MProbe® MSP enables the measurement of film thickness in surface areas as small as 2 µm. The excellent measurement capability of the MProbe® system is further demonstrated with the use of small spots and imaging that expands the wide range of measurable applications.
Small spot measurement is a necessary procedure in thousands of applications and these applications can be, roughly, separated into three types:
- Measurement of patterned films (patterned wafers, MEMS, etc.)
This is a conventional, straight forward, and well-understood application for small spot measurement. However, due to the instances of film thickness measurements on the bottom of deep wells proving somewhat complicated, we reach a better understanding of the MProbe system’s capabilities.
- Measurement of coatings on a rough surface and/or highly laterally nonuniform films
Using small spot measurements in this application may seem counter-intuitive. However, a small spot enables the minimizing variation of film thickness within a measured area. From this perspective, it is very similar to the measurement of patterned films.
- Measurement of coatings on a non-planar surface (pins, wires, etc.)
The rationale for using a small spot, in this case, differs in comparison to the previous two methods. We want to achieve the same measurement conditions as when measuring on a flat surface. This is accomplished by making the measurement spot as small as possible in contrast to the diameter of the measured sample.
Thickness Measurement in High Aspect Ratio Well (Measurement of Patterned Films)
MProbe® Vis MSP system is used for the measurement of oxide (SiO2) layers on the bottom of round wells. The diameters of the various wells are 80 µm, 40 µm, 20 µm, and 10 µm – all 100 µm deep. After lithography, the top surface of the sample (Si) had a photoresist layer left. The round well’s geometry was chosen because it is more difficult to measure in comparison to square wells, or vias of the same geometry.
For this measurement, a 20x APO objective with long working distance (35 mm) is used while the measurement spot size is ~20 um. The reason for using this objective is low NA (0.29), which is crucial for good measurement in the deep well.
Figure 1. Light geometry of the measurement in the deep well (all dimensions are in µm).
1 – Direction of the incident light from the objective
2 – Focusing position (middle of Si layer) for optimal measurement The goal is to measure the thickness of Si oxide at the bottom. There is residual PR on top of Si.
The light geometry of the measurement in a deep well is exhibited in Fig. 1. It demonstrates that at the top of the well the beam diameter is ~ 50 μm, so it can fit rather easily in the 80 μm, and the obstruction effect in the 40 μm geometry is minor. These results are clear and thus confirmed experimentally. This image illustrates the significance of the low NA of the objective.
Figure 2. Measurement of 80 μm Round/circular well (aspect ratio -1.25). (a) The fit of measured data to the model. (b) Use of the thick film algorithm.
Utilizing the same configuration for measurements with 20 µm and 10 µm will cause some of the light to be reflected from the PR layer while only a small portion of the light reaches the bottom of the well. Therefore, we ask the question, is it possible to still measure oxide thickness?
Figure 3. Measurement spot covering areas S1 and S2 with thicknesses T1 and T2, respectively. In our case, S1 and S2 represent oxide (on the bottom of the well) and PR (on top of the Si).
Reflectance from a sample (Fig.1) is r(λ) = S1r1(λ) + S2r2 (λ), where r1,2 are reflectances from areas S1 and S2 . r1,2 = |r1,2 |e-iφ1,2 are complex numbers that possess amplitude and phase (φ1,2). When the reflected light reaches the photodetector – reflectance from area S1 and S2 merge and are converted to intensity.
Reflectance r1,2 are added as complex numbers i.e., as vectors. This procedure is known as convolution and the phase information is retained and converted into a variation of intensity vs. wavelength. The signal can be decomposed during the data analysis to determine original T1, T2 thicknesses, and phase or intensity is never averaged. This process is illustrated in Fig 4.
Figure 4. Transformation of the reflectance from a sample.
The initial peak, corresponding to oxide thickness decreases with the decrease of diameter but it is still possible to measure the thickness. Utilizing the configuration as proposed it is possible to achieve a measurement of thickness on the bottom of the deep well as small as 10 µm with 10x aspect ratio.
Figure 5. Measurement of (a) 20 μm well and (b) 10 μm. The first peak, corresponding to oxide thickness decreases with the decrease of diameter but the thickness can still be measured.
Coating Thickness Measurement on Metal Surface
Rough/Brushed Metal Surface
Typically, it is not possible to measure a coating on a rough metal surface using a standard 0.5 mm spot. Due to roughness, highly non-uniform coatings possess a broad range of randomly distributed thicknesses. The complexity of signals from varying thicknesses average the phase and in effect destroy interference making thickness measurement impossible. By reducing the measurement spot size we curb the variation of thickness within a specific measurement area. For cases where the thickness is relatively thin (<1 µm) – this is the only conceivable way to achieve a reliable measurement.
Various colors on the sample (Fig. 6) correspond to different thicknesses. It is possible to measure coating thickness with 40 µm spot Fig. 6 (b) but the fit of the model compared to measured data is not great. There are still several different thicknesses present within a confined measurement spot. Thus, using an even smaller 10 µm spot Fig. 6 (a) enhances the fit and facilitates a more accurate measurement.
Figure 6. Measurement of the coating on a spray-coated brushed metal. (a) Measurement spot 10 μm, (b) Measurement spot 40 μm.
Smooth Metal Surface
In most cases, measuring the thickness of a coating on a smooth metal surface does not require a small spot. Yet, there are some coatings (such as dip coating) that are extremely non-uniform and have some localized flaws. Therefore, utilizing a small spot facilitates more reliable and accurate measurements (see Fig. 7).
Figure 7. Measurement of dip-coated metal sheet with a smooth surface. Measurement spot size 10 μm is positioned in the uniform area.
Coating Thickness Measurement on Cylindrical Surfaces
Several applications demand coating thickness measurements on cylindrical surfaces e.g. coatings on orthopedic pins, coatings on the vascular stent, coatings inside syringes, etc. Therefore, it is essential, in each case, to use a small measurement spot. The size of the spot must be small enough in contrast to the curvature radius; this is in order for the measurement area to be approximated as a flat surface.
Measurement of Coating Thickness on Stent
Figure 8. DES vascular stent, wire diameter 0.7 mm.
Figure 9. Measurement location – red circle indicates the measurement spot area (10 μm). The surface is ablated to increase the retention of the medicated polymer.
Figure 10. Measurement of the stent coating– thickness 4.35 μm.
Figure 11. SEM measurement of the stents coating shows a perfect match to MProbe thickness measurement: 4.35 μm.
Measurement of Coating Inside Syringes
Prefillable glass syringe cartridges are generally washed, siliconized, sterilized, and packaged by the manufacturer. Siliconization of the syringe barrels is crucial as it performs as a lubricant and allows the plunger to glide smoothly. Furthermore, its hydrophobic layer prevents interactions between the drugs and the glass surface. Therefore, if siliconization is either insufficient or excessive it can lead to various problems. Most siliconization in the modern production line is carried out using a “baked-on” technique where silicon emulsion is sprayed and subsequently baked to produce a permanent layer. Control of the thickness and uniformity of the silicon layer is a vital component of the product quality control throughout production.
Figure 12. Syringe cartridge is placed on the MProbeVis-MSP table.
Figure 13. Reflectance spectra taken at different points on the syringe indicate significant variation in coating thickness.
Figure 14. Silicon thickness measurement at different points on the Syringe. Image Credit: Semiconsoft Inc.
Figure 15. Variation of silicon coating thickness along the 3 ml syringe barrel.
MProbe® MSP is a modular system that can be utilized across various wavelength ranges and measurement spot sizes. Customization of configuration is possible for individual application requirements to facilitate accurate measurements that you can depend on. The system merges the capabilities of an inspection microscope with non-contact optical thickness measurements.
This information has been sourced, reviewed, and adapted from materials provided by SemiconSoft.
For more information on this source, please visit SemiconSoft.