Method of controlling plasma etch process

Details Method of controlling plasma etch process Method of controlling plasma etch process

2002-02-14

2004-03-09

Vigushin, John B. (Department: 2829)

Semiconductor device manufacturing: process

Including control responsive to sensed condition

Optical characteristic sensed

C438S005000, C438S007000, C438S008000, C438S014000, C438S016000, C216S058000, C216S059000, C216S060000, C216S061000

Reexamination Certificate

active

06703250

ABSTRACT:

FIELD OF THE INVENTION
This invention generally relates to plasma etching and more particularly to a method for real-time monitoring of reactant-product concentration ratios including in-situ adjustment of etching parameters.
BACKGROUND OF THE INVENTION
In semiconductor fabrication, various layers of insulating material, semiconducting material and conducting material are formed to produce a multilayer semiconductor device. The layers are patterned to create features that taken together, form elements such as transistors, capacitors, and resistors. These elements are then interconnected to achieve a desired electrical function, thereby producing an integrated circuit (IC) device. The formation and patterning of the various device layers may be accomplished using various fabrication techniques including oxidation, implantation, deposition, epitaxial growth of silicon, lithography, etching, and planarization.
One important etching procedure for forming semiconductor features includes reactive ion etching (RIE) performed by a plasma of reactive ions. In the field of plasma etching, optical emission spectroscopy (OES) has been widely used for end-point detection by monitoring the intensity of wavelengths of either reactive species or etch products. More specifically, during an RIE process, plasma discharge materials, such as etchant, neutral, and reactive ions in the plasma, are continuously excited by electrons and collisions, giving off emissions ranging from ultraviolet to infrared radiation. An optical emission spectrometer diffracts this light into its component wavelengths and determines the intensity at a particular wavelength. Since each species emits light at a wavelength characteristic only of that species, it is possible to associate a certain wavelength with a particular species, and to use this information to detect an etching endpoint.
Also useful for monitoring reactants and products in a plasma etching process includes mass spectroscopy, where a molecule is broken into fragments and then analyzed typically using a magnetic field to determine the mass of the constituent fragments.
In forming a feature in a layer of material in a multilayer semiconductor device, one will typically etch away one type of material included in one layer of the device until another different layer of material which typically has a low selectivity to etching is reached, for example, an etch stop layer. Since the etch stop layer is typically a different material having a low reactivity with the reactive ion species, a discernible change in either or both the concentration of reactive ion species and the reactively etched species may be easily observed. Since the concentration of given plasma species is proportional to an emitted wavelength of light, the concentration of a given species may be tracked by monitoring the intensity at a given wavelength.
One problem with end-point detection using OES or any other concentration monitoring method occurs where no change in material such as a material interface is present during the etching process thereby providing no readily discernable change in reactive species or reactively etched (product) species. However, there are several applications where no material interface is present yet a known feature etching depth, for example, a trench or via etching depth into a monolithic substrate, for example, silicon or dielectric material, is required.
For example,
FIGS. 1A and 1B
show exemplary semiconductor features where an etching depth into a monolithic material is required, yet not material interface is present to detect an end point. FIG
1
A shows typical shallow trench isolation features (STI) e.g.,
12
A,
12
B, etched into a silicon oxide containing material layer
14
, for example a low-k carbon doped oxide.
FIG. 1B
, shows typical dual damascene structures where an etching stop layer is not included at the interface of the via opening e.g.,
16
A,
16
B and the trench opening e.g.,
18
A,
10
B. Although there are several approaches to etching a dual damascene structure, the structure in FIG
1
B for example, may be made by first etching the via opening e.g.,
16
A, and then etching the trench line opening e.g.,
18
A into dielectric layer
20
overlying the via opening e.g.,
16
A by a two step photolithographic patterning process. In both of these cases no material interface is present yet, a known etching depth is critical to successful device performance.
Several approaches have been proposed to overcome the lack of a clearly discernible material interface including tracking the amount of either or both reactants and plasma etching product species during the etching process. For example, statistical techniques of the prior art including typically single out one or more individual plasma species components to monitor, the selection of the components determined by statistical analysis of previous etching procedures. In general, the prior art methods use OES to monitor one or more individually selected wavelengths of plasma species together with a statistical analysis of the results of the optical spectrum of a product produced or reactive ion species consumed. In this regard reference is made to U.S. Pat. Nos. 5,288,367, 5,658,423, and 6,238,937, which are incorporated herein by reference.
One problem with this approach includes the several assumptions involved in performing multi-variant statistical analyses, for example, selecting the number and type of components to be monitored in a principal components analysis (PCA). For example, the number and type of analyzed components may change when the pattern density to be etched changes although the material and the desired etch depth and profile remain the same. For example,
FIGS. 1C and 1D
show an exemplary increased density in an etching pattern, for example STI features
12
A,
12
B,
12
C and dual damascene features
16
A-C,
18
A-C compared to
FIGS. 1A and 1B
, respectively. With an increased etching pattern density, which may change from device to device, the corresponding ratio of masked to unmasked etching areas change thereby altering expected behaviors of etching products and reactants. As the density of patterns to be etched varies from application to application, a ratio of masked material to unmasked material likewise varies with the various depths and profiles thus altering expected concentrations of both reactive species and etch products over time and making, for example, the selection of the number and type of components in a PCA analysis a complicated process.
In addition, the prior art method of comparing individually monitored components to previous etching processes increases at an exponential rate given the large number of variable parameters including pattern etching densities. Such methods have proven to be difficult to implement in real-time processing systems given the large number of variable parameters.
Another problem are the various plasma etching apparatus available for use, each of the etching apparatus having different behaviors in terms of species concentration evolution over time. For example, the use of high-density, inductively-coupled plasma ICP) or cyclotron resonance (ECR) , differ from conventional RIE in plasma formation causing significant differences in plasma behavior. Consequently, models developed for one process according to prior art statistical analysis methods are likely restricted to a particular etching process.
These and other shortcomings demonstrate a need in the semiconductor processing art to develop an improved method and system for monitoring an etching process including real-time analysis and feedback to alter in-situ etching parameters to improve a plasma etching process.
It is therefore an object of the invention to develop an improved method and system for monitoring an etching process including real-time analysis and feedback to alter in-situ etching parameters to improve a plasma etching process while overcoming other shortcomings and deficiencies in the prior art.
SUMMARY OF THE INVENTION
To achieve the foregoing and othe

Affiliated with

Also associated with

Rating

Method of controlling plasma etch process 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 of controlling plasma etch process, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method of controlling plasma etch process will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-3225525