Critical dimension

from Wikipedia, the free encyclopedia

Under critical dimension ( English critical dimension , CD), rare critical dimension or critical dimension , is understood in the semiconductor and microsystems technology , one or more of the circuit designer and process engineer defined sizes in a test structure, the size of systematic information on the quality of the production of a Process step allows, for example a line width or the diameter of contact holes. In addition to checking the overlay offset, checking the CD value is one of the most important steps in the production of microelectronic products.

Measurement

The measurement of the CD value can be carried out using different measurement methods, depending on the structure size and the effort that you want to invest in the measurement. In the early stages of microelectronics, when the typical structure sizes were still in the micrometer range and above, the CD was determined using optical microscopes . With the steady shrinking of the structures, this was no longer possible due to the diffraction effects and other techniques were used. Today, CD measurements are usually carried out using scanning electron microscopy with low acceleration voltages (U B, max.  ≈ 1.5 keV) and automatic image analysis. The complex shape of a structure is initially described in the form of a simplified model, for example a trapezoidal cross-section. The intensity profile of the secondary electrons obtained from the scanning electron microscope image is then assigned to the position of the top edge ( top cd ), the bottom ( bottom cd ) or another prominent point of a trench structure and the CD value is determined at these points. This modeling or the choice of the evaluation algorithm is also one of the biggest sources of error for the determination of the CD value, so that the methods used should be verified with other measuring techniques and should only be changed with caution during production control. Other methods are scatterometry , also known as optical CD measurement, or atomic force microscopes with specially shaped tips.

Scatterometry is a relatively new form of CD measurement in which the spectra of light reflected by a periodic trench or line structure is evaluated. This reflection is then compared with known / calculated spectra of a structural model and the structural parameters, e.g. B. the CD value at the trench opening determined. Scatterometry offers the advantage of a fast measurement which, unlike scanning electron microscopy, for example, does not stress or damage the photoresist or layers made of sensitive low-k dielectrics . It also offers the option of determining other structural parameters, such as layer thicknesses or flank angles. However, these possibilities become practical due to the complexity of the structural model and the associated exponentially increasing effort of the compensation calculation or the size of the spectrum database used. In addition, sufficient measured values ​​or parameters (angle of incidence, wavelength , polarization , etc.) must be available so that the spectrum comparison is really unambiguous.

CD determination using atomic force microscopy is a very slow and complex method that is more likely to be regarded as experimental or intended for individual measurements. A specially shaped tip is guided along the structural wall and the structural floor. In this way, a three-dimensional profile of the structure is obtained and the CD values ​​can be calculated. With a so-called boot tip , it is also possible to measure undercut structural profiles.

Typical measurement structures

In production control, the CD measurement is usually carried out in special test structures for process setting, so-called tuning forks , i.e. a fork-shaped structure for process coordination. They consist of dense, semi-dense and isolated trenches or lines, for example in or made of photoresist. The different structure densities are important because they have a direct influence on the exposure process. The structures are usually oriented in the X or Y direction on the substrate parallel to the scribing frame. Similar structures consisting of dense, semi-dense and isolated hole structures also exist for via planes. These test structures are usually located in the scratch frame of the dies. In addition, there are also (mostly scaled-down) test structures in the area of ​​the chip itself. In addition, some critical parameters are also measured directly on the circuit structures, for example in the case of SRAM blocks, the gate structure or distances in the case of multiple structuring.

Importance and commitment

In industrial production, the measurement and evaluation of CD values ​​takes place mainly in the area of ​​photolithography and etching of structures. Nowadays the measured values ​​flow directly into the process control of subsequent batches of the same process step. In addition to the absolute value of the CD, the statistical distribution is also important for the function of a circuit during production. For example, the CD value of the gate electrode directly influences the electrical properties of the transistor, for example a transistor with a shorter gate length can switch faster. In order to ensure the electrical function of the entire circuit, the CD value should, on average, correspond to the target value from the design (compliance with leakage currents, etc.) and spread around this value as little as possible, otherwise the transistors switch at different speeds and run time problems with one another or with one another occur with the clock signal . The CD measurement is not only carried out during the production of integrated circuits and microsystems, but also during the production of the photomasks required for this , which are produced using similar processes.

The developed photoresist structure is measured in photolithography. In production control, these are the aforementioned structures of dense, semi-dense and isolated trenches or lines (in the FEoL and the conductor track levels ) and holes (the later vias between the conductor track levels). By measuring structures of different densities, working points for exposure are determined and monitored, because in today's photolithography in the lower wavelength range, the CD value for dense and isolated structures with the same dose and focus is usually different. This shift should be minimized as far as possible during production. In addition, there are special CD measurements for the evaluation and process setting, for example of manufacturing weak points caused by OPC . The CD measurement of photoresist structures is suitable because focus and exposure dose, two of the main process influences in photolithography, shape the shape and size of the structures and thus have a direct influence on the CD value. As a rule, measurements are taken in a central area of ​​the top and bottom of the photoresist, since the flank angles of the photoresist structures are in most cases very steep and therefore do not require a separate evaluation in production.

When etching structures, for example contact holes, structured photoresist layers are generally used as a mask for the etching of an underlying layer. The parameters of interest in this area are usually the CD value at the top and bottom of the structure. These values ​​are important, among other things, for assessing the flank angle, important for a subsequent production of a conductive metallic seed layer for a subsequent electrochemical deposition of the conductor tracks, and generally the size of the contact hole at the bottom of the hole. Furthermore, the uniformity of the CD values ​​over the entire substrate, i.e. from the center of the wafer to the edge, is an important control value in all areas in order to enable the same electrical properties of the integrated circuits in all areas.

literature

  • Chris Mack: Fundamental Principles of Optical Lithography: The Science of Microfabrication . 1st edition. John Wiley & Sons, 2007, ISBN 978-0-470-01893-4 , Section 8.2 Critical Dimension Control , p. 299–314 (Further information on the influence of system components on the CD value in photoresist structures).

Individual evidence

  1. ^ P. Rai-Choudhury: Handbook of Microlithography, Micromachining, and Microfabrication: Microlithography . SPIE Press, 1997, ISBN 978-0-8194-2378-8 .
  2. ↑ a b cf. CG Frasc, W. Hiifikr-Grohne, E. Buhr, K. Hahm, H. Bosse: Analysis and Comparison of CD-SEM Edge Operators: A Contribution to Feature Width Metrology . In: Günter Wilkening, Ludger Koenders (Ed.): Nanoscale Calibration Standards and Methods . John Wiley & Sons, 2006, ISBN 978-3-527-60687-0 , pp. 385-403 , doi : 10.1002 / 3527606661.ch29 .
  3. ↑ Chris Mack: Fundamental Principles of Optical Lithography: The Science of Microfabrication . 1st edition. John Wiley & Sons, 2007, ISBN 978-0-470-01893-4 , pp. 307-310 .
  4. ^ A b Robert Doering, Yoshio Nishi: Handbook of Semiconductor Manufacturing Technology, Second Edition . CRC Press, 2007, ISBN 978-1-57444-675-3 , Section 24.2 Metrology for Lithography Processes: Critical Dimension Measurement and Overlay Control , p. 24-5-24-17 .
  5. ^ Robert Doering, Yoshio Nishi: Handbook of Semiconductor Manufacturing Technology, Second Edition . CRC Press, 2007, ISBN 978-1-57444-675-3 , Section 24.2 Metrology for Lithography Processes: Critical Dimension Measurement and Overlay Control , p. 24-11-24-12 .