Niska dyfuzja tlenu w warstwach barierowych EVOH dla rur PERT

Oxygen ingress through the pipe wall of a hydronic heating circuit causes corrosion of cast iron and steel components, pump cavitation, and sludge formation in closed loops. Polyethylene of raised temperature resistance (PE-RT) is specified for hot- and cold-water installations under ISO 22391-2, and the multilayer pipe itself is classified under ISO 21003-2 when an EVOH barrier layer is included. Unmodified PE-RT exhibits an oxygen transmission coefficient in the range 10²–10³ cm³·mm/(m²·day·atm) when measured at 23°C and 0% RH by the coulometric method of ASTM D3985-17. To meet the barrier requirement of DIN 4726, which limits oxygen permeation in surface heating pipes to 0.1 g/(m³·d) at 40°C water temperature, a five-layer construction is commonly extruded: PE-RT/tie resin/EVOH/tie resin/PE-RT. The EVOH layer, a vinyl alcohol copolymer with ethylene content between 27 mol% and 44 mol%, reduces oxygen transport by approximately three to four orders of magnitude under dry conditions because the high density of hydroxyl groups forms interchain hydrogen bonds that reduce free volume and oxygen solubility. However, the same hydroxyl groups absorb water from the service environment, and the resulting plasticization raises both oxygen diffusivity and oxygen solubility. This trade-off imposes constraints on layer placement, ethylene content selection, and barrier layer thickness that are not resolved by static oxygen transmission data alone.

How does relative humidity reverse the oxygen barrier of an EVOH interlayer?

Oxygen transport in EVOH follows a solution-diffusion mechanism in which the permeability coefficient is the product of the diffusion coefficient and the solubility coefficient. At 20°C and 0% RH, a 38 mol% ethylene EVOH grade may show an oxygen transmission rate of 0.4–0.8 cm³·20 µm/(m²·day·atm) under ASTM D3985-17. At 85% RH, published industrial datasheet values for the same grade typically rise to 10–20 cm³·20 µm/(m²·day·atm), representing a 15- to 50-fold loss of barrier. The increase is nonlinear: below 30% RH the effect is modest, while above 65% RH water clusters begin to disrupt hydrogen bonding and the activated diffusion process becomes concentration-dependent. In a PE-RT pipe, the EVOH layer is not exposed directly to liquid water unless the inner PE-RT layer is breached, but water vapour from the heated aqueous phase transmits through the inner PE-RT layer and condenses at the tie-layer interface. At service temperatures of 60–70°C, the equilibrium moisture uptake in EVOH is lower than at room temperature because water sorption is exothermic, yet the higher thermal energy increases diffusional jump frequency. Published studies report that the apparent activation energy for oxygen permeation in dry EVOH is in the range 35–45 kJ/mol, while in water-plasticized EVOH the effective activation energy may fall to 20–30 kJ/mol. A barrier layer that passes the 40°C oxygen diffusion test of DIN 4726 as a dry extruded pipe may therefore allow excessive oxygen ingress after long-term hydrothermal ageing once the EVOH reaches equilibrium moisture content. The technical response is to evaluate oxygen transmission after conditioning at 80°C water for 1000 h or after saturation in 90% RH air for 30 days, because the lower ethylene grades that are most effective at 0% RH also exhibit the greatest relative humidity sensitivity.

Before specifying an EVOH layer position, the moisture pathway through the pipe wall must be quantified. An outer EVOH layer sees ambient humidity directly, while an interlayer buried between two PE-RT layers sees water vapour that has first permeated through the inner PE-RT layer. The tie resin used to bond EVOH to PE-RT is typically a maleic anhydride-grafted linear low-density polyethylene or PE-RT carrier with a graft level between 0.5 wt% and 1.5 wt% maleic anhydride; the anhydride ring opens and reacts with hydroxyl groups on EVOH at the melt interface to form ester or acid linkages depending on processing conditions. Adhesion measured on coextruded pipe according to ISO 17454 or client-specific peel protocols generally exceeds 25 N/25 mm for proper tie-layer thicknesses of 0.05–0.10 mm. When tie-layer thickness falls below 0.03 mm, the low-viscosity grafted layer may channel or break during spiral mandrel distribution, leading to intermittent delamination that is not visible on the pipe surface but appears as blisters after pressure cycling at 95°C. The inner PE-RT layer also serves as a moisture barrier to reduce the rate of water vapour reaching the EVOH; a thicker inner layer delays moisture plasticization but increases the diffusion path for oxygen already dissolved in the pipe wall. Published data for this specific configuration is limited, but industrial coextrusion experience indicates that an inner PE-RT layer of at least 0.3–0.5 mm is required to avoid rapid EVOH saturation during initial hot-water operation.

Coextrusion melt rheology, thermal degradation thresholds, and thickness control

EVOH is a shear-sensitive polyol with a narrower processing window than PE-RT. Commercial EVOH for pipe extrusion is melt-blended with a small amount of lubricant and acid scavenger, but melt temperatures above 240°C initiate dehydration of vicinal hydroxyl groups, acetic acid evolution, and crosslinking to form gels. The practical melt-temperature window for continuous coextrusion is therefore 190–230°C for 32–38 mol% ethylene grades, a window of only 40°C and effectively ±5°C around the target when die-head temperature fluctuations and shear heating are included. Extruder screw design for the EVOH stream should use a low-compression barrier screw with an L/D ratio of 25:1 to 30:1, compression ratio 2.5:1 to 3.0:1, and no Maddock mixing sections with tight clearances; spiral mandrel dies should limit shear rate in the EVOH layer to below 300 s⁻¹ to avoid viscous heating and shark-skin at the melt interface. The melt flow rate of EVOH grades measured under ISO 1133-1:2022 at 190°C and 2.16 kg ranges from 1.5 g/10 min for 27 mol% ethylene to 6–14 g/10 min for 44 mol% ethylene; PE-RT pipe grades typically exhibit an MFR of 0.5–1.0 g/10 min under 190°C/2.16 kg, so a substantial viscosity mismatch exists. Tie resins are formulated with lower viscosity than both PE-RT and EVOH to promote interfacial wetting, but excessive tie-layer flow can cause layer thickness variation. Coextrusion feedblocks used for pipe are usually ring-type distributors with five or seven spiral channels; the EVOH layer thickness specified by the pipe manufacturer is commonly 0.05–0.15 mm for outside diameters from 16 mm to 32 mm, corresponding to 4–8% of total wall thickness. Direct inline measurement of EVOH layer thickness is performed with ultrasonic or terahertz gauge systems with a resolution of 5 µm, because post-extrusion destructive microscopy cannot correct drift in real time. When layer thickness varies by more than ±10%, localized thin spots become oxygen diffusion short circuits, and the effective oxygen permeation of the pipe increases disproportionately because permeation is thickness-normalized but real pipes fail at the minimum thickness point.

On production-scale lines, thermal degradation of EVOH produces hard black specks that can bridge die gaps and cause circular scoring on the EVOH layer. The first sign is a rise in die pressure upstream of the barrier layer by more than 10 bar over 8 h of continuous operation, followed by an increase in optical haze of the EVOH layer when viewed through a peeled outer PE-RT layer. Because degraded EVOH releases acetic acid, purge procedures use low-melt-index LDPE or PE-RT with a melt temperature of 210°C for 20–30 min before shutdown; purging with polyamide or polyacetal is incompatible and must be avoided. EVOH pellets must be dried to below 0.05 wt% moisture using a desiccant dryer with a dew point of -30°C or lower; undried pellets release steam at the melt, producing microbubbles in the barrier layer that reduce effective thickness and increase oxygen transmission. On continuous lines, batch-to-batch variation in EVOH water content is a more common cause of intermittent oxygen permeation failures than variation in pipe wall thickness. The dryer hopper should be sized for 4–8 h residence time at 80°C; drying above 90°C can cause pellets to soften and bridge in the hopper, while drying below 70°C does not reach the required equilibrium moisture in the core of the pellet.

When does additional EVOH thickness stop lowering oxygen transmission?

In ideal solution-diffusion transport through a homogeneous film, oxygen permeation is inversely proportional to layer thickness, so doubling the EVOH layer from 0.10 mm to 0.20 mm would halve the oxygen transmission rate. The linear inverse relationship holds only when the EVOH layer is continuous, void-free, and at uniform moisture content. In coextruded PE-RT pipe, each of these assumptions breaks down at a different point. A 0.03 mm EVOH layer is near the minimum continuous thickness for spiral mandrel distribution at pipe diameters above 25 mm; below this, layer breakup and gauge variation dominate and oxygen transmission becomes highly variable. At thicknesses above 0.15 mm, the barrier benefit is often smaller than predicted because the outer PE-RT layers and tie layers impose additional diffusive resistance, and the EVOH layer develops a radial moisture gradient that is not accounted for in single-layer film tests. Industrial coextrusion lines for 16–32 mm PE-RT pipe typically target 0.08–0.12 mm EVOH for 38 mol% ethylene grades, which balances the DIN 4726 oxygen diffusion requirement against pipe flexibility and cost. Above 0.20 mm, the EVOH layer stiffens the pipe wall and can increase flexural modulus beyond the field-bend tolerance specified in installation standards; moreover, the higher EVOH volume increases the risk of gel-related pinholes because the layer spends more time in the die at elevated temperature. Published data for this specific configuration is limited, but process-scale observations indicate that increasing EVOH thickness beyond 0.20 mm does not proportionally reduce oxygen ingress after 1000 h of wet ageing because moisture plasticization of the layer governs the remaining transport rate.

Values represent typical commercial datasheet ranges for unpigmented EVOH film specimens, not specification maxima for finished pipe. The oxygen transmission rate of the finished PE-RT pipe is also influenced by layer thickness, adhesion, and the moisture content of the inner and outer polyolefin layers.

Whole-pipe oxygen diffusion testing exposes localized barrier defects

Two distinct measurement philosophies are used for oxygen barrier pipe. The first is a coulometric oxygen transmission rate measurement on flat films or extruded tape according to ASTM D3985-17; this method provides a permeation coefficient but does not capture pipe curvature, weld-line effects, or layer-to-layer adhesion. The second is a whole-pipe oxygen diffusion test in which the pipe is filled with deoxygenated water, sealed with oxygen-tight closures, and exposed to air at 40°C or 70°C depending on the standard. DIN 4726 specifies an oxygen permeation limit of 0.1 g/(m³·d) at 40°C water for surface heating pipes, and the test is typically run for 30–60 days with periodic sampling of dissolved oxygen. A limitation of the whole-pipe method is that it integrates oxygen ingress through the mandrel-formed inner surface and the outer surface, so a localized defect may pass if the average permeation remains below the limit. For this reason, internal production control on coextrusion lines often pairs the whole-pipe test with a high-resolution layer thickness scan and a peel test on a sacrificial length every 4 h. The coulometric method on pipe sections can be adapted by sealing the outer surface and applying a nitrogen carrier gas inside, but published data for this specific configuration is limited because end-seal leakage often exceeds the oxygen transmission of the EVOH barrier and masks the measurement.

Across continuous multi-shift coextrusion campaigns, the most difficult parameter to hold constant is not the EVOH melt temperature but the moisture content and thickness distribution of the tie layers. A batch of tie resin stored at high humidity can introduce interfacial water that weakens the anhydride-hydroxyl reaction and reduces peel strength below 10 N/25 mm, even when the EVOH layer remains intact. In continuous operation, the EVOH extruder is started only after the PE-RT and tie-layer streams have stabilized at full output for 15 min; if EVOH is introduced too early, low melt-temperature PE-RT can cause the EVOH layer to freeze at the die mandrel and form a spiral fold that later becomes a radial oxygen leak path. Cross-section microscopy every 4 h is used to verify layer concentricity, and the EVOH layer is measured at four radial positions because eccentricity of more than 0.02 mm creates a thin side that dominates oxygen diffusion. Where national specifications require whole-pipe oxygen testing after installation, a failed pipe cannot be remediated by external coatings on the inner surface; the barrier layer must be replaced or the pipe section cut out.