Optics: measuring and testing – By light interference
Reexamination Certificate
1999-12-30
2003-02-25
Font, Frank G. (Department: 2877)
Optics: measuring and testing
By light interference
C356S517000
Reexamination Certificate
active
06525820
ABSTRACT:
FIELD OF THE INVENTION
The present invention relates to an apparatus for in situ real time monitoring of atomic and molecular fluxes by using Rayleigh scattering. The flux can be generated by a suitable source, such as an effusion cell used during molecular beam epitaxy. The present device uses: a light source, such as a helium neon laser for generating a monochromatic coherent beam of light; a source for emitting a flux of atoms or molecules; a high precision mirror assembly capable of providing a sufficient number of reflections though the flux beam to produce a detectable interference pattern; and an interference pattern detector to track the changes in flux.
BACKGROUND OF THE INVENTION
There are various techniques available for depositing high purity materials on substrates by either atomic or molecular growth. Such techniques include, but are not limited to, molecular beam epitaxy (MBE), chemical beam epitaxy (CBE), chemical vapor deposition (CVD), organometallic CVD (OMCVD), plasma enhanced CVD (PECVD), physical or RF sputtering, and laser ablation. All of these techniques require an adequate means for measuring the rate and composition of the material being deposited. The ideal means for such measurements is a flux monitor. To illustrate the need for an improved flux monitor system, one can consider MBE, an established technique for fabricating high quality electronic and opto-electronic devices in both research and industrial environments. The advantages of MBE over other growth techniques include the ability to produce high purity materials with controlled composition, layer thickness, dopant concentrations, and structure. MBE growth is achieved by directing the output of one or more effusion cells onto a heated substrate in an ultra-high vacuum chamber. The effusion cell's output typically consists of atoms of the desired growth and doping constituents to be deposited. The rates at which the atoms fluxes are emitted determine the stoichiometry and growth rate of the substrate overlayers.
A high level of interest in MBE has been driven by the advantages that III-V and II-VI semiconductors have over silicon semiconductors namely, increased carrier mobilities, optical transitions, lattice matching and band gap engineering. The increased carrier (both hole and electron) mobilities enable devices fabricated from these other materials to operate with faster switching times than their silicon counterparts. The use of ternary and quaternary semiconductors allows exotic overlayers and multilayer structures to be grown and lattice matched to substrates, thus minimizing the stress and its effect on device properties. Band gap engineering, accomplished by altering the stoichiometry of ternary or quaternary semiconductors, is used to produce semiconductor lasers, high speed transistors, optical switches and other devices by creating population inversion layers, or sinks, or traps for carriers.
Exploiting the true potential of these advantages requires precise control over the stoichiometry and thickness of the deposited layers which, in turn, requires detailed knowledge of the amount of material being deposited and the rate at which it is deposited. Additionally, it would be extremely helpful to have these pieces of information while the sample is still being grown so that any corrections that need to be made to the fluxes can be made prior to removing the sample from the growth chamber for ex-situ analysis and testing.
Several different techniques are currently in use to monitor growth. These techniques are be either direct or indirect techniques. Direct techniques monitor what happens on the substrate as material is deposited, while indirect techniques monitor the material that is sent to the substrate.
Direct techniques include reflection high-energy electron diffraction (RHEED), ellipsometry, auger electron spectroscopy (AES), and x-ray photoelectron spectroscopy (XPS). These have the advantage that they directly determine either the amount of material deposited, or the surface composition. This is also their weakness. Such measurements are made after the material has been deposited, thereby making it impossible for a priori control of the deposition.
RHEED utilizes a grazing incidence, high energy (~10 keV) electron beam scattered from the target surface to produce an interference pattern on a phosphor screen. The symmetry of the diffraction pattern generated from what is essentially a two-dimensional crystal structure is very useful in tracking both the overall growth rate of the sample and the growth mode. Knowledge of the rate of growth and the growth mode allows calculation of the incident flux. The intensity of the zero-order spot can be monitored for oscillations produced by periodic variations in surface roughness produced when a new layer is grown. The period between oscillations is equal to that required to deposit a single monolayer (ML) of material assuming that the growth mode is layer-by-layer. For other growth modes, such as step flow or for very thick layers, the oscillations can be weak, or not present at all.
AES is commonly used for compositional determination in situ. AES uses a high energy (~10-15 keV) electron to ionize an atom to remove a core electron. A second electron decays into the core hole and imparts its energy to a third electron that escapes from the sample and is captured by an energy analyzer. This technique is very sensitive to local changes in the chemical environment and is used to detect not only the species present but also its bonding arrangement. This process relies on multiple electron transitions to generate a signal, and because of this, the signal strength is relatively small. To offset this small signal the detector must be placed in close proximity to the sample and long acquisition times (10+ seconds) are required. Positioning the detector in close proximity to the sample results in material being deposited on the detector which alters the measured kinetic energies of the ejected electrons and eventually causes it to fail. For these reasons AES analysis is typically done outside of the main growth chamber.
Photoemission spectroscopy is very similar to AES, it utilizes a beam of monochromatic x-rays to probe tightly bound core electrons. The photoemitted electrons escape from the sample and are collected in an energy dispersive spectrometer like that used in AES. Their energy is characteristic of both the species from which they were emitted and their local chemical environment. Differences in binding energies allow differentiation between surface and bulk atoms and provide insight into their bonding configuration. XPS can readily identify compositional changes on the order of 1%. The disadvantage of this method, in addition to problems associated with the detector discussed above, is that the electrons emitted can have high kinetic energies, thus giving them long escape depths that decreases the surface sensitivity of the technique.
Another conventional technique, ellipsometry, utilizes a laser beam, or light from a high pressure Xe lamp that passes through a polarizer and is incident on the surface at or near Brewsters angle. The reflected beam passes through a rotating analyzer and is detected. The ratio of reflection coefficients, as a function of ellipsometric angle, is determined. The data is presented by graphing the real component versus the imaginary components of the ratio. Ellipsometry relies on the difference between the refractive index of the overlayer and that of the substrate. For this reason, ellipsometry is ineffective for determination of growth of homoepitaxial films, or other films, where the indexes of refraction are equal. Ellipsometry requires a long data acquisition time in order to obtain sufficient signal for reliable results, thereby rendering it unsuitable for use as a real-time control technique.
Indirect methods, such as conventional optical flux monitoring (OFM), utilize atomic absorption and are capable of determining the flux of the incident material. Multiple monochromatic beams are
Font Frank G.
Gradient Technology
Natividad Philip
Naylor Henry E.
LandOfFree
Rayleigh scattering optical flux monitor does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Rayleigh scattering optical flux monitor, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Rayleigh scattering optical flux monitor will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3180444