WCN thin film deposition method with adjustable work function

Abstract

The invention discloses a work function adjustable WCN thin film deposition method which comprises the following steps: firstly, heating a tungsten precursor to 80-150 DEG C, and preheating a precursor conveying pipeline and an ALD cavity; forming a monatomic layer WN thin film through NH3 plasma and tungsten precursor reaction deposition, and repeating X times to form X layers of WN thin films; forming a monatomic layer WC thin film through reaction deposition of H2 plasma and a tungsten precursor, and repeating the step Y times to form a Y-layer WC thin film; and repeating the steps for Z times until the WCN thin film with the expected thickness is obtained. According to the WCN thin film prepared through the method of alternately depositing the WC thin film and the WN thin film, atom combination between WCN can be achieved, the existing mode that the proportion of C to N in the WCN is adjusted by adjusting the deposition technological parameters of the single WCN thin film is replaced, the content of C and N can be accurately adjusted, and therefore the work function can be adjusted.

Description

technical field [0001] The invention relates to the technical field of semiconductor functional thin film preparation, in particular to a WCN thin film deposition method with adjustable work function. Background technique [0002] The gate electrodes of traditional CMOS devices are doped P-type polysilicon and N-type polysilicon, but as the device size decreases, polysilicon electrodes encounter more and more problems: polysilicon depletion effect, RC delay response, polysilicon and Fermi level pinning problem in high-K dielectric layers. The solution to these problems is to use metal electrodes instead of polycrystalline electrodes. The work function of metal electrodes is required to be able to match NMOS and PMOS to achieve a low threshold voltage. Generally, the work function of NMOS electrode is ≈4ev, and the work function of PMOS electrode is ≈5ev. However, using different metals as CMOS electrodes will increase the difficulty of process integration, and will also ca...

AI Technical Summary

Problems solved by technology

[0002] Traditional CMOS device gate electrodes use doped P-type polysilicon and N-type polysilicon, but with the reduction of device size, polysilicon electrodes encounter more and more problems: polysilicon depletion effect, RC delayed response, polysilicon and Fermi Level Pinning in High-K Dielectric Layers

However, this method is difficult to prepare, the content of C and N is difficult to control, and it is difficult to achieve large-scale mass production

There are mainly the following WCN thin film composition adjustment methods: (1) Adjust the ratio of C and N in the thin film by adjusting the flow ratio of nitrogen and hydrogen in the reducing plasma gas in the ALD process (nitrogen and hydrogen flow into the cavity at the same time) (like figure 1 As shown), the disadvantage is that the work function of the film is strictly controlled by the process parameters. Once the process drifts, it will lead to the change of the work function of the film, which will lead to the failure of the device; (2) micro-adjust the C and The composition ratio of N (such as figure 2 shown), the disadvantage is that only fine-tuning of the work function can be achieved, and the adjustment range is not controlled; (3) by adding a step of hydrogen plasma treatment in each ALD deposition cycle (such as image 3 shown), so that each deposition cycle obtains a layer of WCN, and the ratio of C and N in each film is controlled by controlling the treatment time of hydrogen plasma. The disadvantage is that the final film composition is controlled by process parameters, especially hydrogen Once the plasma processing power and duration drift, it will have an impact on the work function of the film

Images