Semiconductor device manufacturing: process – Coating with electrically or thermally conductive material – To form ohmic contact to semiconductive material
Reexamination Certificate
2002-07-29
2003-11-18
Nguyen, Ha Tran (Department: 2812)
Semiconductor device manufacturing: process
Coating with electrically or thermally conductive material
To form ohmic contact to semiconductive material
C438S679000, C438S685000, C700S121000
Reexamination Certificate
active
06649521
ABSTRACT:
BACKGROUND OF THE INVENTION
FIELD OF THE INVENTION
The invention relates to a method for determining relevant deposition parameters in i-PVD processes, in which metal atoms and metal ions are transported in a gas plasma, under the influence of electromagnetic fields, to a substrate and are deposited thereon.
The main difficulty in the PVD processes remains that of ensuring sufficiently conformal deposition of very thin films that are only a few nanometers thick in vias and trenches with an increasing aspect ratio. Such a problem is combated by the development of ionized PVD processes (“i-PVD”). Such processes promote the steep angle of incidence of the particles, which generally are of metal or metal compounds, on a substrate of a wafer. In the i-PVD processes, the transport of metal ions is optimized by the generation of a higher plasma density in combination with a greater ionization rate through the action of electromagnetic fields. The plasma used is typically an Ar
+
plasma. In the text that follows, Ar
+
is used as a representative of carrier gases in the plasma. The metal ions are produced by ionization of metal atoms, in the case of hollow cathode magnetron (HCP) or self-ionized plasma (SIP) within a special magnetron, or in the case of ionized metal plasma (IMP) below the target, by an inductively operated plasma. Moreover, the transport of the metal ions to the substrate and the energy distribution thereof is influenced by an alternating voltage applied to the substrate, which is referred to as the substrate bias. Examples of relevant metals or metal compounds are Ti, TiN, Al, Ta, TaN, Cu, W, WN, Ni, Co, Fe, etc.
Optimization of such a novel i-PVD process has the purpose of ensuring the required topological properties, such as conformity, base and side wall coverage, the morphological properties, such as density, grain size distribution, texture, and the electrical properties, such as the conductivity of the ultra thin, multifunctional layers that are to be deposited under the more stringent geometric requirements of future technologies.
To optimize the procedure for certain processes for producing identical process conditions in reactors made by different manufacturers and to allow the process window to be evaluated at all, it is a central objective to determine the particle flows f(Ar
+
) of Ar
+
ions, f(M
+
) of M
+
ions and f(M) of neutral metal (M). If there is an absence of or a low substrate bias voltage (U
(bias)
), a different ratio i=f(M
+
)/f(M) results in different degrees of base coverage, while in the case of a significant substrate bias voltage (with ion energies greater than the threshold energy required for re-sputtering), in addition to the ratio i, the flow of energy ions r=f(Ar
+
)/(f(Ar
+
)+f(M
+
)+f(M)) leads to effects such as re-sputtering, which increases the side wall coverage, beveling (i.e., preferential re-sputtering at edges), changed microstructure, and changed electrical properties. Therefore, the substrate bias voltage U
(bias)
is a further important variable.
Because in the case of a low substrate bias voltage the deposition rate works out as d≈f(M
+
)+f(M), the relevant process parameters that determine the deposition properties in question are d, i, r, and U
(bias)
. In such a context, the angular distribution of the incident particles plays a minor role because, in the i-PVD processes, unlike in conventional PVD processes, the neutral particles are virtually isotropic and the ionized particles have a strong forward directionality. The ionized particles have an opening angle that changes only slightly with the substrate bias voltage U
(bias)
.
The substrate bias voltage U
(bias)
can, in principle, be determined by an electrical measuring configuration by measuring the temporal mean of the substrate potential against the plasma potential. In practice, however, such a possibility does not exist on commercially available i-PVD installations. Therefore, a different determination method is desirable for the substrate bias voltage.
While in the absence of or with a low substrate bias voltage, the deposition rate is given by f(M
+
)+f(M), the determination of the abovementioned ratio i and of the flow r of the energy ions requires separation of the ionized flows of the total particle flow and a distinction between f(Ar
+
) and f(M
+
), for example, based upon mass. This can be achieved with a mass spectrometer that is directly connected to the i-PVD chamber. However, such a method cannot be employed in a manufacturing environment. Therefore, hitherto, in practice rapid and accurate determination of the relevant process parameters has not been possible.
Accordingly, there is currently no suitable way of providing an installation with a type of control configuration on the basis of which it can decide whether or not relevant process parameters are moving out of the desired target corridor.
SUMMARY OF THE INVENTION
It is accordingly an object of the invention to provide a method for determining the relevant ion and particle flows in i-pvd processes that overcomes the hereinafore-mentioned disadvantages of the heretofore-known devices and methods of this general type and that allows rapid and accurate determination of the deposition parameters that are, in this case, of relevance, including in manufacturing, without expensive refitting or upgrading of commercial i-PVD installations.
With the foregoing and other objects in view, there is provided, in accordance with the invention, a method for determining relevant deposition parameters in i-PVD processes, including the steps of transporting metal atoms and metal ions in a gas plasma, under an influence of electromagnetic fields, to a substrate and depositing the metal atoms and the metal ions on the substrate by (A) calculating reaction rates for desired reagents of the gas plasma and of a material to be deposited, the material selected from at least one of the group consisting of a metal and a metal compound, (B) simulating an edge coverage of a predetermined structure with the deposited material based upon the reaction rates calculated in step (A) with simultaneous systematic variation of the relevant deposition parameters, (C) compiling, from results of the simulation of step (B), variant tables for the relevant deposition parameters for the predetermined structure, (D) executing an i-PVD process in a selected gas plasma and using a selected material over the predetermined structure, (E) experimentally verifying the simulated edge coverage for the product produced in step (D) with an imaging process, and (F) determining the relevant deposition parameters by comparing the experimental verification produced in step (E) with the variant tables compiled in step (C).
The invention includes a method by which the particle flows d, i, r, and, if appropriate, also the substrate bias voltage U
(bias)
, can be read from a single experimental verification, obtained by an imaging process, of a predetermined structure formed by i-PVD. Such a method allows the important process parameters to be determined without altering the reactor and with little outlay.
In recent times, it has become possible to calculate the reaction rates of Ar
+,
M
+
, and M on the substrate that is to be metallized with sufficient accuracy using molecular dynamics methods (for copper cf. D. G. Coronell et al., Applied Physics Letters No. 73 (1998), 3860). The inventors themselves have used such molecular dynamics methods to calculate the reaction rates for Cu, Al, Ti, and Ta. Given knowledge of these reaction rates, it is possible, for various assumptions about the values of d, i, r, and U
(bias)
, to simulate the edge coverage of a contact hole or a similar structure (cf. U. Hansen et al., Physical Review, B 62 (2000) 2869). It has been found that the edge coverage of such a structure has characteristic differences for all these variants. Conversely, if the edge coverages have been calculated for all these
Kersch Alfred
Ruf Alexander
Greenberg Laurence A.
Infineon - Technologies AG
Locher Ralph E.
Nguyen Ha Tran
Stemer Werner H.
LandOfFree
Method for determining the relevant ion and particle flows... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method for determining the relevant ion and particle flows..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method for determining the relevant ion and particle flows... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3182984