Measurement of surface roughness by AFM
Project Description
AFM as high resolution instruments are used very often to determine roughness parameters of technical surfaces. However, up to now - as for the optical microscope - there exit no rules or guidance how to perform the measurements and how to analyze them in order to achieve a satisfying degree of comparability. Furthermore our experience e.g. among members of the German VDI FA 3.41 guideline committee has shown that the determined parameters depend critically on the properties of the tip, the digital resolution of the area investigated as defined by the scan range and the number of data points and lines, the dynamic behaviour of the instrument and the settings of the control parameters, plus the data treatment prior to the roughness analysis, e.g. leveling and filtering, and finally the software used for the determination of the roughness parameters.
Within this project we will investigate some samples which might be useful to verify the instruments' properties for areal measurements. A draft guideline under development in VDI FA 3.41 committee is available for this European comparison in this field. In order to exclude the influence of different approaches to AFM image analysis, the raw data of the participants' measurements will be sent to the coordinator for central analysis with the same software and the same analysis strategy. Additionally, it might be agreed to send the used cantilevers to the coordinators for high-resolution SEM inspection.
The samples and the guidelines investigated within this project should be a good base to set-up international standards in ISO/TC 201 or ISO/TC213.
Investigation of roughness measurement will be done by using two artifacts of step hight (PTB step height standards of nom. heights of 20 nm and 40 nm), a flatness sample, two samples with Ge-Si-nanodots (nearly isotropic roughness), lapped Si chips (three samples of the novel RnS-series by Simetrics company, with nominal Sq-values from some ten to a few hundred nm), Si chip with monoatomic steps. Optionally: an ion-track membrane (nom. diameter of holes 50 nm), resolution standard of type RS-N by Simetrics (chip with 9 one-dimensional gratings of different pitch, the finest nom. 300 nm)
Progress Report 2024-09-12
It was originally agreed to circulate 7 samples, and a circulation schedule was drafted for 2014/2015, but shortly before the start several participants requested postponement back then, as they faced technical problems at their instruments. Furthermore, the number of 7 samples envisaged in the draft Technical Protocol was regarded as too many, taking into account that high-quality AFM measurements are known to be time-consuming. The start was halted, and in the meantime the Technical Protocol was reduced to 5 samples, which was approved by the participants. BEV (Austria) withdrew before the relauch, while NPL(UK) joined, and NIST (USA) signalled to take part as external partner from another RMO at the end of the circulation.
Relaunched in June 2018, with reduced measurement effort and tight conditions: one month per institute only, institute to be skipped if not ready. At this stage, DFM (Denmark) had to withdraw, as their instrumentation was not ready in the envisaged timeslot.
The relaunched circulation passed smoothly in 2018/2019 without major delays and without changes in the circulation plan: 9 European NMIs measured within the assigned timeslots, control measurement were done at PTB in late 2019. It was found that the polished Si chip, circulated for noise and flatness measurements, got heavily contaminated during circulation, complicating analysis - but all other samples proved to be still sufficiently clean and stable so that obtained results can be compared well.
NIST measurements had to be postponed repeatedly due to the pandemic, and finally cancelled in 2023 as it was found that the sample quality is gradually deteriorating due to aging/contamination.
Analysis: Two-fold in parallel to identify possible bias by operator
1) central analysis by pilot lab according to uniform procedures – all participating European NMIs have submitted raw data (partly delayed due to pandemic, data fully analysed by pilot)
2) analysis by participants – most reports received well in time, last missing reports received 2023
The central analysis by pilot was completed in mid-2023 and Draft A written, which is sent to the participating NMIs in Sep. 2024. It might still be discussed which further roughness parameters besides Sa might be presented and discussed in the report.
Download project presentation given at TC Length meeting 2023 >>
Progress Report 2018-10-31
The progress report can be downloaded here >>