Sophisticated integrated circuit processing of silicon materials has been promoted as the optimum methodology for miniaturized mechanical and chemical sensors for over two decades. However, in micro biomedical applications, advantages of using Si are often not as clear as in mechanical sensors. The high cost of Si, difficulties in packaging, need for modularity and biocompatibility all encourage investigation of non-silicon materials and semi-continuous processing. Metal/polymer hybrid structures are preferred as the substrate materials in this application.
The two manufacturing techniques demonstrated here involve non-silicon materials and a modular methodology. They are based on robust printed circuit board and flexible printed circuit [FPC] high volume fabrication techniques using polyimide base films for millimeter scale devices and photoreactive dry film resists for sub-millimeter devices. The core concept shared between the designs is a structure having sensor sites and their electrical contacts on opposing sides of the substrate. This separation of fluid chemistry on one side and dry electrical contact on the other improves reliability, ultimate packaging simplicity, and ease of use. Chemical sensors fabricated by each of these processes as well as the corresponding sensor performance are presented.
The two-sided paradigm improves upon single-sided devices by allowing simpler, smaller and higher-yield fabrication of multi-purpose sensor arrays. Finished product yield is additionally enhanced by modularity. That is, each sensor type is created on it’s own sheet, independent of the other sensor types. At final assembly, many sensors of different types can be cut from different sheets and combined into any desired array configuration using contemporary pick and place equipment. A schematic of the two-sided sensor structure is shown in Figure 1.
The first approach by Packard-Hughes Interconnect, OSU and Microbionics Inc. for creating these structures is shown in Figure 2. In this case a method of fabricating high volume FPC with integrated contacts (Gold Dot Technology) is leveraged to create a biosensor that can be produced in volumes of greater than a quarter million per batch and at a low assembled cost.
A second approach explored by OSU, and Microbionics Inc. has been demonstrated to reduce the sensor size to below 100 μm at similar costs. This is accomplished as shown in Figure 3 by defining the sensor geometry in dry photoresist sheets using a simplified two-layer alignment technique that requires exposure from only one side.
Selection of the appropriate technique is application dependent. Preliminary results of potassium ion and chloride sensors are presented.