AFM Image Quality

The smaller the radius of curvature, the smaller the feature that can be resolved. A sharper tip will be able to laterally resolve smaller features than a dull tip with a larger radius of curvature.

The accumulation of debris on the end of the tip can also dull the tip and result in image distortion:

A dull or dirty tip may not affect the measurement of the vertical dimensions of these samples.

The ability to image steep sidewalls on a sample surface is determined by the sidewall angles of the probe tip. The tip is not able to profile sides of surfaces steeper than the sidewall angle of the tip. Consequently, when scanning across features that are steeper than the sidewall angle of the tip, the sidewall angle in the images will reflect the sidewall angle of the tip. For more information, see Probe Tip Geometry in the Probe and Sample Guide.

As discussed in the above section on tip shape effects, the radius of curvature of the end of the tip will determine the highest lateral resolution obtainable with a specific tip. The sidewall angles of the tip will also determine its ability to probe high-aspect-ratio features.

The Samples per Line parameter determines the number of data points present in the image in the X and Y scan direction. These data points are commonly called pixels. Available choices are 512, 256, and 128. Pixelization affects resolution in that you cannot resolve features smaller than the pixel size of the image.

Regardless of the pixel size, the feedback loop is sampling the topography many times at each pixel. The data displayed at each pixel is the average of the sampling iterations by the feedback loop over the pixel area.

The resolution in the vertical direction is not determined by the tip shape; it is primarily determined by the resolution of the vertical scanner movement, which is <1 Å.

As in the lateral dimension, the number of data points in the vertical direction limits the size of the smallest resolvable height change. The number of data points in the vertical dimension is determined by the conversion of 16 bits over the full vertical range of the scanner. The Z limit parameter may be used to increase the sampling resolution in the vertical dimension to image sub-Ångstrom height changes with large Z range scanners.

The overall system noise will generally be the primary limiting factor in the highest obtainable vertical resolution that can be acquired with your SPM. This may be a result of combined effects from electrical, mechanical, and acoustic noise sources.

The AFM noise typically ranges from 0.3 to ~1Å RMS (depending on the type of SPM and SPM environment).

If the tip becomes worn or if debris attaches itself to the end of the tip, the features in the image may all have the same shape. What is really being imaged is the worn shape of the tip or the shape of the debris, not the morphology of the surface features.

Double or multiple tip images are formed with a tip with two or more end points that contact the sample while imaging.

Loose debris on the sample surface can cause loss of image resolution and produce streaking in the image.

The image on the left is an example of loss of resolution due to build up of contamination on the tip when scanning from bottom to top. The small elongated features are represented as larger, rounded features until the debris detaches from the tip near the top of the scan. The image on the right is an example of skips and streaking caused by loose debris on the sample surface. Often, loose debris can be swept out of the image area after a few scans, making it possible to acquire a relatively clean image. Skips can be removed from a captured image with software.

Interference between the incident and reflected light from the sample surface can produce a sinusoidal pattern on the image with a period typically ranging between 1.5 –2.5 μm:

This artifact is most often seen in Contact mode on highly reflective surfaces in force calibration plots and occasionally in TappingMode images. It can usually be reduced or eliminated by adjusting the laser alignment so that more light reflects off the back of the cantilever and less light reflects off the sample surface, or by using a cantilever with a more reflective coating (MESP, TESPA).

When the feedback parameters are not adjusted properly, the tip will not trace down the back side of the features in TappingMode AFM. The image on the left was scanned from left to right (trace) producing tails on the colloidal gold spheres. The image on the right is the same area imaged with a lower setpointOperator-selected threshold used as the target of the feedback control loop. voltage, increased gainsThe gains are scaling factors for the error signal to determine how quickly corrections should be applied in the feedback loop. There are two important gain factors: the proportional gain scales to the current value of the error signal, and the integral gain scales the aggregate of the past values of the error signal (the "area under the curve")., and slower scan rate.

In TappingMode AFM, when operating on the high-frequency side of the resonance peak, rings may appear around raised features that make them look like they are “surrounded by water,” as in the image of Ti grains on the left. Decreasing the drive frequencyFrequency of the signal used to oscillate a tip in TappingMode. during imaging can eliminate this artifact, as shown in the image on the right.

NOTE: When reducing the drive frequency, the setpointOperator-selected threshold used as the target of the feedback control loop. voltage may need to be reduced, too.

The arch-shaped bow in the scanner becomes visible at large scan sizes. The bow in this image (left) can be removed by performing a second order planefit in X and Y (right).

At large scan sizes, the bow in the scanner may take on an S-shaped appearance. The bow in this image (left) can be removed by performing a third order planefit in X and Y (right).