The present invention is concerned with a method of manufacturing small elements. The preferred embodiments of the invention are particularly suitable for the manufacture of small (chip scale) components or microcomponents. The typical chip-scale component size is in the range 0.2 mm to a few mm with features of down to 0.01 mm. The term microcomponent is typically used to describe components which are not visible without the use of an optical microscope (e.g., typically within size range of 10−4 and 10−7 metres). Micro-components may be used in micro-structural devices.
Electronic microcomponents are typically made as arrays of components on a silicon substrate. It is more efficient to make a number of elements on the same substrate. The processes for creating an array of elements are well known and include, for example, photolithography. The microcomponents are formed as an array of connected elements which are separated from each other before being used.
U.S. Pat. No. 5,824,595 discusses a method in which an array of electronic elements are created on a silicon substrate, and the elements are separated from each other by etching of the substrate.
A problem with the etching process disclosed in U.S. Pat. No. 5,824,595 is that separate elements end up loose having been etched from the carrier. Due to their small size, they are prone to clump together and are difficult to separate without damage. There is therefore a low yield. For example, the typical yield for the separation of cross-shaped gold bonding preforms of the type shown in
The present invention provides method and apparatus as defined in independent claims 1, 2 and 4. Preferred features of the invention are set out in the dependent claims.
Preferred embodiments of the present invention allow one to remove components or elements one at a time in a controlled and/or controllable manner. This means that it is possible to prevent the formation of a mass or conglomerate of mixed up components.
Traceability of individual elements is also improved as the array format is kept right up to the point where an individual part or element is used. This means that a user or a system can determine and/or monitor which particular element or component is taken from where and then where it is placed.
Preferred embodiments of the invention will now be described, by way of example only, with reference to the attached figures in which:
a to 4d illustrate a separation method embodying the invention for separating an element from the array of
An array 1 of connected components 2 may be (see
In the known processes such as that described in U.S. Pat. No. 5,824,595 the individual elements in the array of elements are then separated by a chemical etching processes. This results in a jumble of elements and the disadvantages discussed above.
In the embodiment of the invention illustrated in
Embodiments of the invention are concerned with the selective and controllable application of energy to selected tabs so as to allow the separation of a selected or selected elements from an array of inter-connected elements. In an alternative embodiment of the invention having non-conductive tabs, the tabs may be removed, for example, by a blade or by laser ablation.
In the embodiment described with reference to FIGS. 1 to 4, the element to be separated from an array is a conducting preform 2 designed to provide an electrical connection to a semiconductor die. The invention is, however, applicable to any components formed in arrays of components. It is therefore also of application to the separation of the semiconductor die itself, which is subject to the same problem in production as the preforms, namely, difficulty in handling the diodes when individually separated. As an example of the use of the invention in connection with the manufacture of semiconductor components, we shall now describe further embodiments of the invention concerned with arrays of Gunn diodes. As mentioned above, the invention is, however, equally applicable to arrays of other elements or components.
As currently manufactured, the process for manufacturing a Gunn diode begins with a semiconductor wafer on which epitaxial layers are grown, which is metallised to provide a bonding layer for subsequent plating. Heat sinks are plated onto the metallised layer, the wafer is then etched from the other side to thin it, and gold contacts are plated onto the other side of the thinned wafer in register with the heat sinks. The individual Gunn diodes are then etched to form a mesa, and are then separated from each other.
In accordance with embodiments of the invention, this process is modified to improve the ability to handle the Gunn diodes after the separation process. Thus, after the wafer with epitaxial layers is metallised, and the areas are suitably defined by means such as a photoresist mask, an additional layer of gold is plated on by, for example, electroplating, in order to form a mesh 9 together with a first layer of heat sink 10 (only one being indicated by a reference numeral) for the Gunn diodes and tabs 11 (only four being indicated by a reference numeral) for joining the heat sinks of the Gunn diodes to the mesh. In a first form of the second array shown in
A typical manufactured and separated Gunn diode is shown in
A suitable thickness for the first plated layer is between 5 and 10 μm (microns), for example, 6 μm (microns), and for the total thickness of both layers between 30 and 50 μm, for example, 40 μm.
In use, the tabs are broken to separate the individual Gunn diodes, and this may be done mechanically, for example, with a blade, by laser, or electrically.
In the case of mechanical separation, the variant of
In the case of electrical separation, the tabs 3 of
A suitable circuit for applying a pulse of current, through a pick-up tool 8 of the type described above, to melt the tabs 3 surrounding the preform 2 of
It may be found convenient to step repeatedly the finished wafer to bring the next Gunn diode to be separated over a workstation provided with means to connect one pole of the capacitor to the heat sink of the Gunn diode to be separated, the other pole being in electrical contact with the mesh 9.
While the mesh 9 is shown in
Further, it is possible to dispense with the mesh 9 altogether, so that the heat sinks are connected to each other by the tabs only. This would be possible for the arrangement shown in
Number | Date | Country | Kind |
---|---|---|---|
0406639.5 | Mar 2004 | GB | national |
Filing Document | Filing Date | Country | Kind | 371c Date |
---|---|---|---|---|
PCT/GB05/01178 | 3/24/2005 | WO | 7/2/2007 |