PET Film Barrier: Metallising, OTR and WVTR Targets

Polyethylene Terephthalate (PET) film is manufactured by sequential biaxial orientation into thicknesses from 10 µm to 23 µm for vacuum web metallisation. Aluminium vapour from resistively heated intermetallic composite boats is condensed on the moving polyester web at chamber pressures between 1 × 10⁻³ mbar and 5 × 10⁻⁴ mbar; the resulting metal layer reduces oxygen transmission from an unmetallised benchmark of approximately 1.1 × 10² cm³/(m²·day·atm) at 23 °C and 0 % RH to ≤0.5 cm³/(m²·day·atm) when the aluminium optical density is held between 2.0 and 3.0. Water vapour transmission rate measured per ASTM F1249-20 at 38 °C and 90 % RH falls from roughly 3.0 × 10¹ g/(m²·day) to ≤1.0 g/(m²·day). For the specification class PET Film Barrier: Metallising, OTR and WVTR Targets, the final values are controlled by end use, lamination sequence, web surface defect density, and metallising process conditions rather than by the polyester base film alone.

What determines the OTR floor in vacuum-metallised PET?

The oxygen transmission rate of metallised PET is not determined by an oxygen-scavenging reaction; it is determined by the continuity of the vacuum-deposited aluminium layer. A conventional aluminium thickness of 30–50 nm is obtained at a transmission optical density of 2.0–3.0 on a 12 µm film. Below 2.0, aluminium growth is dominated by isolated island formation, and the oxygen flux through exposed PET can remain above 10 cm³/(m²·day·atm). At an optical density above 2.5, residual oxygen permeation is controlled primarily by pinholes, spatter particles, and stress-induced microcracks in the aluminium layer rather than by metal thickness. Dry snack laminates accept ≤0.5 cm³/(m²·day·atm); oxygen-sensitive beverage and pharmaceutical strip packs specify ≤0.1 cm³/(m²·day·atm), which requires an optical density above 2.5, substrate surface roughness Ra below 0.05 µm, and pinhole densities typically below 10 m⁻² in the metallised roll. Oxygen transmission is measured coulometrically according to ASTM D3985-17 after conditioning at 23 °C ± 2 °C and 0 % RH ± 2 % to eliminate moisture transient interference.

WVTR values are not uniform across a full metallised roll. On production-scale vacuum web coaters with widths from 1,000 mm to 2,450 mm, aluminium vapour is generated from a row of resistively heated boats; unless the boat power zones are individually regulated, the web edges receive a thinner deposit than the centre. This condition appears after winding as an edge-to-centre WVTR gradient, with edge samples failing a ≤1.0 g/(m²·day) requirement while centre samples pass. Closed-loop optical density scanners and movable edge masks reduce the gradient, but published data for the exact acceptable cross-web standard deviation is limited because each metalliser geometry has a different vapour plume shape. Converter specifications should therefore include both average OTR/WVTR and individual sample maxima across the web; specifying only a centre-panel value can mask edge-band failures on long runs.

Additional WVTR variation arises from aluminium spatter and from blocking during rewinding. If the freshly metallised film is wound with excessive tension, the brittle aluminium layer can craze under contact stress and create microcracks that raise OTR and WVTR after slitting. High-humidity WVTR testing itself is sensitive to defect location because water vapour transport through the exposed PET regions is orders of magnitude higher than through intact aluminium; a single pinhole can therefore shift the measured value above a strict limit even when the average optical density is acceptable.

When moisture values below 0.1 g/(m²·day) become necessary

When a package requires WVTR below 0.1 g/(m²·day) at 38 °C and 90 % RH, a single-pass aluminium-metallised PET layer on a 12 µm substrate rarely provides a sufficient safety margin because the measured water vapour transmission is defect-dominated. Moisture-sensitive pharmaceutical products and flexible organic electronic devices generally require either a thicker aluminium overlayer beyond an optical density of 3.0, an additional clear barrier layer such as vacuum-deposited aluminium oxide, or lamination to a second metallised film. When oxygen targets fall below 0.01 cm³/(m²·day·atm), aluminium-metallised PET alone is not an appropriate sole barrier solution; the pinholes and microcracks generated during converting and heat-sealing dominate the residual permeation. Published data for this specific configuration is limited because high-barrier laminates are often proprietary, but the general trend is that the aluminium layer moves the OTR and WVTR targets into the 10⁻¹ to 10⁰ range rather than into the 10⁻³ range required for foil replacement.

OTR and WVTR targets on the metallised PET itself are not sufficient to predict the barrier of the finished laminate. Solvent-based adhesives used for laminating metallised PET to polyethylene or cast polypropylene can retain ester or alcohol solvents; these residuals migrate into the PET and plasticise the amorphous regions, increasing WVTR by 0.2–0.5 g/(m²·day) compared with the as-made metallised film. Laminators therefore specify a two-step drying profile with final web temperatures in the 70–90 °C range and residual solvent limits below 5 mg/m² for high-barrier structures. Water-based or solventless adhesives eliminate some of this effect but require longer cure times before barrier testing, because premature testing can record transient moisture from uncured adhesive components.

Aluminium optical density alone does not guarantee barrier stability in flexed packaging

Flex durability is a critical barrier-stability parameter because the aluminium layer is a brittle inorganic coating on a ductile polyester core. The Gelbo flex procedure described in ASTM F392 applies repeated twisting and crushing; metallised PET can exhibit an increase in OTR from ≤0.5 cm³/(m²·day·atm) to above 5 cm³/(m²·day·atm) after 10–20 flex cycles when the aluminium develops through-thickness cracks. This failure mode is observed in stand-up pouches in which reverse-printed metallised PET is laminated to a polyethylene sealant; barrier loss is concentrated at crease lines and gusset folds. Converters that require flex-stable OTR targets select a higher aluminium thickness, reduce the flex severity at the package design stage, or replace the metallised layer with a transparent oxide barrier that has lower crack propagation. For metallised PET, the flex test is performed before and after barrier measurement, and the result is reported as a flexed OTR value, not as an as-made value.

Web handling, surface tension and aluminium nucleation

Adhesion between aluminium and biaxially oriented PET depends on surface polar groups and mechanical anchoring rather than on bulk chemical bond formation. Untreated polyester surfaces commonly exhibit a wetting tension below 40 mN/m; in-line plasma or corona treatment before metallisation raises this to above 44 mN/m measured by ISO 8296:2003. Chemically primed PET films use an adhesion-promoting coating that stabilises metal adhesion after heat-seal exposure and is specified when cross-hatch adhesion results must remain at class 0 or 1 according to ISO 2409:2013.

On a roll-to-roll evaporator, aluminium wire is fed into the boats at a rate matched to the target optical density. The boat temperature is maintained in the region of 1,300–1,450 °C; excessive temperature increases aluminium spatter, while insufficient temperature produces incomplete evaporation and uneven vapour distribution. The web speed is then adjusted within a typical production window of 4–12 m/s so that the required aluminium thickness is achieved in one pass. For 12 µm PET, unwind tension is held at approximately 100–200 N/m to maintain web tracking without stretching the film; tension excursions above this range can create cold-flow deformation and subsequent barrier loss at the reel edges. Thin-gauge films require proportionally lower tension because the cross-sectional area decreases directly with gauge.

Desiccant-free oral solid dose packaging tests the lower WVTR boundary. In this structure, metallised PET is laminated to a heat-sealable low-density polyethylene layer; the WVTR target is commonly set at ≤0.5 g/(m²·day) at 38 °C and 90 % RH, with an OTR target of ≤0.1 cm³/(m²·day·atm) at 23 °C and 0 % RH. The converter must verify that the seal layer does not create microchannels during polyethylene extrusion coating because extrusion temperatures in the 280–320 °C range can thermally stress the aluminium layer and raise both OTR and WVTR; moisture barrier acceptance is therefore measured on the final laminate, not on the metallised PET alone.

If the formulation requires WVTR below 0.1 g/(m²·day), the structure falls outside the reliability window of single-pass metallised PET and requires a higher-barrier replacement such as aluminium foil laminate or a multi-layer oxide-coated film.