Substrate glass surface roughness measurement quantifies the nanometer-scale texture that directly determines thin-film adhesion, defect density, and optical quality in display, semiconductor, and photovoltaic products. This technical guide explains the key roughness parameters (Ra, Rz, Sa, Sq), the standard test methods used to evaluate them—including stylus profilometry, atomic force microscopy, and white-light interferometry—and how accredited third-party laboratories perform substrate glass roughness determination in accordance with ISO and GB/T standards.
Technical Overview
Substrate glass is the foundational platform on which thin-film transistors, organic light-emitting layers, touch sensors, and functional coatings are deposited in modern display and semiconductor manufacturing. Because device layers on these substrates are often only tens to hundreds of nanometers thick, even sub-nanometer irregularities on the glass surface can propagate through the film stack, causing line breakage in metal wiring, non-uniform liquid crystal alignment, increased optical scattering, and degraded device yield. Surface roughness determination is therefore a critical quality control step for incoming substrate inspection, process qualification, and failure analysis.
Substrate glass is typically produced by the float process, the fusion draw process, or the slot draw process, each of which leaves a characteristic surface texture. High-quality alkali-free display glass commonly achieves an arithmetic mean roughness (Ra) below 0.5 nm, with premium fusion-drawn sheets reaching approximately 0.1 to 0.2 nm. In contrast, etched, textured, or chemically treated glass may exhibit roughness values ranging from several nanometers to several hundred nanometers. Accurate measurement across this wide range requires appropriately selected instruments, filtering conditions, and evaluation parameters as defined in international standards such as ISO 4287 and ISO 25178, together with the corresponding national standards GB/T 1031, GB/T 3505, and GB/T 10610.
Testing Samples
Display Glass Substrates
Alkali-free borosilicate and aluminosilicate glass sheets used as TFT-LCD and OLED panel substrates, supplied in standard mother-glass sizes or as cut coupons. Both as-formed surfaces and post-cleaned surfaces can be evaluated to verify incoming quality and cleaning effectiveness.
Cover Glass and Chemically Strengthened Glass
Aluminosilicate cover glass for touch panels and protective windows, including ion-exchanged and ground/polished variants, where roughness influences optical clarity, coating adhesion, and fingerprint resistance.
Semiconductor Glass Wafers and Carriers
Glass wafers, carrier substrates, and through-glass-via (TGV) interposer blanks used in advanced packaging, where surface texture affects bonding quality, lithographic focus, and film uniformity.
Photovoltaic and Coated Glass
Solar cover glass, superstrate glass with textured surfaces, and glass bearing transparent conductive oxide (TCO) or other functional coatings, where roughness is deliberately engineered for light trapping or controlled to preserve coating integrity.
Specialty and Processed Glass
Etched, matted, sandblasted, or laser-treated glass surfaces, as well as development-stage glass coupons submitted for research and process optimization studies.
Testing Items
Substrate glass surface roughness determination typically covers the following measurands and analytical outputs:
- Profile (2D) roughness parameters: Ra (arithmetic mean deviation), Rq (root-mean-square deviation), Rz (maximum height of the profile), Rt (total height), Rsk (skewness), and Rku (kurtosis), evaluated in accordance with ISO 4287.
- Areal (3D) surface texture parameters: Sa, Sq, Sz, Ssk, and related height parameters, evaluated in accordance with the ISO 25178 series.
- Power spectral density (PSD) analysis to characterize the spatial frequency distribution of surface irregularities, which is especially valuable for evaluating polishing quality and detecting periodic process marks.
- Waviness (Wa) separation using Gaussian filtering to distinguish long-wavelength form deviations from short-wavelength roughness.
- Uniformity mapping across the glass sheet, including center-to-edge comparison and statistical summaries (mean, standard deviation, maximum) of multiple measurement locations.
Testing Methods
Stylus Profilometry (Contact Method)
A diamond-tipped stylus is traversed across the glass surface while a sensor records vertical displacement, generating a high-resolution height profile. Measurement conditions follow ISO 3274 for instrument characteristics and ISO 4288 (with GB/T 10610 as the corresponding national standard) for sampling length and evaluation length selection. Typical settings include a stylus tip radius of 2 to 5 micrometers, a low measuring force below 1 mN to prevent surface damage, and cut-off lengths selected according to the expected roughness range. This method offers excellent vertical resolution and direct traceability but is limited to line profiles and requires care on delicate coatings.
Atomic Force Microscopy (AFM)
An atomically sharp probe scanned across the surface in contact or tapping mode provides true three-dimensional topography with sub-nanometer vertical resolution, making AFM the reference technique for the smoothest display-grade substrates. Scan areas are typically selected between 1 × 1 micrometer and 10 × 10 micrometers, and areal parameters are computed in accordance with ISO 25178. AFM results serve as the benchmark for validating measurements made by faster production-line instruments.
White-Light Scanning Interferometry
Optical profilers based on vertical scanning (VSI) or phase-shifting interferometry (PSI) non-destructively capture large areal topography maps within seconds, following the measurement methodology described in ISO 25178-6 series documents. PSI mode delivers sub-nanometer resolution on smooth glass, while VSI mode accommodates rougher etched or textured surfaces. The non-contact nature of this method eliminates the risk of scratching soft or thin coatings.
Confocal Optical Microscopy
Laser scanning or imaging confocal microscopes reconstruct surface height from optical sectioning, offering a useful combination of large field of view and good vertical resolution. This method is suitable for stepped surfaces, coated regions, and areas where both roughness and local geometry must be characterized, and it is standardized within the ISO 25178-6 framework for areal surface texture measurement.
Measurement Assurance
All methods are supported by rigorous measurement assurance practices: instruments are calibrated against certified reference specimens in accordance with ISO 5436-1 and ISO 5436-2, measurements are performed on vibration-isolated tables in temperature-controlled environments, and Gaussian filtering with documented cut-off values is applied consistently. Repeat measurements at multiple locations provide the statistical basis for acceptance decisions.
Applications
Substrate glass surface roughness data are used throughout the electronics and glass industries for a variety of purposes:
- Incoming quality inspection of display glass substrates at panel fabrication facilities, confirming that supplier surfaces meet specified Ra or Sa limits before thin-film deposition.
- Process qualification and optimization of polishing, cleaning, etching, and coating operations, where roughness trends reveal tool drift or chemistry changes.
- New product development for next-generation OLED, micro-LED, and semiconductor packaging substrates, where increasingly smooth surfaces are required for fine-pattern lithography and layer transfer processes.
- Touch panel and cover glass manufacturing, where surface texture affects anti-reflective coating performance and tactile quality.
- Photovoltaic glass engineering, balancing light-trapping texture against coating compatibility and encapsulant adhesion.
- Failure analysis and supplier comparison studies, supported by PSD characterization and detailed topographic maps in test reports issued by accredited third-party laboratories.
FAQ
What roughness level is typical for display substrate glass? High-quality alkali-free display glass normally exhibits Ra values below 0.5 nm, and premium fusion-drawn substrates can reach 0.1 to 0.2 nm, which is why AFM and phase-shifting interferometry are often required for reliable measurement at this scale.
Which method should be chosen for a given sample? For the smoothest as-formed or polished substrates, AFM or PSI interferometry provides the required sub-nanometer resolution; for etched or textured surfaces, stylus profilometry or VSI interferometry covers the larger amplitude range; and for coated or fragile surfaces, non-contact optical methods are preferred to avoid damage.
How are 2D and 3D parameters related? Ra and Sa are both arithmetic mean deviations, but Ra is derived from a single line profile while Sa is computed over a measured area. Areal parameters generally give a more representative picture of surface texture and are increasingly specified in modern drawings referencing ISO 25178.
What sample preparation is required? Samples should be cleaned with lint-free wipes and appropriate solvents or pressurized air to remove particles, and measurement areas must be free of fingerprints and dust. Coupons cut from mother glass should be deburred and handled with gloves or vacuum tools to preserve edge and surface integrity.
How is measurement traceability ensured? Instruments are calibrated with certified roughness reference specimens traceable to national metrology institutes, and measurement uncertainty budgets account for probe geometry, filter settings, environmental vibration, and temperature effects, giving clients confidence in reported values and pass/fail decisions.