Teflon®

Technical Information

Nonstick & Industrial Coatings

Teflon® is a registered trademark of DuPont.Use of DuPont trademarks subject to License Agreement and qualification.

��������������������������� ��������������������

Permeation—Its Effects onTeflon® Fluoropolymer Coatings

����������Fluoropolymers are virtually inert substances thatcan withstand high temperatures. These two proper-ties make them ideal for use as coatings to protectmetals from corrosive attack. It is important to real-ize that fluoropolymer coatings are used where otherpolymers could not possibly survive, that is, underextremely hostile conditions. Under such conditions,permeation assumes a much more important role inthe determination of performance.

This technical bulletin deals with the effects of per-meation, not the theory itself (Fick’s first law,Henry’s Law, etc.). Further, it deals with the effectsof permeation in the context of the harsh chemicalenvironment to which coated parts are exposed (andnot, for example, to the common thin plastic wrapsused in the food packaging industry). The first por-tion of this bulletin discusses permeation throughfree-standing films, which relates to plastics andcomposites such as dual laminates. In the secondportion, the effects of permeation are addressed asrelated to coatings bonded to metal substrates.

Permeation describes the transfer of gases andvapors in barrier materials such as polymeric plas-tics. As shown in Figure 1, the process involves:(1) dissolving the penetrant in the barrier material,(2) diffusion of dissolved penetrant through the ma-terial as a result of the concentration gradient, and(3) evaporation of the penetrant from the oppositeside of the material.

Permeation is generally regarded as an importantconsideration in determining the performance ofplastics or composites, and for good reason. Allpolymers are permeable, and structures such asdual laminates or sheet linings are essentially free-standing polymeric materials.

Coated metal vessels, on the other hand, are acomposite consisting of a relatively thin coatingbonded to a metal shell. The major constituent ofthe coating is a fluoropolymer, which is permeable.The steel shell, however, is not permeable (at leastnot to the gases and vapors encountered in chemicalprocessing).

����� ���

• Penetration of a barrier (polymer layer)by chemical vapors or gases.

• Function of the particular permeant:

– Solubility in the polymer

– Diffusion through the polymer’sintermolecular spaces

2

Thus, whereas the concern for plastic and compositestructures relates to the exterior emanation of chemi-cal vapors that have permeated the plastic, the con-cern for coated metal vessels relates to the chemicalattack on the steel from vapors permeating thefluoropolymer coating. This bulletin addresses theseconcerns and provides a discussion on coatings tech-nology that helps explain how bonded linings madefrom fluoropolymer coatings are ultimately perme-ated, and how new advances in coatings technologyallows us to manage these permeation effects.

� ������� ������������ ���Permeation involves a combination of physical andchemical factors1 as shown in Table 1. Note thatall the Factors in Table 1 are represented as a posi-tive (+) change, i.e., an increase. The effect of thischange can be an increase (+) or decrease (–) inthe rate of permeation, depending on the respectivefactor. For example, increasing the concentration,temperature, or pressure increases the rate of perme-ation. Increasing the polymer thickness decreasesthe rate of permeation.

Table 1Permeation Variables

Effect onFactor Change Permeation

Permeant Concentration + +Temperature + +Pressure + +Permeant/Polymer Chem. Similarity + +Voids in Polymer + +Permeant Size/Shape + –Polymer Thickness + –Polymer Crystallinity + –Polymer Chain Stiffness + –Polymer Interchain Forces + –

In general, crystallinity can range from the perfectorder of carbon atoms in a diamond to completelynoncrystalline, or amorphous, substances such asglass. Polyvinylchloride (PVC) polymers are consid-ered noncrystalline. Fluoropolymers and polypropy-lene resins are semicrystalline, meaning they havedomains (separate regions) of crystalline and amor-phous character. The ratio of these domains is deter-mined by the chemical makeup of the polymer aswell as how it was processed (rapid cooling, forexample, freezes the polymer chains in an amor-phous state whereas slow cooling allows the mol-ecules to arrange themselves in crystalline patterns).As the crystallinity increases, the related intermo-lecular volume decreases. This tighter packing of themolecules restricts the permeant from passingthrough.

Large size molecules, especially those with morecomplex, branched configurations are likely to besterically hindered and thus diffuse slower thansmall molecules with simple configurations.

If the permeant is chemically similar to the barriercoating, i.e., has similar chemically functional groupsand similar polarity, it will likely be more solublein the polymer. This will cause the polymer toswell, resulting in an increase in the molecularspace between the polymer chains and thus increasethe rate of permeation. Note that the effect of chemi-cal similarity counteracts that of crystallinity.

Increasing the polymer chain stiffness imparts resis-tance to bending. Increasing the interchain forces(Van der Waals, hydrogen bonding) increases themolecular attraction between adjacent polymer mol-ecules. Both of these factors result in lower molecu-lar mobility. The polymer molecules are less likelyto separate to allow the permeant molecule to passthrough, thus decreasing the rate of permeation.

Voids obviously relate to the quality of the polymerand the fabrication process. Voids are not consid-ered in permeation theory, but they are a fact ofcommercial life.

In addition to these factors, consideration must alsobe given to environmental stress cracking. Stresscracks can occur when a polymer is under a smallload condition for a prolonged time. Crystallinity,molecular weight, absorption of chemicals, me-chanical or thermal stress, and processing conditionscan all affect stress cracking2. Stress cracks, likevoids, allow the mass transport of chemicals throughthe coating.

Figure 1. The Process of Permeation

Evaporation(Outside)

Dissolution

Barrier Material

Diffusion

HighConcentration

(Inside)

3

Figure 4. Methyl Ethyl Ketone Permeation,250 µm (10 mil) thick

����� ���� � Permeation data is often presented for water vapor(H2O) as shown in the representative examplesFigures 2 and 3 below. The rate of diffusion ismeasured against the film thickness of the polymerfilms, at some defined temperature. Other kinds ofgraphs show how permeation varies with tempera-ture, at a given film thickness. The atmosphericgases (O2, CO2, and N2) are often presented thisway. Figure 4 represents an example of a moretechnical treatment of data3. The empirical dataquantify and confirm, as was stated above, thatpermeation decreases with increasing film thicknessand increases with increasing temperature.

It is important to exercise some care in the interpre-tation of these kinds of data. Casual observationmay lead to the conclusion that Product A is moreor less permeable than Product B, but upon furtherconsideration it becomes clear that such a conclu-sion generalizes a variety of factors that the datamay not necessarily support.

For example, the data are only valid for the sub-stances measured, namely water vapor, MEK, orthe atmospheric gases, and only under the conditionsof the test design used for the measurements. If thesewere the only substances involved in the process,and if the process had some correlation to thetest method, the data would be more meaningful.But most processes involve other substances, forwhich permeation test data are not available, andalso mixtures of other substances for which perme-ation test data are even less likely to be available.

Because there are a variety of factors that can affectpermeation, any generalizations based on simplewater vapor, gas, or single chemical data, therefore,are likely to be misleading.

Permeation is a rate function. It is a measure of thequantity (grams) of material that traveled througha given area of film (cm2) over some time interval(hours). The time interval actually used in the labo-ratory test is mathematically normalized to 24 hoursto make it easier to compare materials, but keep inmind that the graphs illustrate how fast a substancediffuses through the film. Saying that Film A ismore or less permeable than Film B is accurate, aslong as it is understood that the rate of diffusion isbeing discussed. Most processes operate longer than24 hours, so over extended time intervals any slightdifferences in the rates of permeation between twomaterials become less important. It’s just a matterof time before the effects of permeation make theirpresence known in either case.

Figure 2. Water Vapor Transmission Rate ofTeflon® FEP Film at 40°C (104°F) perASTM E96 (Modified)

Figure 3. Moisture Vapor Permeability Rate vs.Thickness at 60°C (140°F)

0.35

0.40

0.30

0.25

0.20

0.15

0.10

0.05

00

(0)50(2)

100(4)

150(6)

200(8)

250(10)

300(12)

Thickness, µm (mil)

Tran

smis

sio

n R

ate,

g/1

00 in

2 /24

hr

Lo

g P

erm

eati

on

Rat

e

Inverse Temperature x10 3, K 1

3.0

2.5

2.0

1.5

1.0

0.5

02.8 2.9 3.0 3.1 3.2 3.3

ECTFE

3.4

ETFEPTFE

0.09

0.08

0.07

0.06

0.05

0.04

0.03

0.02

0.01

90 25 50 75 100 125 150

Halar ResinK= 0.006

FEPK= 0.004

PVF2K= 0.003

∆P = 134 mmHg (90%RH)

Thickness, mil

F(

Gra

ms

24 H

r 10

0 in

2)

= K

∆P 1

Source: DuPont

Source: Ausimont Literature, 6/89 GB204

Source: DuPont

4

Figure 5. Chemical Reactions

The graphs in Figures 2 and 3 show a dramatic dropin permeability with increasing film thickness, thevalue of which depends on the temperature. Note,however, that the permeation rates at the higherthicknesses are not zero. In other words, the datadoes not say the fluoropolymer resins are imperme-able. Thicker films retard the rate of permeation, butdo not stop it. Permeation is unavoidable in free-standing plastic films, including fluoropoly-mer filmsas thick as 4,450 µm (180 mil). But it can be man-aged.

����� ���� � � �������• Little useful, hard data available

• Thin film (<250 µm) data is typically misleading

• Atmospheric gas data is typically misleading(N2, O2, CO2, H2O)

• Mixtures can permeate differently and/or selec-tively vs. single chemicals

• Laboratory data recommended only for qualitativecomparison

• Best data from test of actual component in exactprocess conditions

����� ������������� ��� �����A fundamental assumption in the formal treatmentof permeability is the absence of chemical reactivitybetween the penetrant chemical and the polymermaterial through which it diffuses. As illustrated inFigure 5, chemical reactions with the polymer itselfor with fillers or other additives can create voids,which then become pathways for the mass transportof the penetrants. This is the case with many organicpolymers in chemical service, especially at elevatedtemperatures. What finally emerges from the otherside can be an entirely different species.

If chemical reactivity is found to occur, permeabilityis no longer the sole issue. For example, Figure 6illustrates a failure of fiberglass reinforced plastic(FRP). FRP is constructed from fiberglass clothimpregnated with a nonfluoropolymer resin, usuallya phenolic, modified phenolic, and sometimes a vi-nyl ester. The FRP duct as shown is used in theSemiconductor Industry to exhaust the chemical by-products from the etching processes for the manu-facture of computer chips. Here, a white crystallinematerial is observed emerging. The composition ofthe white material is unknown, so it is not possibleto say whether it is a permeant chemical from theprocess or some other substance that resulted froma chemical reaction within the FRP itself. In eithercase, however, the duct is leaking and may representa potential hazard to personnel.

Other plastics such as polypropylene often fail bybecoming brittle, and are then prone to cracking.This type of failure relates to the poorer chemicalresistance of polypropylene polymer.

These kinds of observations characterize the com-plexity of the situation in the Chemical ProcessingIndustry, where it is often difficult to distinguishthe effects of permeation from those of chemicalcompatibility.

����� ��������������������������� ����Unlike the common plastics just discussed, Teflon®

fluoropolymers are exceptionally resistant to chemi-cal attack, even at high temperatures. The likelihoodof a chemical reaction between the permeant withthe pure polymer itself, therefore, is very low. Thus,as illustrated on the left side of Figure 7, there is ahigh probability that the process of permeation willleave both the permeant and the film unchanged.

Figure 6. FRP Failure

SinglePermeant

(A)Mixture(C+D)

(B) (E)

Different Species May Emerge

BulkPolymer

Chemical AttackCauses Voids

PermeationPermeants Can React Inside the Film

Photo courtesy of Fab-Tech, Inc., Colchester, VT

5

Figure 7. Permeation in Fluoropolymer Films vs. Coatings

But what happens when this irresistible force ofpermeation reaches the impenetrable wall of a steelvessel? It stops—at least in the sense that it doesnot emerge from the other side. At this point thepermeant chemical has few options. Depending onits degree of chemical similarity, it will becomeabsorbed to some extent within the polymer until asaturation point is reached. However, since the typesof process chemicals and the conditions inside thevessel (temperature, pressure, and concentration)fluctuate over time, molecular diffusion continu-ously occurs in both directions.

� ����������������It is important to distinguish a coating from a lining.A “lining” refers to a protective layer of material,and usually a thick layer of material, applied to theinside surface of a vessel. It does not matter whatkind of material is applied, or how it is applied.For example, the lining can be made from extrudedsheets of fluoropolymer (or other plastic), thermo-formed to the interior contours of the vessel, andglued in place (with welded seams). Or it canbe a Teflon® coating applied and baked on to theinterior surface. In both cases the vessel is saidto be lined because the material is on the inside ofthe vessel. On the other hand, the agitator bladefor that vessel is described as “coated” (or havinga Teflon® coating)—where the film covers theoutside surface.

The distinction between (interior) linings and coat-ings is important because the failure mechanismsare drastically different. The effect of permeationon linings is far more severe than on coatings.

But to better understand this, it is first necessaryto know something about the technology offluoropolymer coatings.

��������������� ���� �������Fluoropolymers have a high degree of nonstickcharacter, thus by themselves do not adhere well tometal (or other) substrates. A primer coat is neces-sary to achieve adhesion, and by default it cannotbe a fluoropolymer. There are a variety of tough,temperature-resistant organic resins that have excel-lent adhesion to metals and that also have goodchemical resistance. A primer, therefore, containsa blend of the fluoropolymer with another organicresin. When applied to the metal and fusion bonded(baked at high temperature), the organic resinbonds with the steel. The fluoropolymer componentsurrounds the organic resin molecules, providinga shield from chemical attack. It also fuses withthe fluoropolymer topcoat subsequently applied tocreate a chemically resistant surface layer. Thisconstitutes the coating system.

The development of the adhesive bonds with themetal substrate depends on how well the substratewas prepared prior to application of the coating.Contaminants such as dust, oils (even from finger-prints), or ionic residues from aqueous cleaningsolutions that remain on the surface will prevent theprimer from bonding at those points. Rougheningthe surface, preferably by grit blasting, increases thesurface area and provides additional sites for adhe-sive bonding to occur. A clean, roughened surface,therefore, is essential for good adhesion.

Permeant

Fluoropolymer Film

Time

Permeant DiffusesUnchanged Through Film

BulkPolymer

Permeant

Fluoropolymer Coating

TimePermeation StoppedAt Steel Substrate

Steel Substrate

CoatingFilm

PermeationFilms vs. Coatings

6

Another critical factor that affects adhesion is thebaking process. Fluoropolymer resins are thermo-plastic. Melt-processable varieties (including ETFE,PFA, and FEP) liquefy above their melting points,but the viscosity of the molten resin is very high.It takes time for this viscous fluid to flow togetherto form a properly fused, cohesive film. From anadhesion standpoint, enough time is also needed forthe molten resin to flow into and wet the valleysof the rough surface profile and for the organic resinin the primer to form the adhesive bonds. An ad-equate dwell time at or above a specific temperature(depending on the particular resin) is essential,therefore, to develop the full integrity of the coating.

From a practical standpoint, metal parts come indifferent sizes and masses, and the temperaturesin baking ovens are seldom constant everywherewithin. These variations can significantly affect thequality of the bake, and therefore the performanceof the final coating. To compensate for these effects,a robust coating system is needed, i.e., one havingthe thermal stability to withstand prolonged expo-sures at high bake temperatures without drips, sags,blisters, or polymer degradation.

No coating process is perfect (and this applies to allcoatings, not just fluoropolymers). A single coat willalmost always contain pinholes, and pinholes allowgross mass transport of chemicals through the film(which makes the discussion of permeation effectsacademic). Fortunately, multiple coats cover up thepinholes, and routine quality control spark-testingon the final system assures this is so. Micropores,however, are more insidious. Micropores can be anymicroscopic void in the coating. They arise fromminute impurities inevitably encountered duringnormal production conditions and are more com-monly found at the interface between the primer andthe metal substrate due to contamination from pro-cessing and handling. Spark tests will not detectthem. As will be discussed later, these microporesare believed to be the locus of severe failure of thelining in service.

!�������������� ��������������������� ����Regardless of differences in the rates of permeation,penetrant chemicals will ultimately reach the metalinterface where they can react with the molecularfunctionalities in the primer that form the adhesivebonds. If the adhesive bonds were destroyed thecoating would obviously lift away from the metal.In practice, however, this is seldom the case becauseof the chemical shielding effect due to the presenceof the fluoropolymer. What usually happens is that

the adhesive bonds become solvated (hydrated inaqueous systems)4. Although they remain mostlyintact they may be considerably weakened. Theeffect of hydration is temperature dependent, almostinsignificant at room temperature and increasing inseverity as the boiling point of water is approached.At the boiling point, where water is now in equilib-rium with its vapor, the rate of permeation increasesdrastically. Thus water alone is capable of signifi-cantly weakening the organic resin adhesive bonds.Peel strength values, for example, show a significantdecrease after exposure to several hours of immer-sion in boiling water. However, a day or so afterexposure the peel strength values show considerablerecovery, although not to their original values.

The discussion above describes what happens whena coating (as on the agitator blade) is exposed to apermeation situation. The factors listed in Table 1,for a specific exposure situation, determine theoverall performance of the coating. In practice,fluoropolymer coatings on parts that are immersedin some chemical mixture perform very well for along time (unless the coating was poorly appliedinitially). Most failures involve other factors, suchas abrasion resistance where the coating is mechani-cally removed slowly over time.

!�������������� ��������������������� ����"���# ���������The situation is much different when the same coat-ing system is used as an interior lining. In this case,after some period of exposure at elevated tempera-tures, the lining will start to blister. The numberand size of these blisters can vary. Sometimes manysmall blisters form, as illustrated by a clear PFAlining in Figure 8. Over time, these small blisterswill continue to grow and (as evident in Figure 8)begin to merge into larger ones. Sometimes only oneor a few very large blisters form initially. In eithercase, given sufficient time of exposure, large sec-tions of lining can completely delaminate from themetal surface. Note, however, that this phenomenonis only observed when the liquid inside the vessel isat a high temperature.

Obviously, this phenomenon cannot be explainedby the mere hydrolytic weakening of primer bondsdue to permeation. Coatings do not fall off agitatorblades, nor then should they delaminate from theinterior wall of a vessel. The fact that they do re-quires a more complex explanation, and the mecha-nism postulated for lining failure includes, in additionto permeation, a combination of osmosis and ther-modynamics.

7

As permeant chemicals and moisture travel throughthe film they dissolve any ionic impurities thatare present and, eventually, transport them to amicropore, as schematically illustrated in Figure 9.Micropores within the film itself are not criticalbecause the plastic matrix is a nonreactive fluoro-polymer. At the metal interface, however, they aremuch more critical because this is where the adhe-sive bonds are. The accumulation of ionic speciesinside the micropore at this point is driven by thetemperature differential between the warm coatingsurface on the interior of the vessel (tProcess) and thesteel with a cooler exterior temperature (tAmbient).Fluoropolymers, being excellent thermal insulators,allow this differential to exist. The gaseous per-meant, therefore, can condense to a liquid. Thenet result is the creation of a region of higher ionicconcentration, which sets up an osmotic cell5. Theosmotic pressure is strong enough to disbond theprimer coat and thus lift the film from the metal.

Over time the area of disbonding continues toenlarge, eventually forming a visible blister.Remembering that permeation and the hydrolyticweakening of primer bonds increase with increasingtemperature, the disbonding effect from moisture isaccelerated as the temperature increases. Upon

reaching the boiling point of water, catastrophic fail-ure (severe blistering or total delamination) can occurwithin a few days for a poorly designed or poorlyapplied coating system (at 1,000 µm [40 mil] thick-ness; thinner films fail faster). Thus, whenfluoropolymer coatings fail as linings in a chemicalexposure, they almost always do so by delaminationfrom the metal substrate.

In the laboratory, the Atlas Cell Test (also calleda Blind Flange Test), Figure 10, is probably thebest prognosticator of the performance of a liningsystem. This device simulates the conditions of anactual lined vessel, especially the thermodynamicsbecause the process temperature of the contentsinside the vessel can be carefully controlled andmeasured against the ambient air temperature.

Coated (lined) panels are clamped to each openside of the glass vessel and sealed with a rubbergasket. Any liquid or mixture can be tested byadding a sufficient amount to cover the bottomtwo-thirds of the panels so that the effect on thecoating in the liquid and vapor phases, as wellas at their interface, can be observed. The apparatusadditionally includes a heating element to heat theliquid, a thermometer to measure temperature, anda reflux condenser to return condensed vapors. Thecell can be operated for months with little attention,although it is often more frequently dismantled tobetter observe progress.

Boiling water is always used as the first test mediumbecause it is safe to handle and quite effective indefining performance. Typically, most fluoropoly-mer linings of 1,000 µm (40 mil) thickness start toblister after only 2–3 weeks (335–500 hours) whenexposed to boiling water. The pictures in Figures 8and 11 were obtained from this test. If a coatingcandidate shows promise, additional acids andbases or other chemicals can then be tested.

Figure 9. Mechanism of Lining Failure

Figure 8. Initial Blisters in Lining

Permeation Effectson Fluoropolymer Linings

• Permeation Stops at Steel Substrate

• Attacks/Weakens Primer Bond

• Osmotic Pressure DisbondsCoating—Forms Blister(or in worst case, severefilm delamination)

Blister

Permeant

Topcoat

Primer

Substrate

Time Micropore t Ambient

∆t

t Process

8

Although the Atlas Cell Test closely simulatesactual field parameters and conditions, there is noconsensus on its correlation to actual field perfor-mance. One possible reason is that most fluoropoly-mer linings evaluated to date only last a relativelyshort time in this test (2–3 weeks). It may be thatthis level of performance is simply inadequate tocompensate for the actual variables such as the spe-cific chemicals, their concentrations, and especiallythe temperatures and pressures of real processes.However, as will be discussed next, new technolo-gies that are now lasting much longer in this test arecurrently under field evaluation and the expectationis that a better correlation will be found.

$ � ���������� ���%��������� ��������������It is obvious by now that regardless of the slightlydifferent rates of permeation through various materi-als, some species will eventually get through to theinterface between the metal substrate and the coat-ing or lining and cause problems. The problem isespecially acute for linings because they are themost expensive to install, yet most likely to failprematurely.

It does not matter which kind of fluoropolymer isused; they are all permeable. Thus the only recoursefor the corrosion engineer who requires a fluoro-polymer lining for chemical protection is to use athicker one. Thick linings do slow the permeationprocess and extend the useful life of the lined vessel.

For example, many vessels are lined with fluoro-polymer sheet linings that are 1.5 to 2.25 mm (60to 90 mil) thick. However, sheet linings are veryexpensive and have their own limitations (namelyseam cracking and disbonding of the neoprene orepoxy adhesive).

Fluoropolymer coatings used for linings are lessexpensive, but until recently there were no com-mercial products that performed reliably. This isbecause it is difficult to apply coatings on the interi-ors of vessels and the maximum obtainable filmbuild is generally only around 1,000 µm (40 mil).Furthermore, the currently available coatings arelimited to essentially pure fluoropolymers (althoughsome applicators add small amounts of pigmentsfor color). After final curing, these coatings have thesame basic properties as extruded films—includingthe same degree of permeability. Thus, since even2 mm thick sheet linings have limited service life-times, it would be unreasonable to expect a 1 mmthick coating to perform as well—it certainly wouldnot be expected to perform any better.

The challenge, therefore, is to focus on the effectsof permeability, and that inevitably leads to coatingstechnology. Coatings technology has the uniqueadvantage of versatility. Unlike sheet linings madeof pure bulk polymers, for example, where theonly means of retarding permeation is to makethem thicker, coatings can be modified with specialadditives to enhance a desired property withoutsacrificing the basic performance properties of the

Figure 10. Atlas Cell Test Apparatus Figure 11. Atlas Cell Test Panel, 1,000 µm (40 mil)PFA, 2-Week Boiling Water Exposure

9

fluoropolymer. In the case of primers, additionalother nonfluoropolymer resins are added to provideexcellent adhesion, as well as pigments and otherfillers that act as barriers to permeation. The resultis a strongly bonded primer coat that better resiststhe effects of hydrolytic weakening from permeation.

Special pigments are selected for the topcoats thatsignificantly retard permeation by forcing the vaporsto penetrate the film using a zigzag pathway, asillustrated in Figure 12. Adding such fillers hasthe same effect as increasing the thickness of thefluoropolymer, but it is more efficient and far morecost effective. Such additives, of course, must bechosen judiciously because the wrong choice canactually increase the rate of permeation by formingchannels between the nonstick fluoropolymer resinand the surface of the additive. It is the coatingsystem, therefore, that determines performance, notjust the particular polymer upon which it is based.

DuPont has over 40 years of coating experience ona vast array of miscellaneous industrial parts usedin a wide variety of service conditions, as well ason millions of frying pans coated with Teflon® andSilverStone® nonstick. This long-term success hasrelied on developing primer technologies that formexceptionally strong bonds with metal substratesand topcoats that are both durable and functional.It is this expertise that was incorporated into thedesign of the Teflon® coatings and linings dedicatedto the Chemical Processing Industry, and is whatbrought new products such as the Aqueous andPowder “Ruby Red” coatings to this market. As theAtlas Cell Test performance shown in Figures 13and 14 attests, these new “Ruby Reds” are far supe-rior in performance than any other fluoropolymercoating ever tested by DuPont.

Figure 12. Special Fillers Force Permeants to Take a Much Longer, Zigzag Path

Filler = Barrier

Chemicals

Unfilled

Chemicals

Substrate

Clear Topcoat

Barrier Midcoat

Primer

Thick, ClearTopcoat

10

Figure 13. Atlas Cell Tests, Boiling Water, 1,000 µm PFA Linings

Figure 14. Ruby Red Prototype, 445 µm (17.5 mil) Thick, After 1-Year Atlas Cell Exposure inBoiling 0.05N Hydrochloric Acid

Standard PFA after two weeks Ruby red after two months

11

&��� ��All polymers are permeable and, although many ofthe factors that affect permeability are known, thereis often no test data available that are meaningfulfor the specific conditions of a specific process.Using laboratory test data based on simple atmo-spheric gases or other individual chemicals to pre-dict performance in harsh chemical environmentsencountered in the Chemical Processing Industrycan be risky. Chemical compatibility of the plasticsand any additives with the process chemicals mustalso be considered, because chemical attack can befar more detrimental to performance than mere per-meation. Fluoropolymers have outstanding chemicalresistance. When integrated into a well-designedand properly applied coating system, Teflon® coat-ings from DuPont provide a superior alternative toany of the competitive coatings based on the pureresins and approach the performance of sheet lin-ings. These new coatings clearly demonstrate thatthe effects of permeation can be managed. Andwhile they may not be the final solution to the per-meation problem, the evidence so far strongly indi-cates they are a quantum leap better in performancethan anything else on the market.

����������1. Buxton, L. W., Spohn, P. D, and Will, R. R.,

“Understanding How Molecules PermeateThrough Solid Substances,” SAE TechnicalPaper Series, 940164 (1994).

2. Imbalzano, J. F., Washburn, D. N., and Metha,P. M., “Basics of Permeation and EnvironmentalStress Cracking in Relation to Fluoropolymers,”DuPont Technical Information BulletinH-24240-1, pp. 4–5.

3. Buxton, L. W., et.al., “Fluoropolymer-LinedChemical Systems and Permeation,” DuPontTechnical Information Bulletin H-54485-1.

4. Wu, S., Polymer Interface and Adhesion,Marcel Dekker, Inc., New York, 1982.

5. Zolin, B. I., “Protective Coatings: ProtectiveLining Performance,” Chemical EngineeringProgress, Vol. 66, No. 8, 1970.

Teflon®

Only by DuPont(1/01) RWJ68 Printed in U.S.A.Reorder No.: H-88495

��������������� ����������������� ����%DuPontTeflon® Nonstick & Industrial CoatingsChestnut Run PlazaP.O. Box 80702Wilmington, DE 19880-0702

'())*�++,-./,/� 0%�'1)2*�133-(3)2

����DuPont de Nemours (Belgium)A. Spinoystraat 6B-2800 MechelenBelgiumTel.: 33-15-441188Fax: 33-15-441160

�� �DuPont China, Ltd.Room 1122,New World Office Building

(East Wing)Salisbury RoadKowloon, Hong KongTel.: 852-2734-5459Fax: 852-2368-3512

��� � �DuPont Australia, Ltd.254 Canterbury RoadBayswater, Victoria 3153AustraliaTel.: 61-3-9721-5617Fax: 61-3-9721-5690

DuPont Korea4/5th Floor Asia Tower#726 Yeoksam-dong,Kangnam-kuSeoul, KoreaTel.: 82-2-2222-5385Fax: 82-2-2222-5478

����DuPont K. K. (Teflon® Finishes)4th Floor, Chiyoda Honsha Building5-18 Sarugaku-cho, 1-chomeChiyoda-ku, Tokyo, 101 JapanTel.: 81-3-5281-5888Fax: 81-3-5281-5899

The information set forth herein is furnished free of charge and is based on technical data that DuPont believes to be reliable. It is intended for use by personshaving technical skill, at their own discretion and risk. The handling precaution information contained herein is given with the understanding that those usingit will satisfy themselves that their particular conditions of use present no health or safety hazards. Because conditions of product use are outside our control,we make no warranties, express or implied, and assume no liability in connection with any use of this information. As with any material, evaluation of anycompound under end-use conditions prior to specification is essential. Nothing herein is to be taken as a license to operate under or a recommendation to infringeany patents.

CAUTION: Do not use in medical applications involving permanent implantation in the human body. For other medical applications, see “DuPont MedicalCaution Statement,” H-50102.