Etching Pet

8/8/2019 Etching Pet

1/15

JOURNAL OF POLYMER SCIENCE VOL. 57 , PAGES 993-1007 (1962)

Etching of Polyethylene TerephthalateWILLIAM P. BAKER, JR., Film Department, Experimental StationLaboratmy, E . I . du Pont de Nemours and Company, Inc., Wilmington,

Delaware

INTRODUCTIONThe molecular structure (0.1 p ) has not been adequately explored, al-though some work with filmshas been reported recently in the literature byNielsenl and by Kodov2and co-workers.

This paper is largely concerned with the demonstration that organizedvolumes of three different orders of magnitude (all greater than molecular)exist in polyethylene terephthalate films and that at least one of these isprobably general for all polymeric films subjected to anisotropic stressduring processing. Polyethylene terephthalate was chosen as an arche-type because of the number of differently processed films available forexamination and because its molecular structure was suitable for etchingby chemical degradation.

EXPERIMENTALPolyethylene terephthalate film samples are etched by n-propylamine

at convenient rates at room temperature by chemical degradation intoethylene glycol and N,N'-n-propyl terephthalamide. Infrared studies ofthe reaction products confmned that the degradation was uncomplicatedby side reactions. Since both the glycol and amide are soluble in excessamine, the products are removed from the reaction site. It isprobable thatthe rate of etching can be controlled over a wide range by variation of thesteric and basic properties of the amine.An everpresent concern in etching work is that the developed structuresbe characteristic of the substrate itself and not of the substrate-etchingreagent system. Sufficient film samples fabricated by different processeswere examined to afford assurance that our data are representative of thefilm (therefore, its processing) and not of spurious effects. This is partic-ularly important in this case, because n-propylamine does cause someincrease in the crystallinity of heat-set polyethylene terephthalate filmsand a large increase in the crystallinity in amorphous films of like composi-tion. Although this occurrence proved of no importance in this work,

993


2/15

994 W. P. BAKER, JR .it may be significant in other cases in which the rate of relaxation of struc-ture is much greater than the rate of degradation.

A l-in. by 2-in. samplewas suspended without tension in 100 ml. n-propylamine (purification ofEastman Yellow Label product was found unnecessary) in a closed con-tainer at room temperature. After a desired time interval, the samplewas removed, rinsed in fresh propylamine, rinsed thoroughly in water,and dried. Control experiments demonstrated that the etched patternswere not affected by the clean-up procedure or sample size. The etchedsample was then examined by optical microscopy and by electron micros-copy. Optical microscopes employed were: American Optical Model1583 phase microscope, American Optical Model 337047 polarizing micro-scope, and Reichert research polarizing microscope Zetopan-Pol. Elec-tron micrographs were obtained with an RCA electron microscope afterstandard replication and shadowing techniques.

The majority of film samples described in this work were stretched on aspecially designed apparatus which applied the stress homogeneously. Theuniform application of stress is important because without i t only generaltendencies may be observed and not the regularity of structures reportedin this paper. Unless otherwise stated, the base film was amorphous( p = 1.33), unoriented polyethylene terephthalate (v i = 0.55 f 0.03;3OoC., 0.5% in CF3COOH). This base film etched into random patterns,roughly similar to those of sun-dried mud.

A typical etching experiment is detailed below.

RESULTSA. One-way Stretched Film

All one-way stretched films were found to have rather evenly spacedlong (centimeters), parallel lines of uniform width (1-7 microns) etched intoboth surfaces in the direction of stretch. In general, the lines on one surfacewere not coincident with those on the opposite surface. Qualitatively, iin-der the same stretching conditions (temperature, rate of stretch, time of resi-dence, type of restrain), the number of lines per uni t of width varied di-rectly with the stretch ratio, while the line widths concomitantly variedinversely. A quantitative correlation of the number of lines with stretchratio was not feasible because the stretching apparatus was limited to a 4Xstretch and structural regularity sufficient for analysis did not occur until a21/2x ratio had been reached. It did appear, however, that over the nar-row stretch ratios suitable for analysis a definite correlation existed betweennumber of lines and stretch ratio (and, by inference, orientation). A typi-cal etched sample is shown in Figure 1, the film having been stretched 4Xwith lateral restraint.

As the time of etching was allowed to increase, the films exhibited bothdelamination and fibrillation characteristics, being degraded into long rib-bons of uniform width composed of layers of micron-sized thickness.


3/15

ETCHING OF TEREPHTHALATE 995

Fig. 1 . Etched, nonheabset film, stretched one way 4X with lateral restraint.

Fig. 2. Etched, heat-aet film, stretched one way 4X with lateral restraint.

The number of lines (see Fig. 2) was increased greatly by heat settingat 185OC. for 15h.,ncreasing by a factor of 3-10 depending upon theparticular sample. Stretching without lateral restraint produced morelines than stretching with lateral restraint under the same conditions andstretch ratio. Heat setting a narrow strip of film with two-way restraintcaused the etched lines to change from parallel to the stretch direction(long dimension) to perpendicular to the stretch direction.


4/15

996 w. P.BAKER, JR.B. Heavy-Gage, Two-way Stretched Polyethylene Terephthalate

(3 mil and Greater)Film samples of polyethylene terephthalate of various thicknesses pre-

pared by stretching 3.5X in one direction followed by 3.5X in a perpendic-ular direction then were heat-set a t 185OC. These samples were found toetch into long, parallel lines, generally uniform in width and spacing. Thelines were parallel to the first stretch direction and were connected byperpendicular lines which did not cross them; i.e., the etched surface re-sembled a brick wall rather than a grid. The unbroken lines were cen-timeters long and 3-7 microns wide, the ultimate width depending moreon the individual sample than the etching time. The brick size averaged100 microns by 150 microns, the long dimension lying in the first stretchdirection. Since experiments with one-way stretched film had demon-strated that the long line direction coincided with stretch direction, itwas expected that the edges would behave differently due t o the influenceof the stress patterns around the clamps. In the edge area the long linesdid not run in the direction of stretch, but instead ran a t angles up to 45'to the direction of stretch.It was found that the two-way stretched polyethylene terephthalatesamples were composed of laminae parallel to the film plane, each resem-bling a brick wall when viewed after etching. The subsurface lamina wascomposed of smaller bricks than the surface lamina, usually by a factor of

The cross-sectionalarea of interior bricks varied from the same as that of the penultimatelayer to a graduated decreasing size as the distance from the film centerdecreased. The number of laminae varied with the film thickness and thefilm forming processes, sometimes being as great as 25. The brick thick-

to 3 / 4 , thus indicating a more oriented material.

Pig .3. Etche d, heat-set film, two-way stretched 3.5X by 3.5X surface view.


5/15

ETCHING OF TEREPHTHALATE 991

Fig. 4. Etched, heatrset film, two-way stretched 3.5X by 3.5X edge view.

Fig. 5. Etch pattern in interior of heat-set film, wo-way stretched 3.5X by 3 . 5 X .ness varied with position of the lamina, but an average thickness was ofthe order of 12microns. A normal surface after etching is shown in Figure3, and a typical edge view is shown in Figure 4.As the etching of a film sample progressed, the volumes surrounding the

bricks were etched deeper and deeper into the film until the second laminawas reached. The material between the two laminae was then eatenaway, so that the surface bricks became detached, fell off, and exposed thesecond lamina. At this time, the surface bricks were still rectangular and,thus; much more resistant to basic attack than the mortar. The bricks


6/15

998 W. P. BAKER, JR.were highly crystalline as judged by density (1.42g./cc.) and by x-raypowder photographs, but were not close to a single crystal classification.

When etching of material between the first and second layer of bricksoccurred, a unique type of structure was commonly found. It consistedof long, parallel ribbons of roughly constant width and spacing. Theseribbons ran in the first stretch direction and possessed either a crest ortrough running along the middle. The resemblance to metallurgical slipplanes was striking. This typeof structure was never found to occur singly-either quite a few were pres-ent or none at all.

A typical example is shown in Figure 5 .

Fig.6. Etch pattern in interior of heat-set film, two-way stretched 3.5X by 3.5X.As etching progressed into the second lamina, either ovals or spirals were

eaten into the bottom of the bricks of that lamina, Examination of thespiral structure by the optical microscope and by the x-ray microscope(kindly done by Dr. P. H. Geil of the Polychemicals Department, Experi-mental Station, of the du Pont C o . ) suggested the spiral was caused by aflow pattern occurring during extrusion and not by a unique molecular or-dering. The same conclusion was reached concerning the origin of theovals.

The above is the general etching pattern for heavy gage, two-waystretched, heat-set polyethylene terephthalate films. Although the pre-dominant orientation of both surfaces was in the first stretch direction, largeareas of the surface beyond the clamp influence were oriented at angles aslarge as 45' from this direction. Since the orientation of each surface was

A typical spiral is shown in Figure 6.


7/15

ETCHING O F TEREPIITHALATE 999independent of the other, the orientation between the surfaces differed byas much as 90' in some regions. All surfaces contained faults of two types.One type resembled a geological fault, where the bricks had slipped by eachother along a line roughly perpendicular to the first stretch direction. Theother fault type consisted of an abrupt change in orientation wherein1-20 brick rows running at some angle such as 45' was inserted into thenormal arrangement. Occasionally the brick size differed on either sideof the fault. These faults were usually only one layer deep. Thus, ori-entation differences exist both between surfaces and on each surface.

To ascertain that the laminar structure existed before etching, a filmsample was placed on edge and a microscope was focused on the top edge.The sample was illuminated below the top edge by light incident on thefilm surface. Reflected light travelling through the film clearly showedthe laminae on the top edge. The sample was cut in half, and one halfwas etched into the usual laminar pattern and replaced in its former posi-tion. The etched laminae coincided with those of the unetched samplehalf.

C. Light-Gage, Two-way Stretched Polyethylene Terephthalate(1 mil or Less)Generally, the light-gage films degraded in a manner undiscernable under

the light microscope, and the degradation of sample was followed by weightloss. Occasionally, a fuzzy brick pattern seemed to be present, but i tcould not be resolved. A few 1-mil samples did show a brick pattern ofsmall dimensions (10 by 15 p) , but these were infrequent. It was tenta-tively concluded that these thin-gage films did contain the bricklike struc-ture of heavy-gage film, but that this structure was of smaller dimensionsbecause of higher orientation and a thickness effect.

Etching of intermediate gages such as two mils brought forth the usualbrick structure, but the disparity of brick size was large, a row of 1 by 2 pbricks running adjacent to a row of 100 by 50p bricks.

D. Solvent-CrystallizedFilmsChloroform is known t o be a solvent for amorphous polyethylene tereph-

thalate, although an unsuitable one, because it crystallizes the polymerunder most experimental conditions. When placed in contact with a par-tially crystallized film of two-way stretched, heat-set polyethylene tereph-thalate, no dissolution was found to occur, and the per cent crystallinityincreased. Etching studies were made of chloroform-crystallized samplesto determine whether such crystallization caused an alteration in the struc-ture. It was found that no gross structural changes as detected by etchingpatterns were caused by this type of crystallization. The brick wall pat-terns remained after the treatment, but were reduced in size by aboutone third.


8/15

lo00 W. P. BAKER, JR.E. Etching of Mechanically Damaged Films

Two-way stretched, heat-set polyethylene terephthalate films were slowlycreased through 180 and allowed to relax for several hours before place-ment in propylamine. The samples were completely eaten through a t thecrease before any noticeable etching had occurred in other parts of the film.The etching in the vicinity of the crease consisted of long lines, roughlyparallel to the crease direction, regardless of the spatial relationship be-tween crease and stretch direction.

Two-way stretched, heat-set polyethylene terephthalate films were tornand etched and were etched and torn; in both cases the torn regions wereexamined under the light and electron microscopes. Extensive damagedone in the immediate vicinity of the tear prevented firm conclusions.These exminations did indicate that the tears occurred along the lines be-tween bricks and received their steplike character from splittings betweenlaminae, i.e., the tears appeared to follow along volumes containing themost rapidly degraded polymer. This suggested that a breakup of theregular brick wall structure should increase tear strength. A piece of filmwas simultaneously stretched 3X by 3X, but the stretching directionswere changed by 45 every 33% of ultimate stretch. Etching of the result-ant film after heat setting showed highly oriented local areas, the orienta-tion directions of the various areas being random with respect to each other.This film possessed a higher tear strength than the usual heat-set filmstretched 3X by 3X simultaneously.The difference between surface and interior patterns suggested that thesurface lamina had some independence from the interior laminae. Tocheck this, various designs (circles, squares, parallel lines) were cut intounstretched, amorphous, polyethylene terephthalate films to a depthgreater than one lamina. The cut films were one- and two-way stretchedand then etched. The patterns traversed the cuts with no shift of orien-tation. In this case, the patterns were apparently templated from theinterior before the -formationof laminae, a situation, perhaps, not true ofall stretching processes.

F. Rates of EtchingThe rates of etching of various films were followed by weight loss de-

terminations and periodic viscometric measurements. Since the reactionwas heterogeneous and the character of the film surfaces differed, the mini-mal data collected were not considered worthy of extensive interpretation.In general, the rates depended upon the amount of stretch, the density,and the thickness. Initial rates fell into three groups: 4%/hr., 3.2%/hr.,and 1.6%/hr. The highest rates were characteristic of film of low stretchratio. Since the rates were dependent upon a t least three variables, thethree rather narrow groupings were probably the result of sampling arti-facts rather than distinct film types. Often, long induction periods wereobserved.


9/15

ETCHING OF TEREPHTHALATE 1001G. Etching of Polystyrene

To establish that crystallinity was not essential for the structures ob-served by etching in polyethylene terephthalate films, amorphous atacticpolystyrene film, which had been stretched about 1QX by 6X, was ex-posed to a poor polystyrene solvent. Viewing of the film so treated undera microscope revealed a surface pattern similar to that etched into one-waystretched polyethylene terephthalate films.

H. Dyeing and Stress-Crackingof Polyethylene TerephlhalateAttempts to stress crack or solvent crack the etch pattern into two-vay

stretched, heat-set films of polyethylene terephthalate were unsuccessful.To achieve such cracking, it was necessary to apply a stress of greater mag-nitude than that residual in the film; as a result, the crack patterns ob-tained were those characteristic of the applied stress.

Cursory attempts to dye stretched heat-set film of polyethylene tereph-thalate without concomitant crystallization failed. It was thought khatthe volumes which were preferentially attacked chemically might also dyepreferentially.

I. Examination of Etched, Two-way Stretched, Heat-SetPolyethylene TerephthalateFilm By Electron MicroscopyEtched samples bf polyethylene terephthalate film which had been

stretched 3.5X by 3.5X and heat set a t 185OC. were replicated and elec-

Fig. 7. Electron micrograph of etched, hea tset film,two-way tretched 3.5X by 3.5X.


10/15

1002 W. P. BAKER, JR .

Fig. 8. Electron micrograph of etched, heat-set film,wo-way stretched 3.5X by 3.5X.tron micrographed by standard techniques. The replications demon-strated that the first layer bricks were composed of spheres about 500 A.in diameter. Uneven etching made it difficult to determine the arrange-ment of these spheres, but occasionally they were seen to exist as rowsaligned in the stretch direction. The material beneath the first layer wascomposed of somewhat larger spheres (700-1000 A. diameter). The sur-face of the first layer of bricks was quite smooth, except for hexagonlikeemergents (presumably exuded cyclic trimer3) and occasional randomlyswirled areas of unknown origin. Typical electron micrographs of etchedfilm are shown in Figures 7 and 8.

DISCUSSION AND CONCLUSIONSThis study o the gross structures of anisotropically stretched films is

in too embryonic a stage to permit an origin and meaning of the structuresto be assigned with cd d e n c e . The mechanism of the appearance of thesestructures and the relationship of these structures t o organization in thefilms are not clearly defined. The extent to which the etching reagent maybe enhancing organizations existing in the film and the relationship of thisphenomena to the craze cracking described by Nielsen' are not known.However, it is often advantageous to advance a hypothesis based on avail-able data for use as a point of departure. In the following paragraphs sucha hypothesis of the structure of anisotropically stretched films of poly-ethylene terephthalate will be proposed. Much more extensive and variedinvestigation will be required to establish the degree of correctness of thepicture.

There is a large regular structure of micron size in the anisotropicallystretched films; this structure is built into the film duriag the fabrication


11/15

ETCHING OF TEREPHTHALATE 1003processes and represents the distribution of strain caused by the fabricatingstresses. The micron-sized structures are built of spheres, 500-1000 A.in diameter. Spatially ordered over a long period, the spheres are com-posed of highly ordered molecular units and may move as rheologicalunits. Theetching studies of various films showed that the etched patterns appearedto be a measure of the direction and extent of orientation. That the poly-mer in these patterns was attacked preferentially suggested that it existedat a higher energy level than adjacent material. This higher energy con-tent could be the result of higher strain because the film tended to distributethe easily degradable structures uniformly at all opportunities just as itmight be expected to distribute strain uniformly if given opportunity.Thus, although the total easily degraded mass remained about the same,the number of lines per unit length of one-way stretched film increasedfrom film stretched with restraint to film stretched without restraint.Similarly, the subsequent heat-setting step or crystallization by chloro-form caused an increase in the number of lines.

To demonstrate that the more readily degradable volumes were actuallyof higher energy content and inherent in the film, we attempted to stress-crack film along the lines of strain already in the film. As reported, thisapproach failed. However, the tear characteristics as revealed by etchingof two-way stretched, heat-set polyethylene terephthalate films are inaccord with this concept. It may be that these volumes are strained andhighly organized. Thus, although the films are initially hard to tear,once the tear is started, the tear strength is considerably lower. Thismight be the case if highly organized volumes of higher energy contentthan their environment were disrupted by the started tear and becamedisorganized volumes of excess energy.

Heat of solution experiments were employed to measure the relativeenthalpy of stretched films. This work, which will be described elsewhere,showed that the enthalpy of a 4x by 4x sequentially stretched film (high-est ratio available) was 5.8 cal./g. higher than that for unstretched filmof the same crystallinity. It is tempting to assign a large share of this tothe etched volumes.

The spherical building blocks were readily discernible from electron mi-crographs. Uneven etching allows little to be said concerning their ar-rangement except that frequently they were observed to lie in a row in thestretch direction. The camposition and rheolagical identity are inferredfrom the studies of Heffelfinger.' He found by x-ray techniques that g o u pof crystallites probably moved together as units and studied the molecularorganization within these groups.Ultimately, the significance of structures will reside in their ability tocontrol various properties of film. Reakoning somewhat intuitively, onemay infer that the ordered close-packed constituents of the spheres andthe rheological identity of the spheres make the spheres the strongest

The experimental evidence for the above will now be considered.


12/15

1004 W. . BAKER, JR.II 'LODULUS

I I I


13/15

ETCHING O F TEREPHTHALATE 1OQ5voids, having high permeability for the attacking reagent and numerousdegrees of freedom. Since permeability of these. films does not changemuch with orientation and rapid redistribution occurs whenever possible,the alternative case is weakened.N o property dependence upon the sphere arrangement was determinedbecause of sample limitations. However, it would seem odd if the sizeand packing of the spherical organizations influenced properties onlynegligibly. The origin of these organizations is unknown, but their sizeis compatible with the size of rheological units, which have been postdatedto exist in polymer melts.

The origin of the laminar structure found in stretched polyethyleneterephthalate films is unknown. It is possible that it is generated byanisotropic heat transfer during fabrication by the anisotropic stfucturesput into the film.It is thought tha t a t least the micron-sized structures described herein aregeneral to all anisotropically stressed films. Work by Dr. R. M. Ikeda ofthis laboratory has indicated the existence of these structures in polyvinylchloride and cellophane, and these structures have been shown to be presentin polyvinyl fluoride.6

In conclusion, it should be stressed that to see the regularity of thesestructures a most carefully designed stretching apparatus is necessarybecause the applied stress must be uniform.

It is a pleasure to acknowledge the many helpful discussions throughout the programfrom the following people: Dr. L. F. Beste, Pioneering Research Laboratory; ProfessorGeorge S. Hammond, California Institute of Technology; Dr. C. J. Heffelfbger, MylarReaearch and Development Laboratory, du Pont Company, Circleville, Ohio; Dr. S. S.Tor, Engineering Research Laboratory, du Pont Company; Dr. W. J. Pangbnis, Dr.F. P. Gay, and Dr. R. D. Pruett of this laboratory. The author is indebted toMr. . F.Godlewski of the Central Research Department, du Pont Company, for replications andelectron micrographs.

References1. Nielson, L.E., . A p p l . Polymer Sci., 1,24(1959).2. Berestneva, G. L., and P. V. Kozlov, Vpokomolelrulyarnye Soedineniya, 2, 18543. Giuffria, R., J . Polymer Sci., 51,427 (1961).4. HeffeEnger, C. J.,J . Polymer Sci.,47,304 (1961).5. Osborn,R.O., Film Department, Yerkes Research Laboratory, du Pont Company,

(1960).

Buffalo, N. Y., unpublished work.Synopsis

A degradative etching technique has been employed with polyethylene terephthalatefilms to demonstrate the existence of several different structures, the size of which varyfrom hundreds of microns to hundreds of Angstroms. The micron-sized structures me afunction of the processing and orientation of the film and occur in regular geometricpatterns. Thus, parallel stretch-direction lines are etched into one-way stretched film,the number of lines being dependent upon film history. Etched, two-way stretched6lms reveal a lamellate composition, each layer possessing a brick wall-like structure.This general structural pattern has been found in other types of film by similar meth-


14/15

1006 W. P. BAKER, JR.ods. Examination of the etched areas by electron microscopy demonstrates th at themicron-sized blocks are composed of spheres, 700 A. in diameter, which are alignedover moderate distances.

RbW 6Une technique de gravure par dbgradation a C tC appliquCe8. des films de tCrCphtalatede polyethylene pour prouver lexistence de plusieura structures diffCrentea dont lagrandeur varie de centainea de microns 8. des centainea dAngstroms. Lee structures dedimensions en microns sont en fonction de la preparation e t de lorientation du film etsont presentes sous des formes g6om6triquea rCguli8res. Des lignea de direction dCtire-ment parallele sont gravees dans des films CtirCs dans une seule direction. Le nombre delignes depend du traitement du film. Dea films CtirCs dans deux directions montrent unecomposition lamellaire dont chaque couche a une structure rCticulCe. Cette structuregCn6rale a C t k retrouv6e dans dautres films. Lexamen des surfaces graveea par micro-scopie Clectronique montre que les blocs de plusieurs microns sont composCs de spheresde 700 A. alignCes sur des distances modCrCes.

ZusammenfassungEin Atzverfahren durch Abbau wurde an Polyathylenterephthalatfilmen zum Nach-

weis dea Bestehens mehrerer, verschiedener Strukturen, deren Groeae im Bereich voneinigen hundert Mikron bis einigen hundert Angstrom lei& angewendet. Die Struk-turen im Mikron-Grossenordnungsbereich entstehen bei der Verarbeitung und Orientier-ung des Films und treten in regelmiissigen, geometrischen Anordnungen auf. So lassensich in einseitig gereckte Filme parallele Linien in der Reckungarichtung atzen, wobei ihreAnzahl von der Vorgeschichte des Films abhangt. Zweiseitig gereckte Filme weisen beider Atzung einen lamellaren Aufbau auf, wobei jede Schichte eine ziegelmauerartigeStruktur beaitzt. Dieses dgemeine Strukturbi ld wurde auch bei anderen Filmtypenmit ahnlichen Methoden festgeatellt. Eine elektronenmikroskopiche Untersuchung derAtzflachen zeigt, d m die Blocke im Mikronbereich aus Kugeln mit einem Durchmesservon 700 A. zusammengesetzt sind. welche iiber miiasige Entfernqngen angeordnet sind.

DiscussionB. Rosen (Westinghouse Ele-tric Corporation, Pittsbirrph, Pa.) I have a question

concerning your implication in using the term etching. Etching usually has theconnotation of a lorally selective reaction or solution. Could it be th at you are reallyconcerned with a fracture? I dont understand the role of a uniform stress in a selectiveetching. Could your material be anisotropic and have points or areas which are goodstress concentrators? If this sample were first fractured, subsequent reaction might bepreferentially located a t these new surfaces. Have you eliminated the possibilities ofenvironmental stress cracking, whether the stress were applied or internally locallyresidual? Or is this a solvent cracking, by which is meant the introduction of transientstresses due to a diffusion-controlled localization of swelling, perhaps complicated byresidual stresses or inhomogeneities?W. P. Baker: The structures are not discerned until chemical degradation hm oc-curred, evidenced by weight loss. Thus, initially chemical degradation is required andit occurs selectively in the regions eaten from the film surface. When enough materialhas been removed to weaken these areas to the necessary degree, it is quite possible tha ta fracturing process is involved, perhaps abetted by solvent cracking.

The uniform stress mentioned refers only to the application of the stress (freedom fromclamp effects, uneven stretching, thermal gradients). This uniform application of strewresults in an anisotropic film because the stress is applied in only one or two directions.It is thought th at residual strains are left in the film by the stretching process and tha tthese strained regions are preferentially attacked chemically.


15/15

ETC HING O F TER EP HTHALATE 1007I. Goodman (Imperial Chemical Industries Ltd., Fibres Division, Harrowgate, York-shire, England): 1 . Are structures similar t o those described produced a1.so in the pro-

gressive dissolution of polyethylene terephthalate films by alkaline (e.g., NaOH) treat-ment? 2. To what extent are the structures described characteristic of the chemicaletching method used?W. P. Baker: 1. Alkaline treatment of films of polyethylene terephthalate has notyielded structures similar to those described in this work. After sufficient contact withaqueous sodium hydroxide, large areas of surface in the form of thin flakes were removed.Probably the rate-controlling step was the approach of the anion from its high dielectricmedium, water, to the hydrophobic polyethylene terephthalate of low dielectric con-stant. The flaking may be caused by solvent cracking after a small amount ofetching by hydroxide ion has occurred. 2. The structures described in the paper havebeen found after etching with several organic amines. It has been reported [G. L. Ber-estneva and P. V. Kozlov, Vysokmlekulyarnye Soed ina iya , 2,1854 (1960)l hat struc-tures which appear to be similar have been produced b y several organic acids (phenols).This variety of reagents and the work reported in the paper strongly suggest that thestructures are characteristic of the substrate rather than of the etching method.R. S. Stein (University of Massachusetts, Amherst, Mass.): It seems that , if thesestructures were associated with residual strain, birefringence patterns should be ob-served with the polarizing microscope. Were they?W. P. Baker: Only cursory attempts were made to observe these structures by bire-fringence patterns, and they failed, possibly because of diffraction of light by manyunderlying noncoincident layers.

Etching Pet

Documents