The present invention concerns a method of manufacturing multi-level, silicon, micromechanical parts and the parts thereby obtained, used in particular within the field of the watchmaking industry.
Current techniques for manufacturing silicon, micromechanical parts allow the verticality of the sides of the parts to be sufficiently well controlled for the part to be used as a timepiece mechanism component. However, it remains very difficult to manufacture multi-level parts from a single silicon wafer, using current techniques.
An alternative method of producing multi-level parts is proposed here. This method discloses how to obtain multi-level, silicon parts solely from single level silicon parts using current methods.
It is thus an object of the present invention to provide a simple and economic solution for manufacturing multi-level, micromechanical parts from silicon wafers available on the market, said parts, whether they be fixed or mobile, being used in particular for making a timepiece movement.
The invention therefore concerns a method consisting in:
This method structures each element of the final part with precision, using proven techniques for wafers whose thickness is less than 1 mm.
In step c) various positioning means are possible to ensure precise positioning of the two elements that will form the final part, prior to performing heat oxidation to connect the two elements physically to each other. In the following detailed description, two examples are given of positioning means, which rely either on marking means, which are formed in each wafer at the same time that the elements are structured, with the elements then being linked by bridges of material; or by using a template for positioning the wafers on top of each other or elements that are separate from the wafers.
Step d) of oxidising the interface between the wafers can be performed by a dry method with dry oxygen, or by a wet method with water vapour, for example in a furnace heated to a temperature of between 900° C. and 1200° C. in accordance with techniques that are well known to those skilled in the art and reported for example in the work, “Semiconductor devices: physics and technology—editions John Wiley & Sons, ISBN 0-471-874-8-01.01.1985 p. 341-355”.
This oxidation can also be performed by any other method known to those skilled in the art, for example by laser amplified oxidation.
It is also possible to obtain a part that has more than two levels by machining a third silicon wafer that can be integrated in step c) of the method, or after step e) by repeating steps c), d) and e).
The method according to the invention makes it easy to manufacture multi-level, silicon, micromechanical parts, such as bearings, plates or slotted bridges, reduction wheel sets for a timepiece movement, etc. . . .
Other features and advantages of the present invention will appear more clearly in the following description of example embodiments, given by way of non-limiting illustration, with reference to the annexed drawings, in which:
In a first step a) the first element 3, but preferably a plurality of said first elements 3, are machined in a first silicon wafer 1. In fact, starting with a commercially available silicon wafer, whose diameter may vary from between 75 mm to approximately 300 mm, it is possible to structure a plurality of elements having a diameter of less than 2 mm. The thickness of this wafer is generally less than 1 mm, which is compatible with timepiece dimensions, but this thickness may be brought to the precise desired dimension by etching. Preferably, it will be ensured that one surface of the wafer remains unpolished, whether the wafer is sold like this or subjected to a slight abrasive treatment.
Elements 3 are structured in wafer 1 by methods known to those skilled in the art by photolithography and etching through the mask formed. The etching can be carried out in accordance with various methods that are also known within the field of silicon machining. The method that provides the best aspect ratio, i.e. sides cut perpendicular to the surface of the part, is the RIE technique (Reactive Ion Etching).
Since this technique, and the adaptations necessary for each particular case, is well known to those skilled in the art, it will not be described further here.
In this first step, bridges of material 5 are provided to hold elements 3 connected to their support wafer 1. This structuring also forms elements in wafer 1 for positioning said wafer 1 relative to wafer 2. By way of example, these positioning means are formed by two bores 6, with the reference 7 designating other positioning means that will be explained below.
It is clear that bridges of material 5 between the elements of each wafer are not necessarily required depending upon the position used.
In step b), the same process as in step a) is used from a second wafer 2 to form second elements 4, namely the “endstone”, and positioning means 6, which coincide with positioning means 6 of the first wafer. In this example, elements 3 and 4 have the same diameter.
When wafers 1 and 2 used in steps a) and b) have at least one unpolished surface, they can be assembled without any prior surface treatment by placing the unpolished surfaces against each other. If the two surfaces of each wafer are polished, a prior surface treatment will preferably be carried out to make the surfaces that will have to be placed against each other either slightly rough, or coated with a silicon oxide film that is slighter thicker than the native oxide.
Step c), as shown in the left part of the diagram, consists in positioning the two wafers 1, 2 one on top of the other and securing this position for example by means of pins 16 to form an assembly 11.
The right part of the diagram illustrates a second positioning example. The second elements 4 are first of all separated from second wafer 2 by breaking bridges of material 5, or they are obtained without any bridges of material 5, then positioned in corresponding slots of a quartz template 18. The first wafer 1 is then applied by being positioned using a slot 7 machined in the edge thereof, said slot 7 engaging in a snug 17 of template 18 to form an assembly 13. By using a template 18 that has slots of suitable depth, the first elements 3 can also be separated from the first wafer 1 to be placed in template 18.
In step d) a series of assemblies 11 or 13 is placed on a carriage 15, which is introduced into a furnace 8 heated to a temperature of between 900° C. and 1200° C. and in which an oxidising mixture 9 circulates for around 2 to 4 hours. This oxidising step, which will connect the two wafers 1 and 2 by siloxane type connections can be performed, either by a dry method with a mixture containing an inert carrier gas and dry oxygen, or by a wet method replacing oxygen with water vapour.
At the end of this step, carriage 15 is removed from furnace 8, assemblies 11 or 13 are allowed to return to ambient temperature, then the parts 10 thereby formed are separated by breaking any residual bridges of material 5. The bearing 10 thus obtained is then ready to be used by being mounted in a setting 12, to receive a pivot 14 of a timepiece movement wheel set.
The examples that have just been described concern the manufacture of a micromechanical part with two levels, but it is clear that there could be a larger number of levels. In order to do this, one need only start with three or more silicon wafers, or repeat the method at the end of step e).
The invention is not limited to the examples that have just been described and those skilled in the art can implement the method to obtain other micromechanical parts.
The invention is not limited to parts of small dimensions either. It may advantageously be implemented for parts of larger dimensions, such as timepiece movement plates or bridges including recesses. The thickness of such parts is of the order of 2 mm and they could not be made from a single commercially available silicon wafer.
Filing Document | Filing Date | Country | Kind | 371c Date |
---|---|---|---|---|
PCT/EP07/61800 | 11/1/2007 | WO | 00 | 5/11/2009 |