Excessive Dyne Level Drop in High Slip PE Film

Question: We extrude and corona treat high-slip PE blown film. In a recent evaluation, the film went from a COF of .420 hot off the line to .188 five days later. The treat level went from over 50 to below 36. This seems like an excessive amount of treat loss. Any comments or suggestions?

Answer: That is an unusual amount of treatment loss, but on the other hand, trying to treat freshly blown PE to 50 dynes/cm in one step is pushing things – usually the film extruder will treat to about 38 to 44 dynes/cm, and the printer/converter will then bump treat as needed. The higher the treat level, the faster the treat loss, especially on high-slip films.

The degree to which the COF (and dyne level) dropped in five days suggests that you are using a traditional low molecular weight amide slip agent, such as erucamide or oleamide. These are intended to migrate quickly to the surface, especially on treated film, where the surface polarity attracts the slip molecules. Secondary amides such as oleyl palmitamide have approximately twice the molecular weight, and subsequently bloom less aggressively. They are also amenable to higher process temperatures without degradation, and tend to offer somewhat more stable results from corona treatment.

As a rule of thumb, films extruded from resin blends containing these compounds should condition for at least a day or two to allow the migration to take its course, with a corresponding COF reduction. After this conditioning period, blooming and COF reduction will continue, but at a much slower rate.

There is an alternative to these compounds: non-migratory slip agents which have a molecular weight 30 to 50 times that of the traditional amide formulations. These compounds are too massive to bloom to the surface, so their effect on COF is more or less immediate, and stable. As such, they will also have a vastly reduced effect on corona treatment loss. Another advantage is that they are stable at higher processing temperatures than are the amide-based slip agents.

Based on this information, my suggestion is twofold: Consider treating to a slightly lower dyne level initially, and investigate the feasibility of non-migratory slip agents, or at least of the secondary amide formulations.

Unusually High Dyne Level Results on Aluminum

Question: We have determined that a dyne level result of about 42 to 44 qualifies an aluminum surface for our bonding operations. We get this dyne level pretty consistently from our cleaning line, but recently we had a case where the surface wetted out all the way to 50 dynes/cm. It’s hard to imagine that the cleaning process was actually providing that clean of a surface. Any ideas on this?

Answer: For most aluminum alloys, a reading of 42 to 44 dynes/cm reflects a relatively clean surface, essentially free of oils or other problematic contaminants. So, your historical experience that this result predicts good adhesion is not surprising.

As to how you could end up with a reading considerably higher for no obvious reason, I have a few thoughts. First, is it possible the test fluids became contaminated from exposure to oils or other surface contaminants? If you had previously tested samples that failed at the usual dyne levels (in other words, were not clean), then re-used the same test markers – or somehow re-introduced contaminants into bottles of test fluid – their surface tension would be affected. Specifically, it would decrease due to incorporation of the low surface tension contaminant into the reagent. This would cause “higher” readings.

Second, did you perform the test starting at lower dyne levels, which wetted for considerably more than two seconds, then worked upwards until you determined the dyne level which started to bead up in just a couple of seconds? Based on your usual results, I would recommend starting at about 34 to 36 dynes/cm to ensure full wetting at the outset of the test. Finding the transition point from wetting to beading is important: if you start the test at too high a dyne level, spurious wetting can occur, which will invalidate results. This is why we stress that the test is finished as soon as you find the dyne level which beads up in 2 seconds or less. If this is the case, it suggests that your parts actually have a low surface energy, and will likely prove difficult to bond.

Finally, if your cleaning process uses solvents or surfactants (detergents), it is possible that the rinse cycle was ineffective. This would leave a film of low surface tension solvent or surfactant on the part, which could become solubilized in the dyne solutions, dropping their surface tension as soon as they contact the surface. This is the same effect as the first possibility noted above, but from a different cause. The way to test for this possibility would be to hand-rinse the parts and re-test. If the results return to your expected values, that would be an indication that the cleaning solution is not being entirely rinsed from the parts. In that case, the first thing I would look at is the purity of the rinse water. Distilled water has a surface tension of 72 dynes/cm at 20° C – if your rinse water has a lower surface tension, it is contaminated.

Effect of Surface Roughness on Dyne Testing

Question: We have a customer that uses your test markers to measure the surface energy of aluminum before bonding it to a composite. They find the test works great on a smooth surface, but on rougher surfaces they are unsure of results, and don’t think they are accurate. Is there a rule of thumb on how smooth a surface needs to be for surface energy testing?

Answer: I don’t know that you could really come up with a “rule” as far as roughness goes, but there is an obvious problem with textured surfaces: the test fluid will tend to settle into the “valleys” so wetting vs. beading will become more and more difficult to gauge as the roughness increases. This will be as true for plastics, composites, glass, and other materials as it is for metals. It’s important to keep in mind that the dyne test is based on the behavior of a retreating liquid/solid contact line. When this line is on anything other than a horizontal plane, gravity comes into play, either aiding or abetting the retreat of the liquid.

One thing is certain – the testing of this material should be done with bottled test fluids, applied as lightly as possible (in terms of both amount of fluid used and pressure applied) with a cotton swab. Test markers really will not be controllable enough, and will tend to flood the valleys with test fluid. Also, as testing of metals is usually performed to evaluate surface cleanliness, test markers are not a good option, as surface contaminants can affect results (for details on this, please see our discussion here).

Using a strong magnifier will be helpful – I’d look for signs that the fluid is creeping away from even the valleys, and tending to aggregate in micro-puddles rather than coating the entire low area. This discernment may be easiest at the perimeter of the test area. Also, while it may be rather interpretive rather than an absolute indicator, if the high spots on the surface retain a thin film of test fluid, that is a strong suggestion that wetting has been achieved.

Testing PET for the Presence of a Silicone Coating

Question: We convert silicone-coated PET, and test it to make sure we are working with the coated side. At present we’re testing with 30 dyne and 50 dyne inks. Any comments on this, and do you have a test procedure for this application?

Answer: First, the test procedure, which covers virtually all polymers, is available at https://www.accudynetest.com/qctest.html.

Untreated, uncoated PET generally has a surface energy of about 43 dynes/cm. As such, if the PET surface has not been modified, I would expect the 50 dyne/cm test fluid to bead more or less instantaneously when applied to the material. The 30 dyne/cm test fluid would wet out for a long period of time – permanently as likely as not.

For silicone-coated PET, it is obvious the 50 dyne/cm test fluid should bead instantly, as silicone compounds have surface energies ranging in the 20s. In most cases the 30 dyne/cm test fluid would do the same, but some formulations may result in a surface slightly higher than 30 dynes/cm, so some short-lived wetting may be observed.

I would suggest testing at 36 dynes/cm. Print-primed PET has a surface energy of about 38 dynes/cm; virgin PET is at 43, and surface-treated PET should be at 48 dynes/cm or higher. All these surfaces should show wetting for at least 4 seconds, and perhaps even permanently, when tested at 36 dynes/cm. The presence of any silicone compound would cause immediate beading at this dyne level.

Cleaning and Evaluating Drawdown Rods

Question: We use your wire-wound draw down bars here, and have a couple of questions on their care, and how to evaluate their condition. First, please let us know the recommended care and cleaning instructions. Would blasting with baking soda be safe and effective? Second, what is the best way to determine their overall condition?

Answer: Thanks for the question. The most important factor in keeping Mayer rods clean is to use an appropriate solvent immediately after every use. For non-aggressive, low viscosity fluids, simply wiping with a solvent-wetted soft, lint-free cloth may be all that is needed. In other cases, solvent immersion in an ultrasonic cleaning tank, along with brisk scrubbing with a very fine bristle brass brush, may be required. Please keep in mind that this abrasive method may cause burrs on the wire, which will affect subsequent coating performance.

It is important that the final rinse or wipe be done with a liquid such as water or high purity isopropyl alcohol, which will not leave a residue on the surface.

Once cleaned, it is imperative that metering rods be thoroughly dried before re-use, or the residual solvent may interact with the wet film coating when next used. Please note that compressed air – a fast, non-contact drying technique – generally contains trace amounts of oil, and should not be employed.

I don’t necessarily recommend blasting them with baking soda – or any other media – as even with stainless steel there is a potential for some degree of etching, which one would expect would be most pronounced at the “high” spots on the wire winding. This would reduce the area between the windings, with a concommitant reduction of coating thickness. However, if other methods have proven unsatisfactory, I suppose it would be worth a try, especially if the coating formulation is acidic. I would certainly want to check the resulting wet film thickness and finish after trying this! Other than those potential effects, the worst that is likely to happen is that the wire loses its weld or simply breaks and unwinds. If you try this, start with the largest wire sizes first.

As to evaluation of the rod’s condition, precise measurement of the wet film thickness applied would be most important . Evaluating the surface finish quality of the dried (or cured) coating would also be important. If you have a good microscope, that would be an effective inspection tool in this case.

Finally, as the cost of replacement rods is relatively modest, if there is any doubt about quality, it is probably best to simply replace the rods, rather than putting too much time and effort into trying to maintain them forever.

Test Marker Results Seem Inconclusive

Question: I used your test markers to test some film extruded from a masterbatch we are developing, and the results seemed inconclusive. It was hard to tell if the ink was beading or maintaining its integrity. Any comments?

Answer: If the test marker results seem inconclusive, it sounds like the surface energy of the film varies from spot to spot. There are several possible reasons for this.

First, if there are spots of contamination on the surface (fingerprints, transfer of contaminants from a process roll, contact of the sample with a foreign object, etc.), the contaminated spots will definitely have a lower surface energy, which would cause the swath of test ink you applied to show inconsistency in wetting vs. beading.

In the case of a polymer blend, or a polymer modified with multiple additives, it is possible that the blend is not being adequately dispersed in the extruder barrel, causing the various components of the formulation to segregate in the melt. In this case, concentrations of lower surface energy additives – or of lower surface energy polymers in a blend – will naturally bloom to the surface, creating “puddles” of low energy scattered across the film. Blends that include additives or modifiers with incompatible solubility coefficients could have the same problem: Even if they blend in the extruder barrel, they may disassociate before setting in the final film structure.

If the film is surface treated, it is possible that the treatment itself is contributing to the variation in dyne level. This case usually manifests in patterns that can be traced back to the mechanics and geometry of the treater.

To properly investigate any of these causes, it would be best to perform the dyne test using the drawdown method, which allows you to evaluate a relatively large area of film surface all in one pass. Patterns of variation will appear, which will be a valuable clue in determining the root cause of the problem. Also, by testing over a variety of dyne levels, you may be able to determine what might be called surface energy topography: just as an example, you may find that some randomly dispersed areas are testing at only about 32 dynes/cm, whereas the majority of the surface holds at 38. That might suggest blooming of low surface energy partials at those locations. And, in the case of film which has not been surface treated, it may provide a clue to which constituent is not being properly dispersed.

Even by using the simple method of applying the test fluids with cotton swabs, some patterns are easily recognizable. For example, spreading 38 dyne/cm test fluid over and around a fingerprint on a 44 dyne/cm surface will show the print as clearly as a forensics lab photo.

Dyne Testing at Elevated Temperatures and/or Humidity Levels

Question: I know that ideally dyne testing should be done under standard laboratory conditions, but as we test in our shop, which is not air conditioned, that is not possible for us – we sometimes reach temperatures as high as 40° Celsius, at high humidity. How do we deal with this?

Answer: Ideally, we do recommend testing at about 20° to 25° Celsius, with humidity in the range of 40% to 60% RH, Some variation from these conditions is not likely to affect results meaningfully, as the surface tension of liquids and the surface energy of solids are similarly affected by temperature changes, as shown in the following table. But with more seriously elevated temperatures, caution should be employed.

Material Surface Tension(a) Change per °C
2-ethoxyethanol 28.8 -0.13
40 dyne/cm test fluid 40.0 -0.14
Formamide 57.0 -0.15
Water 72.7 -0.21
Nylon 6-6 42.2 -0.065
PC 44.0 -0.060
PE 31.6 -0.057
PET 39.0 -0.065
PMMA 37.5 -0.076
PP 30.5 -0.058
PS 34.0 -0.072
PTFE 19.4 -0.058

Surface tension, and change per degree Celsius are shown in dynes/cm (equivalent to mJ/m2).

(a) Critical surface tension in dynes/cm at 20° to 25° C, generally determined by the Zisman method (regression of the cosine of the contact angle), or by the wetting tension method, using solutions of 2-ethoxyethanol and formamide per ASTM Std. D-2578. A more complete list of polymers is available here, and a more complete list of liquids is available here.

The largest numeric effect would be seen when testing at very high treat levels, where the test solutions are formulated from formamide and water. In the most extreme case – polyethylene treated to be water-wettable at 72 dynes/cm – the net drift works out to 0.153 dyne/cm per degree Celsius. This translates to an effect on test results of roughly -3 dynes/cm at 40° C, as the substrate surface energy would be reduced by only 1.14 dynes/cm, whereas the test solution (100% reagent grade water) would be reduced by 4.2 dynes/cm. It can be argued that even in this extreme case, an error of 3 dynes/cm at a treat level of around 70 dynes/cm would not be critical in most instances. But even at a more typical treat level of 40 dynes/cm, the net drift is still 0.083 dyne/cm per degree Celsius, resulting in an effect of about 1.7 dynes/cm, which could be significant.

So much for the hard data. Please note that this analysis assumes that both the test solution and the substrate have been stabilized at the elevated ambient temperature. If this is not ensured, then one component or the other will obviously be affected more by the temperature change, and test results could deviate from those that would be obtained under laboratory conditions in a less predictable fashion.

However, elevated temperatures pose their own problems. Surface tension test fluids will tend to degrade faster at elevated temperatures, and, while open, bottles will be far more prone to preferential evaporation of 2-ethoxyethanol, which has a higher evaporation rate. When testing with ACCU DYNE TESTTM Marker Pens, the test fluid may pass somewhat more readily through the spring-loaded valve tip – this can have an effect due to a tendency towards gravitational wetting when an excess of test fluid is applied. Also, the rate of evaporation of the test fluids once applied to the surface will increase, so the two second timeframe on which the test is based could come into question. Finally, solubility parameters are affected by temperature as well, so the chemical affinity of the fluids to the surface may be changed.

With regard to the substrate, the rate of crosslinking may be affected, which could have an impact on the surface energy when tested vs. its level over time. Finally, elevated temperature (as well as humidity) levels will tend to accelerate treat loss of the substrate. The mobility of surface-blooming additives will be enhanced, and transfer of the treatment from the treated to untreated side of a film once it is wound will also be more pronounced. This will not only drop the dyne level, but can also lead to blocking of the film when it is paid off from the roll during further processing operations. The latter effects are perhaps the best arguments for limiting product exposure to extreme environments to whatever extent is possible.

As to RH, it is best to avoid excessive humidity, as it can cause higher variability in test results. Also, if there is any moisture on the surface, it will absorb into the test solution, changing its surface tension and invalidating the test. Nylons especially may be prone to this, as they absorb water vapor far mote readily than most other polymers.

Given the range of physicochemical effects which can have an impact on the accuracy of test results produced at elevated temperature and or humidity, if product end-use requirements are rigorous, such as in the automotive industry and for many medical applications, it would be prudent to set up an experimental study to correlate dyne testing results obtained in the shop vs. those obtained under laboratory conditions. You can randomly split your samples into two groups. On one set, test at the machine as usual. For the other set, remove to the laboratory, allow the material to stabilize under that environment, then perform the same dyne test (with a separate set of test fluids or test markers which are to be kept in the lab!). Results can be compared readily. If there is a meaningful difference, I would also suggest looking closely at which set of results best predicts end-use performance or adherence to customer specs. Considering all the factors discussed above, you might even find that the results from the elevated temperature testing are the better predictor of product suitability. I wouldn’t want to bet against it a priori, even though such an outcome does seem rather counter-intuitive.

Overtreatment of TPO

Question: We supply solvent-borne automotive coatings, and recommend that our customers flame treat their TPO components to 48 to 60 dynes/cm for best adhesion and durability. We have seen adhesion failures at higher dyne levels. Would you expect that? And, would your test markers be able to identify overtreatment?

Answer: First, you are correct that any polymer can be over-treated by either flame or corona. The mechanistic details of just what happens to the material’s surface layers are undoubtedly different for the two treating methods. But what basically occurs in either case is the surface layer gets etched and oxidized to the point where it may be water-wettable, but it has been so decimated by the aggressive treatment environment that it no longer anchors well to the bulk of the polymer. The paint adheres well to the surface layer, but the entire paint/surface layer will easily lift away from the bulk of the polymer.

Over-treatment will not cause a decrease in dyne testing results unless too much pressure or abrasive force is used during the test – this would have the same effect as mentioned above, with the untreated underlayer exposed at the surface. So, as long as a light application pressure is used, as directed in the instructions, ACCU DYNE TESTTM Marker Pens can definitely identify the excessive levels of surface treatment that you have found to cause defects.

The problems that arise from over-treated surfaces suggest that it should be standard practice to establish a realistic maximum treatment level, as measured by the dyne test, as well as a lower one. Your suggested range of 48 to 60 dynes/cm sounds reasonable, but I would think that for most applications you’d do fine with a surface energy of 44 dynes/cm or so for solvent-based paints. The presence and concentration of additives and pigments could affect this minimum, especially with thick parts, where there is a large polymer bulk compared to surface area. And, for waterborne or energy cured coatings, the required dyne level would increase substantially.

Finally, as this is a rigorous application (automotive finishing), I would recommend that your customer consider an experimentally designed study of treatment parameters and measured dyne levels vs. end-use quality and durability metrics. Tightening the window of optimal treatment level could prove commercially beneficial.

What Dyne Levels Should I Be Testing At?

Question: My ink requires a minimum of 40 dynes/cm, and my film supplier says they run their poly at 56 dynes/cm. Should I use a 40 as a threshold test, or does the marker dyne level need to match the material? What range should I be using?

Answer: If the poly were actually at 56 dynes/cm when you tested it, a test with a 40 dyne/cm test marker would probably wet out so well and have such an attraction to the treated surface that it would permanently mark the film. That would tell you that the surface was way higher than 40 dynes/cm – if it was actually only 40 dynes/cm, the test fluid would start to bead up within two seconds or so. But there’s a lot more to the story than that.

Polymers lose treatment – especially when induced by corona treatment – over time and with downstream processing, so if a film tests at 56 dynes/cm at the end of your supplier’s extrusion line, you might find it to have a surface energy of as low as 44 dynes/cm a few weeks later, when you are ready to print it. (Please don’t take these treat loss numbers as gospel – they are for explanatory purposes only!) After a few months of storage in a hot and humid environment, it may well have dropped to below 40 dynes/cm. Slip agents are especially problematic when it comes to treatment loss over time, especially at elevated temperatures.

ACCU DYNE TESTTM Marker Pens measure surface energy by testing over a range of dyne levels, starting with a low enough level that you expect it will wet out for at least several seconds. In your case, I would recommend starting at 34 dynes/cm, and having the ability to test up to 56 dynes/cm, which is at the high end of what you are likely to ever see.

While your ink supplier probably is correct about 40 dynes/cm being the minimum acceptable treat level, it will still be instructive to test filmstock as it goes to press. Keeping records of dyne levels (and any comments testers may report) along with other process data may prove to be a valuable tool in troubleshooting at some point. Also, you may find that with higher substrate dyne levels you are able to increase press speed somewhat, and/or improve print quality. Quantifying these relationships can streamline your operation and ultimately reduce costs by enabling you to develop better specifications for your purchased rollstock.

TSCA Review

Question: I am conducting an environmental review on ACCU DYNE TEST Marker Pens, and the SDS does not have the information I need for approval. I need to verify that “all the components of the product are either listed or exempt from listing under the TSCA section 8(b) chemical inventory list.” Your SDS lists the TSCA inventory but references only the TSCA section 5(a) (significant new use) for one ingredient. Can you please provide the requested information?

Answer: All constituents of ACCU DYNE TESTTM Marker Pens and surface tension test fluids (both use the same formulations) are listed in the TSCA inventory. For full information on 2-ethoxyethanol (CAS 110-80-5), please see https://www.federalregister.gov/documents/2005/11/29/05-23421/2-ethoxyethanol-2-ethoxyethanol-acetate-2-methoxyethanol-and-2-methoxyethanol-acetate-significant.