Next Article in Journal
SurfILE: An Open-Source Python Package for Surface Topography Analysis
Next Article in Special Issue
Advancing Molecular Spectroscopy Efficiency with Extensive Parallelism
Previous Article in Journal
A Statistical Approach for Characterizing the Behaviour of Roughness Parameters Measured by a Multi-Physics Instrument on Ground Surface Topographies: Four Novel Indicators
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Metrology
Volume 4
Issue 4
10.3390/metrology4040040
metrology-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Investigation of the Measurement Systems’ Suitability for the Non-Destructive Measurement of Complex Polymer-Based Micro and Nanostructures

by
Selina Burkert
1,*,
Lukas Schwörer
1,
Tim Schubert
2,
Jana Grundmann
3,
David Stein
4 and
Andreas Heinrich
1
1
Aalen School of Applied Photonics, Aalen University of Applied Science, 73430 Aalen, Germany
2
Institute for Materials Research, Aalen University of Applied Science, 73430 Aalen, Germany
3
Physikalisch-Technische Bundesanstalt, 38116 Braunschweig, Germany
4
Institute of Physics, University of Augsburg, 86150 Augsburg, Germany
*
Author to whom correspondence should be addressed.
Metrology 2024, 4(4), 673-694; https://doi.org/10.3390/metrology4040040
Submission received: 26 August 2024 / Revised: 13 November 2024 / Accepted: 14 November 2024 / Published: 21 November 2024
(This article belongs to the Special Issue Advancements in Optical Measurement Devices and Technologies)
Browse Figures
Versions Notes

Abstract

:
In the fabrication of optical polymer-based components, such as diffractive gratings and waveguides, high throughput and high precision are required. The non-destructive evaluation of these complex polymer-based structures is a significant challenge. Different measurement techniques can measure the structure geometry directly or via its functionality indirectly. This study investigates various measurement techniques aimed at assessing these structures from 200 nm up to 20 µm. Environmental scanning electron microscopy (ESEM), white light interferometry (WLI), atomic force microscopy (AFM), micro computed tomography (µCT), optical coherence tomography (OCT), phase contrast microscopy (PCM), and Mueller matrix ellipsometry (MME) are investigated for their practical limits of lateral resolution and aspect ratio. The impact of the specimens’ complexity factors, including structure width and aspect ratio, on measurement quality is discussed. A particular focus of this study is on the suitability of different measurement systems for evaluating undercuts and enclosed structures while considering structure size, slant angle, and cover thickness. The aim is to discuss the specific advantages of the individual measurement systems and their application areas in order to be able to quickly select suitable measurement systems for a non-destructive evaluation of polymer-based micro and nanostructures.

1. Introduction

In micro and nanofabrication, polymer-based structures are used to maximize the functionalization of the smallest possible devices [1]. In optics, for example, diffractive gratings are used for light coupling and decoupling, as well as structures for light guiding (waveguides) [2,3,4] or intended phase shifting (holograms) [5,6]. Typical structure sizes here are in the sub-wavelength range from 200 nm to a single-digit micrometer range. The complexity of these structures increases when the gratings are slanted and overhangs are generated [7]. The quality of diffractive and refractive elements directly influences light transmission and imaging efficiency [8,9,10]. There are other applications that are based on polymer structures that are larger than those of a single-digit micrometer range, such as biofluidic devices. For example, small channels and filter elements are manufactured for lab-on-chip devices and rapid diagnostics [11,12]. The size of the structure varies from rather small channels with a width of 3 µm [13] to 8 µm [14] up to larger channels of 200 µm [15]. Therefore, precise manufacturing of the chips is a prerequisite for precise diagnostics [16,17].
As already outlined, the precision of micro and nanofabrication is very important with respect to the given examples of optical and biofluidic systems. In order to be able to assess the fabrication processes, measurements and evaluation of the functionalized structures are required. Measurements can be made directly through optical [18] or tactile [19] measurement of the geometry or indirectly through the structure function. The more complex the structures are, the more difficult it is to make a reliable assessment.
A complexity factor is the geometry of the structure. As structures become smaller, microscopic imaging becomes increasingly difficult [20]. Especially if the measuring systems are limited by diffraction, the structures of diffractive gratings are sometimes smaller than the resolution limit of a microscope objective [21]. Another complexity factor is the ratio of structure height and width, the so-called aspect ratio. The greater the aspect ratio, the more difficult it is to detect the interior of the structure, both optically and in a tactile way. A high numerical aperture is required for optical detection [22] and the smallest possible tip for tactile detection [23]. The complexity factor is also increased as soon as structures with overhangs or enclosed structures within a polymer solid are to be measured and evaluated. For this purpose, the measurement system must be able to penetrate the polymer solid [24] and to visualize the undercuts and cavities or derive information about the geometric quality via the optical function of the structures [8,25]. In summary, complex polymer structures are those that are very small (<1 µm), have a large aspect ratio (>1), have overhangs, or are enclosed.
Different measurement techniques have their strengths and weaknesses that can be utilized in the evaluation of complex polymer structures without damaging the structure during measurement. In this study, we investigate different measurement techniques such as environmental scanning electron microscope (ESEM), white light interferometer (WLI), atomic force microscope (AFM), micro computed tomography (µCT), optical coherence tomography (OCT), phase contrast microscope (PCM), and Mueller matrix ellipsometer (MME). First, the limits of lateral resolution and maximum aspect ratio are examined using pillar structures. These pillar structures were written into a commercial acrylate-based resin using two-photon polymerization (2PP) [26]. It will be discussed for which measurement systems the complexity factors, structure width from 200 nm up to 20 µm and aspect ratio 0.1 up to 2.86, have a definable influence on the measurement quality. Furthermore, it will be investigated which measurement systems are suitable for the measurement of undercuts and enclosed structures and where the limits are by taking the structure size and the covering layer height into account. The specific advantages of the individual measuring systems and their application areas are discussed in order to be able to quickly select a suitable measurement system for a given polymer structure.

2. Measurement Systems and Methods

A WLI, AFM, OCT, ESEM, µCT, PCM, and MME were used to investigate polymer-based micro and nanostructures. The applied systems and the respective main settings are summarized in Table 1. For a better understanding, the measurement principle of the used systems and, if necessary, the post-processing method for data acquisition are described below.

2.1. White Light Interferometer

The used WLI is based on the measuring principle of a Mirau interferometer with a broadband light source (spectral width 120 nm) and a 50× Mirau objective. During the WLI measurement, the Mirau objective is moved by a piezo actuator along the optical axis. The interference signal is processed in a frequency domain analysis (FDA [27,28]) and a three-dimensional profile is reconstructed as shown in Figure 3a. WLI measurements are particularly suitable for flat structures in the sub-micrometer range [18].

2.2. Atomic Force Microscope

In an AFM measurement, a measuring tip scans the sample. The individual AFM techniques differ in how the sample is scanned by the tip and which measurement variables are gained. Using frequency-modulated AFMs, the tip oscillates and the oscillation amplitude or the phase shift of the tip oscillation is acquired with or without contact to the sample [29,30,31,32]. In the case of scanning tunneling AFM, an additional reference voltage is applied between the tip and the sample and the tunnel current is measured to reconstruct the height [29,33]. Using the constant force measurement mode, the mechanical deflection of the tip (ContAl-G-10, BudgetSensors, Sofia, Bulgaria, tip radius approx. 10 nm) on the sample’s surface is detected by a laser triangulation and compensated by a piezo actuator. The voltage of the piezo actuator, that is needed to compensate for the tip deflection in order to keep the force constant, is recorded [32,34]. The recorded voltage values of the piezo actors are reconstructed to a metric height profile as shown in Figure 3b using a calibration factor. An AFM can be used to resolve structure sizes below the microscopic diffraction limit [19,35].

2.3. Optical Coherence Tomograph

The used spectral domain OCT is based on a broadband point light source, which is directed onto the measurement object and a reference mirror via a beam splitter. The reflection from the mirror and the measurement object is guided back into a spectrometer via the same beam splitter. In the spectral domain analysis, the metric distances of the reflection layers at a certain point of the sample are reconstructed [36]. The sample is scanned laterally with a scanning unit, and a three-dimensional data stack of the reflection layers is generated. Figure 3c shows the recorded surface of the sample based on the summed stack of all vertically recorded reflection intensities. The height data need to be calibrated for an absolute measurement. OCT is particularly suitable for material examinations, as it can detect material layers [24] or defects [37] below the polymer surface based on refraction differences.

2.4. Environmental Scanning Electron Microscope

In the ESEM measurement, the sample is scanned by an electron beam. Atoms in the sample react by emitting secondary electrons (SE), among other things. The SE are captured by a SE-detector. Non-conductive materials, such as polymers, charge up particularly quickly with the primary electrons. To avoid charging defects during an ESEM measurement, the chamber vacuum is set as low as possible (e.g., 50 Pa) or water is actively introduced into the vacuum chamber to discharge the sample surface. Using the ESEM, structures below the usual light microscopic resolution can be captured in two dimensions [38,39]. Cross-sections of the structures are often recorded to image undercuts [7,40]. An exemplary ESEM measurement is shown in Figure 3d. The gray values represent the detected SE at the respective position in the scan field and no information about the structure height can be gained. Since the structure edges have a significantly higher SE dose as illustrated in Figure 1a and in the respective top view (b), the position of the edges can be used to determine the structure width by binarizing the ESEM image as shown (c). The edges are detected (1) or the entire structure of the pillar is detected (2).

2.5. Mirco Computed Thomograph

In a conventional µCT measurement, the object is rotated in an X-ray beam, and the transmitted X-rays are captured by a detector. The single detector images at different rotation angles are combined to form a three-dimensional data stack. For a better resolution down to the sub-micrometer range, magnification lenses are placed in front of the detector of the utilized µCT as standard [41]. Figure 3e shows an image from the top view. Similarly to OCT, µCT is also suitable for sub-surface material analysis of polymers as demonstrated in [42,43].

2.6. Phase Contrast Microscope

The PCM is based on a conventional bright-field microscope. In addition, it has a phase ring for phase shifting by a quarter wavelength (λ/4) and a ring aperture between the light source and condenser lens. This allows phase shifts to be visualized as intensity differences, as shown in Figure 3f, for example. Thus, the phase shift ∆φ is caused by a change in the propagation time through the sample as visualized in Figure 2a. If the focus is well set, the structure can be well identified in the top view (b). Similarly, to the ESEM images, the PCM images are binarized to detect the edges and calculate the diameters of the structures (c). The visualization of phase shifts can be used in the material analysis of polymers and blends, as demonstrated in [44,45].

2.7. Imaging Mueller Matrix Ellipsometer

The imaging MME system combines conventional ellipsometry and bright-field microscopy. The main measurement parameter is the Mueller matrix M. The 4 × 4 Mueller matrix describes the change of an electromagnetic wave after interaction with a sample based on the change in the initial Stokes parameters Si and final Stokes parameters Sf, as can be described by [46]:
Sf = M Si
Instead of averaging the sample’s polarization behavior at the entire illumination spot of the ellipsometer, the Mueller matrix values are calculated for each pixel of the camera image using the imaging MME. Due to the calculation of the local polarization behavior, non-periodic structures and structures that are smaller than the illumination spot can be evaluated [47]. Figure 3f shows the sample’s microscope image and the respective normalized value M32 of the 4 × 4 Mueller matrix for example. A change in geometry can be detected indirectly by a shift in the local polarization behavior of the structure. Comparable indirect measurement techniques for polymer nanostructures are scatterometry [10] or diffraction tests [35].

3. Polymer-Based Test Patterns

3.1. Pillar Arrays

In the investigation of the practical limits of the lateral resolution and the aspect ratio, a pillar test pattern is used, as shown in Figure 4. A pillar as given in (a) has a diameter of width w and a height h that ranges from the base to the top of the pillar. The base width corresponds to twice the pillar width 2w and the base height b is 1.5 µm for all pillar arrays. The pillars are rotationally symmetric as shown in (b). A pillar array consists of integer multiples of a pillar from (b) within a field of 120 µm, as illustrated in (c). To check the directional dependence of the measurement systems, a distinction is made between horizontal slices in the x-direction (blue) and vertical slices in the y-direction (orange).
The pillar arrays illustrated in Figure 4 were realized using two-photon polymerization (2PP, introduced in [26,48]). In the manufacturing process, the pillars were written with a 2PP system (QuantumX, Nanoscribe, Karlsruhe, Germany) in a 2PP polymer (IP-Dip, Nanoscribe, refractive index n 1.55 @ 590 nm [49]) on a 2″ borosilicate wafer (MicroChemicals, Ulm, Germany, refractive index n 1.52 @ 590 nm). To improve the adhesion of the 2PP polymer to the borosilicate wafers, the wafers were pretreated with oxygen plasma (Zepto BLS-W6, Electronic Diener, Ebhausen, Germany) and a silanizing adhesion promoter (Primer20, MicroResist, Berlin, Germany). In Table 2, the chosen pillar widths w and heights h are defined concerning the manufacturing process. To achieve the maximum resolution, the pillars are a multiple of the smallest voxel width of 200 nm of the 2PP system. Examining eight different pillar widths ranging from 0.2 µm to 20 µm with an exponential step size (w = 0.1 e0.66 n), n = [1:8]) ensures a comprehensive investigation, even below the resolution limit of diffraction-limited systems. When varying the height h from 0.5 µm to 4 µm, the aspect ratio a of h/w results in 0.1 to 5. In the design of the experiment, 26 pillar arrays were manufactured and their respective indexes are written in brackets in Table 2.
A laser scanning microscope image of the realized pillar test structure is shown in Figure 5. According to this figure, arrays with an aspect ratio of 5 (like the arrays 7, 14, and 21) are defectively manufactured in the selected 2PP polymer. Therefore, the test range is restricted to aspect ratios of 0.1 to 2.86.

3.2. Slanted Test Pattern

Test patterns with slanted gratings are used to analyze the measurability of the undercuts. The slanted test pattern in Figure 6a is used to investigate the lower limit at which undercuts can be detected with respect to the size of the structure w and the slanted angle α. The structure width w and the height h are always equal. Hence, the aspect ratio always remains one. Structures with a height h of 1, 2, and 3 µm as well as 10, 20, and 30 µm (factor 10 larger) are tested. The slant angle α is varied in steps of 0° to 40° in 10°. A slanted grating element is 12 µm or 120 µm in the y direction with respect to (b). Three slanted grating elements were placed in the x direction. A captured overview of the realized structures is provided in (c). The slanted polymer structure was produced as already described in the manufacturing process of the pillar test structure using the 2PP technique. In the analysis, the slanted test samples are captured from the top, and a slice in the x direction is used for evaluation.

3.3. Covered Pillar Test Pattern

Covered pillars are used to assess the possible penetration depths of measurement systems through a polymer layer (IP-Dip, n = 1.55). The design of the covered pillars is visualized in Figure 7. As shown in (a), the pillars are hollow and have a constant width w and height h of 4 µm. The covering layer thickness hc varies from 0.5 µm to 2 µm in 0.5 µm increments. The 2PP structure is connected to the substrate via base pillars in order to be able to rinse the cavities underneath the covering layer. The base pillars also have a height hb and a diameter of 4 µm. They are offset symmetrically to the structure pillars as shown in (b) and (c). The pillars fill a square of 80 µm and the different cover heights are tagged on the top side as displayed in (c). For a better understanding, a three-dimensional representation of the designed structure is given in (d).

4. Results and Discussion

4.1. Pillar Analysis: Investigation of the Lateral Resolution and Aspect Ratio Limits

For an assessment of the limits with respect to the lateral resolution and the aspect ratio, the pillars’ test range and all successful measurements of the measurement systems (WLI, AFM, OCT, ESEM, µCT, PCM, MME) are plotted in Figure 8. The measurements were classified as successful if the pillars were identifiable based on the diameter at half height (FWHM) or at half intensity (using the described image processing methods such as binarization). According to the comparison in Figure 8, the smallest structures from a width of 0.4 µm can be measured with the ESEM. The AFM and OCT measure structures starting at 0.8 µm. The WLI, µCT, PCM, and MME detect structures with a width of at least 1.4 µm. Pillars with a maximum aspect ratio of 2.86 can be visualized using the WLI, OCT, ESEM, PCM, and MME. Due to the end of the test range, it is possible that higher aspect ratios can be measured. This can be investigated by further tests with higher aspect ratios.
At this point, the µCT measurement is remarkable in contrast to other systems’ measurements. The smallest pillar array that could be measured by the µCT was the fourth array with a width of 1.4 µm and a height of 0.5 µm. The corresponding µCT images are shown in Figure 9a and the respective slice in (b). The biggest challenge of the µCT measurements is the low contrast of the 2PP polymer. An attempt was made to compensate for this using a long exposure time, power optimization (loss of contrast at high power), and measuring ‘in phase’. In the end, long exposure times must be selected that exceed 48 h per pillar array so that the contrast is sufficient and the pillars can be reconstructed. Another way to improve the contrast of the polymer would be to enrich the polymer with high atomic number elements, e.g., barium or silver, and increase its absorption [50].
Figure 8 shows that several measurement systems are suitable, especially for larger structure widths (>1.4 µm). However, the measurement quality of the measurement systems differs. In order to be able to select the most suitable measurement system for a certain measurements task, the measurement quality is investigated. In the following, the deviation from the expected value (residuals) and the measurement repeatability (standard deviation) of all measurement systems (except MME) are examined as a function of the structure width in Figure 10. Horizontal (►) and vertical (▲) measurements are separated by their symbols.
The percentage residuals as a function of the structure width for each comparable measurement system are shown in Figure 10. The respective aspect ratio of the measurement can be assigned via the color and the legend of the measurement. The higher the aspect ratio, the lighter the respective color. Analyzing the AFM and ESEM measurements, it is noticeable that the horizontal and vertical measured values including the standard deviation do not overlap. The difference is significant and is examined in more detail. The AFM and ESEM measurement images and the respective statistical evaluation of the third array (w = 0.8 µm, a = 0.63) are shown in Figure 11a,b. These measurements of the same structure clarify that the 2PP pillars are not ideally round. In the analysis of the 0.8 µm pillars, the vertical diameter is 20% (0.16 µm of w = 0.8 µm) smaller than the horizontal diameter. Theoretically, this would be 40% for the 0.4 µm pillars (0.16 µm of w = 0.4 µm). This is confirmed exactly in the ESEM analysis. Therefore, in the measuring range below 1.4 µm, statements can only be made about the repeatability based on the standard deviation. It is assumed that the absolute deviation of 160 nm affects every pillar due to the 2PP manufacturing. However, this deviation is marginal for pillar widths of 1.4 µm and wider, as it is 10% of the total width or less. In that case, it cannot be distinguished from the measurement inaccuracy due to the scattering of the measured values. From a pillar width of 1.4 µm, both the residuals and the standard deviation can be used for evaluation.
In the measurement range up to 1.4 µm, the standard deviation of the OCT is the largest reaching more than 20% (e.g., w = 0.8 µm: 23%, w = 1.4 µm: 27%). Compared to the other measuring systems, this is more than twice as large as the maximum deviation of WLI, AFM, and ESEM. Therefore, it is recommended to use OCT for a width of 2.8 µm at least. At a structure size of 1.4 µm, the measured values of all measurement systems overlap within their standard deviation, and the measurement quality does not differ significantly. In the measurement range from a structure width of 5 µm and wider, the measured values of the WLI and the ESEM are the most stable. The deviation from the expected value is less than 10%. The standard deviation is less than 5% and is negligible. From a structure width of 20 µm, the residuals of the PCM are less than 10% as well. In contrast, the AFM and OCT measurements have residuals of up to 25% (w = 5 µm) and −26% (w = 10 µm) in this range. Consequently, the use of WLI and ESEM is recommended for structure widths equal to or greater than 5 µm and the PCM for structures equal to or greater than 20 µm.
Analyzing the residuals and the standard deviation as a function of the structure width, the impact of the aspect ratio has been neglected. Therefore, the impact of the aspect ratio on the reproducibility (standard deviation) is displayed in Figure 12. Up to an aspect ratio of 0.2, the standard deviation of all measuring systems is less than 5%. Subsequently, this deviation can be neglected. In addition, this means that there is no difference between the measurement systems in terms of their repeatability up to an aspect ratio of 0.2. The standard deviation increases on average as the aspect ratio of the measurement system increases according to the dashed trend lines. The measurements of the AFM (slope m = 2.4) and OCT (slope m = 10) are most influenced by an increasing aspect ratio. A linear relationship between aspect ratio and standard deviation could not be demonstrated for any measurement system. The deviation of the measurement points from the trend line is too large, and the coefficient of determination R2 does not reach a minimum target value of 0.95. In order to improve the measurement accuracy, it is proposed to measure with a maximum of 10% standard deviation. Accordingly, OCT and PCM would be suitable for structures with an aspect ratio up to 0.4.
The limitations and recommendations resulting from the investigation of the measurement quality based on the measurement deviation and repeatability are applied below to the valid measurements map in Figure 13. In addition, application areas are defined. An application area extends from the lower resolution limit (left limit) via the maximum aspect ratio (upper limit) to the maximum measurable structure size (right limit). It is assumed that the maximum structure width is equal to the edge length of the field of view of the measurement system measuring the maximum aspect ratio. Restrictions limit the application areas based on the investigation of the measurement quality. With respect to Figure 13, the smallest structures with a width less than 1.4 µm can be measured by ESEM and AFM. Medium-sized structures can be measured with OCT. If the structure widths of a sample vary greatly in width, the WLI, µCT, and PCM are suitable because they cover a large structure size range. Structures with high aspect ratios greater than two can preferably be examined with the ESEM, WLI, and MME.

4.2. Undercut and Enclosed Structure Analysis: Evaluation of the Minimum Structure Size and Maximum Cover Heights

4.2.1. White Light Interferometer

The WLI measurements of the chosen slanted gratings and the covered pillars are shown in Figure 14. In (a), the 40° inclined structures with a structure size of 1 µm, 2 µm, and 3 µm are captured from the top view. The respective slice is shown below. In the slice, the ground plane and structure plateaus are connected via less isolated measuring points. These isolated measurement points indicate that the measurement resolution is not sufficient to reconstruct the slanted edge or undercuts for these structure sizes. The larger slanted edges with a height of 10 µm, 20 µm, and 30 µm with a 40° angle in (b) can be reconstructed. Based on the measurements of the larger slanted gratings in (b) and the enclosed structures with different cover heights hc (0.5 µm–2 µm) in each quadrant in (c), it is clear that the WLI used can only look at a structure from the top and not into or through the structure. This is due to the signal recording of the measurement system. In the specific frequency domain analysis, only the interference signal of the sample’s surface is used to calculate the height values [28]. Therefore, the underlying surface cannot be reconstructed.
The reconstruction quality of the WLI depends on the slant angle of the gratings as shown in Figure 15. The density of the measurement points is only sufficient to reconstruct the edges from a slant angle of at least 30°. The reflected light at the edges must be captured by the WLI objective (ZeGage, Zygo, Weiterstadt, Germany, NA 0.5) to enable the reconstruction of slanted edges. Using a high numerical aperture (NA) objective, smaller slant angles can be captured and reconstructed. The exact slant limit could be tested using a sinusoidal structure since it has a continuous change in inclination as demonstrated in [51].

4.2.2. Atomic Force Microscope

The AFM measurements of the slanted gratings and the covered pillars are shown in Figure 16. Only the smaller slanted structures with a size of 1 µm in (a), 2 µm in (b), 3 µm in (c) could be scanned. For example, the measurement of the samples with a 40° slant angle are displayed in Figure 16. The scanned edges in (a) to (c) are blurred and partially distorted, which indicates that the tip is stuck during movement. The standard tip (Tap300Al-G, BudgetSensors, Sofia, Bulgaria, tip height: 17 µm, half cone angles 20°–30°, radius 10 nm) is limited by its geometry. In measurements of larger structures, the tip would have remained stuck in the structures until it broke. According to Figure 16b,c, two tip behaviors can be distinguished. In the first, the tip touches the edges of the structure with its side surfaces, and the edges vanish as shown in (b) and (c) in red-lined area 1. In the scan lines below, there are tip displacements similar to tip contamination artifacts [52,53]. In the second behavior, the tip jumps over the edges as given in (b) and (c) in area 2. The structures look as if they have broken off, although the structures are defect-free. The phenomena are comparable to known misinterpretations by tip-sample convolution [52]. Since measurement defects cannot be sufficiently distinguished from slanted edges, a standard AFM is not recommended for slanted structures, even in the smaller measurement range below 3 µm. However, there are special AFMs with individual tips that are designed for high aspect ratios [23] as well as undercuts and sidewalls [54,55]. The covered pillars of Figure 16d cannot be measured using a basic tactile measurement technique. One possibility would be to measure these covered pillars with contact resonance atomic force microscopy as presented in [56]. The ripples in the slices are caused by the 2PP manufacturing of the cover layer and can also be found in the measurement of the WLI in Figure 14c (different scaling).

4.2.3. Optical Coherence Tomograph

The slanted gratings with a size of 10 µm, 20 µm, and 30 µm could be captured by the OCT as shown based on the 40° structures in Figure 17a–c. Smaller slanted structures (1 µm–3 µm) could not be imaged. Especially in the processed slices of the OCT images, the structures appear several times with low intensity (2). These artifacts are caused by the autocorrelation signal which leads to double interpretations. They do not influence the dominant measurement signal (1). Therefore, only the measurement signal (1) is analyzed. The reflection strength at the slanted edges decreases compared to horizontal interfaces, as the reflection cannot be completely captured by the objective. Similarly, to the WLI, the measurability of the edges depends on the NA of the objective. Compared to the WLI, however, undercuts and covered structures can also be measured. In the slice of Figure 17d, the interfaces of the covered structures are visible. At this point, it is important to mention that the scaling of the height refers to air. The average pillar width is 3.7 ± 0.2 µm, measured in the horizontal direction along the blue slice line. In this measurement, the measured pillar widths are independent of the cover thickness hc (left side: 0.5 µm, right side: 2 µm). It was proven that the measurement with the OCT through a 2 µm thick polymer is successful. It is assumed that much thicker layers than 2 µm can be penetrated since the system has a limiting scan depth of 600 µm and comparable systems (Fourier Domain OCTs) have been shown to penetrate much thicker layers than the tested 2 µm ([57], crack in 1500 µm depth; [58], 50 µm film thickness). A measurement can be carried out without any difficulties as long as the refraction effects along the structure edges still allow for a reliable measurement of the enclosed structure.

4.2.4. Environmental Scanning Electron Microscope

Using the ESEM, the samples are tilted by 30° in order to detect undercuts based on their lateral profile. The captured ESEM images of the 1 µm, 2 µm, and 3 µm structures as well as the 30 µm structure with a slant angle of 40° are shown in Figure 18a,b. The profile can be reconstructed using the measured distances and recalculating the vertical scale with the factor of the tilt angle. A reconstruction or measurement of the undercuts in the middle of the structure is impossible using the intensity image. Cavity structures such as the covered pillars in Figure 18c cannot be detected unless the cover layer is thin enough that its charge density differs from the surrounding surface areas. At a cover thickness of 0.5 µm (top left), different intensities can be seen at the location of the hollow pillars. It is assumed that the difference in intensity is due to the waviness of the cover layer, which was also seen in the WLI (Figure 14c) and AFM images (Figure 16d).

4.2.5. Phase Contrast Microscope

The PCM could capture all slanted structures from 1 µm to 30 µm. The slanted gratings from 1 µm to 3 µm with a slant angle of 20°–40° and the 30 µm structure with 40° slant are shown in Figure 19a,b. The smaller structures in (a) suffer under diffraction effects at the edges. Since the PCM measurement is based on transmitted light, the measurements are susceptible to diffraction effects, which lead to vanished structure edges, especially of slanted structures, as can also be seen in (a). The greater the angle of inclination, the less clearly the geometry is displayed. Due to the refraction of light at the structure’s edges and the phase shift filters of the PCM, only the left side of the slanted grating elements appears bright. The right side with the undercut appears dark. As a result, the PCM light image of (b) implies that the structure is tilted to the left, although it is tilted to the right as indicated by the red reference line in the image. Consequently, slanted gratings cannot be captured reliably with the PCM. In contrast, the covered pillars are reliably recognizable in the PCM images in (c). Depending on the focus position, the hollow pillars or the base pillars appear sharp. In (c), the focus was set on the hollow pillars. The intensity slice through the top view image does not contain any helpful information; therefore, it is not shown. Using the image in the top view, an average pillar width of 3.6 ± 0.2 µm could be measured. This is consistent with the OCT measurements (3.7 ± 0.2 µm). The possible penetration depth of the PCM depends on the transmission properties of the polymer. The limit of the PCM’s penetration depth could not be identified using a maximum cover layer of 2 µm.

4.2.6. Imagining Mueller Matrix Ellipsometer

The imaging MME enables an analysis of the polarization properties of the slanted gratings as shown for the smaller structures of 1 µm–3 µm and a slant angle of 10°–40° in Figure 20a and the corresponding Mueller matrix in (b). In this case, the Mueller matrix element M32 was selected due to its high contrast to determine the influence of the slanted edges on the structure’s polarization properties. The different slanted gratings (0°–40°) of the first slanted grating image (SG1) in (b) show in the Mueller matrix element M32 that all 1 µm slanted gratings (left column) do not differ in their polarization properties. To confirm this statement, two additional 2PP slanted gratings SG2 in (c) and SG3 in (d) were printed and measured. Consequently, it is assumed that the geometry of 1 µm gratings with slant angles from 0° to 40° does not differ significantly. However, the polarization properties of the 2 µm and 3 µm structures change with an increasing slant angle and an increasing structure width. For example, as the slant angle increases, the areas with higher Mueller matrix values become larger, and more red pixels (especially at the gratings’ edges) can be observed in more detail in (b). If the polarization property of a reference structure (same structure, same material) is known, the shape fidelity and the polarization property of the test structure can be compared using the Mueller matrices. In addition, a simulation of the polarization behavior can be applied for a more reliable structural evaluation in order to be independent of system-induced influences [47].
As before, the covered pillars were also measured with the imaging MME to investigate the influence of the cover thickness. According to the covered pillars in Figure 21a and the corresponding Mueller matrix elements M33 and M34 in (b) and (c), the values fluctuate over the entire structure area and no significant difference can be seen in the quadrants with different cover heights. That implies that the polarization properties of the hollow pillars differ over the entire test structure independently from the cover height. Thus, it could be shown that the MME is not significantly influenced by the thickness of the cover in a range of 0.5 µm to 2 µm. The difference between the basic pillar probe of Figure 3g and the covered pillar probe in the polarization property can be seen at the edges of the pillars. In comparison, the edges of the covered pillars are not captured, only the position of the pillar can be estimated due to the dominant polarization change in the middle of the pillar. It is assumed that this difference is due to the inverse geometry (hollow instead of filled pillars), the additional base pillars, and the wet development process of the 2PP structure. If the cavities cannot be completely rinsed out during the wet development process, the propagation and the polarization behavior are changed.

4.2.7. Interim Summary of the Undercut and Enclosed Structure Analysis

The investigations of complex polymer structures with slant angles of 0° to 40° and structure heights of 1 µm to 30 µm show that the application areas of the individual measurement systems differ as summarized in Figure 22. Accordingly, ESEM and MME measurements are suitable for evaluating small slanted structures with a height of less than 10 µm. The slanted edges of structures larger than 10 µm and slant angles equal to and greater than 30° can be reliably captured by WLI and OCT. The suitability of WLI and OCT to detect slanted edges depends on the numerical aperture of their objectives. It is assumed that the µCT can also reliably reconstruct slanted edges from a structure size of 10 µm since its resolution is comparable to the WLI’s and OCT’s (with respect to the pillar analysis in Figure 8). Due to the sample rotation during the measurement, it should be possible to record the edges regardless of their angles. The AFM and PCM are not suitable for evaluating slanted structures.
The investigation results of covered structures with a cover height of 0.5 µm to 2 µm are illustrated in Figure 23. It could be shown that the tested cover height is not a challenge for the OCT, PCM, and MME and that the limit of penetration was not reached. OCT and µCT in particular are suitable for penetrating deeper polymers. Comparable OCT systems were able to detect objects in thicker polymers of 50 µm [58] and 1500 µm [57]. The used µCT penetrates through a 500 µm thick borosilicate waver (n = 1.52 @ 590 nm, not X-ray related) and a 1.5 µm polymer layer (IP-Dip, n = 1.55 @ 590 nm, not X-ray related) during the measurement of the basic pillars. Comparable µCT systems penetrate polymers based on bisphenylglycidyl dimethacrylate (Bis-GMA) and triethylene glycol dimethacrylate (TEGDMA) with thicknesses of 1000 µm [59] and more than 5000 µm [60]. If necessary, for the µCT measurement, the polymer must be enriched with particles that support X-ray absorption [50,60]. The WLI, AFM, and ESEM only measure the surfaces and are not suitable for measuring structures below the surface. At this point it is important to note that the measurement capability in terms of penetration always depends on the sample’s material. We used a standard acrylate (IP-Dip) and in principle the results are transferable to all materials that have similar absorption and scattering properties at the measurement system’s specific wavelength.

5. Conclusions

In the investigations, WLI, AFM, OCT, ESEM, PCM, and imaging MME are utilized for the non-destructive evaluation of complex polymer-based micro and nanostructures. According to the pillar analysis, the practical limits of the lateral resolution and aspect ratio were identified for each measurement system, and application areas were defined with respect to the measurement quality. The ESEM and AFM were suitable for measuring the smallest structures down to 0.4 µm, while the WLI, OCT, µCT, PCM, and MME were effective for larger structures starting in the single-digit micrometer range. The WLI, µCT, and PCM in particular cover a wide range of structural width variance. Furthermore, some analysis results of slanted and enclosed structures should be highlighted. According to the analysis, the ESEM (with a tilted sample) and the MME are suitable for measuring slanted edges from 0° to 40° on small structures that are equal or larger than 1 µm. The OCT, µCT, PCM, and imaging MME are reliably suitable for penetrating materials with similar absorption and scattering properties as the standard acrylate (IP-Dip) used, up to a thickness of 2 µm. The OCT and µCT are especially suitable measurement systems for measuring in thicker polymer layers up to the millimeter range. On the basis of the defined application areas, suitable non-destructive measurement systems can be quickly selected to reliably measure polymer structures depending on their specific geometry.

Author Contributions

Conceptualization, S.B. and A.H.; methodology, S.B., L.S., T.S., D.S. and J.G.; validation, S.B.; writing—original draft preparation, S.B.; writing—review and editing, A.H. All authors have read and agreed to the published version of the manuscript.

Funding

This work was supported by German Federal Ministry of Education and Research (BMBF FH Kooperativ grant no. 13FH080KX1 and KMU Innovativ grant no. 01IS20080C). The publication is funded by Aalen University of Applied Sciences.

Data Availability Statement

Data are contained within the article.

Conflicts of Interest

The authors declare no conflicts of interest.

References

  1. Yang, L.; Wei, J.; Ma, Z.; Song, P.; Ma, J.; Zhao, Y.; Huang, Z.; Zhang, M.; Yang, F.; Wang, X. The Fabrication of Micro/Nano Structures by Laser Machining. Nanomaterials 2019, 9, 1789. [Google Scholar] [CrossRef] [PubMed]
  2. Panusa, G.; Pu, Y.; Wang, J.; Moser, C.; Psaltis, D. Fabrication of Sub-Micron Polymer Waveguides through Two-Photon Polymerization in Polydimethylsiloxane. Polymers 2020, 12, 2485. [Google Scholar] [CrossRef] [PubMed]
  3. Mattelin, M.-A.; Radosavljevic, A.; Missinne, J.; Cuypers, D.; Kommeren, S.; Vandael, J.; ter Meulen, J.M.; Verduyckt, L.; van Steenberge, G. Fabrication and replication of high efficiency blazed gratings with grayscale electron beam lithography and UV nanoimprint lithography. In Proceedings of the Advanced Fabrication Technologies for Micro/Nano Optics and Photonics XIII, San Francisco, CA, USA, 2–5 February 2020. [Google Scholar] [CrossRef]
  4. Lee, K.D.; Ahn, S.W.; Kim, S.H.; Lee, S.H.; Park, J.D.; Yoon, P.W.; Kim, D.H.; Lee, S.S. Nanoimprint technology for nano-structured optical devices. Curr. Appl. Phys. 2006, 6, e149–e153. [Google Scholar] [CrossRef]
  5. Jang, C.; Mercier, O.; Bang, K.; Li, G.; Zhao, Y.; Lanman, D. Design and fabrication of freeform holographic optical elements. ACM Trans. Graph. 2020, 39, 1–15. [Google Scholar] [CrossRef]
  6. Lu, W.-G.; Xiao, R.; Liu, J.; Wang, L.; Zhong, H.; Wang, Y. Large-area rainbow holographic diffraction gratings on a curved surface using transferred photopolymer films. Opt. Lett. 2018, 43, 675–678. [Google Scholar] [CrossRef]
  7. Levola, T.; Laakkonen, P. Replicated slanted gratings with a high refractive index material for in and outcoupling of light. Opt. Express 2007, 15, 2067–2074. [Google Scholar] [CrossRef]
  8. Jin, G.; Liu, W.; Ye, Z.; Jia, W.; Xie, Y.; Zhou, C. High Efficiency Polarization-Independent Slanted Grating for RGB Bands. IEEE Photonics J. 2021, 13, 5100108. [Google Scholar] [CrossRef]
  9. Li, H.; Peng, X.; Guan, C.; Hu, H. Ultra-Precision Cutting and Characterization of Reflective Convex Spherical Blazed Grating Elements. Micromachines 2022, 13, 1115. [Google Scholar] [CrossRef]
  10. Al-Assaad, R.M. Physical characterization of nanoimprinted polymer nanostructures using visible light angular scatterometry. J. Micro/Nanolith. MEMS MOEMS 2008, 7, 13008. [Google Scholar] [CrossRef]
  11. Varshney, M.; Li, Y.; Srinivasan, B.; Tung, S.; Erf, G.; Slavik, M.F.; Ying, Y.; Fang, W. A Microfluidic Filter Biochip-Based Chemiluminescence Biosensing Method for Detection of Escherichia coli O157:H7. Trans. ASABE 2006, 49, 2061–2068. [Google Scholar] [CrossRef]
  12. Wengenmayr, R. Zellen unter Druck. Max Planck Forsch. 2021. [Google Scholar]
  13. Tsai, C.-H.D.; Tanaka, J.; Kaneko, M.; Horade, M.; Ito, H.; Taniguchi, T.; Ohtani, T.; Sakata, Y. An On-Chip RBC Deformability Checker Significantly Improves Velocity-Deformation Correlation. Micromachines 2016, 7, 176. [Google Scholar] [CrossRef] [PubMed]
  14. Keshavarz Motamed, P.; Abouali, H.; Poudineh, M.; Maftoon, N. Experimental measurement and numerical modeling of deformation behavior of breast cancer cells passing through constricted microfluidic channels. Microsyst. Nanoeng. 2024, 10, 7. [Google Scholar] [CrossRef] [PubMed]
  15. Hongbin, Y.; Guangya, Z.; Siong, C.F.; Shouhua, W.; Feiwen, L. Novel polydimethylsiloxane (PDMS) based microchannel fabrication method for lab-on-a-chip application. Sens. Actuators B Chem. 2009, 137, 754–761. [Google Scholar] [CrossRef]
  16. An, L.; Ji, F.; Zhao, E.; Liu, Y.; Liu, Y. Measuring cell deformation by microfluidics. Front. Bioeng. Biotechnol. 2023, 11, 1214544. [Google Scholar] [CrossRef]
  17. Kim, D.S.; Lee, S.H.; Ahn, C.H.; Lee, J.Y.; Kwon, T.H. Disposable integrated microfluidic biochip for blood typing by plastic microinjection moulding. Lab Chip 2006, 6, 794–802. [Google Scholar] [CrossRef]
  18. Memon, I.; Shen, Y.; Khan, A.; Woidt, C.; Hillmer, H. Highly uniform residual layers for arrays of 3D nanoimprinted cavities in Fabry–Pérot-filter-array-based nanospectrometers. Appl. Nanosci. 2016, 6, 599–606. [Google Scholar] [CrossRef]
  19. Peter Amalathas, A.; Alkaisi, M.M. Efficient light trapping nanopyramid structures for solar cells patterned using UV nanoimprint lithography. Mater. Sci. Semicond. Process. 2017, 57, 54–58. [Google Scholar] [CrossRef]
  20. Köhler, H. On Abbe’s Theory of Image Formation in the Microscope. Opt. Acta Int. J. Opt. 1981, 28, 1691–1701. [Google Scholar] [CrossRef]
  21. Kim, J.U.; Lee, S.; Kim, T. Recent Advances in Unconventional Lithography for Challenging 3D Hierarchical Structures and Their Applications. J. Nanomater. 2016, 2016, 7602395. [Google Scholar] [CrossRef]
  22. Lehmann, P.; Niehues, J.; Tereschenko, S. 3-D Optical Interference Microscopy at the Lateral Resolution. Int. J. Optomechatron. 2014, 8, 231–241. [Google Scholar] [CrossRef]
  23. Morimoto, T.; Kuroda, H.; Minomoto, Y.; Nagano, Y.; Kembo, Y.; Hosaka, S. Atomic Force Microscopy for High Aspect Ratio Structure Metrology. Jpn. J. Appl. Phys. 2002, 41, 4238–4241. [Google Scholar] [CrossRef]
  24. Meemon, P.; Yao, J.; Lee, K.-S.; Thompson, K.P.; Ponting, M.; Baer, E.; Rolland, J.P. Optical Coherence Tomography Enabling Non Destructive Metrology of Layered Polymeric GRIN Material. Sci. Rep. 2013, 3, 1709. [Google Scholar] [CrossRef]
  25. Lu, J.C.; Liu, Q.; Huang, S. Research on slanted trapezoidal surface relief grating. In Proceedings of the Holography, Diffractive Optics, and Applications IX, Hangzhou, China, 20–23 October 2019; Zhou, C., Sheng, Y., Cao, L., Eds.; SPIE: Bellingham, WA, USA, 2019; p. 81. ISBN 9781510630932. [Google Scholar]
  26. Wang, H.; Zhang, W.; Ladika, D.; Yu, H.; Gailevičius, D.; Wang, H.; Pan, C.-F.; Nair, P.N.S.; Ke, Y.; Mori, T.; et al. Two-Photon Polymerization Lithography for Optics and Photonics: Fundamentals, Materials, Technologies, and Applications. Adv. Funct. Mater. 2023, 33, 2214211. [Google Scholar] [CrossRef]
  27. de Groot, P. Method and Apparatus for Surface Topography Measurement by Spatial-Frequency Analysis of Interferograms. U.S. Patent 5,398,113, 14 March 1995. Available online: https://patents.google.com/patent/US5398113A/en (accessed on 5 June 2024).
  28. de Groot, P.; Colonna de Lega, X.; Kramer, J.; Turzhitsky, M. Determination of fringe order in white-light interference microscopy. Appl. Opt. 2002, 41, 4571–4578. [Google Scholar] [CrossRef]
  29. Sugimoto, Y.; Nakajima, Y.; Sawada, D.; Morita, K.; Abe, M.; Morita, S. Simultaneous AFM and STM measurements on the Si(111)−(7 × 7) surface. Phys. Rev. B 2010, 81, 245322. [Google Scholar] [CrossRef]
  30. Cerreta, A.; Vobornik, D.; Di Santo, G.; Tobenas, S.; Alonso-Sarduy, L.; Adamcik, J.; Dietler, G. FM-AFM constant height imaging and force curves: High resolution study of DNA-tip interactions. J. Mol. Recognit. 2012, 25, 486–493. [Google Scholar] [CrossRef]
  31. Albrecht, F.; Bischoff, F.; Auwärter, W.; Barth, J.V.; Repp, J. Direct Identification and Determination of Conformational Response in Adsorbed Individual Nonplanar Molecular Species Using Noncontact Atomic Force Microscopy. Nano Lett. 2016, 16, 7703–7709. [Google Scholar] [CrossRef]
  32. Dufrêne, Y.F.; Ando, T.; Garcia, R.; Alsteens, D.; Martinez-Martin, D.; Engel, A.; Gerber, C.; Müller, D.J. Imaging modes of atomic force microscopy for application in molecular and cell biology. Nat. Nanotechnol. 2017, 12, 295–307. [Google Scholar] [CrossRef]
  33. Bryant, P.J.; Miller, R.G.; Yang, R. Scanning tunneling and atomic force microscopy combined. Appl. Phys. Lett. 1988, 52, 2233–2235. [Google Scholar] [CrossRef]
  34. Thorlabs, Inc. EDU-AFM1/M Manual English. Available online: https://www.thorlabs.com/thorproduct.cfm?partnumber=EDU-AFM1/M (accessed on 9 June 2024).
  35. Hasko, D.; Kovác, J.; Satka, A.; Drzík, M.; Uherek, F.; Hubbard, G.; Allsopp, D.W.E. Structural and optical characterization of photonics structures prepared by nanoimprint technology. In Proceedings of the Photonics, Devices, and Systems IV, Prague, Czech Republic, 27–29 September 2008; Tománek, P., Senderáková, D., Hrabovský, M., Eds.; SPIE: Bellingham, WA, USA, 2008; p. 713824. [Google Scholar]
  36. VDI/VDE 5565; Optical Coherence Tomography (OCT): Process Descriptions. Verein Deutscher Ingenieure e.V.: Düsseldorf, Germany, 2023.
  37. Stifter, D.; Wiesauer, K.; Wurm, M.; Schlotthauer, E.; Kastner, J.; Pircher, M.; Götzinger, E.; Hitzenberger, C.K. Investigation of polymer and polymer/fibre composite materials with optical coherence tomography. Meas. Sci. Technol. 2008, 19, 74011. [Google Scholar] [CrossRef]
  38. Osari, K.; Unno, N.; Taniguchi, J.; Machinaga, K.; Ohsaki, T.; Sakai, N. Evaluation of filling behavior on UV nanoimprint lithography using release coating. Microelectron. Eng. 2010, 87, 918–921. [Google Scholar] [CrossRef]
  39. Schleunitz, A.; Spreu, C.; Vogler, M.; Atasoy, H.; Schift, H. Combining nanoimprint lithography and a molecular weight selective thermal reflow for the generation of mixed 3D structures. J. Vac. Sci. Technol. B Nanotechnol. Microelectron. Mater. Process. Meas. Phenom. 2011, 29, 06FC01. [Google Scholar] [CrossRef]
  40. LaFratta, C.N.; Li, L.; Fourkas, J.T. Soft-lithographic replication of 3D microstructures with closed loops. Proc. Natl. Acad. Sci. USA 2006, 103, 8589–8594. [Google Scholar] [CrossRef]
  41. Carl Zeiss Microscopy GmbH. Extend the Limits of Your Exploration: ZEISS Xradia 610 and 620 Versa. Product Brochure. 2024. Available online: https://www.zeiss.com/microscopy/de/produkte/roentgenmikroskopie/xradia-versa.html#accordionItem-818906634 (accessed on 5 June 2024).
  42. Szewczykowski, P.P.; Skarzynski, L. Application of the X-ray micro-computed tomography to the analysis of the structure of polymeric materials. Polimery 2019, 64, 12–22. [Google Scholar] [CrossRef]
  43. Cangül, S.; Adıgüzel, Ö.; Sağmak, S.; Evran, B. A micro-computed tomography evaluation of the change in volume of different bulk-fill composite materials caused by polymerization shrinkage. Int. Dent. Res. 2021, 11, 75–82. [Google Scholar] [CrossRef]
  44. Xie, X.-M.; Kong, X.-M.; Xiao, T.-J.; Yang, Y.; Gao, N.; Tanioka, A. Interface-Induced Coarsening Process in Polymer Blends. J. Colloid Interface Sci. 2001, 234, 24–27. [Google Scholar] [CrossRef]
  45. Xu, X.; Zhu, T.; Yan, X.; Zhang, C. Temporal evolution of phase morphology of polypropylene/poly(ethylene octene) elastomer binary polymer blends by phase contrast microscope. J. Appl. Polym. Sci. 2007, 104, 2778–2784. [Google Scholar] [CrossRef]
  46. Azzam, R.M. Photopolarimetric measurement of the Mueller matrix by Fourier analysis of a single detected signal. Opt. Lett. 1978, 2, 148. [Google Scholar] [CrossRef]
  47. Käseberg, T.; Grundmann, J.; Siefke, T.; Klapetek, P.; Valtr, M.; Kroker, S.; Bodermann, B. Mueller Matrix Ellipsometric Approach on the Imaging of Sub-Wavelength Nanostructures. Front. Phys. 2022, 9, 814559. [Google Scholar] [CrossRef]
  48. Cumpston, B.H.; Ananthavel, S.P.; Barlow, S.; Dyer, D.L.; Ehrlich, J.E.; Erskine, L.L.; Heikal, A.A.; Kuebler, S.M.; Lee, I.-Y.S.; McCord-Maughon, D.; et al. Two-photon polymerization initiators for three-dimensional optical data storage and microfabrication. Nature 1999, 398, 51–54. [Google Scholar] [CrossRef]
  49. Schmid, M.; Ludescher, D.; Giessen, H. Optical properties of photoresists for femtosecond 3D printing: Refractive index, extinction, luminescence-dose dependence, aging, heat treatment and comparison between 1-photon and 2-photon exposure. Opt. Mater. Express 2019, 9, 4564. [Google Scholar] [CrossRef]
  50. Crica, L.E.; Wengenroth, J.; Tiainen, H.; Ionita, M.; Haugen, H.J. Enhanced X-ray absorption for micro-CT analysis of low density polymers. J. Biomater. Sci. Polym. Ed. 2016, 27, 805–823. [Google Scholar] [CrossRef] [PubMed]
  51. Gao, F.; Leach, R.K.; Petzing, J.; Coupland, J.M. Surface measurement errors using commercial scanning white light interferometers. Meas. Sci. Technol. 2008, 19, 15303. [Google Scholar] [CrossRef]
  52. Gołek, F.; Mazur, P.; Ryszka, Z.; Zuber, S. AFM image artifacts. Appl. Surf. Sci. 2014, 304, 11–19. [Google Scholar] [CrossRef]
  53. Nie, H.Y.; McIntyre, N.S. Unstable amplitude and noisy image induced by tip contamination in dynamic force mode atomic force microscopy. Rev. Sci. Instrum. 2007, 78, 23701. [Google Scholar] [CrossRef]
  54. Cho, S.-J.; Ahn, B.-W.; Kim, J.; Lee, J.-M.; Hua, Y.; Yoo, Y.K.; Park, S. Three-dimensional imaging of undercut and sidewall structures by atomic force microscopy. Rev. Sci. Instrum. 2011, 82, 23707. [Google Scholar] [CrossRef]
  55. Murayama, K.; Gonda, S.; Koyanagi, H.; Terasawa, T.; Hosaka, S. Side-Wall Measurement using Tilt-Scanning Method in Atomic Force Microscope. Jpn. J. Appl. Phys. 2006, 45, 5423. [Google Scholar] [CrossRef]
  56. Wang, W.; Ma, C.; Chen, Y. Measurement of undercut etching by contact resonance atomic force microscopy. Appl. Phys. Lett. 2020, 117, 023103. [Google Scholar] [CrossRef]
  57. Liu, P.; Groves, R.M.; Benedictus, R. Optical Coherence Tomography for the Study of Polymer and Polymer Matrix Composites. Strain 2014, 50, 436–443. [Google Scholar] [CrossRef]
  58. Yao, J.; Rolland, J.P. Optical Coherence Tomography for Polymer Film Evaluation; IntechOpen: London, UK, 2020. [Google Scholar] [CrossRef]
  59. Sun, J.; Lin-Gibson, S. X-ray microcomputed tomography for measuring polymerization shrinkage of polymeric dental composites. Dent. Mater. 2008, 24, 228–234. [Google Scholar] [CrossRef]
  60. Cho, E.; Sadr, A.; Inai, N.; Tagami, J. Evaluation of resin composite polymerization by three dimensional micro-CT imaging and nanoindentation. Dent. Mater. 2011, 27, 1070–1078. [Google Scholar] [CrossRef]
Metrology 04 00040 g001
Figure 1. The shot with primary electrons causes secondary electrons to be emitted at the sample surface (a). During the scan of the sample surface, the dose of emitted secondary electrons is significantly higher at the structure edges than at the surface areas (b). In order to determine the width of the structure, the ESEM images are binarized to identify the edges of the structure. The width of the structure w is equal to the distance between the outer edges (1) or to the diameter of the circular surface (2) (c).
Figure 1. The shot with primary electrons causes secondary electrons to be emitted at the sample surface (a). During the scan of the sample surface, the dose of emitted secondary electrons is significantly higher at the structure edges than at the surface areas (b). In order to determine the width of the structure, the ESEM images are binarized to identify the edges of the structure. The width of the structure w is equal to the distance between the outer edges (1) or to the diameter of the circular surface (2) (c).
Metrology 04 00040 g001
Metrology 04 00040 g002
Figure 2. The phase difference in the propagation through the sample ∆φ (a) is visible in the intensity image of the top view (b). After binarization, the structure width w is equal to the diameter of the white circle (c).
Figure 2. The phase difference in the propagation through the sample ∆φ (a) is visible in the intensity image of the top view (b). After binarization, the structure width w is equal to the diameter of the white circle (c).
Metrology 04 00040 g002
Metrology 04 00040 g003aMetrology 04 00040 g003b
Figure 3. The overview shows examples of the captured images of the investigated measurement systems WLI (a), AFM (b), OCT (c), ESEM (d), µCT (e), PCM (f), and MME (g). The respective measurement characteristics such as measurement outputs and the field of view are compared using selected pillars. All images show pillars with a width of 2.8 µm and a height of 0.5 µm. The scaling of the height was adapted to the individual measurement in order to clearly display the measurement characteristics.
Figure 3. The overview shows examples of the captured images of the investigated measurement systems WLI (a), AFM (b), OCT (c), ESEM (d), µCT (e), PCM (f), and MME (g). The respective measurement characteristics such as measurement outputs and the field of view are compared using selected pillars. All images show pillars with a width of 2.8 µm and a height of 0.5 µm. The scaling of the height was adapted to the individual measurement in order to clearly display the measurement characteristics.
Metrology 04 00040 g003aMetrology 04 00040 g003b
Metrology 04 00040 g004
Figure 4. A polymer-based pillar array is used to investigate the practical limits of the lateral resolution and aspect ratio. The pillar width w varies from 0.2 µm to 20 µm and the height h from 0.5 µm to 4 µm. The base layer height b of 1.5 µm is constant (a). The pillars are rotationally symmetrical and measured in the horizontal (blue) and vertical (orange) directions (b,c). An integer multiple of pillars is arranged in a square array of 120 µm (c).
Figure 4. A polymer-based pillar array is used to investigate the practical limits of the lateral resolution and aspect ratio. The pillar width w varies from 0.2 µm to 20 µm and the height h from 0.5 µm to 4 µm. The base layer height b of 1.5 µm is constant (a). The pillars are rotationally symmetrical and measured in the horizontal (blue) and vertical (orange) directions (b,c). An integer multiple of pillars is arranged in a square array of 120 µm (c).
Metrology 04 00040 g004
Metrology 04 00040 g005
Figure 5. The 2PP pillar test structure captured by a laser scanning microscope consists of 26 pillar arrays. Structures with an aspect ratio of 5 (array 7, 14, and 21) are excluded from the investigation due to their defects.
Figure 5. The 2PP pillar test structure captured by a laser scanning microscope consists of 26 pillar arrays. Structures with an aspect ratio of 5 (array 7, 14, and 21) are excluded from the investigation due to their defects.
Metrology 04 00040 g005
Metrology 04 00040 g006
Figure 6. The slanted gratings vary in height h (=width w) between 1 µm and 30 µm. The slant angle is in the range of 0° to 40° (a). At least three elements are investigated based on the top and the cross-section view in the x-direction (blue line) (b). The microscope image of the slanted gratings shows the arrangement of the sample (c).
Figure 6. The slanted gratings vary in height h (=width w) between 1 µm and 30 µm. The slant angle is in the range of 0° to 40° (a). At least three elements are investigated based on the top and the cross-section view in the x-direction (blue line) (b). The microscope image of the slanted gratings shows the arrangement of the sample (c).
Metrology 04 00040 g006
Metrology 04 00040 g007
Figure 7. The coved pillar structure is based on hollow pillars (width w = 4 µm, height h = 4 µm) and a varying cover thickness hc. The cover thickness hc ranges from 0.5 µm to 2 µm. To be able to flush the hollow pillars free during the 2PP manufacturing process, the structure is supported by base pillars (height hb = 4 µm) (a). The pillars are symmetrically offset (b,c). The covered pillars are distributed over an area of 80 µm × 80 µm and the quadrants have the given cover thicknesses hc (c). The three-dimensional slice shows how the hollow and base pillars are arranged in relation to each other and where the cover thickness hc is measured (d).
Figure 7. The coved pillar structure is based on hollow pillars (width w = 4 µm, height h = 4 µm) and a varying cover thickness hc. The cover thickness hc ranges from 0.5 µm to 2 µm. To be able to flush the hollow pillars free during the 2PP manufacturing process, the structure is supported by base pillars (height hb = 4 µm) (a). The pillars are symmetrically offset (b,c). The covered pillars are distributed over an area of 80 µm × 80 µm and the quadrants have the given cover thicknesses hc (c). The three-dimensional slice shows how the hollow and base pillars are arranged in relation to each other and where the cover thickness hc is measured (d).
Metrology 04 00040 g007
Metrology 04 00040 g008
Figure 8. In the pillar analysis, the plotted pillar dimensions could be measured by the referenced measurement systems.
Figure 8. In the pillar analysis, the plotted pillar dimensions could be measured by the referenced measurement systems.
Metrology 04 00040 g008
Metrology 04 00040 g009
Figure 9. The µCT images of the pillar with a width of 1.4 and a height of 0.5 µm are shown. The individual pillars are visible in the top view (a). The cross-section at the blue line is shown in (b). There are slight diffraction effects.
Figure 9. The µCT images of the pillar with a width of 1.4 and a height of 0.5 µm are shown. The individual pillars are visible in the top view (a). The cross-section at the blue line is shown in (b). There are slight diffraction effects.
Metrology 04 00040 g009
Metrology 04 00040 g010
Figure 10. The measurement deviation corresponding to the structure width of comparable geometrical-based systems (WLI, AFM, OCT, ESEM, PCM) is shown to evaluate the measurement quality. Horizontal (►) and vertical (▲) measurements are plotted separately. The aspect ratio a of the measured pillars can be classified using the color bar.
Figure 10. The measurement deviation corresponding to the structure width of comparable geometrical-based systems (WLI, AFM, OCT, ESEM, PCM) is shown to evaluate the measurement quality. Horizontal (►) and vertical (▲) measurements are plotted separately. The aspect ratio a of the measured pillars can be classified using the color bar.
Metrology 04 00040 g010
Metrology 04 00040 g011
Figure 11. The WLI image (a) and the ESEM image (b) show pillars with an expected width of 0.8 µm and a height of 0.5 µm. The measurment lines are marked in red. The offset between the horizontal and vertical diameters is 20% of the expected structure width of 0.8 µm.
Figure 11. The WLI image (a) and the ESEM image (b) show pillars with an expected width of 0.8 µm and a height of 0.5 µm. The measurment lines are marked in red. The offset between the horizontal and vertical diameters is 20% of the expected structure width of 0.8 µm.
Metrology 04 00040 g011
Metrology 04 00040 g012
Figure 12. The influence of the aspect ratio is evaluated on the basis of the repeatability (standard deviation) of the measurements. The greater the gradient m of the trend line, the greater the impact of the aspect ratio on the measurement quality.
Figure 12. The influence of the aspect ratio is evaluated on the basis of the repeatability (standard deviation) of the measurements. The greater the gradient m of the trend line, the greater the impact of the aspect ratio on the measurement quality.
Metrology 04 00040 g012
Metrology 04 00040 g013
Figure 13. In the map of the application areas, the colored dots indicate valid measurements of the respective measurement systems. The application areas were defined taking into account the quality of the measurement. The application areas range from the lower resolution limit (left limit) to the maximum aspect ratio (upper limit) to the maximum structure width (right limit).
Figure 13. In the map of the application areas, the colored dots indicate valid measurements of the respective measurement systems. The application areas were defined taking into account the quality of the measurement. The application areas range from the lower resolution limit (left limit) to the maximum aspect ratio (upper limit) to the maximum structure width (right limit).
Metrology 04 00040 g013
Metrology 04 00040 g014
Figure 14. The slices of the 1 µm to 3 µm slanted structures with a 40° slant angle show that the edges cannot be reconstructed sufficiently since only a few measurement points are detected (a). The comparable structures with a 10 µm to 30 µm height have reliably measured edges on the top sides (b). The undercuts could not be captured. This also becomes clear in the image of the covered pillars (c), as no undercuts can be captured below a transparent surface. The steps in the slice view of (c) are due to the different cover heights hc on top of the quadrants ranging from 0.5 µm to 2 µm.
Figure 14. The slices of the 1 µm to 3 µm slanted structures with a 40° slant angle show that the edges cannot be reconstructed sufficiently since only a few measurement points are detected (a). The comparable structures with a 10 µm to 30 µm height have reliably measured edges on the top sides (b). The undercuts could not be captured. This also becomes clear in the image of the covered pillars (c), as no undercuts can be captured below a transparent surface. The steps in the slice view of (c) are due to the different cover heights hc on top of the quadrants ranging from 0.5 µm to 2 µm.
Metrology 04 00040 g014
Metrology 04 00040 g015
Figure 15. The structures 10 µm to 30 µm are compared with respect to their slant angles. From a slant angle of 30° the edges are captured sufficiently. The minimum slant angle depends on the maximum angle of incidence (numerical aperture) of the WLI objective.
Figure 15. The structures 10 µm to 30 µm are compared with respect to their slant angles. From a slant angle of 30° the edges are captured sufficiently. The minimum slant angle depends on the maximum angle of incidence (numerical aperture) of the WLI objective.
Metrology 04 00040 g015
Metrology 04 00040 g016
Figure 16. The AFM measurements and the respective slices of the 40° slanted gratings with a structure size of 1 µm (a), 2 µm (b), and 3 µm (c) demonstrate the challenge of a tactile measurement. The AFM tip either sticks with its flange (red area 1) or jumps from the structure top to ground level (red area 2). The measurement of the covered pillars (d) shows that no enclosed structures can be detected, but the vertical resolution is very high and small height deviations (<100 nm) can be detected. In (a), the height scaling has been changed in order to improve the analysis of the structures.
Figure 16. The AFM measurements and the respective slices of the 40° slanted gratings with a structure size of 1 µm (a), 2 µm (b), and 3 µm (c) demonstrate the challenge of a tactile measurement. The AFM tip either sticks with its flange (red area 1) or jumps from the structure top to ground level (red area 2). The measurement of the covered pillars (d) shows that no enclosed structures can be detected, but the vertical resolution is very high and small height deviations (<100 nm) can be detected. In (a), the height scaling has been changed in order to improve the analysis of the structures.
Metrology 04 00040 g016
Metrology 04 00040 g017
Figure 17. The top view of the OCT measurements and their cross-sections of the 10 µm (a), 20 µm, (b) and 30 µm (c) structures show that the undercuts of the slanted structures can be captured by OCT. The enclosed pillars (d) have a cover thickness of 0.5 to 2 µm. In this measurement range, no significant impact of the cover thickness on the measurement accuracy could be proven.
Figure 17. The top view of the OCT measurements and their cross-sections of the 10 µm (a), 20 µm, (b) and 30 µm (c) structures show that the undercuts of the slanted structures can be captured by OCT. The enclosed pillars (d) have a cover thickness of 0.5 to 2 µm. In this measurement range, no significant impact of the cover thickness on the measurement accuracy could be proven.
Metrology 04 00040 g017
Metrology 04 00040 g018
Figure 18. The ESEM images show small structures (1 µm, 2 µm, and 3 µm) with different tilt angles in (a) and a large structure (30 µm, 40°) in (b). In the ESEM images of (a,b), the sample was tilted by 30°. The covered pillars in (c) are captured from the top view. In the two right-hand quadrants (top: hc = 2 µm, bottom: hc = 1.5 µm), a difference in charge density can be seen on the structure’s surface. This charge difference was caused by ESEM scans of the left side taken shortly before.
Figure 18. The ESEM images show small structures (1 µm, 2 µm, and 3 µm) with different tilt angles in (a) and a large structure (30 µm, 40°) in (b). In the ESEM images of (a,b), the sample was tilted by 30°. The covered pillars in (c) are captured from the top view. In the two right-hand quadrants (top: hc = 2 µm, bottom: hc = 1.5 µm), a difference in charge density can be seen on the structure’s surface. This charge difference was caused by ESEM scans of the left side taken shortly before.
Metrology 04 00040 g018
Metrology 04 00040 g019
Figure 19. The PCM image of the small structures (1 µm, 2 µm, and 3 µm) shows increased diffraction effects while increasing the slant angle (a). According to the image of the large structure (30 µm, 40°), it seems like the slanted structure is tilted to the left, although it is tilted to the right. A reference slice has been drawn in red corresponding to the position of the PCM image’s slanted gratings (b). In contrast, the covered pillars are recognizable. When testing a 2 µm cover height (top right), the maximum penetration depth of the PCM has not yet been reached (c).
Figure 19. The PCM image of the small structures (1 µm, 2 µm, and 3 µm) shows increased diffraction effects while increasing the slant angle (a). According to the image of the large structure (30 µm, 40°), it seems like the slanted structure is tilted to the left, although it is tilted to the right. A reference slice has been drawn in red corresponding to the position of the PCM image’s slanted gratings (b). In contrast, the covered pillars are recognizable. When testing a 2 µm cover height (top right), the maximum penetration depth of the PCM has not yet been reached (c).
Metrology 04 00040 g019
Metrology 04 00040 g020
Figure 20. The intensity image in (a) shows the 1 µm, 2 µm, and 3 µm structures (left to right) with slant angles from 0° to 40° (top to bottom). The respective polarization property is displayed by the local M32 values of the Mueller matrix in (b). In comparison, two additional samples SG2 (c) and SG3 (d) were measured. They show the same polarization behavior as SG1. This proves that the polarization behavior can be measured with imaging MME from a structure width of 1 µm, even with slanted gratings.
Figure 20. The intensity image in (a) shows the 1 µm, 2 µm, and 3 µm structures (left to right) with slant angles from 0° to 40° (top to bottom). The respective polarization property is displayed by the local M32 values of the Mueller matrix in (b). In comparison, two additional samples SG2 (c) and SG3 (d) were measured. They show the same polarization behavior as SG1. This proves that the polarization behavior can be measured with imaging MME from a structure width of 1 µm, even with slanted gratings.
Metrology 04 00040 g020
Metrology 04 00040 g021
Figure 21. The polarization properties of the covered pillar structure in (a) are represented by the Mueller matrix elements M33 and M34 in (b,c) using the imaging MME. The polarization properties of the covered pillars vary across the entire structure surface, regardless of the cover height of 0.5 µm to 2 µm. Consequently, it has been demonstrated that the MME is not significantly affected by the cover thickness.
Figure 21. The polarization properties of the covered pillar structure in (a) are represented by the Mueller matrix elements M33 and M34 in (b,c) using the imaging MME. The polarization properties of the covered pillars vary across the entire structure surface, regardless of the cover height of 0.5 µm to 2 µm. Consequently, it has been demonstrated that the MME is not significantly affected by the cover thickness.
Metrology 04 00040 g021
Metrology 04 00040 g022
Figure 22. The map shows the application areas of WLI, AFM, OCT, ESEM, µCT, PCM, and MME with respect to their suitability to capture slanted edges in a specific structure size range. The dashed measurement range of the µCT is based on the assumption that structures of 1.4 µm and larger can be resolved and that the slant angle does not influence the measurement quality due to the rotation during the measurement.
Figure 22. The map shows the application areas of WLI, AFM, OCT, ESEM, µCT, PCM, and MME with respect to their suitability to capture slanted edges in a specific structure size range. The dashed measurement range of the µCT is based on the assumption that structures of 1.4 µm and larger can be resolved and that the slant angle does not influence the measurement quality due to the rotation during the measurement.
Metrology 04 00040 g022
Metrology 04 00040 g023
Figure 23. The OCT, µCT, PCM, and imaging MME can penetrate polymer layers of 2 µm without a significant loss of measurement accuracy. OCT and µCT in particular are suitable for deep penetration of polymer layers according to the literature * [57,58] ** [59,60]. The WLI, AFM, and ESEM are only suitable for measuring structures on surfaces.
Figure 23. The OCT, µCT, PCM, and imaging MME can penetrate polymer layers of 2 µm without a significant loss of measurement accuracy. OCT and µCT in particular are suitable for deep penetration of polymer layers according to the literature * [57,58] ** [59,60]. The WLI, AFM, and ESEM are only suitable for measuring structures on surfaces.
Metrology 04 00040 g023
Table 1. Used measurement systems and their main settings in the investigation using polymer-based structures.
Table 1. Used measurement systems and their main settings in the investigation using polymer-based structures.
SystemNameManufacturerDimensionsMain Settings
WLINewView8300Zygo, Weiterstadt, Germany3d50×, NA 0.5, 1024 × 1024 px
AFMEDU-AFM1/MThorlabs, New Jersey, Vereinigte Staaten3d250 × 250 px, 200 px/s, different magnifications
OCTSpectral domain OCT according to VDI/VDE 5565non-commercial 3d50×, NA 0.8, FOV 0.1 (factor)
ESEMPrisma ESEMThermo Fisher Scientific, Massachusetts, Vereinigte Staaten2dLDV, 5 kV, 50 Pa, different magnifications
µCTXradia Versa 610Zeiss, Oberkochen, Germany3d40×, 80 kV, 10 W, different magnifications
PCMBZ-X800Keyence, Ōsaka, Japan2d40×, 960 × 720 px
MMEImaging Mueller matrix ellipsometry, Accurion EP4Park Systems, Suwon, Korea2d50×, incidence angle 40°, 500 nm
Table 2. Pillar geometry.
Table 2. Pillar geometry.
Width w [µm]
0.20.40.81.42.85.010.020.0
Height h [µm]0.52.50 (1)1.25 (2)0.63 (3)0.36 (4)0.18 (5)0.10 (6)
1.05.00 (7)2.50 (8)1.25 (9)0.71 (10)0.36 (11)0.20 (12)0.10 (13)
2.0 5.00 (14)2.50 (15)1.43 (16)0.71 (17)0.40 (18)0.20 (19)0.10 (20)
4.0 5.00 (21)2.86 (22)1.43 (23)0.80 (24)0.40 (25)0.20 (26)
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Burkert, S.; Schwörer, L.; Schubert, T.; Grundmann, J.; Stein, D.; Heinrich, A. Investigation of the Measurement Systems’ Suitability for the Non-Destructive Measurement of Complex Polymer-Based Micro and Nanostructures. Metrology 2024, 4, 673-694. https://doi.org/10.3390/metrology4040040

AMA Style

Burkert S, Schwörer L, Schubert T, Grundmann J, Stein D, Heinrich A. Investigation of the Measurement Systems’ Suitability for the Non-Destructive Measurement of Complex Polymer-Based Micro and Nanostructures. Metrology. 2024; 4(4):673-694. https://doi.org/10.3390/metrology4040040

Chicago/Turabian Style

Burkert, Selina, Lukas Schwörer, Tim Schubert, Jana Grundmann, David Stein, and Andreas Heinrich. 2024. "Investigation of the Measurement Systems’ Suitability for the Non-Destructive Measurement of Complex Polymer-Based Micro and Nanostructures" Metrology 4, no. 4: 673-694. https://doi.org/10.3390/metrology4040040

APA Style

Burkert, S., Schwörer, L., Schubert, T., Grundmann, J., Stein, D., & Heinrich, A. (2024). Investigation of the Measurement Systems’ Suitability for the Non-Destructive Measurement of Complex Polymer-Based Micro and Nanostructures. Metrology, 4(4), 673-694. https://doi.org/10.3390/metrology4040040

Article Metrics

Back to TopTop