AU593827B2 - Beverage can - Google Patents

Description

r ir? t i i: i i n Ynemr*~^^*ram~ .U~~jr;l~lrUr~~~SUU~ COMMONWEALTH OF AUSTRALIA PATENT ACT 1952 COMPLETE

SPECIFICATION

(Original) FOR OFFICE USE 9 Class Int. Class Application Number: Lodged: 61053/ Complete Specification Lodged: Accepted: Published: Priority: Related Art: kgt 1h1;: Rocument cot e t1 Sa'2O! 49 nd i coIrrec

FICE

s P1 16 e tI t Name of Applicant: Address of Applicant: 'SHELIrJO1X-COMPANY II V c, ti; 7 iir i

C

Actual Inventor(s) t 4 Address for Service:

I

900-r -oisiana, V A V- 8-oy Houstonr Texas. 7700.1, t -Uni t ed States of-America.

Walter Richard

SCHMEAL

Surendra Nath SINGHAL King Him LO DAVIES COLLISON, Patent Attorneys, 1 Little Collins Street, Melbourne, 3000.

Complete Specification for the invention entitled: "BEVERAGE

CAN"

The following statement is a full description of this invention, including the best method of poerforming it known to us 1- 7 -r 1a- "BEVERAGE CAN" The present invention seeks to provide a fiber wound container plastic beverage can which will withstand pressure from carbonated beverages, which has low swelling (or stretching) properties, and which has a fabrication cost less than that of a typically used aluminum can.

Numerous sizes, shapes and forms of beverage cans are available to the consumer. Many cans are made of aluminum, in particular those which must withstand high pressure conditions of carbonated beverages, such as soft drinks, and thus rely on the strength of the aluminum as a can wall material. Various shapes, forms and sizes of beverage bottles are also available, but this invention is addressed to cans and not bottles.

The performance criteria for carbonated beverage r1 v t S 5- 5 cans are: 1) A can must have a low permeability so as to ensure a shelf life (during storage at, for example, supermarkets and home) for up to 4 months. This means a maximum loss of 16% CO 2 for the typical carbonated beverage can.

2) A can must be strong enough to withstand a pressure of 0.6 MPa gauge (87 psig) at a 350C (950F) liquid temperature. This is so that cans stored in hot warehouses or automobiles for extended periods of time will not burst, split, crack or otherwise fail at a pressure below the buckling stability limit of the can lid. This pressure is about 0.6 MPa gauge (87 psig) for a conventional pull tab aluminum lid.

3) A can should be able to withstand axial loads which occur when cans are stacked in warehouses and during end seaming of tops and bottoms.

4) The volume stability of a can is very important. The can must retain cosmetic and structural integrity, to allow performance in the vending distribution k,, -2channel. Thus, the can should have nonexistent or low creep tendency at 35 0 C (950F) liquid temperature and 0.6 MPa gauge (87 psig).

Performance criteria 2) and 4) are the most important and the most difficult to meet when fabricating a carbonated beverage can.

Heretofore, the fabrication of plastic cans has been difficult, US-A-4,379,014 discloses a plastic beverage can which is fabricated by winding thermoplastic tape around a mandrel and then reinforcing the tape. A single, homogeneous polyester sheet material is used to construct a container, presumably for vessels capable of withstanding q internal pressure, for example those containing carbonated beverages. As taught, the homogeneous oriented sheet is wrapped monodirectionally and continuously onto a mandrel to form a cylinder. As taught this document there is no possibility of separating two important functions of the container wall: 1) to provide a continuous vessel wall to prevent leakage, and 2) to provide the necessary strength when pressurized. Because the sheet material functions as both the strength material and the homogeneous and continuous finished container wall, the sheet material must be continuous to prevent leakage.

Further, US-A-4379014 does not teach the use of barrier layers to prevent carbonation loss in beverage containers. It is well known in the art of beverage containers formed of polyester material that carbonation loss is significant performance factor, especially in smaller sized containers (0.47 litres (16 U.S. fluid ounces) or less) where the ratio of the volume of the carbonated contents to the surface area of the container is high. In homogeneous, thin wall polyester vessels as taught in US-A-4379014 such a barrier layer would be required for carbonated beverage containers in order to meet shelf life requirements in smaller sized containers.

3 US-A-3662944 discloses a batch process for making Scontainers one at a time, rather than a continuous process.

This patent also does not teach the fiber wrap angles which must be employed in making a, fiber wound plastic beverage can. Also required in US-A-3662944 is a plastic layer, with or without an aluminum foil "label" layer, which is necessary to support the wound fibers to impart the rigidity necessary of such a container.

US-A-2873226 teaches another batch process wherein a polystyrene inner form and a thermoset resin system is used as an adhesive.

US-A-3735895 teaches a non-pressurized vessel with glass fibers bound by epoxy resin.

US-A-3098582 teaches a batch process where a 15 curing resin system is used as an adhesive.

The present invention provides a carbonated beverage can which has a barrier layer to give good product t 4 shelf life and low permeability, and which has sufficient strength to withstand the pressures of a carbonated beverage at the temperatures to which it might be subjected, sufficient axial load bearing stiffness and strength so that the top closure may be joined and the cans may be stacked in warehouses, good volume stability, i.e. no excessive swelling, which retains cosmetic and structural integrity, 25 and which feels like a can and will gain consumer acceptance. Furthermore, the can can be continuously fabricated.

*According to the invention there is provided a beverage can for carbonated beverages which comprises a can wall made of a thermoplastic core and at least one fiber l3; wound adhesive layer on said thermoplastic core, the can wall comprising a barrier layer between said i thermoplastic core and said at least one fiber adhesive wound layer, or contained within said thermoplastic core or between adjacent fiber adhesive wound layers.

r 11 c b 2 -4 The beverage can of the invention thus has a can wall made of thermoplastic materials and is provided with ends which may be made of thermoplastic or metallic materials. The can is capable of withstanding pressure by the use of a continuously fabricated homogeneous core providing the container body/wall (function to prevent leakage) upon which: fiber strands (strength function) are wound/applied, said fibers not necessarily being homogeneous in their coverage of the container body/wall, and is also capable of resisting volume swelling.

The invention will be described in more detail by reference to the accompanying drawings, in which: Fig. 1 is a cross-section, partly in perspective, of the wall of carbonated beverage can fabricated in accordance with the invention; Fig. 2 is a flow sheet which illustrates the basic process steps involved in fabricating a filled beverage can in accordance with the invention; Fig. 3 is a schematic side elevation illustrating the step of core making of the process summarised in Fig. 2; Fig. 4 is a schematic side elevation illustrating the step of fiber winding a core in the process summarised in Fig. 2; 25 Fig. 5 is a schematic side elevation illustrating the step of labelling, cutting and bottom joining a fiber-wound core in the process summarised in Fig. 2; and Fig. 6 is a graph showing the relationship between fiber coverage per layer of a fiber wound core and its burst strength.

Referring now to Fig. 1 there is shown a fiber wound soft drink can which has a film wound wall which, beginning at the innermost layer, is fabricated of a polypropylene film 1, a polypropylene film 2 and a t I t 4 II

V

'4411

I

c ;u

L

I c i

J

5 polypropylene film 3, which are contiguous one with another and which are bonded by adhesive layers 4 between the layers of films 1 and 2 and films 2 and 3. An extruded core may be used instead of a film wound core as illustrated in Fig. i.

Adjacent to film 3 is a barrier layer 5 of latex "Saran" (copolymer of vinylidene chloride). Adjacent barrier layer is a layer of polyethylene terephthate (PET) fiber 6.

Fiber layer 6 is bonded by an adhesive layer 7 to a second PET fiber layer 8. Fiber layer 8 is in turn bonded by adhesive layer 9 to a label film Fig. 2 is a flow sheet illustrating one method of fabrication of the fiber wound beverage can. In Step 1, (core making), lubricant 11, polypropylene films i, 2 and 3, adhesive 4, and latex Saran 5 are combined to make the core.

15 In Step 2 (fiber winding), wound PET fiber layers 6 and 8 and adhesive 7 are added to the core. In Step 3 (labeling and cutting) a label film 10 is applied and a can length is cut from the fiber wound core. In Step 4 (bottom joining), can bottoms 12 and joining materials 13 are used to attach 20 the bottoms 12 to the can. In Step 5 (filling), cleaning materials 14 and beverage 15 are used to clean the can and fill it with the desired beverage. In Step 6 (top joining), tops 16 and joining materials 17 are used to join the tops 16 to the can.

*r 9 4 4 eL 4t 4 4 4t~ 4 4 4 25 Fig. 3 illustrates Step 1 (core making) in more detail. A mandrel 18 is (ending at 18A) is affixed with a belt drive 19 to rotate the mandrel 18. A torpedo control mechanism 20 is attached to the end of madrel 18 and a source of air 21 is provided to torpedo contaol mechanism 20. The mandrel 18 is lubricated from lubricant tank 22 and adhesive from adhesive tank 23 is applied to polypropylene films 1 and 2. Films i, 2 and 3 are sequentially wrapped at an appropriate angle around mandrel 18. Films i, 2 and 3 are then treated by film treater 24. At some point after treatment of films 1, 2 and 3, mandrel 18 ends. "Saran" i; ei Q is i thnaple o S- 6 latex from tank 25 is then applied to I applicator 26. The films progress thr emerges as core 28.

Core 28 then passes to Step S shown in Fig. 4. Core 28 passes throu Sstage 29, adhesive 7 is applied from a I core 28 and a spiral winder 31 winds t stationary fiber stage 29 onto core 28 through fiber dryer 32 and belt puller wound core 34. A torpedo seal 35 resi core 34 and is attached to the torpedc the moving films by an ough dryer 27 and 2 (fiber winding) as igh a stationary fiber dhesive tank 30 to he fibers from Core 28 is passed 33 and emerges as des inside the wound control mechanism i1 a

(I.

I LI

V

by a stationary torpedo control wire which passes through a hole in the center of the mandrel 18 and also through the center of the moving core 28. The purpose of the seal 35 is 15 to retain pressure to prevent failure of the core 28 due to torques applied by the stationary fiber stage 29 and the spiral winder 31.

As shown in Fig. 5 which illustrate Step 3, the wound core 34 is wrapped by label film 10 provided with adhesive 9. Traveling saw 35 cuts wound core 34 into 1.2 meter (four feet) lengths 36 and transfers lengths 36 onto a conveyor 37 and then through a slit saw 38 provided with a plurality of heads to cut the wound core 34 into appropriate lengths for the cans. The bodies of the can 39 are then 25 sent to Step 4 as by a conveyor to have bottoms 12 joined, to be filled with beverage 15, and to have tops 16 joined.

The core may be made of polypropylene, polyethylene, polyketone, polyethylene terephthalate (PET), paper or some other suitable material or combination of materials. The core may be constructed by extrusion, or core winding of films onto a lubricated fixed mandrel. The core may be of any thickness but 0.25 0.76 mm (0.010 0.030 inches) is preferred. The core should be homogeneous enough to prevent leakage and also rigid enough to allow the application of the fibers at stationary and spiral winding

A

7 -7stages 29 and 31.

Example 1 Core Making A 0.33 mm (13 mil) extruded polypropylene core of Sabout 6.6 cm (2.6 inch) outside diameter was pressurized with air and torque was applied to simulate winding torques.

The torque to failure is noted below in Table 1 as a function of the applied internal air pressure. Bursting occurred at (40 psi).

TABLE 1 APPLIED PRESSURE kPa (psi) Torque at Failure Nm (inch-lb) 0 1.7 19.2 (170) {i 103 (15) 25.4 (225) 124 (18) 28.2 (250) 276 (40) burst The thermoplastic core may be filled with inert material or recycled plastic for the purpose of allowing faster extrusion and lowering cost. Examples of filling materials are CaCO 3 mica and talc. A barrier to reduce the 20 rate of diffusion of water and gases such as CO 2 and 02 may be added to the core by coextrusion, or by application contiguous to or within the fiber-adhesive layer. Adhesives may be applied to either side of the barrier. The purpose of the barriers is to enhance the utility of the container.

Examples of such materials are SARAN (copolymer of vinylidene chloride) and EVOH (hydrolzed copolymers of vinyl S acetate and ethylene). These may be applied by spraying such material in a latex form onto the exterior of the core or extruding the barrier within the base core material, or *,J as an integral or laminated film in the film wound core construction. The coextrusion may include additional layers of adhesives on either or both sides of the barrier.

As example might be ABA, ABC, or ACBCA, where A is the base core material, B is a barrier and C is an adhesive. The f a$I L* 8 first letter designates the internal layer and the last letter the external layer.

Various fibers may be used in this invention.

Suitable fibers are those of glass, polyethylene terephthalate (PET), nylon, rayon, or cotton. These fibers are characterized as having very high elastic modulus, excellent creep properties, very high breaking strength, high initial yield strength and low cost. Such fiber prop rties are crucial to the design of an economical/high performance beverage container to replace metal cans.

Fiber strand dimensions are characterized by their weight per unit length (or its reciprocal) in terms of yield, in units of meters per kg (yards per pound) or in terms of denier. Denier is a measure of strand diameter for 4 15 a given fiber type.

Each fiber strand is made up of many small filaments held together with sizing material. For example, Celanese Type x 785 PET fiber has a strand diameter calculated to be 0.37 mm (14.5 mils) when in a cylindrical state and it is made up of 74 filaments of 0.036 mm (1.4 mil) diameter each. When this fiber is wound under tension, j it flattens out on the mandrel to a width of 0.77 mm about (30.5 mils) and a height of about 0.076 mm (3 mils). This S flattening is very beneficial from a performance/cost perspective since it allows the core to be covered by thin, economic layers of PET fibers. These fibers inhi /it creep t. L cof the underlying thermoplastic core.

It is quite noteworthy that it is not necessary completely to cover the thermoplastic core with fibers to gain the benefit of their superior machanical properties.

Fibers wound about a 6.6 cm (2.6 inch) diameter core at 60 0 C for example make a complete revolution in 12 cm (4.72 inches) of cylinder length. This is the band width for a 600 winding angle.

Preferably two layers are used, each with a

'I

m -9if i Is 4i 0! I 4 I a Cf

C-

aP winding angle with respect to the axis of the cylinder which 0 is the negative of the other, viz, +60 This creates a balance of forces which prevents twisting of the cylinder when under pressure.

In each layer, 136 fibers of a width of 0.77 mm of (30.5 mils) would be required for a complete coverage of the core, calculated by dividing the band width by the fiber width and multiplying by the sine of the winding angle.

Example 2 Fiber Winding Fibers were wound on a 0.33 mm (13 mil) polypropylene extruded tube using a McClean Anderson Model i, filament winder at Westhollow Research Center. Test set A was carried out with glass and PET fiber available from previous filament winding and pultrusion work. Test sets B 15 and C were carried out with Celanese Type 785 PET seatbelt cord of 93.2 tex (840 denier). Dimensions of this fiber and other typical candidate fibers which were on hand, PPG, ECG 75 glass, and Allied 1W69 denier PET are given in Table 2.

Static burst tests were carried out in the existing pipe testing apparatus according to a modification of the ASTM test standards Test No. D 1599. Test cylinders were cut to lengths of at least 25.4 cm (ten inches) from the wound specimens. Special end plugs were designed after trials of numerous designs. The specimens were immersed in the water bath, filled with water, and pressurized with nitrogen rapidly until they burst.

Long term creep tests were carried out by pressuring specimens to 0.38 MPa gauge (55 psig) at room temperature in an ambient air environment. The conditions after a few seconds represent the initial conditions after can filling and it is the swelling which occurs after this period which is relevant to creep performance.

Short term high temperature creep tests were also carried out. Specimens were immersed in the hot water bath at 43 C (110°F) filled with water, and pressurized to 0.69

I

i r; d~~ b

I

i~i i0 MPa gauge (100 psig) with nitrogen. This condition was more severe than required for a carbonated beverage container to emphasize problem areas which require further study, thus there was no adherence to the ASTM test specification of I 5 holding the speciment for thirty (30) minutes before beginning measurements. The change in volume with time was recorded by photography. The first volume swell picture was taken at the time that the pressure gauge on the waterbath holding tank indicated 0.69 MPa gauge (100 psig).

The radius of the specimen at various axial positions and the height were recorded, and the volume calculated. Base conditions were chosen after temperature and pressure were established (arbitrarily five minutes into the test) for S 1 evaluation of long term creep and volume swell. Volume 15 changes after this time represent the effects of creep under rtti t,-L the constant, very severe conditions of 0.69 MPa gauge and 00 43 C (100 psig and 110°F)

RESULTS

Test Set A 20 The first set, Set A, was carried out to screen r It effects of fiber thickness, balanced versus unbalanced fiber J tlayers and the fole of adhesive for glass and PET fibers.

All wound specimens passed the quick burst test by exceeding *o 0.69 MPa gauge and 43 C (100 psig pressure at 110 F) and none exhibited observable volume expansion, as seen in Table 3.

t, "Presence or absence of adhesive, presence or absence of balanced layers of fibers, and fiber thickness were varied. Unbalanced specimens exhibited some twisting.

The fiber winding angle was 890. it, A 0.05 mm (two mil) outer surface of Hercules T-503 oriented polypropylene film was used for handling purposes.

1:1 I-t

'I

11

-A,

i I 4 I Of r 4 I t

RESULTS

Test Sets B and C Test Set B was designed to provide performance data on PET fiber wound layers on the polypropylene tube combined in accordance with the expected stress and strain response of the composite container. Test Set B was undertaken to test the use of 840 denier PET fiber. This fiber was selected becuse it was relatively fine in diameter (Tire cord is typically 111 tex (1,000 denier)) and has an enhanced tendency to flatten. Flattening occurs because it is composed of individual filaments which are fewer in number, but thicker than those used in tire cord. The filament diameter was 0.036 mm (1.4 mils). Fiber surface coverage was varied and a winding angle of +60 degrees was used although various different winding angles may be calculated and used other than 600. The calculated stresses in a pressurized cylinder are such that those in the hoop direction are twice those in the axial direction. The optimum angle of orientation of the fibers to avoid bursting can be determined based upon the stiffness and strength properties of both the polypropylene tube and the PET fibers so as to provide the required strength ratio in the hoop and in the axial directions. The analysis shows such an angle to be about 60 degrees from the axis of the cylindrical container.

Test Set C was developed to improve fiber winding techniques and provide a better set of specimens for long term creep test, and seaming tests and adhesion tests.

Specimens of surface coverage of PET fibers of 0-71% per layer were tested in Set B described in Table 4, while specimens of 37-100% coverage per layer were tested in Set C, described in Table 5. All test specimens in both sets B and C were constructed by winding two layers of fiber about the core in a balanced fashion, with winding angles of +600 and -600 with respect to the axis.

t 12 It was observed that 136 strands of fiber covered 100% of the core. Surface coverage per layer was measured by counting the number of strands on a specimen and dividing by 136. Fiber layer thicknesses were measured by caliphers and found to be 0.15 to 0.18 mm (6 to 7 mils) for two layers for the high coverage cases. (See Tables 4 and Fiber raw materials costs were estimated by weighing the amount of fiber on a specimen. A PET fiber price of US $1.40 per 0.453 kg (1 lb) was used.

Specimens in Test B were subject to static burst after completing a 4-hour creep test at 43 0 C (110 Burst pressure equaled or exceeded 0.69 MPa gauge (100 psig) for o all specimens tested of surface coverage of 25% per layer or 4. greater (See Table Burst pressure for the 71% coverage per layer specimen exceeded 2.07 MPa gauge (300 psig).

II One sample of 53% fiber coverage per layer from test set B and three samples 54.5%, 72% and 100% fiber f coverage per layer were subjected to long term creep tests for 18 weeks at room temperature and 0.38 MPa gauge psig) internal nitrogen pressure. Samples exhibited between 0-4% volume swell after an equilibration period of several i ,minutes. Growth was in the radial rather than the axial direction. Swelling in the equilibration period was 1-2%.

This swelling may occur during filling of containers and should be taken into account in can design.

It is important that excellent resistance to t swelling is exhibited by samples of lower fiber coverage as well as higher fiber coverage. This illustrates the surprising capability of the fibers to inhibit creep of the thermoplastic core even when not in direct contact at every location on the surface.

Specimens from sets B and C were subject to short 0 0 term creep tests at 43 C and 0.69 MPa gauge (110 F and 100 psig). Volumetric swelling was noted over three time periods, zero to 5 seconds, 5 seconds to 5 minutes, and ft 13 minutes to 4 hours. The "Pressurization" time period represents pressurization to 0.69 MPa gauge (100 psig), a second "Early Period" to establish temperature equilibrium, and finally a "Terminal Period" of long duration.

A summary of test results show the following ranges of swelling% Time Period Volumetric Swell 0.12% Cover agc Per Layer Per Pressurization (0-5 sec) Bur Early Period (5 sec-5 min) Terminal Period (5 min-4 hr) Lays st 36-53% 70-90% Coverage Coverage ~r Per Layer 4-5% 7-14% 2-11% 0-3% 1k

I

S.

Rejection of one specific data point in the Early Period for the 36% per layer coverage specimen would reduce the volumetric swell reported herein for the Terminal Period. Results for each specimen are for the Terminal Period and are presented in Tables 4 and Initial volume swell is a function of fiber coverage, and can be controlled through it, but surprisingly, long duration volume swell is low even for a structure with as little as 53% fiber coverage per layer.

The volume swell at 35% fiber coverage may be reduced further by either wrapping some of the fibers in an axial direction, providing a protective layer, such as an adhesive, over or under the fibers, use of an outer wrap impermeable to water, or use of a fiber grade with improved mechanical properties. Pig. 6 is a graph of burst pressure versus PET fiber coverage and supports a claim of 20% coverage as a lower limit of acceptability. A burst pressure requirement is established for the tested specimens of 0.6 MPa gauge at 35 0 C (87 psig at 95OF) equivalent to aluminum lid design requirements. Interpolation of test data suggests a fiber coverage of about 20% per layer is 2I -14 r: needed to achieve this pressure requirement. Note that tent conditions are severe, 43 0 C compared to the required 35 0

C.

Note that the specifications for a commercial generic PET bottle rquire adequate performance at only 35 0

C

rather than 43 0 C although the time period is longer. It hacn been found in practice that the liquid temperature in a container seldom exceeds 35 0 C because of the long time required for warming even when exposed to very high amblenL temperatures. Specimens with fiber coverage of 70-90% exhibited volumetric swell no greater than 6% after pressurization to 0.69 MPa gauge at 43 0 C the very severro conditions of testing.

Photomicrographs of 64x magnification were made of the surface of the polypropylene core after the short term creep tests were completed for specimens of 37% and coverage per layer. only in the specimen of lower coverage could the indentations created by previous contact with PflT fibers be observed.

The lack of indentations of fibers into the polypropylene core for the higher surface coverage caze suggests that creep of polypropylene relative to the PET fibers did not occur.

In conclusion, test specimens in which PMT fibr~r coverage per layer was as low as 35% exhibited burst 25 pressures over 1.21 MPa gauge (175 psig) after immersion in water at 43 0 C for four hours. This is well in excess of ths vapor pressure of 0.6 MPa gauge of carbonated beverage at 0 C and in excess of the burst pressure of 0.28 MPa gauge pslg) of specimens uncovered by PET fiber. Long term creep tests run for .18 weeks at room temperature and 0.38 MPa gauge (55 psig) pressure with specimens of 53-100% fiber coverage showed that volume swell of 1-4% occurred after a brief equilibration period, with no apparent correlation with fiber coverage. Short term creep tests run for four hours at severe conditions of 43 0 C and 0.69 MPa gauge showed if

I

'I

I

7 2 P 15 volume swell of 0-3% for specimens of 70-90% coverage per layer and 2-11% for specimens of 35-53% coverage per layer after a five minute equilibration period.

,t t) I UI 0 C IU A B rr cs *r r r a a a h~ll- Material Thickness Glass

PET

Fiber Type TABLE 2 DESCRIPTION OF FIBERS AVAILABLE FOR TESTING Standard Filament S Size Diameter Cylii yd/lb denier tex (mils) mm Calci trand Dimensions ndrical Diameter ulated Measured s) mm (mils) mm 5)0.75 (mil; (29.

Layer (when wound) (mils) mm 0.23 OCF type 30 P346BX675 PPG ECG75 Allied 1W74 Allied 21 strand material a Allied 5 strand materiala Celanese 785 Seat Belt Fiber Allied 1W69-LO-500 675 6622 736 (0.525)0.013 7,500 4,470 576 1000 64 (0.375)0.0096 ll (1)0.0254 (9)0.23 (15.9)0.40 (73)1.85 (10.1)0.26 212 21,000 2333 850 5,000 556 (1.4)0.036 (1)0.0254 (10)0.25 (36)0.91 5,320 894 840 93.3 500 55.6 (14.5)0.37 (11.2)0.28 (2.5-4)0.06-0.1 a) These materials were created by Allied from the 1W74 base material.

I 1 Ar.

b* a.

ro 01 0c 0 0 O* 000 00 O C O 0 '0 00 0 C 0 CJ o o O o

BALANC

FIBER WRAPS LAYERS

TA]

FIRST SCREEN FIBER

FIBER

ED BAND a)

THICKNESS

(mils) mm (6)0.15 (6)0.15 (12)0.30 (12)0.30 (12)0.30 (12)0.30 BLE 3 TEST RESULTS (SET A) Glass 1 1 2 2 2 2 no no yes yes no no ADHESIVEb) yes no no yes yes no no yes

QUICK

BURST

Pass? (Psig) yes (162) yes (152) yes (172) yes (147) yes (167) yes (200) yes (162) yes (117) MPa gauge 1.12 1.05 1.19 1.01 1.15 1.38 1.12 0.81

PET

Balanced means each wrap at angle. An 890 angle was used in tests.

Adhesive was Shell Kraton 1107 pressure sensitive.

The is a 0.05 mm (2 mil) outer surface of Hercules T-503 OPP film on all specimens for handling purposes.

I

TABLE 4 FIBER TEST PROGRAM (SET B)

SURFACE

COVERAGE

PER LAYER (All specimens

WINDING

STRANDS PASSES PER PASS PER PER LAYER LAYER consist of

FIBER

THICKNESS

(TWO LAYER (mils) mm balanced layers ot 93

STATIS

S) BURST (psig) MPa' gauge

A--

.3 tex PET

SHORT

TERM b) SWELLb) AT 43°C fiber wound at

FIBER

RAW MATERIAL

COST

U.S. dcents

/CAN

60 60' LONG TERM SWELL AT

ROOM

TEMPERATUREc) (6-7)0.15-0.18 (>300) >2.07 (6)0.15 (>200) >1.38 (3-4)0.076-0.1 (4)0.1 (175) 1.21 0.1-0.13 (100) 0.69 <0.28 0-3% 2-5%

HIGH

H IGH (burst) 1.50 1.12 0.75 0.75 0.75 0.54 0.22 3-4% <0.28 After creep test at 43 0 C (110 F) Change in volume between 5 minutes and 4 hours Change in volume between 5 minutes and 18 weeks Bases $1.40/0.453 kg price for PET fiber

J

34

_A

rr rrr 31 C L a

C.

TABLE FIBER TEST PROGRAM (SET C) (All specimens consist of balanced layers of 93.3 tex PET fiber wound at 600)

SURFACE

COVERAGE

PER LAYER

STRANDS

PER PASS PER LAYER

WINDING

PASSES

PER

LAYER

FIBER

THICKNESS

(TWO LAYERS) (mils) mm

SHORT

TERM

SWEL

1

AT

43 0

C

a LONG TERM SWELL AT

ROOM

TEMPERATUREb)

FIBER

RAW MATERIAL

COST

U.S. cents

/CAN

100 72 54.5 37 (7.5)0.19 (6.5)0.165 (6.0)0.15 (3.5)0.09 0.1% 0-1% 4-11% 2-3% 1-3% 1-3% 2.00 1.88 1.43 1.00 0.7 Change in Change in Basis PET volume between 5 minutes and 4 hours volume between 5 minutes and 18 weeks fibers at U.S. $1.40/0.453 kg r

Claims (8)

1. A beverage can for carbonated beverages which comprises a can wall made of a thermoplastic core and at least one fiber wound adhesive layer on said thermoplastic core, the can wall comprising a barrier layer between said thermoplastic core and said at least one fiber adhesive wound layer, or contained within said thermoplastic core or between adjacent fiber adhesive wound layers.

2. A beverage can according to claim 1, wherein the thermoplastic core is of polypropylene, polyethylene terephthlate, paper, polyethylene or a polyketone and is either spirally wound or extruded.

3. A beverage can according to claim 1 or 2 wherein said fiber adhesive wound layer is of glass, polyethylene terephthlate, nylon, rayon or cotton.

4. A beverage can according to claim 1, 2 or 3 wherein said barrier layer is a vinylidene chloride copolymer or a hydrolyzed copolymer of vinyl acetate and ethylene.

A beverage can according to claim 1, wherein said thermoplastic core is of polypropylene, the fiber adhesive wound layer is of polyethylene terephthalate and wherein the barrier layer is of a hydrolyzed copolymer of vinyl acetate and ethylene.

6. A beverage can according to any one of the preceding claims wherein said'core is extruded.

7. A beverage can according to any one of the preceding claims wherein each fiber wound adhesive layer provides at least 20% coverage.

8. A beverage can according to claim 1 substantially as described herein. i t i i K, ~21 ii Li 11 iv I t t4.l -21- described with reference to the dr w-'s or examples. The hte described invention in all its Dated this 11th day of AUGUST, 1986 SHELL OIL COM4PANY By Its Patent Attorneys DAVIES COLLISON C' t A a I.aq a *14