Polyethylene terephthalate (PET) bottle manufacturing is often discussed as a packaging problem, but the actual cost behaviour is determined by melt-phase thermal history, stretch-induced orientation, and gas pressure management. A two-stage line separates preform injection moulding from reheat stretch blow moulding; the split is not merely logistical. It determines where molecular weight loss, scrap, and electrical load accumulate. For a 500 ml carbonated soft drink bottle with a preform weight of 24 g, resin dominates the variable cost. Published cost models from equipment suppliers and resin producers typically assign 60–70% of the variable bottle cost to PET resin, 15–20% to conversion inputs, and the remainder to logistics, secondary packaging, and quality release testing. The following sections isolate the main cost positions.

Running a PET Bottling Line: Where the Cost Actually Sits

The cost concentration differs by line configuration. In hot-fill and carbonated soft drink applications, preforms are heavier and resin share increases. In still water bottles below 12 g, conversion cost and scrap become proportionally larger. The table below summarises the typical cost and energy distribution for a 25,000 bottles/h two-stage stretch blow moulding line processing 24 g preforms at 70% mechanical efficiency. Ranges reflect published technical bulletins and plant energy audits; line-specific values vary with altitude, cooling-water temperature, resin supplier, and preform design.

Resin cost is rarely fixed; it follows formula pricing over purified terephthalic acid and monoethylene glycol. A 50 € per metric ton movement in PET resin changes the cost of a 24 g preform by 0.0012 €. On a 25,000 bottles/h line running 6,000 hours per year, that equals 180,000 € per year for every 50 € change. This is why preform weight and scrap control have more leverage than utility optimisation alone.

Cooling-water demand for injection moulds and hydraulic systems is another hidden cost. A preform injection mould for a 72-cavity system may require 80–120 m³/h of chilled water at 8–12 °C. If the chiller condenser is air-cooled and ambient air enters above 35 °C, the electrical load can rise by 15–20% because the refrigeration lift increases. Water-cooled chillers with evaporative towers improve coefficient of performance but consume make-up water. Published data for water consumption varies with water hardness and bleed rate; typical evaporative systems require 2–4 L of make-up water per kilowatt-hour of rejected heat. The cost sits not in the water itself but in treatment, pump energy, and fouling of mould channels.

Compressed air is the most misallocated cost in blow moulding. In a rotary stretch blow moulding machine, high-pressure air at 30–38 bar is used for final bottle formation, but the energy stored in the bottle cavity is exhausted after each cycle unless recovery circuits are installed. The pressure-time integral rather than peak pressure governs material distribution: for a 24 g preform, final blowing pressure above 38 bar rarely improves sidewall orientation but increases compressor heat load and maintenance frequency. Air recovery systems can return 35–50% of the exhaust energy, but their efficiency collapses if the blow valve exhaust path has a backpressure above 0.5 bar or if the recovery receiver volume is less than 3–5 times the cavity volume. On production lines, the largest compressed-air losses typically occur at quick-exhaust valves, pinholes in pneumatic seals, and condensation traps left open during sanitation. These losses are not visible in moulding parameters but appear in the electricity bill as a 20–30% increase in compressor-specific energy.

Preform injection moulding cost is not solely determined by cycle time. High-cavitation systems are sensitive to hot-runner balance, gate freeze, and cushion consistency. The injection barrel is usually a reciprocating screw with 20:1–24:1 L/D and a compression ratio of 2.5:1 to 3:1, designed to limit shear heating and acetaldehyde generation. Shot-to-shot consistency on a 48-cavity or 72-cavity preform platform depends on cushion control between 3 and 6 mm. Below 3 mm, cushion variation can produce short shots; above 6 mm, residence time increases and acetaldehyde rises. Holding pressure is maintained until gate freeze but must be balanced against mould opening: overpacking the gate area creates gate crystallinity and higher static electricity, while underpacking increases sink marks and weight variation. Preform weight variation above ±0.1 g per cavity on a 72-cavity tool shifts material distribution in the blown bottle and raises top-load variability, a defect that cannot be corrected at the filler.

What Process Instabilities Emerge When Intrinsic Viscosity Falls Below 0.70 dL/g?

Bottle-grade PET resin for carbonated soft drinks is generally supplied at 0.80 ± 0.02 dL/g when measured in 60/40 phenol/1,1,2,2-tetrachloroethane according to ASTM D4603-18. During injection moulding, hydrolytic chain scission is driven by residual moisture. At moisture contents above 50 ppm, the IV loss per pass can increase from 0.015 dL/g to more than 0.04 dL/g. Desiccant dryers are therefore set to deliver air with a dew point of −40 °C to −50 °C and a residence time of 4–6 hours for virgin pellet at 160–170 °C. Hopper and conveying lines must be sealed; a 1 m³/h leak of ambient air at 25 °C and 60% relative humidity can introduce enough moisture to raise dried resin moisture above specification within 30 minutes.

Thermal degradation also produces acetaldehyde. Residual acetaldehyde measured by ASTM F2013-10 in preform wall is controlled below 5 ppm for taste-sensitive still water and below 8 ppm for many carbonated beverages, though customer specifications vary. Acetaldehyde generation is temperature- and time-dependent; barrel residence times above 5 minutes at 285 °C increase acetaldehyde sharply. On high-cavitation preform systems, the hot runner must hold melt temperature within ±1.5 °C across all gates. If a gate is 2–3 °C cooler than adjacent gates, the affected preform can exhibit different weight, gate crystallinity, and blow-moulding behaviour. The cost appears later as intermittent blowouts, carbonation loss, or top-load failures, rather than as an immediate injection alarm.

Acetaldehyde is a sensory defect with an odour threshold in water of roughly 20–50 µg/L. Therefore, preform residual acetaldehyde alone is not the only criterion; migration into water is governed by time, temperature, and bottle surface-to-volume ratio. This is why preform acetaldehyde specifications are stricter than resin acetaldehyde specifications.

When Lightweighting Moves the Bottleneck from Injection to Blow Molding

Lightweighting changes the cost position. For a 500 ml still water bottle at 8–10 g, the material cost is lower, but the wall thickness in the label panel can fall below 0.15 mm. Under top load and drop impact, performance is dominated by stretch-induced orientation and sidewall thickness distribution. The preform design must therefore be matched to the reheat profile. PET absorbs infrared radiation and reheat ovens must produce a surface temperature gradient of 100–115 °C for the preform body while keeping the neck below 60 °C to prevent deformation. Above 115 °C, spherulitic crystallisation competes with strain-induced orientation, causing haze and loss of bottle burst strength.

In stretch blow moulding, axial stretch ratios of 2.8–3.2 and radial stretch ratios of 3.5–4.0 are common. When preform weight drops below 10 g, the processing window narrows; the difference between insufficient stretch and stress whitening can be ±3 °C in reheat temperature. The blow-pressure curve becomes the main cost lever. A 50 ms reduction in exhaust vent time per cycle can raise output by 8–12% on high-speed rotary machines, but only if vent silencers and air-recovery pipework do not create backpressure above 0.5 bar. In practice, many lines operate at 25–30 bar blowing pressure for lightweight water bottles; carbonated soft drink bottles may require 35–38 bar to force the sidewall into the mould detail and reduce base clearance variation.

Top-load strength is measured according to ASTM D2659-16 at 23 °C and 50% relative humidity on blow-moulded containers. For lightweight bottles, top-load above 250 N is often required, but the value depends on neck finish and wall thickness. If sidewall orientation is insufficient, top-load values scatter and the filler can crush bottles during capping.

Recycled PET shifts the cost equation away from virgin resin price. Post-consumer flake after hot wash and solid-state polycondensation can be blended at 25–50% with virgin resin. The actual cost is not captured by flake price alone; variability in flake IV (0.68–0.74 dL/g), colour b*, and residual cap/liner contamination requires near-infrared sorting and metal detection upstream. For food-contact bottles, FDA requires the recycled plastic to be of suitable purity under 21 CFR 177.1630 and any letter of no objection; EU regulation 10/2011 sets overall migration below 10 mg/dm². These compliance costs sit in documentation, batch quarantine, and testing. Published data for specific rPET blend processing windows is limited because flake source and SSP history differ by supplier.