Let’s first understand what device is called a flip chip? Generally speaking, such devices have the following characteristics:
1. The substrate is silicon;
2. The electrical surface and solder bumps are on the lower surface of the device;
3. The ball pitch is generally 4-14mil, the ball diameter is 2.5-8mil, and the overall size is 1-27mm;
4. Underfill is required after assembling on the substrate.
In fact, the reason why the flip chip is called “flip chip” is relative to the traditional wire bonding connection method (Wire Bonding) and the post-balling process. The electrical side of the traditional chip connected to the substrate by wire bonding faces up, while the electrical side of the flip chip faces down, which is equivalent to turning the former over, so it is called “flip chip”. After the chip is implanted on the wafer (Wafer), it needs to be turned over and sent to the placement machine for easy placement. Because of this turning process, it is called “flip chip”.
1. Assembly process of flip chip for PCB board
In semiconductor back-end assembly plants, there are now two methods of module assembly. In two reflow processes, SMT devices are first assembled on a separate SMT line consisting of a screen printer, placement machine and the first reflow oven. The partially assembled modules are then processed through a second production line consisting of a flip-chip mounter for PCB boards and a reflow oven. The underfill process is completed in a dedicated underfill production line, or in combination with a flip-chip production line for PCB boards.
2. Introduction to the assembly process of flip chip for PCB board
Compared with other IC devices, such as BGA, CSP, etc., the flip-chip assembly process for PCB boards has its particularity. This process introduces a flux process and an underfill process. Flip-chip mounting of PCBs on solder paste is not an acceptable method of assembly because of flux residues (impact on reliability) and the risk of bridging. The industry has introduced no-clean flux, and the chip dip flux process has become a widely used flux technology. The main alternative at present is to use no-clean flux, dip the device in the flux film, let the device solder balls dip a certain amount of flux, then mount the device on the substrate, and then reflow; Pre-applied on the substrate, and then the device is mounted and reflow soldered. The flux acts to secure the device prior to reflow, and to wet the solder surface during the reflow process to enhance solderability.
After the PCB board is soldered with flip-chip, a glue (usually epoxy resin material) needs to be filled between the bottom of the device and the substrate. Underfill is divided into fluidity and no-follow underfill based on the “capillary flow principle”.
The above-mentioned flip-chip assembly process for PCB board is for C4 devices (device solder bump material is SnPb, SnAg, SnCu or SnAgCu). Another process is to use anisotropic conductive adhesive (ACF) to assemble flip chips for PCB boards. Anisotropic conductive adhesive is applied on the substrate in advance, and the patch head uses a high pressure to mount the device on the substrate, and at the same time, heat the device to cure the conductive adhesive. This process requires the placement machine to have very high precision, and the placement head has high pressure and heating functions. This process tends to be used for the assembly of non-C4 devices where the bump material is Au or others. Here, we mainly discuss the C4 process. The following table lists several ways of flip-chip bumping and connection on the substrate for PCB boards.
Flip-chip PCB board with flip-chip geometry can be described by a “small” word: small diameter of solder balls (as small as 0.05mm), small spacing between solder balls (as small as 0.1mm), and small size (1mm2). To obtain a satisfactory assembly yield, it brings challenges to the placement equipment and its process. As the diameter of the solder balls shrinks, the placement accuracy requirements are getting higher and higher. At present, the accuracy of 12μm or even 10μm is more and more common. The graphics processing capability of the camera of the SMD equipment is also very critical. Small ball diameters and small ball pitches require higher pixel cameras to process.
With the passage of time, the size of high-performance chips continues to increase, the number of solder bumps continues to increase, and the substrate becomes thinner and thinner. In order to improve product reliability, underfill has become necessary.