Scatterometry

The scatterometry (dt. About scattered radiation measurement ) is a non-destructive method for the analysis of particles and periodic surface structures using elastic scattering and diffraction of electromagnetic waves (often visible light ). The sizes of structures can reach into the nanometer range, that is to say also below the Abbe resolution limit (see resolution ) of the light used.

In addition to evaluating the diffraction of visible light, there are also similar methods that generally use electromagnetic radiation of shorter wavelengths, e.g. B. X-rays , or use particle radiation . In German, these methods are generally not referred to as scatterometry, but rather as diffractometry , for example neutron and X-ray diffractometry .

Classification

Principle of 2θ scatterometry on a trench structure with the representation of typical parameters

Scatterometric methods can be divided into the following sub-methods with regard to the radiation source used and the type of detection:

1. Detection type / measurement setup
   1. 2θ-scatterometry: measurement of the zeroth diffraction order of the reflected / scattered light by varying the angle of incidence θ ( English angle-resolved scatterometry ) as well as in consideration of the s- and p- polarization (similar to the ellipsometry ). Originally only monochromatic light was used here.
   2. Scatterometry with normal incidence ( English normal-incidence scatterometry ): Measurement with normal incidence of light with variation of the wavelength, sometimes also in connection with the evaluation of the ± 1. Diffraction order
   3. Dome scatterometry: measurement of higher diffraction orders using a diffusing dome ( English dome )
   4. Fourier (transform) scatterometry
2. Radiation source
   1. Spectral scatterometry: Scatterometry using a broadband radiation source and the detection of a larger spectral range
   2. Laser scatterometry: Scatterometry using a laser or monochromatic light

In addition, there are scatterometry methods that can be viewed as a combination of different basic shapes or that have been supplemented by additional parameters, for example by a further angle of rotation in the sample plane (φ scatterometry).

functionality

In the following, the functionality of the method will be briefly described using the example of a depth measurement using spectral scatterometry. For the measurement, a field several tens of micrometers in size with a periodic line structure is irradiated with visible light and the intensity and / or the polarization state of the light reflected on the structures is detected. As a rule, only the zeroth diffraction order is considered. Depending on the method, the light detection can take place as a function of the angle, the wavelength or both. The detected spectrum depends on the structure parameters such as the line width, the depth, the flank angles of the side walls or the material, and is characteristic of the diffractive structure in one or more areas. A direct inverse determination of the structural parameters from the measured spectrum is not possible, however, since the necessary equations cannot be solved analytically. The inverse determination is therefore carried out indirectly with the aid of a compensation calculation on a previously created model that describes the essential target shape of the structure. The adjustment calculation therefore only determines the parameters roughly recorded in the model more precisely.

application

Scatterometric methods are based on the elastic scattering and diffraction of light on particles and non-planar surfaces. They are therefore used on the one hand to detect particles in gases, liquids or on surfaces and on the other hand to characterize periodic or random surface structures.

Spectral scatterometry has become an important method for production monitoring in semiconductor technology in recent years . There she serves u. a. to characterize periodic profiles such as the depth or side angles of etched structures and the critical dimension (CD); the method is therefore also referred to in English as optical critical dimension metrology (OCD metrology ). In addition, it is also suitable for determining the overlay offset ( English diffraction based overlay , DBO) or for measuring buried structures, which opens up new possibilities for process control , especially with advanced techniques such as FinFETs .

As an optical measuring method, it can be integrated into other optical measuring devices, such as devices for determining the overlay offset or the layer thickness (e.g. reflectometer , ellipsometer), and also in process systems. The latter enables very fast feedback to the process control system (sometimes in real time ) and thus better process control.

A disadvantage of scatterometry measurements is the relatively high space requirement for the test areas, these are often 50 microns x 50 microns in size and can therefore normally only in the scribe line ( English scribe line placed between the actual chips). In-die measurements are therefore very rare, since valuable chip area would be occupied here. In certain cases, however, periodic structures of the circuits can also be used for the measurement, for example larger DRAM or SRAM blocks on the chip. An important disadvantage, which is particularly important for CD measurements, is that OCD measurements generally do not allow the measurement of isolated line structures. In addition to the CD value from dense line-trench structures, such information is required in today's photolithographic processes for the assessment of the exposure dose influence. Therefore, in the industrial environment, the critical dimension is usually not measured using scatterometric methods, but rather using scanning electron microscopes .

Advantages and disadvantages

Scatterometric measurements are non-destructive and non-contact methods. For use in process control in the manufacture of semiconductor products, they are characterized by a high level of sensitivity, even to small changes in material or structure, and are usually relatively easy to implement in terms of system technology. Compared to alternative methods such as atomic force microscopes , the scatterometric determination of profile depths is significantly faster, less prone to failure in high throughput operation and also offers the possibility of determining further profile parameters such as line width, side wall angle and layer thickness. However, it is disadvantageous that the measurements of the desired material and geometry data take place indirectly, that is, they require an adaptation calculation of a previously defined model. A direct determination of the material data of an unknown system is usually not possible. Larger deviations can possibly only be determined with larger error values. In addition, the determination of isolated line widths, as they are often required for process control in photolithography, is not possible.

Individual evidence

1. ^ Scatterometry. University of Stuttgart, accessed on March 11, 2015 . 
2. ^ Gary S. May, Costas J. Spanos: Fundamentals of Semiconductor Manufacturing and Process Control . John Wiley & Sons, 2006, ISBN 0-471-78406-0 , pp. 97 ff .
3. ↑ Thomas Schuster: Simulation of light diffraction on cross-grating structures and their application in scatterometry . 2010, Section 5.3 Historical development and variants of scatterometry , urn : nbn: de: bsz: 93-opus-51081 (dissertation, University of Stuttgart, 2010).
4. ↑ Lifeng Chi: Nanotechnology: Volume 8: Nanostructured Surfaces . John Wiley & Sons, 2010, ISBN 978-3-527-31739-4 , Section: Optical Critical Dimension Metrology: Scatterometry , pp. 181 ff .
5. ^ Peter Van Zant: Microchip Fabrication: A Practical Guide to Semiconductor Processing . 5th edition. Mcgraw-Hill Professional, 2004, ISBN 0-07-143241-8 , pp. 454 .
6. ^ Dieter K. Schroder: Semiconductor Material and Device Characterization . 3. Edition. John Wiley & Sons, 2006, ISBN 0-471-73906-5 , pp. 601 .