Answers are grounded in this paper and cite their sources.
With the number of Type 1 diabetes patients increasing across the globe, Continuous Glucose Monitors (CGM) have emerged as a key component in the management of this worldwide epidemic. 1.5M patients in the US alone are currently living with type 1 diabetes and this number is expected to grow to 50M by 2050. Unlike the prevalent solution of finger pricking multiple times a day, CGMs stay in contact with the body and take continuous blood glucose measurements making it much easier and safer for the patients. A CGM solution was developed that is implanted under the patient's skin and takes measurements every 5 minutes, sending the data to patient's smartphone. The patient wears a transmitter right outside/on top of where the sensor is implanted. This transmitter wirelessly powers the sensor while also relaying the data collected by the sensor to the app on the patient's smartphone. This paper describes the key design, process, material, and equipment challenges involved in developing and qualifying this unique implantable sensor's package/module portion, and the innovative packaging solutions devised to overcome these challenges.The module consists of a ferrite-based thick film printed antenna, two identical ASICs with 8 photodiodes per ASIC, 2 UV and 2 Blue LEDs flip-chip mounted on each of the two ASICs, two capacitors mounted on each of the two ASICs, while the ASICs are wire bonded to the bond fingers on the ferrite antenna. The photodiodes on the ASICs are coated with custom optical filters. The ASICs are attached to the antenna using die attach film. In the absence of any leadframe or substrate, this ferrite-based antenna – which works as a Near Field Communication (NFC) antenna for data transfer from the sensor to the transmitter and for power transfer from the transmitter to the sensor – acts as the carrier or the substrate. There are two passive components mounted on to the antenna as well. Finally, glob top encapsulation is dispensed on the non-photodiode portions of the ASICs.As is evident from the unique structure of this sensor package/module, there were many obvious challenges to be overcome. As there is no leadframe or substrate strip, to make it compatible with a typical assembly line, the individual antennae were reconfigured into a strip by placing them on a metal carrier with double-sided tape. Since the custom optical filters on the ASIC photodiodes could not withstand the harsh chemicals involved in a typical Under Bump Metal (UBM) deposition (bumping) process, standard flip chip process could not be used for attaching LEDs and capacitors. Solder can't wet the standard Al pad metal on the ASICs, hence a unique solution was proposed and successfully used as an interface between solder and Al metal pad. After applying this unique interface on top of the Al bond pads, solder paste was applied with jetting/screen printing. Due to a narrow 80um gap between the LED attach pads on the ASIC, it was found to be extremely challenging to avoid solder bridging when using unit-level solder jetting. To overcome this issue, a wafer-level process was devised where the unique interface application, followed by solder screen printing, followed by flip chip LED attach were all performed on wafer-level. The unequal sizes of the two LED terminals made it difficult to achieve balanced solder joints and controlled LED tilt. The wafer had to be then background to the required thickness and special care was taken to select appropriate tape material and process parameters to prevent damage to the LEDs and capacitors on the front side of the wafer. These wafer-level process steps are described in detail including solder screen printing and LED attach DOEs to achieve the optimum LED shear, joint stand-off height, and tilt.Multiple assembly challenges and corresponding solutions related to the complexities of antenna are also briefly touched upon. Prime among them were antenna warpage, poor die shear due to non-flat antenna top surface (thick traces of printed coil resulted in a corrugated surface), and wire bonding challenges on thick and soft antenna metallization. The resulting optimized module passed the unique set of target reliability requirements arising from being implanted inside the human body.
No takes yet. Share an insight, caveat, or question.
Mehrotra et al. (2024) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: