Substrate glass is the foundational material for flat-panel displays and emerging glass-core semiconductor packaging, and its phase transition temperature directly defines the thermal budget of every downstream manufacturing process. Accurate determination of the glass transition temperature (Tg) and the related transformation-range points—strain point, annealing point, and softening point—is essential to guarantee dimensional stability, minimize thermal compaction, and prevent deformation during device fabrication. This article presents the measurement principles, applicable international standards, sample requirements, and practical applications of substrate glass phase transition temperature testing.
Technical Overview
Substrate glass is an amorphous material that does not exhibit a sharp first-order melting transition like crystalline solids. Instead, when heated, it passes through a reversible transformation range in which the rigid, glassy state gradually converts into a viscoelastic, undercooled liquid state. The reference point at the midpoint of this transformation is commonly designated the glass transition temperature (Tg), while the lower and upper boundaries of the range are functionally defined by the strain point and the softening point.
Because the glass transition is kinetic rather than a true thermodynamic phase change, the measured transition temperature depends on the thermal history of the specimen, the heating or cooling rate, and the physical property being monitored (heat capacity, thermal expansion, or viscosity). For substrate glass—typically alkali-free aluminosilicate compositions produced by the fusion draw, float, or slot-draw process—these temperatures typically fall between approximately 550 °C and 850 °C, a range that overlaps directly with high-temperature display processes such as poly-silicon crystallization, thin-film encapsulation, sealing, and annealing. Precise knowledge of the transformation range therefore determines whether a given glass grade can survive a process without warping, compaction, or irreversible dimensional drift.
Key temperature points routinely evaluated in substrate glass characterization include:
- Glass transition temperature (Tg): the midpoint of the transformation range, determined by calorimetric, dilatometric, or thermomechanical methods.
- Strain point: the temperature at which viscosity reaches approximately 1014.5 dPa·s; below this point, internal stresses relax only over impractically long times.
- Annealing point: the temperature at which viscosity reaches approximately 1013 dPa·s; internal stresses are relieved within minutes, making this the reference temperature for stress-relief annealing.
- Softening point (Littleton point): the temperature corresponding to a viscosity of approximately 107.6 dPa·s, marking the onset of visible viscous deformation under its own weight.
Testing Samples
Alkali-Free Display Substrate Glass
High-generation TFT-LCD and OLED substrate glass, produced by fusion draw or float processes in generations up to Gen 10.5 and beyond. These aluminosilicate compositions contain virtually no alkali oxides to prevent mobile-ion contamination of transistor arrays, and their transformation range must safely exceed all panel fabrication temperatures. Specimens are typically cut from full-size mother glass into coupons or coupons of suitable dimensions for the selected analytical technique.
Glass-Core Substrates for Semiconductor Advanced Packaging
Glass wafers and panels with through-glass vias (TGV), intended to replace organic build-up substrates and silicon interposers in high-density packaging. Because these substrates experience repeated thermal cycling during deposition, via filling, lamination, and reflow, the phase transition temperature and the thermal expansion behavior around the transformation range are critical acceptance criteria.
Cover Glass and Chemically Strengthened Glass
Sodium-containing aluminosilicate cover glass for touch devices undergoes ion-exchange strengthening; while the compressive layer is formed well below the transformation range, accurate Tg data are required to define safe strengthening, cutting, and bending windows.
Specialty Borosilicate and High-Silica Glass
Borosilicate and high-silica glasses used in hermetic packaging, optical components, and high-temperature carrier applications, where a lower or higher transformation range must be verified against specification limits.
Testing Items
Typical test items reported in a substrate glass phase transition temperature determination include:
- Glass transition temperature (Tg): reported as onset, midpoint (half-height or inflection), and end-set values from the transition step.
- Strain point and annealing point: determined from fiber elongation or beam-bending viscosity measurements or estimated from dilatometric data.
- Softening point: determined by fiber elongation (Littleton method) or calculated from viscosity–temperature curves.
- Mean coefficient of thermal expansion (CTE) below and above Tg, together with the differential expansion step at the transformation range.
- Dilatometric softening temperature and the temperature of maximum expansion-rate change.
- Thermal compaction (irreversible shrinkage) behavior, often evaluated by repeated cycling to temperatures approaching the transformation range.
- Step height in heat capacity (ΔCp) across the transition, where the calorimetric signal is sufficiently pronounced.
Testing Methods
Differential Scanning Calorimetry (DSC)
DSC detects the glass transition as a step-like change in heat capacity. Measurements are performed in accordance with ISO 11357-2 or ASTM E1356 using specimens of roughly 10–30 mg, a controlled heating rate of 10–20 K/min, and a nitrogen or dry-air purge. Because substrate glasses are highly durable, low-alkali silicate systems with a comparatively weak calorimetric step, high-sensitivity instruments, repeatability checks, and slow scanning rates are often required. The midpoint of the transition step is reported as Tg.
Differential Thermal Analysis (DTA)
DTA monitors the temperature difference between the specimen and an inert reference and can locate the transformation range under the same general conditions covered by ASTM E1356. It is frequently used for screening and for detecting concurrent crystallization or devitrification events that appear as exothermic peaks above Tg.
Thermomechanical Analysis (TMA)
TMA tracks the expansion of a glass specimen with a low-force quartz probe and identifies Tg from the kink in the expansion curve where the CTE abruptly increases. The method follows ASTM E1545 or ISO 11359-2 and is well suited to substrate glass because the dilatometric signal is pronounced even when the calorimetric step is weak. TMA also yields the CTE on both sides of the transition and an estimate of the dilatometric softening temperature.
Push-Rod Dilatometry
Push-rod dilatometry per ASTM E228 provides high-accuracy expansion curves on larger specimens, from which the transformation range, expansion coefficients, and dilatometric softening point are extracted. It is the reference technique for CTE-related acceptance testing of substrate glass grades.
Fiber Elongation and Beam-Bending Viscometry
The viscosity-defined reference points are measured by fiber elongation (ASTM C336 for annealing and strain points; ASTM C338 for the Littleton softening point; ISO 7884 series for viscosity by fiber drawing) or by beam-bending viscometry (ASTM C598). These methods anchor the transformation range to defined viscosity values, which is essential for process simulation of sag, compaction, and forming operations.
Regardless of the technique, accredited third-party laboratories operate under an ISO/IEC 17025 quality system, apply standardized heating rates and atmospheres, condition specimens to a common thermal history before measurement, and report results together with the method, scanning rate, and evaluation convention so that data from different sources remain comparable.
Applications
Reliable phase transition temperature data for substrate glass support a wide range of industrial decisions:
- Display manufacturing: defining the maximum allowable temperature for TFT deposition, crystallization, encapsulation, and laser annealing while keeping thermal compaction of the mother glass within dimensional tolerances.
- Advanced semiconductor packaging: selecting glass-core substrate grades whose transformation range safely exceeds solder reflow and deposition temperatures, and validating warpage budgets for panel-level packaging.
- Sealing and assembly engineering: matching the softening behavior of substrate glass with frit seals, solder glasses, and low-melting sealing glasses to achieve hermetic joints without substrate distortion.
- Annealing and stress-management schedules: setting hold temperatures relative to the annealing and strain points to relieve residual stresses introduced by cutting, grinding, and thermal cycling.
- Incoming quality control: verifying that production lots of substrate glass conform to the transformation-range specification agreed with the supplier, using standardized methods for lot acceptance.
- Material development: comparing candidate glass compositions during formulation and melting trials, and generating viscosity–temperature input data for forming-process simulation.
FAQ
What is the typical phase transition temperature of display substrate glass?
Alkali-free display substrates generally show a Tg in the region of 600–700 °C, with strain points around 600–670 °C and softening points near 730–850 °C depending on the exact aluminosilicate formulation. Exact values must always be confirmed by testing, as they vary between manufacturers and glass grades.
Why do DSC and TMA sometimes give different Tg values for the same glass?
The glass transition is rate- and property-dependent: DSC senses the change in heat capacity, while TMA and dilatometry sense the change in thermal expansion. Because the two properties shift across the transformation range at slightly different rates, and because results depend on heating rate and thermal history, small systematic differences between methods are normal. Consistency is achieved by always stating the method and evaluation convention.
How does heating rate affect the result?
Faster heating rates shift the apparent Tg to higher temperatures, while slower rates or prior thermal conditioning shift it lower. Standardized rates of 10–20 K/min for DSC and 3–5 K/min for dilatometry are therefore specified and reported with the result.
What sample forms are required for testing?
DSC requires small fragments or discs of approximately 10–30 mg; TMA and dilatometry require flat, parallel-ground coupons typically a few millimeters thick; fiber elongation requires drawn fibers of specified diameter. All specimens should be defect-free, cleaned, and thermally conditioned before measurement.
How is Tg different from the strain, annealing, and softening points?
Tg is a single reference temperature marking the midpoint of the transformation range, whereas the strain, annealing, and softening points are engineering reference temperatures tied to defined viscosities. Together they describe the complete thermal window within which the substrate glass changes from a rigid solid to a deformable viscous body.
Which standards govern substrate glass phase transition testing?
The principal standards are ISO 11357-2 and ASTM E1356 (DSC), ASTM E1545 and ISO 11359-2 (TMA), ASTM E228 (dilatometry), ASTM C336, ASTM C338, ASTM C598, and the ISO 7884 series (viscosity-defined points), with laboratory competence typically demonstrated under ISO/IEC 17025 accreditation.