CA1071369A - Polytetrafluoroethylene molding powder - Google Patents

Abstract

POLYTETRAFLUORETHYLENE MOLDING POWER
Abstract of the Disclosure Finely divided polytetrafluoroethylene molding powder is disclosed having a remarkable comination of high apparent density and high moldability, the latter being re-flected as a small difference inspecific gravity (.DELTA.SG) of standard moldings molded at high and low preform pressures.
For example, moldings powders having an average particle diam-eter(d50) of no greater than 60, a .DELTA.SG5-1 of no greater than 30 and apparent density of at least 600 g/l are disclosed.
These molding powders are obtained from unique process steps involving either the fine powder type of polytetrafluoro-ethylene of the granular type of polytetrafluoroethylene as the starting material. The fine powder type is compacted to a high degree, followed by partially decompacting the resul-tant compacted fine powder to the molding powder particle size desired. When the granular type is the starting material, this is comminuted to the molding powder particle size desired, followed by seperating the molding powdrer into hard and soft material fractions, followed by re-grinding teh relatively coarse, hard fraction by itself to an average particle size of at least as samall as the original combined fractions.
The resuslant molding powder is useful in the manner of previous polytetrafluoethylene molding powders with the advantage of requiring reduced mold height andmolding pres-sure toacheive equivalent quality moldings.

Description

~ his lnvention relate~ to poly~iet;ra~luoroeth~rlene ..
molding powder.
PRIOR ART

'" ' - .
.
`' ' " - - ' . .
' ' ' ' ' .
' : ` ' '' .

.

.
.
. ...
.. , ., .. , .. . ~.. .. . :

' . ~ . .. .

. ~ . .

. . .. ~. . ...... . ~ .. .. ~
, , , , . ~ , ~ . .. . . i ... . .

: ~ ; . ~ ~ ~

. .-, -:- - ~ . . - . ,, , , -: -Two general types o~ polgt~tra~luoroethylene (PTFE) are available, the granular type9 usually called molding -powder, and the fine powder ~ype which ls obtained ~rom aqueous dispersion polymerization. ~oth ~ypes are made by pressuring tetra~luoroeth~lene into a stirred aqueous solution OI Iree radical polymeriæation initiator, under polymerizing conditions. In aqueous dispersion polymerization, the stir-ring is ~low enough and sufficient dispersing a~ent is present t.hat the PTFE is ob~ained ~n the f`o~n of colloidal~ize pa~t-icles less than one micron in diame~er, which remain dispersed in -~he aquecus medium when polymeriæation is stopped. ~on ~
. coagula~ion o~ the particles and drying, the ~ine powder type ..
: o~ PT~E ls obtained.
In granular polymerization, the stirring ~or agita- :
tion) is su~icien~ly rapid to cause coagulation o~ the pol~-mer particles during the polymerization. Dispersing agent is generally not presen~ except perh~ps in smaller than dispersion-stabilizing amounts ~or the d~eren~ purpose disclosed in U.S. Patent No. 3,245,972 to Anderson et al~ : -20 When polymeriz~ion is stc~ppedl the re~ul~ant granular poly-mer i8 in the fo~n of relatively coarse particles, sorne 1000 microns in diameter and larger. No~lly, this polymer is sub~ected to coarse or ~ine grlnding ko provide the molding powders commercially a~ailable.
: These two di~eren~ types o~ PTFE have quite di~-~erent and mutuall~r exclusive moldin~ characteristics. PTFE
fine powder is ~abricated by blending ~ith an oil lubrication aid in about an 80:20 parts by weight proportion and the resultant pasty mass is extruded, generally at room temper-30 2ture~ this process being called paste extrusion. PT~E

~, .
~ ~ .
. ~, .~. , , molding powder ls ~abricated by (a) pressing in a moldg ~ollowed by sintering of the resultant preform without ap-plication of pressure, or (b) ram ext~sion, which in~olves ramming the powder through a heated orlfice~ which si~ters the powder under pressure. PTFE ~ine powder is not fabricab1e except as small moldings ~gener~lly less than 30 grams) by the pre~orm/Pree sinter techniqueg or by r.am ex~ru~ion~ and conversely, PTFE molding powder is not paste extrudable.
Because o~ the dif~erent polymeFlzation techniques involved in rnaking the two types o~ PTF~ and their differen~
methods o* ~br~cation leading generally to the application o~ these types ~n d~ferent ~ields, the tech~ology of these two ~ype~ has been separate and independent ~rom one another.
An exception to this is U.S. Patent No. 3,087,921 to ~athews and Rober~s, which d~sclos~s the making o~ PTFE molding powder having good hand~ing characteristics and high appare~t density by subjecting either a previousl~ available PTFE mold-ing powder or PT~E ~ine powder to the steps o~ (a) compacting the PTFE to a den~t~ o~ at least 2~l5 g/cc at pres~ing ..
conditions of 50 to 300C. and pressures o~ 1000 psi ~70 kg/
cm2) to 3000 p~i (211 kg/cmZ), (b) cooling the compacted polymer, and (c) comm~nuti~g the cooled, compacted polymer to particles having an average par~lcle diameter o~ smaller than ~ .
1000 microns and pre~er~b~y ~rom 200 to 500 mlcrons ~the wet sie~2 d5~ particle size o~ ~50 to 650 microns ~or uncompacted PTFE ~ine powder is the pa~ticle size of the loose a~glomerates fo~med by coagulating the colloidal size po~yme~ particles)O
While this treatmen~ increa~es the powder flow for the PTFE
~ine powder ~rom ~n essen~ially no-~low condition to 17 g/sec ~....

.

~ ~ 3 - . . .
.
, ' ~

., - ~. - ~ : . .- , 3~
and increases apparent density ~rom 400 to 600 g/l ~o 800 to 1000 g/l, the tensile strength of the PTFE molding powder made ~rom the PTFE fine powder is only 1650 psi (116 kg~cm2~, which is poor at best, especially as compared to the minimum tensile strength o~ 4000 psi (280 kg/cm2) Por ASTM Type IV~
which is the highest quality PTFE molding powder.
THE PRESENT INVENTION
The present inve~tion provides a high qualit~ poly tetra~luoroeth~lene molding powder which can be obtalned ~om PTFE ~ine powder or .~rom special techniques appli~d to existing PTFE molding powder~ Speci~ically, the polytetra-~luoroethylene ~olding powder of the present invention can be characterized as ha~ing a specific ~ur~ace area o~ at least 1.5m2/gJ as being ~inely divided as evidenced by an average particle diameter of less than 100 microns, and having the combination o~ high moldability and high apparent density, as evidenced by an apparent density o~ at least 5O0 g/l and ; rela~ed to moldability by the follo~ng equation:
; (13 Apparent denslty ~ 5OO ~ 3.00 ( ~ SG5 1) wherein ~SG5 1 is 1000 tlmes the di~Pe~ence in Sp2C~.EiC
gravities o~ the sintered molding m~de at pre~orqn pressures o~ lO00 psi (70 kg/cm2) and 5000 psi (352 kg/cm2) (the proce-dure ~or determining ~SG5 1 is described further hereinaEter) wherein ~SG5 1 is no greater than 75.
The la,rger the difference be~ween t~e specific gravity values at 70 kg/cm2 and 352 kg/cm2, or in o~her words the higher tl~e ~,S~5 1 value, the more voidy would be the sin~ered a~icle ~de from the low pre~sure preform. T~s voidiness would resu~ in reduced tensile and dielectric 30 stren~th and thus poor qualit~r o~ the sintered a~ticle. In co~ercial practice, voiày sintered article can o~ten be preven~ed by the use o~ high pre~o~n pressures bu~ this re~u~res more ma~siv~, a~d thus more expen~ive~ pre~orming equipment.
Thus, the lo~er the ~ SG~ he lower is the voidiness and the better is the quality o~ the si~tered article~ Low ~ S~5 1 values are thus indicative o~ high quality moldings made ~rom the molding powder, or in other w~rds~ high moldabllit~, The ~ SG5 ~ value is also re~erred to hPrein as moldability index ~at 5-1 unless otherwlse indi.-ca~ed~ ~ pre~er~ed moldability index ~or molding powders Or ~his in~ention ls no greater than 60~
The low moldability indexeæ exhibited b~ the molding powders o~ the present in~ention in combina~ion with their 8mall particle si~e correspond to high tensile strengths for ob~ectæ made there~rom, whlch have a tensile strength o~ `~
at least 3500 psi (2~5 kg/cm2) and pre~erably at least 4000 .; psi (280 kg/cm2). The s~andard tensile strength test is ; done on sintered objec~ m~lded at 50Q0 psi (3~2 kg/cm~) .' pref~rm pressure. The low mo.~dability indexe~ ~or molding ~:~
powders of the p~2sent l~e~tion e~ableæ these ten~ile s~rengths to be achieved at only 1000 psl (70 kgJcm2) preform pressure~.
By way o~ comparison, the molda~ility ~ndex o~ the Ma~hewæ and Rober~s molding pow~er which is made from densi- ~ -~ied granular pol~mer and which h~ a hlgher tensile ~trength ~:
than the ~nolding po~der made ~rom ~ensified dispersion poly-m~r (2150 psi v. 1650 psl) is 89 as rep~:~ed in comparative ....
Ex~rnple 8 o~ Roberts and Ander~on, U..S. Pa~ent No. 3,7663133.
This hi~h mold~bili~y index i5 ob~ined from ~h~ ference between ~peci~ic grari~ies at pre~o:~m pre~sures which are clo~er to~ether, vi~. 2000 psi and 5000 psl, than the test used in the present in~en~i~n. At the pr~form pre~ure~ o~
1000 psi and 5000 psi used in the prese~ vention, the .

moldability ~ ndex o~ the M~thews and Robe~s product would be much higher than 89 . To illustrate, the lowest mold-abillty index value reported ~rom a ~gh apparent denslty molding powder (565 g/l) in U.S. Patent No. 3,245,972 to Anderson, Edens, and Larson i~ 15, based on speci~ic ~rav-ities taken on samples made at 2000 and 5000 psi prefo~n pressure ( ~SG5_2). At pre~ox~n pressures o~ lO00 anà 5000 psig this moldabil~ty ind~x (o~ 15) increa~es to about 75.
A molding po~Yder o~ the presen~ invention that has a mold~
10 abili~y index ~5-l) of 75 has an extremeïy high and desir-; able apparent densit~ o~ at least 700 g/l.
Figure 1 o~ the drawings acco~panying this inven-tion s~o~Ys a graph o~ apparen~ d~nsil;y versu~ ~SG5 1 (calcula~ed ~rom pre~orms pressed ~t 1000 psi (70 kg/cm2) and 5000 psi (352 kg/cm2)). The n~nbers plotted in the graph correspond to the molding pow~ers O:e the Exa~ples disclosed later herein.
~ he letters shown on the graph are located at points where other high per~oxmance ~ine ground PTFE molding powders) prior ~o the presen~ rer~tion, :Eall in t~ms of their~appare~t densities and ~ SG5 l~8~ The molding powders are as follows:

A. ~Algo~lon'l F-2 (trade mark of Mo~teca~inl Edison) B. ~'Hostaflon~ TF-17 (tr~de mark of Farbwerke Hoechst) C~ ~Pol~lon~l M12 (trade mark o~ Daikin Kogyo) D. ~'Fluon~ G }63 (trade mark of I~CoI~) E. "Halon~ G l 80 (trad0n~.ark of Allied Che~icals~, UOS. Pa~ent Nv~ 356L~,984) F. ~e~lon~ 7A ~rade mark o~ Du Po~t) G. "Te~lon~ trade ~ark o:E Du Po}~tj ~I. Exa~le l, U,, S. P~te~t 3, 690y 569 I. S~n~ple g of Example 4 of U.S~ Pat. 3,690,569 The average particle d~ameter o~ all these moldlng powders ~alls ~ithin the range 10 to 100 microns. The AndersonJ ~ :
Bdens and Larson molding powder is not included i.n the graph because it is not finely ground; the co~rse grinding prac~iced therein gives an average particle diameter in the range o~
about 400 to 500 microns. Finely ground molding powders have the advantage over coarsel~ ground resins of being pre~orm-able and free sinterable to moldings Q~ improved mechanical and electrical propert,ies. In a~dition, the finer ground ~olding powders are more suitable ~or blending with p~rti-culate flllers ~Q give ~illed moldlng powder~ which are wlde~y used ~or their property advantages, eæpecially wear ~.
.. .
resistance.
e da~a points sho~ by numbers and letters in th~ graph are about at midpoint o~ their reæpecti~e numbers and letters.
Curve 1 in Fig~ 1 is the line represented b~ equa- . .
tion (1). Curve 2 in Fi~. 1 is the line and lower ~oundary o~ the pre~erred AD vs. ~ SG5_1 relation~hip represented -1 20 by the followln~; e~uation:
~ 2) AD - 600 ~ 3.00( ~S~
Cu~ve 3 in Fig. 1 is the line represented by the following equation:
(3) AD 400 -~ 3~00( ~SG
From Fig. 1 it is seen th~t except ~or molding powder I whlch is considered und~slr&ble in U~S. Pat.
3,690,~69, all the lettered molding powders fall below cur~e 3, well removed ~rom ~he molding powders o~ the present invent:ion ~Jhich ~all on ~r above curve 1. The slope o~ cu~ve 30 3 approximately corr2sponds to the e~:~ect of fi~er grinding to in~prove moldability (low ~,SG5 1 value) which resul~s ln 7 :

:, decreasing apparent density~ Th~ ~m~ller the average particle diameter for these moldin~ powdersJ the lower is their ap-parent density. ~his is the ef~ect of the very fine grinding disclosed in Kometani et al. U.S. Patent No. 3~7261483r Figures 2 and 3 are photomicro~raphs at a magnifi-cation lOOX of crossections of skived tapes skived ~rom billet~ of sintered molding powder. In Figure 2~ the mold- ;
ing powder making up the skived tape is "Halon" G-30. The ligh~ colored spot~ in these Figures are voids in the tape, which render them unsuitable ~or some applications, such as insulation of electric~l wire and cable.
In Figure 3, the molding powder is representative o~ the molding powder o~ Example 25 before agitation with water, and the benef'icial e-~fect of its high mo-dabllit~
(low ~SG5 1) is shown by the absence o~ voids in the t~pe.
~ igure 4 is a graph showing the variation o~
SG5_1 ~ith average particle diameter o~ molding powder ;~
derived ~rom PTFE ~ine powderO
The molding powders o~ the present invention are obtainable ~rom diverse sources, from the aqueous dispersion or fine powder type of polytetrafluoroeth~lene and ~rom the ~ranular or molding powder type o~ polytetra~luoroethylene.
MOLDING POWDER OF THE PR~SENT INVEMTION
; _ MADE FROM PTFE FINE POWDF.R
With respect to aqueous dispersion PTFE ~s the source~ this starting mater~al is known in the ar~ disclosed ~or example in the a~oremen~ioned Mathews and Roberts patent and in greater detail in U~S. Patent No~ 2,5599752 to B~rry.
~'his t~pe o~ PTFE is u~ed in the coagulated fOrm9 o~ten called "~ine powder."
,. . .

~ -8 ~
~ ' .

7~
The firs-t step in the conversion o~ this coagulated aqueous dispersion PT~ to molding powder is to compact this P~E under high pressure at room temperature~ i.e. 20 ~o 30C., in a pressure device~ such as a molding press or compacting rolls. Pressures ranging ~rom 8000 psi (5~2 kg/cm2) to 50,000 psi (3513 kg/cm2) can be used. The coagu-lated aqueous dispersion PTF~ can be wet at the time of compaction, i.e. still contain some o~ the aqueous polymer-ization medium therein, or can be dry.
The second step is to break up, i.e. partially decompact, the compact into particles having an average diameter less than 100 microns. Although this decompacting is not considered grinding because the pa~icles making up the compact are already much smaller than the particles resultin~ from decompacting~ conventional grinding equipment can be used to do the ~ecompactingO Such equipment includes high speed cutter mills run in water, such as the "Taylor Stiles Glant Mill" ~Taylor Stiles Co.), which decompact in water; and ~luid energy mills, such as a "Micronizer"
~Sturtevant M111 Co.), which de~ompact ~he compact in the dry state. Prior to ~eeding the compact to such mills, it may have to be broken into relativel~ coarse chunks o~ a slze which can be fed to the mill. If wet milling is done~
this is followed by drying. The resultant partially decom-pacted material is the molding powder o~ the present invention.
The molding powder of the present inven~ion pre-pared by thls route o~ compaction and decompaction o~ the ~ine powder type of PTFE is disti.n~uished ~rom the starting : :
; ~ine p~wder in several ways. First~ the produc~ of the inven~io~ ha~ a pr form porosity at a pre~orm pressure of 1000 psi (70 kg/cm ) of no greater than 0.20 a~d pre~erably '', '.
: ~ _g_ :
. ., :

.
.. . . . . - :: , les~ than 0.17. In ~act, many molding powders obtained from this route have a porosity belo~ 0~15, showing excellent low pressure preformabil~ty. In con~rast, the fine powder start-ing materials ha~e a porosity well above 0.20~ indicating poor low pressure prefoxmability. The signi~icance of this di~ference in porosity values is that while the ~ine powder is not sinterable to massive a~ticles ~ithout crac~ing~ the molding powder obtained therefrom b~ the presen~ invention is sin~erable without cracking. Second~ mold shrinkage at 1000 psi preform pressure (~S(1000)) of the fine powder start-ing material is greater than 8.o~ ~hile %S(1000) of the product of the process is less than 8.o and preferably no greater than 7.2. This lower shrinkage is an advantage in that it is less di-~icult to design a mold to make a product of a given size. Third~ mold shrinkage at 5000 psi pre~orm pressure (~ S (5000)) of fine powder will be substantially reduced in the moldin~ powder made therefrom~ Generally, the % S ~5000) o~ fine powder is greater than 3.7, and tne molding powder of the present invention obtained therefrom has a ~ S (5000) o~ no grea~er than 3.7~ ~inally, the product of the lnvention is moldable by molding po~der techniques, while ~ine powder tends to stLck to the mold and to crack on pre~orming or sintering.
Typical molding powder o~ the in~ention derived ; ~rom fine powder as described hereinbefore has been molded ex~ensively, the results indicating that typical molding pow-der is con~arable to a high quality ~inely ground granular PTFE and in some ~espects superior. Commercial fine po~der PTFE is not sultable ~or molding by pre~orming and sintering because it sticks to the mold and cracks badly. ~undreds of -10 _ . ~ ;

5.72 cm, 7.62 cm, and 10.15 cm diame~er cylinders have been made from the mold~ng powder of the p~esent in~entio~
derived ~rom fine powderg and no mold ~ticking has been observed. The resul~ant sintered cylinders7 including the 10.15 cm cylinders (weighing o.9Q8 k~), have been just as crack-~ree a~ controls made ~rom high qu~lity ~inely ground PTFE molding powder (made from moldlng powder, l.e,, granular r~in) available hereto~ore. No c~ac~s were ~ound in 5.72 cm or 7062 cm c~linders.
~ Moldings made from the molding powder derlved ~rom Pine powder ha~e a very smooth surface, and ~ape slsived ~rom these moldings ha~ a uni~orm void-~ree ~ppearance as shown in Fig. 3. This molding powder sinters to a clear~ transparent, sel~-supporting melt, whereas flnely-ground granul~r resin available here~ofore gives a cloudy melt. A clear melt i8 advantageous, because the user can look in the o~ren ~nd see whether sintering is complete3 iOe., the melt is clear, a~er which the coollng cycle can be startedO The high ~uali~y o~
the skived t~pes is illus~ra~ed b~ the~ havlng a dielectric strength in exces~ o~ 1800 vol~s/mil (700 kv/cm) on 5 mil (127 micron) thick sam~les prepared as desc~.bed below~ :.
To $11u~trate the good mechanical and electrical properties of a molding powder of the present invention derived from ~ine powder~ the molding powder ha~ a tensile ` .
strength of 4550 psi (320 kg/cm23, an elong~tion o~ 320~o and a dielectric strength of 1880 volts/mil (7~0 kv/cm), measured on 5 mil (127 m~cron) tape skived ~rom 5.72 cm diameter solid -`
billets sintered ~or 5 hours at 380~ ænd cooled ~t 2-3C/ . `
mi~ute. Elongation ænd dielectric ætrength cor~are with ~ 3 side~by-æid~ con~rols o~ other representative PTFE molding - po~er~ as ~ollows:

.

3~
Dielectric Strength Resin Elon~,ation ~ kv~.
Typical molding pow~er of this inven~ion derived from fine powder 320 740 Represen~ative commercial ~inely ground molding pswder A klolding Powder E
(~lg- 1) 285 729 Mold~ng Powder F
(~ig- l) 280 828 Preforming of the billet was done at 176 kg/cm2. When the preform pressure was only 70 kg/cm2, the moldlng powder still ga~e a dielectric strength su~erior to that of commercially available ~inely ground molding powder, owing to the improved physical t~ni~ormity of the tape as is observable, for example~
by comparing the tape of Fig. 3 ~ith the tape o~ Fig. 2.
Specifically, at 70 kg/cm2 pre~o~n pressure, 5-mil (127 micron) thick tape ~kived ~rom the sintered pre~orm o~
molding powder o~ the present invention exhibited a dielec tric strength o~ 768 k vvl~s/~m, as compared to only 433 ~ :
kv/cm for molding powder B and 295 kv/cm ~or molding powder above.
MOLDING POWDER OF THE PRESENT INVENI'ION
MADE FRO~ PTFE MOLDING POWDER
~G~ANULAR RF.S ~
__ _ W~th respect to the granular type PTFE as the sta~ting material for moldin~ p~wder o~ the present in~en- :
tion~ it has ~een discovered that t.he granular ~ype of PT~E consists o~ t~o fractions~ one of ~Jhich is sof~ and the other o~ which is har~ intimately associated with one another, The so~t fraction is re~erred to herein as alpha ', -. ' . .

resin and the hard ~raction as beta resin. As part o~
the presen-t invention~ it has been found that raw ("as polymerized") granular resins contaln fractions o~ alpha and beta resin in proportions depending on polymerlzation condltions. For example~ by increasing the percent solids to which the polymerization is conducted, the proportion of beta fraction is increased.
In the commercial fine grind~ng of granular PTFE, it ha~ also been ~ound that the two fractions tend to become dissocia~ed from one another as separate particles. Specifi-cally, the alpha resin grinds more rapidly~ i.e. about lOX
as ~ast, than the beta resin so that when the grinding mill is shut down a~ter lengthy running~ the resin that is in the recycle line because it is oversized is primarily beta resin.
This resin in the recycle line is called the mill residue when the mill is shut down and representq a ~e~J small pro-portion of the total ~eed to the mill, depending on the running time for the mill. Because this mill residue was stlll coarse or large in particle ~ize relative to the aYerage diameter present in ~he desired product of the mill~
the mill res~due was heretofore discarded. This residue ~rom the making o~ the major ~inelg ground molding powders -~
hereto~ore available had the follo~ing characteristics:

, '~`
` ' .

~7~3~

~:;
rl rl r-l =
r-l 0 5-~ ~
r-l ~ O O
~rl N ~ ,~, r l 1 ~ CU
O ~ CO r~ L) r l S:~ ~ Q ~r1 r~ ~ O,~ ,~:
,~ I ~D r-l C)~ C~l ~ U~ ~
~3 0 ~CU C) O
O r-l =
rl = rl ~3 ~rl 1 ~ r-l 1) W r~ rl h rl rl ~ ~ ~ C; ~
~; ¢ $ Q) a~ ~ o r~l ~rl O
~ ~ O
~~rl !~ ~ S:~ ~ .
U~V~ O ~ rl ~ ~ O~ r_l , 1~1 O I--1 0 ~H ~3 CU CO ~ a) :1 P~h r l ~1 ~ ~ ,r~,~ cq O ~ ~ ~0 OLS~ ~ O ' ~_ ~ ~ ~r ~ U~ ~ N 1 r~ .
~ ~ ~ ~ ¢ ~ (~
O f~ l~t~D O
V ~ eH e,~ ~ ' ' ~ ~0 ~ r~ rl ~ ~' ~
1~ r I ,~ ,~

~ ~
rl ~ ~ ~H 0~ :
~1 ~ h n~ o U~ S
r CO O
~H ~ ~ tn ~1 ~P 0~ 0 Il) 031~:1 tlL ~~J Lf~ rl ~ ~ 4r1 a~ /11 bO ~1 ~r-l ~S) O O Q~
rl od rd c~ r~ In N ~ Q~ C~ 5 V:l rl ~0 ~ ~rl ~d a) s ~ O ~ i r~; 0~2~ r~

. P:; rl ~ ~q ~
io - ~o ~o ~ ~d ~W o ;~
~ rl ~~ rQ
O ~ 10 0 ~ ~ N J ~ S~ eh 5 ~ rl r rc ~
S-- b~ r~ 1 0 1--I
03 C~ ~r~

rl r~ rl 03 r1 ~ r~l N~) i~ ~d V~ ¢
,~

:

The molding powder o~ the present invention a~ broadly defined differs from these mill resldues by having a higher speci~ic surface area and higher tensile strength. 1~ le the molding powders made in the grinding operation that produced these mill residues had relativel~ high speci~ic sur~ace areas, the mill residues had very low specific sur~ace areas. rl'he increase in the maximum specific sur~ace area of the mill residue from 1024 to the minimum o~ 1.5 m2/g for the molding pow~er of the present in~en~ion is an increase of a~ l~ast 25~. This corresponds to a minlmum decrease in par~icle size of at least that proportion. The desired finely ground molding powders obtained ~rom the mills were a mixture o~
predominately beta resin containing some alpha resin.
To make the molding powder of the present in~ention, (a~ the beta resin is first separated from a mixture o-~ beta resin and alpha resin and (b) then this separated beta resin is conver~ed by ~urther milling to the improved molding powder having the desirable combination of properties stated hereinbefore. ~he bettsr the degree o~ separation in step 20 (&), the higher will the apparent densit~ be for a given degree o~ milling in step (b).
While mill residue repre~ent~ a ~eparation o-f sorts in which onl~ an in~initesimal proportion of isolated beta resin is obta~ned in a commercial milling operation, the separation i~ more economically done by classi~ing finely ground molding powder into approxlmate alpha and beta ~ractions, ~rhich ~s possible because the harder-to-grind beta fraction has a greater average particle diameter than the alpha ~raction after a limite~ grinding time. A~ter the alpha ~raction is suf~iciently grouna to lea~e the mill, o~ly the beta ~raction ' ~ :

' :.

3~

remains, thereby accompli~hing the ~eparation o~ the alpha ~raction from the beta ~raction.
The re~ultant beta fraction obtained b~ this classi-ficatlon is too coarse, as in the case o~ mill residue, to have good moldability propert~es by itself. mis separated beta ~raction or m~ll residue of beta resin ls then sub~ected to ~urther fine grinding to increase lts spec~fic ~ur~ace area an~ ten~ile ~trength. It then become~ the product o~
the inventlon. Thi~ ~ine ground beta resin molding powder, 10 contrary to usual experience, ha~ a hlgh appar~nt den~ity as stated in the descriptlon o~ molding powder o* the present invention hereinbefore. Thi~ i s made po~31ble by th~ low alpha re~in content or absence of the alpha resin ln the fine ground beta re~in.
Examples o~ grinding mill~ for this fine grlndlng that can be u~ed are grinding mills obtainable und~r the trade marks "Micronizer", "Hurrican~ Mill" (Mlcrocyclomat Co.) (U.S. Patent NOr 2,936,301~, and 'IJet-O-Mlzer" (U.S.
Patent NoO 3,640 J 984)o me ~inely ground molding powder and the mill residue o~ta~ned therefrom having a ma~or proportion of both beta resin and alpha resin cannot be ~inely ground lnto the apparent density/moldabil$ty index region o~ the pre~ent : lnvention because the apparent denslty of the alpha ~raction i~ relatlvely low to begin with~ and the f~ner grind~ng would amount to overgrinding o~ any al~ha res~.n pr~sentl causing a ~urther lowering of the ~pparent density o~ the overall molding powder. The ef~ect o~ fine grinding in produclng low apparent den~ity 1~ seen for mold$ng powders A-G in Figure 13 the low apparen~ densit~es ~temming ~rom the alpha resin ~ontent o~ the e moldlng powder~.
'~ ,';
,:
~ -16- ~

7~

In addition to relative hardness that distlnguishes particles o~ beta reæin ~rom those o~ alpha resin, the beta resin particles are also distinguishable by sh~pe, i.e~, the ; finely ground beta resin particles are smooth sur~aced and have the general shape o~ ~latt~ned ~pheres wh~reas the finely ground alpha resin particles are flu*~y in appearance. In addition~ indi~idual p~r~icles of ~he molding powder o~ the present i~en~lon which consîsts of fine ground beta resin are characteristically birefrigent showing the mal~ese cross typlcal ~ single crystals under a microscope illumina~ed wi~h polarized light~ whereas alpha particles do not show~ -this characteristic appearance. The particles o~ ~he moldin~
powder derlved from fine powder according to this in~ention also exh~bit this birefringence.
In addition ~o high AD and low ~ SG5 l' molding ; powder o~ ~inely ground beta resin is notable ~or its lowi~
anisotropic behavior during moldlng. This means thRt a preform of the molding powder shrinks relatively uni~mly i~ all directionæ when sintered. This makes it easier for the user to design and use a mold to obtain a sintered part of the deslred dimension~. Anisotropic behavlor is measured as (S). Desirable low anisotropic behavior i~ indicated by an ~S) absolu~e value o~ less than o.8 (i.e., between -o.8 and t oo8) and preferably less than 0.5 (i.e~, between -0.5 and +0.5) for the molding powders of the prese~t inve~tion.
The closer the (S) value is to æero9 the better i5 the molding powder in this respect. Other finely ground g~an-ular reslns have greater (S) ~alues~ e.g., -1.23 and -l~O
for molding pow~ers F and E o-~ Figure 1. ~ -The molding powders of the p~esent in~en~ion, ~hether derived ~rom fine po~rder or by finer grinding OI coarse beta : 17- -: ' ~
`

resin particles, have somewhat better flowability ~powder ~low) than the ~ine ground resin of U.S~ P~tent No. 2~936,301 to Thomas and T~allace. Powder ~low can be ~urther improved to form a free-~lowing molding pow~er by forming agglomerates of the molding powder, e.g., average agg~omerate diameter o~ 300 to 1000 mlcrons, using water, an organic wetting liquid or an immiscible mixture thereo~ as agglomeration medla.
The molding po~rs o~ the p~esent ~nvention in the preferred sense have ~ moldability index ( ~ SG5 1) of no greater than 30 and an average part~cle diameter o~ no greater than 60 microns. Molding powder o~ ~inely ground beta resin preferably has a ~ SG5 1 of less than 1~ and a~erage par~lcle diameter o~ le~s than 20 microns. Some o~ these molding powders exhibit ~ SG~ } values o~ zero~ ~hich meanæ that the molding powder has extremely good si~erabili-ty at 1000 psi (70 kg/cm2)~ Such molding powders of the present invention can be pre~ormed at pressures as low as 500 psi (35 kg/cm2) or even 200 psi (14 kg/cm ) and still be sintered ~o dense, strong moldings. Pre~erabl~, the molding powder of the ~0 prese~ inve~tion has an aver~ge particle diameter o~ at ~: least 10 microns.
The molding powders o~ the preæent inven~ion : cons~st o~ high molecular weight PTFE~ which means that these molding powders are fabricable by the n~n-melt ~abrication processes use~ on the con~entional PTFE molding powders.
One measure o~ high molecular weight is that the molding powder o~ the present invent-ton has an apparent melt viscosity ~:
of at least 1 x 109 poises at 380~C.
The PTFE ~r~m which the molding powder of the ; 30 prese~ invention is made can also contain a sm~ll proportion o~ ~luori~ated terminally unsaturated comonomer content, e~g.
, . .

' :; ,., .3~i~

O.Ol to 0.5~ based on the total weight of the copol~ner.
Pre~erred comonomers are the per~luoroalkenes and perfluoro-(alkyl vinyl ethers), each containing 3 to 8 carbon ~tom~.
The comonomer contributes high tou~hness and high flex li~e ~o the molding powder.
TEST PROC~DU~S
Tesk results reported herein are dete~mined accord-ing to the following test procedures:
Apparen~ ~ensit~ (AD) is the uncompressed apparen~
density o~ the powder and ls m~asured by ASTM D-l457-6g~
withou~ separ&ting and reconstituting the sam~le. Theor0ti-cally the max~mum apparent density of the molding powd~r that could be obtained i~ the particles thereof were spheres o~ di~ferent slzes to giv0 ~ maximum packing density, would be 1942 g/l.
Calc. AD - Calculated apparent density is deter-mined by measuring the volume occupied by the pow~er co~pris-ing 1 cm3 o~ pore-~ree polymer con~ained in the specimen tube used for the subsieve siæe test. This vol~me iS te~med 20 ~he bulk factor~, Calc. AD ~ 2285,Jbulk factor. The units o~
apparent density are grams/liter~ m e value obtained in thls measurement is always close to but not necessarily identical with the value ~or apparent density as measured by the proce-dur~ of AS~M lL~57. The reason *or use o~ calc~ ~D rather than the ASTM procedure ~s the requirement for 2.285 grams of sam~le rather than a quan~ity that ma~ be as large as 200 grams.
~pecific Sur~ac~ Area (SSA) - The speci~;c ~ur~ace , area of a Bample ~ powder is the number o~ square meters of sur~ace per g~m ~ polymer as measured by a nl~rogen adsorp-tion ~echnique. The measurements of ~his paramet~r were all ` made using a modif~e~ g~s chromatographic technique wlth the : lnstrumen~ be~ng calibrated ~or each ~un wlth a sample o~ ~
-19- ; `
'' ~''.

polytetrafluoroethylene on which the sur~ace area ~s measured using the ~tandard BET procedure des¢ribed by ~arr and Anhor~
in Chapter XII o~ "Scienti~ic and ~ndus~rial Glassblowing and Laboratory Techniques" published in 1949 by Ins~rument Publish-ing Company. All of the molding powders of' the presen~ in~en-- tion prepared on the Examples herein have an SSA o~ at least 1.5 m2/g~ PTFE fine powder as a s~arting material has an S~A
o~ at least 9 m2/g, and af~er the co~paction and decom~action æ~eps, the SSA is s~ill we1.l above 1~5 m /g, Subsieve siz0 (SSS~ - This is a number expressed in microns determined on an apparatus made by the Fisher Scien-tific Company and called a subsieve sizer. The procedure is essentially that described in ASTM Standard B-330-58T using a porosit~ value in the determination of 0~55 using a sample size for unf'illed resin of 2.28 grams. SSS is a measure o~
air pe~neability, whieh is a ~unction of particle ~ize and porosity. ~or series of' samples in which porosity does not vary, SSS is a measure of average particle size.
SSG - The s~andara speci~ic gravity is measured on samples prefo:Yqned at 5000 psi (or 3~2 kg/cm2).
Th~ test procedure is described in ASTM D 1~57-69, except that the pre~o~ming die used is 2.86 cm in diameter and a ch~rge o~ 12.Q g of pol~nner is used. The sintering cycle -:
includ~ a step o~ heating up the specimen from 300 to 380C.
~ a~ 2C/min~ A~r the speci:~ied 30 minu~es at 380C, the oven is cooled to 295C. at 1C. per minute and held at this temper~
.. .
ature ~or 25 minutes a~ter which the specimen is removea ~nd cooled to room tsmperature and the standard speci~ic gravity is : determined as ~pecified ~n D 1457-69. SG ~lOOO)means that the 30 SSG procedure w~s ~ollowed except that the speci~ic gravit~ is :
~ .:
:, ~

` ' ~
, . . . - ~ ~ . : .

~7~

determined on a sintered pre~'orm molded a~ lOOO psi instead o:~ 5000 psi. S~G increases wi-th the rate of crys~alliza~ion~
and (for homopolymers at least) rate o~ crystallization decreases with increas;ng molecular weight. m us SSG measure-ments before and a~ter a process give a me~sure of molecular wei~;h~ change due to that process.
~ SG5 1 - Delta speci~ic gravi~y (mold~bility index 5~1) in~ol~es determination o~ the specific gra~ity of a sintered specimen prepared as in the SSG procedure except that the preform pressure used is 1000 psi (70 kg/cm2).
SG5 1 ~ 1000 (SSG (5000 psi pre~o~m pressure) - SG (1000 psi preform pressure))O When the term moldabilit~ index is used for ~ SG5 2~ such moldabilit~ index is de~ined as 1000 x (difference between the speci~ic gravity of moldings made at 5000 and 2000 psi preform pressures).
~ S (5000) - The percent shrinkage is the percent decrease in diameter between the pre~orm and ~inal sintered piece of the test specimen used to determine SSG Mith the measurement bein~ carried out in the direction perpendicular to the direction of applied preform pressures ~late~al change).
A value obtained ~or % S varles appreciab~ with prefo~m pres-sure and even with details o~ application of the preform pressure. ~ S (1000) is the same with a preform pressure of ~000 psi (70 kg/cm2) instead o~ 5000 psi ~352 kg/cm2).
(S) is the symbol used for a constan~ in an equ~tion used to predict lateral and axial dimensional changes during sintering. (S) is a measure of the elastic memory or ~ibrous-ness o~ the polymer par~icles and has been shown e~perimen~ally to be essen~ially constant ~ith pre~o~m pressure rather than 30 ~arying widely as does percen~ shrinkage. ~f on~ knows the , .
~ -21-: .
. . .
:

void volume o~ a pre~orm it is possible to calculate dimen-sional changes in both the axial and lateral direction of a molding by usin~ (S) according to the e~uations 4a and 4b below. The value of (S) is determined ~rom %S as shown in equation 5 below~ usin~ a pre~orm pressure of 352 kg/cm2.
The closer ~S) is to zero, the more isotropic is the behav-ior of the sampleO

Eq. 4a - calculate later~l change % shrinkage = 100 ~ 1 - [ ~ ~pre~orm density si~~~ea~a~r~y) x [1 -~ 0.01 (S)] :K (o.g667 ~
0.1025 yrG - ~ o.o844 ~G2)~ ~

~ , where ~ G - Void fraction o~ the pre~orm.
To a good approximatlon ~G = 1 - (pre~orm density/2.285) Eq . 4b - calculat e axial rhange growth = 100 ~ preform densityjsintered den x ~ 1/(1 ~ 0.01 (S))2] x (1.06g - 0.224 o.l979 ~G )] - 1 }

Eq. 5 - calcul~te (S) - me~ure o~ elastic memory o~ particlers in mold (S) = lOO ~ OoOl (5~ S)~
s ered densi y) x (o.9667 ~ 0.1025 ~G ~ o.o844 ~G 2) ~ _ 1 }

Porosity - The porosity is the void ~raction in (cm3 voids)/(cm3 total ~olume) o:E the prefo~qn used to prepare the specimen ~or the SG ~1000) dete~nination as dePined aboveO
It is indic~ive of the prefo:~mability of a :resin.
Tensile ~trengkh - is the stress at r~pture in kg/cm2 of ox~nal cres~-séctlonal are~ o:f a tensile strength test ,'' ' ':
-22- ;

: :' , - : . . .......

L3~

specimen of the dimensions speci~ied in ASTM D-1457-69 pre-formed at 5000 psi (or 352 kg/cm ) and sintered according to the schedule described under SSG, unless otherwise specified.
AEF (anisotropic expansion factor) is a measure of the dimensional change obtained on sintering. The value ls obtained as follows: Twelve grams of powder is weighed into a 1-1/8" (2.86 cm) diameter mold and compressed to 352 kg/cm2 during 1 minute, held ~or 2 minutes, and then released. The diameter and height oP the pre~orm are measured and the pre-form is sintered by the same sintering c~cle as under SSG.
The sin~ered thickness and diameter are then obtained and anisotropic expansion factor is then the value o~ ~ -TS/rrp _ S/ p ~ ', when Ts and r~ are thickness of sintered resin and preform, respectively.
when Ds and Dp are diameter of sintered piece and prePorm, respectively.

~ E - is the percent elongat~on of the Tensile Strength (TS) test specimen at rupture.

Powder Flow - 'rhe polymer sample is u~ed to fill a vertical polytetra~luoroe~hylene pipe 2~.8 cm high and 5.08 cm in diameter and having a 6 mesh screen attached across the base of the pipe. 'rhe pipe is sub~ected to ~ibration having a frequency of 675 cycles/minute and an amplitude of 0.762 cm.
The amoun~ o~ powder ~lowlng through the screen is continu-ously weighed and recorded. From the resulting curve the powder flow is calculatea as grams/second.
Particle size disclosed herein unless otherwlse spec-ified is the weight average particle diameter (d50~ f the molding p~wder determined b~ the wet sieve procedure disclosed .. ~. '~' ' ' .

.. . ,, ,. , ~ . .

.~ !l A'~ I ~ .
~ L~ ~ ~ ~t ~

in U.S. Patent 2,936~301~ Standard sieves ~or wet sieve analysis are not readily available in ~izes smaller th~n 37 microns and the wet sieve method is not ~pplicable to very small particles~ The weight average particle diameter of particles smaller than 37 microns is determined by the "Micromerograph" method descrlbed in U.S. Patent 3~265J679~
unless otherwise indicated herein. Results ~rom "Micromero-graph" dete~inations are in units of d (microns) x ~here rho is the density o~ the particleO This density is 10 not known but i~ believed to vary wlth particle slze and type (alpha or beta resln). The density is expec~ed to vary ;~
~rom about o.8 to 2.28. The corresponding s~uare root values vary from about 0.9 to 1.5 and the actual average size in microns, therefore, is usually somewhat less than the value o~ d ~ o reported. In most instances, particle size values obtained by one o~ these tests were confirmed qualitatively by optical microæcopy.
The average particle diameter, d50, f agglomerated powders is dete~mined by the wet~sieving procedure of ASTM ~ .
20 D-1457-6~ but selecting a set o~ sieves in the square root o~ 2 series starting with 1000 microns recommended by the Int,ernational Standards Organizati.on. The particle size of the basic or primary par~icles of PTFE ~ine powder is deter-mined by observation through an elect~ron microscope.
App~rent melt viscos~ty is calculated by measuring the tensile creep of a sintered piece held a~ 380aC~ Specifi-cally, 12 g. o~ molding powder ls placed in a 7~6 cm diameter mold between 0 152 cm rubbe~ cauls and pap0r space~s. The mold is then heated at 100C~ ~or 1 hour~ Pressure is then ; 30 slowly applied on the mold until a value of 140.6 kg/cm2 is ~ obtained. This pressure is held for 5 minutes and then , ..
~ -24-.' ~ ~,, released slowly. A~ter the sample disc is removed from the mold and separ~ted from the cauls and paper spacers3 it is sintered at 380C for 30 minutes. The oven is then cooled to 290C. at a rate of abou~ 1C~ a minute and the sample is removed. A crack-free rectangular sliver with the ~ollowing dimensions is cut: 0.152 to 0~165 cm. wide~ 0.152 to 0.165 cm. thick~ and at least 6 cmr long. The dimension~ are measured accurately and the cross sectional area is calcu-lated. The sample sliver is attached at each end ~o quartz rods by wrapping with sil~er-coated copper w~re. The dis-tance between wrappings is 4.0 cm. This quartz rod-sample assembl~ is placed in a columnar oven where the 4 cm. test length is brought to a temperature of 380 ~ 2C. ~ weight is then attached to the bottom quartæ rod to give a total weight suspended from the sample sliver o~ about 4 g. The elongation measurements vs. time are obtained, and the best average value for the creep curve in the interval between 30 and 60 minutes ~s measured. me speci~ic melt viscosity~
which may be better called apparent mel~ viscosity, is then calculated from the relationship .1 WLtg :
app = ~ - _ 3(dLT/dt)AT
where ~ app - (apparent) melt viscosit~ in shear~ poises W = tensile load on sample~ g = length of sample (at 380C~) cms. (l~32 cm) g = gravitational constan~ 980 cm./sec.2 (d~ /dt~ = rate of elongation o~ sample under load = slope o~ elongation ~s. time plot, cm.~sec.
AT = cross-sectional area of sample (at 380C.)~
cm2 - area increases 37~ at 380C. over that at room tempera~ure ,~

~.

~ .

.. ..

~7~

Examples o~ molding powders of the presen-t i~ve-n-tion are as ~ollows (parts and percents are by weight unless otherwise indicated):
Exanrple 1 In this experiment, a 20.3 cm diameter stainless steel Micronizer air mill was used. This is Model o8-5057~
manufactured by Jet-Pulverizer Company, Palmyra, New Jersey.
I~ is operated adiabaticall~ - i.e., without addlng or remov-ing heat - using 2830 l/min ~iltered compressed air in~roduced at 25C~ and 6.67 kg/cm2.
The ~eed polymer was TEFLON~ 7A ~luorocarbon resin 11967). Durlng a four-mlnute period, 200 g o~ the -feed resin was gradually introduced into the mll~ at a uniform ~eed rate of 50 g/min. This feed ra~e w~s experimentally determined as providing the maximum separation o~ beta resin from a~pha resin. When introduction o~ the polymer ~eed was complete~ ~ `
the mill ~as operated for one minute with no polymer ~eed to remo~e most of the remaining alpha resirl as e~fluent. The tota~ e~luen~ and the residue remo~ed ~rom the micronizer chamber a~ter shutdown were ~pproximatel~ equal ~n weight.
This experimen~ was repea~ed several times and the residues (coarse beta) were combined and ~ed into another run in the same e~uipment. This time the polyme~ feed rate was 30 g/min. The resultant e~luent (rnolding powder o~ ~ne ground beta resin of this invention) was abou~ 70~ o~ the ~eed : - and weighèd 354 g and is the molding powder o:E this invention.
The characterization o~ the products is in Table I.

, ~ ""
.:.. . . . ...... .. .... . , - . .,., .. . -.. : - ...... . . . ..

3~

~
~
~ a) ~3 q~ o~
O-rl ~
5~ ~ ~ CO
L~
P; ~ o ,~ ~ ~o C~l o ,, . . . . CU
~:; tB O P~ rl ~ O N C~ ~ N rl 1~ I C ~ ~h LS~
~m ~Do s~
rl ::~
ID rl ~ bO
::

h ~ H ~rl . . .

tl3 ~ 0 Ll~
~ O r~ D N O
m ,~ O c~ N .C~\ ~n ~ ~q ~ ~ ~

~ ~ O
~1 CO ~ o CO CJ o æ :
.~ .
.
~ t, ~ ~rl ' b o C~ V
X ~t ~Q~rl rl ~t ~ rl g g ~ ~
C) ~c~ h ,r.

,~U:2 ~ ^ ~ '~ O ~ N
t3 ~ ~ ~ ~ Ul lS~
~' ~ ~ ~ q ~ O ~.q ~' 1 ' :` 27-'':.. ~' ~ 7~36;9 The molding powder of the presen~ invention made in this Exa~ple still has h~gh apparent densi~y and has a much better moldability as indicated by a ~ SG5 l ~ zero~
which is indicative of a tensile strength in exces~ of 4000 psi (280 kg/cm2). In actual fact~ the speci~ic gravity o~
the sintered molding made by pre~orming at lO00 psi (70 kg/cm2) was 0.0007 g/cc higher than that made at 5000 psi (352 kg/cm2) pre~o~m pressure, which probably r~presents the degree of e~perim0ntal accuracy o~ the test method. The significance o~ zero ~SG5 1 is that this molding pow~er can be pre~ormed at very low pressures to give prefo~ms which will sinter to high quali~y moldings. To illustrate~ to obtain a positive value o~ ~ SGg the low pressure preform pressure would have to be lowered~ e.g., to 700 psi or 500 psi (49 kg/cm2) or (35 kg/cm2). Preformability at this low pressure is unlaue in the molding powder art. The high degree of compactne~s o~ the prefo~s is indlcated by the poro~ity at 70 kg/cm2 being 0.15. This porosity value is much less than ~or the fine powder type of PT~E and is similar to poro~ity values o~ other finely ~round PT~E mold-ing powders, but the ~interabil-lt~ as indicated by SG5 1 of zero is much better than other ~inely ground molding powders o~ similarly high apparent densities. Shrinl~age values, S(1000) oP 6.32~, S(5000) of 2.9 ~ and AEF of 10124 ~or the molding powder o~ this inven~ion all compare favor~
ably with commercial ~inely ground PTFE molding powders.
; The particles o~ molding powder made in this Example exhibited bire~ringence when viewed through a polarizing mi croscope ~

.':
' ' "' Exam~?le 2 ;~
A. Examp].e l was repeated to obtain three pounds (1.46 kg) of molding powder of the present invention made of finely ground be~a resinJ wi~h the results shown in Ta~le II.

', .:, ~'.'' :~ .' ' . .

~:
a) ~1 m~
~R ~
S~ tQ
~ Q~
o s~
V
a) h bD ~ ~ O
O -rl O O C~ ri CU ~i 0 0 0 o (r~
O bO
~: ~a : .

~rl ~
~ ' ' ~~~
~1 0 .. .

q O
~ ~ :

H ~ ..
O 0~ C~l ~ ~' ~ ~ ~ r~ 0 r~ l N CU O ~ ~ -. E~ ~i O N (:~i j =t ~ O 1-1 1 0 0 0 ~ ~~ N
., q~

~ r~ ~ ~ C~ ~ ~
(~:; O e-- CO (!i C~ C~ t~ ~. ri ~ CU ~ ~N
~ U~ ~

.
a~ :

O
O ~ ~ N~ ~ O N~ N~ ~ ~

¢ ,~ ~ ~ O O C~J ''' ':
r~ G ~
..~
. .~ ,, : .

The e~luent of 70% alpha resin is the product obtained from the separation of alpha resin from coarse beta resin. This coarse beta resin was then ~inely ~round to get the final ef~luent which is the molding powder of finely ground beta resin of the present invention. The e~fluent of 70% alpha resin exhibits excellent moldabilit~ as charac-terized by the low ~ SG5 1 value o~ 2.7, but this improve-ment is obtained at the e~pense o~ decreased apparent density~
increased particle fibrousness (increased tS) value)~ and increased shrinkage. In contrast, the ~inal e~fluent ~rom regrinding the coarse beta resin has even better moldability than the starting TE~LON~ 7A~molding powder with an even higher apparent density, decreas~d particle fibrousness, and at no significant sacrifice in shrinkage characteristics.
~ The particles o~ the molding powder o~ this invention made ; in this Example 2A exhibited bi-refringence when viewed through a polarizing microscope.
B. This experiment was carried out using the fluid energy mill and polytetrafluoroethylene resin feed described in Example 1. During a four-minute period~ 200 g o~ the feed resin was introduced into the mill at a uni~orm feed rate o~ 50 g/min. When the introduction of the polymer ~eed was complete the alr ~low ~o the mill was shut of~ and the produc-t recelver changed. The mill was started up and run for two minutes wi~h no feed~ and then shu-t o~. The product receiver was changed an~ the mill was run again for two minutes collecting 18.5 gr~ms of unifbrmly shaped pa~ticles having an average particle diameter o~ about 15 microns and exhibiting biref~ingence when examined with a polarizing microscope.

.
.
.. . . . , ,, .. ..... ... .. , , ..... . .. -~ . In this experiment, the micronizer was run as in Paragraph B. until the 200 g o~ ~eed pol~mer was intro-duced into the mill. The air ~low was then shu~ off and the product fraction was isolated. The mill was opened and :~
the residue in the mill was removed. mis operation was repeated ~hree times.
The three product ~rac~ion~ tot&lling 202 ~rams were combined and fed to the clean mill. This feed material is resin that had already gone through the mill one time and consisted o~ a mixture o~ 80 percent alpha and 20 percent beta resin. A~ter the once ground rnaterial had been in~ro-duced into the mill, the mill was shut do~, the product receiver changed and the mill run ~or ~our rni~utes~ On opening the mill, 13.2 grams o~ residue resin was recovered.
This materlal is the ground beta resin of the in~ention. It had a subsieve size of 6.o and a calculated apparent density of 671 g/l~ Microscopic examination wi~h polarized light showed it to consist o~ uni~orm small (about 10 micron in aYerage diameter) birefringent partlcles.
.:
E~ _s 3-14 These exa~ples ~how the making o~ molding powder of the present invention ~rom various PTFE ~ine powder starting materials (Fine Powder E is used in Examples 21 and 26).

.
.

:: .

3~

Fine Powder De~cription_ _ A PTFE homopolymer, SSG c a. 2 . 220 (U. S. Patent No. 1, 559 ,752) B PTF:ES modif ied by hexaf'luoro-propylene (U. S. Patent ~o.
3g 142~6653 C PTFE modi~ied by hexa~luoro-propylene (l~wer mol Wt.
than A, U0 S. Paterlt No.
3, 14:~, 665) D PTFE modifled by per~luoro-propyl vinyl ether (U. S. Patent No. 3~819,594) E PTFE homopolymer, S5G 2.167 All the3e starting materlals ~ere large agglomerate~ (d50 400-600 micron~) o~ tiny ba~ic particles (0.1 - 0.5 microns).
The starting ~ine powder (coag~lated and dried aqueou~ dispersion PTFE) was compacted in a laboratory press ~ at 25 C. and at various pre~sures~ me mold wa~ a cyllnder ; 20 5.72 cm in diameter and the fine powder charge wa~ 100 g.
The partial decompaction wa~ acco~plished in 3.785 l. Waring ~lendor (Model Mo. CB 5)9 a high-~peed blade type mixlng device, equipped wlth a 12.1 cm diameter blade, 6~35 mm wide and 3017 mm thick leading edge, unles~ otherw~se indicated. me bro~d races of the blade mo~e in ~ plané
perpendicular to the ~ertlcal ~h~fto me resin particles are 3truck by the 3~17 mm thick, blunt leading edges of the . :
blade a3 it rotat~s. Dur~ ng the finishing, temperature was mea~ured with a thermocouple in the ælurry and controlled by circulating ice water on hot water through the ~acket o~ the .~ blender. Thl~ apparatu~ wa~ u~ed at high speed for Examples 3-8, 12, 13. and l4, a~d their control experlment~
, '.,"- "
, ,'".
o33- .
..

~7~3G9 In some e}~pe:riments, a standard blade supplied t~th the Waring ~lendor (Examples 9, lO and ll and their control) was used in place of the flat bla~e. In othe-r experiments, a 0.947 1. ~aring ~lendor ~as used with its standard blade (Example 8 and its con~rol). The data tables show whether the high or low speed o~ the blender was used.
- It is estimated that with the 12,~ cm b~ade, the peripheral speed was 76.3 m/sec at high speed and ~5.8 m/sec at low speed.
In each partial d ~ mpac~ion step, water at 30 C
and the compacted ~ine powder cylinder were charged to the blend~r, enough water being used to ~i~e a~ou~ 10 - 20~
solids in the blender. The time of partial decompacti.on is shown for each experiment. The molding powder w~s separated and dried at 120 C. for 16 hours. Further details o~ these experiments and results are shown in Table III:

, ~ ' " '.

, .

: ~ ".`' : .' ' :
: -34~

:: ( . ; - .

~fll7~3~

~;
~ . .
o S I I I I I I I I i I I ~D ~ ~ ~ ~ I r-U~
o ~ o u~ ~1 3 1 I I . . I ~ co a~ ' ~ ~ 3 ~ ~ ~ ~ O OC~ 3 ._ O ~ ~ O O (~ ~ O ~ /~ ~ 3 _~-1 o I o I O I O ~J O I O O O C~ O O O
O I I I ~ I I I I I ,:
O
O~ o Ir~ o L~ O ~ J L~
r-l ~I r-l Q
O C~ J IS~ ~1 3 ~ CO ~ O ~1 ~ 3 L~
n ..
~J ~ ~J CO N O~ N Cl~ ~ ~) L~ 3 ~0 ~D O O
¢~ ~I . ~ r~ ~ . o r~ ~1 ¢ ~I ~1 ~1 ~I r~l r-l r-i ~1 ~J r-l ~1 ~1 ~1 ~1 ~ r-i ~1 ~1 U~ CO N ~I3 ~ r~ O r-t ~ If~
g ~1~`J r-l N ~I N ~J N r~ N r~ N ~ r-l ~1 ~1 ~I r-l ~1 H " 2 rl O o o o o o oo o o o o o o c o o o ,_, ~1 r~ Ia:) ~ ~ c~ ~u ~D CO CO ~ r~ m 1-- ~ .=S ~ ~1 ~ O
. . ~ r~ r~ r-l O O~ O ~D O O i a~ L~
¢ ~ r-l N Lr~ L~
E~ , '~¦ o O o~ O ,r.~ r~ ) 1~ N N ~ CC~ ~ ~I N r-l N r~
c~5 N N ~D ~ N N ~~D N N r~ r~ r~ N N N N r~
tn~J r~lr~ l N N r~J N ~J N r.~J N N N
~J ~ N N N N N N N ~1 N N N N N N N N
. .
~1~D r~ ) J Lr~ o~O ~ ~ r-1 0 3 t~ ~1 0 0 ~D ~1 ¢ ~ LS~ n ,~ c ~ ~ ~I t- co ~ ~ ~O ~
bl) 3 t-- Ln ro J' L~ 3 L~ Ln ~0 Ln CO CO 1~ t~
. ~ I , ~ m ~ m ~
.,, ~ .. , ~, ~ d I o , o I o , o ~ o I N r-l O N ~1 0 N
~a o~ ~ r ~
a~ ' 1~1 Cq E~
U N N ~I N ~ N N N N N N . N
~i O tD ~ I ~ 0 I N I
~3-~ p, ts I u~ I ~n I u~ I n I Ln I Lr~ Lr~ Lr~ Ln Lr~ n u~ .
..

h ~l ~l O ~ h ; ¢ ¢ m a~ C~ V ~ ~ ~ ¢ cC ¢ ¢ ~ ~ ¢ ~ ¢
~rl O ~ Q~
`~ :
aDr~ r I ~1 ~ ~ r-l : . ~ O O O O O O
h ~ h 1~ h CO
~ ~1 ,~ ,~ ~ ~1 : ~ O o o o O O
dC3 V C~

:; 3 ~ : :

The fine po~er starting materials A, ~, C and D
all exhibit low ~ SG5 1 values~ but these materials are not ~abricable by molding powder tec~niques because o~ the ten-denc~J of the ~ine po~er to stick to the pre~o~m mold and o~ the preform of the ~ine powder to crack upon sinte~.ng.
This poor molding ~uality is caused by the high porosi~y (greater than 0~2~) of p~e~o~ms o~ the ~ine powder~ high shrl~a~e of the sintered ~ine powder ~% S10OO greater than 9.0)and high (S) absolute value oP greater than 1Ø
On the other hand, the Examples shown in Table III
are good to excellent molding powder~3 having low pre~orm poro.sity o~ less than 0.20~ in m~st cases less than 0.17, and in many cases less than 0,15, (S) absolute value less : :
than o.8~ and smaller Q~F. In addit~on, these molding powders all have a considerably higher apparen~ density than prior art ~inely ground molding powders. Examples 8~1~3 which are conducted under di~eren~ partial decompac~
tion conditions, show generally that the more vigorous or extensive the decompaction is, the smaller is ~he SG5 1 value (improved moldability).
E~ les ~ ..
These Exa~ples show the e~fect in Table I~ of varying compaction pressure in the procedure used ~or Ex~le 3, using :~ine powder A B~ the st~rting material, ,'~ ' :

, :

36 .

: ~ ' ' ' ' ', L31~9 ~ o ~ :
~ U~ o .,~
al ~ ~o ~ ,( Lr~ 3 o t C~J O ~ ~ O C ~ D O
~O ~ ~ Lf~ I -~S ~ U~ ~D

~r Lr\ ~
~ ~o ~ o a~ t~ ~ L~
Q,~ o ~) ~ ~ o r~ 1~ ~ s o F; L~ L~ ~ 3 L~ . :

S: L 3 h~1 0 ~ ~ ~ o ri ~~ l C
~d ¢
~a~
I c~
. - . .
. ~ . . ~

~ ¦ --¦ I I o ~ o o u~
.. ~ C~
~ , E~
.
. . ~
,~, .~ . .
i N

X S~

- ' U~ ~rl ~ ' ,' o o 3 ~
u ~ ~ ~ .~ ~ ~

a ¢ ~ q P~
.
.~, ~ - .

.
: -: ` -37- :

6~

Co~parison molding powder A~ which læ not a prior art molding powder, is in~erior because of its low apparent density. As khe compaction pressure increases to 352 kg/cm and then to gre~ter than 562 kg/cm2, apparen~ density -lncreases, moldabil^lty improves and im~rovements in such p~operties as porosity and anisotropic expansion are obtalned.
E a~les 17 and 18 These Examples show ln Table V the e~ect of ~ary-ing partial decompaction temperature in the procedure used for Example 3 using flne powder D as the starting material.
,''' ~ '~

-38'-3~

~¦ 3~1 ~I r1 N N ~1 ~ i~i U~ O ~O~ O ~1 ~ ~ o P ~ ~ iY~ ~I Lr I
i~ ~ if~
~C
. .

r o , 1 ' a~ ~o 0 t~l , ~ O 1~ Lr~ O i~J
a) ~1 ~ i,~ ~ ~ O ~iY~ O ~1 ~ i~ ~D O ', 1~1 ., ~ . , '.
i~ O~ ~ ~1 O ~ 00 h ~ is~ o (:`J o ~ ~ o ~,i ~ ~ ~t o . ~ ~ 0 ~1 Ir~ I
~' ~ C) n~ C~ : ' ' ~ , :
.~: .
: ., ~ ' ~ ~ ~ co .~ o , t-- ~r ~ .
~ . . . . . . I
O I ~ I I ~ . I ' ' ,.
.: :
C~ " .
' ~ . O ' ': :', ., ~, , I , U~ , ~
` , . . O ~i o h O i_ :

o a~ a ~ <J ~
.

~ ' :: . , :: , , , , 3 9_ :
: -~
., . , ... , .. ,.,. , , , .. . -, . , . - . - , -These experimen~s show that various decompac-tlon temperatures can be used. In ~he case o~ Comparison B
molding powder~ a higher compaction pressure or shorter decompaction time would provide higher AD molding powder upon partial deco~paction.

Examples 19 and 20 ~ .
In these Examples, the procedure o~ Example 3 was repeated except a~ indicated in Table VI below and except that after coagulation the ~ine powder was not dried so th~t at the time o~ compaction~ the ~ine powder A con-tained 30~ by weight o~ water.
'r.ABrP VI
con~Y~ol E~.~O
Compaction kg/cm2 -~ 528 528 Deco~paction time (min) -- 4 10 ~D - g/l 506 713 671 SSG 2.222 2.21~ 2.220 a S ~ -1 0.7 18.5 2~2 Porosity 0.256 0.168 0.175 AEF 1.170 1.108 17124 S (5000) 5.6 3.6 3.7 (1000) 10~2 6.4 6.9 SSS microns 3.0 16 6.5 ~-(S) -2.26 -0.40 ~o.76 -4- ;
'..

- - . . .. , : . . -. .

3~;~
Ex~les 21-23 These examples show that lt is not necessary to use a Waring Blender for partial decompaction. In thes~
- experiments, each fine powder starting material is compacted in a press at 562 kg/cm2 pressure at 25C. Type E is the ~ine powder used for Example 21 and ~ype D is the fine powder used ~or Examples 22 and 23. The compacted ~ine powder T~pe E ~or Example 21 and Type D ~or Examples 22 and 23 were each mixed with water and ~ed con~i~uously to a Taylor-Stiles cutter~ Model TS-o6, with 15.2 cm rotor blades~ operating at a rotor speed o~ 9600 rpm and equipped with a screen across the discharge of the cutter. For Examples 21 and 22, the ~ ;
screen was a 30P screen (manufacturer's designation) in which the screen has a thickness of 0.21 mm, a minimum hole diameter of 0.33 mm and 14.5 percent open area. The hole has a pro~ile starting at about o~83 microns on one sur~ace o~ the screen and then decreasing to the minimum .indica~e~at the opposite side. For Example 23 the screen was a 60R VER0 screen (manufacturer's designatlon~. T~is screen has a thickness o~ 0.20 mm with a m~nimum hole diameter o~ 0~13 mm and 8 perc~nt open area. The holes have a rounded pro~
going ~rom about 0.42 mm diameter at one surface to the minimum give~ ab~v-e and then increasing again to the other sur~ace. The feed slurry at 8~Co was passed through the ;
cutter at a water flow rate o~ 1360 liters/hour and polymer feed rate o~ 22.7 kilograms/hour.
The product was separated ~rom the wat~r by ~lotation~ diluted with ten parts o~ fresh water and agi-tated in a slurry tank ~or 30 minutes at room temperature ~ 30 according to ~.S. 3,690~69. The slurry tank ~as eq~pped '' ~- .
~,, .

.:
~: ,, 3~ ~

with vertical ba~fles to increase turbulence and jacketed to control the temperature. The tank had a diameter of 45.7 cm and a height o~ 45.7 cm. The agitator blades were rectangular and four in number, each measuring 22.9 cm in diameter and 5.1 cm in height, with a ~5 pitch. The agi~
tator speed was 400 ~pm. Further details and results 0 these e~periments are ~hown in Table VII:

-~2-36~

~ o ,_ ~
CU C\J O r i ~ ~~ 1~ 0 ~ j ~ Lr~ o a, ~D
O L~ ~ C~ O
~ C~l ~ O ~ ~ U~

O O
~.1 ,1 O ~ ~ ~ 0LS~ t-- O
L~ r I C~ N ~I C`.l cO C--Orl U g CU O O ~i ~ (~ ~ .
X--' h O
~i O CO O ~
. ~ ~ ~O r-l O ~ L~ O
X C~i ~o O ~

H hO O j:
41 ~ ~ ~
O ~,.. ': :' E~~ X 1~ CJ ~

-- s~
o ::
~: :

:

o ~ o o rl O O ~3 ::
l: .
:` ~

-~3 -While the control ~ine po~ers in the Table exhibit high apparent density and low ~SG5 1~ these fine powders ~re not moldable by the usual molding powder ~abrication tech-niques, due in part to the sticky nature o~ the fine powder type of PT~ and in part to ~he high (S) value and porosity thereof. The molding powders obtained ~rom these ~ine powders are moldable like PTFE molding powder.
Examples 24 a d 25 These examples show in Table VIII that with the same feed resi~ (~ine powder A)~ comparable results can be obtained by Waring B~endor partial decompaction a~ter 562 kg/cm precompaction (~xample 24) and Taylor-Stiles partial deco~paction ~ollowed by aqueous agita~ion, followlng pre-compaction at 1760 kg/cm2 ~Example 25~.
TABLE VIII
~x. 24 Ex. 25 (Same as Ex. 3) Compaction kg/cm 562 1760 -20temp- C. 25 25 DecompactionWaring Blendo~ Taylor-Stiles 10 min. at 30C. plus cold water . wash AD-g/l. 713 620 SSG 2.220 2.226 ~SG5 1 11.3 1.2 Porosity 0.158 0.152 AEF 1. o89 1.101 ~ S(5000) 2.9 3.4 % S(1000) 6.o 6.2 SSS-microns 14 0.2 (S) -0.22 -0.1~ -.~. .
'. .

` -Exam~les 26 and 27 _ ___ _ ._ ___ _ These examples show in Table IX that the precom-pacted fine powder can be pa~tially decompacted by alr grlnding rather than decompacting under water. A 20~3 cm fluid energy mill was used, as described ~n U.S. Patent No. 3,726~48~. The precompacted resin was broken up in a shredder through a screen having 6.35 mm ape~tures so the resin could be fed ~o the air mill.
TABIE IX :~
__ _ Ex. 26 Ex.27 Fine po~er ~
starting material E B ;i.~ :
Compaction 2 pressure, kg/cm 562 3515 Compaction ; temperature~ C 25 25 Fluid energy mill feed rate, kg/hr 31.1 70.5 :
-: air inlet pressure~ ~g/cm2 7.03 7.6 feed air pressure, kg/cm2 7~74 7,74 ; ~r flow~ l/m 2830 2830 ~eed air ~:
temperature, C* ca25 ca25 Product SSG 2.169 750 5 1 4.5 3 ..
Porosity l 103 1 17~
% S (5000) 2,7 2.~ .;
:~ ~ S ~10003 5.g 6.~ :
~ SSS, microns - 12.1 ; (S) -0,03 -0.19 .
-45~ : .

:: .

. . . . . .. ..

~7~3~
Example 28 In this Example~ a series of compactions was carried ou~ at 562 kg/cm pressure and at 25C. using fine powder A
and partial decompaction was done to a varying degree to obtain the follo~ ng data for the resultant molding powder of this invention.
d50 mlcrons ~ 5~1 ____ .
66 Sl ; 29 5~4 This data shows the general relationship of increasing mold-ability with decreasing particle si2e. These data are plotted in Fig. ~ in which the high d50 region is plotted ; from the ~ollowing experimentally determined information:
at d50 f 206 microns, ~SG5 1 ~ 168; at d50 f 170 microns, G5 l ~ 147~ at d50 of 90 microns3 ~ SG5 l f 115. From ~ig. 4, it ls a}so possi~le to determine d50 particle size from the determi~ation of ~SG5 1 on the molding powder. For - example, at ~SG~ 1 ~ from 0 to 75 ~hich enco~passes Examples 3-27 herein, the d50 part~cle size of the molding powders of the present invention is from about 30 to 70 microns.
Examples 2g-3 ~ e~ation _____ Molding pow~er o~ the present invention o~ ~inely ground beta polymer was agglomerated by stirring with tetra-chloroethylene and w~ter at 25C. for about 15 minu~es in a

2-l~ter glass resin kettle ~itted with four one-hal~ inch baffles and a ~tirrer set to operate a ~our-bladed l~5-degrea down~dra~t agi~a~or at 2000 rpm~ The solven~ : PTFE ratio

3~

(ml sol~rent: g PT~E) is shown in the Table X and ~he PTFE:
water weight ratio ~as about 1:10. One hundred grams o~ the ground beta resin was used in e~ch e~peri.men~. The product after separation and drying, had the Iollowing character-istics shown in ~he Table X. Properties of the ~inely ground bata resins used as the sta~ting material are included u~lder the heading con~rol.

' .
~''' : ', ..
. .

'..

~. .. . . , -.. .. ..... . : . . - . ~, . ,:; . .

3~ ~

C~
a~ ~ ~ L~
=~ N , ~rl O ~ l O ~

;o ~IS~ Q C~ -o ,~ N
. ;:
~ l O ' '.
o~ i O
~ ~ ~ ~ ~r-l N (Y) 'f~ ~
(~1 . . ~ . o N
~r) O r~l O C~ ~
O
' , .

~ ~I N ~U~ Lr~ L(`\ N t~
t~ O ~ l O C~
~}
~' ~
rd ~ :~ N C~
.~ ~1 ~ c~ o~ ~ ~~1 ~1 ~ ~ O . . . ., . ,~
~ ~ lr) O '~ C~ l ~3 ., O
., , bD CO ~
~'~ ~ ~ ~ ~ ~1 ~ ~ ~ C~ ~ ~1 ,1 N Ori C\J V C~
. cO~
IS~ (X) h ~ ~ -O ~ O 0~

~ ~ s3 5~ ? : -a) ~ ~ ~, c~
: ~ Q ~ u~ r U~ ~ ~ g ~ ~ O ~ "~
O ~ ~ o ~n ~ ~ . .
~:1 V3 ¢ V~ Cq ~ E~
. '.
',. :
~8 ''' Example 35 - Agglomeration __ _ Typical moldin~ powder derived ~rom fine powder A
was agglomerated in the 45.7 cm stirred tank o~ Examples 21-23. The char~e was 4.54 kg polymer, 38.14 kg water, and 1200 cc tetrachloroethylene. It was agita-ted at 860 rpm ~or 30 minutes at 25C. The product, a~ter separation and drying, had good sinte~ability and excellent AD and powder flow as shown by the following properties:

Calc ~D-g/1 875 Powder flow, g/sec 31 SSG 2.217 - -~ S (5000) 3.18 d50-microns (Av. agglom.size) 590 Tensile strength/elongation, kg/cm2/~
70 kg/cm2 preform pressure 167/227 141 kg/cm2 It " 204/326 352 kg~cm2 It ll 245/453 Using the same equipmen~ as in Example 35, an experiment was conducted chargin~ 6.82 kg of product made by pa~tial decompac~lon of compacted Resin F, 38.1 kg water, and 3000 ml tetrachloroethylene~ The mixture was ~tirred at 860 rpm for 30 minutes at 25C , separated~
and dried. Follvwing are its propertles: ;

.,. ' ,:

. . .
,'.
~9 :~ .
'~ - ,, ~7~31~

Calc AD-g/l 8~o Powder f`low g/sec 28 SSG 2.175 5-l 4 % S (5000) 2.92 d50-microns (~v. ag~lom. si.ze) 225 Tensile strength/elongation, kg/cm2/%
70.3 kg/cm2 preform pressure 198/292 140 . 6 kg/cm2 ll 'l255/316 352 kg/cm2 l~ "280/374 ' .

, , -, , .
~ `, , ~.
..
' . .' L ~ ~

: .

. :

, , . - .

:~ -50- ~
: ~ ' .
. :

. -. : - . . .. . .. . .

Claims (15)

The embodiments of the invention in which an exclu-sive property or privilege is claimed are defined as follows:

1. Polytetraflyoroethylene molding powder having a specific surface area of at least 1.5 m2/g, an average particle diameter of less than 100 microns, and an apparent density of at least 500 g/l, with the proviso that the apparent density is at least as great as the value calcu-lated from the equation Apparent density ? 500 g/l+3.00(.DELTA.SG5-1) wherein .DELTA.SG5-1 is 1000 times the difference in specific gravities of sintered moldings made at preform pressures of 1000 (70 kg/cm2) psi and 5000 psi (352 kg/cm2) wherein said .DELTA.SG5-1 is no greater than 75.

2. The polytetrafluoroethylene molding powder of Claim 1 wherein .DELTA.SG5-1 is zero.

3. The polytetrafluoroethylene molding powder of Claim 1 wherein .DELTA.SG5-1 is no greater than 30.

4. The polytetrafluoroethylene molding powder of Claim 3 wherein the apparent density is at least 600 g/l.

5. The polytetrafluoroethylene molding powder of Claim 1 wherein said apparent density is at least 600 g/l.

6. The polytetrafluoroethylene molding powder of Claim 5 wherein said apparent density is at least as great as the value calculated from equation Apparent density ? 600 g/l + 3.00(.DELTA.SG5-1)

7. The polytetrafluoroethylene molding powder of Claim 1 having a porosity of no greater than 0.20 and an (S) absolute value of less than 0.80.

8. The polytetrafluoroethylene molding powder of Claim 1 having a % S (1000) of less than 8.o.

9. The polytetrafluoroethylerle molding powder of Claim 1 wherein the individual particles thereof exhibit birefringence when viewed microscopically with polarized light.

10. The polytetrafluoroethylene molding powder of Claim 1 in the form of free-flowing agglomerates having an average agglomerate diameter of from 300 to 1000 microns.

11. The polytetrafluoroethylene molding powder of Claim 1 having an average particle diameter of at least 10 microns.

12. The polytetrafluoroethylene molding powder of Claim 11 having .DELTA.SG5-1 of less than 15.

13. The polyrtetrafluoroethylene molding powder of Claim 12 having an average particle diameter of 10 to 20 microns.

14. The polytetrafluoroethylene molding powder of Claim 1 having a tensile strength of at least 3500 psi (245 kg/cm2) at 1000 psi (70 kg/cm2) preform pressure.

15. A process for making polytetrafluoroethylene molding powder from polytetrafluoroethylene granular resin which comprises (a) separating beta granular resin from a mixture of beta granular resin and alpha granular resin, and (b) milling the separated beta granular resin in the absence of alpha granular resin to increase its tensile strength and its specific surface area to at least 1.5 m2/g, and to have an average particle diameter of less than 100 microns.