CA1041263A - Reinforced formaldehyde-based molding materials - Google Patents

Abstract

REINFORCED FORMALDEHYDE BASED
MOLDING MATERIALS

ABSTRACT OF THE DISCLOSURE

This invention is concerned with increasing the impact strength and other properties of formaldehyde-based thermosetting molding resins. This is effected by compounding short length fibers of poly(vinyl alcohol) with the resins. The compounds possess excellent moldability particularly where molding is achieved under ram pressure, as with injection and transfer molding.

S P E C I F I C A T I O N

I.

Description

This invention relates to fiber reinforced formaldehyde-based thermosetting resin compositions, to methods for producing shaped articles, and to the resulting shaped articles. More particularly, this invention is concerned with fiber reinforced formaldehyde-based ~hermosetting resin compositions which possess excellent processability and melt t. -flow characteristics, and hence, can be effectively . and advantageously em~loyed in all of the known processes for molding thermosetting resins and produce shaped molded products having excellent physical properties, particularly high impact strengths, excellent surface appearance and unique ` low water absorption.
In recent years, considerable progress has been made in the technology of molding formaldehyde-based thermosetting resins. A most significant development has been the recent advances in the screw injection molding process enabling direct in-line injection molding of fast curing thermosets.
It is now possible in many instances to injection - mold parts from thermosetting resin co~pounds at curing cycles even shorter than the normal cooling cycles required for thermoplastics, especially when molding thirk walled or large parts. The reason is that thermosetting materials cure during molding and can be discharged from a hot mold; in contrast, thermoplastic materials must be cooled in the mold before discharge. A concurrent development has ~.j~

~824 ~ v ~3V
been the recent advances made in mold designs for thermosets which are referred to in the art as '~arm"
ori'cold runner", "runnerless" or "hot manif~Ld"
moldsO These new mold designs minimize ~he amount of crosslinked scrap formed in the sprue and runner section of She in~ection or transfer moldsO Sinr.e crosslinked scrap is not reprocessable, the new mold designs have enabled substantial s~vings~in raw materialsO As a result of the lower process-and.
raw material costs made possible by these~evelop~
ments~ molded parts-from thermosetting resin composite~ are becoming increasingly compè~itive with injection moided parts made from the so called high performance thermoplastic resinsO
The injection and tran~fer mol~ing processe,~
require excellent "mol~ability" in order to realize 'the potential of ~ast molding cycles a~d to enable proper mold ill out of intricate parts with thermo ~etting compound~p The "moldability" of reinforced thermose~ting res~n composition~ is very d~ff~cult to ch~rac~erize in terms of fund~mental rheological parameters for several reasonso First, the m~terials are generally non~Newtonian in the sense that the viscosity decreases with increasing r~tes of shearO
Second, they are frequently Bingh~m plastics in tha~ they have a yield point, i~eO 9 no flow occurs un~il the shear stress exceeds a certain critical (yield stress) levelO Third, in c~se of fiber reinforcement, the flow ls Yeriously impaired when 3~

1~ 6;~
the length of any reinforcing fibers used in the compound approaches the cross sectional dimensions of the ac~ual flow channels. Last, but not least, since thermosetting materials are heat reactive, the rheological properties change continually during molding and flow testing until the material iS completely crosslinked and incapable of flow.
It has, therefore, become standard practice in ~he art ~o characterize the flow behavior of thPrmosets in terms of more or less empirical "moldability" tests. For the purpose of this invention, the molding behavior of thermosetting resin composites is characteri~ed by a flow test (hereinafter called "Spiral Flow") similar to that described by K.R. Hoffman and E.R. Fiala in paper XXIV-2 from the Annual Technical Conference of the Society of Plastics Engineers, Vol. 12, 1966 entitled "A Simple Ram Following Apparatus Applied to Spiral Flow of Plastic Molding Compounds" with the following modifications:
(a) the cross-section of the flow channel is 0.125" x 0.375"; (b) the molding material is charged to the apparatus as a preform preheated to 121C;
(c) a mold temperature of 168C is used and (d) the transfer pressure is 8,800 psi on the ram. For the purpose of this invention, the "moldability" of a thermosetting compound is characterized by the number of inches the material is capable of flowing within the mold channel before setting up under the ,. , ,^ /, , i, ,.. --"" .. ,, . ", .y " . , ., , , . " " ,, , ,. , ,, " ,, ~.

conditions of this test. I~ has been empirically established that good performance in injection molding requires a tPst flow equal to or greater that 24 inches, and good transfer molding requires a test flow equal to or greater than 18 inches in this test. Materials having spiral test flows below 15 inches are generally suitable only for the less demanding compression molding techniques.
Thermosetting polymers such as phenol-formaldehyde, melamine-formaldehyde and urea-form-aldehyde resins generally have high modulus, excellent dimensional stability, good thermal and chemical resistance and low cost. Their greatest single deficiency is brittleness as reflected in high notch sensitivity, low fracture toughness and low impact strength. For this reason, these resins are almost invariably reinforced for improved impact strength and toughness in their practical applications. The reinforcing techniques used range from the use of elastomers which are at least partially soluble in ~he resin such as the use of nitrile rubbers in phenol-formaldehyde resin;
- elastomeric inclusions in the form of a separate discrete rubber phase in a continuous resin phase such ; as the use of CTBN (carboxyl terminated poly(butadiene-acrylonitrile)copolymer) rubber particles in epoxy resins; macerated fabrics and coarse cord fibers as used in compression moldable phenolics; non-r - 8824 "

woven felts and fabrics or organic and inorganic fibers impregnated with liquid ~hermosetting resins - such as the use of sisal matting impregnated with liquid polyester resins; woven fabrics such as the `~, use of cotton fabrics and glass fabri~s in making , reinforced laminates from phenolic and polyester ` resins; etc.
These approaches have serious practical limitations. The use of soluble elastomers lowers the modulus and the heat distortion tem~erature (e.g~, hot rigidity) which results in warpage when parts are removed from the hot mold and thus limits the uses for which the compositions can be employed.
The use of rubber inclusions as a discrete phase causes less reduction in modulus and heat distortion temperature since the continuous resin phase remains unplasticized; however, the impact reinforcement caused by the rubber particles is negated when fillers are added because the solid particles interfere with the mechanism by which the rubber particles provide toughness reinforcement; also, the presence of the rubber phase impairs the oxidative and thermal stability of the composite. The use of , . . .
~ long fibers and m~era~ed fabrics, including woven ; and non-woven fabrics severely limits the general processing and molding characteristics of the thermo-setting composites as well as impairs the surface appearance of the slolded parts.

' 6~ .

. . .

,. ..

It has long~been recognized- in the art of fiber reinforcement t~t both compouTId vis~oglty and impact re~is~ance increase markedly~ with in~
creasing fiber length and fiber concentrationO For example, general discussions of these phenomena are given by P, Robi~scher~ et al" in Phenollc Resins, Publo by Iliffe, Londorl, 1950, pages 104-106 and by ToSo Carswell ~n ~, Publo by Inter-science, New York, 1947~ pages 86-980 Da~a-on the effeet of glass fiber length on the impact strength of fiberglass reinforced polyesters are reported by RoB~ White ln Premix Moldin~ New York, 1964 (Reinhold), pages 60-61 and the e~fects of the leng~h and conceritrat~o;i of organic iEibers on the impact strength of phenolic resins are di~scussed by R~Ho Mo~her and JoBo Gr:Lffin in the artiicleo "Physical Stren~th Properties ~f Molded F~er-Resin Preformed Mater~ ", 1l7~m l'la~clcs; pages 147-152, Febru~ry 194So In ~greernen~ with the above re~erences, it has b~en the general experience in the art of iber reinforced ormaldehyde-based resins that Izod impact 8trengths of the order of on~ foot pound per inch of notcL; or ~reater requires the use of ~ber~ ~t le~st r/8 ", ~preferably at least 1/4 " in length or fiber c~oncentratiqns greater t~i~n 40 ~t, %, prefer~bly greater than 50 wto %~ or bothO At these fiber length~ and/or fiber concentrations, the . .
molding materials are difficult to prepare, the bulk densities o~ the materials are very low and automatic eeding to molding machines is cumbersome or impossible~ Even more importantly, the moldability of such compounds is too poor to enable molding by transfer and injection molding, the most eeonomical molding processesO In fact3 current high ~mpact for0aldehyde basea resin compounds are made with long fibérs (> ll47'~ in high concentrations ~>40 wt, %) ~nd have such llmited melt flow characteristics that they can only be compression molded into objects requiring very slight flow, Specifically, fiber reinforced formaldehyde-based molding materials have heretofore not been commercially a~ailable which have Izod i~pact s~rengths greater than one foot pound per inch of notch ~nd at the same time having molding ~haracteristlcs suit~ble for in~ectlon molding.
These ~roblems have now been overcome by - .
the present invention which provides formaldehyde-ba~ed molding material compo~itlons which are capable of being injection and trans~er molded with good toughness in the final products as characteri~ed by an Izod impact strength in excess of one foot pound per inch of notchO The present invention further provldes excellent surface of the final molded products, far superior to the surface attain-: able with coarse fibers or macerated fabricsO Also, the compositions of ~he present invention provide ~ .

~ Z~ 3 molded products with lower water absorp~ion than has generally been posslble ln the past for products with similar toughnes~O
These and other obJectives are achieved by admixing with formaldehyde based resins such as phenol-formaldehyde resins; bisphenol-formaldehyde reslns; melamine-formaldehyde resins; and urea-orma1dehyde resins; and mixtur~s thereof; from about 5 to about 50 weight %, preferably from about 10 ~o about 40 weight ~0; bAsed on the weight of the mixture, of poly(vinyl alcohol) fibers having a weigh~ average fiber length from abou~ 1/100" to about lt8", preferably from about 1/50" to about 1/10~, and having a f~lament deni.er from aboùt lt2 to about 100, preferably from about 3 to about 30. For example, when poly(vinyl alcohol) fibers (6 denler), as~short as 1/25" ln length, are used in concentrations from 10 to 40 wto % in the above resin matrices, Izod impact strengt~s ranging from about one ~o about three foot pounds per inch of notch can.be achieved while at the same ~me maintaining spiral test flow values in excess of twenty-four inches. The final moldings have excellent surface appearance andilow water absorption,`
Without wishing to be bound by any theory or mechan~sm, it;is presently believed ~hat t;he surpri8ing ability of very short poly~vinyl alcohol~ ("PV-OH~') flbers in providing exceptional impact reinforcement in formaldehyde-based resin systems is due to fu~ion ~ 3 of the fi~ers into a three-dimensional network during molding, In other words, the fibers remain as dis- -crete fibers during the flow stages of the injection or transfer molding processes while they= fuse together in situ into a more or less continuous non-woven fiber mat durlng the high heat and pressure exposure in the mold. According~y, at the stage where good flow ~s required, the short fibers cause relatively lit~le lmpairmentof the flow properti~s whereas at the stage where long fibers are required for impact reinforcement, the originally short fibers fu~e into very long and intermeshed mat structurec.
In order for this phenomenon to oecur, it appears that the fibers must be able to fuse together .
but not completely dissolve in th~e resin matrix which would result in complete destruction of the fiber structure. Al~o, it i5 belie~ed that the pre~ence of a certain amount of water or other suit-able fiber ~olv~t~ng ~ent in the resin matrix acilitates fiber interfusion over a broad molding temper~ture rsngeO For example, water is genera~ed during the condensation and curing of the above resin matrices and the water generated under the conditions of high temp~erature and pressure is believed to assist the ^thermal fusion of the ~ibers and achieve interfusion without loss of the mechanical integrity of the fibers over a broad range of molding temperatures.

10.

La~ 3 A~cordingly, the temperature during molding should be above that-necessary to cause fiber inter-.

fusion, yet below that causing com~lete fi~er melting and loss of integrity of the fibers. In the presence of small amounts of water, neat poly(vinyl alcohol) fibers (i.e. without any matrix resin) will bond readily to each other under application of pressure at a temperature range of 100~C to about 220C. How-ever, at temperatures in excess of about 230C, the fibPrs will fuse into a solid sheet with li~tle or no retention of the original fiber structure. Accord-ingly, due to the fiber fusion tPmperature range for PV-OH fiber reinforced formaldehyde-based resins, the actual molding temperatures should remain in the range of 100-220C.
The compositions of this invention are :`
~ also critical in certain aspects, to wit, the poly-`~ (vinyl alcohol) fiber length and the weight percent of the fiber as expressed above. If the fiber length is too short, then the impact reinforcement becomes very inefficient and if the fiber length is too long, then the flow characteristics of the composition will be inadequate for injection or transfer molding. If the fiber loading is less than that speciied, then little of the type o reinforcement characteristic of this invention is noted and if it is higher, then the ~-flow behavior of the composition again becomes unsuit-a~le for injection and transfer molding.
The discovery of in-situ fiber fusion of poly(vinyl alcohol) fibers in formaldehyde~based ~' .
11.

,. .

4~L~263 ~
resin systems during molding is not per se the invention of this disclosure. The fiber-fusion phenomenon will occur at any fiber length abvve a certain minimum fiber `, loading and still the compound could have very limited utility. This invention is concerned with making a ., .
moldable formaldehyde-based ther setting resin compound ' with exacting flow characteristics to meet the stric~
demands of transfer and injection molding and produce ' a molded product which possesses a unique combination of desirable properties, e.g. high impact strengths, excellent surface appearance and low water absorption.
The general utility of poly(vinyl alcohol) fibers and fabrics as reinforcement for thermosetting resins is well known in the art of fiber reinforcement. However, the results discussed herein have not hereto~ore been recognized. The novel com~ositions defined herein provide uniquP combinations of properties which have long been sought in the art of molding thermosets but which have herPtofore been unattainable.
The advantages of this invention are not restricted to making articles by transfer and injection molding. The com~ositions of this invention can often be employed advantageously in molding techniques which are less demanding than injection and transfer molding.
: For example, short poly(vinyl alcohol) fibers of, say, 1/25" - 1/16" length can be added to an adhesive or binder resin with comparative ease and without seriously ,, .

~ ~ 8824 Lf~3 impairing the flow charac~eristics of the resin, Never-theless, the desirable high impact characteristics are still achieved when the composition is molded, Thus, the fiber reinforced formalde~de-based resin compositions of this invention are useful for thermosetting adhesives and binders where improved fracture toughness and reduced notch sensitivity are desired. In contrast, addition of large concentrations of long fibers has been necessary in the past to achieve adequate improvement in toughness.
This has not been practical in adhesive and binder resin systems because of the severe impairment of flow properties.
It is generally desirable to use poly(vinyl alcohol) fibers of equal length, i.e. monodispersed fibers. However, even if the original fibers all were of the same length, a certain amount of fiber attri~ion often occurs during compounding and processing so that ~` the final molding compound usually has a distribution of fiber lengths. For the purpose of this invention, the fiber length for polydispersed fibers is understood as the weight average fiber length which is defined by the following equation :; .
W~
:, O ~ W
,,, ~ i n ;- where Wi is the weight of fibers having a length ~ i.
If all fibers have the same diameter~ the weight average can instead be determined from the following equation ~,ni R i2 ~ ~ n1 R L

:

~ 13, 88~4 4~ 3 where nL is the nu~iber o~ fibers having a leng~h ~i.
indicate~ summation of over all ~ber lengths.
Also, ~or the purpose of the presen'c invention9 ~he f~bar l~ h is the actual fiber length occurring in the finished m~3ldlng ma~erial no ma~er what ~he length might b~ of the fibers originally added to the resin.
This fiber length di~tribution can be d~t~rmined from the uncured moldin8 material by s~parating the f~bers from :~ the uncured resin. For exampleg the flbers may be iso~ -10 lated by 8elective sol~rent extraetion of the uncured re~in matrix. In cases where other fil:Lers are present in the moldlng material, the poly(~inyl alcohol) fibers may be eeparated rom such other fil~rs by phy~ical means æuch as segregation by density (e.g. uslng a liquid with a ~exl~lty interm~diate betwee~ ~hose of the fibe~ and 'che other ~iller39 by screenin~ etcO, depending on the particular iller composition.. Im some cases, ~he poly~
(vi~yl alcohol) fiber len~th distributiLon can be deter~
mined mlcro~copically even wlthout separ~tion o:E ~he 20 flberfi .
The poly(virlyl alcohol) :Eibers may bc incorporated into the various formaldehyde~based re8ins by a variety of means ot~ the purpose of preparing molding m~terialsO
In the ca~e of urea arld mel~mine~formaldehyde re~ins, fiber~, fillers and other additive~ such as pigments, lubr~cant~, etc., are commonly adted to a 30~60 wt. %
~qu~3ou8 solu~iorl of the resinO The wet mixture i~

~40 subseqllently roll milled or sheared in an internal m~Ser, dried and granulated. Simllar technigues can be used with phenol and bis-A ~ormaldehyde resins w~ereby ~he compound~
lng agen~s are first impreg~ed wi~h the resin in liquid or solutlon fo~n. T~e short le~gth o~ poly~vinyl alcohol) fibers u~ed in the compositiorls of ~his invention 8i.111plifie9 these blending processes in comparison ~o the current high lmpac~ composi~:ions which are based on long fibers, macera~ed fabrics, e~cO
Hot mixing of solventless resin with ~ibers, fi~lers, pigm~nts, lubrica~t 3 hardeners, etcO ~ ls extens~vsly used for preparlng standard phenolic molding matèrials. This procedure is particularly desirable for making co1npositions covered by this inven'cion becau~e of the~ ready procegsability of the short leDgth poly(vinyl alcohol) fibersO The process is carried out on ~wo roll mills, banbury mi~ers or ~crew mixing m~chlne~O The final moldi~g m~terials may be prepar~d in granular or powder f~rm, a~ hot or cold comp~cted t~blets or as ex~ruded pelletæ~ It i8 also possible ~o prepare flber concentr~es or masterbatches which æubsequently can be dilu~ced prior to mo~ding with starldard molding;
materials having le~s or ever no poly(vinyl alcohol~
f~ber content~ F~ber m~s~e~batches m~y be prep~red at any fiber loadings by appropriate m~ans, but preferab~y co~in from 20~50 wtc % of poly(vinyl alcohol) ibersO
Th~ opportunity to use a~y on~ or more of the above 15 .

, , 8824 .-~4~6~ :
processing techniques stems principally from the short fiber lengths and moderate loadings of poly(vinyl alcohol) fibers characteristic of the compositions of this invention.
It has Pven been found possible to blend the short poly(vinyl alcohol) fibers externally with standard granular or powder form molding materials containing resin by simple mechanical blending. Such molding material/fiber blends have been fo~nd to feed surprisingly well to screw injection molding machines without bridging in the feed hopper or creating conveyance problems in the feed section of the screw. In screw injection molding all of the above blending techniques--including simple external fiber addition to a powdered or granular molding material--has been found to give excellent results for the compos-itions of this invention.
Poly(Vinyl 'Alcohol) Fibers Poly(vinyl alcohol~ is normally made by ~, saponification of poly(vinyl acetate) in a methanol solution. The precipitated poly(vinyl alcohol) is ~' 20 next refined and dissolved in ho~ water to provide a spinning solution. This solution is then extruded through spinnerettes into a coagulating solution such as an aqueous solution of sodium sulphate whereby the fibers are formed. The fibers are then heat treated and '- acetalized to rendex the, fibers insoluble in water. The ~' fibers of the present invention have deniers ranging from one half to one hundred, preferably from three to thirty.

.

16.

3 :~
The denier of a textile :Eiber is the weight in grams of nine thousand md~ters of ~he fiber l'ilamen~c Th~ curable pheTlolic resin componellt of the con~positions of the pres~n~c invention are the condensa~ion products of a phenol with an aldehyde. Suoh condensation product~ are divided into two cla~ses, resoles and novolaksO These two types of resins are discussed in order below.
Resole Resins Resole resins are most reque~tly producad by ~h~ conslensation of phenols and aldehydes u~der alkaline cot~d~tionsO Resole~ difer from novolaks i~
that polynuclear methylolc subqtituted pherlols are formed as intermedlates in resoles~ A resole produced by the ~; condetlsation of phenol with formaldehyde most likely proceeds through ~n intermediate having the following illustrated type struc~ure ~, CH;~OH CH 2OH

HC)~ ~z-OGCH2~0H
, \~ ~ ., CH20H ' ~20H' In a ~yp~cal synthesis, resoles are prepared by he~ting o~e mole o phenol with 105 moles of formalde~
hyde under alkaline condltions.
~:: The resole resins are prepared by the conden-8ation of phenols with ormaldehyde, or more generally, by the reaction of ~ phenolic compound, havlng two or 8~2 ~hree reactive aromatic ring hydrogen positions, with an aldehyde or aldehyde-liberating or engendering compound capable of ~ndergoing phenol-aldehyd2 condensationO
Illustrati~e of phenQls are phenol, cresol, xylsnol, al~yl phenol~ such as ethylphenol, bu~ylphenol, nonyl-phenol, dodecylphenol, isopropylmethoxyphe 1, chloro-phenol, resorcinol, hydro~uinone, naphthol, 2,2-bis(p-hydroxyphenyl)propane(bisphenol A), and ~he like, and mi~cture~ o~ such phenolsO Bi~phenol A--formaldehyde resins 10 are describ~d further in the followingO Illustrative of aldehyde~ are formald~hyde, paraform, acetaldehyde, ~crslein, crotonaldehyde, furfural 5 and the like .
Phenol--furfural r~sins are particularly valuable ~or in~ection moldi~g co~posites b0cause they can b~ held 81ightly below the molding temperature for extended periods wlthou~ hardening, yet they cure quickly whcn the t~mperat~re i8 elevatedO Illu~trative of aldehyde-liber-~ting compounds is 1~3~5-trioxaneO Illustrative o~ the ald~hyde engendering agent8 is hexamethylene ~etrami~eO
; 20 Ketones such a~ ace~one are al60 c~pab~e of condensing with the ph~nolic compounds to ~orm phenolic resins.
The . ondensation of a phenol and an aldehyde 1~ conduct~d in the pre~ence o~ ~lkaline reagen~c~ such as sotium carbonate, ~odium acetate, sodium hydrox~de, am~nonium hydroxide, bariu~ hydroxide, c~lcium hydroxida, and the like. When the condensatlon reaction is com-ple~ed, if desiréd, the water and other volatile m~terials can be removed by distillation, and the ca~alyst neu~ralized.
The resole resins are 'cermed heat curable or one-step reæin~O That is, under the application of heat these resins progressively polymerize until they are finally in~oluble, lnfusible and completely cured.
For the purpose of ~he present invention the curable phenolic resins are considered those which have not so advanced in polymerization that they have become infusibleO
Novolak R~ins Th~ novolak res~ns are prepared in a manner similar ~o that used to prepare the resole resins. One distinguishing exoeptlon in this preparation is the use o~ an acidic media to catalyze the reaction instead o~ an alkaliIIe media as is the case with ~he resole~.
:~ The phenolic compounds useful in novolaks as well as ~he aldehydes are the same as are useful in the preparation o~ r~sole r~sins. When less than Rix moles of formalde- ~:
hyde are us~d per seven moles of phenol, the products 20 ar~ perman~ntly fusible and ~olubleO Theæe are the novolak resins. The novolaks have a diferen~ structure than the re~ol~s as ~s illustrated by the no~olak conde~-sation products of phenol with forn~ldehyde:
(a) OH - OH
~=C~2~, , .

,, .

;~ ha~63 HO ~ CH~ ~ OH

OH
HO ~ H2 -~ In a typical synthesis, novolaks are prepared by heating one mole of phenol with 0,5 mole of form-aldehyde under acidic conditions~ The temperature at .
which the reaction is conduoted is generally from about 25C, to about 175C.
It should be noted that the novolak resins are not heat curable per se and are, therefore, often referred to as two-step resins. Novolak resins are cured in the presence of curing agents or hardeners such as formaldehyde, hexamethylene tetramine, para-for~aldehyde, ethylenediamine-formaldehyde, and the ~, ~ e.
:
Therefore, for the purposes of the present invention curable novolak resins are considered to be novolak resins in the presence of a suitable curing -~ agent. Such curing agents are generally used in amounts of from l to 30 parts by weight per 100 parts of novolak resin. Curing is generally effected by heat.
Phenolic molding resins prepared by condensing ~n aldehyde such as formaldehyde with a mi~ture of a phenol and up to 100 parts by weight o~ an amine ~uch 20.

urea, melamine ox aniline per hundred parts phenol, provide highly desirable proper~les when used lnJche molding compositi~ns~ These resins are commonly known as melamine~phenolic resins, urea~phenolic res~ns, aniline~
phenolic resins and the like 9 depending on the difying compoundO
The amino phenolic~aldehyde resins such as the urea~phenolic resins,^the melamine~phenolic resins, and the aniline~phenQlic r`es~s can be used in the molding 10 compositions of thLs inven~ion, Bis~henol-Formaldehyde Resins The chemistry of bisphellol formaldehyde resins is similar to that of the phenol-formaldehyde resins with the exception that bisphenol~ are used instead of the simple phenolsO Because the para- :
--, po~itions in the bisphenols are blocked, they are less :~ reactive than the simple phenols; hence, more v~gorous ccnditions are needed to react with formaldehydeO
~` Bisphenol A is a particular me~iber of ~he bisphenol family which may be represented by the following - ~, formulaO
: ' C~3 . :

H(~X~OH H~ ~OH

Gener~l Bispheno~ls _ Bisphenol A
/2,2-bis(4-hydroxyphenyl)-propane/

~824 1~4~LZ~
In the case of bisphenol A, X is isopropylidene. X can represent other divalent organic or inorganic residues such as oxygen, sulfur, methylene, ethylidene, sulfonyl, sulfoxide, and the like.
Bisphenol A-formaldehyde resins may be produced in either heat-curable sr non-curable form in analogy with the resole or novolak resins discussed earlier. The heat-curable resins are prepared under alkaline conditions (caustic is preferred~ at a bisphenol ~ 10 A-formaldehyde ratio between 1:2 and 1:3, and at a -~ reaction temperature from 90 to 100C. and higher. The crude resin is acidified to slightly acid (pH 5 to 7~
and dehydrated under vacuum to give a colorless, solid resin which softens at about 85C.
The non-heat-reactive resins are produced under acid conditions, usually at a bisphe~ol A-formaldehyde ratio of 1:1 to L:1.5, at a reaction temperature of about lOO~C. The crude resin is nearly neutralized and dehydrated in the same manner described above. The non-heat-reactive resins are also colorless and soften J~: at about 131C. Like the novolak phenolics, they can be :`:
cured by "hexa" at elevated temperatur2s.
Bisphenol-formaldehyde resins offer important advantages over phenol formaldehyde resins, in that (a) they are nearly colorless and light stable`enabling formul-ation of light colored molding materials and (b~ they have excellent resistance to premature hardening at Z~:i3 temperatures slightly below the molding temperature, yet they cure quickly at the proper temperatureO This latter feature is of par~icular commercial importance ~or injection and trans~er molding applicationsO The compos-ition of such bi~phenol A-foxmaldehyde resins per se is not the invention of the applicants hereofO
Urea-Formaldeh~de Resins Urea, or carbamide, is made commercially from ammonia and carbon dioxide unde~ high pressure and temperatureO The chemistry of reactions between urea and formaldehyde is very complex, and the nature of the condensation products is still no~ fully understood~ :
Under neutral or alkaline cond~tions, it is known that : urea reacts with formaldehyde to produce mon~methylol urea and symmetrical dimethylol urea depending on ~he -~
: ratio of the reactantsO
,~ / , ~ ~FlCH20H
C = O ~ -- o '~ ; \ 23H~ n20H
20 Monomethylol Urea symO-Dlmethylol Urea They may be regarded as the monomers of the urea-~onmal-dehyde resins. Upon acidification and heating, these monomers undergo condensation and crosslinking reac~ions resulting in insoluble, in~usible productsO
On the other hand, under acid conditions9 bi~amides are formedO The reactions are highly pH -dependent and difficult to control.

2 3 D

882~ .
~(14~ 3 In the produc~ion of urea molding resins, the more readily controllable alkaline process is usually employed. Aqueous solutions of urea and form aldehyde are mixed in a reaction kettle at a ratio usually between 1:1.3 and 1:108, toge~her with a suitable alkaline catalyst such as ammonia, ealcium hydroxide, alkanolamine . or sodium hydroxidet Condensation is usually carried out - in glass-lined or stainle~s steel reactorsO
; :~ Produ~ts formed during the initlal stages of the re~ction are wa'cer~sol~ble and consist mainly of mono-and dlmethylol ureasO The reaction mixture is then heated ~; to boiling and held until it becomes a viscous mass. The liquid resin is then filtered and flllers are usually , addedO The resulting damp mixture is then screened and ~; : driedO In the conventional a~t practlces, the coarse . powder is subsequently blended with o~her compounding . ~ .
ingredients; for example, pigments or dyes, lubricants, curing agents and stabilizers, in ball mills to reduce the mixture to a ~ine, uniform powderO
Commonly used fillers for reinforcing urea molding compounds include pulp flock (c~-cellulose~ and-wood flockO An acid curing agent of either the direct or the latent type is usedt The former are true acids and are effec~ive at all ~emperaturesO Typical examples are inorganic and organic acids and acid salts such as sodium bisulfate~ The latter type~ on the other hand~
form ~cid products only at elevated temperatures and, 24.

~ 3 usually, in co~junction with moisture liberated during ::
the mold~ng operationO Consequently, they exhibit no crosslinking activities at ambient temperaturesO Typical latent catalysts include salts and esters which generate acidity upon heating, such as amine hydrochlorides and dime~hyl oxalate, respe~kivelyO
Urea-form~ldehyde resins are often formulated with some urea, thiourea or substi~uted ureas to prevent excessive shrinkage and crackillg of the molded parts~

:~ , They also function as plasticizers and help insure uniform flow and curingO
~ r .~ h ~ ~ Lr~ ~
. Melamine is commonly produced by condensation . of dicyandiamide (~Idicy~) at high temperature and moderate pressure in-the presence of ammoni.aO The overall reaction ~; is represented by the following equation: ;~

3 H2N-C(=NH)-NHCN - > 2 H2W ~ NH~

t Dicyandiamide Melamine Melamine, like urea, reacts with ~ormaldehyde under neutral or sligh~ly alkaline conditions ~o produce methylol melamines, The latter may be regarded as the monomers of melamine-formaldehyde resinst The condensation reaction is usually carried out in corrosion-resistant equipmentO Due ~o the low solubility of melamine in wa~er ~ .

2~ 3 and aqueous ~ormaldehyde at ambient ~emperatures, cond~n-sation is usually conduc~ed at temperatures of 90C. or higher. A melamine-formaldehyde ratio of 1:2 to 1:3 is usually employedO Water may be removed by vac~um distillation at low temperatures to keep the resin from . prem~ture gelling in the reactorO
;` Melamine lding compounds are manufactured ~ by processes ~imilar to those described for ureaO These `: resins cure more rapldly than urea resins at the molding 10 temperaturesO They may sometimes be combined with slightly alicaline fillers or with other resins, such as p~enol-; f~nmaldehyde resins which set-up under alkaline condit ions .
The pure resins 3 like the urea-formaldehyde resins, are colorless h~nce, molding compounds can be produced in light shadesO M~lamine resins are usually reinforced wi~h ~ ~cellulose and asbestos materials.
Ihe bl~ndin~ of poly(vinyl alcohol) fibers with the ~ormaldehyde-based resin can be achieved by the ordinary practices in the art~ The resultLng molding material need not contain special ingredientsO However, formulators in the art frequently use additives in thermosetting molding compositions which enXance such properties as mold release, hot rigidity, cure rate and the likeO Though the proper ~eiection of them will provide a mol.ding compound having certain optimum usage, ~heir ~ fonm~ no part of thi~ inven~ion~

2~o ,~

~24 This embodiment illustrates the comparatîve ~:
effects of fiber leng~ch on composite moldability and toughness . for a two ste~ phenol-forrnaldehyde ~ (novolak3 resin rein~orced with poly(vinyl alco~ol), polyester and sisal fibers, respectively.
The novolak re~in used in preparing the composites was an acid catalyzed novolak having a phenol to fo~mal dehyde mole ratio of 104 and a ring and ball softening polnt (ASTM E 2858T~ o:E 88Co The fibers used were: poly(vinyl alcohol) fibers of 6 denlers and 708 tenacity; polyester fibers of 3 deniers and 6 0 3 ~enacity and natural sisal fibers of about 200-400 deniers and having a tenacity of about 403 grams per denlerO The fibers were precision cut nt various nominal length~ to achi.eve fiber samples with narrow leng~h distributionsO
Molding materials were prepared using ~he following base ~ormulation: novolak (50 parts), hexa-methylene tetraamlnR ("hexa") (10 parts), woodflour (20 parts) and reinforcing fibers (20 parts), all parts belng determined by weight of material, The materials were compounded by fluxing a mixture of powdered novolak and woodflour on a 6 x 12 inch two roll mill (front roll 105Co~ back roll 145C~)o After a band was ormed on the front roll, the flbers were gradually added to the ~ill using a roll nip opening of about 1/8 inch ~o minimize fiber attrltion and making frequent cuts in the - 8824 .

6;31 sheet to assure good dlsparsion of the fibersO After a uniform sheet was obtained, hexa was added and the compound was now roll milled with frequent blending cuts .~or one and one half minute to advance the resin and assure good mixing of the hardenerO The sheat was then cut off, cooled and granula~edO
The average fiber length for each compound was n~xt determined from a representat~ve sample by first extracting the phenol--formaldehyde resin with ace~oneO
10. The reinforcing fibers were then separa~ed from the wood-flour filler by solutions having densities intermediate between those of the fibers and ehe wood flour and the distrlbutions of actual fiber lengths were determined microscopicallyO The weight aver,age fiber length for each sample was then calculated as discussed aboveO
The moldability was measured for each molding material sample by the spiral flow te~t previously dis-cussed and the results ~re shown in Figure 1 as a unction of th~ weight average fiber lengthO It is seen that the 20 ~piral flow and thus the moldability of the compounds drop off sharply as the fiber length increases~ These results were relatively insensitive to the particular type of fiber usedO
The impact strength was de~ermined from 1/8"
~hick plaques made by compression molding at 170Co for five minute8 at a mold pressure of 2~000 p~io Test .

28~

8~24 ~ 3 specime"s of 1/8" x 1/2!' x 2'l were machined from the plaques, notched and tested for impact strength as pre~cribed in ASTM method ~D 2560 The Izod impact result~ plott~d as a ~unction of the wei~ht average fiber lengths for the three fibers are shown in Figure 2~ The data show that the impact strength in ~ll cases in¢re~ses with the fiber lengthu However, the poly(vinyl alcohol) fibers show a surprisingly steep increase in impact reinforcement at very small fiber lengths in c~mparison ~o the other ~ibersO
It is believed that this dramatic departure f~om the impact reinforcement behavior of other fibers :, in formaldehyde-based resins i~ due to a unique fusion mechanism whereby the poly(vinyl alcohol) fibers inter-fuse into a three-dimensional ne~work structure during moldingO This thermal fusion is probably assisted by wal~e~ and arm~onia released durmg the curing of the resin, Thus, the exceptional reinforcement observed for very sm311 fiber lengths is ascribed to in-situ forma~ion o~ much longer fibers during molding by fusion explaining the unique insensltivity sf the impact strength to ~he in~tial fiber leng~hO
A comparison of Figures 1 and 2 shows tha~
the discovery of this unique behavior of poly(vinyl alcohol~
fibers in phenol-formaldehyde resins has enabled the simultaneous achievement of high impact strength (~one ft. lb/~n. notched Izod) and mold~bility suitable for ~824 ~ 3 screw injec~on molding ( ~24 inches spiral flow) by uSing PV-OH fibers at moderate loadings ~20 w~O %) and short fiber lengths ( ~ 1/8 inch~O

This embodiment II illustrate~ ~he comparative impact reinfoxcement e~flciency of a variety of organic and inorganic f~bers aæ compared to short poly(vinyl alcohol).
fibers in a phenol-~ormaldehyde resin~
Molding materials were prepared, molded and tes~ed for Izod impact strength a~ described in embodi-ment Ic In addition, the ball drop impact strength was measured by dropping a 174go steel ball on an inverted 2 inch ASTM compressîon ms:)lded cup (ASTM D 731~o The test is started at a height of two inches and continued a~ two inch incremen~s until the cup breaksO The ~mpact strength is then reported as the ball drop heigbt at which failure occurs~
The results are shown in Table l-below:
TABL~ 1 ~L~5 Nominal Notched Fiber FiberIæod Ball Drop Ye~ Denier ~ nohes Poly(vinyl alcohol) 1/25 8 1,4 40 1/5 6 lo 7 42 Polyacrylonitrile 1/4 3 0O4 23 (Acrylic fiber) Vinyl chloride/
Acrylonitrile Copolymer (Modacrylic fiber) 1/4 3 0.3 : 30.

~824 ~4~Z~3 `
TABLE 1: (con~inued~
~e~s~
~: Nominal Notched Fi~er Fiber Izod Ball Drop Fiber T~72~ Denier ft lbs/in inches Polyhexamethylene adipamide 1/20 6 004 19 Regenerated Cellulose 1/4 3 0.5 23 -(Rayon) Pulp Flock - 004 19 Ceramic 1/2 - 0O3 11 Fiber glass, E glass type K 1/8 - 0O6 22 It is seen that the short (1/25") poly(vinyl alcoht)l) fibers have a much higher impact reinforcing efficiency than any of the o~her organic and inorganic fibers in spite of the greater nominal length of the other ~iberæO The surprisingly small decrease in impact ~ .
~: 20 reinforcement which is encou~t~red in going from l/51' to 1/25" fiber length or poly~vinyl alcohol) fibers is again cle~rly apparent from the data in.Table lo It should be not~d that the fiber lengths quoted in this example are original fiber lengths, i~eO the "nGm~nal fiber lengths" (hereinafter abbreviated "nfl") ~or the various fibers before being added to the resinO
Since sti~f and brittle fibers such as cer~mic and glass fibers suffer much more fiber breakage during compounding and molding than do the ~ofter and tougher fibers, the actual fiber leDgths in the final composite wi~l be quite d~fferent ~r~m tho~e quotedO N~vertheless, fiber ~ttrition ~824 ~ 0a~63 is more or less unavoidable in most ~ompounding operations a~ well as in more demanding molding and extrusion operations such as screw injection molding, transfer molding and ram extrusîon. Accordingly, the da~ are representative of what is experienced in actual practice with the various reinforcing fibers~

: Table 2 illustrates the efect of molding ~emperatures on the impact strength of the phenol-form-aldehyde molding materials of embodiment I reinforced wi h poly(vinyl alcohol), polyester and sisal fibers:

: Notched Izod_ImPact S~ren~th (ft. ]hs/in. of notch) PV-OH Fibers Polyester Sisal Moldin~ Temperature ~ . 1/4" nfl 1/4" nfl 140 1 . 3 6 160 1 . 38 0 . 60 Oo 70 180 1036 0.61 0,75 200 1.54 0.33 0O57 220 1, 03 0 . 24 0 o 53 ~; 230 0.41* - -240 - 0.23 0.53 260 - 0O22* 0,50 ~Approaching or exceeding the fiber melting poin~, :
Within the temperature range of about 140 ~o about 200QC., which covers ~he molding temperatures most 32.

- 882~
~ U4~2~3 co~nonly used for ormaIdehyde-based molding compounds, the short poly(vinyl alcohol) fibers provide much more -: impact reinforcement than the much longer polyester and sisal fibers. Above about 220C~" which is the melting poînt of poly(vinyl alcohol) fibers, ~here is a sharp drop ln impact strength due to disintegra~ion of ~che fiber~. The impact ~tren8th of the polyester and sisal rainforced materials decreases gradually with increasing molding temperatures and remain at low le~els throughout the tempera~ure~ studiedO
- ~IV-As poin~ed ou~ above, the unique impact rein-forcement provlded by ~hort PV-OH fibers in formaldehyde-based resins is believed to be due to a .solvent assi~ted ~hermal fuslon of the short PV-O~ fibers wi~h each other into a con~inuous nonwoven, three-dimçnsional fiber net-work during moldingO In order to substantia~e this suggestion, the fotlowing experiments were run: PV-OH
fibers (3 denier, l/8l' nfl) were divided into three por~ion~. The ~irs~ portion was left untreated; the second portion was soaked in water at room ~emperature ~or five mlnutes, filtered off and dried on a filter funnel for five minutes at room temperature; the third portion was pre-treat~d as the second portion except that a one percent aqueous 601ution of ammonium hydroxide was ~sed instead of ~ater to simulate the curing environment o~ a Hexa cured novolak compound. Samples of each flb2r 33, 3L~)9~1Z~;3 portion were then pressed betw~en aluminum foil in a ; hydraul~c p~egs a~ a wide ran~e of temperatures and ~he resulting fiber pressings were inspec~ed ~or fiber~fiber fusion, The results are tabulated in the following Table 3.
TABLE 3 ::
Fiber Pretreatmen~ and LR~el Pressing 1% Aqueous Temperature, C, None H20 NH40H Solution 120 n~ne ~parti~l 140 none ~pa~
160 none ~extensiLve~
180 none ~--~xe~nsive~
200 partial ~----extensi~e ~
220 ~fiber structu~e destroy~d~D

It was found that untreated~ OH fibers do ~ --no~ ~use together at points of fiber-fiber con~act : ~ ulltil near or abo~ 200Co where they-may become weake~ed due to thermally i~duced de~ompo~i~ionsD However, upon 20 add~Ltion of wa'cer or a diLu'ce solution of a~nonium hydroxide, fiber fusion occurred evén a~ the lowest : ~emperature ~ested (120C)o At 220C. or above, sufficient fiber melting occurred in all cases so tha~
the original fiber structure was destroyed, In commercial molding practice, a broad molding temperature range is an important requirement, and, as 3~, 216~
shown above, such a broad molding ~emperature range exists for ormaldehyde-based resin materials of ~his invention, e.q.~ about 140Co ~O abou~ 200Co Bei~g inherently non-fusible, sisal fibers do not exhibit this phe~omenon. The same is true of other `common cellulosic fibers such as co~ton and pulp ~locks.
Although polyester fibers are thermoplastic and undergo mel~ing (the meltlng point for polyethylene terephthalate ~6 about 250C.), they are also partially soluble in : 10 ~ormaldehyde-based resins. A~cordingly, at higher molding temperatures, the fibers lose their integrity through dissolution and/or decomposition by the matrix re in and/or ~y resin additivesO A similar behavior is observed w~th other thermoplastic fibers such as poly-amide5, p~lyacrylonitrile and vi~yl chloride/acrylonitrile copolymer fibers in formaldehyde-based resins.
" -V-It is interesting to note the contrastirlg effect of poly(vi~yl alcohol) and sisal f~bexs on the 20 notched Izod imp~c~ strength of a polyester resin, i.~. -a ~hermosetting resin which is not of the formaldehyde-based type and does not cure by condensation. A polyester resiTI made rom maleic acid and ethylene glycol was co~ibined with PV-O~t and sisal ~iber~ a~ descr~beà in embodiment I, each precision cut (n1) ~co 1/25 inch and :~ 1/4 inch length. The following base formula in parts 35.

.

~ 3 by weight, was used for making the molding materials: the polyester resin (64 parts~, styrene monomer (16 parts), dicumyl peroxide (1.6 parts) and reinforcing fibers ~18 parts). The compounds were'mixed on the two roll mill at conditions similar to those of embodiment I
except that a front roll tempeirature of 93 C. and a back roll temperature of 72C. were used, The materials were then compression molded C160C. for 20 minutes at 2000 psi mold pressure~ and tested for impact strength with ~he following results:

TAB'LE 4 Composite Notched `I'z'od'ImP'a`ct' Stren~'th Re`infbrcing Fiber ft. ~s/in, 1/25" pol~(vinyl alcohol~ 1. 3 1/4" " ~ 2, 9 1/ 4" s isal 2, 3 It is seen that in a polyester resin matrix, ', the impact reinforcement provided by sisal and PV-OH
~' 20 fibers is nearly equîvalent. Also, there is a sharp drop in îm~act strength as the PV-OH fiber length (nfl) -decreases from 1/4" to 1/25" in contrast to the behavior in formaldehyde-based resins. These differences are ascribed to the lack of fusion assisting solvents suc~
as water in the polyester/styrene resin system which ; makes the thermal fusion mechanism of the poly(vinyl alcohol) fibers much less effective'here than in the formaldehyde based resins. It should be noted that 36 ~

~lZ63 ; polyester resins generally give higher impact strength and are easier to reinforce for improved toughness as compared to formaldeh~de-based resins due to differences in the mechanical properties of the two classes of resins, -VI-This embodiment illustrates the effects of s~ort poly~vinyl alcohol~ fibers used as impact reinforce~
ment for a conventional one-step phenol-formalde~yde resin (resole).
The resin used in preparing the following composites was a base catalyzed resole having a phenol to formaldehyde mole ratio of 0.63 and a ring and ball j softening point ~ASTM E 2858T) of 88C. The fibers used were 8 denier poly(vinyl alcohol~ fibers having a tenacity of 8.5 grams/denier. The fibers were precision cut to a nominal fiber length of 1/25". For comparison, wood pulp flo~k having an average length of about 1/25"
was used as an alternate reinforcing fiber.
The following base formula~ in parts by weight, was used in preparing the molding materials: resole resin (50 parts~, chrysotile asbestos floats (23 parts); lubricants, pigments ~7 parts) and reinforcing ibers (20 parts~. The ~ same procedure as described in embodiment I was used in ; compounding the materials except that the total roll mixing time was one and a half minute.
The two materials had the following properties:
TABLE S
PV-OHRulp Flock Spiral Flow, inches 31 30 Notched Izod, ft. lbs/in.1~34 0.42 Ball Drop Impact, inches 42 18 , - ~

~ 2~ 3 The short PV-OH fibers provide efficient impact reinforcement in a one-step phenol-formaldehyde composite~
-VII-Table VI compares the effects of short poly-(vinyl alcohol) fibers and some other fibers as impact reinforcement for a bisphenol-formaldehyde resin:
The resin used in preparing the following co~posites was a base catalyzed bisphenol A-formaldehyde ' resole having a bisphenol A~formaldehyde mole ratio of O.4 and a ring and ball softening temperature of 85C.
The fibers and fillers used are listed below together with the impact properties of the resulting compression molded composites:

~, Co osites Filler System, Wt. % A B C
Woodflour - 23 Pulp flock - 8 Asbestos floats - 30 : 20 PV-OH fibers (1/25" J 6 denier~20 - -:: Notched Izod Impact Strength, ; ft-lbs/in. 1.6 0.35 0.25 :j:
; Ball Drop Impact Strength, in. 40 15 14 The short PV-OH fibers outperform traditional filler reinforcements in a bisphenol A-formaldehyde r~sin composite.
-VIII-Table 7 shows in a comparative manner the . superiority of short poly(vinyl alcohol) fibers as 3~.

` ~824 impact reinforcement for a urea-formaldehyde resin:
The resin used in Table VII was a commercial urea-formaldehyde molding resin called Beetle - Alpha Urea, Type Granule 75 NF, manufactured by American Cyanamid Company. The reinforcing fibers used were 1/25" PV-OH
fibers (6 denier) and 1/10" sisal fibers, respectively.
The following base formulation, in parts by weight, was used for preparing the molding materials: t~e urea-form-aldehyde r~sin (50 parts), woodflour (28 parts~, reinforc-- 10 ing fibers (20 parts) and lu~ricant (2 parts~. The molding materials were prepared by dry blending and intimate mixing in a high speed blender. 1/8" thick specimens were compression molded at 165C. fQr 3 min.
under a pressure of 2,000 psi and tested for impact strength as described above. The composite properties were:

~LE 7 lt25" PV--OH 1/10" Sisal ~` Izod impact strength, ft lbs/ L.6 0.6 in. of notch Ball drop impact, in. to 32 20 break -IX-Table 8 demonstrates the superiority of short poly(vinyl alcohol~ fibers as ~mpact reinforcement for a melamine-formaldeh~de resin:

.
, .

39 ~

The resin used was a commercial melamine-TM
formaldehyde resin called Cymel - Melamine Type Granule HS
manufactured by ~merican Cyanamid Company. The molding materials were of the same composition and were prepared as described in embodime~t VIII except-that the melamine resi~ was substituted for the urea resin.
The composîte properties were:

1/25'' PV-OH 1/10" Sisal Izod impact strength, ft lbslin.
~, notch 1.6 0.8 Ball drop impact, in. 40 32 X

To illustrate the effect of poly(vinyl alcohol~
fiber denier (i.e. fiber diameter~ on the impact strength of a phenol-formaldehyde composite, the following was done:
- A series of molding materials were prepared as described in embodiment VI except that poly(vin~l alcohol~
fibers of various deniers were used. Then test specimens were prepared by screw injection molding as described in embodiment XII infra, and the Izod impact strength were determined (ASTM D 256~ ~ith the following results:

40.

TABLl~ 9 : Nominal Notched Izod Fiber Length Fiber Filament l~ e5_~5~Y~
y~ inches . Denier ~t. lbs/in~
., PV-OH 1/25 1 O~ 9 ` " 1/8 3 1.1 " 1/25 4 1, 2 ~ ~ " 1/25 6 1.7 ". ~ " 1/25 8 1,5 10Pulp ~loc . l/25 _ 005 ~ Although fibers of a wide range of deniers can i be uset, deniers withln about 1/2 to about 100 and, - ~ preferably from abou~ 3 to. about 30 have the best overall ~ combin~tlon of propertias as reinforcing fibers fox screw ::j in~ection and transer moldable formaldehyde-based ~ compo ites-:..

Table 11 characterlz~ the moldabili~y of ~;:. a se~ies of poly(vinyl alcohol) fiber reinforced phenol~
~, .
~; 20 ormaldehyde (novolak) composi~e$ having vary~ng ~iber : loading~ and fibex lengthc in terms of the 6piral flow tes~, Using the proced~re~, the two-step novolak resin and the poly(vinyl aloohol~ fibers described in :~
embodiment I, the following molding materials were pre-pared:

~824 ~041263 TABLl: 10 Novo lak 43 43 43 43 43 Hexa 8 8 8 8 8 ~igments, lubrlçants 6 ~ 6 6 6 Pulp f ~oc lû
A~bestos ~oats 3~ 30 ~ 13 3 ~l~y ~ 3 3 - - -Poly(vinyl alcohol)fib~rs ~ 10 20 30 6,0 ., ~: :10 ~ormulation ~o. 1 2 3 4 5 The re6ul~s o~ spiral 10w and impact testing ~e ~l~own in T~bl~ 11 below:

Nom~nal Notchad Ball- Inject-F~ Fiber Fiber Izod Drop Splral ion 2t~0n ~o. Lqngth Loadin~ Impact Impact flow Mold-. . ¢~ c~s _~ ~t~b,~ lnches inches ~ ~:
: ~ : JL Con~rol - 0,4 18 30 Excelle~t 2 1/2$ 10 0. 8 30 32 Excell~nt 20 ~ : 3 ~ 1/25 20 lo 5 40 33 Excellent

4 1/25 30 1 o 7 48 31~ ~xcellent ;~
. ~
1/2$ 40 .2.1 55 25 Falr, 2 1/5 1~ 00 9 32 24 Fair 3 1/5 20 1.7 42 14 Very Poor ^
~:The da~a in Table 11 show 1:he de~irability o:~
using short fiber~; especially at high iber loadings, 1;o aohieve high impac~ oharac~eristics. Tha ~niq~e ability or~ poly(vinyl alcohol) fibers to provide high impact ~O , ~.
. ~ ~

~ Z 6 3 reinforcement a~ fiber leng~ps sufficiently short to pro~ide good moldability in a phenol-form~ldehyde resln is clearly apparent.

. The following illustrates ~he ac~ual screw injection moldlng performance of the poly(viny1 alcohol)_phenol-~ormaldehyde composition~ characterized by the spiral flow test.
~: Moldi~g compound~ containing 20 wt. % of 1/25'1 . 10 and 1/5" poly(vinyl alcohol) ~ibers as well as a control formulation without PV-OH ~ibers (Table 10) were prepared as described earlier and in~ectio~ m~lded on a Model 175 New ~ritain Screw Injec~ion M~ldin.g Machine having the specificatio~ shown in Table 12 below:
~ABLE 12 ~ Screw diamO ~ inO 2.16 - ~ozzle diameter, ln. 5/32 : ~cr~w, speed, rpm 27-145 :~ 20 Iner~mant~ of speed~ rpm 5 Injection stroke, in. 5 Hydraulic lihe pres~ure, psi 1250 Injection pre~sure, psi 19,300 (max,) The in~ection molding machine was equipped wlth an ASTM te~t specimen mold having the followin~ s.pecifi-cation~: .

43.

~82 ; :~L04;~ 63 TABIE 2.~
ASTM TEST SPECIMI~N MQLD
Uold:
Fi~re-cavity ASTM specimen^ mold consisting of or~e disc 1/8 x 4 in.; one disc, 1/8 x 2 in.; one bar, 1/4 x 1/2 x 5 in~,`one bar, 1/2 x 1/2 x 5 inO; and one tensile bar 1/8 x 3/4 x 8-1/2 lllu Length, 4 inc; diam~ter, 7/16 inO tapering to 3/16 in~
Runnere: Recltangular~ 1/2 x 1/8 inO tapered to gates~
Shot Weigh~: 138 g, gerleral-purpose phenolic~

T~e compoullds were injection n~lded at the followiTlg operating conditions:
TABIE XIV
Mo:Ld temperature, C. 172 Bar~el tempe~Cature, CO98 Injsction pre8sure, psi1250 Gure ~ime, seconds 90 ~ OveraJ.l Cycle, ~econds97 The ~ompou~d~ with snd without ~0 Wto % of 1/25" :~
20 poly(vinyl.alcohol) fib~rs gav~ excellent mold f~ out and the p~xts had high surface glos$ and smoothneqs.
In contraqt, ~che compound containing 20 wt. v/~ of l/
f~ber~ ga.ve ~ncomplete mold fill out and w~s judged unsuitable for general injection moldirlg appllcatiorls... ~
The results of physical tests of the molded parts are ~:
~hown in Table 15:

~'~0 ' ' 8824 , .

26;~

PY-OH fibers (1/25"), wt. ,V/~ 0 20 "
-` Notched Izod Im~act, ft-lbs/in. 0.4 1.7 Flexural Streng~h, psi 15,000 12,000 Flexural Modulus, psi 1,500,0001,270,000 Heat Distortion Temp., C, , (264 psi) 192 189 more than four fold i~prove~ent in impact strength has thus been achieved with the use of 20 w~. % of ~/25"
poly(vinyl alcohol? fibers while at the same ~ime retaining or exceeding the injection moldability and appearance of a m~re conventional type of formulation, Also, the moldability evaluations by t~e spiral flow test were substantiated in these experiments.

-XIII-, . ' .
The following illustrates the transfer moldin~
performance of the polyCvinyl alco~ol~/phenol-formaldehyde ~- compositions characterized by the spiral flow test.
,", The molding compositions discussed in XI above, containing 20 and 40 wt. % of 1/25" and 20 wt. % of 1/5" poly~vinyl alcohol) fibers were plunger molded on a 100 ton Hydro Power Machine using-the ASTM specimen mold described in Table 13. The following operating conditions were used:

TABLE '16 Preform tem,perature, C. 122 Mold temperature, C. 166 Mold pressure, psi 800-1200 psi Cure time, seconds 90 45 ~

88~4 ~)4~LZ~i3 The c~mpounds containing ~0 and 40 wto % of the lt25" fibers both ~howed good mold fill-out and parts with excellent surface appearaTlce could be made wi~h relat-lve ea e~ In contrast, ~he compound containing 20 ~. % of the l/5" fibers gave incomplete fill-out of th~ mold and was judged unsllitable for gen~ral transfer molding appliea~ionsO
The molded specimen~ were tested fo~ physical proper~ies with ~he following rasul~s:
~ABI.E 1.7 OH ~ibers (l/25") 3 wto % 20 40 No~c~ed Izod I~pact, :Et:-lbs/inO106 2~5 Flexural Strength, psi ll, 00û -Flexural Modulus, psi 1,340,000 - ;~
Heat l)isto~tlon Temperature, CO
(26~ p3~ 9 - It 18 seen that ~ood mo1dabiliey in the transfer molding process can ~e achieved wi~h high impact phenol-form-, ' : , , a1dehyde resins compo~i~e~ usin~ short poly(vinyl alcohol) 20 fibersO A}so~ it i8 seen ehat transfer moldlng i8 ~ome-what les~ demand~ng in flow charac~erist~cs than i9 the screw injectio~ molding process and the moldability predica~ions by the spiral ~low data are verified by the8e ~xperimentsO
A~ shown below, ~hort poly(vinyl alcohol) fibers can be used effPctively for boosting the impa~t str~ngth of formulations containing other reinforcing fibers such 4~ O

as glass fibers or sisal fibers for applications where the handling and appearance problems-with these fibers can be tolerated. Also, such other reinforcing fibers can be used to provide desirable combinations of physical :
properties beyond those attainable with poly(vinyl alcohol) -~ fibers alone.
- Table 18 showq the impact properties of a 20 wt. % fiber reinforced phenolic molding material using blends of 1/25" poly(vinyl alcohol) and 1/10"
sisal fibers~

PV-OH fibers, wt. % O 5 10 20 Sisal fibers, wt. % 20 15 10 0 , Notched Izod Impact Strength (ft-lbs/in.) 0.6 0.9 1.2 1,8 Ball Drop Impact Strength 28 34 38 42 (inches) Table 19 shows the physical properties of a 20 wt. ~/~ fiber reinforced phenolic molding material using blends of 1/25" poly(vinyl alcohol) and 1/8"
E-glass fibers (Type K~:
~ABLE``l9 , PV-OH fibers, wt. % 0 5 10 15 20 Fiberglass, wt. % 20 15 10 5 0 Specific Gravity 1.641.59 1.54 1.48 1.46 Ball Drop Impact Strength Inches 22.5 27 32.5 38 42 I Notched Izod Impact Strength (ft-lbs/in.) 0.6 0.9 1.2 1.4 1.5 Tensile Strength, psi 7,5706,9606,490 6,060 6,490 Flexural Strength, psi 16,800 13,800 13,300 11,100 11,100 Modulus of Elasticity, pBi X 10-6 1.901.671.601.421.34 ,Z~
A rather surprising and useful behavior of poly(vinyl alcohol) fiber reinforced phenol-formaldehyde molding compositions is markedly improved moisture resistance compared with compounds without these fibers. Table 20 shows the results of water absorption tests (ASTM D 570) on a woodflour-illed and a mineral-filled 2-step phenol~
orma1dehyde molding compound with and without poly(vinyl alcohol) fiber reinorcement.
~, ;~ TABLE 2 0 Poly(vinyl ~` Formulation Filler Type alcohol~ Fiber Water Absorption No. (% by Weight) (% by Weight) Gain in Wt., %
(a.~ Woodflour (38%) none 0.60 (b.) Woodflour (18~Jo~ 20 0.14 (c.) Asbestos (31%) ~;` Pulp flock (9%) none 0.14 (d.~ Asbestos (23.5%) 20 0.07 A significant improvement in moisture resistance ;~ is evident in each of the two molding formulations upon : 20 additi~on of poly (vinyl alcohol~ fibers. Becau~e poly (vinyl alcohol~ resins, not fibers, are known to be rather water absorbent, their presence impairs rather than improves the moisture resistance of phenol-fonmaldehyde molding compounds. For example, incorporation of one percent by weight of a poly(vinyl alcohol) resin to a phenol-formaldehyde molding compound increases its water abæorption by 35% in contrast to the marked impxovement shown in Table 20~
: '' ~ .

"

48.

,-i3 Consequently, the improved moisture resistance in the present system is ascribed to good fiber/ma~rix adhesion which prevents capillary migration of water through the fiber/matrix interstices of t~e molding compound. This behavior could have significant effect on the retention of wet electrical and mechanical properties of poly(vinyl alcohol) fiber reinforced materials.

~ .
~ .

-' ,, , ~, ~,'' ` .
, :~';. .
( ' 49,

Claims (11)

1. The method of making thermosetting resin molded articles having improved impact strength which com-prises providing a thermosetting formaldehyde-based molding material comprising a mixture containing a thermosetting for maldehyde based resin from the group consisting of a phenol-formaldehyde resin, a bisphenol-formaldehyde resin, a urea-formaldehyde resin, a melamine-formaldehyde resin, and mixtures of them and fusible, water insoluble poly(vinyl alcohol) fibers, said fibers being present in the mixture in an amount of from about 5 to about 50 weight per cent, based on the weight of the mixture, and having a weight average fiber length of from about 1/100 inch to about 1/8 inch, which material has a Spiral Flow of at least 18 inches, and injection or transfer molding said molding material at a temperature of 120°C. to 200°C., causing said fibers to interfuse, and producing a molded article having an Izod strength greater than about 1 foot pound per inch of notch.

2. The method of claim 1 wherein the formaldehyde-based resin is a phenol-formaldehyde resin.

3. The method of claim 2 wherein the resin is a novolak.

4. The method of claim 2 wherein the resin is a resole.

5. The method of claim 1 wherein the resin is a bisphenol-formaldehyde resin.

6. The method of claim 5 wherein the resin is a bisphenol A-formaldehyde resin.

7. The method of claim 6 wherein the resin is alkaline catalyzed.

8. The method of claim 1 wherein the resin is an urea-formaldehyde resin.

9. The method of claim 1 wherein the resin is a melamine-formaldehyde resin.

10. The method of claim 1 wherein the extrusion is effected under heat and pressure.

11. The method of claim 10 wherein the extrusion is effected by action of a ram.