DektakXT Surface Profilers - Automatic Step Detection Feature for DektakXT Surface Profilers from Bruker

Topics Covered

Introduction
Automated Step Detection
How Step Detection Works
Setting Up Step Detection
General Settings
Every Step Settings
First Step Settings

Introduction

DektakXT stylus profilers provide accurate step height measurements from several nanometers through hundreds of microns, with repeatability down to 7.5 Angstroms. A unique Step Detection feature is now included with all DektakXT surface profilers to automatically locate and measure single or multiple step heights. Step Detection is especially useful in applications where repetitive measurements are performed on like samples.

This document explains how to use automatic Step Detection for fast, repeatable step height measurements.

Automated Step Detection

Most stylus profilers require an operator to manually set measurement positions and parameters for individual step measurements-a time-consuming operation with low repeatability.

DektakXT Step Detection lets you set parameters to automatically level scan data, detect multiple steps, and measure their heights, as in Figure 1. These parameters ensure that each step height is measured correctly regardless of the start and end points of the scan, and without intervention from the operator between scans.

Figure 1. Multiple steps automatically measured with Step Detection software. Found edges are shown as green dots.

How Step Detection Works

The Step Detection algorithm follows these steps:

1. A measurement is made.
2. The data is smoothed and leveled based on your settings.
3. Each step is detected and measured based on your settings.

The Smoothing, Cursor Positions and Step Description settings greatly affect the number of steps that the algorithm will locate. Careful setup will lead to the most accurate step detection and analytical calculations.

Setting Up Step Detection

Figure 2 shows a typical scan across a surface step approximately 1µm high by 700µm wide. The sample is slightly tilted, with approximately 60Å roughness on the upper and lower surfaces. Step Detection will automatically level and smooth such data to obtain accurate step height measurements.

Figure 2. Typical raw scan of a step.

To make a step measurement, start by taking a sample scan of the feature(s) to be measured. From this scan you can determine the correct scan length, vertical measurement range, stylus force and the number of data points required to achieve the desired horizontal resolution.

Knowing these correct values, scan the step(s).

Next, choose Analysis>Step Detection to open the Step Detection parameters screen. You can also right click in the Plot Summary or Scan Summary screens. The General Settings tab (Figure 3) includes parameters for the overall scan and for data leveling. The Every Step and First Step tabs (Figures 4 and 5) show parameters that define valid steps and determine the parameters to report.

Figure 3. Step Detection General settings. Automatic Leveling parameters per settings (Right).

Figure 4. Every Step settings.

Figure 5. First Step settings

General Settings

1. Detection Method. For most applications, the Every Step method is used, to automatically detect and measure every step within the scan length. The First Step method will measure only the first step encountered.
2. Detection Range determines the portion of the scan that will be analyzed for valid steps. In Figure 3, only the first 2000µm of the scan, where the feature of interest is located, will be analyzed.
3. Check Automatic Leveling to level scan data prior to step analysis. The R and M Cursor parameters define the cursors' position relative to the first found edge (not necessarily a step), distance from the edge, and width, as per Figure 4. These cursor settings apply only for leveling, not for data analysis. Typically the R cursor is placed before the first edge, and the M cursor follows the edge.
4. Check Enable Step Detection, then click Apply to analyze the current data set. The screen will update to show the step analysis for the current data.
5. Check Save Changes to Scan Routine to perform Step Detection, with the current settings, each time the current Scan Routine is used.

Every Step Settings

Click the Every Step tab to set the following parameters:

1. Smoothing determines the minimum slope change between adjacent data points that will be considered an "edge." A smaller Smoothing value increases sensitivity to smaller steps; a larger value filters noise to limit analysis to distinct, welldefined steps.
2. Min and Max Step Heights The Step Detection algorithm places cursors on either side of each found "edge," based on the cursor settings in the Analytical Functions box (below). If the ASH between these cursors falls within these Min and Max values, the edge will be considered a valid step.
3. "+Steps" and "-Steps" lets you select whether positive steps (above the reference) and/or negative steps (below the reference) will be analyzed.
4. Analytical Functions define which analysis parameters will be reported for each valid step. Check the box for each analysis that should be reported. Set the Distance to Step for the R and M cursors, relative to the leading edge of the step. Note that each analysis can have its own cursor locations. For Average Step Height (ASH) calculations, you can also set cursor Widths to determine how much scan data will be averaged along the top and bottom of each step.
5. Check Compute Average to report the average of each selected parameter over all valid steps. This feature is particularly useful for characterizing the depths of V-grooves or the height of bumps in an array, as well as for traditional step heights.

First Step Settings

On occasion you may want to analyze a single step using multiple sets of cursor positions. In this case you would check the First Step method under General Settings, then click the First Step tab to set these parameters:

1. Smoothing determines the minimum slope change between adjacent data points that will be considered an "edge." A smaller Smoothing value increases sensitivity to smaller steps; a larger value filters noise to limit analysis to distinct, well-defined steps.
2. Step Description defines the expected height and width of a valid step, within the given Tolerance.
3. Distance to Step and Band Width set the location and width of the analytical R and M cursors, relative to the leading edge of the step.
4. "+Steps" and "-Steps" lets you select whether positive steps (above the reference) or negative steps (below the reference) will be analyzed.
5. Analytical Functions define which analysis parameters will be reported for each valid step. Set the locations of the R and M cursors relative to the leading edge of the step. For each analysis you can define up to ten sets of cursor locations and widths to analyze the step at varying distances.
6. Check Compute Average to report the average of each selected parameter across all sets of cursors.

Once you've entered all of the parameters, click Apply or OK to display the measured parameters for each valid step (Figure 6).

Figure 6. Scan profile after step detection. The yellow line shows the data after smoothing.

Figure 6 shows the data from Figure 2 after Step Detection analysis. The scan profile has been leveled, the step has been detected, and the average step height (ASH) has been calculated.

Automated step measurement is a valuable tool for determining the height or depth of single and multiple steps across a wafer or substrate. The ability to program step measurement sequences greatly reduces measurement time as well as operator-to-operator variability. Only DektakXT stylus profilers offer this unique capability for rapid, accurate step height assessment.

This information has been sourced, reviewed and adapted from materials provided by Bruker AXS.

For more information on this source please visit Bruker AXS.