Hello M. Yi Jia!

Your question is very general!

In ALD, the key to control particles is to avoid contact between the various precursors. Enough purging to reach this goal is achieved by adequate chamber design, something that will be shared by every process for a given chamber, and also by designing correctly purging duration depending on precursor sticking. The maximum needed particle count and the size at which you look depends on the type of layer. Clearly if you are looking at high K dielectric for transistor gate and at top of the art technology, you look at particle size in the 30 nm range and at counts like dozens of particles... I am working for ASM and can tell you that our tools are at top manufacturing level! please contact regional ASM office to get our specifications. Have a good quest!

Philippe BAJOLET, ASM France

Mr. Bajolet,

Thank you very much for your answer. It seems to be a systematic work to control the particles. In our demo lab, It is difficult to fulfill all the requirements ,such as clean environment, gas supply, valve, especially the chamber design may need complex simulation. It's very glad to know ASM's leading technology in manufacturing instruments.

Hello Mr. Jia,

Mr, Bajolet already answered you. I would like to add one more point. The particle contamination and size can be affected by the base pressure of the ALD chamber.  For example with base pressure 1^-6mBar has smaller particle size as compare to base pressure of 1^-3mBar. Other than this the purging of the system also varies for precursor to precursor is same chamber is used to grow different oxides. 

Mr. Das,

Thank you for the answer. I got confused about the effect of base pressure on the particle size. Any relevant published paper that I can study? Further more, Mr. Bajolet mentioned purging duration depends on the precursor sticking , dose this connect to specific physical quantity of precursor? How can we compare among those liquid, solid metal-organic compounds?

Have you considered particle contamination on your surface prior to ALD deposition? ALD is conformal, so any particles on the surface prior to deposition i.e. dust particles, will be coated and will be visible in the film, especially if you're after a very smooth surface?