Note: Descriptions are shown in the official language in which they were submitted.
1341016
-1-
POLYMER-METAL BONDED COMPOSITE
AND METHOD OF PRODUCING SAME
FIELLI OF THE INVENTION
This invention relates generally to the field of
bonding polymeric materials to metal materials and
particularly to bonding fluorinated polymers and
polyether resin: to metals, including ferrous-based
metals.
REPORTED DEVELOPMENTS
In the chemical processing industry, as well as
many other industries, a variety of composite
materials are used to fabricate apparatus used
therein. In many instances, metals are used to
provide the structural strength for such apparatus.
However, such apparatus, many times, is required to be
exposed to highly corrosive materials which are being
processed; some of th.i.s exposure is at elevated
temperatures and/or elevated pressures which tend to
exaggerate the corrosive properties of the materials
being processed. It has been found necessary, in many
applications, to protect the metals used in such
apparatus from the effects of corrosion, under varying
1341 01 6
2
conditions of temperature and pressure, particularly
at elevated temperatures, and at increased pressures.
The approach, generally, to this problem has been
to shield the si~ructural metals from corrosive materi
als. This is done by forming a composite by super
imposing other materials onto those surfaces of the
structural meta:Ls which would otherwise be exposed to
corrosive attack, the concept being to only permit
contact of corrosive materials to barrier materials
which will resi:~t the effects of such corrosion which
are formed on the surfaces of underlying or substrate
structural meta:Ls. Of course, such overlayed or
barrier materia:Ls are selected to have relatively
little, or idea:Lly no, reaction to those materials
which otherwise corrosively attack the substrate
structural metals.
One concept of protecting structural metals from
corrosive attack is to bond a glass coating to those
surfaces of the substrate structural metal which are
to be exposed to materials which would corrode those
structural metals if contact were permitted. This
concept has been used for many years and is quite
satisfactory where substantially no flexing of the
structural metal is possible. If flexing is a
possibility, glass-metal composites may encounter
problems as the glass overlay, or barrier coating, of
the composite may tend to crack, thus providing an
avenue for corrosive material to reach the base sub-
strate structural material. Also, glass coated metal
materials tend to be highly susceptible to mechanical
damage. Finally, glass, being an amorphous material,
is not resistant to corrosive attack by various common
chemicals.
1341p16
3
Another concept has been to overlay, or form a
barrier coating, of relatively thin sheets of non-
corrosive metals, such as titanium, tantalum, hafnium,
etc. onto structural_ base metals, for example, mild
0.25% carbon low carbon steel. This concept requires
the bonding of very a}:pensive overlay, or barrier
coat, metals onto othEar_ dissimilar structural base
metals. Not only are the overlay, or barrier coat,
metals expensive:, but the process of bonding, usually
requiring extensive and complicated welding techni-
ques, is very e~c:pensive. Nevertheless, this concept
is commercially used where the apparatus is to be
exposed to a combinat_Lon of extreme corrosion and
extremely elevated temperatures and where extreme
temperature differentials, which occur rapidly, are to
be encountered.
A third concept has been to apply a polymer to
the surface of t:he structural metal, the polymer being
bonded to the surface of the structural metal. This
concept has had some success where the corrosive
effects of the c:orros:ive materials are relatively mild
and where the e7_evated temperatures to be experienced
are modest, being below the heat degradation points of
the polymers usod. Also, this concept has been used
where anti-stic~:ing properties are important, such as
in roll coverings used in dryer rollers, carrier
rollers, etc.
1341016
- 4 -
In an attf~mpt to overcome some of the limitations of
most polymers, in rE=spect to corrosive resistance and:to
limitations on rangE= of elevated temperature use, fluorinated
polymers have been used as overlays, or barrier coats, on base
structural material;. As is well known, fluorinated polymers
exhibit relatively high corrosion-resistance in comparison to
other polymers. Al;~o, fluorinated polymers have a relatively
high operating temperature point of degradation, in comparison
to other polymers. Finally, fluorinated polymers, as well as
other polymers are :relatively much more flexible in comparison
to glass, and are e:~sentially inert to most common chemicals up
to the melting poinl~ of such fluorinated polymers.
In developing polymeric barrier coatings as applied
onto metals, a true composite is formed only where the
materials are bonded together with high integrity bonds. This
is to say that the ==esin used, which is in contact with the
metal substrate, should be as firmly bonded to that metal as
possible.
It is also necessary that microvoids (porosity) of
the coating be essentially eliminated if the polymer barrier
coating of the composite is to be utilized to prevent corrosion
of the underlying substrate metal.
All polymeric barrier coatings, to one extent or
another, are subjeci~ to molecular permeation by gaseous
chemicals. Permeani~ flow is accelerated by elevated
temperatures and by increased pressure. This phenomenon exists
because it is virtu<~lly impossible 'to remove all of the voids
(porosity) in the coating. However, the fewer voids, the less
the permeation. The permeation, of course, is not detrimental
to the resin itself because, hopefully, the polymer selected
for the barrier coai~ is chosen because of its inertness in
respect to corrosive: attack by a particular chemical or set of
r
1341016
- 5 -
chemicals. However, the barrier coat is necessary, in the
first place, becaus~s the chemical or chemicals in question do
corrosively attack she underlying substrate metal, while the
barrier coat should impede this process.
If the bonding of the resin to the substrate metal is
not complete, that is, if a substantial percentage of the resin
is not completely banded to the adjacent metal surface, the
metal at those poini~s is open to attack by the permeants.
Generally, the subsl~rate metal, particularly in the form of
vessel walls, is non-isothermal with respect to the thermal
condition of the co=rrosive medium, which, in the case of
vessels, comprises i~he contents thereof. Thus, the metal
surface acts as a hf=at sink, and the permeants tend to condense
and collect on thos<~ portions of the colder metal surface to
which the resin is not bonded. When this occurs, the corrosive
substance causes note only deterioration of the exposed metal,
but also deterioration of the adjacent metal underneath the
metal-resin particle bonds that do exist. The result may be
that the existing bonds a:re destroyed and delamination occurs.
This phenomenon can show -upon the surface of the barrier coat
as blisters. Such blisters may be caused by gas and/or liquid
build up, beneath the barrier coat, where the barrier coat has
begun to become del<~minat~~d. These blisters indicate that the
metal substrate underneath the blisters is suffering corrosive
attack. Of course, the corrosive attack to the substrate metal
frequency create di;~coloration of that metal as salts and
oxides are formed. This discoloration, or blanching, as well
as blistering, are ~risual:ly detectable on the barrier coat
surface as the underlying corrosion becomes pervasive.
On the other hand, the more resin bonded to the
metal, the less metal substrate surface area there is exposed
to the permeants. ~Cherefore, the less condensation there is
which occurs on the metal. Also, the fewer voids there are
X
_ X341016
- 6 -
within the barrier coating, the less opportunity for the
permeants, gaseous ~~r otherwise, to get through the barrier
coating to the substrate metal.
In additi~~n to increasing the integrity of bonding of
the resin to the substrate metal, thus decreasing voids and
decreasing metal surface available to attack, another approach
is considered desir:~ble. This is basically to increase the
thickness of the barrier coat itself, the theory being that
this will hinder permeation because, simply, the permeants have
a greater distance i~o travel, and the possibility of tortuous
pathways, through interconnected voids, being blocked by resin
is increased because there is more resin between the barrier
coat surface and thc~ underlying metal substrate surface.
Fluoropol~,rmers are well known for their inert
characteristics in :=espect to a wide variety of different
chemicals. In addit=ion, fluropolymers are well known for their
high temperature capabilities relative to other polymers.
Therefore, fluoropo=Lymers are primary candidate materials for
chemical barrier co<~tings.
On the other hand, fluoropolymers characteristically
are very long chain,, high molecular weight, high melt viscosity
polymers with a narrow temperature .range, relative to most
other polymers, between melt and degradation. Fluoropolymers
are also very poor conductors of heat, complicating the
approach to developung heat input to induce melting, thus
producing void free barrier coatings. The combination of these
factors makes proce:~sing of fluoropolymers difficult, if not
impractical, under rnany circumstances. Therefore, although
fluoropolymers may be primary candidate materials for chemical
barrier coatings, they are=_ difficult to process and apply,
which in many circurnstances substantially diminishes this
candidacy.
x
1341016
As implied above, it is known in the art that the
chemical permeability of barrier coating varies inversely with
the thickness of that barrier coating. However, it is quite
difficult to form relatively thick coatings of fluoropolymers
because of their inherent high viscosities which result in low
melt flow and slow fusion characteristics. To bond the
particles of fluoro_oolymer resin to each other and to
underlying substrate materials, the particles must be brought
to above melt temperature but kept below the practical
degradation temperature. Inability to control this process may
result in entrapment of air between particles, ultimately
resulting in the formation of bubbles in the barrier coating.
It is very difficult to control this process when such coatings
are applied to relatively large or complex metal shapes, as it
is difficult to convrol the temperature of each discrete point
of such pieces within a narrow range such that each discrete
point is above the melt point of the fluorinated polymer being
applied but not above the practical degradation point. Also,
it must be ensured ~~hat the surrounding atmosphere, adjacent to
the exposed face of the barrier coat being applied is,
likewise, within thf~ narrow practical range of temperature.
And finally, all particles of the fluorinated polymer across
the thickness of the=_ coating must, likewise, be within that
narrow practical temperature range, notwithstanding the fact
that fluorinated po:Lymers are notorious for poor heat transfer.
Attempts have been made to build up series of thin
coats of fluorinated polymers, as barrier coatings, overlaying
one on another, and using a heating cycle in between each thin
layer to bond it to the previous layer. In commercial
applications, as arc=_ well known, the powdered polymer resin
particles are suspended in a carrier fluid, usually water, and
sprayed in a thin layer, onto the substrate metal, followed by
a heating step. This is followed by a repeat of the cycle,
134~0~
.. s
_8_
many times, each time laying down a 0.001" to 0.010" thick
layer. This method has encountered difficulties as small
quantities of the c,~rrier fluid tend to remain trapped within
the lattice formed :by the powdered resin particles. On
heating, the carrier fluid vaporizes and expands, which can
separate the layers from one another and may prevent bonding.
This can appear as .surface bubbles. However, even when each
sprayed layer is carefully dried, acceptable bonding may not
occur between the t:zin layers of the barrier coating; the
reason for this is :zot clear.
Also, there are major problems in developing and
maintaining a uniform suspension of fluorinated polymer resin
particles in the carrier fluids. A variety of additives in the
form of surfactants, antifoaming agents and other "wetting
aids" and "processing aids" are used in an attempt to overcome
134101 fi '
_g_
these problems. It is bE~:lieved that these additives hinder
the bonding of successfu:l:ly built up thin layers of the
polymers to each other, Even in situations where no bubbling
occurs between such layers.
More recE~ntly, attempts have been made to apply dry
powdered fluorinatErd polymer resins to metal substrates using
elect rostat is deposit io:n, f loccing and f luidized bed
techniques. Although coatings in Excess of 0.012" thickness
have been accomplished, attempts to form coatings up to 0.040"
thickness have failed due to the formation of bubbles and
voids during the heating stage, regardless of whether a single
thick layer is applied or multiple thin layers, interspersed
with heating steps, are attempted. The cause of such failures
is not clear.
The following publications disclose coatings
comprising fluorocarbon polymers: U.S. Patent Nos. 4,064,963
and 4,605,695; U.K. Patent No. 2,051,091, and EPO Publication
No. 10152.
The major problem in using fluorinated polymers as
an overlay, or barrier coat, in composites with substrate
metals, is that it is difficult to produce high integrity
bonding of the fluorinated polymers to the base structural
metals. One composite of fluorinated polymer, overlayed as a
barrier coating onto steels, has been successfully marketed,
under the trade name Fluoroshield*, by W.L. Gore and
Associates, Inc. This composite is believed to be detailed in
British Patent No. 2,05:1,091. Fluoroshield coated metals,
* Trade-mark
1341016
-9a-
however, as will be' later specifically detailed, do not appear
to exhibit the lone-term bonding integrity or chemical
resistance which i~~ deemed necessary, by those with skill in
the field, to ensure the extended higher temperature
corrosion-
134016
- 10 -
resistance necessary for reliable use in chemical processing
equipment.
SUMMARY OF THE INVENTION
In accordance with one aspect of the present
invention, there it; provided a substrate coated with a non-
porous corrosion-resistant reslnou:> coating formed by fusing a
coating composi_tior~ including a polyether resin and comprising
a mayor amount (more than about 50 wt.~) of resin and a minor
amount ( less than C~bout 50 wt . ~ ) of a property-improving
additive, said resin being:
( A ) a f luoroca.rbon res in se leca ed f rom t he group
consist ing of ( 1 ) ~~erf luoroa:lkoxy t: et raf luoroethylene
copolymer resin (PfA), (a?) ethylene-chlorotrifluoro-ethylene
copolymer resin (E-CTFE), (3) ethyJ.ene-tetra-fluoroethylene
copolymer resin (E-TFE), (4) poly(vinylidine fluoride) resin
(PVDF), (5) tetrafluoroet.hylene-hexafluoropropylene copolymer
resin (FEP), (6) poly(chlorotrifluoroethylene) resin (CTFE),
or a mixture of two or more of said fluorocarbon resins; and
also containing, a;a an addit ive, a ceramic powder, the
ceramic powder being a tne~t:al carbide, silicon nitride, boron
nitride, titanium dibor de or aluminum diboride powder.
Another aspect: of i:he present invent ion is the
provision of a coating composition which is capable of being
fused, that is, melted at elevated temperature and then cooled
to form the aforementioned corrosion-resistant resinous
coating. The present invention encompasses within its scope
coat ing composit ions in ~,rhich the resin const ituent is present
in a mayor amount :.in the form of a mixture of resins (A)
1341016
- 11
and/or (B) above combined with a mj.nor amount of the property-
improving additive. In addition, t;he composition includes
within its scope tr.e use of a mixture of additives, for
example, a mixture of two or more of the additives of (C), (D)
and (E) above, and, withj.n the group of additives of (E)
above, a mixture of two or more of such additives.
Accordinl~~ to a further a~>pect of the invention there
is provided a method of forming a <:oating on a metal surface,
which method comprises :fusing a pol.yether-containing resin to
said metal surface to form a base <:oat; fusing a top coat to
said base-coated mEtal surface, saj.d top coat containing (A) a
fluorocarbon resin selected from the group consisting of (1)
perfluoroalkoxy tetrafluoroethylene copolymer resin (PFA), (2)
ethylene-chlorot rif luoro--ethylene copolymer resin ( E-CTFE ) ,
(3) ethylene-tetra-fluoroE~thylene copolymer resin (E-TFE), (4)
poly- ( vinyl idine f I uoridE~ j yes in ( F'VDF ) , ( 5 ) tet raf luoro-
ethylene-hexafluorupropyl.E~ne copolymer resin (FEP), (6)
poly(chlorotrifluoroethylene) resin (CTFE), or a mixture of
two or more of said fluorocarbon resins; and, as an additive,
(B) a ceramic powdery, t:he ceramic powder being a metal
carbide, silicon nitride, boron nitride, titanium diboride or
aluminum diboride ~~owder,.
Speaking generally, the property-improving additive
can be selected to improve various properties of coatings
formed from the co~T~positj_on of the present invention, for
example, properties; such as corrosion-resistance, abrasion-
resistance, and/or bonding characteristics.
~341p16
- 11 a --
The preferred property-improving additive is a metal
carbide, most prefE~rably silicon carbide or zirconium carbide
or a mixture of such carbides, and,. for use with fluorocarbon
resins, also PEEK.
It is ex~~ected that the _Lnvention will be used most
widely in connection with forming coatings on metallis
surfaces, part iculo.rly anon substrates . However, non-metallic
surfaces can be co~~ted ,also with a composition of the present
invention.
The present invention encompasses also a method for
forming a coating from the composition of the present
invention, including process means for the
1341016
- 12 -
application of such a composition to an underlying substrate.
DETAILED DESCRIPTION OF THE INVENTION
The present invention includes a composite of a
polymer barrier coaxing which is highly integrally bonded to a
substrate metal sur:Eace, the coating exhibiting substantially
fewer voids and improved resistance to corrosion and abrasion.
Thus, given a certain thickness of the barrier coating of the
present invention, in comparison with an equal thickness of
barrier coatings found in the prior art, the present invention
is responsible for affecting substantially decreased corrosive
attack to the under:Lying metal substrate at elevated
temperatures and/or pressures over a prolonged period of time.
The present invention also provides a method of producing such
a composite as well as a formulation, the use of which will
produce the barrier coatings, sheets and shaped articles of the
present invention.
As will bc~ described in greater detail below, the
additive-containing polymer compositions of the present
invention may be di~,rided for convenience into three groups
based upon the uses to which they are put: primer coatings,
barrier coatings, and abr,asion/wear resistant functional
coatings.
Primer coatings bond very strongly to the underlying
metal substrate and themselves provide a substrate to which
coatings having other properties may be strongly bonded. In
applications where umprov~ed bonding of a protective resin
overlay is sought, t=he coating composition may be applied as a
"primer coating" to the underlying metal substrate.
Barrier coatings provide a barrier between the
substrate to which t=hey are bonded and a corrosive environment.
In applications where improved corrosion-resistance of a
x
1341 01 6
- 13 -
protective resin overlay is sought, the coating composition may
be applied directly to the metal substrate or over a previously
applied resin coat, for example, over a primer undercoat.
In certain instances, there may be an overlap in the
use to which a given composition may be put, that is, certain
of the compositions bond strongly to a metal substrate and at
the same time provide a superior barrier to chemical attack.
In addition, as a gf=_neral rule, polymer compositions of the
present invention which bond most strongly to the substrate are
useful also as abra;~ion-resistam coatings applied directly to
the substrate or over other polymer coats.
According:Ly, i.n applications where improved abrasion-
resistance of the coating is sought, the coating composition
may be applied as an outermost layer over previously applied
resin layers or in direct contact with the metal substrate.
Compositions useful particularly for forming primer
coatings include (l~ a fluorocarbon resin of (A) above in
admixture with an aciditiv~e of (C) and/or (E) above and (2) a
polyether resin of (B) above in admixture with fluorocarbon
resin, preferably a resin of (A) above, and/or a ceramic powder
of (D) above.
Compositions useful parti~~ularly for forming a
corrosion-resistant barrier coating include a fluorocarbon
resin of (A) above =Ln admixture with an additive of (E) above.
Compositions useful particularly for forming
abrasion- and wear-resistant surfaces or top coatings include a
fluorocarbon and/or polyether resin of (A) and (B) above
respectively in adm=~xture with a non-resin additive as
identified in (D) and (E) above.
1341016
- 14 -
The fluor~~carbon resins and polyether resins of (A)
and of (B) above ar~= known classes of resins. Species of such
resins are available commercially.
The fluorocarbon and polyether resins of (A) and (B)
above are used in the practice of the present invention in fine
particle size form. It is recommended that the particle size
of the resins be about 1 to about 200 microns, preferably about
20 to about 120 microns.
As mentioned above, PFA is a preferred fluorocarbon
resin for use in thc: practice of the present invention.
Examples of commercially available PFA resins are TEFLON*-P
532-5012 PFA powder resin which is manufactured by E. I. DuPont
de Nemours & Company, Inc. of Wilmington, Delaware and which is
described in DuPont Fact Sheet TI-:14-84, and Neoflon* AC-5539
and Neoflon AC-5500 PFA powder resin, both manufactured by
Daikin Industries, htd. of Osaka, Japan.
DuPont recommends said 532-5012 resin for use as an
intermediate resin t=o be overlayed onto other primer systems,
followed by the application of other top coat resins overlayed
onto a coating formed from the 532-5012 resin. Daikin,
likewise, recommend: said AC-5500 o:r AC-5539 resins for use as
a top coat resin to be ov~~rlayed onto an intermediate resin
which, in turn, is overlayed onto a primer resin. (DuPont's
850-300 Teflon primer system, believed to consist of a chromium
oxide-containing po=Lytetrafluroethy:lene, is recommended by both
Daikin and DuPont a:~ a primer resin.) However, the
aforementioned 532-5012 resin and the AC-5500 and 5539 resins
have been found to be
*Trade-mark
i Tr
1341 01 ~
-15-
a quite acceptable resins for forming primer coatings when
used in accordance with t:he present invention.
Examples of other commercially available fluoro-
carbon resins for vse in the practice of the present invention
include AUSIMONT'S HALARF~ 6014 E-CTFE copolymer resin;
DuPont's TEFZELR 5~~2-600() E-TFE copolymer resin; and Kreha
Corporation of America's KF polymer. poly(vinylidine fluoride)
resin (PVDF).
With respect to polyether resins, they have
outstanding mechanical ;properties (flexural and tensile,
resistance to abra~,ion, wear and creep under load) and have
the best radiat ion res ist:ant propert ies of any plast is
material. PolyethE~retherketone (PEEK) is a particularly
useful polyether, 1-iaving a lower oxygen permeability and water
vapor transmission level than the most resistant
fluoropolymers. Pc~lyethE~rketone (PEK) and polyethersulfone
( PES) are also part icularly useful. .
Polyether~ resin:3 also have excellent adhesion to
metals as well as good compatibility with fluorocarbon resins
making polyether resins Excellent primers for top coatings
formed from fluorocarbon resins.
Turning now to a description of the property-
improving additive of thE~ present invention, they are also
known materials . F~referably, the part isle size of the
additive does not exceed the particle size of the resin (A) or
B ) const ituent . I:t appE~ars that the addit ive is present in
the fused coat ing in discrete part icle form. In the case of
an organic addit iv~~, the shape of polymeric part icles are
~34~~~6
-15a-
changed by the heat of 'the fusing process.
With res~~ect to additive (C) above, that is
poly(phenylene sulfide), examples of commercially available
PPS resins which cam be used are Ryton* type
* Trade-mark
~341~1
- 16 -
V-1, P-4 or P-6 as manufactured by the Phillips Chemical
Company or Bartlesv:ille, Oklahoma. The Ryton type V-1 PPS
resin is most preferred. The PPS should be used in fine
particle size form. It is recommended that the particle size
thereof be about 1 1~o about 200 microns, preferably about 10 to
about 100 microns.
With respect to the ceramic powder of additive (D)
above, this include; fine particle size, inorganic crystalline
materials. A cerami~~ powder is characterized typically by its
ability to be converted by sintering into a chemically inert
material. Examples of ceramic powders that can be used as
additive (D) above <~re: refractory carbides such as silicon
carbide, tungsten carbide, molybdenum disilicide and boron
nitride; metal oxidE~s such as alumina, chromic oxide, powdered
quartz, cerium oxidf~, silicon oxide, beryllia and zirconium
oxide; silicon nitride, titanium diboride and aluminium
diboride.
The ceramic powder can be in various forms, for
example, in the form of regularly or irregularly shaped
crystals, whisker fibers, long fibers, and platelets.
Metal carbide powders are a preferred additive for
use in the present .invention. The preferred carbides include
silicon carbide, zirconium carbide, tungsten carbide and boron
carbide, silicon carbide being most preferred.
A consideration in selecting the type of ceramic
powder to be used i;s its resistance to the corrosive effects of
the chemical materi,~l with which the resin composite material
is to be used. It is believed that alpha silicon carbide is
the most corrosive :resistant type of ceramic powder available
in respect to corrosive attack by a very broad range of
chemical materials. Thus, it is highly preferred. In
addition, silicon carbide is a low-cost material. However, for
x
1341p16
- 17 -
a variety of reason:, such as cost factors, etc., another type
of ceramic powder m<~y be selected.
Examples of commercially available silicon carbide
powders are 39 Crysi~olon green silicon carbide flour as
marketed by the Nori~on Company of Worcester, Massachusetts and
Arendahl SIKA SiC powder, marketed through Standard Oil
Electrominerals Co. of Niagara Falls, New York. These are
recommended for use in the practice of the present invention.
The ceram:ic powder to be used is preferably no larger
than the particle size of the resins with which it is to be
mixed and should preferably be within the size range of about 1
micron to about 40 microns, most preferably up to about 5
microns in size.
With respect to the fluorocarbon resin of additive
(D) above, such resin additive can be any one or more of the
fluorocarbon resins of (A) above. Other fluorocarbon resins
can be used also as the property-improving additive for use
with a polyether re:~in of (B) above. Such other resins are
those which include polymers that do not have hydrogen atoms
and in which there are at least three fluorine atoms for each
other halogen atom (for example chlorine) that may be in the
polymer. Polytetra:Eluoroethylene (PTFE) is an example of such
a fluorocarbon resin.
The particle size of a fluorocarbon additive for use
in admixture with a polyether resin of (B) above should
preferably be about 1 to about 200 microns, preferably about 20
to about 120 micron;.
With respc=ct to the non-resinous, property-improving
additives of (E) abcwe, such materials are selected ceramic
powders within the class of ceramic powders of (D) above. The
particle sizes of s,.~ch selected ceramic powders can be liked
__ X341016
- 18 -
those of the genera:Lly described ceramic powders mentioned
above. Similarly, and as also described above, the preferred
additive of (E) abo~,re is a metal carbide most preferably
silicon carbide.
With respE~ct t.o the polyether additive of (E) above,
the particle size thereof can be like those of the generally
described polyether resins mentioned above. The preferred
polyether resin additive of (E) above for use is admixture with
the fluorocarbon re:~in of (A) above is PEEK.
There fol:Low hereafter general descriptions of the
effects that the property-improving additives have on coatings
formed from composii~ions of the present invention and also
general observation: respecting the characteristics of coatings
of the present invention.
In genera:L, it :has been observed, most notably in the
use of ceramic powders, particularly with fluorocarbon resins,
that bond strength between the coating and an underlying metal
substrate increases with increased quantities of ceramic powder
in the composition. On t:he other hand, resistance to corrosion
by chemical attack .LS observed to be highest where relatively
small amounts of ceramic :powder are added to the resin,
corrosion resistance being observed to decrease as amounts of
ceramic powder in the resin are further increased.
As mentioned above, polyether resins have excellent
adhesion to metals as well as good compatibility with
fluorocarbons, making polyethers excellent primers for
fluorocarbon resin-based top coatings. Where the specific need
for the special properties of a polyether resin-based coating
is required in a mei~al protecting coating, polyether resins in
admixture with eithf=_r ceramic powders or fluorocarbon polymers
or both may be used to advantage. Silicon carbide is a
preferred ceramic powder and PFA is a preferred fluoropolymer.
1341 01 6
- 19 -
Wear and .Load-bearing properties of polyether resin-
based coatings are _Lmprov~~d by the addition of ceramic powders,
particularly silicon carbide.
Substitut=Lon of PES for PEEK or PEK is useful where
some temperature and chemical resistance can be sacrificed.
The major advantage of PES is its exceptionally low cost as
compared to both PEK and :PEEK.
Applications of composite mixtures of SiC-containing
polyether resins as coatings to metal roll surfaces which are
subject to high abrasion <~nd wear as well as high nip roll
loading at elevated temperature pro~;ride exceptional life
performance in resistance to damage in applications such as
papermaking, calendering and extrusion lamination, for example,
of plastics employed in packaging and similar industries.
Where release characteristics are desired, a
fluoropolymer, for example=_ PFA, may be added to the formulation
to impart release properties. Other applications for release,
corrosion-barrier, wear- and load-resistant coatings will be
evident to those with expc=_rience in end use application
materials.
In genera7_, the smaller the particle size of the
resin constituent, t:he bel~ter the properties of the coatings.
The major constituent of the composition of the
present invention i~~ a fluorocarbon resin and/or a polyether
resin of (A) and (B) abovf=_, the property-improving additive
being present in a minor amount. Although the additive can be
used in an amount approaclZing 50 wt. % of the composition, it
is preferred that the amount of additive comprise a lesser
amount. The property-imp=roving additive can be used in a bond-
improving amount, preferably about :L to about 40 wt. %. Such
amounts improve also the abrasion-resistance of the coating.
-2°- ~34~0~6
Additives providing an improved resin barrier coating which
inhibits corrosion of an underlying metal substrate should be
used in amounts of about 1 to about 25 wt. %, preferably about
1 to about 20 wt. %; and most preferably about 2 to about 5 wt.
o.
There fol=Low descriptions of preferred embodiments
within the scope of the present invention.
In a preferred embodiment of the present invention, a
perfluoroalkoxy (PF~3) resin is used to form a primer coating.
The primer coating, that which is directly in contact with the
underlying substrate, most typically a metal surface, is a PFA
resin, predominantl~r in a powder sire range of about 1 micron
to about 200 micron:, preferably predominantly in a range of
about 20 microns to about 120 microns, preferably modified with
the addition of a polyeth~=r.
A preferrE=d PFA-based primer coating may be formed
from polyether resin, preferably in an amount of about 2 to
about 40 wt. %, most: preferably about 5 to about 20 wt. %. A
very acceptable prirner coating can be prepared by mixing 15 wt.
% PEEK type 150 PF (Batch No. SPG9-:191p), as manufactured by
ICI America, Inc., with 85 wt. % of Neoflon AC-5500 PFA resin.
An alternative :PFA-based primer-coating may be formed
from PPS which is preferably present in an amount of about 2 to
about 20 wt. %, most. preferably about 5 to about 10 wt. %.
A very acceptable primer coating can be prepared from
a composition comprusing '7 wt. % of Ryton type V-1 PPS resin
and 93 wt. % of Neoi=lon AC-5500 PFA resin.
In another embodiment of the present invention,
ethylene-chlorotrif=Luoroethylene (E-CTFE) copolymer resins,
ethylenetetrafluoroethyle:ne (E-TFE) copolymer resins, or
.~ 134116
- 21 -
poly(vinylidine fluoride) (PVDF) resins are used to form primer
coatings, modified with the addition of a polyether and also a
selected ceramic powder, preferably a metal carbide, and most
preferably silicon carbide or zirconium carbide, in an amount
less than 50 wt. %, preferably in an amount of about 1 to about
25 wt. %, most preferably about 2 to about 20 wt. %.
In anothe:= embodiment of the present invention, a
composite coating i:~ formed from the aforementioned primer
compositions and an overlay or barrier top coating is formed
from a composition comprising PFA and ceramic powder, using as
the PFA TEFLON-P 53:?-5010 PFA powder resin which is marketed by
DuPont (Fact Sheet SCI-13-84.) The ceramic powder is used in an
amount preferably w_Lthin the range of about 0.5 to about 5 wt.
%, most preferably about 1 to about 3 wt. %.
Where the fluorocarbon resins are used to form a
corrosion-resistant barrier coating, they may be modified to
advantage with a se=Lected ceramic powder in an amount of about
0.5 to about 5 wt. '-'s, preferably about 1 to about 3 wt. %, and
most preferably about 2.5 wt. %.
In applications subject to abrasion and wear, an
outer top coating oj= any of PFA, E-CTFE, E-TFE and PVDF in
admixture with less than 50 wt. % o:f ceramic powder, preferably
silicon carbide or s;ircon.ium carbide, may be used to advantage.
A very acc:eptab:le polyether-containing primer coating
of E-CTFE can be prepared from about 10 wt. % 39 CRYSTOLON
green silicon carbide flo,ar (up to 5~ in particle size) and
AUSIMONT'S HALARR 6014 E-CTFE resin.
A very acceptable corrosion-resistant barrier top
coating of E-CTFE can be formed from about 2.5 wt. % of 39
CRYSTOLON green sil_LCOn carbide flour (up to 5~ in particle
'x
.. 1341016'
- 22 -
size) and about 97.5 wt. % of AUSIMONT'S HALARR 6014 E-CTFE
resin.
Additiona:L polyether-containing primer coatings may
be formed f rom about. 2 5 wt . % 3 9 CRYSTOLON s i l icon carbide and
DuPont TEFZEL 532-6000 E-'TFE copolymer. A 5 wt. % SiC-
containing coating of TEFZEL may be used to excellent advantage
as a corrosion-resi:~ting :barrier coating.
A composii~ion comprising 5 wt. % 39 CRYSTOLON silicon
carbide and 95 wt. '-'s Kreha Corporation of America KF polymer
PVDF resin may be u:~ed to advantage in forming a barrier
coating having exce=Llent corrosion-resistant properties.
In a most preferred embodiment of the invention,
there is provided a composite of a build up of PFA mixed with
the addition of about 1 to about 20 wt. %, preferably about 1
to about 5 wt. %, and most preferably about 2 wt. % of a
selected ceramic powder dispersed within the PFA resin, as a
top barrier coating,, overlayed onto and integrally bonded to a
primer coating of PFA/polyether which in turn is overlayed onto
and integrally bonded to a metal substrate, in particular a
metal substrate.
A primer or barrier coating may be formed also from a
composition comprisung PE:K, PEEK, o:r PES admixed with about 1
to less than 50 wt. %, preferably about 2 to about 25 wt. % of
a ceramic powder. ~Che composition can be applied to a metal
substrate in dry powder form, for example, by electrostatic
means used for fluor_opolymer-based coating compositions or by
other known methods such <~s, for ex<~mple, fluidized bed
methods, floccing methods, etc.
The addituon of SiC in the range of about 20 to about
25 wt. % to about 80 to about 75 wt. % of either PEK or PEEK
produces a polyether resin coating composition which can be
'x
1341A16
- 23 -
formed into a coating which exhibits significant reduction in
abrasion, wear and ~~reep under load relative to neat polyether
resin-based coating; which do not contain SiC. The addition of
chemically resistani~ SiC with its exceptional hardness enhances
the already superior mechanical properties of the polyethers.
Polyether resin-based compositions containing either
or both fluoropolymE=_rs and ceramic powders, for example, PFA
and silicon carbide, may be used to advantage when applied as a
coating to the chem:LCal seal and drive portion of agitators
employed in chemica:L vessels for mixing corrosive chemicals, as
such coatings have excellent chemical-resistance and very
desirable wear-, abrasion- and creep-resistance under load,
particularly at ele~rated temperatures.
The same preferentially applied polyether resin-based
composite is also u:~eful when applied to the tips of agitator
blades subject to h_Lgh abrasion and wear, particularly when
exposed to mixing liquids containing abrasives. In such
applications, prefe~~red coatings include a SiC-containing
polyether coating applied directly to the metal substrate or
over SiC-containing fluorocarbon resin-based primer coating.
The present invention encompasses applying an
undercoat of resin/additiwe composition to a substrate and
integrally heat bonding it to the substrate followed by the
application of successively built-up top coat layers and
integrally bonding each, :respective:Ly, to both the undercoat
and each preceding =_ayer of the top coat. The coating
composition may be applied in a dry powder form,
electrostatically, or by a wet spray system, or by other known
methods such as, for example, fluidized bed methods, floccing
methods, rotomoldinc~, and rotolining etc.
The present inv~=ration encompasses also a method of
applying coating cornposition by wet spraying to form both the
1341016
- 24 -
top coating and the primer coating, as well as the process of
forming the aforementioned as a barrier coating and also a
formulation for wet-spraying the coating composition.
PRE-APPLI(~ATION PREPARATION OF TOP COAT RESIN
In prepar:Lng the resin for application to the
substrate, a prefer:=ed procedure is described below for a
fluorocarbon resin <~nd ceramic dry powder mixture of PFA and
silicon carbide.
1. Five micron particle size SiC powder is weighed out
and sieved through <~ series of Tyler mesh screens to break up
agglomerations. The powder is first sieved through a 42 mesh
screen, second through a 100 mesh screen, third through a 325
mesh screen and finally through a 400 mesh screen. The sieving
operation is accomp:Lished by shaking the screens either
manually or preferably mechanically using an automatic sieve
shaker apparatus as is well known to those with skill in the
field. Two wt. o o:E the sieved silicon carbide powder is then
placed in a suitabl<~ container for roller mixing.
2. Ninety-ei<~ht wt. % PFA resin is then added to the
sieved silicon carb:Lde powder in the roller mixing container
and that container :LS sealed. A suitable container for mixing
is a polypropylene ;jar or bottle which can be obtained from a
variety of differeni~ sources as is well known to those with
skill in the field. It does not matter whether the resin is
added to the silicon carbide powder in the jar or the silicon
carbide powder is added to the resin therein. Whatever is most
convenient will suf:Eice.
3. After the two components are placed in the container
~~ ,r
X
-w 1341016
- 25 -
and the container i;~ sealed, the container should be shaken
thoroughly to dispe=rse and separate the particles of silicon
carbide powder to ensure that agglomeration does not occur.
4. The sealed container is then placed on a roller mill
and rolled for aboui~ 0.5 hour to fully and evenly mix the two
components together to produce a homogenous mixture of the
resin and the silicon carbide powder. An acceptable roller
mill for this purpo:~e is a Norton 735 RM Jar Mill, marketed by
the Norton Company, although many competitive devices are also
on the market and readily obtainable as is well known to those
with skill in the f:field.
5. The mixture, now ready fo:r application, is emptied
from the container into t:he fluidized bed hopper of an
electrostatic spraying apparatus. An acceptable electrostatic
spraying apparatus, including a fluidized bed powder container
(hopper), is the Ransburg GEMA 701 unit, marketed by Ransburg
Corporation of Indianapolis, Indiana, although a variety of
competitive models are available from other sources as is well
known to those with skill in the field. This unit is used to
apply the primer re:~in, a;s well as the overlay or top coat
resin mixture, the preparation of which has been described
above.
PREPARATION OF 1'~ETAL SUBSTRATE FOR COATING
Before an~r of the resins can be applied, the
substrate metal must. be prepared. 'Typically a mild, low carbon
steel substrate metal is most commonly used; however, various
other ferrous and non-ferrous metals may be used as the
substrate metal. It. is preferred to employ carbon steel as a
metal substrate because of its low cost, although the surfaces
of other metals may be coated if prepared to accept the primer
-26- 1341~~~
resins of the present invention. The preparation of the
substrate metal sur:Eace is substantially the most important
consideration, as distinguished from the species of metal to be
used as a substrate. Specifically, the substrate metal surface
must be cleaned such that it is free of oils, greases, blasting
grit, water and othE~r contaminants to the degree reasonably
practicable in gene=rally accepted shop conditions in the
coating industry. 'L'his can be done, for example, using
standard solvent cl<~aning techniques as are well known to those
with skill in the a..t. After the surface of the substrate
metal has been cleaned of surface contaminants, the following
procedure may be used for the preparation of a mild low carbon
steel substrate:
6. The steel piece is grit blasted with 3/0 (000) size
silica (sand) which is both fresh (unused) and dry.
Alternatively, a reusable blasting medium such as 24 grit
aluminum oxide can be used. After blasting, the blasted surface
should not be touched with bare hand; it is recommended that
clean gloves be worn. Care should be taken to avoid
contamination of the blasted surface from water, oil, grease,
dirt, etc. The bla:~ted s~~urface should also be inspected at
this point to find <~ny su:rface defects in the metal. If there
are any deep groove;, sharp edges, pinholes or weld defects,
such should be repaured at this point and the surface re-
cleaned and reblasted as above. The grit blasting roughens the
metal surface of the metal and, thus, enhances the bonding of
the primer resin thereto.
7. Within twelve (12) hours of the above described
blasting step, preferably sooner, the metal pieces should be
placed into an oven for heating. An example of an electric
oven which can be u;~ed and which was used in the examples
described hereinafter, is the Ramco Model RT-215 (Serial No.
1341016
- 27 -
813054) as manufact~.~red by Ramco Equipment Corporation of
Hillside, New Jersey.
8. Optionall~~r, the pieces may be blasted a second time,
this time preferabl:~ with an 80 grit size aluminum
oxide/titanium oxide grit, within about 0.5 to 1.0 hour before
they are placed into the oven. This second blasting is
recommended in humid weather to eliminate any rust which may
have formed on the ;surface since the first blasting step.
9. Whether o:r not one or two blasting steps are
utilized, just before the pieces are to be placed into the
oven, the pieces should be vacuum cleaned, using a suction type
vacuum cleaner, to :remove any residue of blasting grit.
Following the vacuum cleaning, the pieces should be thoroughly
brushed with a non-metal :bristle, non-shedding brush. The
blasting steps accomplish two things: firstly, the surface is
cleaned, and, secondly a surface texture is developed which is
most advantageous for and facilitates the bonding of the primer
resin thereto.
10. As the piE=ces are being loaded into the oven, a
thermocouple should be attached to each on a surface of the
metal which is not t=o be ~eoated with a barrier coating. These
thermocouples should be connected to standard apparatus to
enable the monitoring of the temperature of each piece within
the oven.
11. The bare ;~ubsta:nce metal pieces are then "baked" in
the oven. The oven temperature, being set at 760°F, the pieces
must be soaked in the oven for a sufficient time to bring the
temperature of the pieces up to 740°F, as the pieces will never
reach the set temperature of the oven, due to convection,
conductance, etc. a;~sociated with the design characteristics of
1341016
- 28 -
ovens of such type. In addition to this temperature
equalization step, preferably when the temperature of the
pieces has reached '740°F, a timer should be set so that the
pieces are "baked" :in an .air atmosphere for a period of at
least one (1) hour but not more than eighteen (18) hours (to
avoid unnecessary surface oxidation.) The purpose of this
extended "bake-out" time is to drive out essentially all of the
gases, organics and other contaminants which may be trapped
within the interstii~ial metal structure thereof. Higher "bake-
out" temperatures and/or "bake-out" periods may be used. Once
the "bake-out" is finished, the pieces are now ready to be
removed from the oven and sprayed with the primer resin.
APPLICATIN OF FLUOROCARBO:L~
RESIN AND ADDITIVE MIXTURE
The spraying of the resin powders, both the primer
resin and the top coat resin mixture, requires particular care
as control of the substrate temperature ranges are important.
Also control of the ranges of thickness of the coats is
important. Finally, it is important that the thicknesses of
each coating be conl~rolled within defined ranges, from one
section to another <~cross the surfaces of the substrate metal
pieces; that is to ;gay that the spray application of any given
layer of coating should be controlled such that it is not too
thick at any given ~~oint and/or not too thin at any given
point. A procedure used for the application of a PFA/PPS
primer coat to the metal substrate, following "bake-out" of the
pieces is preferable as follows.
A. PRIMER LAYER APPLICATION
12. The piece~~ should be removed from the oven with the
1341n16
- 29 -
temperature of the ~~ieces being at least about 700°F, with the
thermocouples still attached to each piece. The first spraying
of the primer resin should be commenced preferably within
thirty (30) seconds from the time the pieces are removed from
the oven with the temperature of the pieces preferably not
being lower than about 680°F, although it is quite acceptable
that the finishing t=ouches of the spraying may be added as long
as the discrete section of each piece, which is then being
sprayed, is above the resin melting point. Spraying may be
done, simply, but not preferably, until no more of the powder
primer resin melts onto t:he pieces. The melt range of the
PFA/PPS resin mixture is 575°F to 600°F, but the powder resin
will not normally melt as it hits the metal unless the metal is
at about 600°F or above. The thermocouples attached to each
piece only show the temperature of that portion of each piece
which is immediatel~r adja~~ent to the thermocouple attachment
point, while other ;~ectio:ns of each piece may be higher or
lower in temperature, depending on the rate of cooling of each
discrete section of each piece. Thicker sections will cool
more slowly while thin se~~tions wil:1 cool relatively rapidly.
Preferably, the thickness of the primer resin will be within
the range of about 0.002" to about 0.005", although primer coat
thicknesses in the range of about 0.001" to about 0.025" have
been found acceptable. The setting on the Ransburg GEMA 701
electrostatic spray apparatus will vary according to the size
of the pieces, the thickness of each discrete section of the
substrate metal, and the geometry of the pieces. An
appropriate setting for a 1/4" x 8" x 8" mild steel plate is 40
Kv. The primer rerun is preferably applied in a single
spraying, forming a single layer, although in some
circumstances an additional layer or more may need to be
applied. If such i:~ nece;ssary, additional coats of primer
resin may be applied as follows.
X341 X16 '
- 30 -
B. APPLICATIC7N OF .ADDITIONAL "PRIMER" LAYERS
13. After the first layer of primer resin has been
sprayed onto the pic=ces, additional layers may be applied but
are not necessary to improve bond strength. Such additional
layers may be used i.o provide a transition between the metal
substrate which characteristically has a low coefficient of
expansion and top coat material which has a higher coefficient
of expansion. If ail additional layer of primer resin is to be
applied, the pieces are then returned to the oven. The oven
should be set at about 700°F. When the temperature of the
pieces has reached '700°F and all portions of the first coating
have reached the me_Lt phase, as is determined by visual
inspection through a window in the oven (usually requiring a
soaking of about twenty (:20) minutes), and if it is determined
that one or more additional layers of primer resin are
necessary, the pieces are again removed from the oven and a
second layer of prirner resin is sprayed on, overylaying the
first layer. Becau:~e of notably poor heat transfer in
fluorinated polymer;, the first layer will hold the temperature
of the pieces allowung ample time for the second layer to be
sprayed on at this temperature. The actual surface temperature
of the first coat may drop to about 650°F or less, but should
not drop below about. 600°F. The objective is to obtain a primer
resin build up of about 0.002" to about 0.02". In applying the
second layer (and any necessary additional layers) of the
primer resin, the Ransburg GEMA 701 electrostatic spraying
apparatus may be set. at about 30 to 40 Kv. for a 1/4" x 8" x 8"
piece of mild steel..
x
1 3 41 0 1 6 "~
- 31 -
14. Whether o=r not additional layers of primer resin are
applied over the fi=rst layer of primer resin, following the
last layer of prime=r resin sprayed on, the pieces are returned
to the oven which i:~ set at 700°F until inspection determines
that the last layer of primer resin has reached the melt phase.
A procedu:=a used for the application of PFA/SiC top
coat to a "primed" metal substrate is preferably as follows.
C. APPLICATION OF 'TOP COAT LAYERS
15. The pieces are :removed from the oven and sprayed with
the first layer of topcoat resin mixture, using the Ransburg
GEMA 701 electrostatic spraying apparatus, which may be set at
about 30 Kv. for a .L/4" x 8" x 8" piece of mild steel. Care
should preferably be taken to ensure that the temperature of
the pieces being sprayed ahould always remain at or above the
575°F to 600°F melt range of PFA. The thickness of each layer
of top coat resin which ins sprayed should preferably be within
the range of about 0.006" to about 0.01", although layer
thicknesses within t:he range of about 0.001" to about 0.015"
have been found to be acceptable.
16. After the first layer of top coat resin mixture has
been sprayed onto tree pieces, they are placed back into the
oven and heat soaked unti:L the just-sprayed resin coat has
fully melted.
17. SuccessivE: layers of top coat resin mixture are
applied in the same manner, following the specifications set
forth in Step. Nos. 15 and 16 above. The objective is to form
an overall barrier <:oating on the pieces which is at least
0.040" thick but wh__ch may be of a greater thickness. Thus,
as many layers of top coast resin mixture are applied as are
x
-32- 134101fi
necessary to achiev<~ such a thickness. After the last layer of
top coat resin is a~~plied, the pieces are first reduced in
temperature to 550°F' in the oven, by shutting the oven off but
continuing to circu:Late air in the oven with the oven blower.
When the pieces reach 400'°F, they are removed from the oven and
cooled to room tempE=rature, thus being ready to place into
service.
WET SPRAY APPLICATION
Where the=re is a need to form a barrier coat type
composite, in accordance with the present invention, in
relation to the sub;~trate metal of relatively large metal
apparatus, such as banks and pressure vessels, there is a
problem in applying the fluorocarbon resins to the substrate
metal. Normally, in such apparatus, the barrier coating is
needed on the inside walls of tanks, vessels and the like. To
follow the procedures described above would necessitate manual
spray of the interior of a tank or vessel which is at a
temperature in exce;~s of 600°F. It would be impossible to place
a person inside a tank or vessel at such a temperature to
effect the required spraying operations.
It is envisioned that such spraying of the interior
of hot vessels migh~~ be effected by the application of robotics
technology; however, until such is developed, an alternate
approach has been followed for spraying the fluorocarbon
resins, and mixture; thereof with fine crystalline ceramics,
onto substrate metals at ambient temperatures, while still
achieving the high integrity bonding of those resins and resin
mixtures to the substrate metal, and to each other,
substantially free ~~f voids (porosity) .
1341 01 6
- 33 -
To develop the :barrier coating of the composite of
the present invention, each layer of both the barrier resin and
the top coat resin rnay alternatively be applied to a piece when
that piece is at ambient (room) temperature. However, each of
those layers is spr<~yed o:n wet, rather than as a dry powder,
and electrostatic spraying apparatus is not used. A preferred
method of developing the composite of the present invention,
using wet spraying, is as follows.
A. PREPARATION OF .AQUEOUS DISPERSION
Preliminary, a top coat resin mixture in a water
suspension is prepa=ed most preferably comprising: 96.04 wt.
of PFA resin; 0.04 ~Nt. % of Dow Corning Anti-Foam A antifoaming
additive, marketed by Dow Corning Corporation of Midland,
Michigan; 1.96 wt. 'o of 5,~ silicon carbide; and water (in the
laboratory, de-ioni:~ed water may be used but in production,
standard "softened" water, with the minerals removed, is quite
satisfactory.) To prepare a sample of the mixture in a water
suspension, 100 ml of refrigerated, de-ionized water is
deposited into the mixing container of a blaring model 34BL21
high speed blender. The :blender is then turned on and the
speed is adjusted to the :highest speed which will still
maintain a smooth vortex without splashing. Two grams of Triton
X-100 are added to i=he blending water with a standard eye
dropper, a drop at <~ time. Then add 0.05 gram of Anti-Foam A
x
134~~16
- 34 -
to the blending watf~r and reset the Variac control on the
blender to 20-30V. Mix t:he solution at this speed for one (1)
minute. Then reset the blender Variac to 60V and slowly add 2
grams of silicon carbide to the solution. Then set the blender
Variac to 60V-70V and add 98 grams of PFA resin slowly to the
center vortex of the blending solution. If the PFA resin does
not disperse, add ac3ditio:nal water to the blending solution in
5 ml increments until the PFA resin disperses. The dispersion
of the PFA resin is aided by using refrigerated water, although
this is not a neces:~ity. It has been found that this
formulation, in modified form, produces acceptable coatings
where as much as 99.9 grams of PFA resin are added and as
little as 0.1 grams of silicon carbide powder is mixed in.
B. PRE-APPLICATION PREPARATION OF METAL SUBSTRATE
Secondly, the piece to which a barrier coating is to
be bonded is prepared in .exactly the same way as specified
above in No. 6 through 11, with the prior cleaning step, prior
to sand blasting, a:~ specified above, included. Once the
"bake-out" procedure has :been completed, the piece is removed
from the oven and air cooled to below 100°F.
1341016
- 35 -
C. WET SPRAY APPLICATION OF PRIMER COAT
Thirdly, <~ prim~sr resin coat should be applied. A
preferred primer re;~in coat may be the PFA/PPS resin mix
described above or .Lt may be the DuPont 532-5012 resin. Either
of these may be app=Lied electrostatically using the Ransburg
GEMA 701 electrostat=ic spray apparatus by spraying a preferred
thickness of 0.002" to 0.005" per layer of resin to the cold
(room temperature) piece. Then the piece, with the
thermocouple attached thereto as described above, is inserted
back into the oven, set at 700°F. The piece is brought up to
temperature and held there until all portions of the primer
resin layer have reached the melt phase, as determined by
visual inspection through a window in the oven following this,
1341016
- 36 -
the oven is shut ofi_ with the oven blower still running, and
the piece is slow cooled to 550°F, as described above. Finally,
the piece is removed from the oven to rapidly air cool to below
100°F for application of t:he next layer of primer resin, if
necessary. Succeeding layers of electrostatically sprayed
primer resin layers may b~~ applied in exactly the same way.
Applying the primer
'x
1341 01 6
37
res,~n , n this mar.n~ar .r,ay require multiple layers as
the sprayed on czry powder, to some extent, tends to
fall off of the piece, decreasing the thickness of the
layer.
A1'~heugh the electrostatic dry spray method is
preferred, alte.~nati.vely, the primer resin may be
applied wit. L:~>ual_ly, a 0.00?." to 0.005" thi.ckriess of
primer resv~n cant can be applied in a single layer.
mhe top coat re:~in misaure in a water suspension,
described a:Jove, may, in a slightly modified form, be
used as a primer re~_,in. For example, in the case of a
PFA/PPS pr=L"'.er res,~n described above, the only change
necessary _is to reduce. the 96.04 wt. o of PFA resin to
90.00 opt. -'~ anci to add G.04 wt. o of PPS resin. In
preparing a sample mixture, as a preliminary step, 6.2
grams of PPS resin are thoroughly mixed and blended
~~;ith 91.3 grams of :~?I~'~, resin using a roller mill as
described above. T.nen the sample mixture procedure
for the water :~uspensi.on, descr,~bed above, is used
2C except that the PFA/PPS resin mixture is added instead
of the straight PFA resin. The silicon carbide may
be, aptian.ally; taker, out in the primer resin mixture
water suspension.
Prior to spraying on the primer resin mixture
water suspe.~,sion (pri.:~er wet spray) , the "baked-out"
piece, prepared as described above, is removed from
the furnace anal air cooled to 1_ess than 100°F. The
primer wet spray is loaded into a one (1) quart pot
for a Bin'Ls model 1'~ ;pray gun which is used to apply
the primer ;aet ~~pray to the piece, as marketed by
Bin;cs manufacturing Company of l~rarklin Park, Illi-
nois. Prefera._ly, the Binks model spray gun is
equipped with ~ Bin:K~> No. 66SS fluid nozzle, a Binks
~'o. 66SF air nozzle and a Binks No. 65 needle. The
38
primer wet spry in the one (1) quart pot should,
preferably, be shaken 'intermittently, but frequently,
to keep ti:e solids ~._ the primer wet spray in suspen-
sion. mhe air pre~;sure used to apply the primer wet
spray should, preferably, be within the range of about
40 psi to aboL:t 50 p"i.
Tn spray~_ng the piece, new below 100°F in
temperature, preferably at room temperature for ease
of handl i ng, t=he sprG.y should be applied evenly, first
to the critical areas w::ere complex relatively sharp
curvatures, corners, etc. ex it, then to the relati-
vely more ;month, 1e~>s curved, flat, etc. areas. Care
should be tal~:on to avoid running and overspray of the
primer wet spray on the piece. Also, as mentioned
1:~ before, care must beg taken to see that the solids
suspended ,~n the wager, in the primer wet spray, do
not separate during the spraying operation. All
spraying of the primer wet spray should preferably be
done without a break in the operation. Stopping the
operation. will alloH; the primer wet spray on the piece
to dry. When this occurs and the spraying is recom-
menced, the dried material can easily be blown off by
the atomizing air of_ the spray gun. Thus, it is
recommended that the spraying always be done against a
trailing wet edge, as is well known to those with
skid in the vfield. An even ~_ayer of primer wet spray
should be app?ied by spraying by a steady, even
movement or t:~e spray gun. The primer wet spray
shcuid be applied until a single layer in a range of
about 0.00?_" ~o O.Ot?5" thickness is built up, prefer-
ably in a range of about 0.00" to 0.005" as measured
in the wet co~dition. A Nord~~on Wet Film Thickness
Gauge, as mar'~:eted by Nordson Corporation of Amherst,
1341016
39
Ohio, may be used t:o determine the thickness of the
wet sprayed layer.
Once the primer wet spray coating has been
applied, a~ a sing:Lc layer, the piece is preferably
air dried for abo~,~t fifteen (1.5) minutes. Then the
piece shoo-d ire pl aced :in a preheated oven set at
350 ° F, wit'_: a -thermocouple at.t:ached as explained
above. After_ the ~z_ece is so placed in the oven,
without delay, the oven should be reset to 720°F.
When the temperature of the p~_ece reaches 700°F and
the primer wed spr:ay coating, fully dried, and all
surfaces thereof have reached the melt phase, as
determined visually, the oven is turned off, with the
oven blower still r~.nnir~g, and the piece is cooled in
the oven until it reaaches 550"F. At this point, the
piece is removed from the oven and air cooled to below
100°F, preferably to room temperature, for application
of the top coat wet spray.
D. WET SP_T2AY APPLICATIO:v' OF TOP COAT SUSPENSION
The top coat wet spray may be the top coat resin
mixture in a water suspension described above, without
FPS resin and defin_~'~ely with silicon carbide in-
eluded. The v.op coat wet spray is sprayed onto the
piece, prerer~.bly at= room temperature, in layers,
exactly follor'ing the spraying techniques described
above for the prime~~ wet spray resin, except that the
thickness of each layer of top coat wet spray is
greater, preferably :in the range of about 0.010" to
about 0.01", with due care being taken to avoid
running and o-~~erspray. The objective is to build up
a:~ overall ba~~rier coating of at least about 0.040" in
thickness.
._
- 40 -
After each layE=r of top coat wet spray is sprayed
onto the piece, the piece is air dried for fifteen (15) minutes
and placed into a ~>reheat~ed oven set at 350°F, with the
thermocouple attached as described above. Then, without delay,
the oven is reset for 650°F and the temperature of the part is
brought up to the ~~oint where visual inspection assures that
the now dried top coat wet spray layer is in the melt phase, or
until the piece temperature reaches 620°F, whichever comes
first. If the melt phasE: is reached before the piece
temperature reaches 620°F, preferably, continue to heat the
piece for about ten (10) minutes or until the piece has reached
620°F, whichever comes first. Then the oven is shut off with
the blower still running and the piece is cooled to 550°F,
followed by removal of th.e piece from the oven for air cooling
to below 100°F. Th=~s heating-cooling cycle is repeated for each
layer of top coat wet spray applied.
Additional advantages of the present invention are
that the use of a property improving additive, as described
herein, is effective in increasing the rate at which the
coating composition can be fused. By way of background and as
exemplary of the aforementioned, it is known to those with
skill in the field, as exemplified by the two (2) DuPont Fact
Sheets mentioned abcwe, that it is difficult to get good fusion
or sintering of pari~icles of PFA resin to each other, or to
primer systems over metal substrates, at temperatures below
about 700°F within time periods that: can be used practically in
a commercial setting. Thus, it is ~~onventionally recommended
that PFA resins should be fused at about 725°F for about 20
minutes. However, .Lt is also known that PFA, as a fluorinated
polymer, deteriorates (degrades) relatively quickly at
1341016'
41
temperatures of aboL:+. 700°F and above, as fairly rapid
oxidaticn occurs. 'Thus, the objective is to effect a
complete fusion of the resin particles before the
resin. itself degrades. It can be appreciated that it
is relatively diffic::lt to get good bonding of PFA
resin. to metal substrates and to get one layer of PFA
resin bonded t.~ another.
By way of example, neat powdered PFA resin,
app lied directly to a 8" x 8" x 1/4" mild steel plate
brought to a temperature of about 725°F, fuses com-
pletely in abo,~t 20 minutes; at 675°F, fusion is
completed in about 30 minutes; and at 620°F, fusion is
completed in about 40 to about 50 minutes. The
addition of pc;adere.d ceramic material to the resin
1F significantly reduces the time necessary to complete
the fusion of i:he resin particles. When powdered PFA
resin and cermaic material is applied to a mild steel
plate heated to abcut 725°F, fusion of the resin is
completed in about 10 minutes; at 675°F, fusion is
completed in about 15 to about 20 minutes; and at
620°F, fusion '_s completed in about 30 to about 40
minutes.
Thus, a preferred method for forming a fused
coating from t_ue composition of the present invention
includes heating said composition to a temperature for
a period of time no =longer than a predetermined period
of time, said temperature being at least 25 F° below
the temperature at which the resin of said composition
in neat form can be f.'usec~ completely by heating for no
longer than said precLetermined period of time without
substantially degrading said resin.
It has also been observed that where neat
powdered resins are applied to an already resin-coated
metal substrate, the time to complete fusion is
1341A~6
42
increased to even longer periods than those just
recited above. I:t is suspected that this longer
fusion. time ~~s due to the heat insulating property of
the previous_Ly applied neat resin layer. Thus, to
effect compic~te fusion. of the top-most resin layers,
it is necessary to prolong the period over which the
plate is heaved, they-eby degrading the bottom-most
resin layers in closest proximity to the substrate
surface.
The addition of ceramic powders to the resin
speeds t:he fusion process, perhaps due to the improved
heat conductivity of ceramic powder-containing resins.
Thus, a'~1 of the resin layers may be brought to fusion
temperature :r.ore quickly, reducing the exposure time
and concomit:~nt degradation of the bottom-most layers.
It is also n.nown that it is relatively difficult
to achieve b~~ndirg of straight PFA resin to metal
substrates, regardless of the temperature of fusion or
sintering us?d. It has now been determined that the
addition of the above specif=ied quantity ranges of PPS
resin to the PFA resin, surprisingly results in a very
high quality integral bonding of the PFA resin to
substrate metals at temperatures in the range of about
675°F to about 720°F without significant deterioration
(degradation) to the PFA. It is also now known that a
very similar phenomenon occurs in dry sprayed PFA
resin without the addition of PPS resin, but with the
addition of the specified quantity ranges of ceramic
powder, in those same temperature ranges. No explana-
tion of why these phenomena occur is known and no
speculation. thereof is offered herein.
134~~~6
43
EXAMPhES
EFFECT OF ADD=CNG SiC TO PFA
TO ELIMINATE RESIN BUBBLING
In the e:fampl-es which follow, Sample Plates (A-G)
were fabryca'<~d to determine the effects of adding
silicon carbide to PFA resin in respect to the
occurrence of bubbling of top coat layers during the
build up of the barrier coat. In all cases, DuPont
1.0 TEFLON-P 532-~>0~2 PFA resin was used as the primer
resin and Dur~~ort TEFLON-P 532-5010 PFA resin mixed
with silicon c~arb~.de powder (as indicated) was used as
the top coat resin. The sample plates were formed
using 1/~'r" x a3" x. 8" size mild steel plates and
1.5 composites were formed thereon. in accordance with the
foregoing procedures. The results are as follows in
Table 1.
TABLE 1
a:0 Sample
Plate A B C D E F G
Pr imer
Thickness .021 .02_2 .021 .018 .024 .017 .001
25 (inch)
No. of
Layers 2. 2 2 2 2 2 1
-s0 =----____-_--:--_-___-----__--_--_-____________________
Top Coat 100 99.75 99.5 99 98 98 99
PFA wt.o
.5 SiC wt. % 0 0. 2.5 0. 50 1 2 2 2
(Particle
Size) - 5~.c 5h 5~C 5~t 5~ 5u
~.0 Thickness .008 .007 .007 .007 .008 .040 .018
(inch)
1341p~6
44
No. of
Layers 1. 1 1 1 1 7 3
Steel Temp.
(°F) 675 675 675 675 675 675 675
Barrier Coat .029 .029 .028 .025 .032 .057 .019
Total Thick--
;inches)
Bubblinc_~ Yes Slight Very No No No No
Slight
It can }~e seen from the results reported in Table
1 above, tha': when sample plate A was sprayed with a
top coat of :L00~ PFA, it exhibited bubbling after a
sing a 7_ayer of top coat was applied. With the
addition of 0.25 wt. % of silicon carbide to the PFA
top coat, tre bubbling was significantly reduced with
the applecat:ion of a single layer of top coat as shown
in Sample Plate i_s. The single layer of top coat
applied to Sample Plate C had 0.5 wt. o of silicon
carbide added, and the bubbling was reduced to a point
where it waC just barely noticeable upon visual
inspection. In ~>ample Plate D, 1.0 wt. o of silicon
carbide was ;~ddecl to the PFA top coat, and in the
single layer app~.ied, no bubbling was detected. In
Sample Plate E, 2.0 wt. o of silicon carbide was added
to the PFA t~Jp coat, and in the single layer applied,
no bubb'_ing was detected. Sample Plate F was prepared
identically vo Sampie Plate E, except that 6 addition-
al layers of top coat were applied for a total of 7;
again, no bubbling was detected.
6
In the preparation of Sample Plates A-F, a
relatively thwick primer coat was applied, using two
layers in each case. To determine whether or not a
reduction of the thi.c:kness of the primer coat, and a
single layer applicat=ion thereof, would generate
bubbling in the PFA top coat, with 2 wt. % silicon
carbide added, Samp1_e Plate G was prepared, with a
single layer cf primer coat, only .001" in thickness,
being appl;.~ed and then ove.rlayed with 3 layers of
10 PFA/2 wt. ~ S_C mixture top coat resin. Again, no
bubbling was c.etectod. It should also be noted in
regard to Samr~le Flutes D and E that thicknesses of
0.007" - 0.00." of )?FA/SiC mi:ftures, as a top coat,
were applied in single layers without bubbling,
15 whereas, it i_=> known to those with skill in the field
that straight PFA cannot be applied in coating layers
greater than 0.003" thickness without the occurrence
of bubbling. This :is pointed out in DuPont Fact Sheet
TI-13-84 referenced above.
20 EFFECT OF SUBSTRATE TEMPERATURE
Several <zdditional Sample Plates (H-L) were
fabricated wherein composites were formed on under-
lying mild steel substrates in accordance with the
25 foregoing procedures. The object of developing these
samples was t« determine the effect of lowering the
metal substrate temperature in respect to the occur-
rence of bubbling of top coat layers during the build
up of the barrier coat. Again, in all cases, DuPont
30 TEFLON-P 532-5012 PFA resin was used as the primer
resin and DuP~~nt ~'EFLON-P 532-5010 PFA resin mixed
with silicon carbide powder (as indicated) was used as
the top coat resin. The Sample Plates were formed
134~p~ fi
46
using 1/4" x ~" 8" steel The
x size plates.
mild
results are a_, in Table2.
follows
TABLE
?.
Sample Plate H I J K L
Primer Thickness .0:L7 .024 .010 .022 .011
(inch)
2v'o. of Layers 2 3 2 2 2
Toy Coat :L00 99.5 99 98 99
PFA wt.
SiC wt.% 0 0.5 1 2 1
(Particle size) - 5u 5u 5u 5u
Thickness .015 .031 .006 .008 .034
(inch)
No. of Layers 3 5 1 1 7
Steel Temp. ("F) 625 615 625 625 625
Barrier Coat .032 .055 .016 .030 .045
Total Thickr.c~s
"_-5 (Inch)
Bubbling Yes No No No No
4.0 In Sa:npl~~ Plates I-L, where silicon carbide was
mixed with the PFA resin of the top coat, no bubbling
was detecaed where the steel substrate metal tempera-
ture was only raised to a range of 615°F to 625°F
while significant bubbling was noted in Sample Plate H
~.5 which used straight PFA resin, without silicon
1341o~s
47
carbide, as ;she top coat where the steel substrate
metal temperaa~ure was only raised to 625°F.
EFFECT OF' SiC PAR~'ICLE
SIZE ADDED TO TOP COAT
Several additional Sample Plates (M-R) were
fabricated wherein composites were formed on underly-
ing mild stee:L substrates in accordance with the
foregoing pro~:edures. The object of developing these
7_0 samples was to determine the effect of adding larger
sized pa~-ticl~~s of: silicon carbide powder to the PFA
top coat resi.:~ in respect to the occurrence of
bubbling of t~~p coat layers during the build up of the
barrier coat. Again., in all cases, DuPont TEFLON-P
._5 532-5012 PFA :resin was used as the primer resin and
DuPont TEFLON-P 532-5010 PFA resin mixed with silicon
carbide powder (as indicated) was used as the top coat
resin. The sample plates were formed using 1/4" x 8"
x 8" size mii_~3 steel plates. The results are as
a?0 follo~,as .in Ta:bl a ~3 .
TABLE 3
Sample
Plate M N O P Q R
Primer
Thickness .010 .021 .020 .020 .020 .020
(inch)
No. of
Layers 1 3 2 2 3 2
Top Coat 90 99.5 90 80 99 98
PFA wt . o
SiC wt.% 10 0.5 10 20 1 2
~~0 (Particle
Size) 7u 7~ 7u 7~C 14~C 14~t
~3'~~~16
48
Thickness .035 .026 .040 .040 .021 .021
(inch)
NO. Of
Layers 6 5 6 10 4 4
Steel Temp.
(°F) 680 615 600 650 620 620
Barrier Coat .045 .047 .060 .060 .041 .041
Total Thick
ness (Inch)
Bubbling No No No No No No
In regard to Sample Plates M-P, silicon carbide
powder of 7~~, size was mixed with PrFA resin and applied
to form the t:op coat with no bubbles detected therein.
The top coat, of S<~mple Plates Q and R had 14~. sized
silicon carb~_de powder added to the PFA resins before
application. Also, it should be noted in regard to
Table 2 that the temperature of the substrate steel
metal was varied i:~ a range from 600°F to 680°F.
Finally, it :>hould be noted i.n regard to Table 2 that
the wt. ~ of silicon carbide which was mixed with the
PFA resin. to form 'the top co:~t resin mixture was
varied i.n the range of 0.5 wt. % to 20 wt. ~. In none
of these cases, reported in Table 5, did any visibly
detectable bubbling occur.
'34~~16
49
CORROSION RE:~ISTANCE DETERMINATION
FOR FLUOROCA:~BON POLYMER BARRIER
COATING CF THE PRIOR ART
British Patent No. 2,051,091, believed to be the
detailed specification for developing Fluoroshield
coatings, te:?ches using a composition comprising a
mixture of P'rFE (polytetrafluoroethylene) resins, and
PFA resin to form overlays or_ barrier coats. These
dry powder resin mixtures are mixed with a carrier
liquid and, it appears in commercial applications,
glass powder, for wet spray applications. British
Patent No. 2_,051,091 teaches that "To obtain non-
porous coatings it is necessary to densify the applied
coating. This may be accomplished by rolling the
coating prior to heating the coating to coalesce it."
It also teac'zes that '°...a pure PFA coating is
unsuitable a:nd failed to provide a uniform, non-porous
coating."
Althoug":~ British patent No. 2,051,091 discloses
several exam:~les, of what are now known commercially
as Fluoroshield coatings, which were spark tested at
10,000 volts, none of the examples in that patent
discloses actual corrosion tests.
One apparatus which is widely accepted by those
with skill in the field for testing corrosion is the
Atlas Cell, as marketed by Custom Scientific Glass,
Inc. of Elkton, Maryland, U.S.A. Basically, the Atlas
Cell tests materials or exposed material surfaces to
the effects of corros~_on at either ambient or elevated
temperatures, as desired, and for extended periods of
time as dess_red. Various tests have been made on
Fluoroshield coated mild steel samples, with one side
of each being Fluoroshield coated, exposing only that
Fluoroshield coated surface of each of those pieces to
the tests. All test samples were commercially
fi~~~Ot6 _
acquired specimens of Fluoroshield coated mild steel
apparatus. .?~11 samples of the tested Fluoroshield
coated contained finely ground glass powder. The
samples and the test results are as follows in Table
5 4.
TABLE 4
Fluoroshield Corrosive
10 Sample No. _ Material Temperature Time
1. 70 wt. % 252°F 600 Hours
Nitric Acid
15 2. 20 wt. % 220°F 600 Hours
Hvd.rochloric
Acid
20 3. 70 wt. % 252°F 1,000 Hours
Nitric Acid
4. 20 wt. % 220°F 1,000 Hours
25 Hyc!rochloric
Acid
30 The evaluation of each of these samples, follow-
ing the abo~~ a to st.~>, was by subj ective observation, in
each case relative to a sample in accordance with the
present ,~rvention, as described hereinafter, and
comprise the opinions of the inventors of the present
35 invention. That s_<_> to say that each Fluoroshield
coated test sample, recited above, was tested under
equivalent conditions used t:o test a corresponding
non-Fluoroshield c:oat.ing test sample, in accordance
with the present invention, and compared thereto. The
40 results of such tests of non-Fluoroshield coated test
samples are deta.il_ed hereinafter in Tables 6-8. The
51
subjective o~~servations in regard to each Fluoroshield
coating sample are as follows in Table 5.
TABLE 5
Fluoroshield
Sample No. Evaluation
1. Coating severely blanched and fully
blistered.
2. Caating blanched and blistered, slight
degradation of underlying steel
evident, slight delamination evident.
3. Coating severely blanched and fully
blistered, almost total delamination
evident, coating substantially pealed
away from substrate metal, substantial
degradation of underlying substrate
metal evident.
4. Coating severely blanched and fully
blistered, degradation of underlying
steel evident, substantial delamination
evident.
In each of the foregoing Fluoroshield samples, the
blistering ~_s an indication that the overlayed barrier
coating may not be firmly bonded to the underlying
substrate steel and that the coating is beginning to
separate. The blanching or discoloration is an
indication that the corrosive medium has penetrated
into and even through the pores of the overlayed
barrier coating, attacking both additives which have
been mired into the coatings as well as, possibly, the
underlying Cubstrate metal. Delamination indicates
that the bonding of the Fluoroshield coating to the
substrate metal has failed.
'341D16
52
As is demonstrated by the Fluoroshield coating
tests, above, there is substantial room for improve-
ment in both the integrity of the bond between the
overlayed barrier_ fluorinated polymer coating and the
underlying substrate metal, in particular, steel.
Also there is substantial room for improvement in
diminishing porosity in the overlayed barrier fluori-
nated polymer coat.i.ng, irrespective of spark test
evaluations.
The use of At.l.as cell testing for corrosion
testing has been briefly described above in regard to
the testing of Flu.oroshield coated mild steel samples.
An Atlas ce,~l is arranged in the form of a hollow
cylindrical section with both ends being open. Ports
extend through the sides of the cylindrical section
through which instrument sensors and heating elements
are inserted into the hollow of the cylindrical
section. The open ends of the cylindrical section are
capped with the test samples which are to be exposed
to corrosion. Such test samples are normally in the
form of fiat plates, one surface each of which is
abutted against an open end of the cylindrical
secticn. The open ends of the cylindrical section are
capped T,aith the test samples which are to be exposed
to corrosion. Such test samples are normally in the
form of flat plates, one surface each of which is
abutted against an open end of the cylindrical section
and clamped or otherwise mounted thereto so that the
joint between is ~;ealed. Because there are two (2)
open ends to the hollow cylindrical section, two (2)
test samples are used to cap those respective open
ends, thus two (2) test samples are simultaneously and
concurrently subjected to testing by each Atlas cell
test program.
1341 Q1 6
53
Atlas cell test programs are normally set up to
test the effects of a corrosive liquid medium, for
example, an acid, which is introduced into the hollow
cylindrical section after the open ends thereof have
been capped .end sealed with the test pieces. The
corrosive liquid is introduced through one of the open
ports e:ltending through the wall of the cylindrical
section, after which the ports are sealed. The
corrosive li.~uid, either heated or at ambient tempera-
ture, is left within the Atlas cell for extended
periods of time amounting to several hundreds of hours
or more.
In each of the above reported 1,000 hour Atlas
cell tests of Fluoroshield coated mild steel samples,
Fluoroshield Sample Nos. 3 and 4, the opposite end of
the Atlas Cell was capped with a sample plate which
was a composite according to the present invention.
These sample plates, in accordance with the present
invention are described hereinafter as Sample Nos. 13
and 14 which correspond, in their Atlas cell testing,
to Fluoroshield Sample Nos. 3 and 4 respectively. The
sample numbering system is used for convenience in
making cross comparisons; thus Fluoroshield Sample No.
1 corresponds to Sample No. 11 and Fluoroshield Sample
No. 2 corresponds to Sample No. 12, etc.
The preparation of Sample No. 11 through 14 was
in accordance with. the procedures described above in
accordance with th.e present invention. In all cases,
the primer coat resin used was DuPont TEFLON-P 532-
5012 PFA rein and. the top coat resin mixture was 98
wt. % of DuPont TEFLON-P 532-5010 PFA resin mixed with
2 wt. °s of 5~, sized silicon carbide powder. In all
cases the barrier coat overall thickness of Sample
Nos. 11 through 14 exceeded .040", but did not exceed
1341016
54
.060". The details of the preparation of Sample Nos.
11 through 1~= are as follows in Table 6.
TABLE 6
Sample No. Primer Coat Top Coat
11 Heated Heated
Dry (Electro- Dry (Electro-
static) static)
12 Heated Heated
Dry (Electro- Dry (Electro-
static) static)
13 Heated Heated
Dry (Electro- Dry (Electro-
static) static)
14 Heated Heated
Dry (Electro- Dry (Electro-
static) static)
To furtzer explain the nomenclature used in Table
6, above, th~~ term "Heated" in regard to the "Primer
Coat" indicates that the procedure used is that
described ab~we, from the point just following step
No. 5 throug:~ Step No. 11 are followed. The term "Dry
(Electrostatic)" in regard to the "Primer Coat"
indicates that the procedure used is that described
above, from 'the point just following Step No. 11
through Step No. 14. The term "Heated" in regard to
"Top Co<it" indicates that the procedure used is that
described ab~we, in Step No. 14. The term "Dry
(Electrostatic)" in regard to the "Top Coat" indicates
that the procedure used is that described above, in
Step No. 15 through Step No. 17
1341016
Table 7, following, shows the Atlas cell test
conditions which Sample No. 11-14 were subjected to,
each corresponding to the individual test program to
which Fluoro=_:hield Sample Nos. 1-4 were subjected,
5 respectively.
TABLE 7
Corrosive
10 Sample No. Material Temperature Time
11 70 wt. 0 252F 600 Hours
Nitric Acid
15 1?. 20 wt. 0 220F 600 Hours
Hydrochloric
Acid
20 13 70 wt. 0 252F 1000 Hours
Nitric Acid
14 20 wt. % 220F 1000 Hours
25 Hydrochloric
Acid
30 The subjective observations in regard to the evalua-
tion of Samp=Le Nos. 11-14 are as follows in Table 8.
TABLE 8
35 Sample h'o. -_ Evaluation
11. Very slight blanching detected. No
blistering detected.
40 12. No blanching detected. No blistering
detected.
13. Slight blanching detected. No blister-
45 ing detected.
13410'
56
14. No blanching detected. No blistering
detected.
In comparing the evaluations of Table 5 and Table
8, it is clear that all of the Fluoroshield samples
tested were significantly deteriorated and degraded by
the Atlas cell test while none of the samples in
accord with the present invention suffered any
significant deterioration or degradation. Sample Nos.
13 and 14 were further tested, under the same cor-
responding conditions stipulated above in Table 8 for
an additional 300 hours each. In all cases, the
evaluation of the=.e samples, after the additional 300
hour exposures, remained unchanged.
BOND STRENGTH DETERMINATIONS FOR
SiC-CONTAINING E-C'TFE COATINGS
The next group of examples illustrates the effect
on bond strength between the coating and the underly-
ing metal substrate by increasing the concentration of
SiC in the primer coat layer..
Several additional sample plates (S-X) were
fabricated wherein composites were formed on underly-
ing mild steel substrates in accordance with the
procedures described above with the following depar-
tures from the enumerated protocol. The workpieces to
be coated were cleaned by a single grit blasting with
80 grit aluminum oxide (Step 6) and "baked" in an oven
at 600°F (St.ep 11;. A primer coat was applied to
workpieces brought to a temperature of about 500°F
(Step 12). Workpieces were returned to the oven and
brought back up to 500°F before application of each
succeeding coat (Step 16). Coatings were applied
before the workpiece cooled below 465°F (Step 15).
1341016
57
All bond ~~trengt~h determinations were made
according to A~~TM D3167-76 (Reapproved 1981), entitled
"Standard Test Method for Floating Roller Peel
Resistance of Adhesives", with the exception that an
equivalent to t:he fi:rcture for supporting the test
specimen, described :in paragraph 4.2 of ASTM D3167-76,
was used to the same end result.
In sampler S-W, Ausimont's HALARR 6014 ethylene-
chlorotrifluoroethylene copolymer (100 E-CTFE) was
used as a primer coat resin with Norton Company's 39
Crystolon gree:z silicon carbide flour 4647 (1000 grit)
in admixture t:zerewith in the amounts indicated in
Table 9 below. The primer coat layer applied to each
plate was fol7_owed by five ten-mil thick coats of
1~> neat, 1000 E-CTFE, giving a total coating thickness of
53 to 55 mils.
In sample plate X, Ausimont's HALARR 6614 E-CTFE
primer system, believed to contain, as its major
ingredient, E-CTFE, with chromium oxide as a minor
additive, was first applied followed by five applica-
tions of AUSIN:ONT'S HALARR 6014 neat E-CTFE top coat
resin.
Sample plates were formed using 1/4" x 8" x 8"
size mild steal planes. The results of the bond
strength testing are as follows in Table 9.
1349016
58
Table 9
E-CT:E'E COATING BOND STRENGTH TESTING
Sample Plate S T U V W X
Primer Coat
E-CTFE' wt. % 100 95 90 85 75 N. D.3
SiC2 wt. % 0 5 10 15 25 0
Particle Sia - 5~C 5~, 5u 5~,
1.5 Cr203 wt. % 0 0 0 0 0 N.D.'
Thickness approx. 3-5 mil (all)
Top Coat
~:0 E-CTFE wt. % 100% (all)
Thickness 50 mil (all)
No. of layers 5 (all)
a~5 Peel Strength 60 >180° >134' >1104 >1044 75
(pli)
'Ausimont; HALAI~'z 6014 E-CTFE resin.
30 ZNorton Company 39 Crystolon green silicon
carbide flour 4647, Worcester, Mass.
3Ausimont: HALA:2'~ 6614 E-CTFE primer system,
believed to consist. of a major volume of E-CTFE and a
3 5 minor amount o f Cr_ X03 .
' Value :>hown is actually a measurement of the
cohesive strength of the coating. Adhesive (bond)
strength of the coating to the substrate believed to
~i0 be considerably higher than cohesive strength value
shown.
In sample plate S, coated with neat (0 wt. ~ SiC)
.~5 E-CTFE in both the "primer" layer and top coat layers,
bond strength between the "primer" coat layer and
metal substrate wa=. relatively low, around 60
1341016
59
pounds/linear inch (pl_i). Sample plate X, prepared
with the man'ufacturer's recommended primer system,
(HALARR 6614 E-CTF~~~ primer) believed to contain a minor
amount of chromium oxide, fared slightly better than
sample plate S, giving a bond strength between the
primer coat layer and metal substrate of 75 pli, an
improvement of about 25%.
In sample plate T, the addition of only 5 wt. %
SiC to the primer coat resin resulted in a bond
strength value in excess of 180 pli, better than a
300% improvement in bond strength over plate S; 240%
over plate X coated with the manufacturer's recom-
mended primer system. The actual bond strength of the
sample plate T coating to the substrate could not be
precisely determined because the strips of coating
being peeled away from the substrate during testing
ripped apart shortly after peeling was initiated.
Thus, the value shown is actually a measure of the
cohesive strength of the coating being peeled away
from the substrata during the test.
In sample plates U, V, and W, having 10, 15 and
wt. % SiC, respectively, in admixture with the E-
CTFE resin as the primer coat, the bond strength
between the coating and metal substrate was, like
25 sample plate T, so great that it exceeded the cohesive
strength of the strip of coating being peeled away
from the substrate. Thus the peel strength values
expressed in Table 9 for plates T, U, V, and W reflect
the cohesive strength of the resin coating itself; the
adhesive strength of the coating to the substrate is
believed to be substantially in excess of the cohesive
strength.
The apparent decrease in bond strength in
coatings having increased amounts of SiC, actually a
1341p16
decrease in 'the coating cohesive strength, not
adhesive strength, is believed to be due to the
increased brittleness of the primer coat layer brought
about by the elevated SiC content. Thus, as the strip
5 was peeled away f-.rom the substrate during the test,
the more brittle, higher SiC-containing primer coat
layer caused the strip of coating being peeled to tear
more easily, the tear being initiated by a fracture in
the brittle :primer coat.
10 In an effort to determine an actual bond strength
value for Si.C-containing E-CTFE coatings, a sample
plate having a coating corresponding to that applied
to sample plate U was prepared, this time with a piece
of metal screen embedded in the top coat layers. The
15 screen, 6" x. 9" ir,, dimension, was positioned over the
10 wt. o SiC-containing primer coat layer and the top
coat layers applied over the screen in the same manner
as described for the preparation of plate U. The
screen was intended to provide a substantial reinfor-
20 cement of the coating as it was being peeled away from
the substrate. Despite the addition of the reinforc-
ing screen in the strip of coating being peeled away
from the substrate, the strip fractured, stretching
the screen embedded in it, just after the point at
25 which a value of 1.50 pli had been measured.
CORROSION-RESISTArfCE DETERriIINATION
FOR SiC-CONTAININCT E-CTFE COATINGS
In the examples which follow, the corrosion
30 resistance of SiC--containing E-CTFE coatings is
illustrated. Sample plate Y, corresponding to sample
plate U described above, was prepared by applying a 3-
5 mil thick primer coat layer of E-CTFE (HALARR 6014)
having 10 wt. % SiC admixed therewith, followed by
35 five successive.coats of E-CTFE (HALARR 6014) having
1341016
61
2.5 wt. % SiC admixed therewith. A 10 wt. % SiC-
containing E-CTFE primer coat layer was selected on
the basis of the bond strength test results reported
in Table 9 above and was deemed to represent a primer
coat having a preferred bond strength. A 2.5 wt. %
SiC-containing E-fTFE top coat was selected on the
basis of the superior corrosion test results observed
for 2 wt. % SiC-containing PFA top coat resins.
For purposes of r_omparison, sample plate Z was
prepared in the same manner as sample plate X in Table
9 above, (one 3-.5 mil primer layer coat of Ausimont
HALARR 6614 :E-CTFE, believed to contain chromium oxide,
followed by five ~>uccessive 10 mil coats of neat E-
CTFE (Ausimont HAhARR 6014)).
Both sample plates Y and Z were subjected to
Atlas Cell testing as described above (20% HC1 C
220°F) and observations of the respective coatings
made at 300, 600 and 1000 hours. Prior to the Atlas
cell testing, each plate was spark tested for pinholes
in the coating by testing with a Wegener WEG 20 High
Frequency Spark Tester set to 20 KV (AC). The power
level at whv.ch spark testing was conducted was
considerably more demanding of the coatings being
tested than is recommended by the Society of Plastics
Industry (SPI) Test Method for Detecting Faults in
Corrosion RE~sistant Fluoropolymer Coating Systems, No.
FD-128. SPI: recommended test voltages do not exceed
6,000 volts (DC). No pinholes were detected in either
plate spark tested. The results appear in Table 10
below.
1341~1~
62
Table 10
E-CTFE CORROSION RESISTANCE TESTING
Sample Plate
Hours Y Z
300 no change no change
600 :mall blister no change
(3mm) forming
1000 :mall blister in- no change
creasing in size
(l4mm) and beginning
to crack; second
small blister (3mm)
forming
As Table :LO shows, SiC-containing E-CTFE barrier
coatings were unaffected by exposure to hot (220°F) 20
HC1 even after exposure for 1000 hours.
In comparison, the AUSIMONT barrier coat system
began to blister after 600 hours of exposure under the
2~~ same acid conditions. Blistering suggests two modes
of failure for the system: (1) permeation of the top
coat layer_ by 13C1 and (2) insufficient bonding of the
primer coat la:~er to the metal substrate (75 pli),
permitting a direct chemical attack on the underlying
metal substrat~n, lifting the coating, exacerbating
coating failure. No failure was evident after 1000
hours of hot acid exposure in the SiC-containing E-
CTFE system.
BOND STRENGTH DETERMINATION FOR
ZrC-CONTAINING E-CTFE COATING
In this example, the bond strength of a 10 wt. ~
zirconium carbide-containing E-CTFE coating was deter-
mined. The zirconium carbide (Z-1034, a product of
Zerac/Pure, Milwaukee, Wisconsin 53233) was less than
44 a in particle size. Coating application and bond
63
strength testing were carried out as described for
SiC-containing E-CTFE. Bond strength was determined
to be in excess of 190 pli, the value measured just
prior to the cohesive failure of the strip of coating
being peeled fro:~i the substrate.
BOND STRENGTH DETERMINATIONS
FOR SiC-CONTAINING E-TI?E COATINGS
The object of developing these samples was to
determine the effect on bond strength between an
ethylene-tetrafluoroetylene copolymer (E-TFE) coating
and underlying metal substrate of adding SiC in
increasing concentrations to the primer coat layer.
Sample plates were prepared wherein E-TFE composite
barrier coatings were :formed on underlying mild steel
substrates i.n acc:ordance with the procedures described
immediately above for :E-CTFE with the exception that
after grit blast~.ng, workpieces were "baked" at 530°F
(Step 11) and the primer coat and all subsequent top
coats applied with the workpiece at a temperature of
525°F (Step 12).
In samples ~~A-EE, Dupont's TEFZELR ethylene-tetra-
fluoroethylene copolymer (E-TFE) was used as a primer
coat layer with 2Jorton Company's 39 Crystolon green
silicon carbide :Flour 4647 (1000 grit) in admixture
therewith in the amounts indicated in Table 11 below.
Table 11
E-TFE COATING BOND STRENGTH TESTING
Sample Plate AA BB CC DD EE
Primer Coat
E-TFE1 wt. % 100 95 90 85 75
SiC2 wt. 0 0 5 10 15 25
1341016
64
Particle Size - 5~, 5~ 5~C 5~t
Thickness 3-5 mil (all)
Top Coat
E-TFE wt. % 100%
Thickness 27 N.D. 30 30 35
(mils)
No. of layers 10 N.D. 10 10 10
Peel Strenctth 29 N.D. 28 31 37.5
(pli)
1TEFZELR 532-6000 ethylene-tetrafluoroethylene
copolymer sold by Duport.
239 CRYSTOLON green silicon carbide flour 4647 of
Norton Co.
From Table 11, bond strength between E-TFE and
the metal substrate i~> observed to improve measurably
with increased amount~> of SiC. A 23% improvement over
neat E-TFE is o~~taine~i by the addition of 25 wt. %
SiC.
CORROSION-RESISTANCE DETERMINATION
FOR SiC-CONTAINi:NG E-TFE COATINGS
In the exa~~ples which follow, the corrosion
resistance of Si.C-ceni~aining E-TFE coatings is
illustrated. Sample plate GG was prepared
corresponding to samp:Le plate EE above, that is,
having a "primer" coals layer of 75 wt. % E-TFE/25 wt.
SiC and "top" coat :Layers of 95 wt. % E-TFE/5 wt. %
SiC. For comparative purposes, sample plate FF,
corresponding to samp:Le plate AA above, was prepared
having a neat E--TFE coating applied thereto.
E-TFE coatings applied to sample plates FF and GG
and corrosion test re:~ults are summarized in Table 12
below.
1341016
Table 12
E-TFE CORROSION-RESISTANCE TESTING
5 FF GG
Primer Coat
E-TFE wt. % 100 75
10 SiC wt. 0 0 25
Thickness (mil.s) 3-5 3-5
Top Coat
15 E-TFE wt. s 100 95
SiC wt. 0 0 5
Thickness (mil_s) 37 40
20 No. of Coats 10 6
Corrosion Test
300 hours no change no change
25 600 hours no change no change
1000 hours test area has a single pin-
developed hun- hole has devel-
d:reds of pin oped in the
30 holes through- coating
oat due to the
e:Ktensive cracking
o:f the exposed
coating surface
35 After 1000 hours of hot acid (20% HC1 at 220°F)
exposure, the neat E-'rFE coating of sample plate FF
was literally r~_ddled with pinholes (2-3 pinholes/cmz)
with extensive ~iiscre~te micro-cracks throughout, each
crack about 3mm or so in length. In comparison, the
40 SiC-containing ~;-TFE coating of sample plate GG
developed a single pinhole and no evidence of any
cracking after x_000 hours of hot acid exposure.
1341016
66
Pinholes were detected using a WEG 20 Wegener High
Frequency Spark Tester set to 20 KV (AC).
In addition to the superior corrosion-resistance
of the SiC-containing E-TFE coating, the admixture of
SiC was seen to reduce shrinkage of the coating
applied. On a scale of 1 to 5 (1 being no visible
coating shrinkage and. the coating is observed to flow
smoothly around the sample plate edge without
thinning; 5 being severe coating shrinkage and the
coating is observed t:o form shrinkage ridges pulled in
over 1/4" from the edge of the sample plate), neat E-
TFE coatings experienced severe shrinkage, for a
rating of 5, while shrinkage of 95 wt. % E-TFE/5 wt.
SiC coatings (a.pplied over a 3-5 mil primer coat of 75
wt. % E-TFE/25 wt. % SiC) was very low, for a rating
of 2 (slight shrinkage - just starting to thin or pull
in at the corners).
Further, t:he addition of SiC was seen to improve
surface uniformity o:E the applied coating. Neat E-TFE
coatings tended to be uneven with large bumps and
waves unevenly distributed across the coating surface,
giving a mottled effect, whereas the addition of only
5 wt. % SiC rendered E-TFE coat smooth and uniform.
High surface g:'_oss of E-TFE coatings remained
unaffected by i:he addition of SiC.
Still_ furt=her, the addition of SiC to E-TFE was
seen to improve the "buildability" of the
electrostatica:Lly applied coating. Ten coats of neat
E-TFE were required to achieve a 37 mil thick top coat
whereas only 6 coats of 5 wt. % SiC containing E-TFE
were required ~~o achieve a 40 mil thick top coat, an
improvement over nearly 180% (6.6 mils/coat vs. 3.7
mils/coat). The phenomenon observed appears to reside
in the amount ~~f dry E-TFE powder that will adhere to
134101 fi '
67
the sample plate during electrostatic deposition.
SiC-containing E-TFE dry powder was observed to build
to a greater depth than neat E-TFE dry powders. A
possible explanation for the observed phenomenon may
be that the negatively charged resin powder insulates
the relatively positively (i.e. grounded) charged
piece being electrostatically coated. Once insulated,
the charged workpiece cannot attract additional powder
and, in fact, additional powder sprayed on the piece
is repelled or simply falls away. The addition of SiC
to the resin powder may i.mpr_ove the powder conduc-
tivity, thereby permitting a thicker layer of dry
resin powder to be attracted to the substrate before
reaching a thickness great enough to insulate the
underlying substrate.
Overall, the addition of SiC to E-TFE coatings
was seen to improve coating bond strength,
dramatically improve corrosion resistance, markedly
reduce coating shrinkage, significantly improve
surface uniformity, and provide nearly a two-fold
improvement in "buildability".
CORROSION-RESISTANCE DETERMINATION
FOR SiC-CONTAINING PVDF COATINGS
In the following examples, the corrosion
resistance of SiC-containing poly(vinylidene fluoride)
(PVDF) top coats applied over a Cr203-containing PVDF
"primer" coat is illustrated. Sample plates HH and II
were coated accorc.ing to the procedures already
described for E-CTFE with the exception that after
grit blasting, the plates were "baked" at 550°F (Step
12). Between sub~;equent: coating applications, the
plates were returned to 500°F. Top coatings were
applied before thE~ plates cooled below 350°F (Step
15).
?341p16
68
A 3-5 mil thick primer coat of KF Polymer poly-
(vinylidene fluoride) (PVDF), a Kreha Corporation of
America produci:, admixed with 5 wt. % chromium oxide
(Cerac, Inc., Milwaukee, Wisonsin), was applied to
both sample plates tested. To sample plates HH and
II, were applied top coats of PVDF resin having in
admixture therewith 0 and 5 wt. ~ SiC respectively.
Plates HH and II were concurrently subjected to
Atlas cell- tesi~ing (20% HCl at 220°F) and observations
of the coating:> made at 300, 600 and 1000 hour
intervals. Each plate was spark tested at 20 KV (AC)
and found free of pinholes. Results of Atlas cell
tests are summ<~rized in Table 13 below.
1F~ Table 13
CORROSI02J-RESISTANCE DETERMINATION FOR PVDF
Sample Plate HH II
Primer Coat
PVDF1 wt. % 95 95
Cr203 wt . 0 5 5
Particle Siz~a <_10u 510
Thickness 3-5 mils 3-5 mils
Top Coat
PVDF wt. % 100 95
SiC 0 5
Particle S i z a -- 5~C
No. of layers 3 4
Thickness (mil) 45 50
1341416
69
Atlas Cell Test
300 large (llmm) very small
b:Lister (4mm) blister
formed beginning to
form
600 b:Lister enlarged blister
to 15 mm causing enlarged
d:isbonding in to 6mm
area
1000 blister continues pinhole
to enlarge developed in
(23mm); pinhole 6mm blister
in blister
formed.
'PVDF KF Polymer poly(vinylidine)fluoride, Kreha
Corporation of ~~meri.ca.
ZCr203, Cerac, Inc., P.O. Box 1178, Milwaukee,
Wisconsin 53201
As Table 1:3 shows, sample plate II, having a 5
wt. % SiC/95 wt. % PVDF top coat, was much less
susceptible to :glistering than sample plate HH
having a neat P'JDF top coat. The addition of SiC
to the PVDF top coat significantly reduced
permeation by hst HC1 (20% HC1 C 220°F) as
evidenced by the greatly reduced extent of blister-
ing, notwithstanding the use of a Cr203-containing
"primer" coat layer of PVDF.
BOND STRENGTH DETERMINATION FOR SiC AND/OR HIGH
PERFORMANCE THERMOPLASTIC-CONTAINING PFA COATING
In the examples which follow, sample plates
MM, NN, 00, PP, QQ, FR and SS were prepared in
order to compare bonf, strengths for PFA resin
coatings having 0, 10 and 20 wt. % SiC; 20 wt.
polyphenylene sulfide (PPS); 20 and 15 wt.
polyetheretherketone (PEEK); and 10 wt. % SiC in
admixture with 20 wt. ~ PEEK. All samples were
t34t~t6
prepared in the same manner described earlier for
PFA coatings. r,ond strength data for each are
presented in Table 14 below.
5 Table 14
r' PFA COATING BOND STRENGTH TESTING
10 Primer Coat NCI NN 00 PP QQ RR' SS
PFA' 1000 90 80 80 80 85 70
SiC2 0 10 20 0 0 0 10
15 PPS3 0 0 0 20 0 0
PEEK4 0 0 0 0 20 15 20
Thickness 4/19 4/21 4/19 4/33 4/31 3/23 4/26
20 (~ coats/mil)
Bond
Strength <5 10-15 <5 <5 >405 N. D.6 N. D.6
25 1PFA (perfluoroal_koxy resin) , NEOFLON AC-5500 PFA
resin, Daikin Industries, Osaka, Japan.
ZSiC (green silicon carbide flour) 39 CRYSTOLON
4647 (1000 grit; Norton Company, Worcester, Mass.
30 'PPS (polyphenylene sulfide resin) Ryton type V-1;
Philips Chemica_L Co., Bartlesville,
Oklahoma.
4PEEK (polyetheretherketone) VictrexR 150 PF,
Batch No. SP69-:L91P, ICI Americas,
Inc., Wilmington,
35 Delaware 19897.
Sexceeded c:ohesi~re strength of coating.
6bond strength too great to initiate peeling.
40 'Plate RR was At7_ac cell tested
for corrosion
resistance (70 pat. . nitric acid at 225F) and, after
300 hours of te:~ting, shows no
evidence of pinholes or
blistering.
1341 p' 6
71
In comparing bond strength of PFA with the
various additives above indicated, it is clear that 20
wt. % PEEK provides air least an 800 % increase in bond
strength over noat coatings, 20 wt. % PPS and 20 wt.
SiC-containing fFA "primer" coatings, and a 3 to 4
fold increase over a :LO wt. % SiC-containing PFA
"primer" coating.
In the above described examples, microscopic
examination of ~'PS and PEEK-containing PFA coatings
shows that PPS remaina, for the most part, in its
, particulate stage as discrete spheres in the PFA
resin. Where P3?S particles are in direct contact with
the metal substrate, there is some evidence of PPS
flow at the point of contact.
PEEK particles, on the other hand, in the
involved examples appear to flow to a greater extent
at their point of contact with the metal substrate,
appearing as rounded mounds rising from the substrate.
PEEK particles not in contact with the substrate form
interconnected :strings, anchored to the mounds, and
form a matrix through which the PFA resin flows.
Coating composites of about 10 to about 40 wt. %
PEEK/about 90 to about 60 wt. % PFA "primer" coat and
about 2 wt. % SiC/98 wt. % PFA top coat provide a
superior barrier resistant coating resulting from the
vastly superior bond strength of the PEEK-containing
PFA "primer" system and the corrosion resistance pre-
viously demonstrated for a 2 wt. %-containing PFA "top
coat".
In Table 14A below, bond strengths and Atlas Cell
test results ar~~ presented for such composite
coatings. The sample plates TT, UU, W, WW, XX, and
YY were prepare~3 by applying 5 coats of a "primer"
coating composition of PFA resin and 5, 10, 20 and 40
1341 t11 6
72
wt% PEEK or 8 wt:% PPS, and 5 coats of a "top coat"
composition inc7_uding PFA resin and 2 wt% SiC. All
samples were prepared in the same manner described
earlier for PFA coatings.
Table 14A
COMPOSITE COATINGS
BOND STRENGTH AND ATLAS CELL TEST RESULTS
_ Sample
uT U1:J W WW XX YY
Primer Coat
PFA1 95 90 80 80 60 0
PFA2 0 0 0 0 0 92
PEEK' 0 ~0 20 0 40 0
PEEK 5 10 0 2 0 0 0
PPSS 0 0 0 0 0 8
ThlCkneSS 10/36 10/x.210/41 10/3910/36 10/43
COatS~ml_1 )
Bond
Strength 24 38 >40 >40 >40 336
Corrosion
Resistance?
(500 hrs) 0K$ OK OK OK OK OK
(1000 hrs) 0K OK OK OK NO9 NO
1PFA (perfluoroa7_koxy NEOFLON AC-5500 PFA
resin),
resin, Daikin Industries,Osaka, apan.
J
2PFA (perfluoroaT_kox y resin), NEOFLON AC-5600 PFA
resin, Daikin Industries,Osaka, apan.
J
'PEEK (polyetherethe rketone)
VictrexR
150 PF,
Batch
No. SP69-191P, :LCI Americas, Wilmington, Delaware
Inc.,
19897.
~ 34~ o t ~
- 73 -
4PEEK (polyethE:retherketone) VictrexR 450 PF, ICI
Americas, Inc., Wilmington, Delaware 19897.
SPPS (polyphenylene sulfide resin) Ryton type V-1;
Philips Chemical Co., Bartlesville, Oklahoma.
6Lower and higher concentrations of PPS result in
lower bond strength.
7Atlas cell te~;ting conditions for corrosion
resistance (70 wt. % nitric acid at 252°F).
BIndicates slight blanching but no blistering was
detected.
9Indicates that slight blanching and either
blistering or disbonding was detected.
PREPARATION OF SiC-CONTAINING PFA SHEETS
The object of the example which follows was to
demonstrate the preparation of a sheet of SiC-containing PFA
resin.
The sheet was formed on a mild steel plate which had
been cleaned according to the procedures described above,
including grit blasting with 80 grit aluminum oxide grit,
vacuum cleaning and baking in an oven for eight hours to drive
contaminants from t:he steel. The surface of the cleaned plate
was first sprayed with a heat stable release agent (Frekote*
33, a fluoropolymer product manufactured by Frekote, Inc. of
170 W. Spanish River Blvd., Boca Raton, Florida 33431) to
permit a subsequently applied resin coating to be stripped
cleanly away from the underlying substrate, then heated in an
oven to between 680-700°F. The heated, release agent-treated
plate was then sprayed with six 8 to 10 mil thick coats of a 2
wt.
0
*Trade-mark
x
- 73a -
SiC/98 wt. % PFA dry powder mixture to give a total coating
thickness of 60 mils. The plate was repeated to 680-700°F
between each coat. After the final coat was applied, the
coated
x
1341016
74
metal plate was allowed to cool to ambient temperature
and the coating stripped cleanly away from the
substrate in thc~ form of a sheet.
Continuous sheet production is contemplated
through the use of endless thin steel belts, treated
with a suitable relea:~e agent, heated between resin
coating applications by passing the belts through a
series of ovens heated, for example, by both convection
and infrared radiation, the resin compositions being
applied through spray nozzles spaced between ovens.
Stainless steel belts, 18 to 24 gauge thick,
manufactured by Sandv.ik Co. of 1702 Nevins Road, Fair
Lawn, New Jersey 07410 would be suitable for this
purpose.
PREPARATION OF ;>HAPED ARTICLES
ARTICLES OF SiC--CONTAINING PFA
The object of the example which follows was to
demonstrate the preparation of a shaped article
comprised of a :resin/ceramic powder mixture of the
present invention.
In this example, a 3" pipe elbow having a 50 mil
wall thickness ;end comprised of 98 wt. % PFA/2 wt.
SiC was prepared from a two-part mold assembled from a
3" carbon steel pipe elbow, cut in half along its
length. The in;~ide surface of each pipe half was grit
blasted with 80 grit aluminum oxide, the residue
removed and the pipe halves baked in an oven for eight
hours at 7G0°F to remove contaminants. After cooling
the pipe halves to room temperature, they were re-as-
sembled and the interior thereof coated with a release
agent (Frekote 33) and the liquid excess removed.
The release agent-treated mold was returned to the
oven and brought to a temperature of between 680 and
700°F after which the mold was removed from the oven
1341016
and the interior thereof electrostatically sprayed with
the 98 wt. o PF1~/2 wt. % SiC dry powder to form a
coating thereon 8 to 10 mils thick. The thus-coated
mold was return<~d to 'the oven and again heated to 680-
5 700°F until the resin,/SiC powder had fused to a smooth
and glossy film. The coating process was repeated
until the coating was 50 mils thick. After cooling,
the molded piece was tested for pinholes using a WEG 20
spark tester se-t to 50 KV (AC). No pinholes were
10 detected. The mold was disassembled and the PFA/SiC 3"
elbow removed from the mold.
It is cont~Amplated that the fluorocarbon
polymer/additive coatings can be used in a variety of
applications including those for which wear and load
15 resistance, corrosion resistance, and/or release
characteristics is desired.
For example, composite coatings of the present
invention may be applied to the chemical seal and drive
portion of agitators commonly employed in chemical
20 vessels for mixing corrosive chemicals. Particularly
useful in this regard are polyether/fluorocarbon
polymer composites described above. Such composite
coatings may also be applied to the tips of blades of
such agitators which are subject to high abrasion and
25 wear.
Still further, composites of the present invention
may be applied as coatings to metal roll surfaces of
the type found on rol:Lers used in paper making,
calendaring, and extrusion lamination, which rollers
30 are usually subjected to abrasion, wear, and high load.
Many of the conventional primer systems used with
the application of fluorinated polymer coatings to
metal substrates, include chemicals, such as for
example chromic: oxide, which are considered detrimental
13410-16
76
to the environment and definitely are not approved for
use with food stuffs :for human consumption. On the
other hand PFA, PPS, and PVDF, PES and silicon carbide
have been approved by the U.S. Food and Drug
Administration as coating materials which can be used
in the processing of food stuffs for human consumption.
Such approval h<~s likewise been given to many of the
species of crysi:alline ceramics. Thus, the use of the
barrier coating system of the present invention and the
composites formed therewith exhibit an additional
advantage wherein applied to process equipment used in
the preparation of such food stuffs.
