MEMS Pirani Sensors for Plasma Sterilization Applications

Low pressure plasma sterilization has been developed as secure sterilization

methods and has already been put to practical use for medical equipment. Although the mechanisms are not yet fully understood, the following cellular inactivation mechanism has been reported: ultraviolet radiation in the far-ultraviolet and mid-ultraviolet range is capable of damaging nucleic acids (DNA, RNA) and proteins.

In order to produce such ultraviolet radiation the pressure of the gas mixture used

as plasma source should be well controlled in the range of 13 to 130 Pa.

We have developed Pirani sensors which can be used to control the pressure of

the gas mixture. The amplified response of a Pirani sensor to pressure variation is shown in the following figure. It can be seen from the figure that the controlled pressure locates the sensitive range of the Pirani sensor.

Medical Liquid Thermal Flow Sensors

Recently, our thermal flow sensors have been developed

for medical liquid applications. These new applications require low noise, high reproducibility, short response time, battery powered, potable and wearable, and low-cost disposable. Our sensors meet all these requirements because they are fabricated using a CMOS comparable technology and are based on a structure integrating a heater and a thermopile in a single chip.

   The following figure shows a typical characteristic line of 1% NaCl in DI water (stroke-physiological saline solution) flow measured by a liquid thermal flow sensor.

Disposable Water Flow Sensors

    POSIFA Microsystems now provide disposable water thermal flow sensors with outstanding performance.  

     Medical instrumentation has recently been moved outside the hospital and clinic into the home to help keep health care costs down whilst providing a better level of care. Sensors are an important part of the new technology for allowing this to happen.

     At the present, POSIFA’s thermal flow sensors are mainly used for air flow determination in human respiration. However an increasing demand is observed for liquid measurements in smallest flow ranges for medical applications. 

     Compared with traditional flow sensors, the thermal flow sensors have the advantages of highly precise measurement and high repeatability, short response time, low power consumption, smallest dimensions, adhesive-free packaging and low-cost. 

     On the market there are only micro pump systems available that work in a controlled but not in a regulated mode. This means that an extrinsic disturbance will lead to a falsification of the flow rate. It is not feasible that those systems run under disturbance free conditions. Therefore for small, cheap and especially regulated measurement our thermal flow sensor is a satisfying solution.

A Perfume Bottle with a Silicon Based MEMS Vaporizer

     As the vaporizers developed by POSIFA can be combined to comprise a total olfactory system, a variety of olfactory displays can be realized. Compared with several types of developed olfactory display, our system is much light and small. Since the developed olfactory display comprises of mass flow controllers, inkjet devices, solenoid valves, and fans, it is inevitable to form a larger instrument.  Our system is portable and wearable it can be shaped like a portable perfume bottle.

Non-invasive blood glucose monitor 

    A micro-optical-mechanical-electro-system (MOMES)-based non-invasive blood glucose monitor comprises a micromachined infrared optical filter array, a micromachined infrared mechanical modulator array, at least one micromachined infrared tunable filter, and at least one infrared detector. Each optical filter is aligned with a mechanical modulator along its optical axis direction. The optical filter continuously divides a monochromatic infrared light in a wavelength range within 0.8 to 25 micron from an infrared light. The aligned mechanical modulator turns the monochromatic infrared light into an alternating monochromatic infrared light. The tunable filter is aligned with the infrared detector along its optical axis direction. The tunable filter selects the back-diffused alternating monochromatic infrared light emitted from a measured blood subject that is illuminated by the alternating monochromatic infrared light. The infrared detector converts the back-diffused alternating monochromatic infrared light into an alternating electronic signal. Then a photo-integrated circuit (IC) combines with the infrared detector for synchronous detection and amplification of the electronic signal generated by the synchronous detection.

---

w.

Wearable Jewelry with Olfactory Display Based on MEMS Vaporizer

      A MEMS vaporizer based olfactory display can be shaped like a wearable jewelry accessory such as pendant, waist, arm ornament, and so on. The MEMS vaporizer is developing by POSIFA Microsystems using our patented MEMS technologies. MEMS are made up of components between 1 to 100 micrometres in size (i.e. 0.001 to 0.1 mm), and MEMS devices generally range in size from 20 micrometres (20 millionths of a metre) to a millimetre (i.e. 0.02 to 1.0 mm). The following figure shows a pendant with a silicon based MEMS vaporizer.

 

Smart Phone with MEMS Vaporizer

     We are developing MEMS vaporizers based on our patented technology. These devices are portable and wearable and can be attached to a smart phone or other portable electronic devices. Therefore an internet carrying an extensive range of olfactory information will be reality in the near future.

Micromachined vertical vibrating gyroscope 

Abstract A micromachined vertical vibrating gyroscope consists of three single crystal silicon assemblies: an outer single crystal silicon assembly, an intermediate single crystal silicon assembly, and an inner single crystal silicon assembly. The outer assembly includes a plurality of arc-shaped anchors arranged in a circle and extending from a single crystal silicon substrate coated with an insulating annulus thereon. The intermediate assembly is a suspended wheel concentric with the arc-shaped anchors. The inner assembly is a suspended hub concentric with the circle formed by the anchors and having no axle at its center. The three assemblies are connected to each other through several flexures. The intermediate suspended wheel is driven into rotational vibration by lateral comb capacitors. Input angular rates are measured by two vertical capacitors. The gyroscope is fabricated utilizing a bipolar-compatible process comprising steps of buried layer diffusion, selective epitaxial growth and lateral overgrowth, deep reactive ion etching, and porous silicon processing.

---

Claims

What is claimed is:

1. A method for fabricating a micromachined vertical vibrating gyroscope comprising steps of a) providing a lightly doped single crystal silicon substrate; b) forming a heavily doped buried layer in the substrate; c) forming an insulating ring on the surface of the substrate; d) performing selective epitaxial growth to form a vertical epitaxial single crystal silicon layer on the exposed silicon surface of the substrate and a lateral overgrowth single crystal silicon layer on the insulating ring; e) conducting first deep reactive ion etching to form first trenches in the vertical epitaxial layer; f) filling up the first trenches with an insulating material(s); g) conducting second deep reactive ion etching to form second trenches in the vertical epitaxial layer; h) performing anodization to convert the buried layer into a porous silicon layer; i) removing a portion of the porous silicon layer to form suspension structures; j) conducting pre-deposition with a diffusion source; k) conducting post-diffusion in oxygen atmosphere to form a heavily doped single crystal silicon layer on the surfaces of the suspension structures and turn the rest of the porous silicon layer into an oxidized porous silicon layer; l) performing low pressure chemical vapor deposition to form an insulating layer on the surfaces of the suspension structures; and m) depositing a metal layer on the top surfaces and sidewalls of the suspension structures.

2. The method for fabricating a micromachined vertical vibrating gyroscope of claim 1, wherein said insulating ring comprises silicon dioxide and silicon nitride.

3. The method for fabricating a micromachined vertical vibrating gyroscope of claim 1, wherein said insulating ring has a width ranging from 8 to 60 micron.

4. The method for fabricating a micromachined vertical vibrating gyroscope of claim 1, wherein said vertical epitaxial layer has a thickness ranging from 5 to 40 micron.

5. The method for fabricating a micromachined vertical vibrating gyroscope of claim 1, wherein said first trenches have a width ranging from 1 to 3 micron.

6. The method for fabricating a micromachined vertical vibrating gyroscope of claim 1, wherein said second trenches have a width ranging from 1 to 3 micron.

7. The method for fabricating a micromachined vertical vibrating gyroscope of claim 1, wherein said diffusion source is POCl.sub.3.

8. The method for fabricating a micromachined vibrating gyroscope of claim 1, wherein said post-diffusion is carried out in oxygen atmosphere at a temperature ranging from 900 to 1000.degree. C.

---

Description

---

FIELD OF THE INVENTION

The present invention relates to a micromachined gyroscope and more particularly, to a micromachined gyroscope performing lateral sensing of angular rates, having both driving and sensing oscillation modes, and having all components formed from a single silicon crystal of a single silicon wafer.

BACKGROUND OF THE INVENTION

Gyroscopes are used to measure the angular deviation of a guided missile from its desired flight trajectory; to determine the heading of a vehicle for steering; to determine the heading of an automobile as it turns; to indicate the heading and orientation of an airplane during and after a series of maneuvers; or to stabilize and point radar dishes and satellites. Recently, micromachined gyroscopes are receiving increasing attention because of their low cost, small size and high sensitivity. Micromachined vertical or z-axis gyroscopes are used to counteract the rolling effect on a vehicle, and thus, are a preferred stabilization tool for vehicles such as airplanes, ships, and cars.

One developed micromachined vertical gyroscope utilizes a rapidly spinning, heavy mass. These spinning mass gyroscopes require lubrication and eventually wear out.

Another developed micromachined vertical gyroscope is based on vibration mode, but uses polysilicon technology. All suspension structures of this gyroscope are made of a polysilicon layer. To release the suspension structures a thick sacrificial layer is applied beneath the polysilicon layer. After the polysilicon layer is etched through, the sacrificial layer is removed. There are several problems with this gyroscope.

Although a single crystal silicon wafer may be used as a substrate of the gyroscope, the single crystal silicon wafer with several additional layers thereon is no longer suitable for standard microelectronics processing to realize monolithic integration.

An as-deposited polysilicon layer is in compressive strain. Suspension structures formed from the strain polysilicon layer tend to buckle, causing gyroscope instability or inability to work.

Since the surface of a relatively thick polysilicon layer is quite rough an additional polishing step is required before a planar processing process.

It is impossible to deposit a relatively thick polysilicon layer with device quality. This limits the stiffness of the suspension flexures in the vertical direction and makes electrostatic comb drive levitation more difficult to control.

SUMMARY OF THE INVENTION

Accordingly, it is an object of the present invention to provide a micromachined vertical vibrating gyroscope with all suspension structures being of single crystal silicon. The use of single crystal silicon structure eliminates the problems of the prior art caused by utilizing polysilicon as a building material.

It is a further object of the present invention to provide a micromachined vertical vibrating gyroscope with all suspension structures recessed into a processed silicon wafer. These recessed structures are more robust than any structures being out of the plane of the processed silicon wafer.

It is a still further object of the present invention to provide a micromachined vertical vibrating gyroscope capable of being produced by processing a single silicon wafer. Since no other wafer is required, the process is simple and allows batch production.

It is a still further object of the present invention to provide a micromachined vertical vibrating gyroscope capable of being fabricated on a single processing plane all along. This facilitates the use of standard planar processing technologies for integrated circuits.

A still further object of the present invention to provide a micromachined vertical vibrating gyroscope capable of being electronically integrated with other electronics similarly fabricated on the same chip so as to realize monolithic integration.

In accordance with these and other objects, a micromachined vertical vibrating gyroscope is described. The gyroscope consists of three single crystal silicon assemblies: an outer single crystal silicon assembly, an intermediate single crystal silicon assembly, and an inner single crystal silicon assembly. The outer assembly includes a plurality of arc-shaped anchors arranged in a circle and extending from a single crystal silicon substrate coated with an insulating annulus thereon. Each of at least four anchors support a suspension flexure and two suspension fan-shaped stops on its inner edge. The intermediate assembly is a suspension wheel possessing a same center with the circle and having a plurality of combs protruding from its outer edge. The linkage between the outer assembly and the intermediate assembly is realized through the four suspension flexures arranged along two orthogonal axes. The inner assembly is a suspension hub possessing a same center with the circle and no axle at its center. The linkage between the intermediate assembly and the inner assembly is realized through other two flexures arranged along a same axis. The intermediate suspension wheel is driven into rotational vibration by lateral comb capacitors. Each lateral comb capacitor is formed by a combination of a comb protruding from an anchor and a comb protruding from the intermediate wheel. Input angular rates are measured by two vertical capacitors that are formed between the bottom of the inner suspension hub and the interior top surface of the substrate used to support the anchors.

The micromachined vertical vibrating gyroscope is fabricated utilizing a bipolar-compatible process. This process comprises the steps of forming a buried layer, depositing and patterning an insulation layer, and growing an epitaxial layer. The epitaxial layer grown on the buried layer is of single crystal silicon and on the insulation layer is of lateral overgrown single crystal silicon. After converting the buried layer into a porous silicon layer all suspension structures are formed by partially etching into the epitaxial layer and removing the beneath porous silicon layer. The rest of the porous silicon layer is turned into an oxidized porous silicon layer for electrically isolating the anchors located thereon.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 is a partially cut away perspective view of a micromachined vertical vibrating gyroscope in accordance with the present invention;

FIG. 2 is a cross-sectional view of a buried layer formed in a single crystal silicon substrate for fabricating the vertical vibrating gyroscope of FIG. 1 in accordance with the present invention;

FIG. 3 is a cross-sectional view of an epitaxial single crystal silicon layer formed on the silicon exposed regions of the single crystal silicon substrate and a lateral overgrowth single crystal layer formed on an insulating ring of the single crystal silicon substrate for fabricating the vertical vibrating gyroscope of the FIG. 1 in accordance with the present invention;

FIGS. 4A and 4B are cross-sectional views showing an empty trench formed in the epitaxial single crystal silicon layer and showing an etch depth indication cavity formed in the lateral overgrowth single crystal silicon layer for fabricating the vertical vibrating gyroscope of FIG. 1 in accordance with the present invention;

FIG. 5 is a cross-sectional view of an insulating material filled trench formed in the epitaxial single crystal silicon layer for fabricating the vertical vibrating gyroscope of FIG. 1 in accordance with the present invention;

FIG. 6 is a cross-sectional view of a plurality of empty trenches formed in the epitaxial single crystal silicon layer for fabricating the vertical vibrating gyroscope of FIG. 1 in accordance with the present invention;

FIG. 7 is a cross-sectional view of a porous silicon layer replacing the buried layer for fabricating the vertical vibrating gyroscope of FIG. 1 in accordance with the present invention;

FIG. 8 is a cross-sectional view of a plurality of suspension structures formed by partially removing the porous silicon layer and an oxidized porous silicon layer replacing the rest of the porous silicon layer for fabricating the vertical vibrating gyroscope of FIG. 1 in accordance with the present invention;

FIG. 9 is a cross-sectional view of a heavily doped diffusion layer formed on the surfaces of the suspension structures for fabricating the vertical vibrating gyroscope of FIG. 1 in accordance with the present invention;

FIG. 10 is a cross-sectional view of an insulating layer coated on the diffusion layer for fabricating the vertical vibrating gyroscope of FIG. 1 in accordance with the present invention; and

FIG. 11 is a cross-sectional view of a metal layer deposited on the sidewalls of the suspension structures for fabricating the vertical vibrating gyroscope of FIG. 1 in accordance with the present invention.

DETAILED DESCRIPTION OF THE INVENTION

Referring now to FIG. 1, a micromachined vertical vibrating gyroscope 10 includes a single crystal silicon substrate 11 with an epitaxial layer thereon. The epitaxial layer includes three regions: an outer periphery region, an intermediate ring region 18, and an inner circle or hub region 23. The outer periphery region is originally grown from the surface of the substrate, but later the beneath layer turns into an oxidized porous silicon layer 15; an outer ring region 14 is extended from beneath insulating ring 13 that is coated on the surface of the substrate 11. The inner circle region (hub 23) also is originally grown from the surface of the substrate 11, but later the beneath layer is removed to leave an air gap 24 therein. The whole epitaxial layer including the intermediate ring region 18 is single crystal silicon with same crystalline properties as the substrate 11. Also shown are a plurality of arc-shaped anchors 12 that are arranged in a circle and each having several through holes 16 therein, an intermediate vibrating wheel 18 that has a same center with the circle formed by anchors 12, and an inner vibrating hub 23 that also has the same center with the circle formed by anchors 12. The arc-shaped anchors 12 form a circle that is concentric with the circles formed by the intermediate vibrating wheel and hub 23.

Each of the arc-anchors 12 comprises of an outer portion and an inner portion. The outer portion is made of the outer periphery region of the epitaxial layer that is disposed on the oxidized porous silicon layer 15. The inner portion is made of the ring region of the epitaxial layer that is disposed on the insulating ring 13.

There are at least four anchors 12, each supporting a suspended single-crystal silicon flexure 20 and two suspended and perforated fan-shaped single crystal silicon mechanical stops 22 that sandwich the flexure 20. Each of the rest anchors supports a suspended single crystal silicon comb 17 with a plurality of side fingers extending from its one side.

The flexures 20 are arranged to align with the coordinate x-axis and y-axis, respectively, which are set on the substrate plane. The aspect ratio of the flexures 20 is relatively high so that they are stiff in response to the vertical motion and flexible in response to the lateral motion. With these features the wheel 18 is easy to stimulate to rotate about the vertical axis or the coordinate z-axis, but not to move along the coordinate x-axis and y-axis, and rotate about the coordinate x-axis and y-axis.

The wheel 18 is made of the inner circle region of the epitaxial layer that is separated from the substrate 11 by the air gap 24. Along the central line there are two insulating trenches 19 to divide the wheel 18 into two area-equaled half-wheels. The four flexures 20 join the wheel 18 and the anchors 12 so that the anchors 12 indirectly support the wheel 18. There are at least eight suspended single crystal silicon combs 21 extending from the outer edge of the wheel 18 so that each two of which sandwich an adjacent two comb 17. The combs 21 also have a plurality of side fingers extending from their one side. Each comb 21 combines with a comb 17 so that their side fingers are interdigitated.

The hub 23 is also made of the inner circle region of the epitaxial layer that is separated from the substrate 11 by the air gap 24. Along a zigzag line there are insulating trenches 25 to divide the hub 23 into two area-equaled half-hubs. A plurality of damping holes 26 are scattered in the hub 23. There are also two suspended single crystal silicon flexures 27 that join the wheel 18 and the hub 23 so that the anchors 12 also indirectly support the hub 23.

The flexures 27 are arranged along the extending direction of the central symmetric line of the hub 23. One flexure 27 electrically connects a half-hub to two outer bonding pads 34 and 35. The other flexure 27 (not shown in the figure) electrically connects the other half-hub to other two bonding pads (not shown in the figure). The bonding pads are disposed on the anchors 12 that are electrically isolated from the substrate 11 by the insulating ring 13 and the oxidized porous silicon layer 15. The geometric shape of the flexures 27 is designed to be stiff in response to the vertical motion of the hub 23 and flexible in response to the rotation of the hub 23 about their longitudinal direction.

The micromachined vertical vibrating gyroscope 10 further comprises at least four lateral driving capacitors 28 and four lateral monitoring capacitors 29. The lateral drive capacitors 28 and the lateral monitoring capacitors 29 are located in four symmetric fan-shaped regions. Each fan-shaped region contains a lateral driving capacitor and a lateral monitoring capacitor. Each lateral capacitor for driving or monitoring of the vibration of the wheel 18 consists of two combs. One comb extends from an anchor 12 and the other comb extends from the wheel 18. The fingers of the two combs are interdigitated with each other so that each two opposite fingers are faced sidewall to sidewall.

The micromachined vertical vibrating gyroscope 10 further comprises two vertical measuring capacitors 32 and 33. The vertical measuring capacitors are formed from the electrodes attached to the bottom of the hub 23 and to the inner top surface of the substrate 11 respectively. The hub 23 is electrically divided into two area-equaled half-hubs by the insulating trench 25. Each half-hub has an independent electrode attached thereon and electrically connects to two shown outer bonding pads 34 and 35 or other two un-shown outer bonding pads through the shown flexure 27 or the other un-shown flexure.

The electrodes of the vertical measuring capacitors 32 and 33 are formed from a diffusion layer disposed on the bottom surface of the wheel 23 and the top surface of the substrate 11. Different from this, the electrodes of the lateral capacitors 28 and 29 are formed from a metal layer deposited on the sidewalls of the figures of the combs 17 and 21. An insulating layer separates these two kinds of the electrodes (not shown in FIG. 1). Even though both the electrodes of the vertical capacitors 32 and 33 and the lateral capacitors 28 and 29 have common supporting silicon structures they are electrically isolated by this insulating layer.

Under the inner portion of each anchor 12 there is a cavity with the insulating ring 13 on its top. It can be seen that the cavity has two functions. The first function is to prevent the two diffusion layers of the electrodes of the vertical capacitors 32 and 33 from joining together at this point. The second function is to stop the metal layer of the electrodes of the lateral capacitors 28 and 29 to continuously extend to the interior top surface of the substrate 11. Because of this, the lateral capacitors 28 and 29 and vertical capacitors 32 and 33 can be electrically isolated each other.

In operation of the micromachined vertical vibrating gyroscope 10 a voltage is applied to the lateral driving capacitors 28. The intermediate wheel 18 is then stimulated into rotational vibration about the coordinate z-axis that is set to be vertical to the substrate plane. For the rotational vibration of the wheel 18, the flexures 20 provide flexible mechanical support. As the rotational angular becomes too large the stops 22 begin to abate the vibration so as to prevent the flexures 20 from damaging. The lateral monitoring capacitors 29 is used to measure the frequency and amplitude of the rotational vibration of the wheel 18. When the substrate 11 experiences an angular rate about the coordinate x-axis that is set to be perpendicular to the flexures 27 a Coriolis force is induced. The Coriolis force exerts on the inner vibrating hub 23 and causes the hub 23 to be rotationally vibrated about the coordinate y-axis.

In the balance state the two vertical capacitors 32 and 33 are designed to be completely equal. When the hub 23 rotates about the coordinate y-axis, the two vertical capacitors 32 and 33 are no longer equal. If the hub 23 rotates counterclockwise, the capacitance of the vertical capacitor 33 will increase and the capacitance of the vertical capacitor 32 will decrease. If the rotation direction reverses, the difference of the capacitance also reverses. Since the difference of the capacitance of the two vertical capacitors 32 and 33 depends upon the input angular rate, the input angular rate can be determined by measuring the difference of the capacitance of the two vertical capacitors 32 and 33.

The measurement circuit can be adopted in open loop or in close loop. In open loop the amplitude of a carrier signal can be modulated by the difference of the capacitance of the two vertical capacitors 32 and 33. After demodulation with the carrier frequency and the driving signal frequency a DC voltage proportional to the input angular rate can be yielded as the output of the measurement circuit. In close loop the yielded signal is first fed to a rebalance circuit. The rebalance circuit then provides a rebalance voltage applying to the vertical capacitors 32 and 33 to null the rotation of the inner vibrating hub 23 about the coordinate y-axis. The rebalance voltage is proportional to the input angular rate.

The micromachined vertical vibrating gyroscope 10 is fabricated, in accordance with the present invention, utilizing a bipolar compatible process. In this process surface micromachining can be carried out as a number of post-processing steps after completion of a standard bipolar process. The bipolar compatible process, which is diagrammatically illustrated in FIGS. 2-11, begins with a single crystal silicon wafer used as a substrate 101, as shown in FIG. 2. The substrate 101 is an (100) n-type silicon wafer with a resistivity ranging from 1 to 10 ohm-cm, typically being 3 ohm-cm. A 1 micron-thick silicon dioxide layer 102 is thermally grown on the surface of the substrate 101. Photolithography is used to pattern the silicon dioxide layer 102, and then a heavily doped buried n-type or p-type layer 103 is formed in the substrate 101 by thermal diffusion or a combination of ion implantation and thermal annealing. The buried layer 103 has a sheet resistance ranging from 4 to 20 ohm/square, typically being 8 ohm/square, and thickness ranging from 1 to 5 micron, typically being 2 micron.

As shown in FIG. 3, the next step is to perform a selective epitaxial growth process. Before doing that, a composite insulating ring 104 is formed on the surface of the substrate 101. The composite insulation layer is preferably comprised of silicon dioxide and silicon nitride. The silicon dioxide layer is formed by thermal oxidization and has a thickness ranging from 500 to 2000 angstrom, typically being 1000 angstrom. The silicon nitride layer is formed by low-pressure chemical vapor deposition (LPCVD) and has a thickness ranging from 400 to 1500 angstrom, typically being 1000 angstrom. Then, photolithography is conducted to create the composite insulation ring 104. The width of the insulating ring 104 ranges from 8 to 60 micron, typically is 14 micron.

Following patterning, the processed substrate is placed into an inductively heated reduced-pressure CVD pancake-type reactor. The reactor temperature is ramped to 970.degree. C. in an H.sub.2 ambient and a 5 min H.sub.2 bake is performed to remove native oxide from the bottoms of the seed holes. After H.sub.2 bake, 1.5 l/m of HCl is added to the ambient and a 30-s etching is performed. SiH.sub.2 Cl.sub.2 (DCS) is then added to induce selective epitaxial growth of single crystal silicon. Reactor pressure during growth is 40 Torr. Flow rates for SiH.sub.2 Cl.sub.2, HCl, and H.sub.2 are 0.22, 0.66, and 60 l/m, respectively. The epitaxial growth proceeds not only in the vertical direction, but also in the lateral direction. Therefore, a vertical epitaxial layer 105 is grown to cover the buried layer 103 and a lateral overgrowth layer 106 is grown to cover the composite insulation ring 104. Since the lateral epitaxial growth proceeds from the two opposite sides of the insulating ring 104 the formed lateral overgrowth layer meets together at the central line of the insulating ring 104 and results in a shallow trench therein. The resistivity of the epitaxial layer 105 may be varied between 1 and 10 ohm-cm, typically is 3 ohm-cm, and a thickness may be varied between 5 and 40 micron, typically is 10 micron.

Turning now to FIGS. 4A and 4B, these figures illustrate the fabrication process of the insulating trenches of FIG. 1. The trenches 108 and 110 shown in FIGS. 4A and 4B are cut from the processed substrate shown in FIG. 3. FIG. 4A shows the substrate 101, buried layer 103, insulating ring 104, epitaxial layer 105, and lateral overgrowth layer 106. The processed substrate shown in FIG. 4B already contains the substrate 101, buried layer 103, and epitaxial layer 105. As the first step for the fabrication process, a silicon dioxide layer 107 is formed by thermal oxidization on the surface of the epitaxial layer 105 and the lateral overgrowth layer 106. Then the silicon dioxide layer 107 is patterned to create silicon-exposed windows therein.

The next step is to perform deep reactive ion etching (DRIE) using the patterned silicon dioxide layer 107 as a protection mask. As shown in FIG. 5, the formed empty trench 110 is required to lower through the epitaxial layer 105 and reach the buried layer 103. For this purpose, an etch monitoring cavity 108 is created, as shown in FIG. 4A. As the etching goes on, the insulating ring 104 is first exposed on the bottom of the cavity 108 and then an etch step 109 is formed at the edge of the insulating ring 104. The height of the etch step 109 is used as an indication of the etch end. The width of the empty trench 110 ranges from 1 to 3 micron, typically is 2 micron. The width of the cavity 108 may be varied between 50 and 200 micron, typically is 100 micron. The cavity 108 is preferably located in the central region of the processed substrate.

As shown in FIG. 5, the empty trench 110 is turned into a filled trench 113 by being filled up with silicon nitride and polysilicon. The nitride layer 111 is deposited by low pressure chemical vapor deposition (LPCVD). Since LPCVD is a conformable coating the inside surface of the empty trench 110 can be covered uniformly. The thickness of the silicon nitride layer 111 ranges from 800 to 2000 angstrom, typically is 1000 angstrom. The polysilicon layer 112 is deposited also by LPCVD. The thickness of the polysilicon layer 112 is little larger than a half width of the empty trench 110 so that the empty trench is filled up completely.

It should be noted that, although the filled trench 113 of FIG. 5 only corresponds to the insulating trench 25 of FIG. 1, the insulating trench 19 of FIG. 1 is formed in the same process.

The continuing step, as shown in FIG. 6, is to perform DRIE again to create a plurality of trenches in the epitaxial layer 105. To do this a silicon dioxide layer 114 is deposited by LPCVD. Then photolithography is used to pattern the stack layer of the silicon nitride layer 111, polysilicon layer 112, and silicon dioxide layer 114 for creating silicon-exposed windows therein. Using the patterned stack layer as a protection mask, a DRIE process is carried out. This process is to create trenches 115 and 116 that pass through the epitaxial layer 105 and reach the buried layer 103.

It should be noted that for this process another etch monitoring cavity (not shown in the figure) is used to accurately control the depth of the etching.

As shown in FIG. 7, the next step is to convert the buried layer 103 into a porous silicon layer 117. Since the buried layer 103 is heavily doped it is preferable to anodize in HF solution than lightly doped substrate 101 and lightly doped epitaxial layer 105 and lateral overgrowth layer 106.

The used HF solution consists of one third of HF, one third of ethanol and one third of water. The used etching apparatus consists of a double tank cell separated by the processed substrate. For the process the two volumes of the HF solution are individually contacted by Pt grids. The frond side surface of the processed substrate faces the cathodically biased grid while the rear surface faces the anodically biased grid in the other tank.

It should be noted that a plurality of through holes in the inner hub 23 of FIG. 1 provide entries for the HF solution to attack the buried layer 103 so as to shorten the lateral anodization path and reduce the anodization time.

After forming porous silicon in selected areas of the processes substrate, a photoresist layer 118 is applied to cover the trenches 116. Then the processed substrate is immersed into a diluted KOH solution (1% KOH in H.sub.2 O) at room temperature to remove the porous silicon layer 117 in the selective area. This results in suspension components 121, 122, 123, air gap 119, and cavity 120, as shown in FIG. 8.

It should be noted that the suspension components 121, 122, 123 only correspond to the fingers of the combs 17, the immediate vibrating wheel 18, and the inner vibrating hub 23 of FIG. 1, the other suspension structures of FIG. 1, which are not shown in FIG. 8, are also formed in the same process.

After forming the suspension structures, a thin heavily doped diffusion layer 124 is formed on the surfaces of the suspension structures, as shown in FIG. 9. The diffusion process comprises of pre-deposition with POCI.sub.3 at 950.degree. C. for 30 min and post-diffusion at a temperature ranging from 900 to 1000.degree. C., typically 950.degree. C. in oxygen atmosphere for 30 min. During the diffusion step the porous silicon layer 117 is turned into an oxidized porous silicon layer 125. The oxidized porous silicon is similar to the silicon dioxide and can be used as an electrically insulating material.

It should be noted that the insulating ring 104 of the cavity 120 separates the diffusion layer 124 into two independence portions: one portion disposed on the bottom of the suspension structures and the other portion disposed on the interior top surface of the substrate 101.

It also should be noted that a plurality of through holes in the inner hub 23 of the FIG. 1 exist to make the diffusion layer 124 of FIG. 9 more uniform on the bottom of the hub 23 of FIG. 1.

Next, a silicon nitride layer 126 is deposited by LPCVD to coat the surfaces of the suspension structures, as shown in FIG. 10.

Thereafter, a metal layer 127, such as an aluminum layer is deposited by sputtering on the surfaces of the suspension structures. Since sputtering does not achieve conformal coating the aluminum layer 127 only forms on the top surfaces and sidewalls of the suspension structures, as shown in FIG. 11. Therefore, there is no aluminum layer on the bottom of the insulating ring 104.

It should be noted that the suspension structures are coated with three layers thereon: the bottom diffusion layer 124, immediate silicon nitride layer 126, and top aluminum layer 127. The bottom diffusion layer 124 and top aluminum layer 127 are conduction layers, but they are not electrically connected due to the immediate silicon nitride layer 126 inserted between them.

Although the present invention has been disclosed in terms of the preferred embodiments, it will be understood that modifications and variations can be made without departing from the true spirit and scope thereof, as set forth in the following claims. 

Optical switch array assembly for DNA probe synthesis and detection 

Abstract An optical switch array assembly for DNA probe light synthesis and hybridized DNA probe light detection is composed of a silicon substrate, an optical switch array disposed in the substrate, a glass plate mounted on the top of the substrate, and a DNA probe array disposed on the surface of the glass plate. The substrate also contains a driving circuit for forcing each optical switch on and off and an addressing circuit for locating each optical switch. A plurality of holes is disposed in the substrate so that each hole is aligned with an optical switch and guides a light beam to a corresponding optical switch.

---

Inventors:	Tu; Xiang Zheng (Redwood City, CA)
Family ID:	27753376
Appl. No.:	10/083,878
Filed:	February 27, 2002

What is claimed is:

1. An optical switch array assembly comprising: a silicon substrate, an optical switch array disposed in the silicon substrate, a driving circuit integrated in the silicon substrate with the optical switch array and forcing the optical switches on and off, a plurality of holes on the backside of the silicon substrate each aligned with an optical switch and guiding an optical beam to the optical switch, and a glass plate mounted on the top of the silicon substrate.

2. An optical switch array assembly of claim 1 further comprising a plurality of DNA probes disposed on the surface of the glass plate.

3. An optical switch array assembly according to claim 2, where said DNA probes are light-synthesized DNA probes.

4. An optical switch array assembly according to claim 2, where said light beams are directed to sites where said DNA probes are light-synthesized.

5. An optical switch array assembly of claim 1 further comprising a plurality of hybridized DNA probes disposed on the surface of the glass plate.

6. An optical switch array assembly according to claim 5, where said light beams are directed to sites where said hybridized DNA probes are light-detected.

7. An optical switch array assembly comprising: a silicon substrate, an optical switch array disposed in the silicon substrate, wherein each of the switches in said array is a Fabry-Perot cavity based optical switch, a driving circuit integrated in the silicon substrate with the optical switch array and forcing the optical switches on and off, and a plurality of holes on the backside of the silicon substrate each aligned with an optical switch and guiding an optical beam to the optical switch.

---

Description

---

FIELD OF THE INVENTION

This invention relates generally to an optical switch array assembly for DNA probe light synthesis and hybridized DNA probe light detection, and particularly to a micromachined optical switches array assembly that can be selectively actuated for DNA probe light synthesis and hybridized DNA probe light detection.

BACKGROUND OF THE INVENTION

With the advance of the human genome program, there is a strong movement to diagnose diseases and understand life phenomena by understanding living bodies on the basis of DNA. The main objective of DNA diagnosis is the development of a simple, accurate and cheep technique for DNA screening. The newly developed DNA chips represent a powerful technique for DNA screening. DNA chips have small size, allow a large reduction of sample and reagent consumption, are quick and can be used simply by untrained operators.

DNA arrays have been synthesized using light-directed methods. As an example, light-directed synthesis of oligonucleotides employs 5'-protected nucleosidephosphoramidite "building blocks." The 5'-protecting groups may be either photolabile or acid-labile. A plurality of DNA sequences in predefined regions are synthesized by repeated cycles of deprotection and coupling. Coupling occurs only at sites that have been deprotected. Three methods of light-directed synthesis are: use of photolabile protecting groups and direct photodeprotection; use of acid-labile 4,4'-dimethoxytrityl (DMT) protecting groups and a photoresist; use of DMT protecting groups and a polymer film that contains a photoacid generator (PAG). These methods have many process steps similar to those used in semiconductor integrated circuit manufacturing. These methods also often involve the use of masks that have a predefined image pattern that permits the light used for synthesis of the DNA arrays to reach certain regions of a substrate but not others. The pattern formed on the mask is projected onto a substrate to define which portions of the wafer are to be deprotected and which regions remain protected. In many cases a different mask having a particular predetermined image pattern is used for each separate masking step, and synthesis of a substrate containing many chips requires a plurality of masking steps with different image patterns. For example, synthesis of an array of 20 mers typically requires approximately seventy photolithographic steps and related unique masks. So, using present photolithographic systems and methods, a plurality of different image pattern masks must be pre-generated and changed in the photolithographic system at each masking step. A direct write optical lithography system has been developed to improve the cost, quality, and efficiency of DNA array synthesis by providing a maskless optical lithography system and method where predetermined image patterns can be dynamically changed during photolithographic processing. As such, an optical lithography system is provided to include a means for dynamically changing an intended image pattern without using a mask. One such means includes a spatial light modulator that is electronically controlled by a computer to generate unique predetermined image patterns at each photolithographic step in DNA array synthesis. The spatial light modulators can be, for example, micromachined mechanical modulators or microelectronic devices. One type of mechanical modulator is a micro-mirror array that uses small metal mirrors to selectively reflect a light beam to particular individual features; thus causing the individual features to selectively receive light from a light source (i.e., turning light on and off of the individual features).

Another type of mechanical modulator is designed to modulate transmitted rather than reflected light. An example of a transmission spatial light modulator is a liquid crystal display (LCD). There are a number of drawbacks with this direct write optical lithography system. First, the system consists of several optical active mechanical alignment apparatus including a mechanism for aligning and focusing the chip or substrate, such as an x-y translation stage and a stepping-motor-driven translation stage for moving the substrate by a distance equal to the desired center-to-center distance between chips. The mass production of the DNA probe arrays spends much labor and time and therefore they are very expensive. Particularly when the density of the cells where the probes are fixed, respectively, in a probe array is large, it is getting difficult to produce the probe array at a low cost.

Second, a complicated apparatus is required for optical detection of a hybridized DNA probe array. This apparatus may involve among others moving a sample substrate while simultaneously detecting light transmitted from one or more sample sites on the substrate by sequentially tracking the sample sites as they move. A stage, movable in a first direction, supports the substrate. A detector detects light emanating from an examination region delimited by a detection initiation position and a detection termination position. An optical relay structure transmits light from the examination region to the detector. A scanning mechanism simultaneously moves the optical relay structure and the substrate in the first direction. The optical relay structure tracks the substrate between the detection initiation position and the detection termination position.

SUMMARY OF THE INVENTION

The present invention is made for removing the above disadvantages, and an object of the present invention is to provide an optical switch array assembly for DNA probe light synthesis and hybridized DNA probe light detection without any moving apparatus for light alignment and probes tracking.

Another object of the present invention is to provide an optical switch array assembly that is integrated in a substrate with an optical switch array and at least a driving circuit so that not only each site but also each group of sites for DNA probe synthesis can be selectively illuminated.

Still another object of the present invention is to provide an optical switch array that is integrated in a substrate with an optical switch array and at least a driving circuit so that not only each hybridized DNA probe but also each group of hybridized DNA probes can be selectively illuminated for light detection.

Still another object of the present invention is to provide an optical switch array assembly that is integrated in a substrate with an optical switch array and at least a driving circuit so that all sites for DNA probe light synthesis can be directly illuminated without any interference with the reactive liquid.

Still another object of the present invention is to provide an optical switch array assembly that is integrated in a substrate with an optical switch array and at least a driving circuit so that all hybridized DNA probes can be directly illuminated without any cross talk between the adjacent hybridized probes.

Still another object of the present invention is to provide an optical switch array assembly composed of an optical switch array and at least a driving circuit can be batch produced simply using integrated circuit technology and micromachining technology.

Still another object of the present invention is to provide an optical switch array assembly for DNA probe light synthesis and hybridized DNA probe light detection that is simple, cheap and timesaving.

Still another object of the present invention is to provide an optical switch array assembly for DNA probe light synthesis and hybridized DNA probes light detection that can be operated in any laboratory and any hospital.

In order to achieve the above object, in the present invention, an optical switch array assembly is composed of a silicon substrate, an optical switch array disposed in the substrate, a glass plate mounted on the top of the substrate, and a DNA probe array disposed on the surface of the glass plate. The substrate also contains a driving circuit for forcing each optical switch or each group of optical switches on and off and perhaps an addressing circuit for locating each optical switch or each group of optical switches. A plurality of holes is also created in the substrate so that each hole is aligned with an optical switch on the backside and guides an optical beam to the optical switch.

The optical switches are constructed from two parallel thin film mirrors separated by an air gap. As well known, the transmission T of a loss less Fabry-Perot cavity is a function of the reflectivities R.sub.1 and R.sub.2 of the mirrors and of the air gap h between the mirrors:

where .lambda. is the working light wavelength.

This expression has maximum and minimum when the sine in the denominator is respectively zero and one. Thus for h being a multiple of .lambda./4, the transmission becomes

The reflectivity of the Fabry-Perot cavity can be become zero only if the mirrors are of equal reflectivity. In this case the above equation is always equal to one. Thus, to get a maximal contract and a maximum reflectivity of the Fabry-Perot cavity, the reflectivities R1 and R.sub.2 of the thin film mirrors must be as equal as possible and as high as possible.

It is also known that the reflectivities R1 and R2 of the thin film mirrors are maximum if their thickness is an odd multiple of .lambda./(4n), where n is the refractive index of the mirrors material. In this case R.sub.1 and R.sub.2 of the thin film mirrors can be expressed by ##EQU1##

where n.sub.0 is the refractive index of the underlying medium.

In the present invention the thin film mirrors are made of amorphous silicon carbide or plasma enhanced chemical vapor deposition (PECVD) silicon nitride or lower pressure vapor chemical deposition (LPCVD) silicon nitride. For ultraviolet light silicon dioxide, silicon nitride and silicon carbide are transparent, thus there is no light loss due to absorption. It has been reported that the refractive index of amorphous silicon carbide is 2.48 to 2.65, and the refractive index of silicon nitride is 2.0 to 2.1.

For a freestanding thin film membrane the underlying medium is air and n.sub.0 =1. Thus the above equation yields values of 52%-56% for R.sub.1,2 of amorphous silicon carbide and 40-36% for R.sub.1,2 of silicon nitride.

The optical switches based on the Fabry-Perot cavity are operated by electrostatic force. The top mirror of the Fabry-Perot cavity is supported by several flexible beams. Without a voltage applied to the cavity, h equals to odd multiple of .lambda./4 so that the transmission of the cavity reaches minimum or the switch is closed. With a voltage applied to the cavity, the flexible beams bend down and the supported mirror moves towards the bottom mirror so that h equals to even multiple of .lambda./4. As a result, the transmission of the cavity reaches maximum or the optical switch is opened. As shining light is perpendicularly projected on the backside of the substrate, each hole guides a light beam to a corresponding optical switch. When the optical switch is opened or closed the light beam proceeds forward along the extension direction of the hole or is reflected along a reverse direction.

The applied voltage is directed to a set of selective optical switches by the combination of a driving circuit and an addressing circuit. The driving circuit may be an electrical switch array. The addressing circuit may be a shift register circuit. Both the driving circuit and the addressing circuit may be partially or fully integrated in a same silicon substrate with the optical switch array.

Synthesizing DNA array with the optical switch array assembly using the DMT process may take place as follows. First, a computer file is generated and specifies, for each light illumination step, which optical switches in the optical switch array assembly need to be on and which need to be off to generate a particular predetermined light illumination pattern. Second, the glass plate of the optical switch array assembly is coated with photoresist on the synthesis surface and the optical switch array assembly is mounted in a holder or flow cell so that the synthesis surface is applied with DMT-protected nucleotides containing the desired base (adenine (A), cytosine (C), guanine (G), or thymine (T)). The photoresist may be either positive or negative thus allowing deprotection at locations exposed to the light or deprotection at locations not exposed to the light, respectively. Thirdly, the optical switch array is programmed for the appropriate configuration according to the desired predetermined light illumination pattern, a shutter in an arc lamp is opened, the synthesis surface is illuminated for the desired amount of time, and the shutter is closed. Fourth, the photoresist is developed and etched. Exposure of the glass plate to acid then cleaves the DMT protecting groups from regions of the glass plate where the photoresist has been removed. The remaining photoresist is then stripped. Fifth, DMT-protected nucleotides are coupled to the deprotected oligonucleotides. Sixth, the glass plate of the optical switch array assembly is re-coated with photoresist. These steps are repeated until the DNA array synthesis is complete.

In the analysis procedure, at first, all the components in the sample are labeled with tags such as fluorophores or enzymes. They are placed on the glass plate of the optical switch array assembly for hybridization. If the sample has a component being hybridized with probes on the glass plate, the component is held on the corresponding regions.

The positions of the region emitting fluorescence can be determined by selectively opening the corresponding optical switches of the optical switch array assembly. From the positional information of the fluorescence emitting regions, the probe species being hybridized with the sample components can be determined.

BRIEF DESCRIPTION OF THE INVENTION

FIG. 1A is a schematic diagram showing a DNA probe of an optical switch array assembly of the present invention being synthesized by projecting a light beam on a predetermined site by switching an optical switch on.

FIG. 1B is a schematic diagram showing a hybridized DNA probe of an optical switch array assembly of the present invention being detected by projecting a light beam on the hybridized DNA probe by switching an optical switch on.

FIG. 2 is a perspective view of a cell cut from an optical switch array assembly of the present invention, which comprises an optical switch and a DNA probe.

FIG. 3 is a cross-sectional view of an optical switch of the present invention at a first fabrication step in which a first silicon dioxide layer, a first amorphous silicon carbide layer, a second silicon dioxide layer, and a second amorphous silicon carbide layer have been deposited on the surface of a silicon substrate.

FIG. 4 is a cross-sectional view of an optical switch of the present invention at a second fabrication step in which an electrical interconnection, two spacers, and a metal reflective layer have been formed on the two sides of the silicon substrate, respectively.

FIG. 5 is a cross-sectional view of an optical switch of the present invention at a third fabrication step in which a refilled trench has been created in the silicon substrate.

FIG. 6 is a cross-sectional view of an optical switch of the present invention at a fourth fabrication step in which a plane configuration of a Fabry-Perot cavity has been defined in the silicon substrate.

FIG. 7 is a cross-sectional view of an optical switch of the present invention at a fifth fabrication step in which a light beam guiding hole has been created on the backside of the silicon substrate.

FIG. 8 is a cross-sectional view of an optical switch of the present invention at a seventh fabrication step in which an air gap has been formed in the silicon substrate and therefore a Fabry-Perot cavity based optical switch has been constructed.

FIG. 9 is a cross-sectional view of an optical switch of the present invention at an eighth fabrication step in which a glass plate has been mounted on the top of the silicon substrate.

FIG. 10 is a cross-sectional view of an optical switch of the present invention at a ninth fabrication step in which a DNA probe has been synthesized on the surface of the glass plate.

DETAILED DESCRIPTION OF THE INVENTION

An optical switch array assembly of the present invention is shown in FIG. 1A and FIG. 1B. As shown in FIG. 1A, the optical switch array assembly consists of a silicon substrate 101, a plurality of closed optical switches 102 and a plurality of opened optical switches including an opened optical switches 103 that are disposed in the silicon substrate 101, a glass plate 104 mounted on the top of the silicon substrate 101 with the periphery sealed, and a plurality of DNA probes including DNA probe 112 disposed on the surface of the glass plate 104. The optical switch array assembly also comprises a plurality of electrical switches 107 and a plurality of logic gates 108 addressed by a vertical shift register circuit 109 and a horizontal shift register circuit 110 for applying a voltage 111 to the optical switches. The electrical switches, the logic gates, and perhaps the register circuits are disposed in the silicon substrate 101. The backside of the silicon substrate 101 is illuminated by a light source. A plurality of the optical switches including optical switch 102 is off and the light beams indicated by number 106 are reflected back. A plurality of the optical switches including optical switch 103 is on and the light beam indicated by number 105 proceeds forward and reaches the front surface of the glass plate 104. This induces the DNA probes including DNA probe indicated by number 112 being synthesized on the light illuminated sites of the glass plate 104.

As shown in FIG. 1B, the optical switch array assembly has the same indication members as the optical switch array assembly shown in FIG. 1A, except that the DNA probe 112 is hybridized with a fluorophore tag attached sample DNA and therefore become a hybridized DNA probe 113. Under illumination the hybridized DNA probe 113 can emit fluorescence 114. Then the fluorescence can be detected by a photodetector. It is supposed that the kind of the DNA probe 112 is known. So when the hybridized DNA probe 113 is illuminated by selectively opening a corresponding optical switch, the kind of the sample DNA can be determined by addressing the location of the opened optical switch.

A single cell of a DNA probe array assembly of the present invention is shown in FIG. 2. The cell comprises a silicon substrate 201, a first amorphous silicon carbide layer 202, a silicon dioxide layer 203, and a second amorphous silicon carbide layer 204 which are sequentially laminated on the surface of the silicon substrate 301, and a Fabry-Perot cavity disposed in the silicon substrate 201. The Fabry-Perot cavity is constructed of a first amorphous silicon carbide membrane 205, an air gap 206, and a second amorphous silicon carbide membrane 207. The second amorphous silicon carbide membrane 207 is supported by four amorphous silicon carbide beams 208. Each amorphous silicon carbide beam 208 is connected to the silicon substrate 201 through two anchors 209. A metal electrode 210 covers the amorphous silicon carbide beams 208 and the outer portion of the second amorphous silicon carbide membrane 207. The first amorphous silicon membrane 205 is coated with an anti-reflective layer 211 on the backside and aligned with an opened back hole 213. The area occupied by the Fabry-Perot cavity, is surrounded by etch stop walls 214. Each anchor 209 is connected to an etch stop wall 214. A reflective layer 215 is coated on the backside of the silicon substrate 301. A glass plate 217 is mounted on the top of the silicon substrate through two spacers 216. A DNA probe 218 is positioned on the front surface of the glass plate 217 and aligned with the bottom hole 213.

The thicknesses of the first amorphous silicon carbide membrane 205 and the second amorphous silicon carbide membrane 207 are designed to be equal to odd multiple of .lambda./(4n.sub.SiC) so as to set their reflectivities to take maximum value, respectively. The original or static thickness of the air gap 206 is designed to be equal to odd multiple of .lambda./4 so as to set the Fabry-Perot cavity to be in a minimum transmission state or a full reflective state.

Since the resistivity of the amorphous silicon carbine ranges from 10.sup.10 to 10.sup.14 ohm-cm, the electrode 210 can combine with the silicon substrate 201 to form a parallel plate capacitor. When a voltage apply to the capacitor, a resulted electrostatic force pulls the second amorphous silicon carbide membrane 207 down to the first amorphous silicon carbide membrane 205 and therefore reduce the thickness of the air gap 206. As soon as the thickness of the air gap equals to even multiple of .lambda./4, the Fabry-Perot cavity is driven into a maximum transmission state.

The DNA probe 218 is generated by light illumination. The shape and the size of the DNA probe 218 are duplicated from the hole 213 because the light beam is introduced through the hole 213.

As an alternative, the mirrors of the Fabry-Perot cavity are made of PECVD silicon nitride instead of amorphous silicon carbide. In this case the stack structure of amorphous silicon carbide layer--silicon dioxide layer--amorphous silicon layer is replaced by a stack structure of PECVD silicon nitride layer--aluminum--PECVD silicon nitride layer.

As another alternative, the mirrors of the Fabry-Perot cavity are made of LPCVD silicon nitride instead of amorphous silicon carbide. In this case the stack structure of amorphous silicon carbide layer--silicon dioxide layer--amorphous silicon layer is replaced by a stack structure of LPCVD silicon nitride layer--silicon dioxide--LPCVD silicon nitride layer.

A first preferred fabrication process of the DNA probe array assembly of the present invention comprises eight steps as shown in FIG. 3 to FIG. 10. The first step as shown in FIG. 3, is to prepare a silicon substrate 301 and deposit a stack structure of a first silicon dioxide layer 302, a first amorphous silicon carbide layer 303, a second silicon dioxide layer 304, and a second amorphous silicon carbide layer 305. The silicon substrate 301 may have been undergone a standard MOSFET processing process. The MOS process is to fabricate a plurality of MOS circuits including MOSFET switches arrays, MOS gate circuits, and perhaps shift registers circuits (not shown in the figure). The first silicon dioxide layer 302 is deposited by PECVD and has a thickness little larger than .lambda./(4n.sub.SiO.sub..sub.2 ) so that after completing a subsequent hole formation process the thickness is just .lambda./(4n.sub.SiO.sub..sub.1 ). Deposition parameters used are power of 250 W, temperature of 300 centigrade, N.sub.2 O flow of 400 sccm, SiH.sub.4 flow of 40 sccm and pressure of 240 mtorr, resulting in a deposition rate of 600 angstrom/min. The first amorphous silicon carbide layer 303 is deposited by PECVD and has a thickness of odd multiple of .lambda./(4n.sub.SiC). The deposition system used is a dual frequency reactor. Deposition parameters used are 600 mTorr, 10 sccm SiH.sub.4, 250 sccm CH.sub.4, 300 sccm Ar, 300 centigrade, and a power level of 60 W. A resulted amorphous silicon carbide layer 303 has low stress of <50 MPa and high etch resistance. The second silicon dioxide layer 304 is deposited by a same PECVD process as above described and has a thickness of odd multiple of .lambda./4. The second amorphous silicon carbide layer 305 is deposited by a same PECVD as above described and has a thickness of odd multiple of .lambda./4n.sub.SiC). It is preferred that the thickness of the first amorphous silicon carbide layer 303 is much larger than thickness of the second amorphous silicon carbide layer 305 so that the first amorphous silicon carbide layer 303 has a higher mechanical strength.

The second step as shown in FIG. 4, is to create an electrical interconnection 306, a back reflective layer 307, and two spacers 308. To do this, a 1500 angstrom thick aluminum is deposited on the both sides of the silicon substrate 301 by electron beam evaporation. Patterning the aluminum layer is carried out using an aluminum etch solution of H.sub.3 PO.sub.4 :HAC:CH.sub.3 COOH=8:0.5:0.1:1. Then another patterned photoresist layer is formed on the front surface of the silicon substrate 301. Using the patterned aluminum layer as a base, a 3 micron thick aluminum layer is deposited. Removing the photoresist layer under the aluminum layer results in two aluminum tops mounted on the spacer 308. In this way, the thickness of the spacers 308 is increased to be larger than 3 microns, but the thickness of the electrical interconnection 307 is still kept unchanged.

The third step as shown in FIG. 5, is to form a refilled etch stop trench 309 and anchors (not shown in the figure). Firstly, a trench is created in the second amorphous silicon carbide layer 305 and the second silicon dioxide layer 304. Patterning of the second amorphous silicon carbide 305 is done by dry etching in a reactive ion etcher (RIE) with a power of 60 W and a pressure of 0.05 mbar. The gas flows are 70 sccm CF.sub.4, 10 sccm SF.sub.6 and 10 sccm O.sub.2, respectively. In these conditions an etch rate of 800 angstrom/min is obtained. The etch selectivities to oxide and photoresist are 1.2 and 0.7, respectively. The second silicon dioxide layer 304 is then etched in a diluted HF solution. Finally, the trench is refilled with a sandwiched plug. To this end, a lining layer of amorphous silicon is deposited by PECVD using a patterned photoresist layer as a mask. Then a central layer of silicon nitride is deposited by PECVD. The composite layer of the amorphous silicon and the amorphous nitride which is disposed in the outer portion of the refilled trench, is lifted off by resolving the underlying photoresist layer.

The fourth step as shown in FIG. 6, is to define the plane layout of a Fabry-Perot cavity. The second amorphous silicon carbide layer 305 is patterned again. Etching of the amorphous silicon carbide layer 305 is performed by a same process as above described. After etching a window 310 is created and the refilled trench 309 now becomes an etch stop wall 309.

The fifth step as shown in FIG. 7, is to create a hole 311 on the backside of the silicon substrate 301. To do this, a 9-10 micron thick positive photoresist is applied and patterned to form an etch open. The aluminum layer in the open is etched by a same process as above described. Then the revealed silicon in the open is etched in a Bosch etcher. The etch proceeds almost through the silicon substrate 301 and finely stops on the first silicon dioxide layer 302. The diameter of the hole 311 on the bottom ranges from 4 to 30 microns, typically 10 microns. The revealed portion of the first silicon dioxide layer 302 is little etched and therefore can be used as an anti-reflective film 312.

The sixth step as shown in FIG. 8, is to etch the second silicon dioxide layer 304 surrounding by the etch stop walls including etch stop wall 309. As a result, a first amorphous silicon carbide membrane 313, four amorphous silicon carbide beams 314, an air gap 315, and a second amorphous silicon carbide membrane 316 are constructed. The two opposite side edges of each amorphous silicon carbide beam 314 are supported by two anchors formed in the process for creating the refilled trenches 309. The size of the first amorphous silicon carbide membrane 313 ranges from 4.times.4 to 50.times.50 microns, typically 20.times.20 microns. The length and width of the amorphous silicon carbide beams 314 range from 4 to 30 microns, typically 10 microns, and from 1 to 5 microns, typically 2 microns, respectively.

The seventh step as shown in FIG. 9, is to mount a glass plate 317 on the top of the silicon substrate 301 through the spacers 308. The backside of the glass plate 316 may be coated with an anti-reflective layer. The periphery of the glass plate 316 may be sealed to protect the optical switches during a subsequent DNA synthesis process. The glass plate 317 is disposable, but the optical switch array may be used repeatedly. So it is preferable that the sealing means is detachable.

The eighth step as shown in FIG. 10, is to synthesize DNA probes including DNA probe 318 on the surface of the glass plate 317. The DNA probes including DNA probe 318 are synthesized using photolabile protecting groups and direct photodeprotection. In this approach, the surface of the glass plate 317 is modified with photolabile protecting groups. A first group of sites for DNA probe synthesis are illuminated through a first group of the opened optical switches, yielding reactive hydroxyl groups thereon. A 3' activated deoxynucleoside, protected at the 5' hydroxyl with a photolabile group, is then provided to the surface such that coupling occurs at the sites. Following capping, and oxidation, the glass plate 317 is rinsed and a second group of sites are illuminated through a second group of the optical switches to expose additional hydroxyl groups for coupling. A second 5' protected activated deoxynucleoside base is presented to the surface. The selective photodeprotection and coupling cycles are repeated until a desired set of probes is obtained.

As an alternative, the DNA probes including DNA probe 318 are synthesized using acid-labile 4,4'-dimethoxytrityl (DMT) protecting groups and a photoresist.

As another alternative, the DNA probes including DNA probe 318 are synthesized using DMT protecting groups and a polymer film that contains a photoacid generator (PAG).

A second preferred fabrication process of the DNA probe array assembly of the present invention utilizes a stack structure of a first aluminum layer, a first PECVD silicon nitride layer, a second aluminum layer, and a second PECVD silicon nitride layer instead of the stack structure of a first silicon dioxide layer, a first amorphous silicon carbide layer, a second silicon dioxide layer, and a second amorphous silicon carbide layer. The difference between the two processes includes the following items: 1. Deposition parameters used for the PECVD nitride layer are SiH.sub.4 170 sccm, NH.sub.3 30 sccm, temperature 300 centigrade, pressure 450 mtorr, power 250 w, resulting in a deposition rate of 160 angstrom/min. 2. RIE etching parameters used for the PECVD nitride layer are CHF.sub.4 7.5 sccm, N.sub.2 42.5 sccm, power 60W, and pressure 37.5 mtorr, resulting in an etch rate of 250 angstrom/min. 3. The refilled trenches or etch stop walls are refilled with silicon dioxide instead of both amorphous silicon and silicon dioxide. 4. The first aluminum layer revealed by creating a hole needs to be removed by etching, and then a .lambda./(4n.sub.SiO.sub..sub.2 ) thick silicon dioxide layer is deposited thereon to form an anti-reflective film.

A third preferred fabrication process of the DNA probe array assembly of the present invention utilizes a stack structure of a first silicon dioxide layer, a first LPCVD silicon nitride layer, a second silicon dioxide layer, and a second LPCVD silicon nitride layer instead of the stack structure of a first silicon dioxide layer, a first amorphous silicon carbide layer, a second silicon dioxide layer, and a second amorphous silicon carbide layer. The difference between the two processes includes the following items: 1. The MOSFET circuits disposed in the silicon substrate have not undergone a metallization step at the beginning of the fabrication process. 2. Deposition parameters used for the LPCVD nitride layer are SiH.sub.2 Cl.sub.2 170 sccm, NH.sub.3 30 sccm, temperature 850 centigrade, pressure 150 mtorr, power 400 w, resulting in a deposition rate of 90 angstrom/min. 3. RIE etching parameters used for the LPCVD nitride layer are CHF.sub.4 7.5 sccm, N.sub.2 42.5 sccm, power 60W, and pressure 37.5 mtorr, resulting in an etch rate of 250 angstrom/min. 4. Metallization for creating electrical interconnections (including interconnections for prefabricated circuits) and spacers is carried out as a final fabrication step.

The above description is illustrative and not restrictive. Many variations of the invention will become apparent to those of skill the art upon review of this disclosure. Merely by way of preferred embodiment, while the invention is illustrated primarily with regard to a Fabry-Perot cavity based optical switch array assembly, the invention is not so limited. The scope of the invention should, therefore, be determined not with reference to the above description, but instead should be determined with reference to the appended claims along with their full scope of equivalents.