Proportional micromechanical device

[0001] (Background of invention) (1. Field of invention) The present invention generally relates to semiconductor electromechanical microdevices, more specifically microdevices having high aspect ratio geometry and displaceable members with respect to the transducer. [0002] (2. Explanation of related technologies) Fluid valves generally include a fluid port, an actuator, and a movable valve structure that opens and closes the fluid port in response to the actuator. There are many types of fluid valves. Examples of fluid valves include solenoid valves and microvalves manufactured from microfabricated semiconductor materials, such as bimetallic microvalves and encapsulated fluid microvalves. However, there are many problems with each of these valve or microvalve types. [0003] Solenoid valves utilize a coil in the form of a cylinder and generally have a core that can be drawn into the cylinder by a magnetic field that is initiated when an electric current flows through the coil. Solenoid valves are typically used, for example, in conventional antilock braking systems. However, solenoid valves are usually relatively large and heavy. In addition, solenoid valves, such as solenoid valves, often require relatively high currents and can cause spiking of the voltage supply. Solenoid valves can also represent hysteresis, which can represent a non-linear response to electronic inputs. Moreover, the operation of solenoid valves, such as solenoid valves, can be relatively slow due to the relatively long lag time between the delivery of current to such valves and the resulting magnetic field and corresponding forces. .. Also, it is difficult to open or close the solenoid valve only partially, so the solenoid valve is typically used as an on / off valve rather than as a proportional valve. [0004] An exemplary bimetallic microvalve utilizes an actuator manufactured from two materials with different coefficients of thermal expansion. The difference in the coefficient of thermal expansion causes the actuator to bend or straighten as it heats or cools, thereby opening and closing the flow orifice. U.S. Pat. No. 5,058,856 discloses such a bimetallic microvalve with first and second substrates. The first substrate defines the flow orifice and valve seat. The second substrate defines the valve face alongside the flow orifice and also defines the movable actuator. Movable actuators include first and second layers of materials with substantially different coefficients of thermal expansion, such as silicon and nickel layers. The actuator also includes a heating element and is secured at one end so that the actuator can be bent due to differences in the coefficient of thermal expansion by selective heating. The bending of the actuator causes the valve face to disengage from the valve seat or face the valve seat to open and close the valve, thereby controlling the fluid flow through the orifice. [0005] However, one of the problems with such a bimetallic microvalve is that the actuator starts in response to a change in temperature, so that a change in the outside air temperature may cause an unintended start of the microvalve. In addition, the heated element, the actuator, is in contact with the fluid flow, which can cause undesired heating of the fluid in the fluid path, cooling the heater and displacing the actuator. In addition, the displacement of the actuator is also relatively small, generally on the order of 10 ppm / ° C. [0006] An example of an encapsulated fluid microvalve is disclosed in US Pat. No. 4,824,073. Encapsulated fluid microvalves utilize the principle of expansion and pressurization of a constant amount of fluid or gas in a closed cavity when the flexible thin film or diaphragm that forms one or more walls of the cavity is heated to deform. When the encapsulated fluid or gas is heated, the diaphragm is deformed to open and close the port and control the fluid flow through the fluid orifice. Heating of the encapsulated fluid or gas is achieved by a resistance heating element in the cavity such that an electric current flows through the resistance element to generate heat that heats the fluid or gas. [0007] Encapsulated fluid microvalves can generate relatively large forces and can be used, for example, as mass fluid controllers to control large volumes of fluid flow. In addition, encapsulated fluid microvalves may also operate proportionally to provide a proportional range of fluid control. That is, the valve can be controlled to adjust the proportion of fluid flow through the valve depending on the magnitude of the control signal. [0008] However, encapsulated fluid microvalves have a relatively slow response time due to the time required to heat and cool the fluid. In addition, the deformed membrane of the encapsulated fluid microvalve is in contact with the fluid or gas flow path. Therefore, the deformation temperature of the film can affect the temperature of the fluid or gas in the flow path, and the temperature of the fluid or gas in the flow path can affect the temperature of the deformed film. [0009] Moreover, none of the above valves provide flow-force and / or pressure-force compensation that minimizes the effect of fluid flow through the microvalve. [Patent Document 1] Japanese Patent Notice No. 49-039223 December 1974 [Patent Document 2] Patent Announcement No. 49-039224 December 1974 [Patent Document 3] Japanese Patent Publication No. 49-048264 October 1974 [Patent Document 4] Japanese Patent Publication No. 49-048265 October 1974 [Patent Document 5] Japanese Patent Publication No. 06-56014 March 1994 [Patent Document 6] Japanese Patent Publication No. 06-286600 October 1994 [Patent Document 7] Japanese Patent Notice No. 59-2175 January 1984 [Patent Document 8] Japanese Patent Notice No. 59-2175 January 1984 [Patent Document 9] U.S. Pat. No. 3747628 July 1973 [Patent Document 10] U.S. Pat. No. 3860949 January 1975 [Patent Document 11] U.S. Pat. No. 4005454 January 1977 [Patent Document 12] U.S. Pat. No. 4019388 April 1977 [Patent Document 13] U.S. Pat. No. 4023725 May 1977 [Patent Document 14] U.S. Pat. No. 4,152,540 May 1979 [Patent Document 15] U.S. Pat. No. 4181249 January 1980 [Patent Document 16] U.S. Pat. No. 4298023 November 1981 [Patent Document 17] U.S. Pat. No. 4341816 July 1982 [Patent Document 18] U.S. Pat. No. 4434813 March 1984 [Patent Document 19] U.S. Pat. No. 4581624 April 1986 [Patent Document 20] U.S. Pat. No. 4,628576 December 1986 [Patent Document 21] U.S. Pat. No. 4,47013, March 1987 [Patent Document 22] U.S. Pat. No. 4,661835 April 1987 [Patent Document 23] U.S. Pat. No. 4,772,935, September 1988 [Patent Document 24] U.S. Pat. No. 482 1997 April 1989 [Patent Document 25] U.S. Pat. No. 4824073 April 1989 [Patent Document 26] U.S. Pat. No. 4826131 May 1989 [Patent Document 27] U.S. Pat. No. 4,828,184 May 1989 [Patent Document 28] U.S. Pat. No. 4,869,282, September 1989 [Patent Document 29] U.S. Pat. No. 4,938,742, July 1990 [Patent Document 30] U.S. Pat. No. 4943032 July 1990 [Patent Document 31] U.S. Pat. No. 4,959,581 September 1990 [Patent Document 32] U.S. Pat. No. 4,966,646, October 1990 [Patent Document 33] U.S. Pat. No. 5029805 July 1991 [Patent Document 34] U.S. Pat. No. 5037778 August 1991 [Patent Document 35] U.S. Pat. No. 5050838 September 1991 [Patent Document 36] U.S. Pat. No. 5054522 October 1991 [Patent Document 37] U.S. Pat. No. 5058856 October 1991 [Patent Document 38] U.S. Pat. No. 5091914 October 1991 [Patent Document 39] U.S. Pat. No. 5064165 November 1991 [Patent Document 40] U.S. Pat. No. 5065978 November 1991 [Patent Document 41] U.S. Pat. No. 5066533 November 1991 [Patent Document 42] U.S. Pat. No. 5069419 December 1991 [Patent Document 43] U.S. Pat. No. 5074629 December 1991 [Patent Document 44] U.S. Pat. No. 5082242 January 1992 [Patent Document 45] U.S. Pat. No. 5096643 March 1992 [Patent Document 46] U.S. Pat. No. 5,116457 May 1992 [Patent Document 47] U.S. Pat. No. 5131729 July 1992 [Patent Document 48] U.S. Pat. No. 5,133379 July 1992 [Patent Document 49] U.S. Pat. No. 5142781 September 1992 [Patent Document 50] U.S. Pat. No. 5,161774 November 1992 [Patent Document 51] U.S. Pat. No. 5,169,472 December 1992 [Patent Document 52] U.S. Pat. No. 5,176358 January 1993 [Patent Document 53] U.S. Pat. No. 5,177579 January 1993 [Patent Document 54] U.S. Pat. No. 5,178,190 January 1993 [Patent Document 55] U.S. Pat. No. 5,179499, January 1993 [Patent Document 56] U.S. Pat. No. 5,180,623 January 1993 [Patent Document 57] U.S. Pat. No. 5,197517 March 1993 [Patent Document 58] U.S. Pat. No. 5209118 May 1993 [Patent Document 59] U.S. Pat. No. 5,215,244, June 1993 [Patent Document 60] U.S. Pat. No. 5,216,273 June 1993 [Patent Document 61] U.S. Pat. No. 5,217,283, June 1993 [Patent Document 62] U.S. Pat. No. 5,238223, August 1993 [Patent Document 63] U.S. Pat. No. 5,244,537 September 1993 [Patent Document 64] U.S. Pat. No. 5,267589 December 1993 [Patent Document 65] U.S. Pat. No. 5271431 December 1993 [Patent Document 66] U.S. Pat. No. 5271597 December 1993 [Patent Document 67] US Pat. No. 5,309,943 May 1994 [Patent Document 68] U.S. Pat. No. 5325880 July 1994 [Patent Document 69] U.S. Pat. No. 5333831 August 1994 [Patent Document 70] U.S. Pat. No. 5,360,662 August 1994 [Patent Document 71] U.S. Pat. No. 5355712 October 1994 [Patent Document 72] U.S. Pat. No. 5,368,704 November 1994 [Patent Document 73] U.S. Pat. No. 5,400,824 March 1995 [Patent Document 74] U.S. Pat. No. 5,417,235 May 1995 [Patent Document 75] U.S. Pat. No. 5,445,185 August 1995 [Patent Document 76] U.S. Pat. No. 5,553790 September 1996 [Patent Document 77] U.S. Pat. No. 5,567,003 September 1996 [Patent Document 78] U.S. Pat. No. 5,575,533 November 1996 [Patent Document 79] U.S. Pat. No. 5,785,295 July 1998 [Patent Document 80] U.S. Pat. No. 5810325 September 1998 [Patent Document 81] U.S. Pat. No. 58383351 November 1998 [Patent Document 82] U.S. Pat. No. 5,884,605 December 1998 [Patent Document 83] U.S. Pat. No. 5,873,385 February 1999 [Patent Document 84] U.S. Patent No. 5909078 June 1999 [Patent Document 85] U.S. Pat. No. 5,926,955, July 1999 [Patent Document 86] U.S. Pat. No. 5,941608 August 1999 [Patent Document 87] U.S. Pat. No. 6019437 February 2000 [Patent Document 88] U.S. Pat. No. 6,105,737 August 2000 [Patent Document 89] U.S. Pat. No. 6,523,560 February 2003 [Patent Document 90] German Patent No. 2215526 October 1973 [Patent Document 91] German Patent No. 2930779 February 1980 [Patent Document 92] German Patent No. 4101575 July 1992 [Patent Document 93] German Patent No. 44172251 November 1995 [Patent Document 94] German Patent No. 4422942 January 1996 [Patent Document 95] European Patent No. 0250948 January 1988 [Patent Document 96] European Patent No. 0261972 March 1988 [Patent Document 97] UK Patent No. 2238267 Issue May 1991 [Patent Document 98] International Patent No. 9916096 April 1999 [Non-Patent Document 1] Qi, J. and Johnson, W. et al. (Apr. 6-9, 1999) "Flip Chip on Laminate Manufacturability," 1999 International Conference on High Density Packaging and MCMs. Proc. SPEI--Int. Soc. Opt. Eng. (USA), Denver, CO., Pp. 345-352 .. [Non-Patent Document 2] Search Report (Mar. 15, 2000) PCT / US 99/19971 .. [Non-Patent Document 3] Williams, KR et al. (Jun. 7-10, 1999). "A Silicon Microvalve for the Proportional Control of Fluids," Transducers '99 Sendai, Japan., Pp. 1804-1807. [Non-Patent Document 4] Yunkin, VA, et al. (1994). "Highly Anisotropic Selective Reactive Ion Etching of Deep Trenches in Silicon," Microelectronic Engineering 23: 373-376 .. [Non-Patent Document 5] Bartha, JW, et al., (1995). "Low Temperature Etching of Si in High Density Plasma Using SF.sub.6 /O.sub.2," Microelectronic Engineering, 27: 453-456 .. [Non-Patent Document 6] Carpenter Technology Corporation Technical Data sheet for "Carpenter Low Expansion; 42", date Nov. 1980 &lt;http://www.carpenter.idesinc.com/datasheet. asp? e = 181 &amp; u = eVIEW = PRINTER&gt; (visited on Mar. 28, 2002) .. [Non-Patent Document 7] Delphi Automotive Systems product brochure (1997). Variable Bleed Solenoid (VBS) for Transmission, copyright 1997 .. [Non-Patent Document 8] Delphi Automotive Systems product brochure (1998). On / Off Transmission Solenoids, copyright 1998 .. [Non-Patent Document 9] Duffy, James E. (1994). "Automatic Transmission Fundamentals," Modern Automotive Technology, copyright 1994, p. 707 .. [Non-Patent Document 10] Fung, CD, et al. (Nov. 7-8, 1984). "Deep Etching of Silicon Using Plasma," Proceedings of the Workshop on Micromachining and Micropackaging of Transducers, pp. 159-164 .. [Non-Patent Document 11] Houston, PN et al. (Jun. 1-4, 1999). "Low Cost Flip Chip Processing and Reliability of Fast-Flow, Snap-Cure Underfills," 1999 Electronic Components and Technology Conference. San Diego, CA, pp. 61 -70 .. [Non-Patent Document 12] Jonsmann, J. et al. (Jan. 17-21, 1999). "Compliant Electro-thermal Microactuator" Twelfth IEEE International Conference on Micro Electro Mechanical Systems Orlando, Florida, IEEE Technical Digest entitled IEEE Catalog No .: 99CH36291C pp. 588-593 .. [Non-Patent Document 13] Klaasen, EH et al. (1995). "Silicon Fusion Bonding and Deep Reactive Ion Etching; A New Technology for Microstructures," Proc. Tranducers 95 Stockholm Sweden pp. 556-559 .. [Non-Patent Document 14] Konarski, Mark M. (May 31 to Jun. 4, 1998). "Cure Parameter Effects on the Tg and CTE of Flip Encapsulants," 43rd International SAMPE Symposium and Exhibition. Materials and Process Affordability. Keys to the Future. Anaheim, CA, vol. 1 pp. 823-832. [Non-Patent Document 15] Linder, C. et al. (Jun. 1991). "Deep Dry Etching Techniques as a new IC Compatible Tool for Silicon Micromachining," Proceedings, Transducers '91 pp. 524-527 .. [Non-Patent Document 16] Madou, Marc (1997). "Scaling Laws, Actuators, and Power in Miniaturization," Chapter 9 In Fundamentals of Microfabrication CRC Press LLC: Boca Raton, FL., Pp. 405-446 .. [Non-Patent Document 17] Noworolski, JM et al., (1996) "Process for in plane-and out-of-plane single crystal-silicon thermal conductors" Sensors and Actuators 55 (1): 65-69 .. [Non-Patent Document 18] Ohio State Univesity Chemistry Department brochure page on Marc J. Madou [online]. Ohio State University [retrieved on Dec. 31, 2000]. Retrieved from Internet: &lt;URL: http://www.chemistry.ohio-state.edu/resource/pubs/brochure/madou.htm&gt;. [Non-Patent Document 19] Petersen, KE et al. (Jun. 1991). "Surface Micromachined Structures Fabricated with Silicon Fusion Bonding," Proceedings, Transducers '91 pp. 397-399 .. [Non-Patent Document 20] Author Unknown. (Sep./Oct. 1999). "HiTecMetal Group Develops Niche Market for Brazed Laminated Assemblies," Fluid Power Journal 27 .. [Non-Patent Document 21] Ayon, AA et al. (Jun. 1998). "Etching Characteristics and Profile Control in a Time-Multiplexed ICP Etcher," Proc. Of Solid State Sensor and Actuator Workshop Hilton Head SC pp. 41-44 .. [Non-Patent Document 22] Eaton Corporation. "Products: Powertrain," located at &lt;http://www.autocontrols.eaton.com/powertrain_body.html&gt; visited on Sep. 1, 1999. (2 pages) .. [0010] Therefore, there has been a need for microvalves that are small, lightweight, cost effective, easy to manufacture, and have fast reaction times. Also, precise and proportional fluid control with flow-force and / or pressure-force compensation, where the response to the control stimulus input is linear, there is no hysteresis, and the effect of fluid flow through the microvalve is minimized. The provided microvalve has been needed. There has also been a need for valves where the operation of the valve does not significantly heat the fluid or gas flowing through the valve. In addition, there has been a need for microvalves that function independently of outside air temperature. The present invention meets these needs. [0011] (Gist of the invention) According to one aspect of the invention, a semiconductor micromechanical device generally includes a first nearly planar layer and a second nearly planar semiconductor layer. The first and second members hang from the second layer, each floating in a cavity defined by the second layer. The first layer may also define a portion of the cavity. The displaceable structure also floats in the cavity from the first and second floating members. The actuator is operatively connected to a first floating member, which may transmit a force that causes displacement of the displaceable member. [0012] According to another aspect of the invention, there is provided a microstructure of the invention that is utilized as a microvalve containing first, second and third layers. The second layer is fixed between the first layer and the third layer. All three layers are preferably made from substantially the same material. The first layer and / or the third layer may define inlet and exit ports. The second layer defines the flow area sealed by the first and third layers and initiates the inlet and outlet ports, the displaceable member, and the displaceable member to open and close the microvalve. Allows fluid flow to and from the above actuators. The displaceable member and one or more actuators are suspended between the first and third layers. The second layer is preferably heavily doped to reduce resistance. The electrical contacts of the actuator are preferably provided through a third layer. During operation, the current is driven to flow through the actuator through electrical contacts, causing heating and thermal expansion of the actuator. The actuator is placed in relation to the displaceable member, and due to the thermal expansion of the actuator, the displaceable member is placed between the open and closed positions with respect to one of the inlet and outlet ports in the plane of the second layer. Displaced in position. The displacement member has a high aspect ratio (ratio to width of height) and is therefore compliant in the plane of the layer and is upright from the plane. [0013] Therefore, the microdevice of the present invention is small and easy to manufacture. It has virtually no hysteresis, has a linear response, and can respond quickly to input stimuli. More specifically, a small displaceable semiconductor structure can float from the semiconductor layer and move with certain accuracy in the plane of the layer in response to an input stimulus. The displaceable structure can act as a valve that opens and closes the fluid port without heating the fluid as it flows through the port. Since the layers have the same coefficient of thermal expansion, the movement of the displaceable semiconductor structure is not affected by the outside air temperature. [0014] These and other features and advantages of the present invention, along with the drawings, will be understood by the following detailed description. [0015] (Detailed description of preferred embodiments) The present invention includes a semiconductor micromechanical device that includes a semiconductor layer that defines a cavity. The displaceable structure is suspended in the cavity by the first and second members. The first and second members also float in the cavity. At least one floating member acts as an actuator that can displace the displaceable structure within the cavity. The following description is presented to those skilled in the art to allow them to create and use the present invention. A description of a particular application is provided as an example only. Various modifications are apparent to those skilled in the art, and the general principles set forth herein apply to other embodiments and applications without departing from the spirit and scope of the invention. Can be done. Accordingly, the invention is not limited to the embodiments shown, but is consistent with the broadest scope consistent with the principles and features disclosed herein. [0016] With reference to the exemplary drawings of FIGS. 1 and 2, a first preferred embodiment of the microvalve 10 according to the present invention is shown. The microvalve 10 of the first embodiment generally comprises three layers or substrates, namely a first layer 12, a second layer 14, and a third layer 16. The first layer 12 defines an inlet port 20 and an exit port 22. The second layer 14 defines a cavity 24 that is anchored between the first layer 12 and the third layer 16 and contains a fluid region that allows fluid flow between the inlet port 20 and the outlet port 22. To do. The second layer 14 further defines a displaceable member 26 that can displace in response to the thermal actuators 28 and 30 to open and close the inlet port 20. In this embodiment, the displaceable member 26 is extended. Electrical contacts 32a, 32b, 34a, and 34b, which electrically heat the actuators 28 and 30, respectively, are provided in vias through a third or cap layer 16. [0017] If an input such as an electric current is applied through each of the actuators 28 and 30 via electrical contacts 32a-b and 34a-b, each of the actuators 28 and 30 will be in the direction indicated by the arrows D28 and D30, respectively. Can exert power on. The forces in the directions of arrows D28 and D30 displace the displaceable member 26 in the direction indicated by the arrow D26 so that at least a portion of the displaceable member 26 is aligned perpendicular to the inlet port 20. Therefore, the current acts as an input stimulus to start the actuator. The input port 20 is closed at least partially by aligning the displaceable members 26 associated with the inlet port 20 at least partially vertically. The degree of displacement or alignment of the displaceable members 26 may be selected, for example, to control the flow rate of the fluid flow. If no input is applied through actuators 28 and 30, the actuators 28 and 30 can exert a force in the direction opposite to the direction indicated by arrows D28 and D30, respectively, and displace in the direction opposite to the direction indicated by arrow D26. Displacement of the member 26 returns the displaceable member 26 to its normal open position with respect to the inlet port 20. [0018] Alternatively, the microvalve 10 may be formed such that the displaceable member 26 is in the normally closed position associated with the inlet port 20 and is displaceable to open the inlet port 20. In another alternative embodiment, the microvalve 10 is such that the displaceable member 26 is in the normal open or closed position associated with the outlet port 22 and is displaceable to open and close the outlet port 22. Can be formed. [0019] Preferably, the first, second and third layers 12, 14, and 16 are made from silicon or other semiconductor material. Alternatively, the first and / or third layers 12 and 16 can be made from glass (Pyrex), conductive ceramics, steel, aluminum, and / or other metals, or conductive materials. The second layer 14 is preferably a single crystal semiconductor (SCS) because it is strong, flexible, and resistant to performance degradation. [0020] Although the microvalve is generally described herein as an opening / closing inlet port 20, such description is for illustrative purposes only, and apparently the microvalve 10 opens and closes the outlet port 22. Can be easily applied when In addition, the microvalve 10 is described herein as a normally open (NO) valve, but can be readily applied to normally closed (NC) valves. Further, for the sake of clarity and brevity, only the actuator 28 and the corresponding electrical contacts 32a and 32b are generally described, but the description is in the actuator 30 and the electrical contacts 34a and 34b. Correspondingly applicable. [0021] [0021] Although only the recess 18 of the first layer 12 is shown in FIG. 1, preferably the first and third layers 12 and 16 define the shallow recess 18. The recess 18 is defined in the area aligned with the displaceable members 26 and actuators 28 and 30 of the second layer 14, and the displaceable members 26 and actuators 28 and 30 between the first and third layers 12 and 16. Provides clearance for floating and displacement within the planar cavity 24 of the second layer 14. The recess 18 may also be defined in the area aligned with the cavity 24 to further facilitate fluid flow through the cavity 24. Alternatively or additionally, the displaceable members 26 and actuators 28 and 30 of the second layer 14 may be recessed or thinned from the first and third layers 12 and 16 (not shown). Provides a gap between the layers. The recess 18, and / or the displaceable member 26 and the recesses of the actuators 28 and 30, may be at the same depth or at different depths. For example, in one embodiment, the recess 18 and / or recess is between the displaceable member 26 and the inlet port 20 when the displaceable member 26 is lined up above the inlet port 20 to stop fluid flow. A gap of approximately 0.5 μm between the displaceable member 26 and each of the first and third layers 12 and 16 in the region near the inlet 20 to minimize fluid leakage by reducing the distance. Can be provided. In addition, recesses 18 and / or recesses are also other such as between actuators 28 and 30 and each of the first and third layers 12 and 16 to minimize fluid or gas pressure differences. A gap of about 1 μm or less can be provided in the region. [0022] When the microvalve 10 is used as a fluid valve, the recess 18 of the first embodiment is preferably about 0.5 μm deep. Therefore, when the displaceable member 26 is in the closed position, a gap is formed between the recess 18 and the stopper end 42. Leakage of fluid through the microvalve 10 is minimal, even with such spacing. These spaces include them in the cavity 24 and between ribs 48 and 50. If these spaces are filled with fluid, further fluid leakage is further limited by the size of the gap, as well as the tension of the fluid. This small leak has already been demonstrated by the larger 5 μm spacing in conventional solenoid fluid valves currently used in antilock braking systems. Therefore, it is preferable to provide a depth of about 0.5 μm for the recess 18 for the fluid microvalve in certain applications. [0023] The second layer 14 is preferably doped, and more preferably more doped, such as, for example, a more doped P-type single crystal semiconductor (SCS). Doping reduces the resistance of the second layer 14. The low resistance facilitates the increased current flow through the thermal actuators 28 and 30. Currents are applied through actuators 28 and 30 via contacts 32a, b and 34a, b, respectively. By increasing the current applied to the contacts 32a, b and 34a, b, the thermal expansion of the actuators 28 and 30 leads to an increase in the force exerted by the actuators 28 and 30 on the displaceable member 26. [0024] For example, the resistance ρ of the second layer 14 is preferably about 0.001 to 0.01 (.cm), with the desired amount of heating and heating by the ribs by the current flow from one side to the other side of the ribs 48. Heat dissipation is obtained. If the rib 48 is silicon, the rib 48 can withstand temperatures up to 1000 ° C, preferably up to about 500 ° C. Preferably the silicon rib 48 is from 10 ° C. It is heated up to 500 ° C, more preferably between 100 ° C and 200 ° C. Obviously, the start of the microvalve 10 is a rib related to the temperature of the remaining microvalve 10. It relies on 48 heatings and is independent of the outside temperature. [0025] The electrical contacts 32a-b are provided on the third layer 16 and are aligned perpendicular to the thermal actuator 28. The electrical contacts 32a-b provide electrical contacts through the via 35 for the application of current to the actuator 28. The rib 48 acts as a conductive path between the contacts 32a and 32b through the heavily doped second layer 14. The contacts 32a-b are electrical contacts, preferably having a region of a second layer 14 that is insulated except for the current conductive path formed by the ribs 48. Such electrical insulation can be established by providing a groove 36 in the second layer 14 to prevent a short circuit between the electrical contacts 32a and 32b. The groove 36 is filled with a dielectric material to flatten the groove 36. Alternatively, electrical insulation can be established by oxidizing rib 48 to prevent short circuits through the fluid between the electrical contacts 32a and 32b. [0026] Further electrical insulation between ribs 48 and 50 is unnecessary. This is because ribs 48 and 50 provide the lowest resistance path between contacts 32a and 32b and between contacts 34a and 34b, respectively, when each set of ribs 48 and 50 is driven independently. is there. In addition, the first and third layers 12 and 16 have lower doping levels and are minimally electrically conductive, unlike the second layer 14, and the applied current remains in the second layer 14. So electrical insulation between the first and second layers 12 and 14 and between the second and third layers 14 and 16 is also achieved. If layers 12, 14, and 16 contain silicon, the surfaces of layers 12, 14, and 16 are also oxidized to provide additional electrical insulation. [0027] The electrical contacts 32a and 32b are perpendicular to the ribs 48 of the actuator 28 so that one contact (eg 32a) is perpendicular to the rib on one side of the insulating groove 36 and the other contact (eg). , 32b) can be seen to line up perpendicular to the ribs on the other sides of the groove 36. The contacts 34a and 34b are similarly aligned perpendicular to the rib 50 of the actuator 30. It can be understood that such vertical alignment provides a smaller microvalve. [0028] The displaceable member 26 has a first actuator end 40 in contact with the thermal actuators 28 and 30, and a second stopper end 42 arranged and formed to open and close the inlet port 20. The displaceable member 26 has an increased cross-sectional area from the first actuator end 40 to the stopper end 42. The larger the area of the second stopper end 42, the greater the force of the displaceable member 26 to withstand the fluid pressure difference. [0029] In a preferred embodiment, each of the actuators 28 and 30 has shafts 44 and 46 with ribs 48 and 50 extending from the shaft, respectively. The shafts 44 and 46 are generally perpendicular to the displaceable member 26. Therefore, by applying an electric current through the rib 48, the rib 48 is thermally expanded, and then a force in the direction indicated by the arrow D28 is exerted on the shaft 44 against the displaceable member 26. Thus, in this embodiment, the shafts 44 and 46, as well as the ribs 48 and 50, serve as both a floating force of the displaceable member 26 in the cavity 24 and a transmission of displacement forces to the displaceable member 26. Can be understood to constitute. [0030] Further, the thermal actuators 28 and 30 are located in the cavity 24 filled with fluid, but preferably the thermal actuators 28 and 30 are located outside the area of fluid flow between the inlet and outlet ports 20 and 22. Will be done. In general, the fluid in the region outside the fluid flow region does not flow and a vacuum is absolutely necessary to remove this dead volume of the fluid. Thus, when the fluid insulates heat, the dead volume of the fluid also acts as a heat insulator between the thermal actuators 28 and 30 and the fluid flow, thereby preventing the fluid flow from being heated. .. [0031] Actuators 28 and 30 and displaceable members 26 float within the cavity 24 between the first and second layers 12 and 16. Specifically, the rib 48 is anchored or secured to the second layer 14 at one end and is ribbed between the first and third layers 12 and 16 by the second layer 14. 48 will float. The shaft 44 and the displaceable member 26 also float in the same manner, and the shaft 44 is supported by the rib 48 and the displaceable member 26 supported by the shafts 44 and 46. In addition, thermal actuators 28 and 30 have a high aspect ratio (ratio to width of height) formed by deep reactive ion etching (DRIE). The thermal actuators 28 and 30 preferably have an aspect ratio in the range of 1: 1 to 50: 1, more preferably about 20: 1. The displaceable member 26 preferably has an aspect ratio in the range of 0.5: 1 to 50: 1, more preferably about 1: 1. Therefore, each of the floating rib 48, the shaft 44, and the displaceable member 26 is displaceable in the plane of the second layer 14, but the direction of displacement is limited by the respective support structures. [0032] FIG. 3 shows the actuator 28 in more detail. Since the rib 48 is tethered to the second layer 14 at one end, the rib 48 cannot thermally expand towards the tethered end. If anything, the rib 48 can thermally expand towards the shaft 44. Each end of the rib 48 is preferably tipped so that the cross section of the junction between the rib 48 and the second layer 14 and / or the rib 48 and the shaft 44 is smaller. It's getting thinner. The tapered junction allows hinge action and allows the shaft 44 to be displaced. Further, the rib 48 has an angle θ with respect to the perpendicular line of the shaft 44, and when the rib 48 is thermally expanded, the rib 48 moves the shaft 44 toward the first actuator end 40 of the displaceable member 26. Displace. The rib angle θ is preferably relatively small, for example, between 2 ° and 5 °, in order to amplify the displacement of the shaft 44. The small angle θ maximizes the displacement of the shaft 44 to the actuator end 40 of the displaceable member 26 for a given degree of thermal expansion of each rib 48. [0033] Further pairs of ribs 48 may be provided to increase the force exerted by the shaft 44 on the actuator end 4 of the displaceable member 26. For example, the actuator 28 may have 5 pairs of ribs 48, as shown in FIGS. 1 and 2. Obviously, the number of pairs of ribs can be easily changed to obtain the force exerted on the stopper end 42 of the desired displaceable member 26. Preferably, the ribs 48 are provided in pairs with respect to the shaft 44, one opposite to the other, and the forces perpendicular to the shaft 44 exerted by the ribs 48 are offset by the opposite ribs 48. Thus, the net force exerted by the rib 48 is parallel to the shaft 44, which then exerts a force on the actuator end 40 of the displaceable member 26. [0034] In this preferred embodiment, each rib 48 is about 200-2000 μm long, 50-200 μm wide and 400 μm high, thus having an aspect ratio of about 2: 1 to 8: 1. The shaft 44 is preferably 1-2 mm in length and width and has an aspect ratio from about 5: 1 to 10: 1. Further, the displaceable member 26 is preferably about 6 mm long, 250-1000 μm wide and 400 μm high. Thus, by providing 5 pairs of silicon ribs for each of the actuators 28 and 30, and applying a total current of 20 amps through the ribs 48 and 50, the shafts 44 and 46 can displace a force of approximately 15-20 N. It affects the actuator end 40 of the member 26. This force is converted into a force of about 0.5 N and a displacement of 150-200 μm at the stopper end 42 of the displaceable member 26. A displacement of about 400 μm at the stopper end 42 of the displaceable member 26 can be easily achieved by a microvalve of similar dimensions. Such microvalves can have switching times shorter than 10 ms and can withstand fluid pressures up to about 5 kpsi, balancing the pressure of the displaceable member 26, more than 0.5 liters per minute, as described. Can accommodate fluid flow. [0035] Shafts 44 and 46 of the thermal actuators 28 and 30 are arranged with respect to the displaceable member 26 to exert a displacement rotational force on the actuator end 40 upon displacement of the shafts 44 and 46 towards the actuator end 40. In a preferred embodiment, the force of this displacement is essentially a rotational force around the position of member 26 between the first and second shafts 44 and 46, as indicated by arrow D26. As shown in FIG. 1, the actuators 28 and 30 are preferably located on opposite sides of the displaceable member 26 and offset with respect to each other along the axial length of the displaceable member 26. The offset distance can be selected by the desired rotational force at the stopper end 42 of the displaceable member 26, and the displacement. Due to the energy protection, the displacement of the shaft 44 is generally inversely proportional to the force exerted by the shaft 44. That is, as the offset distance becomes larger, the force of displacement at the stopper end 42 becomes larger and the displacement becomes smaller. On the other hand, as the offset distance becomes smaller, the rotational force at the stopper end 42 becomes smaller and the displacement becomes larger. Therefore, the distance at which the thermal actuators 28 and 30 are offset from each other along the axial length of the displaceable member 26 can be selected to achieve the desired equilibrium between the rotational force and the displacement. [0036] In order to start the thermal actuator 28, an electric current is applied to the heating rib 48 between the electrical contacts 32a and 32b, causing thermal expansion of the rib 48. The thermal expansion of the rib 48 causes the displacement shaft 44 to move toward the first actuator end 40 of the displaceable member 26. Similarly, an electric current is simultaneously applied to the heating rib 50 between the electrical contacts 34a, 34b, causing thermal expansion of the rib 50. The thermal expansion of the rib 50 also causes the displacement shaft 46 to move towards the first actuator end 40 of the displaceable member 26. Since the shafts 44 and 46 are offset from each other along the axial length of the displaceable member 26, the displaceable member 26 of the displaceable member 26 approximately in the middle of the shafts 44 and 46 in the plane of the second layer 14. Displace at position. Due to the displacement of the displaceable member 26, the second stopper end 42 is displaced with respect to the inlet port 20 so as to open and close the inlet port 20. [0037] Since the relationship between the amount of applied current and the displacement of the displaceable member 26 with respect to the inlet port 20 of the second stopper end 42 is generally non-hysteretic, the amount of applied current is the microvalve 10. It can be controlled to control the level of flowing fluid flow. Controlling the amount of current applied controls the degree of thermal expansion of the ribs 48 and 50, the displacement of the shafts 44 and 46, the rotation of the displaceable member 26, and thus the inlet port 20 of the second stopper end 42. Displacement with respect to is controlled. Thus, the degree to which the stopper end 42 of the displaceable member 26 opens and closes the flow through the inlet port 20, and the resulting fluid flow through the microvalve 10 is by controlling the amount of applied current. Can be precisely controlled. Alternatively, the current applied to the microvalve can be pulsed to open and close the port. The pulse width modulated input signal can achieve the desired overall fluid flow rate for the same average power delivery, i.e., the same heating. [0038] The shape of the inlet port 20 may be similar to the shape of the stopper end 42 of the displaceable member 26, as shown in FIG. Such a shape maximizes the area of the inlet port 20 exposed by a given displacement of the stopper end 42 of the displaceable member 26. That is, the displacement of the stopper end 42 of the displaceable member 26 is minimized so as to expose a given area of the inlet port 20. The outlet port 22 can be of any suitable shape, preferably the maximum size given by the configuration of the microvalve 10, and the effect of the outlet port 22 on the fluid flow between the inlet and outlet ports 20 and 22. To minimize. Needless to say, any other suitable shape and size of the inlet port 20 and the exit port 22 can be utilized, and the inlet port 20 and the exit port 22 can be of different shapes. [0039] After no current is applied to the electrical contacts 32a and 32b, the actuator can be passively cooled, returning the displaceable member 26 to its open position. Alternatively, when two or more actuators are used, the microvalve 10 may be opened with one actuator and closed with another actuator. For example, it is desirable to provide a heat sink (not shown) that contains any thermally conductive metal or ceramic, such as aluminum, and adheres to the bottom surface of the first layer 12. [0040] Alternative configurations of thermal actuators and displaceable members can be readily applied and adopted in the microvalves of the present invention. The thermal actuator and the displaceable member convert the force exerted by the thermal actuator into the displacement of the displaceable member 26. For example, FIG. 4 is a plan view of the alternative actuators 102 and 104 and the displaceable member 26. Each of the actuators 102 and 104 comprises two or more bars 106 connected to the shafts 108 and 110. The shafts 108 and 110, like the shafts 44 and 46 of the actuators 28 and 30, are located on opposite sides of the displaceable member 26 and are offset from each other to exert a rotational force on the displaceable member 26. Obviously, the displacement of each of the shafts 108 and 110 is equal to the thermal expansion of the bar 106, as there is no amplification of the displacement. Additional bars 106 may be provided to increase the force exerted by the shafts 108 and 110 on the displaceable member 26. [0041] In another configuration, as shown in FIG. 5A, two displaceable members 112 and 114 are placed at an angle to each of the actuators 26 and 28 to displace the stopper 116 in the direction indicated by the arrow 118. The angles of the displaceable members 112 and 114 with respect to the shafts 44 and 46, respectively, can be selected to amplify the displacement of the stopper 116. However, by increasing the displacement of the stopper 116, the force at the stopper 116 is reduced because the relationship between the displacement and the force is, of course, also applied here. [0042] Alternatively, as shown in the partial plan view of FIG. 5B, the actuators 118 and 120 may be located on the same side surface of the displaceable member 26. In this configuration, the displaceable member 26 can be displaced to open the inlet port 20 by starting only the actuator 118. By using only one actuator, the force is halved and the displacement of the displaceable member 26 is halved. However, the advantage of this configuration is that the displaceable member 26 can be displaced to close the inlet port 20 by initiating the actuator 120 before the actuator 118 is passively cooled and returned to its initial position. There is. This is in contrast to a configuration in which the actuator is located on the opposite side of the displaceable member and the displaceable member is displaced to its closed position depending on the passive cooling of the actuator. [0043] Figures 6A-6D show other variants of the actuator (s). As shown in FIG. 6A, the displaceable member 26A can be suspended by the elements 29A and 31A. Either or both of elements 29A and 31A may act as inflatable actuators. For example, the actuator element 29A expands towards the displaceable member 26A upon start, and the actuator element 29A displaces the member 26A around the anchor element 31A. If both elements 29A and 31A act as inflatable actuators, then both elements 29A and 31 will inflate the actuator element along the displaceable member, with respect to the middle of elements 29A and 31, around the pivot point. The member 26A is displaced depending on the degree of. Needless to say, even if both elements 29A and 31 act as actuators, any element can be started without starting the other element. Preferably, the elements 29A and 31 define tapered portions 33A and 37A, respectively, to facilitate displacement of the displaceable member 26A. [0044] Alternatively, as shown in FIG. 6B, the displaceable member 26B may float by the element 29B and by the distal end 31B of the displaceable member. Preferably, only the element 29B acts as an inflatable actuator that expands towards the displaceable member 26B during operation and displaces the member 26A around the distal end 31. However, although undesirable, the distal end 31B can also serve as an expansion actuator. Needless to say, even if both element 29B and distal end 31B act as actuators, either element can be started without starting the other element. Preferably, the element 29B defines a tapered portion 33B to facilitate displacement of the displaceable member 26A with respect to the element 29B. Further, the displaceable member 26B preferably also defines a tapered portion 37B at the distal end 31B such that the cross-sectional area of the displaceable member 26B decreases towards the distal end 31B. [0045] Next, referring to FIG. 6C, the displaceable member 26C may be suspended by one actuator 29C with extension arms 39 and 41. The extension arms 39 and 41 have different cross-sectional areas, for example, the cross-sectional area of the extension arm 39 is smaller than the cross-sectional area of the extension arm 41. Due to the difference in cross-sectional area, the resistance of the extension arm 39 is greater and the thermal expansion at start-up is greater than the thermal expansion of the extension arm 41. Therefore, upon starting the actuator 29C, the displaceable member 26C is displaced farther by the extension arm 39 as compared to by the extension arm 41, causing the member 26C to be linearly displaced in direction D43 and the extension arm 41. And rotate around the pivot near the intersection of the deformable member 26C. Although not shown, either or both of the extension arms 39 and 41 provide a taper portion to facilitate displacement of the deformable member 26C. In this variant, the extension arms 39 and 41, as well as the deformable member 26C, are doped to allow the application and flow of current through them. [0046] As shown in FIG. 6D, the displaceable member 26D can be suspended by two actuators 29D and 31D located on either side of the member 26. Actuator 29D comprises extension arms 39'and 41' having different cross-sectional areas such that the cross-sectional area of the extension arm 39'is smaller than the cross-sectional area of the extension arm 41'. Similarly, the actuator 31D comprises extension arms 39'' and 41'' having different cross-sectional areas such that the cross-sectional area of the extension arm 39'' is smaller than the cross-sectional area of the extension arm 41''. .. Extension arms 39 and 39'and / or extension arms 41 and 41'may or may not have the same cross-sectional area. As mentioned above, due to the difference in cross-sectional area, the resistance of the extension arms 39'and 39'' is higher, and the thermal expansion at start-up is greater than the thermal expansion of the extension arms 41'and 41'', respectively. large. [0047] In addition, the actuators 29D and 31D preferably have the extension arm 41'closer to the extension arm 41'' than the extension arm 39'' and the extension arm 41'' to the extension arm 41'than the extension arm 39'. Arranged so that they are close to each other. Therefore, upon starting the actuators 29D and 31D, the displaceable member 26D is displaced farther by the extension arms 39'and 39'' compared to by the extension arms 41'and 41'', and the member 26D Rotates around a pivot approximately halfway between actuators 29D and 31D. Although not shown, any or all of the extension arms 39', 39', 41' and 41'' are provided with tapered portions to facilitate displacement of the deformable member 26C. Similar to the modified example shown in FIG. 6C, the extension arms 39', 39'', 41'and 41'', and the deformable member 26D are doped to allow the application and flow of current through them. [0048] In yet another embodiment of the invention, one or more sensors (not shown) may be integrally secured to the displaceable element. The sensor can be, for example, a device such as a piezo resistor that changes its electrical properties in the event of a change in stress within the displaceable member if it bends during displacement. The piezoresistive can be located on the surface of the side surface of the rib. Changes in the electrical properties of the piezo resistor can be utilized to sense the displacement or movement of the displaceable member. [0049] Obviously, of the many other configurations of thermal or other types of actuators, such as piezoelectric actuators, electrostatic actuators, or barometric actuators, those most suitable for integrated and displaceable members are of the invention. Easily applied and adopted in the microvalve 10, displacement of the second stopper end 42 can be achieved. For example, one of the two actuators of the microvalve in FIG. 1 can be displaced by a single beam that holds and / or rotates displaceable members. The displaceable member can therefore rotate about the center of rotation, or along the displaceable member, around a pivot approximately midway between the fixed beam and the shaft of the actuator. Therefore, the displaceable member may be displaceable between the open and closed positions by thermal initiation of the actuator alone. [0050] Next, with reference to FIG. 7, a cross-sectional view of an alternative embodiment is shown in which the third layer 16 of the microvalve 10'defines a second inlet port 52 that balances the pressure of the fluid. .. The inlet port 20 and the second inlet port 52 are therefore such that the fluid collides with the opposite surface of the stopper end 42 when the displaceable member 26 is in the closed or open position and between the closed positions. , The fluid can be introduced into the cavity. This compensates or balances the fluid pressure exerted on the displaceable member obtained from the fluid entering the cavity 24, at least in part. The fluid pressure exerted on the stopper end 42 is such that the microvalve 10'is located in the area perpendicular to the inlet port 20 so that the microvalve 10'is placed in the dosed position or between the open and closed positions. Occurs in some cases. As the fluid flows through the inlet port 20, the fluid collides with the surface of the stopper end 42 adjacent to the first layer 12, exerts pressure, and enters the cavity 24. The displaceable member 26 of this preferred embodiment can withstand a fluid pressure of about 300-500 psi when the displaceable member 26 is made of silicon, thus exerting a fluid on the stopper end 42 of the displaceable member 26. It is desirable to compensate for the pressure. However, full compensation is unnecessary because the inherent strength of materials such as silicon can easily withstand pressure imbalances, which are relatively small. Therefore, by providing a second inlet port 52 facing each other, the microvalve 10 can withstand a fluid pressure of several thousand psi. [0051] FIG. 8 is a cross-sectional view of another alternative embodiment including an inlet channel 56 and a second inlet port 52'that compensate or balance the vertically incident fluid force on the displaceable member 26. The inlet channel 56 extends through the first, second, and third layers 12, 14, and 16 and directs flow to the cavity 24 through the second inlet port 52. Thus, the fluid can be introduced through the first layer 12 of the microvalve 10 and can be directed to be introduced into the cavity 24 from the opposite direction. A fourth layer 54 is provided, covering the entrance passage 56 over the third layer 16 and arranging the third layer between the second layer and the fourth layers 14 and 54. .. [0052] Next, referring to the plan view of FIGS. 9 to 13, in addition to the fluid pressure due to the fluid collision force, the stopper end 42 of the displaceable member 26 also has a local flow of the inlet fluid flow face 58. Exposed to physical fitness as well as fluid displacement. The local force of face 58 is caused by the bending of the flow path as the fluid enters the cavity 24 through the inlet port 20 and / or the second inlet port 52. This force stimulates the displaceable member 26 to displace in the direction directed by the fluid force. Therefore, it is also desirable to compensate for the reduction of local forces as well as the fluid perturbation in the same plane, as well as the movement of the displaceable member. [0053] Figures 9 to 13 show various methods and configurations for obtaining compensation for fluid force. Each of the embodiments shown in FIGS. 9 to 13 includes a fluid force coupling surface in which the fluid flowing from the inlet port 20 to the outlet port 22 collides and exerts a coupling force exerted by the fluid on the displaceable member 26. The binding force is the collision of the fluid flow with a different surface than the face 58 (the embodiments shown in FIGS. 9 and 10), or the redirection of the fluid flow towards the face 58, or at least perturbation (FIG. 9). It is caused by any of the embodiments) shown in 11 to 13. [0054] As shown in FIG. 9, the displaceable member 26a further comprises a normally U-shaped expansion portion 60 to form a P-type displaceable member 26a. The U-shaped extension 60 at least partially surrounds or seals the exit port 22. The fluid flow exerts a force on the U-shaped extension 60 to at least partially compensate or balance the local force on the surface 58 of the stopper end 42. The U-shaped extension 60 further surrounds the fluid flow between the stopper end 42 of the displaceable member and the extension 60, thereby also reducing fluid leakage in the microvalve. The end application further allows the pressure outside the enclosure to be relatively constant, resulting in lower or no net pressure from the area outside the enclosure. Other suitable shapes of the extension 60, such as the L-shaped displaceable member, may be utilized to form an h-shaped displaceable member (not shown). [0055] Alternatively, as shown in FIG. 10, the displaceable member 26b may include an extension 62 that is located between the inlet 20 and the outlet 22 rather than surrounding the outlet 22. The fluid flow so that the fluid flow exerts a force on the extension 62 to compensate or balance the local force on the face 58 of the stopper end 42, at least in part. To reorient. [0056] As shown in FIGS. 11-13, one or more members secured to the first layer 12 and / or the third layer 16 are provided in the cavity 24 in an alternative or additional manner to provide a stopper. Compensate or balance the local force on the face 58 of the end 42, at least in part. As shown in FIG. 11, the member 64 may be provided in the cavity 24 and the displaceable member 26a may include an extension 60 that surrounds the fluid flow therein. The combination of the extension 60 and the member 64 can compensate for the force without redirecting the fluid flow at the displaceable member 26a. Alternatively, the baffle 66 may have a curved surface, as shown in FIG. 12, redirecting the flow towards the face 58, thereby compensating for the force. FIG. 13 shows a microvalve with a curved baffle 68 and a baffle 70 that reorients the flow around the baffle 70 to compensate for or balance the fluid force. [0057] FIG. 14 is an exemplary drawing of yet another embodiment of the invention in which the angled outlet 72 and the angled inlet 20a serve as fluid flow compensating members. The angled exit 72 and the angled inlet 20a are shown in FIG. 14, as partially defined by the other layer 73, but only by the first layer 12. May be good. The fluid flows into the cavity 24 through the angled inlet 20a at the inlet angle α in the direction indicated by the arrow 74, and the fluid flows out of the cavity 24 at the outlet angle φ in the direction indicated by the arrow 76. The inlet angle α is controlled by the displacement of the stopper end 42 of the displaceable member 26, but the outlet angle φ is generally constant. The fluid inlet and outlet angles α and φ are selected to balance the fluid forces. Therefore, the fluid outflow flow at the outflow angle φ exerts a force that balances the force exerted by the fluid inlet flow at the inlet angle α. [0058] [0058] Next, with reference to FIG. 15, a cross-sectional view of still another alternative embodiment of the microvalve 78 according to the present invention is shown. The micro valve 78 can be used as a gas valve. For the gas valve 78, the distance between the recess 18 and the stopper end 42 when the displaceable valve is in the closed position is preferably less than 0.5 μm for the fluid microvalve 10. Gas leakage is prevented or minimized by minimizing the spacing, as gas leakage is not reduced by fluid tension, as is the case with fluid valves. The magnitude of the spacing is reduced by reducing the depth of the recesses 18 of the first layer 12 and / or the third layer 16. [0059] Additional or alternative, the magnitude of the spacing can be further reduced by providing a flange 80 on the inlet surface of the stopper end 42 of the displaceable member 26. The flange 80 strengthens the seal between the inlet surface of the stopper end 42 and the inlet 20b when the displaceable member 26 is in the closed position. Preferably, the inlet channel 79 is provided through the first and second layers 12 and 14 to allow the gas flow to flow into the cavity 24 in a direction parallel to the plane of the second layer 14. Orient to cavity 24 through 20b. The inlet port 20b is defined along the side wall of the cavity 24, which is generally perpendicular to the plane of the second layer 14. Thus, the displacement that closes the inlet 20 of the displaceable member 26 in the plane of the second layer 14 also causes the flange 80 to form a better seal to the inlet 20. [0060] The manufacture of the microvalve of this embodiment of the present invention includes a silicon melt bond and a melt bond such as deep reactive ion etching (DRIE). [0061] The melt bond allows the bond of one silicon layer to the other silicon layer, forming a single mechanical structure. The melt bond has been proven to be at the molecular level and provides very high mechanical stiffness. The melt bonding technique is well known. For example, "Surface Micromachined Structures Fabricated with Silicon Fusion Bonding" by KEPetersen, D, Gee, F.Pourahmadi, R.Craddock, J.Brown and L.Christal (Proceedings, Transducers91, June 1992, pp. 397-399). checking ... This document is expressly incorporated herein by reference. [0062] The process of making a silicon microstructure according to a preferred embodiment of the present invention will be described with reference to FIGS. 16a-f. This embodiment uses three silicon wafers. Using three silicon wafers, this process forms a single crystal silicon structure (SCS) microstructure defined as an integral part of the second wafer (corresponding to the second layer 14). The first wafer (corresponding to the first layer 12) and the third wafer (corresponding to the second layer 16) function as carriers of the second wafer. Alternatively, the carrier can be formed, for example, from glass (Pyrex). Although the discussion below only refers to three wafers, it is of course understood that these principles can be applied to the formation of microstructures containing stacks of two or more wafers. [0063] In FIG. 16a, the first wafer is patterned with a photoresist, defining a recessed region formed therein, and the recessed region is, for example, plasma etching, wet etching with KOH or other silicon etchant. Alternatively, it is formed using standard semiconductor techniques such as differential oxide growth. The recessed region can have any shape and, for example, any required depth of less than 0.1 μm to over 100 μm. In this embodiment, this recessed area is about 1 μm deep. [0064] It should be understood that the recessed area does not have to be a single uniform depth. For example, some standard silicon etching steps can be used to produce several different depths that can be used for different mechanical functions. Alternatively or further, as described above, it should be further understood that the second layer is indented (not shown) from the first layer 12 and the third layer 16 and provides clearance between them. Is. Moreover, each of the first wafer surface and the third wafer surface can be either bare silicon or coated with an oxide layer. Further, the substrate of the recessed region can be either bare silicon, silicon oxide, doped silicon, or can be coated with any other thin film that can withstand the next wafer bonding and temperature treatment steps. [0065] As shown in Figure 16b, the inlet port is then etched through the first wafer. Although not shown, the outlet port can be etched through the first wafer at the same time. Alternatively or further, the outlet port can be etched through a third wafer. [0066] In FIG. 16c, the patterned surface of the first wafer is bonded (preferably doped) to the second wafer by a silicon melt bonding (ie, direct bonding) process. The melt bonding technique is well known. For example, "Surface Micromachined Structures Fabricated with Silicon Fusion Bonding" by KEPetersen, D, Gee, F.Pourahmadi, R.Craddock, J.Brown and L.Christel (Proceedings, Transducers91, June 1991, pp. 397-399). checking ... This document is expressly incorporated herein by reference. In the preferred melt bonding technique of the present invention, the first and second wafers are made hydrophilic. That is, they are treated with chemicals (eg, hot nitric acid or hot sulfuric acid, and hydrogen peroxide solutions or other strong oxidants, which allow water to adhere to them). After drying, the next two wafers are placed in an oxidizing atmosphere at a temperature of 400 ° C to 1200 ° C for about 1 hour. [0067] The silicon melt bonding technique described above couples the first and second wafers together without the use of intermediate adhesive materials that may have different coefficients of thermal expansion than single crystal silicon wafers. Further, the melt bonding is carried out by forming an oxide layer or a nitride layer on the bonded surface of one or both of the wafers. [0068] As an alternative to melt bonding, for example, the first wafer and the second wafer can be bonded together with an adhesive such as a photoresist. As another alternative, the first and second wafers are used by alloying the wafers with each other, with most of their surface covered with a metal layer such as gold. When a glass carrier is used in place of the first silicon wafer, the second wafer can be anodic bonded to such a glass carrier. [0069] If necessary, the second wafer is thinned and polished to the thickness required for the particular application. Alternatively, electrochemical etching (ECE) can be used to thin the wafer. The diffusion heater can be introduced into the flat surface of the second layer 14 by diffusion. In addition, any required circuit or other thin film deposits and patterning process can be performed using standard silicon processing techniques. [0070] The second wafer is then patterned for deep reactive ion etching (DRIE), which defines the area of the wafer to be etched. DRIE technology is becoming more and more popular. For example, AAAyon, CCLin, RABraff and MASchmidt, "Etching Characteristics and Profile Control in a Time-Multiplexed ICP Etcher", Proceedings of Solid State Sensor and Actuator Workshop, Hilton Heat Island, "Highly Anisotropic Selective Reactive Ion Etching of Deep Trenches in" Silicon, Microelectronic Engineerring, Vol. 23, 1994, pp. 373-376; C.Linder.T.Tschan, NFde Rooij, "Deep Dry Etching Techniques as a New IC Compatible Tool for Silicon Micromachining," Proceedings, Transducers' 91, June 1991, pp. 524-527; CDFung and JR Linkowski, "Deep Etching of Silicon Using Plasma", Proceedings of the Workshop on Micromachining and Micropackaging of Transducers, November 7-8, 1984, pp. 159-164 Also, JWBartha, J.Greeschner, M.Puech and P.₆/ O₂, Microelectronic Engineering, Vol. 27, 1995, pp. 453-456. Reactive ion etching equipment now allows etching of very deep holes or grooves (deeper than 100 microns), while at the same time having a high aspect ratio (ratio between the depth of the etched region and the width of the etched region). To maintain. It has been found that this device is capable of an aspect ratio of at least 30: 1 for grooves as deep as 300 microns. [0071] In essence, DRIE involves a synergy between chemical etching and ionic impact. The impacted ions chemically react with the silicon surface. The DRIE process advantageously etches at a much faster rate (ie, anisotropic) in the vertical direction than in the lateral direction, regardless of the crystal plane or crystal orientation of the silicon. As a result, relatively deep, substantially vertical grooves or slots are formed in the second wafer of single crystal silicon (SCS). These substantially vertical grooves or slots can be formed anywhere on the second wafer, regardless of the crystallographic orientation within the wafer. As a result, high aspect ratio structures such as capacitive or electrostatic plates can be formed, and arbitrary contour structures such as circles, ellipses and helices can be formed. [0072] As shown in Figure 16d, the DRIE process is used to be fully etched through the second wafer, defining displaceable members and actuators. The DRIE etching process mechanically releases the single crystal silicon (SCS) microstructure formed on the second wafer, which then moves with respect to and on the surface of the second wafer. Will be free. Floating plate / beam structures with an aspect ratio (height / width) greater than or equal to 20: 1 are manufactured using the DRIE process described below. [0073] The inductively coupled plasma source etches silicon using a photoresist or silicon dioxide as a mask. The polymerization of the source gas on the side wall of the etched groove slows down the lateral etching rate, allowing high anisotropy. This etching chemistry is SF at 50 Milittle, for example.₆Is. Oxygenated gases and fluorinated gases commercially available from Surface Technology Systems help provide high Si / photoresist etching rate ratios. The 6 micron photoresist acts as a patterning mask. The photoresist sensitivity is about 50: 1, which allows etching to a depth of 300 μm with a resist of about 6 μm. A "multiplex RIE system" commercially available from Surface Technology Systems (STS) (Redwood City, California) can be used to perform inductively coupled plasma DRIE. Alternatively, it is commercially available from Plasma Therm in St. Petersburg, Florida. [0074] The combination of melt coupling and DRIE allows the construction of three-dimensional structures, such as the microvalves of the present invention. For example, EHKlaassen, K.Petersen, JMNoworoski, J.Logan, NIMaluf, J.Brown, C.Storment, W.McCulley and GTAKovacs, "Silicon Fusion Bonding and Deep Reactive Ion Etching; A New Technology for Microstructures", Proceeding, See Transducers95, Stockholm, Sweden, 1995, pp. 556-559. [0075] In FIG. 16e, the patterned surface of the third wafer is bonded to the second wafer by a silicon melt bonding (ie, direct bonding) process, as described above with reference to FIG. 16c. Although not shown, prior to bonding, the third wafer was treated in the same manner as the first wafer, which defines recessed areas, inlet and / or outlet ports, and through wafer contact holes or vias. [0076] As shown in FIG. 16f, layers of conductive material such as aluminum are formed on the surface of a contact hole or via, for example by sputtering, the surface of a second wafer exposed through the contact hole, and the third wafer. It is deposited on at least a part of the outer flat surface of the. The conductive layer thus forms a bond between the pad and the actuator that allows electrical contact. Any required circuit or other thin film deposition and patterning process can be performed on the third wafer using standard silicon processing techniques. [0077] A large number of modifications can be easily introduced into this process. For example, the first layer 12 and / or the third layer 16 can be made from glass (Pyrex) instead of silicon. Microvalves can be 3 or more wafers or layers or<u style="single"></u>Micromechanical devices can be formed from two or more wafers or layers. Further, shallow cavities can be defined in the second layer 14 in place of or in addition to the first layer 12 and the third layer 16. Alternatively, each layer can be processed separately and then assembled in a coordinated bonding process. As will be appreciated, those skilled in the art can easily make these and many other modifications to the manufacturing process, for example by simply changing the layout. [0078] The microvalves of the present invention may be adapted for use in antilock braking systems, inkjet printing, refrigeration, pilots for large valves (eg automatic transmissions and large industrial valves) as described below. [0079] Now, with reference to FIG. 17, the microvalve 82 may be further adapted to selectively control two inlet ports 84, 86 for fluid to flow to one outlet 22. The openings and closures of the inlet ports 84 and 86 are interdependent. Other details of the microvalve 82 will be understood from the above description with reference to other figures. In particular, when activated, the microvalve 82 may be controlled to open the inlet valve 84 while keeping the inlet valve 86 closed. The reverse is also true. The micro valve 82 can also be controlled to partially open both the inlet valves 84 and 86. Therefore, the microvalve 82 can be used to select a fluid flow from one or two fluid sources. [0080] [0080] As is clear, a number of other fluid flow control integrations can be achieved with the microvalves of the present invention. For example, the single integrated microvalve 87 of the present invention can be used to replace the normally open (NO) and normally closed (NC) solenoid valves used in each wheel of conventional antilock braking systems. A partial schematic of the antilock braking system 100 utilizing a single integrated microvalve and a partial schematic of such a single integrated microvalve 87 are shown in FIGS. 18 and 19, respectively. [0081] Anti-lock braking systems (ABS) have become very common in passenger cars. As shown in FIG. 18, the antilock braking system 100 generally has a wheel speed sensor (not shown) that senses the speed of the wheel 102, the flow of brake fluid to and from the brake caliper 104 of the wheel 102. Normally open (NO) valve 86 and normally closed (NC) valve 84 for control, electronic control unit (ECU) 106, master cylinder 108, which receives input from the wheel speed sensor and outputs a signal to micro valve 87, Also equipped with a pump 110. As shown in FIG. 19, the microvalve 87 defines one outlet port 22 that directs the brake fluid to the brake caliper, and two displaceable members 88, 90, with two inlet ports 84, 86 selective, respectively. Opens and closes. [0082] The normally open inlet 86 allows the brake fluid to flow from the master cylinder 108 to the brake caliper 104 when the driver applies force to the brake pedal 112. Without the ABS system operating, the normally closed inlet 88 is at least substantially closed to the flow of break fluid, and the normally open inlet 86 is when the driver applies force to the brake pedal 112, Allows the flow of break fluid to the brake caliper 104. [0083] However, on slippery roads, there can be inadequate friction or grip between the tires and the road, and when the driver applies force to the brake pedal 112, the brake caliper 104 locks the wheel 102. When the brake caliper 104 locks the wheel 102 and the wheel 102 stops spinning, the wheel 102 skids along a slippery road surface, increasing the braking distance of the vehicle. Therefore, the car keeps moving because of its momentum. In essence, wheel locking by the braking system occurs when the grip between the tire and the road surface is less than the restraining or grip between the wheel and the brake pads. [0084] The antilock braking system reduces the stopping force to a level equal to the grip between the wheel and the road surface by controlling the brake fluid pressure applied until the appropriate level of stopping force is reached. By), reduce or eliminate wheel locking problems. The antilock braking system is activated in response to a wheel speed sensor that senses that the wheel is about to lock. When the antilock braking system is activated, the computer closes the NO valve. When the wheel speed sensor continues to detect the wheel trying to lock even after the NO valve is closed, the computer opens the NC valve and pumps some brake fluid from the wheel cylinder or caliper to the master cylinder. The NC valve repeatedly opens and closes to control the flow of brake fluid until the computer determines that the brakes are in control (ie, when the wheel speed sensor senses that the wheels are no longer trying to lock). Fluid flow control is proportionally achieved by repeated opening and closing of the NC valve. The antilock braking system is achieved only when the brake pedal is continuously depressed. [0085] Traditional anti-lock braking systems in automobiles or passenger cars use two solenoid valves per wheel to control the flow of brake fluid, so we decided to use eight solenoid valves for a typical four-wheeled passenger car. Become. However, the use of solenoid valves has some drawbacks as described above. Proportional solenoid valves are available, but the cost-effective solenoid valves used in antilock braking systems simply switch on and off (binary) so that you get the exact desired level of flow control. It is necessary that the valve be pulsed. Such pulses can be sensed by the driver while the brake pedal is being depressed, which may not be desired. [0086] The microvalve 87, as shown in FIG. 19, can be used to replace the two solenoid valves of a conventional antilock braking system. The displaceable members 88 and 90 are separately controlled by their respective thermal actuators. However, in the antilock braking system, the displaceable members 88 and 90 are not both in the open position at the same time. Therefore, during normal braking operation, while the displaceable member 88 (in the normally closed position with respect to the inlet 84) is in the closed position, the displaceable member (in the normally open position with respect to the inlet 86). 90 is in the open position. Thus, when the driver steps on the brake pedal 112 during normal braking operation, the pump 110 pumps the brake fluid from the master cylinder 108 to the brake caliper 104 through the normally open inlet 86. [0087] When the ECU 106 senses that the wheel 102 exceeds a predetermined threshold, the ECU 106 sends a signal to the microvalve 87 to displace the displaceable member 90 to a closed position with respect to the inlet 86, and the displaceable member 88. Is displaced from its closed position to a position between the open and closed positions with respect to the inlet 84. Displacement of the displaceable member 88 to a position between the open and closed positions with respect to the inlet 84 allows the pump 110 to remove the desired level of brake fluid from the brake caliper 104 to the master cylinder 108. Is possible. Therefore, only one integrated microvalve is used to replace the two conventional binary solenoid valves. [0088] The displaceable member 88 is preferably not pulsed between open / close positions and is preferably displaced between open / close positions that precisely control the desired amount of brake fluid pumped from the brake caliper 104. However, the displaceable member 88 can be pulsed between open and closed positions relative to the inlet 84 to achieve proportional fluid flow control. [0089] Although specific embodiments of the invention are described and illustrated, it is understood that modifications of these embodiments can be made without departing from the intent of the invention. Therefore, the present invention is intended to be defined in the claims described above. [Simple explanation of drawings] FIG. 1 is an exploded perspective view of the first, second and third layers of the proportional microvalve of the embodiments presented by the present invention. FIG. 2 is a cross-sectional view taken along the line 3-3 of FIG. FIG. 3 is a plan view of an actuator having a plate or rib. FIG. 4 is a plan view of an alternative structure of the actuator. FIG. 5A is a plan view of an alternative structure of the actuator. FIG. 5B is a plan view of an alternative structure of the actuator. FIG. 6A is a plan view of an alternative structure of the actuator. FIG. 6B is a plan view of an alternative structure of the actuator. FIG. 6C is a plan view of an alternative structure of the actuator. FIG. 6D is a plan view of an alternative structure of the actuator. FIG. 7 is a cross-sectional view of a microvalve with a second inlet port through which fluid enters from the opposite side to balance pressure. FIG. 8 is a cross-sectional view of another microvalve with a second inlet port through which fluid enters from the opposite side to balance pressure. FIG. 9 is a partial plan view showing a displaceable member including an extension portion that compensates for fluid force. FIG. 10 is a partial plan view showing a displaceable member including an extension portion that compensates for fluid force. FIG. 11 is a partial plan view showing a microvalve of the invention, further including one or more baffles and extensions that reorient the fluid flow. FIG. 12 is a partial plan view showing a microvalve of the invention, further including one or more baffles and extensions that reorient the fluid flow. FIG. 13 is a partial plan view showing a microvalve of the invention, further including one or more baffles and extensions that reorient the fluid flow. FIG. 14 is a partial cross-sectional view of a microvalve with an angled outlet port. FIG. 15 is a cross-sectional view of the proportional gas microvalve of the present invention. FIG. 16a is a diagram showing a flow of a manufacturing process according to the present invention. FIG. 16b is a diagram showing a flow of a manufacturing process according to the present invention. FIG. 16c is a diagram showing a flow of a manufacturing process according to the present invention. FIG. 16d is a diagram showing a flow of a manufacturing process according to the present invention. FIG. 16e is a diagram showing a flow of a manufacturing process according to the present invention. FIG. 16f is a diagram showing a flow of a manufacturing process according to the present invention. FIG. 17 is a plan view of a displaceable member that opens and closes two inlet ports that control fluid flow to the outlet port. FIG. 18 is a schematic diagram of a portion of an antilock braking system that is an example of valve utilization. FIG. 19 is a plan view of a microvalve having two displaceable members that independently open and close two inlet ports that control fluid flow to the outlet port.