Manufacturing method of flash memory

Abstract

The invention provides a manufacturing method of a flash memory. A step of surface recovery processing on a floating grid, which is exposed due to sunken of an active region, at a junction of the active region and a shallow groove isolation region is added after a source region or a drain region is formed in a semiconductor substrate at the bottom of an opening and before a second side wall material is deposited in the opening, the opening is formed after a floating grid layer and a floating grid oxide layer are etched, thus, the damage to a side wall of the floating grid on a sunken region of the active region of the opening can be recovered, a favorable process surface is provided for material deposition of a second side wall, the problem of relatively poor isolation between the floating grid and a source line caused by that the active region at a boundary of a shallow groove isolation structure is sunken can be further solved, and the data maintaining ability of the flash memory is improved.

Description

technical field [0001] The invention relates to the field of semiconductor manufacturing, in particular to a method for manufacturing a flash memory. Background technique [0002] At present, flash memory (Flash memory), also known as flash memory, has become the mainstream of non-volatile memory. Its storage unit is based on the traditional MOS transistor structure, adding a floating gate (Floating Gate, FG) and a layer tunnel oxide (Tunnel Oxide), and use floating gates to store charges to achieve non-volatility of stored content, and shallow trench isolation (STI, Shallow Trench Isolation) structure is required for electrical isolation between memory cells . [0003] A typical manufacturing process of flash memory in the prior art includes FG OX / FG Poly / PAD SiN DEP→STI etch→STI Liner OX / STI HDP&CMP→ / PAD SiN remove→FG SiN DEP→FGPH / SiNetch→FGSP1 DEP / etch→FGPL1 etch→VSS IMP→FGSP2 DEP / etch, as follows: [0004] First, please refer to Figure 1A and Figure 1B , providing...

AI Technical Summary

Problems solved by technology

[0010] It needs to be explained here that in the manufacturing process flow of the above-mentioned flash memory, after the FGSP1 DEP process is completed, the thickness of the oxide spacer material deposited at the bottom of the opening is usually thicker than that of the oxide spacer material deposited on the surface of the interlayer dielectric layer 14. The thickness of the side wall material is much thinner (approx. ), and when the FGSP1etch process is performed to form the sidewall 15, the shallow trench isolation structure 11 at the bottom of the opening will be lost, and when the FGSP1etch process is completed, the shallow trench isolation structure 11 will be lost by about Thus it is likely to expose the top corners of the active region 101 around it, such as Figure 1B As shown by the dotted line box 101 of , the exposed top corner of the active region 101 will be etched in the subsequent FGPL1 etch process and part of it will be lost, thereby forming Figure 1C The corner damage of the active area, that is, the active area recess (ACT recess or ACT Pits) 102, the active area recess 102 may lead to poor step coverage of the side wall material deposited in the FGSP2 DEP process, thereby causing the second side wall 17 The ability to isolate the floating gate FG and the source line is reduced, which weakens the data retention performance of the flash memory, and even causes the data retention failure issue of the flash memory; in addition, the chlorine concentration in the sidewall material deposited by the FGSP2 DEP process It may induce a leakage path between the source line polysilicon (SPL) and the floating gate FG filled in the opening subsequently, impairing the data retention performance of the flash memory

Images