Background

Current trends in precision engineering demand ever higher accuracies for industrial high-end production and measurement equipment. This requires control of positioning systems over measurement ranges from nanometres to millimetres in all 6 degrees of freedom (DoF). The application of high precision motion systems ranges from nanometrology (AFM, SEM) to industrial production technologies (machine tools, CMM, photo lithography) to large scale applications like telescopes. Improved precision engineering tools using thermal insensitive design principles is beneficial not only for tool manufacturers but also for European key-industries, especially smaller companies, in terms of more efficient production processes and reliable, better products, an important condition to ensure a competitive advantage of European industries on the world market. This will also help to reduce the number of defective parts, leading to savings in raw materials, and reductions of machine time per part.

 

Need for the project

Mechatronic motion systems are the basis of most production systems ranging from tool machines for wafer scanners in semiconductor circuit production to robotic applications as well as for associated measurement equipment like coordinate measuring machines and scanning probe microscopes. Positioning to the required precision is challenging under dynamically changing and possibly harsh conditions. A trade-off has to be established in being fast and accurate.

Wafer stepping devices of semiconductor industry are a major example of these demands. Larger wafer diameters and high production throughput poses tighter limits for the involved positioning tools and metrology platforms. More generally, the trend for higher integration in terms of packaging size and multi-functionality, such as in photonics and nanotechnology and for a broad range of applications (telecom, computation, navigation, aerospace), requires precision production equipment.

Needs for improved positioning control range from the nano-scale (metrology frames for AFM) and micro-scale (CMM, tomography stages) to mechatronics positioning in automotive and aerospace systems, including traditional cartesian motion systems as well as more general positioning devices (hexapods, goniometers). Similarly rich is the spectrum of required accuracies (sub-nm to metre scale), dynamics (fast microscopy MHz-scale scanning to low frequency mHz-scale noise spectra of astronomy instrumentation), and simultaneous multi-axes measurement and control. Selection and number of the targeted DoF, measurement range, uncertainty and temporal dynamics depend on the specific applications and the actual on-site ambient conditions. Highest level system validation in 6DoF to the nanometre and microradian can only be achieved in terms of a test facility at measurement laboratory level.

 

Scientific and technical objectives

The JRP focuses on the calibration of motion systems with high uncertainty requirements and the development of methods for the analysis of the associated measurement uncertainty as well as methods for the error mapping for real time corrections. Different approaches and instrumentation developed for the measurement of 6DoF motions as well as important aspects of motion systems like straightness and orthogonality will be analysed and compared. Additionally this JRP aims to improve nano positioning systems for the various future relevant high-tech fields not only by novel hardware technology but also by optimised sensor and actuator components as well as optimised measurement and control strategies. The project is organised in three workpackages:

  • Development and optimisation of interferometric sensors for traceable measurement of position, angle and straightness in cartesian motion systems with uncertainties in the nanometer range.
  • Development and comparison of methods and instruments for straightness and orthogonality measurement
  • Characterisation of motion systems in 6DoF. This work package includes the characterisation of motion stages for nanometrology with uncertainties down to the sub-nanometer range as well as those positioning devices with large angular motion possibilities.

 

Expected results and potential impact

Interferometry:

For traceable measurement of positioning devices, laser interferometers are a key technology. Laser interferometers are therefore widely used to characterize the positioning performance of machine axis. In this project a set of different interferometers will be developed. While in most industrial measurement systems embedded grating encoders are used, the developments in the JRP will base on interferometric techniques to establish a direct link to the definition of the length unit. A fibre based multi-axes capable interferometer for a test bed for nanopositioning devices has been setup and first investigations have been performed. Further investigations of different types of fibre based interferometers is work in progress.

Based on a technology using a CCD chip combined with an FPGA for fast data processing of interference patterns an simple optical setup has been designed to measure for the first time simultaneously all six degrees of freedom of a linear motion axis with a single interface. By using multiple regions of interest on the CCD only two chips will be necessary for all six degrees of freedom. This is intended  for faster characterisation of linear machine axes and tool inspection.

A optical arrangement of a ultra high resolution heterodyne interferometer for straightness measurement will allow for a detailed investigation of the influence of production process related mirror topographies and gradients of the refractive index of air in combination with different opening angles.

To support the work on interferometry a stabilized DBR laser diode at 632 nm wavelength will be developed. Based on a new low noise current source the laser diode has been characterized and will now be stabilized on an iodine absorption line and a material reference, which allows for the partly compensation of the refractive index and thermal dilatation of tools.

Measurement of straightness:

All Interferometers for multi axes positioning systems share the problem, that it is necessary to use plane mirror interferometers at varying positions. Therefore all this interferometric measurements suffer from topographic variations of the mirrors. Different methods for the measurement of straightness and orthogonalit, straightness interferometry, external calibration of the topography, self calibration and deflectometry will be implemented and compared in this project.

The external calibration of the measurement mirrors of CMMs by a Fizeau interferometer is a common method. One challenge of this method is the referencing of the external measured values to the machine coordinate system. Another limitation is the accuracy of the Fizeau interferometers in the order of 5 nm, which can be overcome by the use of deflectometric methods like direct deflectometry, difference deflectometry or exact autocollimation deflectometric scanning, which allows for sub-nm accuracy with a limited lateral resolution. As laser interferometer uses beam diameters of some millimetres only spatial resolutions larger than a fraction of the beam sizes have to be taken into account.

In mask measurement tools the preferred method for correcting straightness and orthogonality errors are self calibration methods by measuring sequentially in rotated and shifted positions of the mask. The limitation of this method is the reproducibility of the mask mounting and of the tool with changed load conditions. A comprehensive analysis of the measurement uncertainty of this method is still missing.

To use deflectometry for straightness measurements, the PTB length comparator has been equipped with a three channel heterodyne interferometer with a fibre coupled light source. This interferometer which is based on a design verified in the iMera+ plus project Nanotrace will allow for optical deflectometric measurements of straightness without the need for a prior calibration of the mirror topography associated with further reproducibility factors. Measurement series have been performed using a straightness encoder manufactured specially for this project by Heidenhain. For the validation of the deflectometric measurements, the scale has additionally measured with rotation and shifts relative to the Y-mirror to allow for self calibration techniques. The reproducibility of the deflectometric measurements were below 5nm. The differences of the averaged deflectometric measurements to the self calibration measurements were below 2.5 nm.

All this methods will be compared also including straightness interferometry to allow for a secure base for the choice of an appropriate method in ultra precision engineering.

Large range AFM metrology:

AFM as a scanning instrument is limited in the measurement range by the necessary time and the stability of the tool and the tip. The project deals with these problems in different ways. Two tools were tested regarding long time stability to perform tip wear investigations. New intelligent scanning strategies and associated data processing will be placed at the disposal in form of libraries. These libraries will also be available as part of the open source software GWYDDION. In the project the control electronics of AFM instruments will be changed to implement this libraries. Another possibility to enhance the measurement range is the use of high speed AFMs. These high speed AFMs in the moment do not make use of traceable metrology. A system developed at the University of Bristol was investigated and offline corrected. The integration of a interferometer for online correction is in progress.

To meet the requirement for more accurate traceable characterization of stages used for nanometrology, a test bed will be constructed using above mentioned newly developed interferometers.

Calibration of positioning devices with large angular motion ranges:

The most complex class of motion systems combines linear and large angular motion like Hexapods or a combination of linear and rotary axes. This allows for more universal solutions for measurement and production and will also become more important in the ultra precision engineering and also for nanotechnologies. The problem is, that proven techniques like laser interferometers cannot be used in the traditional way. One way to overcome this problem is the use of coordinate measurement machines with tactile or optical probing of reference points on the motion stage. This method is time consuming as after every motion step the reference points must be localised one after each other. Another solution is the use of laser tracers which uses balls as angle insensitive reflectors. Tracers are commonly used in large scale applications. By increasing the number of tracers they are able to dynamically measure the motion of a stage in all six degrees of freedom simultaneously. Both methods will be analysed and compared. At METAS first measurements of the position of balls attached to a 2D rotational stage has been performed on a micro coordinate machine. The stability of the position readings were about 1.5 nm RMS and a measurement uncertainty below 80 nm will be achieved.