EP3907565A1 - Method for manufacturing a silicon timepiece component - Google Patents

Abstract

The method of manufacturing a silicon watch component according to the invention comprises the following steps: a) providing a wafer (1) comprising a first layer of silicon (2), a second layer of silicon (3) and , between the two, an intermediate layer of silicon oxide (4); b) etching the first layer of silicon (2) to form therein the watch component (5), at least one connecting element (7) located in a recess (6) of the watch component (5) and at least one fastener (8) ) connecting this connecting element (7) to an interior surface (6a; 12a) of the watch component (5); c) removing the intermediate layer of silicon oxide (4) between the watch component (5) and the second layer of silicon (3) and leaving it at least partially between the connecting element (7) and the second layer of silicon (3); d) applying at least one treatment to the timepiece component (5); e) detaching the watch component (5) from the plate (1) by breaking or removing the clip (8).

Description

The present invention relates to a method of manufacturing a watch component in silicon.

Silicon watch components are generally manufactured by deep reactive ionic etching - a technique also known by the English abbreviation DRIE - of a wafer of silicon-based material. The wafer can be a silicon wafer that is etched in its entire thickness (see for example the patent applications EP 1722281 , EP 2145857 and EP 3181938 ) or a silicon on insulator substrate known as SOI (silicon on insulator) comprising an upper layer of silicon and a lower layer of silicon linked by an intermediate layer of silicon oxide, the upper layer of silicon being that in which the etching is carried out (cf. for example patent applications WO 2019/180177 and WO 2019/180596 ). Compared to a simple silicon wafer, the silicon-on-insulator substrate has the advantage of having a rigid support (the lower layer of silicon, of greater thickness than the upper layer) facilitating its handling and its maintenance and a layer of stop (the intermediate layer of silicon oxide) making it possible to stop the etching.

Regardless of the type of wafer used, multiple components are etched simultaneously into the same wafer, and clips or bridges left during etching keep the components attached to the wafer for other stages of manufacturing. The components are then released from the wafer by breaking or removing the fasteners.

Such fasteners, which connect the periphery of each component to the wafer, can pose a problem, in particular when the periphery of the component is a functional surface which must not see its function disturbed by fastener residues or when the external surface of the component must present a particularly neat appearance, as for example in the case of a needle. In certain cases, also, in particular for components bearing a micro-toothing, the functional external surface does not have free space of sufficient size to be able to insert a sufficiently robust fastener therein.

The patent application WO 2019/166922 proposes a method of manufacturing a watch balance-spring according to which one is provided with a silicon substrate carrying a layer of silicon oxide, one forms through holes in the layer of silicon oxide, one grows by epitaxy a layer of silicon on the layer of silicon oxide, this layer of silicon filling the through holes to form fasteners or bridges of material, balance springs are etched in the layer of silicon, the layer of silicon oxide is removed, balance springs remaining attached to the silicon substrate by said fasteners, the balance springs are subjected to heat treatments and finally the balance springs are detached from the silicon substrate.

With such a method, the spirals remain bonded to the substrate after etching by clips extending out of the plane of the spirals rather than between the outer surface of the last turn and the etching silicon layer as is generally the case. However, this process does not allow the use of commercially available silicon-on-insulator substrates, and growing the silicon layer in which the balance-springs will be formed by epitaxy is a complicated operation.

The present invention aims to remedy the aforementioned drawbacks, or at least to mitigate them, and to this end proposes a manufacturing process according to claim 1, particular embodiments being defined in the dependent claims.

• la figure 1 montre par des vues en coupe schématiques les différentes étapes du procédé de fabrication selon l'invention ;
• la figure 2 montre en vue de dessus un composant horloger gravé dans une couche supérieure de silicium d'une plaquette silicium sur isolant selon un premier exemple de réalisation. Sur cette figure, les traits blancs représentent les vides créés dans la couche supérieure de silicium par la gravure et les zones grises représentent le silicium de la couche supérieure. En plus du composant horloger sont représentés un élément de liaison et des parties sacrificielles gravés dans la couche supérieure de silicium, l'élément de liaison servant à maintenir le composant horloger attaché à la plaquette ;
• la figure 3 montre en vue de dessus le composant horloger de la figure 2 attaché à la plaquette après suppression des parties sacrificielles ;
• la figure 4 montre en vue de dessus un composant horloger gravé dans une couche supérieure de silicium d'une plaquette silicium sur isolant selon un deuxième exemple de réalisation. Sur cette figure, les traits blancs représentent les vides créés dans la couche supérieure de silicium par la gravure et les zones grises représentent le silicium de la couche supérieure. En plus du composant horloger sont représentés un élément de liaison et des parties sacrificielles gravés dans la couche supérieure de silicium, l'élément de liaison servant à maintenir le composant horloger attaché à la plaquette ;
• la figure 5 montre en vue de dessus le composant horloger de la figure 4 attaché à la plaquette après suppression des parties sacrificielles.

Other characteristics and advantages of the present invention will become apparent on reading the following detailed description given with reference to the appended drawings in which:

• the figure 1 shows by schematic sectional views the different steps of the manufacturing process according to the invention;
• the figure 2 shows a top view of a watch component etched in an upper silicon layer of a silicon-on-insulator wafer according to a first exemplary embodiment. In this figure, the white lines represent the voids created in the upper layer of silicon by the etching and the gray areas represent the silicon of the upper layer. In addition to the timepiece component are shown a connecting element and sacrificial parts etched in the upper layer of silicon, the connecting element serving to keep the watch component attached to the wafer;
• the figure 3 shows a top view of the watch component of the figure 2 attached to the wafer after removing the sacrificial parts;
• the figure 4 shows a top view of a horological component etched in an upper silicon layer of a silicon-on-insulator wafer according to a second exemplary embodiment. In this figure, the white lines represent the voids created in the upper layer of silicon by the etching and the gray areas represent the silicon of the upper layer. In addition to the timepiece component are shown a connecting element and sacrificial parts etched in the upper layer of silicon, the connecting element serving to keep the watch component attached to the wafer;
• the figure 5 shows a top view of the watch component of the figure 4 attached to the wafer after removing the sacrificial parts.

The method of manufacturing a silicon watch component according to the invention begins with a first step consisting in providing a wafer 1 of the silicon-on-insulator type ( figure 1 (a) ). This wafer comprises an upper layer of silicon 2, a lower layer of silicon 3 and, between the two, an intermediate layer of silicon oxide 4. The silicon is monocrystalline, polycrystalline or amorphous. It may or may not be doped.

At a second stage of the process ( figure 1 (b) ), the upper layer of silicon 2 is structured to form the watch component therein, designated by 5. To do this, a deep reactive ionic etching is typically carried out preceded by a photolithography operation as described in the patent application. WO 2019/180596 . The intermediate layer of silicon oxide 4 serves to stop the etching. The lower silicon layer 3, preferably thicker than the upper silicon layer 2, for its part serves as a rigid support facilitating the etching as well as the following steps of the process. During this second step, one leaves in a recess 6 of the watch component 5 (cf. figures 1 (b) and 2 ) a part 7 of the upper layer of silicon 2 which will be called “connecting element”, as well as a fastener or bridge of material 8 connecting this connecting element 7 to the wall, 6a, of the recess 6. In order to accelerate the etching, it is also possible to leave, in other recesses and / or outside the watch component 5, sacrificial silicon parts 9 that no attachment connects to the watch component in the upper layer of silicon. 2.

In a third step of the process, called under-etching ( figure 1 (c) ), the intermediate layer of silicon oxide 4 is etched in order to remove it between the watch component 5 and the lower layer of silicon 3. This step can be carried out by controlled chemical attack, preferably with steam, with l using hydrofluoric acid or a mixture of hydrofluoric acid and ethanol. This step separates the watch component 5 from the lower silicon layer 3 and detaches the sacrificial silicon parts 9 which can thus be removed (the figure 3 shows the watch component 5 after removing the sacrificial silicon parts 9). The watch component 5 however remains attached to the lower layer of silicon 3 by successively the attachment 8, the connecting element 7 and a part 10 of the intermediate layer of silicon oxide 4 left between the connecting element 7 and layer 3. In fact, the dimensions of the connecting element 7 are chosen sufficiently large so that after the removal of the oxide under the watch component 5, the fastener 8 and the sacrificial parts 9 and until the under-etching stops, the oxide remains under the connecting element 7.

In a first exemplary embodiment, illustrated in figures 2 and 3 , the recess 6 of the watch component 5 in which the connecting element 7 is located is the central hole of a hub 11 formed by elastic arms allowing elastic mounting of the component on an axis, and the clip 8 connects the 'connecting element 7 to the wall 6a of this central hole. In a second exemplary embodiment, illustrated in figures 4 and 5 , the recess 6 of the watch component in which the connecting element 7 is located separates a rim 12 of the component and one of the elastic arms of the hub 11, and the clip 8 connects the connecting element 7 to the surface internal 12a of the rim 12. In general, the connecting element 7 can be connected to any interior surface of the watch component, such as the interior surface of a rim, a cavity, an axis hole or other assembly hole, the surface of an inner arm, the inner or outer surface of a hub, etc. One could also have several disjoint connecting elements 7 connected by respective fasteners to the component. The / each connecting element 7 could also be connected to the component by several fasteners. The horological component is for example a toothed wheel, as shown in figures 2 to 5 , a cam, an indicator needle or a spiral-type spring or the like.

As the outer surface of the watch component does not need to be connected by a fastener to the rest of the plate, it can be functional over its entire circumference and have a micrometric structure of the micro-tooth type or the like. In addition, the surface condition and appearance of the final component may be optimal since the outer surface will be free of any attachment residue.

A fourth step of the process ( figure 1 (d) ) consists in thermally oxidizing the watch component 5. For this purpose, the wafer 1 with the watch component 5 is placed in an oven to subject it to a temperature between 800 and 1200 ° C and to an oxidizing atmosphere comprising gas oxygen or water vapor until a predetermined thickness of silicon oxide (SiO ₂ ) is obtained on its surfaces. This silicon oxide layer, 13, is formed by consuming silicon to a depth corresponding to approximately 44% of its thickness. The oxidation is stopped before the growth of the silicon oxide layer 13 on the clock component 5 and on the lower silicon layer 3 causes them to touch each other.

The watch component is then deoxidized by removing the silicon oxide layer 13, for example by controlled chemical attack, wet or with steam, using hydrofluoric acid or a mixture of hydrofluoric acid and ethanol. ( figure 1 (e) ). The respective sizes of the watch component 5 and of the lower silicon layer 3 are then reduced with respect to the figure 1 (c) , which increases their spacing d.

Then the oxidation-deoxidation sequence illustrated in figures 1 (d) and 1 (e) is repeated a number of times until a distance d of predetermined value is obtained between the clock component 5 and the lower silicon layer 3.

The oxidation parameters can vary from one oxidation-deoxidation sequence to the next. For example, the assembly 3, 5 can be oxidized for a longer time in the last sequences than in the first since more space is available between the clock component 5 and the lower silicon layer 3 for the growth of the silicon oxide. .

Each oxidation step is stopped before the timepiece component 5 and the lower silicon layer 3 touch each other via the silicon oxide layer 13. In this way, the risk of the timepiece component 5 and the lower silicon layer 3 are attracted to each other and merge when removing the silicon oxide layer 13.

In addition to increasing the spacing d, these oxidation-deoxidation sequences make it possible to attenuate the surface defects of the watch component, in particular the undulations created by the deep reactive ionic etching on the sides of the watch component.

At a next step of the process ( figure 1 (f) ), a permanent silicon oxide layer 14 can be formed on the watch component, by thermal oxidation or deposition, to increase its mechanical strength. The large spacing d obtained by the oxidation-deoxidation sequences make it possible to form an oxide layer 14 of great thickness, for example approximately 3 μm or even more, while the component is still attached to the wafer 1.

Commercial silicon-on-insulator wafers are generally obtained by growing an oxide layer on a silicon wafer and fusion welding another silicon wafer on the oxide layer. Due to the compressive stress that it generates on the first silicon wafer, the oxide layer must have a limited thickness, between 1 and 3 μm at most. We can therefore understand the advantage of being able to increase at will the distance d between the watch component 5 and the lower silicon layer 3 before the formation of the permanent silicon oxide layer 14, as the invention allows.

As an alternative to the permanent silicon oxide layer 14, or on this layer, the watch component 5 could receive a layer of a material having good tribological properties, for example carbon crystallized in the form of diamond (DLC) or crystals. carbon nanotubes, a layer forming an oxygen barrier or a layer, for example of parylene, used to contain debris if the component breaks.

In addition to the oxidation-deoxidation treatments, or as an alternative to them, one could apply to the watch component 5 other treatments such as annealing in a reducing atmosphere according to the teaching of the patent application. CH 702431 .

The horological component 5 is typically part of a batch of identical horological components manufactured simultaneously in the wafer 1. At a last step of the process, the horological components are detached from the wafer 1 by breaking or elimination of the fasteners 8. For this purpose, the watch components can be turned with respect to the plate 1 by means of tweezers until the fasteners 8 are broken, or the latter are cut by laser.

The present invention allows complete work on the watch component (s) while they are linked to the wafer. It is thus possible to achieve high and repeatable quality levels with good uniformity between the different components.

The watch components can be integrated into timepieces such as wristwatches, pocket watches or clocks.

Claims (11)

A method of manufacturing a silicon watch component comprising the following steps: a) providing a wafer (1) comprising a first layer of silicon (2), a second layer of silicon (3) and, between the two, an intermediate layer of silicon oxide (4); b) etching the first layer of silicon (2) to form therein the watch component (5), at least one connecting element (7) located in a recess (6) of the watch component (5) and at least one fastener (8) ) connecting this connecting element (7) to an interior surface (6a; 12a) of the watch component (5); c) removing the intermediate layer of silicon oxide (4) between the watch component (5) and the second layer of silicon (3) and leaving it at least partially between the connecting element (7) and the second layer of silicon (3); d) applying at least one treatment to the timepiece component (5); e) detaching the watch component (5) from the plate (1) by breaking or removing the at least one fastener (8). Method according to Claim 1, characterized in that the recess (6) is a hole intended to serve for the assembly of the watch component (5) with an element, for example an axis. Method according to Claim 1, characterized in that the recess (6) separates a hub (11) and a rim (12) from the watch component (5). Method according to claim 1 or 3, characterized in that the inner surface (12a) is the inner surface of a rim (12) of the watch component (5). Method according to any one of claims 1 to 4, characterized in that step b) comprises deep reactive ion etching. Method according to any one of claims 1 to 5, characterized in that step b) leaves sacrificial parts (9) in the first silicon layer (2) which are then detached by step c). Process according to any one of Claims 1 to 6, characterized in that step d) comprises a thermal oxidation phase followed by a deoxidation phase, the thermal oxidation phase being stopped before the watch component ( 5) and the second silicon layer (3) touch each other. Process according to Claim 7, characterized in that the thermal oxidation and deoxidation phases are repeated one or more times. A method according to any one of claims 1 to 8, characterized in that step d) comprises forming a permanent silicon oxide layer (14). Method according to any one of Claims 1 to 9, characterized in that the horological component is a toothed wheel (5), a needle, a cam or a spring. Method according to any one of claims 1 to 10, characterized in that several said watch components are made from said wafer (1).

Images