Saflex DG

Saflex DG is a plasticised polyvinyl butyral (PVB) interlayer supplied in moisture-barrier-wrapped rolls for structural laminated safety glass. The model designation is Saflex DG; the grade is positioned between standard architectural PVB and stiffer ionoplast interlayers where higher post-glass-breakage retention, controlled flow during autoclaving, and conventional PVB line economics are specified. The interlayer is produced in nominal thicknesses of 0.76 mm and 1.52 mm. The 1.52 mm gauge may be laid up as a single sheet or as two plies of 0.76 mm on wide architectural lines. Roll stock is conditioned at 15–25 °C and 20–35 % RH before slitting; moisture content at the pre-lamination stage is held between 0.40 wt% and 0.50 wt%. Interlayer below 0.40 wt% can exhibit reduced glass adhesion, while interlayer above 0.50 wt% can generate steam bubbles during autoclave heating. If the lay-up room remains above 60 % RH, pre-drying in a desiccant chamber at 25 °C and 10 % RH for 24 h is required. Rolls should be used within 8 h of bag opening or resealed in moisture-barrier packaging.

On a production line with a four-station architectural washer, nip-roller deairing, and a 9.6 m working-length autoclave, cold rolls pulled directly from storage at 10 °C produce worm-shaped air pockets between glass and interlayer. The defect is controlled by conditioning rolls for 48 h at 18 °C and 25 % RH, then completing lay-up within the same shift. This is standard PVB handling but is applied more strictly for Saflex DG because structural laminates cannot accept the edge clouding that may be tolerated in decorative glazing. Short-term transport below 0 °C does not damage the film, but 48 h reconditioning at 20 °C is required before opening to prevent condensation at the roll face. The first 2 m of sheet from an opened roll is typically discarded because of outer-lap moisture pickup.

What autoclave pressure–temperature relation suppresses steam bubble formation in Saflex DG laminates?

Autoclave cure for Saflex DG follows the PVB schedule of 120–140 °C and 10–14 bar for 90–150 min, with exact dwell set by glass thickness, laminate area, and load density. The critical control point is that the interlayer reaches 110 °C while hydraulic pressure remains above 8 bar; otherwise steam bubble re-nucleation occurs at the glass–interlayer interface. A production autoclave with a working diameter of 2.8 m and a charge of 24 panels of 2.0 m × 3.0 m exhibits a centre-to-edge temperature lag of 20–30 K, requiring a 20 min plateau extension when the load includes asymmetric glass stacks or protective interleaving paper. Published data for the exact thermal lag in mixed-thickness charges is limited; therefore each lamination line should validate a probe laminate placed at the centre of the autoclave charge before serial production.

The pre-lamination glass surface temperature window is 70 ± 5 °C. Below 65 °C, nip-roller deairing leaves air pockets; above 75 °C, edge starvation occurs. On a line running 4.8 m × 2.4 m panels at 2.0 m/min, the edge-to-centre glass surface temperature can differ by 8 °C. Glass that is centre-hot may adhere prematurely to the interlayer and wrinkle. The operator maintains the temperature band by adjusting the convection tunnel setpoint in 3 °C increments based on an IR pyrometer reading at the nip exit. This narrow processing window is the principal production conflict for structural PVB laminates, and deviation beyond ±5 °C produces a visible edge defect that is not acceptable under EN ISO 12543-2 visual quality requirements.

In an exterior balustrade of 8 mm fully tempered glass /Saflex DG /8 mm fully tempered glass, the interlayer transfers shear across broken glass fragments after impact and reduces fall-out. The assembly is verified under EN 12600 class 2B2 and EN 14449. On point-fixed systems, drilled glass holes introduce stress concentrations that the interlayer does not eliminate; the interlayer redistributes load across the fractured fragment, but the metal fixing remains structurally dominant. The governing service limit is edge shear stress rather than centre deflection. At panel temperatures above 50 °C, PVB shear stiffness declines, and edge delamination may initiate if the exposed interlayer edge is subject to cyclic thermal expansion or high humidity. Face-fixed metal channels or silicone edge sealing are therefore used on high-rise balustrades.

Edge stability and sustained-load thresholds at elevated service temperature

The creep response of Saflex DG under sustained shear is measured by dynamic mechanical analysis at 1 Hz and by tensile creep tests following ISO 527-1. Representative plasticised PVB storage modulus is 1–8 MPa at 20 °C, falling below 0.5 MPa above 65 °C. For laminated glass carrying permanent structural load, the interlayer service limit is normally 30–45 °C; above this range, mechanical fixings bypassing the interlayer are required. In a laminated glass stair tread with 12 mm glass faces, post-breakage retention is evaluated at 20 °C and at 50 °C; published data for this specific configuration is limited, so full-scale staged-load testing is mandatory before design release.

Compared with standard clear PVB, Saflex DG exhibits higher flow resistance at nip-roll temperatures and less edge thinning during autoclave pressure, which reduces edge pinch-off in staggered-edge laminates. Compared with ionoplast interlayers, it has lower modulus and higher creep at elevated temperature; ionoplast sheet stiffness at 20 °C is generally two orders of magnitude higher than plasticised PVB. Saflex DG is specified where process economics and optical quality of PVB are required and where structural service temperature does not exceed 45 °C under permanent load. Unlike EVA interlayers, Saflex DG requires autoclave pressure and temperature to reach full adhesion; unlike polycarbonate interlayers, it does not require a dedicated polycarbonate lamination line but is not compatible with direct polycarbonate glazing without an edge seal

Saflex DG is not recommended for laminating glass directly to polycarbonate without a polycarbonate-compatible edge seal, because PVB plasticiser migration can stress-crack polycarbonate around drilled holes. The interlayer also requires protection from alkaline cementitious water: exposed edges embedded in wet concrete or grout without a sealant can lose adhesion through plasticiser hydrolysis. Cleaning solvents containing ketones, amines, or acrylate monomers should not contact the exposed interlayer edge. Do not combine the interlayer edge with one-part amine-cured sealants unless adhesion testing on laminated coupons shows no interfacial haze or delamination; some amine catalysts can accelerate local PVB plasticiser oxidation.

When edge starvation appears in a nip-roller pre-lamination line

Edge starvation occurs when interlayer flow under nip pressure reduces edge thickness below 90 % of nominal. On a three-roller pre-lamination line with nip roller durometer 65 Shore A, the defect is observed when glass surface temperature exceeds 80 °C. The resulting low-adhesion band appears as a clear edge line after autoclave. Correction involves reducing nip pressure, lowering pre-lamination temperature to 60–70 °C, or using the heavier 1.52 mm Saflex DG sheet. The higher flow resistance of Saflex DG relative to standard clear PVB reduces edge starvation but does not eliminate it when glass temperature is outside the control window.

Adhesion of Saflex DG to clean float glass is evaluated by the pummel test after conditioning; product-specific limits are set by the glass processor and verified against line-specific standards. Boil testing of laminated coupons in boiling water for 2 h should show no bubbles, no delamination, and no edge defect extending beyond 5 mm from the cut edge. Haze and luminous transmittance of the finished laminate are evaluated under ASTM D1003; a 6 mm low-iron glass /Saflex DG /6 mm low-iron glass build typically achieves luminous transmittance above 87 % when the interlayer is dry and bubble-free. Gel count and particulate defects are inspected on a light box according to EN ISO 12543-2; exact acceptance limits for the DG grade are supplier-controlled.

On a vacuum bag line, the pre-autoclave cycle consists of cold vacuum for 20 min at 20 °C, followed by heating to 90 °C over 45 min while maintaining bag vacuum. Transfer to the autoclave occurs only when the bagged laminate shows no visible air channels. A production run of 200 m² per shift typically loses 1–2 % of interlayer consumption to roll-end moisture discard and edge trim. Rolls are shipped in sealed aluminium-lined foil bags with desiccant, and roll length depends on thickness and slit width; exact packaging details are provided in the order-specific packing list rather than as a fixed catalogue value.

Conformance is documented under EN ISO 12543-2 for laminated glass and laminated safety glass interlayer properties, EN 14449 for impact performance, ANSI Z97.1 for safety glazing materials, and ASTM C1172 for laminated architectural flat glass. Wind load resistance is verified by ASTM E330. High-speed impact compliance for the finished laminate is tested under CPSC 16 CFR 1201. The interlayer itself is not a standalone safety glazing material; compliance attaches to the laminated assembly after glass selection, lay-up, and autoclave cure.

The rheological difference between Saflex DG and standard clear PVB is observed as an increase in plateau modulus between nip-roll and autoclave temperatures. This produces two operational consequences: less interlayer squeeze-out at the edge and higher residual shear stress at the glass–interlayer interface during cooling. Residual stress is measured by photoelastic edge inspection; dense fringe orders near the edge indicate constrained shrinkage. A slow cool above 40 °C at 0.3 K/min is applied to reduce residual stress in thick laminates. In an installed structural glass fin, the exposed edge is inspected under polarised light after 12 months; any propagating delamination beyond 5 mm from the cut edge requires edge sealing or replacement.

Thermal cycling of laminated balcony balustrades between −10 °C and 60 °C creates differential expansion between glass and PVB. The edge shear stress generated by this cycling is highest at the laminate corner, which is also the location of lowest autoclave pressure in some load configurations. In a post-installation audit of overhead glazing with unprotected PVB edges, moisture-driven edge clouding appears first at the lower corner of the panel where water collection occurs. Silicone edge sealing reduces moisture ingress but cannot compensate for autoclave undercure or excessive pre-lamination moisture. The service limit is therefore governed by the combined temperature–humidity history at the exposed edge, not by the centre-of-panel optical clarity alone.