Vacuum Web Coating

Making changes – think carefully first!

    It is very easy to make changes to the vacuum deposition process but the results of the changes may well be different than what we expected.  The reason for this is that we are changing more things than just the one parameter we changed.  Sometimes these additional changes are hidden changes and other time we have to intervene and make the additional changes.

 

    Imagine wanting to increase the deposition rate from one resistance heated boat and choosing to increase the current to the boat.  This will increase the boat temperature and if there is a stable molten pool present in the boat this will evaporate at a faster deposition rate. However this molten pool will then start to shrink because the wire feed rate no longer is matched to the evaporation rate. Thus intervention is required to increase the wire feed to match the increase in boat temperature and increased evaporation rate.  Other changes that are unseen but are also present are the heat load to the substrate will change. The deposition rate will increase, as was desired, which means there will be more material condensing on the web and so the latent heat of condensation will increase. Also the increased temperature of the boat will increase the radiant heat load to the web too.  If there is more aluminium vapour being produced the vapour as it deposits will be able to getter more oxygen and so it is also possible the system pressure will change. If the change in temperature is small the change in gettering rate compared to the overall gettering rate of the whole set of deposition sources will probably make the change in system pressure so small that it will not be seen. If all the boats are changed in temperature at the same time and by are large change then a small pressure change may become visible.  The increase in radiant heat from the higher temperature source will also increase the heating of the chamber walls and other system furniture and this can increase the rate of water vapour outgassing.  Again a small increase in temperature of one boat will be a very small overall change in heat load to the system and so the outgassing rate will be a very small change and again probably be insignificant.

 

    I hope that this example has highlighted how changing one parameter may require some other intervention because of the change and also how other things will also change even if they may be unseen.

    In the above example the one unseen change that can be significant is the increase in heat load to the web.  As most metallizers are working at their maximum throughput a small increase in heat load can tip the process over the edge and result in a greater chance of wrinkles being produced.

 

    Thus if you are considering process changes it becomes important to consider what effect those changes might produce both seen and unseen.  Doing this thinking before making any changes can result in making planned other changes that can counteract the effects of the changes without waiting for the process to go wrong.   This might be as simple as increasing the drum cooling capacity to remove more heat from the web and so allowing the boats to be run hotter. 

 

    It is always good policy to think first rather than problem solving afterwards.

Response to question re. pinholes

Question

As far as the source of the pinholes concerns, besides of having a visual inspection and comparison of the appearance of the pinholes, what other type of analysis could be considered in order to be able to understand and therefore eliminate the problem? What would be the recommendation to address a pinhole problem that is suspected to come from the base film itself and that is revealed by the metallization process?

 

 

Answer

 

As the bulk of pinholes are caused by vacuum coating onto substrates covered in debris that is subsequently moved leaving behind a pinhole the best thing to measure is the quality of the web before it is coated.

 

The difficulty can be that the size of the debris is small, often beyond the resolution of the eye and so it can only be seen under a microscope. As it is difficult to examine under a microscope representative parts of the web surface without damage to the film an alternative is needed.

 

Teknek offer a test kit (it used to be free of charge) where you pass a tacky roll down the web for a predetermined distance (1m or 5m etc) and then clean the tacky roll on a higher tack sheet. The higher tack sheet can then be examined under a microscope and the amount of debris can be measured.  This can be done manually or in some cases can be automated using image analysis software.   Once you start measuring the debris levels before metallizing you can then relate these measurements to the measurement of pinholes after metallizing.  There should be a correlation between higher levels of debris before and higher levels of pinholes after.  There can sometimes be differences as there is also a size component. Bigger debris can be more easily moved after metallization than smaller sized debris and so more bigger debris are more likely to result in pinholes than a larger amount of smaller debris.

 

Once you have built up a correlation between the cleanliness of the web before metallizing and pinholes after metallization you then have the decision to make of whether to clean the web or not.

 

Teknek give these test kit away in the expectation that you will want to clean the roll by using one of their tacky roll cleaning systems which can remove debris down to a size of approximately 0.3 microns.

 

If you decide to clean the web then you can use the test kit to look at the quality of the web before and after cleaning using the tacky roll, to confirm the benefit of cleaning. Also you can also look at the pinholes following metallization and it would be expected that the number of pinholes would be reduced.

 

There are other sources of pinholes or other mechanisms that can increase the number of pinholes even for the same level of debris contamination.  Winding the film hot or with too much tension can lead to blocking or if not quite blocking it can cause movement of web layers against adjacent layers which will increase the amount of pinholes. High tension can also lead to pickoff pinholes where any filler particles protruding from the back surface of the web can be pressed into the metal layer and may pick off the metal if the adhesion is not high and this too leaves behind a pinhole.  Within the deposition process a high level of oxidation of the aluminium feed wire and an unstable molten pool size can lead to a higher level of spitting and this too may be seen as pinholes in the coating.

 

All of these can lead to errors in the measurement of pinholes and so it is only by the accumulation of results that will show accurate trends in substrate quality and final pinhole count.

 

Hence I would spend extra time examining and improving the incoming substrate material in order to improve the metallized product.

 

I hope you find this useful.

Pinholes

We came across one question answered by you ( dtd.Feb.2, 2009. "How are these pin holes formed ?")  on internet regarding the pin holes in metallized PET film.

 

Is that pin holes are like sieve or unmetallized / metal removed portion of the film?

 

How to detect the same is only metal removed area or see through hole.

 

Pl. advice.

 

 

Answer

Pin holes or Pin windows is a catch-all term that is used to describe small areas that at first sight appear to have no metal coating present.  On occasions it is possible to do some analysis on these areas and come up with some information about how they were formed.

 

If there was a particle of debris that was metallized and then removed after metallisation there is usually an area of uncoated polymer were the particle was positioned which has a small area of thinner metallisation surrounding this uncoated area.  This thinner metal is due to the debris blocking the metal flux for part of the time in the deposition zone.  As the debris first enters the deposition zone the area behind the debris is shadowed from the metal flux but as the debris traverses the zone this shadowing becomes less until the centre point where there is minimum shadowing. After the mid point the shadowing then is on the opposite side of the debris and this gets larger until the debris exits the deposition zone.

 

If the debris was not rolled away but slid along there would be an uncoated area that then had a scratch away from the pinhole.  The scratch may have the metal completely removed or may just have the metal thinned.

 

By comparison a pinhole formed by pickoff is likely to have a pinhole with no graded metal thickness but generally has a step change in the metal thickness with either the full metal thickness or zero metal thickness where the metal has been removed.

 

If the pinhole is due to spitting there is likely to be a small particle left at the centre of the pinhole and also some heat damage to the polymer where the particle has impacted.

 

 

 

When the metallized film is held up to a strong light there may be many more pinholes visible than if the film is held up to a weaker light.  The reason for this is that a number of the pinholes may be not really holes in the metal coating but simply areas of thinner metallisation that produce a significant difference in contrast.  Areas where there is residual oligomer on the surface can result in these thinner coated areas.  The oligomer can have a significantly lower melting temperature than the polymer. The oligomer can be heated by the evaporation source and the oligomer may be evaporated by the heat and as the oligomer evaporates it prevents the aluminium form depositing. If there is a lot of oligomer this may prevent all the aluminium from depositing or if there is a lower amount of oligomer then once the oligomer has evaporated there may be some metal deposited but thinner than the rest of the metal deposited on the clean surface.

 

 

With respect to the final performance of the coated polymer then all holes with no metal deposited will have no barrier performance whatsoever. It does not matter how the holes were formed the result is the same, no metal results in no barrier.   Where the metal is thinner the barrier performance will be better than where there is no metal but less than where there is full metal.

 

 

The number of holes is dependant upon the amount of debris on the substrate before starting to metallize the web, the amount of oligomer or other low molecular weight material is present on the web surface, as well as being dependant upon the quality of adhesion and web surface roughness along with other factors such as roll tension and handling quality.

 

 

I hope this gives you a more complete knowledge of pinholes.  

 

Something to look forward to - Converting Quarterly

PRESS RELEASE August 23, 2010 FOR MORE INFORMATION, CONTACT: AIMCAL OFFICE: Tel: 803-802-7820; Fax: 803-802-7821 E-mail: aimcal@aimcal.org; Website: www.aimcal.org

AIMCAL Launches CONVERTING QUARTERLY New technical publication to serve the information needs of web-processing, converting and finishing professionals   Website, E-Newsletter, Social Media Sites to extend new property's technical-information resources to the industry on a 24/7 basis August 23, 2010 - Fort Mill, South Carolina, USA - The Association of International Metallizers, Coaters and Laminators (AIMCAL) today officially launches CONVERTING QUARTERLY, a new technical publication and Website for the web-processing, converting and finishing industry.  The first print issue of CONVERTING QUARTERLY, scheduled to mail in February 2011, will have a circulation of 15,000. As the official magazine of AIMCAL, CONVERTING QUARTERLY will be distributed to converting professionals engaged in the web-processing technologies of solution, extrusion and vacuum coating; laminating; film manufacturing; flexographic and gravure printing; slitting; sheeting and other finishing methods.  Each issue will include several technical Q&A columns and technical-journal articles on web-handling, coating/laminating and finishing processes and products.   "Publishing a technical journal represents a new level of service AIMCAL is providing to its members and the converting industry at large," says Craig Sheppard, executive director of AIMCAL.  "Technical information is always in demand."   "The heart of the magazine will be eight to 12 technical articles covering web handling, web coating, vacuum coating, laminating, printing, slitting and rewinding, and finishing (diecutting, perforating, stamping)," says Mark Spaulding, who has been named associate publisher and editor-in-chief.    "Each article will focus on practical applications of technology with heavy emphasis on problem-solving and troubleshooting," he explains.   Other editorial content consists of a market article, a section on new machines and materials and four to six columns authored by AIMCAL technical consultants and Blogmasters, who will write about technology, methods, new product development and troubleshooting in their specific area of expertise.  The roster of columnists currently includes Dr. Charles Bishop (vacuum web coating), Dr. Edward Cohen (solution coating), Dr. Eldridge M. Mount (extrusion-coating, film manufacturing and vacuum metallizing) and Dr. David R. Roisum (web handling).   "We'll be accepting submissions from converters and the supplier community," says Spaulding. "However," he notes, "content must be technical and noncommercial in nature.  The quarterly journal also will carry advertising."   The print publication will maintain a Web presence at www.convertingquarterly.com, which also will be accessible as a channel from the newly redesigned AIMCAL Website, www.aimcal.org.  A weekly e-Newsletter- CONVERTING QUARTERLY E-NEWS-also is planned and will offer sponsorship opportunities.  It will email on Tuesdays to an initial audience of 7,500 readers and consist of abstracts of up to eight converting industry news stories with Website links to full articles, plus AIMCAL Blog posts, including Spaulding's popular Converting Curmudgeon Blog.  "AIMCAL's existing weekly e-Newsletter will continue with a focus on association and member news," reports Sheppard.    Editorial leadership will be provided by Spaulding, the long-time editor of Reed Business Information's CONVERTING magazine, which ceased publication in April 2010.  A packaging/converting-industry journalist for 28 years, Spaulding's publishing experience includes various roles on PREPARED FOODS and PACKAGING magazines, as well as 15 years as chief editor of CONVERTING.  A frequent conference speaker and moderator, he has addressed a wide range of topics including consumer packaging, coating/laminating, digital printing and flexible-packaging production.  Among his accolades, Spaulding has received the Cahners Medal of Excellence for News Reporting, as well as Regional Awards for News and Cover Design from the American Society of Business Press Editors.  An active member of AIMCAL, Spaulding has judged its Product of the Year competition many times.   Peterson Publications, Topeka, Kansas, a specialist in niche market magazines, is providing advertising, circulation, design, production, printing and mailing support to CONVERTING QUARTERLY.  "This is a very exciting opportunity for us to work with both Craig Sheppard (AIMCAL) and Mark Spaulding (CONVERTING QUARTERLY editor-in-chief)," says Jeff Peterson, president of Peterson Publications.  "We believe we have assembled a team that has both the talent and expertise to create a technical publication with a pleasing visual look, which will fill a significant void within the converting industry."   Cindi Stocker, vice president of sales at Peterson, will represent the magazine for print and online advertising sales.  With more than 30 years of experience in sales and marketing, she has a vast background with publication, outdoor and other media sales.  Her current experience includes the development and coordination of advertising sales for Peterson's graphic arts, manufacturing and travel industry publications both on a national and international basis. Editorial materials should be submitted to the AIMCAL office, aimcal@aimcal.org or Mark Spaulding, 262-697-0525; mark@aimcal.org     Free subscriptions to both print and digital editions of CONVERTING QUARTERLY as well as the weekly CONVERTING QUARTERLY E-NEWS are available to qualified readers.  For more information and to subscribe, visit www.convertingquarterly.com   About AIMCAL AIMCAL represents converters of metallized, coated or laminated flexible substrates and their suppliers.  The international nonprofit trade association also includes a CEMA Division focused on the interests of converting equipment manufacturers.  The group is headquartered in Fort Mill, South Carolina, USA.  Membership information is available online at www.aimcal.org.  New members that join during the third or fourth quarter, receive membership privileges for the remainder of 2010, as well as all of 2011.    About PETERSON PUBLICATIONS Peterson Publications, Inc. has been contracted to handle the publishing of CONVERTING QUARTERLY, including all advertising sales (both print and online), layout and design, editorial assistance, and printing/mailing.  Peterson Publications has more than 15 years of publishing experience and specializes in niche association-based magazines.  Most of its publications are within the industrial B2B marketplace, working with converting-related industries such as graphic arts and plastics.  Cindi Stocker can be reached at 785-271-5801; Fax: 785-271-6404; email: cindi@petersonpublications.com

Answer to questions on Cold Cap & position of cryocoils

We have a Varian make diffusion pump nhs-35. the pumping speed is 50000 ltr/sec. in which we supply the water of 22 deg celcius in the cold cap of diffusion. If go for down temp. in minus side there is any effect in pumping speed of pump.

 

 

The purpose of the cold cap is to help prevent backstreaming of any diffusion pump oil It should not affect the pumping speed.  The way the cold cap works is that any oil mist hitting the cold surface will condense and drip back into the diffusion pump. In this way it prevents the oil mist being knocked back into the vacuum system and also the oil dripping back means that the loss of oil is reduced.

 

Reducing the temperature should have little or no effect as the oil will already have a high sticking coefficient and the temperature is already low enough to convert the oil vapour to a liquid and so reducing the temperature further will not change how much oil is collected.  The oil vapour undergoes random collisions with the gas that is being pumped and that is not changed by the temperature of the cold cap so the amount of oil hitting the cold cap is purely due to the pressure and statistics.

 

 

 

We know that poly cold coils in the mouth of diffusion pump improving the pumping speed but if the coil in opposite side of evacuation of the chamber what effect on the pumping time or not any effect. please giving the suitable answer

 

The cryosurface will work wherever it is sited within the chamber but there it is possible to improve the system performance if they are sited well.

 

To some extent it depends on the system design and the space available to site the cryosurfaces.  Many metallizers have the diffusion pumps hung off one end of the system and this creates a pressure difference from one end of the vessel to the other. If this is the case then the cryosurfaces will see different amounts of gas at one end compared to the other and so the effect will be slightly different.

 

Placing the cryosurfaces over the pumping orifice does ensure that vapour passing through collects water vapour well but it also provides an optical baffle and so can reduce the pumping speed of the diffusion pump too. As the bulk of the gas in the system will be water vapour and the pumping speed of the cryosurface for water vapour is enormous this will not be noticeable as the addition of the cryosurfaces is greater than any loss of pumping from the diffusion pump.  

 

 

The ideal is to distribute the cryosurfaces uniformly across the width of the web. The best performance is also if the cryosurfaces is split with some in the winding zone and some in the deposition zone. This can lead to a lower pressure during deposition and so the quality of the coating in terms of reflectivity can be improved.  This was reported on in a paper by Stenhouse at the AIMCAL Conference back in 2003.

 

I hope these answers are clear and help.

Converting School - European courses in October.

CONVERTINGSCHOOL

Web Coating and Drying (Europe)
Dr. Jonathan Summers
Dr. Harvey Thompson

October 5 - 6, Brussels, Belgium

Film web coating machines are used in a variety of high technology, versatile processes, which can effectively manufacture a wide range of coated products from packaging films, speciality papers to adhesive and cosmetics.

This Web Coating and Drying course is designed to provide the technical understanding required to develop new film web coating products and optimise the conditions of existing products. The instruction will make use of specialist software that attendees can install onto their laptops and use in the practical exercises.

For course outline and description as well as registration, visit the website.
Who will benefit from this course?
The course should benefit a wide range of people working in the film coating and drying fields, including product formulation and development scientists, product scale-up personnel, coating operators, managers seeking an understanding of film web coating and drying processes, and new personnel wishing to become familiar with film coating and drying.

Hotel Information:
Mercure Louvain-la-Neuve
Boulevard de Lauzelle, 61
1348 Louvain-La-Neuve
Belgium
Phone: +32 (0) 10 / 45 07 51
Room Rate Expiration: October 1, 2010
To receive discount room rate, reference group code: AIMCAL

Drives in Web Handling and Converting (Europe)
Clarence Klassen

October 7 - 8, Brussels, Belgium

Web Handling including Winders are among the most complex of drive applications. In this Drives in Web Handling and Converting course learn about the importance of tension control to winding quality rolls. There are a number of options available for tension control. The course will be general and applicable to any brand of drive. Limited high school mathematics will be presented, but emphasis will be on description of the controls. We will not touch on university level control theory (Bode, Nyquist, etc.) in this course.

For course outline and description as well as registration, visit the website.
Who will benefit from this course?
This course is intended for electrical/instrumentation maintenance, drive startup personnel, and junior to intermediate drive systems engineers. Emphasis will be on the practical aspects of drives for winding. A prerequisite is electrical aptitude and enough troubleshooting experience to be familiar with the names for the: parts of your machines, and electrical terms like volts, amps, 3 phase, parameters.

Go to www.convertingschool.com

or

www.aimcal.org

if you want to attend either course.

Why polymer films can differ & hence may respond differently to the same surface treatment

When monomers are polymerised into polymers it is possible for some to add monomer to extend the polymer chain length in different ways. For example in polypropylene the methyl groups can be added on one side of the polymer chain only, known as isotactic polypropylene, or the methyl groups can be added alternating between two sides of the chain, known as syndiotactic polypropylene, or if the methyl groups are added randomly on either side this is known as atactic polypropylene.  Now the two regular versions can both allow polymer chains to pack together well and form crystallites whereas the atactic version does not allow easy packing and so is primarily amorphous.  The aim is to get large amounts of isotactic polypropylene and a high degree of crystallinity and to achieve this polymerisation catalyst is used.  There are various catalysts that can be used and these can affect the chain length, molecular weight, transparency, rigidity and toughness of the polymer.  The temperature and speed of the process can also affect a number of these properties too.

 

Hence we can see that the surface of the polypropylene there can be different proportions of crystalline and amorphous polymer along with some residual monomer. The precise proportions of each will depend on the processing conditions and catalyst used.  In addition there can be other additives used to improve the performance of the polymer such as anti-oxidants, acid scavengers that are used to counteract the acid produced from some catalysts, optical brighteners, slip agents, etc.

 

So we can see that chemically even with a simple polypropylene where the additives are kept at a minimum it is possible that different films will have different proportions of crystalline material of either isotactic or syndiotactic form as well as regions of amorphous polypropylene from the atactic form of polypropylene and some residual monomer that is unpolymerised.

 

If we then add on to this the possibility that the polymer film has been pre-treated where the pre-treatment can include coatings, or surface treatments such as flame, corona or atmospheric plasma treatment or more than one of these pre-treatments the precise chemical composition of the surface can have huge chemical differences.  A coating can completely change the chemistry or one of the plasma based treatments can take the surface and modify it both physically and chemically. This usually means roughening the surface and increasing the oxygen content to improve adhesion. 

 

Thus it is no surprise that a vacuum plasma treatment cannot be standardised so that one pressure, gas mixture and power setting can be expected to improve, or optimise, the adhesion of every polymer surface.   

Add to this the variations in plasma treatment units both in physical dimensions as well as magnetic field strength and positioning of the web and the range of possible treatment becomes even larger.

 

This is why I cannot advise a specific power, gas mixture, pressure to optimise a treatment fro any particular film substrate.  I should always be optimised on each vacuum system and for each specific film.  This is not to say that once done for a variety of different substrates you may well find that many of the optimal settings are very similar but beware of falling into the trap of then assuming that the settings will work for every new substrate.

 

Question re. Decline in metal coating surface energy

Q.

As you know, dyne level on metal side of BOPP metallized film drop very fast within a few days. Do you think dyne level reduction on BOPP film before metallized will help to maintain dyne retention after metallization still high ?. For example, if metallization conducted on BOPP with 40 dyne, dyne retention on metal side will be better than if metallization conducted on BOPP with 44 dyne ?.

 

Answer

The dyne level is measured as a way of indicating the chemical cleanliness of surface.  There will be a dyne level that indicates the surface is clean.  If the dyne level is below this clean level it is an indication that the surface has something contaminating the surface such as low molecular chain length polymer fragments (oligomers) or slip agents.  If the dyne level is above this clean level it is an indication that the polymer chains have undergone chain scission and carbon bonds replaced with oxygen which increases the surface energy.

 

If the surface is plasma treated before metallization it is important that this is continued to be done as this directly affects the metal adhesion to the substrate and any loss of adhesion may cause problems further downstream.

 

Once the metal has been deposited the surface energy of the metal will be around 900 dynes and this will be wound up into a roll and the metal surface will be in direct contact with the reverse side of the roll at whatever surface energy that is at.  The speed with which the metal surface energy falls will be dependant upon the amount of low molecular weight material that remains on the back surface and the mobility of that material.  The shorter the chain length or lower the molecular weight the more mobile the material and the higher the temperature of the roll the faster the migration of material.  So the speed with which the metal surface energy will fall depends on the cleanliness of the surface and the temperature.  If the back surface energy is lower that the clean level then there will be contamination present. Even if the surface energy is at the clean level there will still be oligomer in the bulk polymer and this will still migrate to the surface over time and so ALL metal surfaces will have a surface energy that will fall over time. However it could take slightly longer if the starting point was from a clean surface rather than a contaminated lower energy surface.

 

I hope this clarifies the situation.    

Oil and vacuum systems.

As there are a number of moving parts inside any roll to roll vacuum coater it is important that bearings are lubricated. Similarly of you are using rotary vane and diffusion pumps an essential part of the pumping system is the oil.  However from the standpoint of adhesion of any coating oil is always seen as a potential problem, where the slightest amount of contamination, i.e. one molecular layer and above, results in a loss of adhesion.  Thus it becomes important that any possible source of contamination is minimised.

 

Bearings may be supplied with grease that is not really vacuum compatible and so often bearings start be being cleaned of existing grease and a small amount of vacuum grease added instead. Where sealed bearings are used the grease added may be more than the minimum as the seals will be replaced and so the grease contained.

 

With pumps there is the basic careful use of how the pumps are used such as not trying to pump beyond the pumps capability. Pumping to a low pressure can give more time for oil backstreaming. A low backstreaming rate is good but if the system is left at the low pressure for a long time the accumulation of oil even at a low rate becomes much more noticeable. Hence it is always good policy not to hold systems at a low pressure for too long. If the system has to idle for some time it is better and cleaner to give the system some gas load to pump against where this continuous gas flow in and pumping out sweeps the system clean.

 

Included in the good practice for oil based pumps is maintenance of the oil. Rotary pumps can easily take up any moisture and the oil can change from a transparent pale yellow to a translucent cream liquid where the water has been emulsified and the light scattering changes the liquid from transparent to translucent. This only requires the pump to have the gas ballast used periodically to keep the oil free of moisture.

 

Pumps also work best if the oil level is maintained between the minimum and maximum levels.  I have seen a diffusion pumped system that was still operating with only a fraction of the required oil. The pump down time was slower than it should have been and the tolerance to gas load variations was poor but it would still pump the system down and allow production.  Once the level dropped further the pump then failed to pump the system down at all at which point after various checks it was discovered the it was simple a lack of routine maintenance and checking of oil levels that was the source of the lower performance and then failure to pump. The loss of production time to troubleshoot the problem was greater than the time needed in routine maintenance to check the oil level and keep it within the required tolerance.  

 

The choice of oil can also become important. Oil can be recycled which is can be good practice if the recycling is done well. However some oils end up mixed with other oils and the final recycled oil may have more than just the required oil present. What this means in practice is that when the oil is heated, particularly for the first time, some of the lower boiling point oils can be driven off and can cause a backstreaming problem.  Once they have all been boiled off the remaining oil may well act in the same way as good quality new oil but the inside of the system may well have been contaminated by then.  Thus if you are tempted to use lower cost recycled oils it is worth checking the quality and reproducibility of them to make sure you are not introducing a contamination problem to your process.  

Evaluating and improving adhesion

I often get asked about improving adhesion for specific films.  It really does not matter what type of substrate is used the approach to evaluating and improving adhesion is the same. 

 

The substrate surface can vary between clean to contaminated.  The chemistry of the surface can vary between inert to reactive. The surface can vary from perfectly smooth to rough.

 

If the surface is contaminated it needs to be cleaned. The cleaning needs to take account of the type of contamination. The contamination may be from additives, unpolymerised monomer or polymer fragments as well as powder, dust and atmospheric debris.

 

The cleaning of the gross contamination then enables the chemical modification of the surface where the polymer chains can be broken to provide additional bonding sites where oxygen can be attached and provide a bridging bond directly to the aluminium.

 

In terms of measuring the bonding strength this is more difficult. The best method that is currently being used is the fragmentation test where a sample is stretched until the coating cracking reaches a saturation density of cracks and then the adhesion strength can be calculated.  However this requires a tensile stage for use with a microscope and a skilled operator.  What can be more useful is to use fit-for-purpose tests.  Look at how the coated film is to be used and use tests that can mimic the use and evaluate how well the coating stands up to the test.  Thus if the coating is to be used for a package and needs to have a barrier performance then using a bending test in conjunction with a barrier performance test might be appropriate. Similarly if the film is to be twisted then using a Gelbo twist test and evaluating the coating for delamination or cracking might be enough.  If the coating is to be directly subjected to an adhesive and a shear or pull then the test should reflect this.

 

Improving the adhesion to the optimum may take some time whereas improving the adhesion sufficient to be fit-for-purpose may be easier to achieve. 

 

Ideally the optimum adhesion can be obtained and the one process will then suitable for the most products. Anything less than optimum and it may be necessary to have slight process variations depending on product.