While a previous article focused on uneven wall thickness distribution I’ll explain some of the other problems that processors are dealing with on a day-to-day basis. These will be problems that are part of both the single and two-stage process.

1. Pearlescence and haze

These are really problems of opposite nature that can however be hard to distinguish as both lead to a whitening of the bottle.

Pearlescence is also called stress-whitening, pointing more clearly to its causes. When PET is overstretched its micro structure can break up and the resulting wall shows a number of little white circles, hence the connection to ‘pearls’. Overstretching is a function of both wall thickness and temperature, i.e. a thin wall at a temperature well over the Glass Transition Temperature may perform perfectly well whereas a thicker but colder wall may show pearlescence. Pearlescence always happens on the inside of the bottle because the inside of the preform has to stretch so much more. Once this is understood, a solution can be found to the problem at hand. If the affected area is very thin it should be cooled down and the area underneath it should be heated to bring more material into the whitish part of the bottle. If it is reasonably thick temperature in this section or overall should be increased to allow it stretch more easily. In two-stage this is accomplished by changing the lamp or ventilation settings. In single stage, overall heat is controlled by hold and cooling time while a change in injection speed can be used to change the temperature variation in the preform.

A good indicator of pearlescence is its occurrence in highly stretched areas like the feet in this 15 L container.

Haze develops when PET in either injection or blow is heated beyond about 115° C (240° F) and allowed to cool down for even a short time. At this temperature PET molecules have enough flexibility to form crystals and they will do that because it allows for a lower energy state. In two-stage the responsible lamp can easily be found and must be turned down or ventilation increased. Haze always happens on the outside of the preform and ventilation will cool this part more than the inside. In single-stage, cooling and/or hold time are the timers that are increased combat haze.

2. Insufficient top load

As bottle weights are constantly reduced to save resin top load has become a much more important issue. Top load is measured mostly on empty bottles in a suitable device with adjustable speed. The achieved value is then compared with the static load (how many bottles are on top of the bottom pallet layer) and multiplied with a factor of safety (or ignorance) to account for dynamic loads when a truck goes over a bump and the entire load is accelerated up and down. Multiplying static load by 2,3 or 4 is not a very scientific method and some companies have measured the actual load in a truck with sometimes surprising results. Top load performance depends almost exclusively on wall thickness and bottle design. Bottles collapse at the weakest part of the bottle, be it shoulder, bottom, or body. Processors must know where the bottle failed so they can move more material into this area. One vexing issue is that values of empty bottles are of interest to nobody since empty bottles do not fail because of top loading. Some bottles will show the same behavior whether they are full or empty but others will perform up to 4 times better when they are filled. Brand owners can of course draw their own conclusions but converters should have a discussion about necessary top load with their customers if this is the case.

3. Insufficient base clearance / rocker bottom

All bottles have what is usually referred to as a push-up. The center of the bottom curves inward to a various degree. This is to ensure that the bottle only sits on the outer rim to the base or, in the case of bottles for carbonated beverages, on feet, typically five. This center will always shrink out to some degree. When it is too hot after molding it will move out so much that it starts to protrude beyond the limits of the outer rim and the bottle no longer stands straight, it ‘rocks’. It can also happen that residual air pressure inside the bottle pushes the center of the bottom out when the mold opens, even though this is a rather rare occurrence. On some machines cooling jets blow air at the bottom after they have been molded to prevent the shrinking-out from happening. Otherwise, the processor has to reduce heat to the bottom of the preform or increase cooling time.

When the base is too hot after molding it will shrink out and form a center leg that makes the bottle rock, hence the name ‘rocker bottom’.

For carbonated beverages base clearance, the distance of the center of the bottle bottom to the feet, is of utmost importance. This is because the carbonation in the product can lead to a pressure of up to 70 psi inside the bottle. This pressure then pushes on the center disk of the bottle base that therefore has to have some room to move before it exceeds the feet. Base clearance should be measured in short intervals, typically 2 hours. It is often the one parameter that keeps machines from running faster because of the necessary cooling time.

4. Drop impact failure

Bottles must withstand being dropped from 4 to 5 feet containing water at around 4°C (40°F) to simulate conditions in a household refrigerator. Contrary to polyolefins PET does not break at the thinnest corner (unless there is pearlescence or haze). It may dent but will not break because it is highly oriented in these thin spots. A lack of orientation is therefore the most common cause of failure, most often found in custom bottles with small stretch ratios and more common in single-stage than in two-stage. In two-stage a cooler preform temperature will force more orientation into the material and improve performance. In single-stage cooling or hold time must be added, which will increase cycle time. Hold time has more impact because the preform shrinks away from the core during the cooling phase and is therefore preferred. When preforms are overheated to the point of haze the crystalline areas are more brittle and can also fail. The same remedy applies.