KLA's Filmetrics Optical Reflectance Film Thickness Gauge
Online Seminar Highlights
Highlights
Introduction
Introduction
On November 19, the online seminar “Precision Measurement, Empowering the Future: Principles and Applications of KLA's Filmetrics Optical Reflectance Film Thickness Gauge” co-hosted by Unicorn and industry leader KLA Instruments concluded successfully. The event attracted numerous experts, scholars, and technical professionals from the semiconductor, advanced materials, and research institutions sectors. With 244 registrants and 182 participants fully engaged throughout, the event recorded 8,296 online interactions—concrete figures underscoring the industry's keen interest in precision thin-film measurement technology.
During this knowledge feast, we delved into the core principles of optical reflection technology and comprehensively explored its cutting-edge solutions for addressing critical challenges such as measurement consistency, complex application scenarios, and efficiency enhancement. This article compiles all Q&A covered in the webinar. To watch the live stream replay, click the button below or follow our official account for full access to the replay via articles or video channels.
Scan the QR code to follow the Unicorn official account, or click the button to view the webinar replay.。
Q
&
A
Answers
Why are there spectral reflectometers but no spectral ellipsometers for measuring film thickness? Also, can the four-probe resistance method for film thickness measurement be used if the top layer consists of two different metals?
Currently, KLA's ellipsometer solutions remain automated products supplied to fab customers, and desktop ellipsometers have not yet been incorporated into KLA's instrument division.
When measuring film thickness with a four-probe sheet resistance tester, if two metal films are present, the conductive paths between them cause the measured result to reflect a weighted average of their respective conductivities. To accurately measure a single surface film layer, it is recommended to introduce an insulating layer beneath the film to prevent signal interference.
Compared to ellipsometers or other film thickness measurement technologies, what are the core advantages of optical reflection film thickness gauges?
This is an excellent question you've raised, and it touches on a very core issue. Compared to other technologies like ellipsometers, the core advantages of optical reflectance film thickness gauges lie in their ease of operation, wide measurement range, and non-destructive measurement capability. It functions much like a point-and-shoot camera—extremely user-friendly. Operators typically only need to select the material to quickly obtain stable thickness readings. This makes it ideal for rapid quality control and routine inspections on production lines. In contrast, ellipsometers resemble professional DSLR cameras requiring meticulous calibration. While they can simultaneously capture thickness and refractive index data with higher precision, achieving optimal results demands precise parameter tuning and significantly higher operator expertise. In terms of measurement range, reflectometers also offer greater flexibility. Ellipsometers typically cover 1 nanometer to 10 micrometers, while reflectometers more effectively measure thicknesses from 10 nanometers to 1 millimeter, handling thicker films with ease. More importantly, it employs a non-contact optical method that causes zero sample damage. This preserves product integrity and enables seamless integration into production lines for real-time online monitoring—a capability unmatched by many techniques requiring sample preparation or contact, such as profilometers or scanning electron microscopes. Therefore, if your primary objective is rapid, non-destructive, and wide-range thickness monitoring rather than in-depth investigation of material optical constants, the optical reflectometer demonstrates irreplaceable core advantages in convenience, efficiency, and applicability.
Are there any requirements for the sample before measuring with a film thickness gauge? For example, if the surface is dirty or the sample has a certain degree of roughness, how much does this affect the results?
There will indeed be some impact, but the specifics require analysis. For surface contaminants, they can be considered an additional “film layer.” In practice, if the measured coating itself is thick—such as the approximately 10-micron anodized layer on a phone casing—then nanoscale fingerprints or oils adhering to it typically have negligible impact on results due to the vast thickness disparity. However, when measuring ultra-thin films around 100 nanometers, such minor contamination can cause significant interference. Regarding roughness, its effect is closely tied to the scale we observe: Many samples that appear macroscopically rough may, when measured using equipment capable of producing micro-spot sizes of tens of microns (e.g., the Filmetrics F40 series), reveal a relatively flat point within the minuscule area covered by the spot. This enables effective film thickness data acquisition for seemingly rough samples like cardiac stents or anodized layers on phone casings. The core principle of optical reflection film thickness gauges lies in detecting the optical reflection signals from the upper and lower interfaces of the coating layer. As long as an effective reflection signal can be formed at the measurement point, the measurement is feasible. Naturally, if the overall roughness of the sample is extremely high, preventing the formation of a stable reflection at any point, the measurement becomes difficult to perform.
I understand that elliptical polarizer thickness measurement equipment is currently in use at the fab level. Is optical reflector thickness measurement equipment also applied at the fab level? Thank you!
There are indeed applications. Within semiconductor fabrication facilities (Fabs), optical reflection-type film thickness gauges serve distinct and critical roles depending on the specific process stage. In front-end processes, such as at leading foundries like TSMC and SMIC, ultra-thin dielectric layers (e.g., silicon oxide or silicon nitride at around ten nanometers) approach or fall below the 30-nanometer minimum detection threshold recommended for optical reflectance systems. Consequently, higher-precision ellipsometers are undoubtedly the more suitable choice. However, for numerous layers ranging from hundreds of nanometers to micrometers in thickness, optical reflectance equipment demonstrates significant advantages. A typical application is measuring photoresist thickness. Some older KLA models, such as the OP series, incorporate an integrated design combining ellipsometer and optical reflectance modules. This approach better accommodates measurement requirements across different process segments within the fab. In back-end packaging and wafer-level manufacturing, optical reflectance equipment finds even broader application. This is because the layers in these processes are typically thicker—ranging from hundreds of nanometers for passivated silicon oxide and micrometer-scale polysilicon to auxiliary materials like grinding wax. These thicknesses fall squarely within the efficient and accurate measurement capabilities of optical reflectance systems.
What is the minimum thickness in nanometers that your optical reflection film thickness measurement equipment can detect? Can it measure multi-layer transparent films? What is the maximum number of layers it can measure?
Regarding the thinnest measurable limit, the equipment can indeed support measurements as low as 1 nanometer on paper. However, this is achievable only under extremely ideal conditions. At such scales, optical interference becomes extremely weak, making measurements entirely dependent on reflected signals. Any minor sample warping, surface contamination, or environmental interference can significantly impact results, leading to highly unstable measurements. Therefore, from the perspective of practical application and ensuring result reliability, we generally recommend starting measurements at thicknesses above 30 nanometers.
For multilayer film measurements, the answer is yes—the equipment possesses this capability. In principle, measuring multilayer films involves solving a system of optical equations—each additional layer introduces one more unknown to be solved for. We have actually measured complex coating systems with over ten layers, such as AR coatings. However, it is crucial to note that solving becomes increasingly difficult as the number of layers increases. This is primarily because a core distinction between optical reflectometer thickness gauges and ellipsometers lies in the fact that reflectometers typically do not directly measure refractive index. Instead, they rely on fixed refractive index values from an internal material library to calculate thickness. When dealing with excessive layers, especially in precision optical coatings where refractive index may vary, attempting to simultaneously solve for all layer thicknesses and refractive indices using only reflected spectral signals poses challenges like non-uniqueness of solutions. This makes it difficult to guarantee stable measurement accuracy. Therefore, while we have cases of measuring over ten layers, these represent exceptional circumstances. For routine industrial quality control, we most strongly recommend measuring coating structures with three layers or fewer for the most stable and reliable results. Examples include hard coating/penetration layers in the automotive lighting industry or silicon oxide/silicon nitride/silicon oxide (ONO) structures in semiconductors.
Earlier we mentioned the material library. We often test non-standard or new materials—what if the database lacks parameters? Does the instrument software support us in creating and fitting our own models? Is the process complicated?
For new or non-standard materials, our system offers highly flexible solutions. If no existing parameters are available in the material library, you can easily create your own. This process is straightforward: simply prepare the material's refractive index (n) and extinction coefficient (k) data at different wavelengths in an Excel spreadsheet according to our software's format requirements, then import it directly into the software's material library. During subsequent measurements, the imported data can be directly accessed to calculate thickness. A highly practical approach is to utilize an ellipsometer if available at your facility. Precisely measure the refractive index data of the new material using the ellipsometer, then import these measurements into our system. This ensures seamless data integration and efficient utilization.
Our software incorporates built-in refractive index fitting capabilities, primarily designed for common dielectric materials that conform to classical optical models such as Cauchy's model. These include silicon dioxide, silicon nitride, and numerous metal oxides/nitrides. For these materials, you can utilize the software's auxiliary refractive index fitting capability. However, it should be noted that the precision and accuracy of refractive index fitting fall short of those achieved with a professional ellipsometer. Therefore, we primarily recommend using this feature as a reference. When attempting to fit polymer materials using the Filmetrics model, the software's fitting accuracy becomes insufficient.
What is the minimum image size measurable with the F54? Are all F54 objectives optional? Is the same hardware compatible with objectives of different magnifications?
Theoretically, the F54 can measure down to 5 microns, but to ensure signal quality, measurements are generally recommended above 10 microns. The fundamental reason lies in the measurement's reliance on reflected light signals—smaller spot sizes result in weaker signal intensity and a corresponding decrease in signal-to-noise ratio. Additionally, the practical lower measurement limit is closely related to the substrate material: stable signals may require 20 microns or more on low-reflectivity glass substrates, 10 microns is typically sufficient on silicon wafers, while measurements down to approximately 7 microns are possible on highly reflective metal substrates. Regarding objective configurations, the F54 system offers multiple optional objectives to accommodate diverse requirements. Standard options include 5x, 10x, and 50x microscope objectives covering wavelengths from 400-850nm, suitable for measuring conventional films like silicon oxide and metal oxides. For ultrathin films or specialized applications, reflective objectives such as 10x and 15x models are also available. All objectives are designed on a common hardware platform, enabling seamless interchangeability and compatibility across different magnification levels. Users can flexibly select configurations based on requirements—for instance, employing low-magnification objectives for rapid positioning via wide-field image recognition, while utilizing high-magnification objectives for micro-area thickness measurements in narrow fields. Additionally, beyond standard models, the F54 series includes the F54-XY variant—an upgraded product launched after Filmetrics' acquisition by KLA. This model features enhanced automatic image recognition and positioning capabilities (Deskew and Pattern Rec), making it ideal for semiconductor chip-level applications. It automatically identifies pattern structures on wafers, enabling automatic chip positioning and measurement. This significantly improves the efficiency and accuracy of multi-point thickness detection on wafers with repetitive pattern structures, such as optoelectronic chips. It has been implemented in practical applications at companies including Silan and Medike.
Is the Cauchy model suitable for transparent films? Does it also apply to light reflection in opaque films?
This is an excellent question that touches upon a core concept in optical measurement. First, we must understand “transparent” and “opaque” from an optical perspective rather than visual perception. Our film thickness gauges typically operate across a wavelength range of 190-1700 nanometers, far broader than the 400-780 nm range visible to the human eye. Consequently, many materials that appear opaque under visible light may exhibit transparency across a wider spectral range. A classic example is silicon wafers, which are completely opaque in visible light but become transparent in the near-infrared band. This transparency allows light to pass through and reflect off the underlying interface, enabling measurement by reflective film thickness gauges. Thus, the key to measurement isn't whether a material “appears” transparent, but whether it permits light to penetrate the film itself within the operational wavelength range, thereby capturing the reflected signal from the bottom interface. As long as this condition is met, even materials that appear opaque to the naked eye are equally suitable. Conversely, if a material is completely opaque across the entire spectrum (such as certain thick metals), preventing light from penetrating to the film's base, then the reflectance method cannot measure the thickness of films deposited on it.
We have a rather unique requirement: we need to measure film thickness on a curved surface. Does our equipment have a special measurement mode or accessory for this purpose?
Measuring the thickness of surface coatings on curved surfaces presents a challenging application. The specific approach depends on the curvature of the curved surface. For example, in the automotive lighting industry, hardened coatings are typically applied to various lamp surfaces, which themselves possess a degree of curvature. In such cases, a contact probe can effectively address signal acquisition challenges. As long as sufficient valid signals can be collected, the specific curved surface measurement requirements can be met. Additionally, the F40's micro-area measurement mode can also partially resolve application issues for curved surface coatings.
We often need to test dozens of samples from the same batch to evaluate uniformity. Does your instrument come equipped with an automatic sample stage? How complex is it to set up an automated measurement process?
When measuring multiple samples from the same batch to evaluate coating uniformity, the F50 series coating thickness gauge is an excellent choice. Measurement results for individual samples directly display a schematic diagram of the surface coating thickness, along with maximum value, minimum value, average value, standard deviation, thickness range, uniformity, and other parameters.
Creating an automated measurement recipe is not a particularly complex process. Typically, our engineers provide fully developed measurement recipes directly during application support. Subsequent measurements only require selecting the corresponding recipe and clicking “Start Measurement” to obtain results, including uniformity.
Can epitaxial polysilicon on silicon wafers be measured? Is the measurement of silicon wafers possible because it utilizes infrared light?
Certainly, and it's a very common application. In the silicon wafer manufacturing process at the semiconductor front-end, Poly-Si measurement is a widely demanded process node.
Correct. Silicon wafers are opaque to visible light (400–850 nm) but transparent to infrared light (850–1700 nm), meaning infrared light can penetrate the surface silicon film layer.
We are currently working on flexible OLEDs, which use soft substrates like PET or PI that aren't perfectly flat. Under these conditions, can our equipment still maintain high-precision measurements? Are there any special fixtures or measurement techniques?
We have corresponding fixtures available. The flexible, uneven substrates you mentioned are indeed a common challenge in optical measurement. For such samples, we not only have tailored solutions but also numerous successful case studies within the industry. The core issue lies in how to flatten delicate, flexible samples. Reflectance measurement requires the optical path to be perpendicular to the sample surface. To achieve this, we recommend using specialized fixtures to securely clamp and flatten the flexible substrate, artificially creating a localized ideal measurement plane. This approach is complemented by the Filmetrics F40 series equipment mentioned earlier by Mr. Li. Utilize its micro-spot to precisely target relatively flat areas at the micrometer level for testing. Then, employ multi-point measurements to derive statistical analysis values, ensuring the representativeness and reliability of the results. Therefore, while flexible substrates present unique challenges, achieving high-precision measurements is entirely feasible through this integrated approach of “custom fixtures,” “micro-spot precision targeting,” and “multi-point statistical analysis.”
How well does the optical film thickness gauge adapt to environmental conditions? For example, could factors like environmental vibrations or lighting affect measurement accuracy?
The optical film thickness gauge demonstrates excellent adaptability to environmental factors. Regarding ambient light, the instrument performs background baseline acquisition prior to measurement. Through specialized signal processing algorithms, it effectively identifies and eliminates interference from environmental light sources such as everyday illumination, enabling normal operation without requiring darkroom conditions. In terms of vibration resistance, routine environmental vibrations have minimal impact on the accuracy of individual measurements. Only during prolonged continuous measurements might they cause slight fluctuations in repeatability statistics. Of course, severe vibrations during measurement (such as a violent impact on the workbench) will still interfere with the results. Overall, this device maintains stable operation in most laboratory and industrial environments. In special conditions with significant continuous vibration, configuring a professional anti-vibration table serves as a reliable solution to further enhance measurement stability.
For medical devices like cardiac stents with complex three-dimensional mesh structures, how does your optical reflectance film thickness gauge ensure precise coating thickness measurements across every minute area of the mesh?
For complex structures like cardiac stents, we typically employ Filmetrics' F40 series. Its extremely small measurement spot, combined with our specially developed positioning fixtures, enables precise measurements. Equipped with a microscopic spot as small as 5 microns, the F40 series can accurately target and independently measure each minute area within the stent's mesh. This effectively prevents data distortion caused by oversized spots overlapping multiple structures simultaneously. Building on this foundation, our specially developed jigs for irregularly shaped devices ensure stable sample clamping. They assist operators in rapidly and precisely positioning the measurement spot at each designated critical location, enabling efficient, controlled manual precision measurement. This approach systematically evaluates coating uniformity across three-dimensional structures. For applications like this, contact us to discuss detailed solutions with our technical experts.
Are these displayed images all actual measurement cases? Can you provide sample testing services?
Yes, all displayed images are actual cases measured using our Filmetrics equipment. The application scope of optical reflectance film thickness gauges is extremely broad. Beyond their established use in the semiconductor industry, they can be employed in numerous specialized scenarios—such as real-time measurement of water droplet thickness changes in air, air gap thickness analysis, and even coating thickness measurement on luxury leather goods. We are pleased to offer sample testing support. You may choose to send samples to our laboratory, where we will arrange professional equipment for testing and provide detailed reports. Alternatively, you are welcome to visit our lab for on-site technical discussions and hands-on testing experiences. For any specific requirements, please feel free to contact us for further discussion.
總結(jié)
As demonstrated in this Q&A session, while the challenges of thin-film measurement vary, the pursuit of precision, efficiency, and reliability remains consistent. Whether your questions concern measurement repeatability and consistency, the analysis of complex new materials and structures, or the integration of equipment into automated production lines, they all require solid technical expertise and extensive application experience as their foundation.
If the above discussion has sparked further thought or you are facing similar measurement challenges, our team of application specialists stands ready to assist you. We are committed to transforming your specific requirements into reliable measurement solutions.
We welcome you to initiate one-on-one technical discussions by clicking the button below or contacting us via the provided address and phone number:
Official Account
Service Support
