4.8 Article

Using Hydrogen To Expand the Inherent Substrate Selectivity Window During Tungsten Atomic Layer Deposition

Journal

CHEMISTRY OF MATERIALS
Volume 28, Issue 1, Pages 117-126

Publisher

AMER CHEMICAL SOC
DOI: 10.1021/acs.chemmater.5b03319

Keywords

-

Funding

  1. State of North Carolina
  2. National Science Foundation
  3. Semiconductor Research Corporation (SRC) [2401.001]

Ask authors/readers for more resources

Area-selective thin film deposition is expected to be important in achieving sub-10 nm semiconductor devices, enabling feature patterning, alignment to underlying structures, and edge definition. Atomic layer deposition (ALD) offers advantages over common chemical vapor deposition methods, such as precise thickness control and excellent conformality. Furthermore, several ALD processes show inherent propensity for substrate-dependent nucleation. For example, tungsten ALD using SiH4 (or Si2H6) and WF6 is more energetically favorable on Si than on SiO2, but selectivity is often lost after several ALD cycles. We show that modifying the W ALD process chemistry can decrease the W nucleation rate on SiO2, thereby expanding the ALD selectivity window. Specifically, we find that wadding H-2 during the WF6 dose step helps passivate SiO2 against W nucleation without modifying W growth on silicon. Surface characterization confirms that H2 promotes fluorine passivation of SiO2, likely through surface reactions with HF produced in the gas phase. This passivation affords at least 10 additional W ALD cycles, corresponding to similar to 6 nm of additional W growth, before substantial nucleation occurs on SiO2. We show that reactant modification also reduces undesirable nucleation due to substrate proximity or loading effects in patterned film growth. Further understanding of ALD reaction chemistry and film nucleation will lead to improved selective metal and dielectric film deposition, enabling ALD bottom-up patterning.

Authors

I am an author on this paper
Click your name to claim this paper and add it to your profile.

Reviews

Primary Rating

4.8
Not enough ratings

Secondary Ratings

Novelty
-
Significance
-
Scientific rigor
-
Rate this paper

Recommended

No Data Available
No Data Available