Retailer Sector Innovation: Esterform Packaging Final Report March 2007

Raising the bar in PET bottle lightweighting

Research findings that achieved a new European lightweighting standard for carbonated drinks bottles made from PET.

WRAP works in partnership to encourage and enable businesses and consumers to be more efficient in their use of materials and recycle more things more often. This helps to minimise landfill, reduce carbon emissions and improve our environment.

Contents 1 Executive summary 2 Project scope 3 Lightweighting two litre bottles 4 Lightweighting 500ml bottles 5 Benefit analysis 6 Overall conclusions 7 Appendices 3 5 8 11 14 16 17

While steps have been taken to ensure its accuracy, WRAP cannot accept responsibility or be held liable to any person for any loss or damage arising out of or in connection with this information being accurate, incomplete or misleading. For more detail, please refer to our Terms & Conditions on our website - www.wrap.org.uk

3 Retailer Innovation Final Report March 2007

1 Executive summary

This project was initiated by Esterform Packaging, with the support of WRAPs Innovation Fund, to develop innovative packaging that could reduce household packaging waste originating from the retail sector. While the rate of recovery of bottles is improving rapidly from a low base, reducing the weight of beverage packaging provides a further powerful way of conserving materials, reducing packaging weight and reducing the energy used in production and transport. The cost savings involved also provide an incentive to invest in the new tooling needed to produce the lighter bottles. This project makes use of recent improvements in blow moulding technology, resin technology and preform design as a means of further reducing the weight of PET bottles made in the UK. Two bottles were selected for their potential to become optimum packaging weight examples for the beverage market: a 500ml Carbonated Soft Drinks (CSD) bottle for Radnor Hills. This bottle was initially 25g, and it has been successfully redesigned and produced at the target weight of 20g; and a two litre CSD bottle for a major CSD manufacturer. This bottle was initially 42g, and was successfully redesigned at the target weight of 40g. The reduction in packaging weight achieved directly through this project was 250 tonnes of PET per annum, which is equivalent to the elimination of 8.3 million (30g) bottles from the waste stream.

If the new designs were adopted throughout the UK beverage market for these bottle sizes, the savings of PET being used would be 3,400 tonnes, which is equivalent to 113 million 30g bottles. The weight savings also lead to energy savings during processing. For Esterform Packaging the new preform designs could deliver annual energy savings of over 200MWhr. Energy savings of over 2,810MWhr would be achieved if the two new designs were adopted across the UK beverage industry. The development of the new preforms has led to applications beyond those that were originally intended, extending the weight savings achieved and delivering even greater environmental benefits. This validates the approach adopted in this project, as well as showing the value of disseminating the improved efficiencies throughout the UK beverage industry.

4 Retailer Innovation Final Report March 2007

Contents

1. Executive summary 2. Project scope

2.1 Background 2.2 Aim of the project 2.3 Timescales 2.4 Project partners 2.5 Project methodology 2.6 Manufacturing considerations

3 5
5 5 5 5 6 7

3. Lightweighting two litre bottles

3.1 Preform design 3.2 Typical design points 3.3 Blow moulding trial for preforms 3.4 Creep testing of two litre bottles 3.5 Line trials 3.6 Conclusions on the suitability of lightweight two litre bottles 3.7 Other applications for the lightweight 40g preforms

8
8 8 9 9 10 10 10

4. Lightweighting 500ml bottles

4.1 Preform design 4.2 Blow moulding trial for 20g and 22g preforms for 500ml bottles 4.3 Performance of 20g and 22g 500ml bottles 4.4 Stability tests 4.5 Material redistribution improvement 4.6 On-line testing at Radnor Hills 4.7 Other applications for the 500ml 20g preforms

11
11 11 12 13 13 13 13

5. Benefit analysis
5.1 Energy, carbon and CO2 emissions savings of new preforms designs 5.2 Cost benefit analysis 5.3 Overall benefit analysis

14
14 15 15

6. Overall conclusions 7. Appendices

7.1 Injection blow moulding process 7.2 Two litre preform 41g and 40g designs 7.3 Designs for 20g and 22g preforms for 500ml bottles 7.4 ressure and dimensional stability tests on the 20g, 22g P and 25g bottles 7.5 mproved design for 20g preform for 500ml bottles I 7.6 Bottle variations tested with 20g preform

16 17
17 18 19 20 21 22

5 Retailer Innovation Final Report March 2007

2 Project scope
2.1

Background
This project was developed with the support of WRAPs Innovation Fund, to develop and trial innovative packaging that will reduce household packaging waste originating from the retail sector. According to WRAPs 2006 UK Plastic bottle recycling survey, the total quantity of plastic bottles entering the UK household waste stream is approximately 510,000 tonnes per annum, with an estimated 85,000 tonnes of bottles collected by the end of 2005. This represents a recycling rate of 17%, and is equivalent to 2,125 million plastic bottles . If the same approach were applied to the whole of the beverage market, a 10% weight reduction would result in savings of 19,000 tonnes, which would be equivalent to removing more than 570 million bottles (with an average weight of 30g each) from the waste stream. While the rate of recovery of bottles is improving rapidly from a low base, the approach of lightweighting beverage packaging originating from the retail sector provides a further powerful way to minimise waste at source, and reduce raw material processing and distribution costs. This is because it provides an economic incentive to make the necessary changes to the bottles, provided it is technically feasible to do so. If a weight reduction feasibility study is successful then new tooling would be required, which, while initially expensive, can be amortised over a number of years to recoup the capital outlay. Since virgin PET costs approximately 800 per tonne, each reduction of 100 tonnes would represent a saving of 80,000. The opportunity to reduce the weight of two bottles beyond current levels has been made possible by exploiting recent improvements that have occurred in blow moulding technology, resin technology and preform design.
2.2

The key steps taken in the project were: the re-design of the preforms using the latest techniques available to the PET industry and Esterform Packaging; the manufacture and testing of the new designs in conjunction with preform machine suppliers, blowmoulding machine suppliers and beverage fillers; and testing of the new designs in short- and long-term performance tests and filling trials.
2.3

Timescales
The project was begun in May 2005, and completed at the end of May 2006.
2.4

Project partners
The project was initiated by Esterform Packaging with the collaboration of two of their customers. Technical expertise was provided by Husky, SIG and Sidel, key providers of world-class PET moulding and processing technology. The project was managed by Nextek Limited. The key partners and contacts are shown in Table 1.

Partner Esterform Packaging

Role Leading organisation, preform manufacturer and bottle supplier Project management

Key Contacts Mark Tyne, Director

Nextek Pty Ltd

Professor Edward Kosior William Watkins

Radnor Hills Mineral Water Bottle filler Co. CSD manufacturer Husky Injection Moulding systems (UK) Ltd Sidel SIG Krones Bottle filler Preform design and supply Preform design and blow moulding Preform design and blow moulding Preform design and blow moulding

Aim of the project

This project investigated the technical and commercial feasibility of reducing the weight of two specific PET bottles made in the UK by Esterform Packaging, for two major beverage fillers supplying supermarkets: a 500ml CSD bottle used by Radnor Hills; and a two litre CSD bottle. These two bottles were selected because they were already very competitive in terms of packaging weight in the market place. The target weight for each bottle would set a new optimum packaging weight for these sizes of PET bottles, and as a result, others across the industry would be encouraged to follow suit in light-weighting. The 500ml bottle was initially 25g, and the target weight was 20g (a reduction of 20%). The two litre bottle was initially 42g, and the target weight was 40g (a reduction of 4.8%). The potential reduction in packaging weight that might be achieved through this project would be 250 tonnes of PET per annum for the two product lines in the trials - equivalent to 8.3 million bottles. If the new designs were adopted for the whole of the UK market, then the savings of packaging materials would increase to 3,400 tonnes (equivalent to 113 million bottles).

Stephen Martin Colin Farrant Bill Farrant Karl Hopkins Table 1 Project partners

6 Retailer Innovation Final Report March 2007

2.5

Project methodology
The weight of PET bottles is determined predominantly by the bottle performance requirements and specification of the neck and base. The neck design is influenced by handling requirements, and the closure style used. Different neck finishes can have different weights associated with design features. The design and weight of the base is related to the requirement to withstand the internal pressure of a carbonated product. Hence still water bottles can be lighter than carbonated beverage bottles. The mid-section body of the bottle was the focus for manipulation of design for lightweighting in this project, since the design of the neck and base was unchanged. This lightweighting would be achieved by gaining better material distribution during the blow moulding steps, by exploiting new developments in PET resin (energyabsorbing additives) and improvements in preform heating in blow moulding machines. Esterform Packaging anticipated being able to lightweight the samples of the 500ml bottle from 25g to 22g and possibly 20g (the lightest bottle in Europe is currently 22g. In the UK the 500ml bottles are typically 23-25g). The target for the two litre bottle was a reduction from 42g to 40g. Currently two litre bottles are 41.5- 42g. The project was carried out in the manner described in the bullet points, in order to achieve the best technical result in the shortest time: development of a process checklist; development of new designs via a Computer Aided Design (CAD)system; adaptation of the preform designs to suit bottle and blowing machinery; resolution of any nesting issues though preform redesign via CAD simulation; feedback from machinery companies, Krones, SIG and Husky, received and used in the designs; quotations requested for preforms from Husky; moulds, and/or core pins manufactured by Husky; preforms for use in trials manufactured by Husky; bottles blown and tested by Esterform Packaging; line trials conducted at the laboratories of Esterform Packaging and fillers; approval testing conducted by the filler companies; and successful new designs were to be commercialised by Esterform Packaging for supply to the fillers.

7 Retailer Innovation Final Report March 2007

2.6

Manufacturing considerations

Light may affect some ingredients; high storage temperatures would force CO2 out of solution into headspace, making drink fizzy on opening; would also soften bottle.

Cap is screwed on to required torque, note inner shape of tamperevidence ring.

A carbonated drink in a PET bottle

Diffusion of gases depends on their relative concentrations, so more oxygen outside means it will diffuse inwards, as PET is permeable. One volume is about 2gL-1, pressure per vol is about 1 atmosphere (pressure at sea level, 15C, is 101325 kPa).

Possible effects of light and temperature on bottle and drink. Oxygen will diffuse inwards- may oxidise some flavours.

Labels put on after filling, stretch/heat shrunk to fit; any label adhesive and label ink must not migrate Plastic screw cap through the plastic and with tamper-evidence taint the drink. ring attached.

CO2 loss through bottle wall, about 0.04 volumes per week. About 0.3 volumes of CO2 absorbed in bottle wall in first 3-4 days. Up to 1% water may evaporate through bottle wall over shelf life. Identification symbol. Petalloid base so PET is the only plastic used, easier for recycling.

2L bottle (CO2 content 4 volumes or 8 g L -1) product ageing through oxidation and/or acid hydrolysis or flavour oils

Creep expansion (about 2.5%) in first 3-4 days, plus elastic deformation from internal pressure. These affect choice of labels. Odours from the environment could taint product. Depending on the product, up to 5% flavours may diffuse through bottle wall, and/or be absorbed by the plastic. Components of plastic or materials used in manufacture may migrate into drink (e.g. minute, harmless traces of acetaldehyde, PET oligomer). Bottles made thin for least material (a 2 L bottle is <40g), relies on internal pressure for rigidity. Must be stored away from odours.

Maximum shelf-life of 16-17 weeks before loss of CO2 is significant. Allowed for when filling.

Better quality ingredients could be used.

Allowed for in manufacture.

PET bottles can be recycled, but only with the same plastic; recycled PET can be used in a middle layer between layers of new PET.

1 PET

A PET preform is first stretched downwards (1), then blown outwards (2), i.e. biaxial, before it cools; gives more strength and better barrier properties. 2 2

PET bottles are stretch-blow moulded from an injection-moulded preform. Varying wall thickness to give even wall thickness in blown bottle.

All materials leach some components into their contents, the quantity affected by reactions with the contents, but well within safe limits.

Larger PET bottles (e.g. 5 L) have a smaller surface area to volume ratio and lose a smaller proportion of CO2 and other volatile ingredients than do smaller bottles.

PET bottles are lighter and much less liable to breakage than glass bottles, use less energy for transportation.

Glass bottles would be considered for many similar applications; glass is impermeable, commonly used in 350 mL to 1.5 L sizes, with 2 L glass bottles considered the maximum size for easy and safe handling.

Figure 1 Manufacturing considerations for a carbonated drink in PET bottle

8 Retailer Innovation Final Report March 2007

3 Lightweighting two litre bottles

3.1 3.2

Preform design
The size, shape and surface features, as well as bottle performance requirements of the final bottle shape, determine the shape of a preform. Structural designers use well-known expansion ratios, since a controlled amount of stretching at controlled temperatures determines the physical performance of the bottles. The standard preform used for two litre bottles at a major CSD manufacturer has been 42g for a number of years. This preform has already been optimised and reduced in weight from 44g, and the success of further reduction was not immediately assured. For this reason the project team decided to assess the performance of a 41g preform as well as a 40g preform. The design team developed the new designs through an iterative process, first looking at ways to improve the design to meet the bottle performance requirements by minimising the weight of the relevant neck, mid-section and base parts of the bottle and contracting these into the design of the preform. The preform was then redesigned to meet the heating and expansion requirements for creating the final bottle shape. The new design was then analysed and modified by the machinery specialists from Husky, Sidel, SIG and Krones for ease of moulding and blow moulding. CAD simulation was used to detect and then avoid potential problems such as preform nesting. This process was repeated once the final design was arrived at, to ensure that it represented the best possible design for the target lightweight bottle. The technical project team at Esterform Packaging developed the designs for lightweighted bottles as shown in drawings in Appendix 7.2 An order was placed with Husky to manufacture new moulds, and to mould 5,000 preforms of each design. The blow moulding of the preforms into bottles and the testing of lightweight bottles was carried out at Esterform Packaging. Burst pressure, Carbon Dioxide (CO2) retention creep behaviour tests, blowing trials, and conveying and filling level tests were all carried out. Appendix 7.1 provides a simple diagram of the injection blow moulding process.

Typical design points considered for lightweight product development

Preform design to suit bottle: material distribution establish weight required in neck, body and base; stretch ratios best possible for performance, avoiding creep, CO2 loss etc; and design of preform to suit injection moulding process - wall, flow length, tip thickness etc. Preform design to suit blowing machine: neck damage when tipping or handling preforms; nesting* of preforms; support for collar ledge for roller in-feeds; output speeds; efficiency; palletising; and air conveying. Bottle performance, filling line: handling to filler; labelling, expansion of label panel; and packing, expansion of body. Bottle performance in market place: effect of label panel creep; base performance; and shelf life/CO2 loss. Typical steps to achieve design: establish customer bottle weight now; establish preform weight and design used now; and establish bottle performance that is required by the customer. By using a computer model, investigate whether savings can be made on: bottle neck - known specifications to be worked-to; bottle petaloid base - proven designs and performance weight specification to be followed; and bottle body - are there thick areas that could be reduced etc? From findings, produce proposed preform design Supply preform design to the blowing machine supplier for feedback on: suitability in relation to the finished bottle; confirmation that the blowing machine output speed will not be effected and that it will blow efficiently; confirmation that the preform will handle efficiently through the machine; and any other comments? Supply preform design to Husky for feedback on suitability for good quality and efficient production: confirm that tip thickness is optimum; confirm that wall thickness is sufficient to prevent knit lines at the neck; confirm that the injection pressure required to fill the cavity is not excessive, preventing premature mould wear; and any other comments?

*Nesting = the base of one preform locating in the opening of another preform during manufacture and/or distribution. Can effect production efficiencies.

 Retailer Innovation Final Report March 2007

3.3

Blow moulding trial for preforms

The blow moulding trial of the 41g and 40g preforms for two litre bottles was carried out with staff from Esterform Packaging present. For this initial trial, 500 of each of the 40g and 41g preforms were supplied and run on the Krones blowing machine, producing two litre petaloid bottles. The machine was set up to run the standard 42g preform, to keep the adjustments to a minimum. After this settingup phase, the 41g preform was the first to be trialled, followed by the 40g preform. The following observations were made at the trials: a minor decrease in the temperature at the tip of the preform mould was all that was required. No other changes were necessary; the machine performance was not affected; the base weight achieved was 8.5g/9g, within the normal acceptable specification; the bottles passed the customers burst test at 10 bar; the bottles look good and received a positive response from the customer. the 40g bottles appeared to be produced more easily than the 41g.
3.4

Figure 2 41g 40g Bottles and preforms of 40g and 41g

Creep testing of two litre bottles

The creep behaviour test examines the progressive expansion of a bottle under pressure once it is filled and carbonated. Plastics are visco-elastic materials, and continue to respond to applied pressures over time this behaviour is known as creep. In a lightweight bottle the progressive expansion may affect appearance, CO2 retention and adhesion of an applied label, and so it is important to characterise this behaviour. The creep behaviour of the lightweight bottles was analysed, starting on the 7th October. This was carried out by pressurising the filled bottles with CO2 to 3.5 atmospheres, and monitoring changes in dimensions and pressure. The key factor here was to determine whether the lightweight bottles creep and lose CO2 significantly faster than the standard 42g bottle. The testing was carried out for three months, to simulate expected shelf life. The results for retention of CO2 pressure and circumference are shown below, and indicate that even though there are minor changes due to lightweighting, there is no significant difference between the 3 bottles. These results for both 40g and 41g bottles were within the required CO2 retention and creep performance range, enabling line trials to proceed. 40g Date Day 0 14/10/05 25/10/05 12/11/05 06/12/05 03/01/06 CO2 3.5 3.1 3.19 2.84 2.72 2.54 Circumference 31cm 30.9cm 31.1cm 31.1cm 31.0cm 31.0cm CO2 3.5 3.2 3.19 2.88 2.73 2.61 41g Circumference 31cm 30.8cm 30.9cm 30.9cm 30.8cm 30.9cm CO2 3.5 3.29 3.19 2.95 2.79 2.64 42g Circumference 31cm 30.8cm 31.0cm 31.0cm 31.0cm 30.8cm Table 2 Creep test results, 40g, 41g and 42g preforms

10 Retailer Innovation Final Report March 2007

3.5

Line trials
The initial on-line testing of the two litre bottles was carried out on a small scale and completed with satisfactory results for filling, capping and labelling performance. More extensive trials were planned using preforms in four different weights. Esterform Packaging moulded a further 80,000 preforms to be used in extensive online trials to validate the performance of the lighter preforms. Esterform Packaging made 40,000 additional WRAP preforms, i.e. 20,000 each of the 41g and 40g preforms. In addition, Esterform Packaging modified another preform tool by making new cores, to bring the weight of the preforms to 41.5g and 40g. They made 40,000 preforms, i.e. 20,000 of each of the 41.5g and 40g preforms. This step was taken to investigate whether the tooling costs for lightweighting could be reduced. These preforms were run and then tested for line performance as well as for the creep behaviour of the four types of preforms over a further three months. This large scale trial shows that the two companies are seriously considering ways of turning into commercial reality the potential revealed by the initial test results.
3.6

Conclusions on the suitability of lightweight two litre bottles.

The redesign of the two litre preform and the subsequent test results show that the bottle meets: the physical requirements of a bottle, relating to burst pressure strength and base weight for example; processing protocols (i.e. conditions were little changed from standard settings during blow moulding); line filling, capping and labelling performance requirements; CO2 retention performance requirements; and requirements for resistance to creep under carbonation pressures. These results show that it is highly likely that a 40g preform can be used for two litre bottles.
3.7

Other applications for the lightweight 40g preforms

The 40g preform was also successfully tested as a replacement for a 1.5 litre water bottle that currently uses a 44g preform. The redesign of the preform to a shorter format means that the stretching ratio would be higher and the physical properties improved. The success of this application means that there are further opportunities to look into for this 40g preform, with even further savings in PET that could be achieved.

Figure 3 1.5 litre bottle made using 40g preform instead of 44g preform

44g

40g

11 Retailer Innovation Final Report March 2007

4 Lightweighting 500ml bottles

4.1

Preform design
The typical weight of 500ml bottles in UK is 23g to 25g. Esterform Packaging currently uses 25g preforms, and they anticipated being able to lightweight the 500ml bottle to 22g and possibly 20g. The size, shape and surface features of the final bottle determine the shape of a preform. Designers use wellknown expansion ratios, since a controlled amount of stretching at controlled temperatures determines the physical performance of the bottles. Since the design of the neck is determined by the closure specification, the redesign of the preform to deliver a reduction in weight may result in a preform diameter that is less than the diameter of the neck opening. This situation can lead to nesting of the performs, which can lead to difficulties in unscrambling the preforms as they go from the storage bins into the blow moulding machines. CAD simulation predicted that this could have been the situation with the two lightweighting designs, though the dimensions were changed (i.e. the diameter was increased and the length reduced compared to the ideal guidelines) to avoid this possibility. The preforms were still within the typical design guidelines used to predict axial and longitudinal expansion, in order to achieve acceptable physical properties. The preform designs for the 22g and 20g preforms are shown in Appendix 7.3.
4.2

Figure 4 22g 25g 20g Standard 25g and lightweight 22g and 20g preforms

Blow moulding trial for 20g and 22g preforms for 500ml bottles
The trial of the 20g and 22g preforms was carried out at Esterform Packaging using four-foot and five-foot petaloid bases, and they were compared to the standard 25g bottle. The two base designs can both be used for carbonated bottles, and historically the four-footed base has been the standard and accepted design. The five-footed base provides better material distribution and hence a stronger and also more stable base for the bottle that is less distorted by carbonation pressures. The differences between the preform designs are shown in Figure 4. The redesign of the preforms has resulted in a preform that is shorter but larger in diameter, to avoid nesting problems that would occur if the external diameter were reduced in the lightweighting process. The shape of the bottles is shown in the CAD images in Figure 5. The four variants and the standard moulded bottles are shown in Figures 6 and 7.
20g 4 foot 20g 5 foot 22g 4 foot 22g 5 foot 25g 4 foot

Figure 5 4 foot bottle 5 foot bottle CAD images of the different base designs

Figure 6 End view of 20g and 22g bottles with four- and fivefooted bases, along with the 25g standard bottle

Figure 7 20g and 22g bottles with four- and five-footed bases, along with the 25g standard bottle

12 Retailer Innovation Final Report March 2007

4.3

Performance of 20g and 22g 500ml bottles

The specification for the 25g, four-foot standard bottle is: base weight: 4.5g minimum to 6.0g maximum; panel weight: 7.7g minimum to 9.0g maximum; burst Test: 150 psi minimum; height: 215.4 1mm; and fill height: 505 mls 5 mls at 40 mm; The results for the 20g and 22g preforms (see Table 3), show that the 20g and 22g bottles in the four-foot and five-foot versions meet the physical specifications for the standard bottle with the related lighter weights on the base and panel. Height (mm) specimen 1 2 Mean 20g 4ft 215.5 215.5 215.5 20g 5ft 215.6 215.5 215.6 22g 4ft 215.4 215.4 215.4 22g 5ft 215.3 214.4 215.4 20g 4ft 506.1 506.1 506.1 Fill Height (mm) 20g 5ft 501.6 501.1 501.4 22g 4ft 503.5 503.6 503.6 22g 5ft 498.7 499.3 4.0 20g 4ft 526.3 526.8 526.6 Brimful (mm) 20g 5ft 522.1 521.4 521.8 22g 4ft 524.1 524.3 524.2 22g 5ft 519.4 519.5 51.5

Base weight (g) specimen 1 2 Mean 20g 4ft 4.3 4.1 4.2 20g 5ft 4.3 4.3 4.3 22g 4ft 5.0 5.1 5.1 22g 5ft 5.1 5.1 5.1 20g 4ft 5.7 5.7 5.7

Panel weight (g) 20g 5ft 5.6 5.6 5.6 22g 4ft 6.6 6.6 6.6 22g 5ft 6.6 6.5 6.6 20g 4ft 152 154 153

Burst test (psi) 20g 5ft 148 154 151 22g 4ft 199 199 1 Table 3 Properties of 20g and 22g bottles in four- and five-foot variations. 22g 5ft 196 200 18

13 Retailer Innovation Final Report March 2007

4.4

4.7

Stability tests
The performance of the 20g and 22g bottles was tested for stability under pressurised conditions. The results are shown in Appendix 7.4. The key findings are: the four-foot existing 25g bottle is stable at all pressures; the five-foot petaloid 20g and 22g bottles remained stable at all pressures; the four-foot bottle at 22g started losing stability at 60psi, but was still relatively good at 75psi; the four-foot bottle at 20g started losing stability at 40psi, and was considered very bad at 60psi; the use of the five-footed preform improved the material base distribution; and the material distribution near the neck was not as good as was desired, potentially indicating a need to redesign the preform.
4.5

Other applications for the 500ml, 20g preforms

The improvement in the material distribution has meant that the 20g preform could be considered for other applications as well as the Radnor Hills bottle. Esterform Packaging conducted blowing trials on the bottles shown in Figure 8. Esterform Packaging is currently conducting further tests on the applicability of the lighter preforms. The bottles that could potentially be made from the preform are: the 500ml standard Petaloid bottle, drawing A242B-04-3 in Appendix 7.6. Esterform Packaging has a considerable number of other customers that take this bottle. Although the five-foot version performed better than the four-foot standard, Esterform Packagings customers require the look of the fourfoot; the 500ml Ribbed bottle. Bottles have been successfully blown, and can be used for both still and carbonated products; the 500ml Squound bottle. This bottle is specifically for one of Esterform Packagings customers. Bottles have been successfully blown and Esterform Packaging will meet with the customer to present samples; the 500ml Water Wave bottle. Bottles have been successfully blown; the 500ml Sport bottle. Bottles have been successfully blown; and the 500ml CH Carbonates Petaloid bottle. Bottles have been successfully blown.

Material redistribution improvement

In order to improve the material distribution in the neck section, a redesign of the preform was considered, as shown in Appendix 7.5. Sidel was consulted on the redesign, and the new dimensional specifications for the preform were only slightly different form the original. Husky was also consulted on the newly redesigned preform. Their advice was that further thinning of the neck section would make injection moulding of the preform problematic and could result in poor neck thread definition. As a result of these two discussions a further moulding trial was conducted with new processing conditions. An improved material distribution was achieved. One issue that has arisen is the current preference of Radnor Hills for a four-footed base for stylistic reasons. This may have implications for how far the lightweighting process can be taken, given the limitations of the four-foot base compared to the five-foot base.
4.6

On-line testing at Radnor Hills

The on-line testing of the 20g, 500ml water bottles at Radnor Hills was completed with satisfactory results for filling, capping and labelling performance. Insufficient bottles were available for the assessment of unscrambling, though this is not anticipated to be a problem. The bottles were assessed for carbonation level performance for three months shelf life. Radnor Hills confirmed that the 20g, 500ml bottles had met their filling and shelf life requirements for standard bottles.
squound CH Carbonates Standard Wave Sport

Figure 8 Different bottle styles that could potentially use the 500ml, 20g preform.

14 Retailer Innovation Final Report March 2007

5 Benefit analysis
5.1

Energy, carbon and CO2 emissions savings of new preform designs

The energy consumed in the drying of resin and melting of PET to make a preform and then bottle, is directly related to its weight. The reduction in energy consumption is proportional to the weight savings and can be readily calculated from the weight of PET that avoids being processed. The savings are most significant for the melting of PET. CO2 emissions can be calculated from power consumption data. However, although the average electricity emissions are 0.43 kg CO2/KWh (Carbon Trust, 2005), the figures used here are for marginal electricity, i.e. the electricity production that is turned on or off as a response to changes in demand. In the UK, the marginal fuel is gas, which instead gives a figure for electricity of 0.36 kg CO2/KWh (DTI Energy Trends, 2005). The data in Table 4 shows that the new preform designs could deliver annual energy savings of over 200MWhr. The projection for energy savings was over 2,810MWhr if the technology were to be adopted in the two designs across the UK beverage industry. These savings can also be expressed in terms of the number of UK households that could be powered by these quantities as shown in Table 5. The calculations show that the Esterform Packaging savings are equivalent to the energy consumption of 52 households, and the UK savings of 703 households. The data is based on the annual household energy consumption of 4000kWh for a three-bedroom house in the UK1. Moulding Moulding Energy for preforms Energy to process 1kg PET kWh 0.80 Preform heating kWh 0.0267 Total Saving kWh 0.83 Total Saving kg CO2 0.36 Total Saving kg Carbon 0.273 kg C /kg CO2

Energy savings for 30 million 500ml bottles @ 5g reduction Energy savings for 50 million 2 litre bottles @ 2g reduction Total savings for Esterform (80 million bottles)

120,000 80,000 200,000

4,005 2,670 6,675

124,005 82,670 206,675

44,642 29,761 74,403

12,175 8,117 20,22

Energy savings for 80 million 500ml bottles @ 5g reduction Energy savings for 1500 million 2 litre bottles @ 2g reduction Total savings for UK (1,580 million bottles )
2

320,000 2,400,100 2,720,000

10,680 80,100 0,780

330,680 2,480,100 2,810,780

119,045 892,836 1,011,881

32,467 243,501 275,67 Table 4

Overview: Energy and carbon saving using the lightweight preform, based on data provided by Husky, plus associated CO2 and carbon emissions calculations Annual energy consumption for a three-bed household in UK Total annual energy saving for Esterform Equivalent number of three-bed housholds Total annual energy saving for UK Equivalent number of three-bed housholds Energy saving per 100 tonnes equivalent to number of households 4,000 kWh 206,667 kWh 52 2,810,780 kWh 703 21 Table 5 Energy saving expressed in household energy consumption per year
1 Source: University of Strathclyde, http://www.esru.strath.ac.uk/EandE/web_site/01-02RE_info/hec.htm 2 =3,400 tonnes of PET

15 Retailer Innovation Final Report March 2007

5.2

Material savings for Esterform Packaging

Preform weight (g) Initial 42 25 Final 40 20 Weight saving % 7% 20% Market volume tpa 1500 160 Weight of PET tpa 105 32 Cost saving @800/t 84,000 25,600

Assumptions: Volume of 500ml bottles would double from existing 80 tonnes p.a. to 160 tonnes p.a. Table 6 Simple assessment of material cost saving opportunity by moving to the lighter weight preform.
5.3

Benefit analysis if lightweighting on these bottle sizes was replicated across the sector
The Esterform Packaging project clearly demonstrates that a move to lighter weight preforms brings with it benefits in terms of material savings, energy savings and carbon emissions. If this lightweighting on these bottle sizes was replicated across the whole carbonated soft drinks sector, the savings can be seen in Table 7. Per million 500ml PET bottles (using 20g rather than 25g preforms) PET weight saving PET material cost savings Carbon emission savings Energy savings 5 tonnes 4,000 0.41 tonnes 4,133 kWh Per million 2 litre PET bottles (using 40g rather than 42g preforms) 2 tonnes 1,600
0.10 tonnes

For the UKs 1,580 million PET bottles (500ml and 2 litre bottle sizes combined) 3,400 tonnes 2,720,000
276 tonnes

1,653 kWh

2,810,780 kWh Table 7 Overall benefit analysis

16 Retailer Innovation Final Report March 2007

6 Overall conclusions
The approach taken in this project has been successful due to the latest advances in PET technology being applied to the two bottles selected. The lightweight 20g and 40g designs are currently meeting the key success criteria set by the current bottle specifications at both Esterform Packaging and the fillers plants. New applications for the 40g and 20g preforms have been blown and are also being considered by customers beyond the initial scope of the project. It is expected that this approach will lead to significantly larger weight savings in PET than originally anticipated. This project has delivered additional benefits in savings in the costs of materials and energy input, thereby delivering multiple environmental benefits. This strongly validates the approach used in this project, also highlighting the value in disseminating the improved efficiencies throughout the UK beverage industry.

17 Retailer Innovation Final Report March 2007

7 Appendices
7.1

Injection blow-moulding process

Neck Ring Injection Neck Core Ring

Neck Ring

Neck Ring Blow Mould

Neck Ring

Neck Ring

Injection Cavity

Blow Air

Blow Air

Blow Air
Hotrunner

Blow Air

Bottom Plug

18 Retailer Innovation Final Report March 2007

7.2

Two litre preform 41g and 40g designs

1 Retailer Innovation Final Report March 2007

7.3

Designs for 20g for 22g preforms for 500ml bottles.

20 Retailer Innovation Final Report March 2007

7.4

Pressure and dimensional stability tests on the 20g, 22g and 25g bottles

500ml lightweight bottle for WRAP project

Preforms conforming to drawings (20g) E774-P-05 - (22g) E795-P-02 Bottles conforming to drawings (four-foot) A242-B-04-3 - (five-foot) E803-B-01 Test Temperature of water 27.5 deg C

Test results from first blown bottles 30th November 05:-

Effect of pressure on diameter diameter Bottle weight/style 20g five-foot 20g four-foot 22g five-foot 22g four-foot Existing 25g four-foot 0psi 63 63 63 63 63 diameter 50psi 64.1 64.3 63.7 63.9 63.4 diameter 75psi 65.6 66.1 64.3 64.8 63.6

Base clearance mm at 0psi 20g five-foot 20g four-foot 22g five-foot 22g four-foot Existing 25g four-foot 5.4 5.36 5.84 5.85 5 mm at 50psi 3.4 3.4 3.8 3.8 3.85 mm at 75psi 2.9 2.75 3.38 3.2 3.5

Base performance The five-foot petaloid 20g and 22g remained stable at all pressures. The four-foot at 20g started losing stability at 40psi, and was very bad at 60psi. The four-foot at 22g started losing stability at 60psi, and was still relatively good at 75psi. The existing four-foot bottle at 25g is stable at all pressures.

21 Retailer Innovation Final Report March 2007

7.5

Improved design for 20g preform for 500ml bottles

22 Retailer Innovation Final Report March 2007

7.6

Bottle variations tested with 20g preforms 500ml standard Petaloid bottle, drawing A242-B-04-3.

This is the principal bottle being evaluated for Radnor Springs

Waste & Resources Action Programme

The Old Academy, 21 Horse Fair, Banbury, Oxon OX16 0AH

Tel: 01295 819 900 Fax: 01295 819 911 E-mail: [email protected]

Helpline freephone 0808 100 2040

www.wrap.org.uk/retail

When you have nished with this report please recycle it.