Document J3yRKX3vxGkyO6RDxg1Mzvwb2

Buckuan Laboratories, Inc 934 p.a OCT *8 857 Swelling of Paint Films in Water VII. Latex Vehicles- . F. 1. BROWNE Chemist, Forest Producti Laboratory,' Forest Service, U. S. Department of Agriculture - Discuss** the offoct of foot water emulsion vehicles on the behavior of free films of paints when soaked In distilled water for three days and then redded, when soaked again for three days and redried, and when weathered before soaking for three days and redried. Measured were changes In density, absorption of water, swelling during absorption, shrinkage on redrying, loss In weight while soaking, and loss in weight while weathering. h e b e h a v io r o f f r e e f il ms of in comparing latexes as a group with Tdifferent oil tnd oleoresinous vehi the oil and oleoresinous vehicles previ cles with end without pigments when ously studied and in learning whether soaked in distilled water for 3 days pigments exert similar affects in latex and then redried was reported in the and in oleoresinous vehicles. sixth paper of this series (3).* This The pigments were basic carbonate seventh paper describes similar tests white lead, rutile titanium dioxide, of unpigmented and pigmented films zinc oxide, and magnesium silicate of in latex (water emulsion) vehicles. the same brands and grades used in Vehicles and Paints Tested the previous work with oil and oleo resinous vehicles. Pigmented films Three commercial latex vehicles were chosen: an acrylic-resin emulsion, a polyvinyl acetate emulsion, and a styrene-butadiene emulsions. The emulsions contained necessary plasti cizers, the nature of which was not disclosed by the makers. It should be recognized, of course, that different brands of any one of the three kinds of latex may vary consid erably in their properties, particularly in absorption and swelling. For the present purpose, however, the interest is not primarily in comparison of one kind of latex with another, but rather ' A cootributcd ptpex. Mttauioed ti Mfidnoc. Wu. in eoopmtioc with tbe UcmrsitT of Witcoftiic. Numben io pucotbcm refer to literature cited. TJte Author: Frederick L- browpt received B.Cbem. degree frocs Cornell U.. Pb.D> to colloid ebernirtry from U. of Viacooain. He joined the Forest Producti Lab staff in 1918. helped develop caaein flues for wood airplanes of world War 1. In 19/1-22, be was a National Research FelLow at Wisconsin. Since 10K. Dr. Browne has edited tbe General and Physical Chemistry section of Chemical Abstracts. were made with single-pigment paints only, and always at 39 per cent pig. ment volume. It was not found possi ble, however, to make a white lead paint with the polyvinyl acetate latex, because white lead apparently broke the emulsion and a granular product resulted. The latex paints were strictly experi mental. The desire was to study films composed as nearly as possible of polymer and pigment without the complications of the several additives necessary to make practical paints. Other investigators (2, 4, to 7) indi cate that additives, especially those soluble or partly soluble in water, often greatly increase the absorption of water and swelling of latex paints. Accordingly, no wetting agent, dispers ing agent, antifoaming agent, bodying agent, or preservative was added ex cept as described later in the report. The experimental paints were made in laboratory batches of less than 1 pint, with stxh precautions to avoid foaming as it was possible to take without the use of additives. Most of tbe paints field some foam, and films made from some of them, especially those containing white lead or zinc oxide, retained some air bubbles. Nevertheless all films proved im permeable to dry air under one-third atmosphere pressure except those of zinc oxide in aaylie resin and in poly vinyl acetate latex and titanium dioxide in styrene-fcctadiene latex. Paints con taining one additive, methyl cellulose in the proportion of 4.3 per cent based on the nonvolatile content of the latex, were made with etch of the four pigments in styrene-butadiene latex. For comparison with the experi mental paints, one commercial latex paint was tested. It was made with rutile titanium dioxide and magnesium silicate at 33.3 per cent pigment volume in a polyvinyl acetate latex with methyl cellulose and other cus tomary additives. Preparation of Free Films and Tests Free films of latex vehicle* and paints could not be prepared by spreading on gummed paper because the water iin tbe latex softened the dextrine gum and the coatings usually cracked in an alligator pattern as they dried. Accordingly, all films were spread on tinplate and, after curing or curing followed by artificial weather ing, were stripped by amalgamating the tinplate. All coatings were allowed to stand 30 days after spreading to allow them to cure thoroughly. Those to be weath ered were then exposed to ultra-violet Reprinted from tbe June, 1956, Forest Products Journal (Vol. VI, No. <S), pager 255-240 Forest Products Research Society, P. O. Box 2010, University Station, Madison 5, Wisconsin light from a flaming carbon arc en closed in Corex D glass (1,200 kilo watt-hours of electric input) and sprayed with distilled water three times each hour for a period of 15 days. Unweathered films of each vehicle and paint were subjected to two suc cessive cycles of soaking in distilled water for three days followed by re drying over calcium chloride. This was done because it was observed that some films of latex paint absorb much less water and swell less in the second soaking than in the first, presumably because the first soaking removes solu ble ingredients that stimulate absorp tion and swelling. Weathered films of each vehicle and paint, which were always well leached during the weath ering process, were subjected to one cycle only of soaking for three days and redrying. Dry films of latex paints usually begin to absorb water and swell much more rapidly than oil or oleoresinous paint films. For that reason great care ^ was taken to make the weighings of' dry latex Urns as quickly as possible after immersion. Wet films of latex paints, when blotted to remove free water, were found to lose weight more rapidly by evaporation while being weighed than was the case with films of oil or oleoresinous paints. The wet films of latex paint were therefore placed in taxed weighing bottles immediately after blotting. In all other respects die methods of testing were the same as those previ ously described for die oil and oleo-' resinous paints. Results with Experimental Paints The results with the experimental vehicles and paints are repotted in Table 1. Density: The relatively large in crease in density that was characteristic of oil paints during their initial dry ing period (3) was conspicuously lacking among the latex vehicles and paints. In fact, only 5 of the 18 latex vehicles and paints gained at all in density during drying and curing, and for 4 of the 5, die gun was no greater than 3 per cent. The clear film of acrylic latex was 6.4 per cent denser than the nonvolatile part of the liquid latex, but the films of clear polyvinyl acetate and styrene-butadiene dried and cured without change in density. On the other hand, 11 of the 15 latex paints made films of lower den sity than the nonvolatile part of the liquid paint. Decrease in density dur ing initial drying and curing probably indicates the formation of porous films that enclose voids. This interpretation is consistent with the finding that ail latex paints exhibited less than 100 per cent swelling efficiency, and the swell ing efficiency usually was particularly low when the decrease in density on drying and curing was more than 4 per cent. The densities of die nonvolatile por tions of the liquid latexes and paints were as follows: acrylic latex, 1.09; polyvinyl acetate latex, 1.16; styrenebutadiene latex, 1.02; acrylic paints pigmented with white lead, xine oxide, titanium dioxide, and magnesium sili cate, respectively, 2.80, 2.44, 2.02, and 1.62; polyvinyl acetate paints pig mented with zinc oxide, titanium diox ide, and magesium silicate, respec tively, 2.19, 1.95, and 1.67; styrenebutadiene prints pigmented with white lead, zinc oxufe titanium dioxide, and magnesium silicate, respectively, 2.76, 2.39, 1.97, and 1.37; and styrenebutadiene with methyl cellulose pig mented with white lead, zinc oxide, titanium dioxide, and magnesium sili cate, respectively, 2.77, 2.33, 1.98, and 1.38. When soaked in water for tire fust tim* all filw decreased in density by an amount roughly proportional to the absorption of water except thefilms of titanium dioxide and of zinc oxide in polyvinyl acetate, both of which gained in density despite large absorp tions of water. The two exceptions were the paints whose films decreased most markedly in density during drying and curing. Hence tire gain in density when soaked in water may indicate a reduc tion in volume of voids brought about by release of internal stresses in the films when they are plasticized by absorbed water. When redried after the first soak ing, all but four films regained or, more often, exceeded theu density before soaking. Increase in density and decrease in volume an to be expected as a result of loss of water-soluble in gredients of low density to the soaking water. Of the four films that failed to regain the density before soaking, two fell short by leas titan 2 per cent and contained either no pigment or tire light pigment magnesium silicate. The other two, tine oxide in styrenebutadiene with and without methyl cellulose, fell short by 3.7 and 4.8 peg cent, respectively, and also were larger in volume after tedtying than before soaking. When soaked in water a second time, the density of most films de creased, again rooghly proportionally to the absorption of water. The excep tions were films that had originally dried with a density lets than tint of the nonvolatile in the liquid paint and had not yet regained that density after the first soaking and redrying. When redried after the second soaking, all but three films regained or only very 2 slightly exceeded their density before the second soaking. The three excep tions, titanium dioxide in styrenebutadiene^ zinc oxide in acrylic resin, and magnesium silicate in styrenebutadiene, were films that remained larger b volume after redrybg than before the second soaking. Weathering increased the density of all films except those of zinc oxide in acrylic resin, in ttyrene-butadieoe, and in styrene-butadiene with methyl cellu lose. These films decreased substan tially b density. The three exceptions were notable also for very km swelling efficiencies after weathering, 32, 29, and 40 per cent, respectively, which indicates that weathering increased the volume of voids withb the films. When soaked b water, all but five weathered films decreased b density. All five of tire films that failed to decrease b density were films dot, unlike the others, were still below tire density of tire non-volatile b tire liquid paints even after weathering. Four of tire five were pamts whose snweathered films had failed to de crease in density during one or both of the soaking periods, and four of the five had, as unweathered films, b- oreased in volume after ratifying fol lowing one of tire soaking periods. These five films, then, were porous films tint teodad to become ieta potous when soaked b water. When the weathered films were redried after soaking, they all regained or very nearly ngaiood their density before soaking except one, zinc oxide b poly vinyl acetate, which retained 2.9 per greater volume after w--fcbg The observations of density changes indicate tbit the latex vehicles and paints, although subject to leas changes during film:formation and weathering from oxidation, contraction, and de composition of tire vehicle constitu ents themselves than are the oil and oleoresinous pabts, nevertheless are subject to a greater extent than tire oil paints to changes b film structure, particularly to formation of voids withb tire films and alteration b tire volume of voids. Absorption of Water: Clear films of acrylic mb b both the first and second soakings of unweathered and weathered film were remarkably high in absorption, more than twice as high as the film* of unbodied linseed oil previously reported (3). Polyvinyl acetate film before weathering ab sorbed less than unbodied linseed oil and about ts much at films of bodied linieed oil or of alkyd resin. After weathering, however, polyvinyl acetate film, unlike tire oil and oJeotesinou* films, became very highly absorptive. Unweathered ityrene-butadieoe film in the first soaking was about equal GLD38023 i . A to unweathered polyvinyl acetate, but in the second soaking the absorption by styrene-butadiene film was very low. Failure of the styrene-butadiene film on weathering precluded measurement of its absorption after weathering, which may indicate that weathering greatly increased its absorption as it did that of clear acrylic resin and polyvinyl acetate. Weathering, then, seems to increase the absorption of water by clear latex films. Pigmentation altered the absorption of all latex vehicles, usually signifi cantly. In acrylic latex, the clear film of which was very absorptive, all pig ments except zinc oxide greatly re duced absorption. Zinc oxide increased the absorption if allowance is made for the fact that only 70 per cent of the pigmented film was acrylic resin. Unweathered pigmented films of poly vinyl acetate or styrene-butadiene were always higher in absorption than the dear films of the same latex, and among them, the films with zinc oxide were highest in absorption. Thus zinc oxide seems to have tne same effect of stimulating high absorption in latex paints that it has in oil and oleoresinous paints. White lead, however, behaved dif ferently in the latex paints than it did in oil paints. In aciylic latex, white lead did not reduce absorption nearly as mudi .as titanium dioxide did. Tobl. 1.--DATA FOR FREE FILMS OF DIFFERENT KINDS OF IATEX VBOCIES WITH AND WITHOUT MOMENTS WHEN SOAKED IN WATER FOR THREE DAYS AND REDRIED, WHEN SOAKED AGAIN FOR THREE DAYS AND REDRIED, AND WHEN WEATHERED IEFORE SOAKING FOR THREE DAYS AND REDRIED--CHANGES IN DENSITY) AISORWION OF WATER AND SWELLING IN VOLUME, AREA, AND THICKNESS) SWELLING EFFICIENCY, SHRINKING IN VOtUME, AREA, AND THICK NESS WHEN REDRIED, LOSS IN WEIGHT WHILE SOAKING, AND LOSS IN WEIGHT WHILE WEATHERING tied of latex whlel* Stag* la cycle of testing iWelght In i; Desalt? of fila : while : :weatherlog; Dry ! Swollen *l/Vo * ;lo water Initial :Absorption: Smiling t Smiling t AtMekMii: of NUr i-................................ of fils : A/v*> * I In wot; In {efficiency:--- 4TiTo :TOli swolma: tthlekneae: : : *Vf : t At* t Sedrled tlMi In --j weight Arts :1bletaeM:saikiBg Percent Bile. i percent i Per- {Percent: Percent t Percent j Per- :Percent: Percent :0r, : t cent ; t 3 ; cent j : "*** Styrene-butadiene-- -.Weathered : {Weathered : {Second aoak: ;Weathered : (Weathered : Styrene-butadiene plus ethyl celluloee {Weathered : c u m meus cansuran noNsrr : 1.16 t 1.10 t 1.16 t 6.9 t 57.9 * 57.5 * 92.7 1 9.7 53 i 1<16 ; 1.10 : 1.16 1 : 1.17 : 1.09 1.16 1.0 1.6 { 52.0 68.7 S66-.72 35.3 1 31.9 r 11.6 : 21.7 : : 1.34 : 1.15 : 1.16 i 3.9 ; 8.2 : 6.1 : 5.9 : 5.5 t : 1,16 : 1.16 : 1.16 t 3*7 : 8.1 : 7.8 : 1.0 ; 3.* * 9.1 s 1.17 t 1.12 1.17 * 3.7 18.9 : 18.9 : 28.0 : 15.6 { : 1.02 : 1.00 : 1.00 : k.k : 1.C0 : 1.00 : 1.00 t k.2 : 8.7 ( 6.6 : 9-2 * -6.6 t 3-3 : 3.1 J 2,7 : 0 SU1C CAMOIdS MDS 1SO, UKU-man Man, TOMB* Touts 0.30 *.5 8.9 11.1 2.56 : 9.82 : 5.2 22.7 : 18.7 : 5-1! 10.5 2.6k ; 2.92 t 5.1 ! IS.9 : 13.2 : 3.7 1 6.7 : 2.fl9 : 2.33 : 2.89 r k.k : 17.2 t U.2 : 5.0 S 34.3 1 : 2.7k : 2.6l 1 2,79 ! k.9 s 11.9 : 11.0 t 3.9 : 3.7 : : 2.78 : 2.65 t 2.79 t 1.8 * 9.2 1 8.91 1.5 S 6.3 t : 2.62 : 2.58 1 2.92 ) k.k s 16.7 t It. 1 ; 8 : Ik.6 1 { 2.65 s 8.17 { 2.72 ! 3.9 { 515 *7.* : 3.1 t 12.1 t ; 2.72 : 2.33 ; 2.73 : 3.7 5 35.7 t 3*.7 : 1.7 s 29.1 ! : 2.75 i 2.59 : 2.75 : 3-7 ( 15-0 : 12.7 i 1.3 : 9.5 1 99 99 9!) 9<S 100 99 9* 62 70 9k 92 91 96 92 h l -0.5 { -2.2 : t -.5 J 0 : 0 { 3-8 ; 1 -6.2 J -6.3 : { -.3 t 41.1 t ( -1.2 { 5.9 s t -6.7 * -2.0 : ; -.2 t -.6 : 41,5 1 -.5 J -3*9 1 -2.1 ( -1.6 -6.7 6.6 : 3.6 : 1.2 *3 .5 9.3 .3 9.1 2.8 1 t -1.6 t 0 t : : -.3 t -.3 1 4.6 : : -3*6 1 t -.5 J : -.7 : -.9 s -.1 < -.1 : t -5.6 : 0 t s -1.2 : -1.0 : t -1.5 * -.6 : -1.9 t 0: 1.1 : 2.8 { -.6 { -.5 * -5.1 * -.3 * -1.1. 1 2.0 .1 1.0 5.5 .5 2,0 9.1 1.1 1.0 XJIXX* TIMIIM PICKIDS, inmz-niMn .Aims, naan k u mc c .j o : 2.06 : 2.03 : 2.06 : 1.9 ! 7.0 { 6.1 .* 6-2 : 1.1 : 91 t -1.6 1 -3.1 s 1.1 l 2.6 : 2.06 : 2.03 t 2.09 : 5.0 : 7.3 ( 6.7 t 1.2 : 1,6 { 92 : -.7 ( . -.2 ( -.1 t .8 {Weathered : 3*8 : 2.09 t 2.0k t 2.11 ( 5.3 : 6.5 : 6.1 : 8.6 : 1.9 : 99 -1.2 ( -.6 : -.2 : 1.1 :Piret aoak : : 1.71 : l.9 t 1.77 : 1.8 : 23.0 ( 6.3 { 6.0 1 5.0 V SI : -5.7 t -5.2 : -.6 1 3.6 tBecood aoak: (Weathered : k.2 1.91 j 1.78 : 1.6 : 21.6 { k.2 t 5.2 : -1.7 1 : l.8o : 196 t 1.82 s 3.8 : 15.0 : 13.9 : .1 : U.9 . 19 93 s ;*f* : -1,8 i -.1 s 2.1 { -.1 * .5 t .2 *6 :plret aoak : :Beccc4 aoak: {Weathered t 1-5 : 1.95 : 1.86 i 1.96 1 1. t 1.95 1 t 2.00 S 1.89 t 1.99 * 5.1 1.6 l : * 10.9 6.2 15.2 : 9.9 : t 1.3 1 t 1J.T * 6.9 : 3.2 1 1.1 : .2 1 1,6 1 21 : -3.0 t 2.8 : : 4.6 t 3 * 90 ( 0 t -.2 : 1.6 .8 { .3 Styrene-butadiene plue nethyl celluloee {Weathered : 1.8 : 1.93 : l.6k : 1.96 > 3-7 * 15.3 t Ik.8 t 10.1 : 1.1 ( 1.67 t 1.97 J 3-8 : 12.5 ; 12,0 : 6.2 : 3.t 1 : 2.01 { 1.96 I 2.02 : 3-7 * 5.9 : 5.0 : 1.1 ( 2.7 t 97 25 85 : -6.2 : -1.9 : t -.2 1 : -1.1 t '-.8I!1 -l.k : -.5 : -.5 * 7.8 .2 2.1 UK con, nmss-noMBn nns, namn raxm 0.30 : 2.1.8 : 2.06 1 2.*9 I 2.8 : 50.5 : 95.2 t 2.9 s 39:9 1 ?? t -l.e : -1.5 : -.1 t 3.9 {Weathered : 14.3 2.08 : 2.12 t 2.8 { 17.8 : 10.0 : 8.0 : 39.9 1 i 1.99 i 2.1k : 2.00 s 3-6 s 12.5 .* 13.7 t .6 : 12.1 1 6k 3* ( *2.2 : -.9 * 2.9 : 1.1 : -.9 J -.6 : -.3 t 1.2 : 1.91 i 1-95 ( 2.k0 35 : k9.7 : 12.6 ( 5* < 11.0 1 84 (-25.0 1 -6.2 : -20.0 t 3.7 tBeccad eoak: : 2.k0 1.95 t 2.10 : 2.6 : 6k.2 1 55.2 l . = 52.0 t 66 t -2.2 : 42.9 1 -3.6 : 2.1 :Weathered : 22.5 : 2.67 : 2,11 t 2.58 : 2.6 : 61.7 : 51.3 * .9 : 51.1 83 : *2.9 : -.6 t 1.3 ` 2.1 :Second aoak: . . ; 2.31 : 2.IT : 2.19 : 5.0 . 20.k ( 10.5 : 1.0 : 9.9 , ... : 2.19 : 2.19 : 2.20 : 5-3 ! 17.1 t 8.1 : 2,9 : -9.3 : 52 17 * *3.3 t -3.0 : +6.2 t 1.9 { -.8 j -.8 1 {Weathered t .7 t 2.13 : 2.19 t 2.12 : 6.0 s lfl.7 : 5.1 : -5 : 4.9 > 29 : e.k 1 -.2 : .7 : 7 Styrene-butadiene plue ethyl celluloee {Weathered : 6.9 : 2.10 l 1.97 2.00 : : 2.00 : 2.01 : 2.00 : ; 1.97 : 2.06 : 2.01 : 4.5 : 1.7 ( 1.6 1 27.7 23-k 23.7 : 15*6 t t U.5 t t 9-5 : 9.7 > 3.8 : -.9 s 19.6 t T9 t 5.6 57 19 13 ( el.l : -2.1 : { -.k ( -.5 : : -3.3 t -.1 : k.O ; 4.2 : 2.6 : 7.0 .6 2.7 mmu s s h m s iu c a t i, soKU-Pxamrr worn noon vans 0.30 Polyvinyl acetate {Weathered : tflret aoak : {Second aoak: {Weathered 18-3 7.7 {Weathered Styrcne-butadlcne plus nethyl cellulose {Weathered 1.0 6.9 t 1.57 : 1.V5 : 1.57 t 1.2 : 30.2 t 29.5 : 15.0 i 12.0 t ?| 1.17 : 1.56 : : l.fio t 1-36 : 1.6l 1 1.2 k.6 : 23.5 t 119 1 23.0 t 12.2 : {120 : 20.4 1 8-9 > 6l.1 1 93 101 1.65 s 1.71 1 k.6 ! 8.9 r 6.1 : 4.3 > -2.6 { 9k : 1.71 : l.a : 1.71 : 8.9 i 6.5 ; 2*9 : 5.1 * 95 : 1.72 s 1.61 t 1.73 1 1.1 : 08.2 : 18.2 : 5.1 6.5 10O : i.U : l.kl 1 l.k3 : 1*7 s 13.3 : 10.4 : 3.9 : : l.%3 : l.k3 : l.k2 : k.6 : 8.3 ! 5-2 ! 1.3 t : 1.51 : 1.51 1 1.50 1 5.8 : 30,4 t 6.6 t -T : 3.6 1 3-9 1 *.T 63 to 6k : 1.58 : l.k9 1 1.60 : 3-7 : 20.8 1 20.9 t 7.0 1 : 1.60 : 1.53 : 1.60 : 3.5 : 15.5 : 15.1 t 1.6 : : 1.66 : 1.62 : 1.66 t 3.5 : 6.1 : 6.6 ( .6 t 6.0 t 9.7 1 1.8 * n 106 t -1.3 * 0 : 1.0 t : -.9 : -.2 ( -1.3 ! : -1.1 { 41k .0 t -13.1 : t -6.6 ( -1.2 f : *.k t 0 : j -1.6 t --.8 : -1.7 * -,k ; -5.0 : 1 -3k : { 4.1 ( { -l.l l -.6 : -.2 l -.6 1 -2.5 : 0: -.5 t ( *7.2 : -3.0 : : -.6 1 -.3 : j -2.9 t -.8 : -1.1 { 3 : -2.0 t 2.2 -9 1.6 6.0 .3 1.9 5.8 .1 9-2 .8 30 paint ernebed and needed to beily during weathering to permit further study. 3 GLD38024 whereas in styrene-butadiene, white lead increased absorption more than titanium dioxide did, and in styrenebutadiene containing methyl cellulose, white lead increased absorption much more than even zinc oxide did. Titanium dioxide produced the lowest absorption of any pigment in all latex vehicles except polyvinyl acetate in which still lower absorption was produced by magnesium silicate. Addition of methyl cellulose to pig mented films of styrene-butadiene always increased the absorption of the unweathered films. Absorption during the second soak ing of unweathered films was usually less than during the first soaking. The chief exception was zinc oxide in poly vinyl acetate, for which absorption was distinctly higher in the second than in the first soaking. Absorption by titanium dioxide in acrylic latex was very slightly higher and absorp tion by magnesium silicate in polyvinyl acetate was the same in the second as in the first soaking. Weathering decreased the absorp tion by 9 and increased the absorption by 6 of the 15 pigmented films. The increase on weathering was very great for magnesium silicate in acrylic latex and in polyvinyl acetate, and for white lead in acrylic latex. The de crease was very great for white lead, titanium dioxide, and magnesium sili cate in styrene-butadiene containing methyl cellulose. No explanation for these differences in the effect of weath ering can be offered at present. Swelling: In general, films of latex paints differed from the oil and oleoresinous paints previously reported (3) in that the unweathered latex films never swelled in volume quite so much as their absorption of water during the 3-day soaking periods would indicate. That is, the swelling efficiency was always less than 100 per cent and usually was considerably less. More over, the swelling efficiency for the second soaking period was usually less than that for the first soaking period. Low swelling efficiency is considered indicative of porosity, or voids within the film in which free water can be held and cause no swelling. Published electron photomicrographs (1) of latex films reveal such porosity due to incomplete fusion of the latex droplets and gathering of pigment in the inter faces between droplets. Some of the water-soluble substances leached by the first soaking may likewise come from these interfaces and increase the void volume for the next soaking. Unpigmented latex films gave less evidence of porosity than pigmented films. For the unpigmented films, the swelling efficiency in the first soaking was 99 per cent. In the second soaking of acrylic latex, which lost little soluble material in the first soaking, the swelling efficiency remained at 99 per cent, but for films of polyvinyl acetate and of styrene-butadiene, which had higher solubility in the first soak ing, the efficiency dropped to 96 and 94 per cent, respectively. For weath ered polyvinyl acetate him, however, the swelling efficiency was 100 per cent. Among unweathered pigmented films, only three attained 97 to 99 per cent swelling efficiency in the first soaking: magnesium silicate in acrylic resin and in styrene-butadiene with methyl cellulose, and titanium dioxide in styrene-butadiene with methyl cellu lose. For titanium dioxide in polyvinyl acetate, the efficiency was only 27 per cent. For zinc oxide in acrylic resin, polyvinyl acetate, styrene- butadiene, and styrene-butadiene with methyl cellulose, the swelling efficiency was 90, 86, 52, and 57 per cent, re spectively. In the second soaking pe riod, the swelling efficiency for pig mented films was usually less than in the first soaking, sometimes much less. For three films, however, the two efficiencies were equal and for three the efficiency was somewhat higher in the second than in the first soaking. After weathering, the swelling effi ciency of films pigmented with zinc oxide was usually lower, and that of films with other pigments was some times lower and sometimes higher, than the swelling efficiency or the corresponding unweathered films. There seems to be no clear evidence of correlation with other properties of the films except perhaps that a rela tively large decrease in swelling effi ciency after weathering seems to be associated with a significant decrease in film density. Apparently, weather ing may either increase or decrease the volume of voids in pigmented latex films. All latex films swelled in area dur ing the 3-day soaking periods, but the areal swelling showed even greater lack of proportionality to volumetric swelling than was previously observed with oil and oleoresinous paints (3). Areal swelling ranged from 7 per cent to 107 per cent of volumetric swelling for unweathered films soaked for the first time, from 5 per cent to 87 per cent for unweathered films soaked for the second time, and from 2 per cent to 57 per cent for weathered films. In the second soaking of unweathered films, the areal swelling was usually a lower percentage of the volumetric swelling than in the first soaking, and for weathered films, the areal swelling was usually a still lower percentage of the volumetric swelling. The discrepancy in proportionality of areal to volumetric swelling was compensated by reciprocal variation in the proportionality of swelling in thickness to swelling in volume. In seven extreme cases, the thickness of unweathered films actually decreased during the soaking period. Shrinkage on Redrying: On rediying after the first soaking of unweath ered latex films, all but two shrank to a volume less than that before soaking. Such shrinkage is attributable chiefly to the fact that all films lost appreci able amounts of soluble substances to the soaking water. The films of zinc oxide in slyrene-butadiene and in styrene-butadiene with methyl cellulose remained larger in volume and also lower in density after redrying than before soaking, which indicates an in crease in volume of voids within the films. On redrying after the second soaking, most films shrank still fur ther, though less markedly than after the first soaking, which indicates smaller losses of soluble substances. Films of titanium dioxide and of mag nesium silicate in styrene-butadiene and of zinc oxide in acrylic resin, how ever, gained slightly in volume and lost in density. Redried weathered films likewise ex hibited a slight loss in volume, except for clear acrylic resin and titanium dioxide in styrene-butadiene, which ex hibited no change in volume but a slight loss in density and in soluble substances, and zinc oxide in polyvinyl acetate and in styrene-butadiene, which gained in volume and lost in density. Shrinkage in area during redrying was not at all proportional to shrink age in volume. Most films shrank proportionately less in area and corre spondingly more in thickness than they aid in volume whereas many others shrank proportionately more in area and less in thickness. In this respect, even films of the same paint often behaved differently on redrying after the first and second soaking* of the unweathered film anti after soaking of the weathered film. Toss in Weight While Soaking: During the first soaking of unweath ered films, from 1.2 to 9-4 grams of soluble substances per 100 cubic centi meters of film were lost to the soaking water. The loss was less than 3 grams per 100 cubic centimeters for only 4 of 18 films. During the second soaking the losses were less and ranged from 0.1 to 2.1 grams per 100 cubic centi meters, with only 3 losses over 1 gram per 100 cubic centimeters. Thus most of the soluble material seems to be in the latex films before the first soaking. Whether the loss during the second soaking is due to incomplete leaching the first time or to further formation 4 GL038025 of soluble substances by hydrolysis while the film is in water is as yet uncertain. The loss from weathered films was always less than that from unweathered films when soaked for the first time but greater than that from unweathered films soaked the second time. Styrene-butadiene films contain ing methyl cellulose always lost more soluble material than similarly pig mented films of styrene-butadiene without methyl cellulose at the same stage of the testing program. Pigmentation always increased the solubility of acrylic-resin films but usually decreased the solubility of jjolyvinyl acetate or styrene-butadiene Loss in Weight While Weather ing: All films lost weight during weathering, usually substantially. The losses are attributed mainly to leaching of water-soluble substances by the water spraying in the weathering cycle. Pigmentation greatly altered the weathering losses. Titanium (dioxide caused low weight losses in all latexes. Zinc oxide caused remarkably low loss in styrene-butadiene, larger loss in styrene-butadiene with methyl cellu lose, and very high loss in acrylic resin and in polyvinyl acetate. Magnesium silicate caused very low loss in styrenebutadiene, fairly high losses in styrenebutadiene with methyl cellulose and in polyvinyl acetate, ana very high loss in acrylic resin. With white lead, the loss was moderate in acrylic resin, fairly high in styrene-butadiene, and still higher when the styrene-butadiene con tained methyl cellulose. Weathering losses in films of styrene-butadiene were always higher when the films also contained methyl cellulose. Results with a Commercial Paint Table 2 shows the results of tests on films made with a commercial latex paint intended for exterior use. The density of films prepared with the commercial paint was less than the density of the nonvolatile in the liquid paint, and the swelling efficiency of the unweathered film was less than 100 per cent. The film was therefore somewhat porous, as were some of the experimental paints. The commercial paint was free from foam and made smooth films that contained no ob servable air bubbles. After weathering, the film density was no greater and perhaps a trifle less than before weathing. When soaked in water, the density of both unweathered and weathered films decreased roughly in proportion to the quantity of water absorbed. On redrying, the unweathered film re gained the density before soaking but no more, whereas the weathered film increased nearly to the density of the inous paints. Instead, the latex vehicles nonvolatile in the liquid paint and its and paints tend to form porous films swelling efficiency readied 100 per that contain voids, as indicated often cent. by a density less than that of the non The unweathered film absorbed 75.9 volatile ingredients of the liquid paint per cent of water by volume when and by a swelling efficiency in water soaked for three days. Among the ex of less than 100 per cent. The volume perimental paints, the highest absorp of voids may chance when the films tion for an unweathered film was 64.2 absorb water and dry again or when per cent (zinc oxide in polyvinyl ace the films are weathered artificially. tate). Titanium dioxide in polyvinyl There may be a reduction in void acetate absorbed 23.0 per cent, and volume, perhaps by release of internal magnesium silicate in polyvinyl acetate stresses when water plasticizes the film, absorbed 8.9 per cent. Although the or an increase in void volume, which polyvinyl acetate latexes were not the seems to be particularly likely to occur same in the commercial and experi when the film contains zinc oxide. mental paints, it is reasonable to condude that the additives necessary in Latex films absorb water on soak ing and give it up again on redrying commerdal latex paints materially in crease the absorption of water by their much more rapidly than oil or oleo resinous paints. Latex films, like oil films. In agreement with that conclu and oleoresinous films, vary greatly in sion, the solubility of the unweathered film of commercial paint was relatively the quantity of water they will absorb in three days of soaking, and the high. Weathering reduced the solubil absorption is greatly affected by pig ity and absorption of the film of com mentation. Some pigments increase and mercial paint to about the same levels others decrease the absorption. Among as those of the weathered films of experimental polyvinyl acetate paints the pigments, zinc oxide greatly in creases absorption of latex films just pigmented with titanium dioxide or as it does that of oil and oleoresinous magnesium silicate. films. White lead, which minimizes The remaining data of Table 2 con absorption by oil films, stands midway firm the essential similarity in be among the pigments in its effect in havior of the commercial and experi latex films. Titanium dioxide greatly mental paints when due allowance is reduced absorption in acrylic latex but made for the effect of the additives in was Jess effective in polyvinyl acetate the commercial paint. No measure and styrene-butadiene films. ments were made when the unweath ered film of commercial paint was subjected to a second soaking and redrying, but it was observed that absorption and swelling were much , less in a second soaking than in the first. Absorption is greater in styrenebutadiene films when they also con tain methyl cellulose than when they are free from that water-soluble sub stance. Likewise, absorption is less for all latex films during a second soaking than it was in the first, presumably Conclusions because the first soaking removes much water-soluble material. Artificial weath During film formation and curing, ering seems to increase the absorption latex vehicles and paints undergo little of some latex films and to reduce that or no shrinkage in volume or increase of others; perhaps there is a balance in density from the oxidation, contrac between production of water-soluble tion, and loss of decomposition prod decomposition products and their ucts that characterize oil and oleores- removal by leaching. Tobl. 2.--DATA FOR FREE FI IMS OF I COMMERCIAL FOLYVINYL ACETATE IATEX FAINT BEFORE A ID AFTER WEATHERING Item Density of nonvolatile in liquid paint, 2.18 Density of dry film...........-............................... .................................... . Density of film when soaked in water................. ........................ Density of redried film.......--.-- .......................... ------ Initial thickness of film, mils----------------- ---------- - --.............. .. Absorption of water, per sent by volume............................................ Swelling in volume, percent.................................. -......................... Swelling in area.........._........--....................... Swelling in thickness, per cent.........................................--................. Swelling efficiency, per cent.................. ........................................ Change in volume on rodrying, per cent..................... Change in area on redoing, per cent......................... .................. . Change in thickness on redrjnag, per cent................................ ---- Lose in weight during weathering, grams per 100 cubic centimeters. Lorn in weight during soaking, grams per 100 cubic centimeters-- Unweathered WwtlMrad film film 2.08 1.62 2.08 8.8 76.0 78.8 86.2 26.0 07 --8.8 4-4.8 --7.8 7.6 8.07 8.02 2.15 8.2 18.6 12.6 5.1 8.1 100 --4.7 --0.6 --4.8 9.4 1.6 5 GLD3802A The volumetric swelling of htex films, especially pigmented films, is usually slightly to considerably less than the volume of absorbed water, probably because some absorbed water is held in voids within the film, where it causes no swelling. The void volume is often increased by leaching the first time the film is soaked and redtied, so that the swelling efficiency becomes still lower in a second soaking. On weathering, some latex films become more porous and others less so, as indicated by the swelling efficiency. Zinc oxide, however, usually increases the porosity. When latex films are redried after soaking, the volume tends to be less than before soaking in accordance with the loss of soluble substance to the soaking water. Sometimes, however, especially if the film contains zinc oxide, such shrinkage may be more than offset by increase in void volume so that the redried volume exceeds the volume before soaking. With latex films, as with oil and oleoresinous films, swelling and shrinking in area may be proportionately larger or smaller than swelling and shrinking in volume, with compensating variation in the changes in film thickness. Much more soluble matter is leached from htex films during the first soak ing in water than during the second soaking, though the fact that there was always some further loss in the second soaking suggests that there may be slow formation of soluble substances by hydrolysis while the film is in water. The solubility and loss of weight of weathered films show that decomposition goes on during weath ering and is materially affected by the presence and nature of pigments. Ti tanium dioxide seems to reduce the loss of weight during weathering. Zinc oxide and magnesium silicate seem to increase the weight loss from acrylic and polyvinyl acetate films but reduce the loss from styrene-butadiene films. The absorption and swelling of the experimental paints used in this study were probably much less than is to be considered normal for practical latex paints because of the exclusion of water-soluble or water-absorptive addi tives from the experimental paints, except for the methyl cellulose in some of the styrene-butadiene paints. The inclusion of methyl cellulose increased absorption and swelling, and the one commercial latex paint tested, which contained methyl cellulose and other customary additives, was much higher in absorption and swelling than the experimental paints. Literature Cited X. Bell, S. H. 1955. The structure of paint films. Jour, of Oil and Colour Chemists Assoc. V. 38, pp. 595-62}. 2. Brown, G. L. and Scullin, J. P. 1953. Water penetration of emulsion polymer films. Ind. and Eng. Cbem. V. 45, pp. 743-5. 3. Browne, F. L. 1956. Swelling of paint films in water. VI. Effects of different oil or oleoresinous vehicles. Forest Prod ucts Journal Vi (4): 152-9- (Previous papers in the series were published in Jour, of FPRS, HI (5): 108-24; IV (6): 391--400; Forest Products Journal V (1): 92-6; V (2): 142-6; V (3): 192-200. 4. Fletcher, A. C., Hirsch, S. H. A., and Mayne, J. E. O. 1954. Prevention of defects in emulsion paints. Jour, of Oil and Colour Chemists Assoc. V. 37, pp. 300-16. 5. Mandelkeni, Leo, and Long, F. A. 1951. Rate of sorption of organic vapors by films of cellulose acetate. Jour, of Poly mer Science. V. 6, pp. 457-69. 6. Meier, K. and Schulz, C. 1955. Pigment electrolyte content in relation to film .swelling. Farbc und Lack. V. 61, pp. 220 7. Woodbridge, R. J. 1955. Evaluation of emulsion paints. Jour, of Oil and Colour Chemists. Assoc. V. 38, pp. 285-99. 6 GLD3B027