Today’s high-resolution laser scanning confocal microscopes (LSCM), also referred to simply as laser scanning microscopes, are powerful high-magnification surface metrology instruments with a vast application range, from simple metal surfaces to high-end electronic components. Software capabilities make this possible by taking high-resolution surface images and producing accurate, repeatable nanometer measurements. This article will discuss the basic functions of a laser scanning microscope, as well as the software capabilities that enable manipulation of the imaging aspects after hardware acquisition.

Hardware Overview

Figure 1. The Olympus LEXT® OLS5000 laser scanning microscope

The modern laser scanning microscope has come a long way from its early predecessors, which took up an entire lab table. The latest models (Figure 1) are compact enough to fit on a tabletop and are smaller than monitors typically used for imaging. The base unit, called the “scan head,” holds the laser scanning module. Most systems have both a laser scanning light path and a LED white-light imaging path to enable high-resolution laser imaging, while capturing full-color surface images. To improve surface lateral resolution, the laser is typically in the lower wavelengths, around 405-408 nm, but a low-cost diode laser is used to keep the system price competitive.

In the past, deep ultraviolet (DUV) microscopes would use 248 nm wavelengths to obtain the same lateral resolution provided by new laser scanning microscopes. Such a low wavelength required two things to function: first, a very expensive objective lens to correct the low wavelength; and second, a specialized analog camera to produce images of that same wavelength. This resulted in a costly microscope system with magnification limited to 1000x. Another drawback is that DUV microscopes do not have the capability for surface topography 3D imaging.

The laser scanning microscope uses a scanning design called “beam scanning,” where the laser image path is scanned in a raster pattern on the surface of the sample. This technique captures the surface image and enables magnification control by reducing the scan area to smaller sizes, thus increasing magnification without any resolution loss. To generate a scan, a variety of scanning mirrors are used, including acousto-optic deflectors (AOD), polygon mirrors, and resonant galvano mirrors. Galvano mirrors help maximize the image field of view. These mirrors also support relatively large oscillation angles, which improves lower magnification of large fields of view. Because its speed is determined by mechanical resonant frequency, this mechanism is limited in terms of speed compared to other scanning mechanisms. However, recent resonant galvano mirrors can acquire several one-megapixel images every second.

Figure 2. Intensity curve of confocal and non-confocal over travel distance in Z.

Mirrors manufactured with micro-electromechanical system (MEMS) technology have also been developed, enabling a reduction in device size. MEMS scanners are a combination of movable plate, torsion bar, and support frame made by etching a single monocrystalline silicon board. The movable plate has coils driven by a magnetic circuit. Two-dimensional scanning can be accomplished by using a high-speed scanning mechanism and combining it with a relatively low-speed scanning mechanism in the Y-direction. A non-resonant galvano mirror is often used in the Y-direction scanning mechanism, partly as a matter of convenience.

After the laser is scanned by the mirrors, it is directed through the image path and objective lens. It then returns through the optical path to pass through the confocal pinhole. This technique improves the signal-to-noise ratio by removing any out-of-focus signal, allowing only the most focused points of light to pass through the pinhole. This focused light is directed to a photomultiplier tube to digitally transfer the intensity values to an amplifier.

In addition to the high signal-to-noise ratio, the confocal technique improves surface resolution by creating a very low depth of focus image for each focus point. Figure 2 illustrates this effect, where the vertical axis shows the output from the detector after the light passes through the pinhole, and the horizontal axis indicates the travel distance in the Z-direction. This output was acquired while moving the sample and objective lens relative to each other in the Z-direction, without two-dimensionally scanning with the confocal optical system. This waveform is called an I-Z curve. Comparing the non-confocal output with the pinhole removed and the confocal output acquired under the same conditions reveals that the confocal optical system shows a steep waveform.

Figure 3. Theoretical image of different Z planes captured in a single extended image.

An image of a stepped sample where all height positions are in focus, called an extended focus image, can be obtained by scanning the laser light in two dimensions while relatively moving the sample and objective lens and saving the brightest light intensity value of each pixel. Figure 3 shows how the sample’s extended focus image is captured. When scanning the laser light across the top face (Figure 3) in two dimensions and focusing it, blurry images are eliminated, resulting in the square shape being captured. When the laser light is moved in the Z-direction and the second face is focused on, the smallest L-shape is created. By sequentially repeating this step and then capturing and overlapping the image of each face, an extended focus image with a deep focal depth (where every face of the sample is brought into focus at high resolving power in the horizontal direction) can be created.

Figure 4. Theoretical image of measurements in Z across a line on the extended image.

In the confocal optical system, the Z-position with the maximum intensity indicates the height information of the sample surface at that point. Figure 4 shows how height information is captured. The process of recording the height of a sample is similar to capturing an extended focus image – i.e. the user can move the sample and objective lens relative to each other and save the information on the brightest, most intense Z-position for each pixel while moving from height Z1 to Z2. This makes it possible to obtain the surface profile of the sample in the image acquisition area and perform a variety of analyses based on this information.