Chip-side solution
1. Reduce the size of the chip size reduction
On the surface, is the problem of layout design, it seems that as long as the need to design a smaller map can be resolved. However, the chip size can continue indefinitely down? the answer is negative. There are several reasons to limit the extent of chip size reduction:
(1) packaging processing restrictions. During the packaging process, two factors limit the size of the chip. First, the nozzle limit. Solid crystal need to draw the chip, the chip must be greater than the short edge of the nozzle diameter. There are currently cost-effective nozzle diameter of 80um or so. Second, wire restrictions. First, the wire electrode that chip electrode must be large enough, otherwise the reliability of wire can not be guaranteed, the industry reported minimum electrode diameter 45um; followed by the distance between the electrodes must be large enough, otherwise the two must be mutual interference between the wire.
(2) chip processing restrictions. Chip processing, there are two restrictions. One is the layout layout limitations. In addition to the limitations of the package end, the size of the electrode, the electrode spacing is required, the electrode and the MESA distance, the width of the road, the boundaries of different layers have their limits, the current characteristics of the chip, SD process capability, lithography processing capacity Determines the scope of the specific restrictions. In general, the minimum distance from the P electrode to the edge of the chip is limited to 14 μm or more.
The second is the ability to crack processing constraints. SD scribing + mechanical lobe process has a limit, the chip size is too small may not be lobed. When the diameter of the wafer increases from 2 to 4 inches, or to 6 inches in the future, the difficulty of dicing the lobes increases, that is, the chip size can be processed increases. 4 inch chip, for example, if the chip length of less than 90μm short side, aspect ratio greater than 1.5: 1, the yield loss will increase significantly.
Based on the above reasons, I boldly predicted that the chip size reduced to 17mil2, the chip design and process capability close to the limit, the basic no longer reduce the space, unless the chip technology solutions have a big breakthrough.
2. Brightness boost
Brightness enhancement is the eternal theme of the chip side. Chip factory through the extension of the program to optimize the quantum effect, through the chip structure adjustment to enhance the external quantum effect.
However, on the one hand chip size will inevitably lead to reduce the size of the light-emitting area, chip brightness decreased. On the other hand, small pitch screen dot pitch narrow, single-chip brightness demand has declined. There is a complementary relationship between the two, but to leave the bottom line. At present, chip-side in order to reduce costs, mainly in the structure to do subtraction, which usually have to pay the cost of reducing brightness, so how trade-off is the industry to pay attention to.
3. Consistency at low current
The so-called small current, is relatively common indoor, outdoor chip trial current is. As shown in the following diagram, the chip I-V curve, the conventional indoor, outdoor chip work in the linear work area, the current is large. The small pitch LED chip needs to work near the zero point of the non-linear work area, the current is too small.
In the non-linear work area, LED chip by the semiconductor switching threshold, the difference between the chip is more obvious. It is easy to see that the discrepancy of the nonlinear workspace is much larger than that of the linear workspace, because of the analysis of the brightness and wavelength dispersion of the high-volume chips. This is the inherent challenge of the chip side.
The way to deal with this problem is to optimize the direction of the epitaxy to reduce the lower limit of the main linear work area; followed by the chip on the optical optimization, the different characteristics of the chip to distinguish.
4. Parasitic capacitance consistency
At present, there is no condition on the chip to directly measure the capacitance of the chip. Capacitance characteristics and the relationship between conventional measurement items is not clear, the industry has to be summed up. Chip-side optimization of the direction of the epitaxy on the one adjustment, one sub-file on the electrical refinement, but the cost is high, not recommended.
5. Reliability
Chip-side reliability can be described in the chip package and the parameters of the aging process. But overall, the reliability of the chip on the screen after the impact of factors, focusing on ESD and IR two.
ESD refers to antistatic ability. According to IC industry reports, more than 50% of the chip failure and ESD. To improve chip reliability, must enhance the ESD capability. However, in the same wafer, the same chip structure, the chip size will inevitably bring smaller ESD capacity weakened. This is directly related to the current density and chip capacitance characteristics and can not be resisted.
IR is the reverse leakage current, usually measured at a fixed reverse voltage reverse current value of the chip. IR reflects the number of defects within the chip. The larger the IR value, the more the chip is defective.
To enhance ESD performance and IR performance, must be epitaxial structure and chip structure to make more optimization. In the chip sub-file, through strict sub-file standard, you can effectively put the ESD performance and weak IR chip removed, thereby enhancing the chip on the screen after the reliability.
