Processes for the production of organometallic compounds

Abstract

This invention relates to processes for the production of organometallic compounds represented by the formula M(L)3 wherein M is a Group VIII metal, e.g., ruthenium, and L is the same or different and represents a substituted or unsubstituted amidinato group or a substituted or unsubstituted amidinato-like group, which process comprises (i) reacting a substituted or unsubstituted metal source compound, e.g., ruthenium (II) compound, with a substituted or unsubstituted amidinate or amidinate-like compound in the presence of a solvent and under reaction conditions sufficient to produce a reaction mixture comprising said organometallic compound, e.g., ruthenium (III) compound, and (ii) separating said organometallic compound from said reaction mixture. The organometallic compounds are useful in semiconductor applications as chemical vapor or atomic layer deposition precursors for film depositions.

Description

FIELD OF THE INVENTION [0001] This invention relates to processes for producing organometallic amidinate compounds, a method for producing a film or coating from the organometallic amidinate compounds, and ruthenium amidinate compounds that are hydrogen reducible and deposit in a self-limiting manner. The organometallic amidinate compounds are useful in semiconductor applications as chemical vapor or atomic layer deposition precursors for film depositions. BACKGROUND OF THE INVENTION [0002] In existing semiconductor devices, transistors communicate with one another via an elaborate series of copper interconnects connected through a series of metal layers above the transistor. To minimize the capacitive coupling between these interconnects, the space between is occupied by a material with a low dielectric constant (i.e., low-K materials). To prevent the diffusion of copper into this low-K material, a composite barrier is put in place. Current practices use physical vapor deposition t...

AI Technical Summary

Benefits of technology

[0021] This invention relates in particular to depositions involving amidinate-based ruthenium precursors. These precursors may have advantages over the other known precursors, especially when utilized in tandem with other ‘next-generation’ materials (e.g., hafnium, tantalum and molybdenum). These ruthenium-containing materials can be used for a variety of purposes such as dielectrics, barriers, and electrodes, and in many cases show improved properties (thermal stability, desired morphology, less diffusion, lower leakage, less charge trapping, and the like) than the non-ruthenium containing films. These amidinate-based ruthenium precursors may be deposited by atomic layer deposition employing a hydrogen reduction pathway in a self-limiting manner, thereby enabling use of ruthenium as a barrier / adhesion layer in conjunction with tantalum nitride in BEOL liner applications. Such amidinate-based ruthenium precursors deposited in a self-limiting manner by atomic layer deposition may enable conformal film growth over high aspect ratio trench architectures in a reducing environment.

[0022] The invention has several advantages. For example, the processes of the invention are useful in generating organometallic compound precursors that have varied chemical structures and physical properties. Films generated from the organometallic compound precursors can be deposited in a self-limiting manner with a short incubation time, and the films deposited from the organometallic compound precursors exhibit good smoothness.

Problems solved by technology

This will result in a decrease of the surface area to volume ratio of the interconnect, concomitant with an increase in the volume occupied by the diffusion barrier.

As the barrier occupies more of the interconnect channel space, the effective resistivity of the interconnect increases for two reasons: first, decrease in the size of the interconnect and second, copper / barrier surface scattering of electrons becomes a more critical issue.

Unfortunately, no chemistries exist that can deposit tantalum metal using atomic layer deposition.

Without tantalum, copper delaminates from the tantalum nitride film compromising device performance.

Oxygen based chemistries are incompatible with a BEOL integration sequence since the presence of trace amounts of oxygen within the deposited film could diffuse into the copper channel resulting in the formation of copper oxides compromising device performance.

In addition to being hydrogen reducible, chemistries should deposit in a self-limiting manner.

The problem is that there are no known suitable hydrogen reducible ruthenium complexes of sufficient volatility for use as atomic layer deposition precursors, and as such, no self-limiting, hydrogen reducible precursors have been identified.

The economics associated with such processes together with the rigid requirements of the electronics industry make the synthesis of organometallic precursors challenging.

Images