602 on the top surface of membrane layer 406 is formed below gap 122 . Channel 602 allows membrane layer 406 to be more flexible so it can translate under smaller pressure differences. In the fabrication of sensor 600 , spacer 102 , oxide layer 404 , and membrane layer 406 can be made from a single SOI substrate. Channel 602 is formed by etching the top surface of membrane layer 406 , and then forming oxide layer 408 and bonding sensor layer 410 above of oxide layer 408 .

Various other adaptations and combinations of features of the embodiments disclosed are within the scope of the invention. Instead of using single or double SOI substrates to form the sensors, multiple silicon substrate can be bonded together to form the sensors where the oxide layer is grown or deposited on the substrates. Furthermore, although the movable electrode is described as being enclosed by the stationary electrode, the design can be modified where the movable electrode encloses the stationary electrode. Numerous embodiments are encompassed by the following claims.