Fragrance-induced environmental stress cracking in thick-wall cap geometries is distinguished from ordinary detergent stress cracking by the thermodynamic compatibility of terpene-rich fragrance oils with semicrystalline polyolefins and by the high local tensile stresses generated at the sealing plug root during closure removal. A 38 mm continuous-thread cap with a bridge wall of 2.8–3.5 mm, a sealing plug wall of 2.2–2.6 mm, and a plug-to-bridge transition radius below 0.20 mm produces a stress concentration factor that can exceed 2.0 when the cap is torqued onto a glass or PET container finish. The removal torque applied by a consumer may be only 1.8–2.5 N·m, but the resulting thread normal force is amplified by the cap’s interference geometry; the local tensile stress at the inner plug root can reach 8–12 MPa depending upon thread flank angle, plug engagement depth, and the stiffness of the finish. When a fragrance concentrate containing d-limonene, linalool, citral, or benzyl acetate contacts the stressed plug surface, the amorphous tie-chain regions absorb the low-molar-mass species, the craze initiation stress decreases, and radial cracks propagate from the inner bore toward the external knurl within 14–28 days of ambient aging. The failure is therefore not a simple chemical attack but a coupled mechano-sorptive process; published data for this specific cap configuration is limited, so qualification studies must be carried out on the moulded closure rather than on a standard tensile plaque. The hoop stress component in a thin-walled cylinder is calculated as σ_h = p·D/(2t), but the more relevant stress in a thick-wall cap is the assembly stress from interference and thread torque, which cannot be derived from internal pressure alone. Moulded-in residual stress from an injection gate can be additive to this assembly stress, especially when a central gate places a frozen-in orientation zone adjacent to the plug root. On a 32-cavity hot-runner mould with a 24:1 L/D injection unit and a 0.8 mm valve-gate orifice, cycle-to-cycle gate-seal variation may produce cavities with elevated local birefringence, and those cavities are typically the first to exhibit plug-root cracks when exposed to fragrance concentrate.

How do sealing plug penetration depth and gate location redistribute tensile stress?

Sealing plug penetration depth determines the length over which the plug root is subjected to bending and hoop stress. In a cap with a plug penetration of 8.0 mm and a plug wall thickness of 2.5 mm, a removal torque of 2.0 N·m creates a combined stress state at the root comprising membrane tension, bending, and hoop expansion. Finite element analyses generally show that the maximum first principal stress moves from the outer thread flank to the inner plug root as the plug-to-bridge transition radius decreases below 0.15 mm. The gate location interacts directly with this stress field because the frozen-in flow orientation near a valve gate creates anisotropic craze resistance; a cap gated at the centre of the top panel concentrates gate vestige stress at the same zone where the plug root stress is highest, whereas a side-gated cap shifts the high-orientation region to the skirt and can reduce plug-root craze initiation. In a 32-cavity hot-runner mould with a 0.8 mm valve-gate orifice and a gate-seal time of 2.5 s, the local moulded-in stress near the gate can exceed 15 MPa when the moulded part is measured by photoelastic fringe order; this stress is additive with assembly stress and accelerates fragrance-induced cracking. Caps with a gradually increasing plug wall thickness from 1.8 mm at the tip to 3.0 mm at the root and a root radius of 0.5 mm exhibit improved stress distribution; however, such geometry may violate tamper-evidence or child-resistant design requirements and increase cycle time because the thicker root requires longer cooling. The transition from the top panel to the skirt is another discontinuity that must be analysed separately from the plug root; a sharp transition below 0.25 mm can initiate cracking at the outer knurl when the cap is squeezed during removal. For child-resistant closures tested to ISO 8317, the push-down force of 40–80 N precedes torque application and increases axial compression on the plug root, which changes the multiaxial stress state from simple hoop tension to combined compression-shear. Under those conditions, a crack that would normally initiate in the hoop direction may instead propagate at 45° to the thread axis, following the maximum shear plane.

Solvent Ingestion, Crystallinity Gradients, and Craze Fibrillation in a Constrained Bridge

Fragrance species transport through a thick-wall cap bridge follows a non-Fickian concentration profile because the outer surface is exposed to air, the inner surface is exposed to fragrance vapour or liquid, and the polymer contains a crystallinity gradient created by slow cooling. A 3.0 mm wall of polypropylene homopolymer processed at a mould temperature of 20°C develops a skin layer with lower crystallinity and higher orientation than the core, and the equilibrium sorption of d-limonene is lower in the crystalline core than in the amorphous skin. The glass transition temperature of the plasticised amorphous phase can be suppressed by 2–5°C; although the bulk heat deflection temperature remains acceptable under ISO 75-2, the local craze resistance in the amorphous regions is reduced. Stress cracking occurs when the applied tensile stress exceeds the craze initiation stress of the plasticised tie-chain network. In a standard ESC test using a bent strip per ASTM D1693-15 with Igepal CO-630 at 50°C, the F₅₀ failure time for a high-molecular-weight high-density polyethylene may exceed 1000 h; but when the medium is replaced with a fragrance oil containing 1.0 wt% d-limonene, the same material may fail in less than 100 h because the solubility parameter of d-limonene is closer to that of the polyolefin than the standard wetting agent. This discrepancy is the central reason why generic detergent ESC data cannot be used to approve a fragrance closure. Testing must include the actual fragrance formulation and the moulded cap geometry because the cap bridge represents a diffusion path length that is one to two orders of magnitude greater than a standard test specimen. In a thick wall, the diffusion time scale increases approximately with the square of thickness, so a 3.0 mm wall may require weeks to approach saturation under ambient exposure, while a 1.75 mm ASTM test bar may equilibrate within days. The delayed saturation can falsely suggest a resistant material in short-term tests, only for cracks to appear after packaging has been shipped.

Material selection for thick-wall fragrance closures is constrained by the inverse relationship between environmental stress cracking resistance and flowability. A high-density polyethylene grade with a bimodal molecular-weight distribution and a melt flow rate of 0.25–0.45 g/10 min at 190°C/2.16 kg per ISO 1133-1:2022 offers excellent craze resistance but requires higher melt temperature and injection speed to fill a 32-cavity cap mould. A polypropylene impact copolymer with an ethylene content of 6–12 wt% provides better melt flow and acceptable impact resistance, but the polyethylene-rich domains absorb fragrance oils and can promote cavitation at the sealing plug root. A homopolymer polypropylene with a melt flow rate of 8–12 g/10 min at 230°C/2.16 kg fills thin sections readily but has lower environmental stress cracking resistance and tends to shatter when a sharp notch is present. The moulded cap should be evaluated for top-load force per ASTM D2659-16, for removal torque retention per ASTM D2063-05, and for craze initiation after immersion in a fragrance simulant; published data for this specific configuration is limited, so safety factors of at least 1.5 on measured failure time are common in package qualification protocols. High-density polyethylene closures with a density of 0.948–0.954 g/cm³ measured by ISO 1183-1 usually provide better ESCR than medium-density grades, but the denser crystal network can increase stiffness to the point where a child-resistant closure requires excessive push-down force. The closure manufacturer must therefore maintain a narrow density window because a shift of 0.005 g/cm³ can alter both torque removal force and ESCR.

When Fragrance Concentrate Lot Variation Demands Narrow Moulding Windows

Fragrance concentrate lot variation changes the aggressive solvent load without changing the cap geometry. A shift from a parfum concentrate containing 0.5 wt% d-limonene to one containing 2.0 wt% d-limonene may reduce the observed crack initiation time by more than half if the cap is moulded at the low end of the melt temperature window. In production trials on a 32-cavity hot-runner tool, a melt temperature of 220°C with a holding pressure of 65 MPa and a mould coolant temperature of 25°C may produce no cracking after 30 days, while increasing the melt temperature to 240°C to improve surface gloss can increase residual orientation near the gate and trigger cracking after 7 days in the same fragrance medium. The processing window therefore is not defined solely by the resin; it is defined by the intersection of the fragrance solubility parameter, the cap root radius, and the residual stress profile. A gate-seal time below 2.0 s for a 0.8 mm valve gate can create gate blowback and a highly oriented weld line at the gate vestige; a gate-seal time above 3.5 s overpacks the plug root and increases local residual stress. The acceptable cycle time may be as narrow as 9.0–11.5 s for a 3.0 mm bridge wall, depending upon cooling channel pressure and mould temperature. Thermal imaging of ejected parts, combined with photoelastic stress analysis, is necessary to identify which cavities in a multi-cavity tool produce parts with the highest residual stress. The table below summarises the processing window that is commonly evaluated during qualification of thick-wall polypropylene closures destined for fragrance service.

Post-mould conditioning and assembly practices also determine whether thick-wall caps survive fragrance contact. A cap ejected at 80°C and immediately packed into polyethylene bags can retain volatile residual stress and absorb external fragrance vapour before dimensional stabilisation. Conditioning for 48 h at 23±2°C and 50±5% RH per ISO 291 allows shrinkage and stress relaxation to stabilise, but it does not remove moulded-in tensile stress caused by gate orientation. Applying a lining compound or a sealant can introduce an additional solvent system at the plug root, and some linings contain volatile plasticisers that reduce polyolefin craze resistance. The closure torque retention test should be performed after the cap has been aged with the actual product for at least 30 days at ambient temperature and for 72 h at 50°C to detect accelerated environmental attack. A child-resistant closure certified to ISO 8317 must maintain its push-down force and removal torque after exposure; if the fragrance causes plug-root cracks, the torque retention may drop below the standard’s minimum acceptance value because the cracked plug no longer develops the same interference. Therefore, the operational boundary is not a single environmental condition but a matrix of fragrance contact, closure torque, moulded-in stress, and material formulation. The test matrix below cross-references the standard methods most commonly applied when a fragrance-exposed closure exhibits field cracking and a root cause must be documented.

In a production environment, the decision to quarantine a specific lot of caps after fragrance contact is often based on a combination of removal torque loss, tensile yield-stress retention, and stereomicroscopic examination at 10× magnification of the plug-root radius. A visible craze network at the inner bore, even before a full fracture, indicates that the plasticised amorphous phase has undergone fibrillation. At that stage the closure may still pass a leak test, but the residual load-bearing capacity is reduced. For high-density polyethylene caps with a nominal bridge thickness of 3.0 mm, craze initiation is not uniformly distributed across the circumference; it commonly localises at the gate vestige, at a weld line formed around a core pin, or at the thinnest part of the bridge opposite a vent slot. Each of these positions corresponds to a local reduction in craze resistance caused by orientation, degraded additive concentration, or geometrical stress concentration. The cap supplier and fragrance manufacturer must therefore exchange data on fragrance composition, closure torque settings, injection gate location, mould temperature, and the relevant ESCR grade of the resin. Without this combined data set, no single test can reliably predict packaging shelf life for a thick-wall fragrance cap.