Specific Process Knowledge/Characterization/Tencor P17

Tencor P17 Stylus Profiler

The P17 Stylus Profiler from KLA Tencor is used in a similar manner to the Dektak XTA for profiling surfaces with structures in the micro- and submicrometer range as well as for measuring stress.

Compared to the DektakXT, the P17 has more advanced options for stress measurements: It allows the user to measure a stress map with up to 5° radial resolution. Programming and analyzing a sequence of predefined scans in fixed locations on a wafer is also easier and the manual for doing it much better for the P17 than for the DektakXT. A disadvantage of the P17 is that is can be hard to locate structures as the maximum field of view of the camera is 1x1.5 mm. We recommend having a map of the sample design available so you can easily locate the features of interest. Otherwise the P17 is easy to use, fast, and accurate, just like the DektakXT.

The user manual, quality control procedure and results, technical information and contact information can be found in LabManager:

P17 page in LabManager

Process information

Info about measurement accuracy.

Using the analysis software: Apex software access & tips

Info about making 2D stress measurements (requires login). Most users will want to make 3D stress measurements, which is described in the regular Nanolab manual for the P17, also found on LabManager. It counts as "3D" if you wish to make 2 perpendicular scans. A "2D" stress scan means a single line scan per wafer - the software does not even allow manual rotation of the stage for this type of scan.

Acceptance test

The acceptance test carried out for the P17: File:Acceptance test p17 final_no-names.pdf
Vendor's own conformity test results after installation (requires login).

Equipment performance and process related parameters

Front of the P17 profiler located in cleanroom F-2.

Purpose	Profiler for measuring microstructures	Single line profiles Wafer mapping Stress measurements by measuring the wafer bow Surface roughness on a line scan
Performance	Scan range X Y	Line scan X: 20 µm to 200 mm in a single scan. No stitching. Map scan XY: In principle any rectangle that can be inscribed in a 200 mm circle, but resolution is limited to max. 4 million points and scanning is slow. In practice to get good resolution scan a very small area (e.g., 100 x 500 µm)
	Scan range Z	50 nm to 900 µm. It is possible to measure smaller steps but not recommended as the results may not be accurate.
	Resolution X Y	Down to 0.025 µm in theory, but the tip radius is 2 µm, so the meaningful resolution is at the same order of magnitude
	Resolution Z	0.01 Å, 0.08 Å, or 0.6 Å according to the manufacturer for ranges 13 µm, 131 µm, and 1 mm. Note the smallest of these values are purely theoretical as they are far below the lab's noise level.
	Height accuracy z (95 % confidence)	~ 2 % for the smallest range for a 1 micron step and ~ 1 % for a 25 micron step for well-defined steps that are easy to measure (read about reducing and estimating the measurement uncertainty here)
	Max scan depth as a function of trench width W	0.87(W[µm]-2µm) = tan(60^o)/2(W[µm]-2µm) (empirically validated by Nanolab staff)
Hardware settings	Tip radius	2 µm 60^o cone
Substrates	Substrate size	Up to 8"
	Substrate materials allowed	All materials that do not stick to the tip or leave residues on the chuck. For very soft materials we recommend the Dektak150, see below.