Document zBeKb7djBjDV2Gmb1w1VZbQR

N22989 S jL s jp i^ 7 V ' /4 9 S 7 - 7 5 0007-SWP-000114328 -9- #187 - Physical Properties of Pigments The CPVC of the simples of 1403 Ti02 in raw linseed oil (2) exposed to various degrees of relative humidity ns reported in last month's report were run again after 5 weeks of standing. The results' obtained were all slightly higher than the original values, but no sub stantial change in the packing characteristics has occurred. The data for the original determinations and after five weeks is presented in the following table: CPVC Relative Humidity Per Cent Moisture Original After 5 VW.'i eeks 0? (dried) 20? 50? 75? 100? 0 0.06? 0.12? 0,17? 0.34? 37? 38? 32? 54? 31? 52? 29? 31? 25? 27? . -L< From the table it can-be seen that any tendency toward equilib.'; rium of stability of the system 1403/moisture/(2). has apparently occurred J - before ' the' first CPVC determination, which was made about two hours after ,.'/ the samples hod been ground 16 hours on ball mills.' -Another series of "CPVC determinations on the same "samples is contemplated after further ,i;* standing for several months. jti -The porosity of double draw-downs of the systems 20 F in (3109), . (40) and'(2) were studied at PVC above and below the CPVC.' The double draw-downs. were prepared by.making a normal series of draw-downs of paints of increasing ,PVC, straddling the.CPVC as determined previously for the vpigment/binder system in question.,'-These.films were then thoroughly dried, .after .'which; a. similar ;'draw-down!series was made oyer the first, in a per pendicular direction. ..In this system each.sample of various PVC's is ' vrespectiveiy above nnd vbelow.reach;. of the other sanples.;." The results 'for ." . /the porosity of these douhle`*draw-doims" obtained- by jthe' static conductivity '' method are represented.in the.following tables for/20'.F in (3109) at PVC's iof 40?, ,45?, 50?, 55?, .60? and, 65?j 20 F in (48) .at PVC's of the same values viand . 20 F, in (2) at PVC 's .of ;35?, 40?, 45?, 50?, 55? end 60? respectively.' .The .CPyc for the. 20 ,F/(3109) `-combination is 44?, for 20:F/(48) is 46? and 20 F/(2) ,' 43?. ^ The 'greater,the:''Conductiyity,'. the foore porouB the film, vertical headings.,indicate !the first draw-downs, the horizontal headings "the 'second drawidowns. .. ' -. . Charts follow.. *I I ; - -i i 0007-SWP-044958 0007-SWP-000114329 -10- fflB7 - Cent. Conductivity - Amperes x 10 20 F in (3109) PVC % Single 40 45 50 55 60 65 40 33 39 : 44 43 55 60 60 45 600 20 37 S-t 83 140 140 50 1500 55 ` 120 215 500 270 250 55 1500 60 250 600 750 1COO 1000 60 1500 . 52' '270 670 750 . 1000 1000 65 1500 80 300 - 750 1000 1000 1000 20 F in (43) PVC t Single 40 45 50 55 60 65 40 45 , 50 ' ' 55 60 ' 65 '79 25 1200 1700 2700 3000 20 F in (2) ' 10 9 23 27 40 57 9 1 10 17 21 67 27 ioo 30 200 25' 300 25 . 39 . 51 -.17 ` 27 ' 33 82 120 ' 160 . 135 . ,185 22Q ' ?0u : 428 428,- 440 625 7^ 15` PVC i . Single 35 40 ' 45 50 ` ' 56 60 . :35 40 f.45 :. `50 55 60 86 66 o' 270 2000 ' 2000 ' 2000 ` 17 - 40 . . 48 . 86 92 107 . 22 ,/ ' 50 .V : 37 : " . . 33 7 :: . 3iV/ v , "24' ?'' 60 ' : 93 107 V 176 7 '67' 143 .' -V : 25C V: -. 250 ;' 3oo . ,136 i ' -260 ! 460 *-i i':430 - 545. - - ' .136 230 -.750 '"'5007 .665 176 - 250 ' - 1000 ' 860 -1200 r; _ A three dimensional representation ,of-the .data .forj,;2Q/:F in i :-. :; ; (3109) is shown in Figure I.-' It is pbyibus^frpa the.abcye datar:.that.;a V.- 1 porous.film[is' sealed to,a 'great extent-.by "subsequent application,of a '( . * second coat,`v'even if this `film in itself " shows'porosity when applied' I singly.r Data` is. being' obtained for other pigment/binder combinations 'and 7 " V a more , explicit analysis will be presented Then the results are ^obtained. : 'J. A paper was presented on December 26 at the A.A.A.S. meeting in Chicago on "Some Aspects of Critical Pigment Volume Relationships." - 0007-SWP-044959 0007-SWP-000114330 < 1 . 0007-SWP-000114331 #162 - Studies in Flat Wall Paint Formulation Flat Wall Paints Work has been under way to reformulate Flat-Tone White in order to conserve titanium calcium. Several variations have been made, replacing 100 and 200 pounds of 152 dry with 1403 dry and/or 2228 dry. The differ ence in bulking value between the two pigmentations was corrected by the use of additional 188 dry. The work to date has indicated that the higher amount of inert required to correct for pigment bulk has a tendency to decrease flow. Work will be continued. Requests have been received from Cleveland for ah evaluation of National Lead's One Coat Flat White and du Pont's Interior Hat Wall Paint, along with Flat-Tone and Paintercraft One Coat Flat Wall Paint #10. Sam ples of du Pont's and National's flat paints were obtained from the east coast and from the Chicago area, as material sold in the east, may be some what different from that sold in and around Chicago. A sample of Pratt and Lambert's Lyt-All Flowing Flat was also included in this evaluation, since this material is quite popular in Chicago, As indicated by the data below, the du Pont and Dutch Boy flats are quite low in sheen compared with FLat-Tone and Paintercraft One Coat Flat #10. However, due to the low sheen, they are less uni form in appear ance* There appears to be some difference in formulation between the two Dutch Boy Flats, . The sample purchased in'the east is very, good in washability, whereas the Chicago sample was found to be poor in this respect; The twp du Pont samples were also somewhat different in formulation as shown by the difference id hiding. . As indicated by the data below, we believe that flat-Tche is fairly representative of competitive flats such as Dutch BOy., du Pont or Pratt and Lambert's; Paintercraft Flat Wall #10 is equal1 to Dutch Boy for hiding and .superior in brushing qualities; flow and uniformity of sheen. Chart follows* 0007-SWP-044961 F la t Wall P aint Evaluation -27- Ocpr rbo PS h Pp_* ' CaD per aip5t oHO"10PpHD_f*H<MO3c04c*t*t ; *4 ^G_*0 O>1 S'?? *|tr <c?n> s_g c(+- opg *TJ P t+ Sp-' tp> m H O '?fc o op P?3 *& OOt*#hPWMHI So3fSt? C1 t PC* "4 <t <> CP Ol Oj Cn . : o '0 3 : fJ5 n oo a aopo. m P- 0 a o Ho> a H* Cf M H>< 8 PS .H ^ Eow' 2W <oo5 p CL ^1 hs 0' 83 & M s**? L :>. ,e< *> H> <J Sf.? EoM' Sp? 5Oo > P CL *r ? P tJ iO o cc <ft r- / o tK > - r .. TI.f. ' 'j X .1 ; > . O o, >>? rr-s* j25^ ;. X ," S > -; > ' .. / ] 1 OH . p , f;t* .I;!hp&* civ9Dpia* i* H vlj \U sc .. vr. " " '"^ - ,H . H iV T* / K1-il -' .**> ' ` .i |b!ho* g - !=L. j j - i s '*' : . - * - f-.-. ., 1 : 33 O c b FjoT ^ |P < wC5 ' CT -! r (Q . **'<0 V O: - C* ;:v. ,-.?5 09 -i y cr t'* *D H *TJoc Oi .fy P O 0007-SWP-044963 0007-SWP-000114334 J__tm //162 -- Conti -29- The following Semi-Lustres were made using the varnishes as indicated below: Varnish Semi- Under Viscosity - #3 . Brush- Dry- Lustre Observation Fresh O'Nite 5 Davs ,27 Days ing___ Flow Ing- Sheen :'. .461 L 433 L K. r; 460 L CV 7038 . eVE-981 CVR ,981-2 49B LL CVR 981-5 *' 20" 16.6" 21.6" 42.4" 23" 22.2" -- 53" 24" 20.8" 34 days 24" :18.7" 27 days 26.2" . 23" 3 days ..si" - . Good " '" ` 1* ' OK It >1 11 Experimental Enameloid Batches OK 11 11 If OK It It It t` j . Varnish v> : Enamel- Under Viscosity - #3 : old Observation Fresh 01Nite 34 Dnvs Brushing Flow Drvine Gloss % .* 430 L % $ ;-431 L (2848)E 37" CVR 981 23.8" . 463 L CVR 981-2 . - . 500 L - CVR 981-4 44.6" .24" 515 L . CVR 901-6 .- 35.4" 45" 26.8" 48" 3 days 28.8" 46" 40.8" 26" 33 Days 41.4" 19 days 25.7" . 4 duys 39" Good Good Fair + Good Fair + Good " " " " OK II It u Tt Good It i; u Based, on these tests, it appears that a satisfactory varnish of this type can be made for Semi-Lustre and Ennmeloid tints. However, further TOrk must be done to establish ..the correct body on these varuisb'-v >-.nd to .determine tha after-body te.'idencies of enamels saue with these liquids Porch and Floor Enamels The Vehicle Research Department submitted the fallowing "A" resin tarnishes for evaluation 'in j Porch and Floor Enamels, The varnishes coded CVR 894-2 , 896-2 , 903-2 and 906-2 are rechecks on small cooks origi"hlly submitted to this department for evaluation. 0007-SWP-044964 0007-SWP-000114335 -30- &62 - Coat. Varnish CVR 894-2 CVR 896-2 CVR 963 CVR 958 CVR 961 CVR 903-2 CVR 906-2 Viscosity F-G F EFFF F Color 10 12 + 11-12 11 12 8 10 A. V. 5.05 4.1 5.05 4.3 5.3 7.8 9.7 Oil length 18 19 25 18 18 15 15 H.V.M. 50? 50? 50? 50? 50? 50? Composition o' "A" Resin Varnishes (Cn N. V.M.V 50? CVR 894-2 CVR 096-2 CVR 963 CVR 958 CVR 961 CVR 903-2 CVR 906-2 A Resin 41.07 41.82 34.13 41.90 41.87 46.08 46,08 (44) 6.35 58.18 65.87 - 29,13 - - (48) 52.5B - -- -- - (140) - - - < 23.50 11.76 - - (1273) - - - 34.60 17.24 - - ySv>. ' (3017) - - - - 53.92 - Kettle lehy-- - - - - - 53.92 drated Castor Porch and Floor Enamels were made with each "A? resin varnish listed above. These were compared with regular Porch and Floor Enamel (formula 7-2-47) based on (6173) and (4950)G end a similar formulation using AV 170 and (2641) G. The various enamels were checked for general performance and ' the experimental data is reported in the chart below# The enamels were made using a drier concentration of 0*2$ lead, 0.02? manganese and 0.02? cobalt as recommended by the supplier of this particular resin. However, it was found necessary to double the drier concentration in a pigmented enamel such as Porch and Floor in order to obtain satisfactory initial drying and freedom from tack after drying over night. The enamels are to be exposed on yellow pine panels at Chicago and Florida at 45 south. 0007-SWP-044965 ... 0007-SWP-000114336 il S2 - Cont. -31 Porch and Floor Enamel Dark Gray Formula 7-2-47_________ . Viscosity - #3 O'Nite OJ No. Varnishes Under Test Fresh 4 Days Set-'bo-touch Drv Glosi 504 L Reg. Porch and Floor 53,5" Enamel - Rex 153; formula 7-2-47; var nishes (6173) and (4950)G. 50.2" 505 L Reg. Porch and Floor 26.6" Enamel - Rex 153, formula 7-2-47; var nishes AV 170 and (284l)G. 27.3" 69 minutes Very good; Good very slight tack. 57 11 Very good; Good very slight tack. 506 L (4100) - (4103) 82" 85" 35 N Very good; Good very slight tack. 507 L CVR 894-2 ("A" Resin Varnish - 18 gal lon varnish.) 36.8" 34.9" 63 ft Very good; Good slight tack. 508 L CVR 896-2 ("A" Resin linseed - 18 gal lon varnish.) , 33" 33.7" 76 ft 509 L CVR 963 ("A" Resin 28" 28" Linseed (44) - 5 gallon varnish.) 110 11 510 L CVR 958 ("A" Resin 33.2" 37.2" no ft Chinawood, soya - 18 gallon varnish.) 0- 511 L CVR 961 ("A" Resin 35.8" 39.5" no U Linseed, CTO, soya - 18 gallon varnish.) Very good; Good slight tack. Fair + Slight tack. Good Good slight tack. Good Good; slight tack. Good 512 L CVR 905-2 ("A" Pesin (3017) - 15 gallon varnish.) 37.6" 37" 62 " Very good; Good very slight tack. 513 1 CVR 906-2 ("A" Resin Kettle Dehydrated Castor - 15 gallon varnish.) 31.7" 34.4" 71 Very good; Good very slight tack. Note: Brushing and flow characteristics were satisfactory on all "A" resin Porch and Floor Enamels listed tbove. 0007-SWP-044966 0007-SWP-000114337 -32- Porch and Floor Enamel Dark GrayFormula 7-2-47 Continued It Soap Test 24 Hr. Water Test Recovery Recovery Varnishes Under Test Whitening After 2 Hrs. Whitening After 2 Hrs, Reg. Porch and Floor Very Enamel - Rex 153, slight formula 7-2-47; var nishes (6173) and (4950)G. No recovery Very slight; No recov slight blis ery. tering. 506 L Reg. Porch and Floor .Slight + Enamel - Rex 153, formula 7-2-47, var nishes AV 170 and (2841) G . (4100) - (4103) Slight + if tt ti Very slight; Complete no blister recovery, ing. it Very slight; Slight definite recovery blistering. 507 L 508 L 509 L CVR 894-2 ("A" Reain Varnish - 18 gal lon varnish.) CVR 896-2 ("A" Resin linseed - 18 gal lon varnish.) CVR 963 ("A" Resin . Linseed (44) 25 gallon varnish.) Slight Slight Slight ii tt ii n Very slight; No re slight blis covery . tering. n Very slight; No re slight blis covery. tering. tt Very slight; Complete no blister recovery. ing. j. 510 L 512 L 513 L CVR 958 (rAn Resin Chino.wood, soya 18 gallon varnish.) CVR 961 (,:A: Resin Iinteed. OWO, soya 18 gallon varnish.) CVR 903-2 ("A" Resin (3017) - 15 gallon varnish.) CVR 906-2 ( 'ATM Resin Kettle Dehydrated Castor - 15 gallon Very slight Very slight Very slight Very slight ii ti ii it H Very slight; Complete no blister recovery. ing. tt Definite- No re slight blis covery. tering. ti Very slight; Slight definite recovers'. blistering. M Very slight; Slight definite recovery. blistering. i 0007-SWP-044967 0007-SWP-000114338 #162 - Studies in Flat T.'all Paint Formulation After one month's storage all samples of Flat-Rite Undercoater mentioned in the December, 1946 report on a study of settling, are in satisfactory condition except for one sample containing AM Ac 118.5 C. Determinations were made using our special penetrometer and by spatula; The scale of the penetrometer, which reads from 0 to 100, is; adjusted so that the lever arm pointer just passes the 100 unit gradual tion when the plunger makes contact with the bottom of the paint con tainer. The rate of penetration of the plunger is in inverse ratio to the extent of settling. As there was no appreciable variation in the rate of penetration until the plunger approached very near to the hottom of the containers of all samples studied, the data has been lumped to show the rate of penetration over the scale ranges 24 to 94 apd 94 to 100 as units. The rate of penetration was reduced noticeably after the con trol pointer passed 94 on the scale. The control sample with three pounds calcium stearate showed the most settling, followed closely hy the batches carrying Acto 450, and two pounds of aluminum stearate 547 dry, respectively. Six pounds of calcium stearate showed definitely improved suspension properties, while the batches containing Armeeh C and AM Ac 118.5 C show the best suspension. TL.: latter two agents lack interest since they show low surface activity and cause unsatisfactory body characteristics. The observations made by the spatula method tend in general to confirm those of the penetrometer test except in the case of the materials containing Acto 450 and 547 dry. On the spatula test, Acto 450 material appeared to have a bead of softer and more fluffy settled material which was washed away more completely by the flow of material down the blade.- See experimental data below: Settling Test - Rex 50 Flat-Rite Undercoater Agent AM Ac Control 6# n Acto 450 Armeen C 118.5 C 3 1826 1826 547 Viscosity, #3 - fresh Viscosity, #3-1 month 55" 38" 33" 38.5" 35.6" 58" . 34" 32". 57" 39" 47" 46" Penetrometer 24-94 units 172.6" M rate per unit 2.47" 94 - 100 units 27.4" " rate per unit 4.57" 11 rating 5 170.6" 2.43" 18.4" 3.07" 1 197.6" 2.81" 19.6" 3.27" 2 174.0" 2.48" 32.7" 5.44" 6 171.5" 171.8" 2.44" 2.45" 21.4" 24.5" 3.58" 4.08" 34 Spatula te3t Full bead Bead Bead Spatula rating 4 22 fo-tings show relative not quantitative values. i = least settled 6 = most !sett-.-d Large bead 6 Slight Large bead bead 15 0007-SWP-044968 -15jfL82 - Cont. Standard viscosity Semi-Lustre 25 - 30 seconds, #3 orifice. Standard viscosity Enaroeloid, range from 26 to 36 seconds, #3 orifice. #187 - Physical Properties of Pigments A preliminary study was made of the methods available for tfye separation of pigments into particle size fractions. For this purpose, the Roller Analyzer of the Victor Chemical Compa:iy was placed at our disposal and several experimental runs were made. It was found that .the method, although slow in operation, was satisfactory for the separation of pigments of larger particle size than about two microns provided'no serious agglomeration of the particles occurred, while for fractionation of particles such as Ti02, where the average size runs below this value and where dispersion of agglomerates is diffic It, the method was not practical. This is chiefly because the.agglomerates rather than,indi vidual particles were fractionated-. . ... To avoid.this difficulty, the Noebel liquid elutriator pf.the Auxiliaries Research Laboratory was tried to ascertain whether a more complete dispersion of individual TiO-> particles in water was possible. The test runs using water, without a dispersing agent, indicated thqt ' serious agglomeration of the particles occurred, while the addition pf ' 0,5$ by weight of sodium pyrophosphate gave satisfactory results as far as dispersion was concerned. The flow rates must be so extremely: S^ow, .'.however, to obtain satisfactory separation of particles of the average v-;size of Ti02, that the method is impractical from this standpoint a? : far as larger quantity separations are -concerned. The rate was about ;'n 1 cc per. minute, using an elutriator chamber of 20 cm diameter. Us^ng -.air instead of water, increases the flow rate from the above to abpqt i,. 6,000 cc .per minute, so that from this standpoint the air elutriator' would seem very much superior to the liquid elutriator, provided d}.s- ; persion ,of the agglomerates con he achieved. Studies of the possible o. application of an electrostatic method for this purpose are underlay. rf .. . . - , ' pr- - A series of samples were made up, and runs made to determine . by the permeability nethod the "critical" p.v.c. of two pigments, 1403 -- and 20 p, and a mixture of these, in two oils,'(2) and (48) and a mixture of .these oils. The samples were made up to 25$, 30$, 35$, 40$, 45$, 50$, 55$, 60$ and 65$ by volume of pigment in vehicle, using (63) as thiijner and grinding in tall mills for 16 hours. The samples so obtained were drawn down on the standard permeability paper base, dried, and the per- ^ability determined, using Anhydrone as a desiccant inside the cups and 3.01 F. air of about 95$ R.H. outside. The runs were carried out for . about <J hours and reductions made to milligrams moisture permeated per hour, no corrections being deemed necessary for the thickness of film due to the big differences in permeation obtained with increasing p.v.c. The gloss characteristics were also run. The results obtained up to the present time are set forth in the following table; 0007-SWP-044969 0007-SWP-000114340 hi -16- ; Hi R7 - Cont. Permeability and Gloss vs. P.V.C. >lume Pigment Permea bility 1403 TiOn (2) (48) Permea Gloss bility Gloss 20 F Mg Silicate (2) ' (48) Permea Permea bility Gloss bility Glosg 25 85 3 76 93 57 63 66 8 -V`-- 30 307 0 75 90 ; 54 38 64 85 35 . s- 40 363 0 77 63 57 18 54 71 355 0 175 57 67 8 61 28 45 ; 366 0 200 52 129 5 62 11 sq; 55 360 374 0 331 0 337 53 228 54 327 6 102 6 189 8 8 60 372 0 337 - 49 307 7 270 8 65 370 0 341 53 375 8 370 9 Work on permeabilities is being continued, long drying times for the samples being necessary, however, as no drier was used in making up the samples. Oil absorption data were obtained on the above pigments and oils (in an attempt to correlate oil absorption with criticsl p.v.c. The data . to date are not complete so that positive conclusions are not possible at tithe present time. Work on the selective adsorption of drying oils from solution .oh pigments was continued. It was ascertained that the relative "wetness" Jof the pigments plays a large roll in the amount of oil adsorbed and that ..probably an isotherm will have to be run for each oil and pigment concerned before a complete picture of the physical activity concerned can be deter.mined. #177 - Pigment Studies At the request of the Bnw Materials Standardization Committee we have tested a sample of Dicalite L-10 as an alternate for diatomaceous silica as 1552 dry in Rex 450 Undercoater. Test batches were found to give standard characteristics. 0007-SWP-044970 0007-SWP-000114341 -17- #177 - Cont. Opacity 'Study Research Bulletin RC-81 concerned itself, among other things, with the opacity of pigmented systems extended rath 20 dry magnesium silicate. It was found that with magnesium silicate so long as there was sufficient liquid binder and the extender pigment was properly dispersed, the extender functioned as vehicle solids for hiding effects. Work has now been completed which studied the effects obtained with other .extender pigments. Results are given below. Each paint is formulated with 1403 dry at an H.P.V.C. of 12,4$ with a total p.v.c. of -20$ using added extender pigment. Expressed in terms pf 100 gallons of paint this would constitute a pigmentation of approximately 330 pounds 1403 dry and 150 pounds magnesium silicate. Substitutions were made as volume replacements of the magnesium silicate content. Paint No. Extender Pigment Opacity Sq.Ftt/Gcl. at 95$ C.R. I*B 1761 1762 1763 1764 1765 566 dry 1817 dry 213 dry 2209 dry 20 F Natural Ground Whiting Surfex - pptd. CaC03 Blanc Fixe . Multifex MM Magnesium Silicate 543 536 550 508 540 It will be recognised from the above that, within the limits of experimental error, the opacities developed were similar,regardless of the extender pigment used. Work also was carried on to study the effect of brightness . upon the ppacity development of white enamels. Separate batches pf a standard pnaraeloid White were shaded with a methyl violet toner and with a lamp black base. The brightness was reduced in approximately 1$ steps to determine what opacity change is effected. Brightness So.Ft./Gal, at 95$ C.R. Lamp Black Series Methyl Violet 87.2/5 Base Paint 86,7 Base Point 86.0 85.0 85.0 82.0 61.0 00.0 461 523 (ioo) 545 (1042) 565 (1082) 533 (111.52) 614 (117.32) 629 (1202) 463 485 (1002) 512 (105.52' 536 (110/52' 556 (114.52 0007-SWP-044971 0007-SWP-000114342 As might be anticipated, there is a iv.nl opacity build-up vdth lowered brightness. Hhether this is sufficient incentive to shade our v/hites is questionable. It was also found that.an enamel shaded to 805i brightness with either of these bases could be readily used as a tint base for any of our standard shades. The study of dispersion and milling effects on the opacity of flat wall paints was made. Results were negative, since it was not possible in these tests to determine any progressive change which practical tests have proved so definitely. Further work is contem plated. No work was done during the month on the following projects: #178 - S"itudies1 r of Dark Colored House Paints #148 - Examination of White Lead #112 - Fire Retardant Paints . #185 - Pretreatment of Metal Surfaces for Painting 0007-SWP-044972 0007-SWP-000114343 ' -7- jfios - Cont. Metal Protective Paints A report has been written covering the first 21 weeks' exposure of the Exterior Alternate Immersion Cooperative Tests undertaken in conjunction rdth the group on. Immersion Testing - Subcommittee VII, Committee D-l A.S.T.M. This test has previously been outlined in the monthly report for August, 1S46. To date all but five of the alternate immersion, refrigeration, exterior expo sure panels have been retired with a rust rating of 5.. Exterior panels will be continued on exposure until similar ratings are reached. Panels 9511-9526 have been retired after 12 weeks' tide range expo sure in Florida., 'These panels include tests of various whitings in 52P18 on HC1 acid pickled panels and the same paints on Manganese Phospholene, Carborustex, or HC1 - chromic acid treated panels.,, Of the whitings. Mineral Writ ing - 566 dry and Southwark #1 Whiting are the best, Sno-Float and Atomite next best, with York Whiting, Surfex, and Multifex poorest.. All whitings per formed better than the regular inert - 20 F dry, magnesium silicate. All of the special cleaning and treating methods resulted in somewhat better perform ance than did the straight HC1 acid pickling.. Panels have been exposed at Florida, at Chicago, , and in accelerated tests in the laboratory comparing the performance of Pigment E to 12054 dry zinc chromate in both 52P18, the Navy's zinc dmr--te primer,, and in Kromik, Rex 16174.. Pigment E is a barium potassium chromate developed by National ` Lead Company Research Laboratories. #177 - Pigment Studies The study of white pigment opacities has been continued to deter; mine the nature of vehicle effects on opacity and to lay some ground work or a theory to explain and to predict the effectB of vehicle change on opacity development,. Working with two pigments - 461 L dry lithopone and 1403 dry titan.ium oxide, the opacities of these materials, have been determined when dis persed in several vehicle systems. The results are as follows, expressing hiding as square feet per pound at 95$ contrast ratio: 461 L 1403 Dry 20% HPVC 35$ HPVC 2$ HPVC 35$ HPVC Vehicle Hiding Brightness Hiding Brightness Hiding Brightness Hiding Brightness (2) 40.6 (2)/(48) 49.6 (48) 47.0 49.4 6 \siU3; 051.2C I'; (3109)/(41OO1))4477.2 (3880)12 65.4 85.9 85.3 84.3 01,5 82,9 81.2 72.3 30.2 45.6 39.4 44.6 39.5 41.0 50.5 89.8 87.5 86.3 84.0 85.8 85.8 78.1 132.0 127.0 146.0 171.5 175.0 165.0 191.0 91.1 89.2 90.1 87 .0 87.2 86.3 81.2 97.3 89.4 100.0 97.3 122,5 92.5 107.0 39. 0 89. 1 39. C 37, 5 33 r 2 e7. e 5r. X 0007-SWP-044973 0007-SWP-000114344 8- - f!77 - Coni. Work is under way to determine the refractive indices of the dry binders used above. Thus far there is no obvious theory to explain the opa city differences which have been found. Evaluation of 1307 dry TiOP A comparative evaluation of the above pigment showed it to be quite similar to 1403 dry in chalking rate on our accelerated test. This test, in which the titanium oxides are ground in 1-uar and the films are exposed in the Weather-One tor, is only indicative, but in the past it has served quite well to index titanium pigments for chalking rates. The water soluble salt content of 1307 dry is high as indicated by specific resistance measurements. There was some concern that this might causa nail rusting if the pigment viare used in house paints and house paint priners. An iron corro sion test was made by making a 10% slurry of each pigment with distilled water. . Hiis was boiled 5 minutes, then a freshly sanded cold.rolled steel panel wa3 immersed in each slurry, At 20 hours* exposure results wore as follows: Slurry Ohms Soccific Resistance d H of Slurry* Panel Condition 1403 dry 6,050 8.4 Considerable rust 1307 " 1,490 ,9.0 Slight " 20 MSK dry 4,610 8.1 Very bad " 20 F dry 8,360 8.4 Bad ' " ^Measured after the panel exposure. Paint tests were made in Rexes 450, 471 and 710. In all these tests 1307 dry appeared normal for viscosity, brushing, drying, and 'gloss. #187 - Physical Properties of Pigments The January report introduced a study being made to determine by the permeability method the critical.PVC of 1405 ""02 and 20 F magnesium silicate in the individual oils, (2) and (48), and in a mixture of the two. Data accumulated in the progress of this work is given below. VJA2L repre sents a lsl mixture by volume of (2) and (48), while WAPl is a lsl mixture by volume of 1403 and 20 F. (Chart follows.) 0007-SWP-044974 0007-SWP-000114345 i___ n IiMi-- -9- #187 - 'Co nt. Vol. p Piement 1403 in WA2L Perneability GIo s e 25 46 93 30 46 93 35 48 76 40 203 69 . 45 312 63 50 321 66 55 362 64 20 F In WA2L Permea bility Oloss WAP1 i (2) Permea bility Gloss WAP1 in WA2L Permea bility Gloss 39 89 49 74 45 88 35 85 52 41 47 78 33 83 57 17 43 51 34 56 135 9 53 31 31 27 283 8 105 13 81 25 375 B 259 12 135 26 590 8 345 10 60 350 64 234 26 385 8 372 10 65 353 60 287 26 377 10 The CPVC (critical pigment volume concentration) was considered to ' be that point on the graph of the permeability and gloss vs, PVC curves where the permeability and gloss lines intersect, when the shanks of both ore re duced to tho same horizontal plane. The CPVC's for the material so far run ML WA2L M 20 F 46.5* 46,5% 41* WAP1 44* 44% 38* 1403 ' 58% 58% 26.5* The permeability work is being continued with various other combinations of pigments and vehicles. Several preliminary runs have been made on other, shorter methods in attempting to obtain the CPVC of a pigment-vehicle combination but no definite conclusions have been reached to date. $182 - Study of Tiroroved Mixing and Grinding Fractions _ AV-195, a temporary replacement vehicle for (?109j and (4100) i:. the Bnoraeloid lino, T/as tested as to its adaptability as a grind:.a.;-; vehicle. 0007-SWP-044975 0007-SWP-000114346 w- a,; f' -18- ^162 - Studies in Flat Wall Paint Formulation Semi-Lustre The Vehicle Research Department submitted the following varnishes for evaluation in Somi-Lustre tints: CVR-755 (2957) E Standard formula "A." Body I - ' CVR-748 (2957)E Special. (3316) substituted for linseed. Body H-I. CVB-753 (2957)E Special. Resin 50/50 Velsicol/ester gum and (3316) for linseed. Body H +. CVR-754 (2957)E Special. Resin-all Velsicol and (3316) for linseed. Body I, v. CVR-753 was used in the vehicle formulation for Rex 16270, card ,#11, /which consists of 21 gallons (2381) and 22 gallons (2957)E, with a special /titanium calcium-zinc oxide pigmentation. Since the experimental varnish /produced very thin bodied paint which was slow drying, no further work was done with these varnishes.. Vehicle Research then submitted CVR-759 having ..R-body, which also gave a thin paint. Extra inert increased the body but ^reduced the sheen below standard. -V r> ; ' CVR-761-2, an experimental tall oil-soya-CVtO-naleic Velsicol van quish, body S-T, was also submitted by the Vehicle Research Department, It -:-was compared with (2853)G in Semi-Lustre'Ivory, Alternate C, 7-17-46, with /the following results: Fresh 24 Hours Sheen K Regular formula Rex 201 21" 23.2" 54 K As-615 K but CVR 761-2 15.5'' Id. 6" 54 if K As 616 K but + 50# 2188 24" 27.6" 48 Work is continuing to achieve the right viscosity and sheen. Seeding In Enameloid *' Inspection of 0.000 to 0,003 inch films drawn down from c. stearate s'; free Rex 710 enamel and two others, containing 3 pounds of calcium pnd alun:f inum stearate per 100 gallons respectively, shows aluminum stearat^ to be most active in promoting seeding after aging one month at room tenperature as well as at alternate cycling between room and below zero tempera tures. The enamels carry 2147 dry Ti03 in the regular (3109)-- (4100) type vehicle. 0007-SWP-044976 0007-SWP-000114347 Fresh 1 Z weeks Room temp .Cycled. 1 1 1 month Room temp Cycled 1 1 Ratings: 1 = Very slight seeding 10 = Bad seeding 5# Stearate/lOO Gal. Calcium Aluminum 23 Z3 Z3 2-1/2 2-1/2 4 4 < No work was done during the month on the following projects: 65 - One Coat House PaintV V' #160 - Painting of Bricks ' i ic : 0007-SWP-044977 0007-SWP-000114348 -3- tm. Cont. It should be noted that the above "post-war1' Kronik was a tentative formula in which the vehicle consisted of approximately 50$ Dehydrol, 25$ (48), 12-l/2$ (2), and 12-l/2$ Ester Gum (1157) tjnsed on solids. This approaches very closely the vehicle used in card $17, Bex 16174. The vehicle of Rex 16174, card $19, which shows much worse rusting and blistering under approximately the sene exposure conditions is 75$ (48), 11-1/2$ (2), end 13-1/2$ (284l)G based on solids. ' . After approximately 600 hours on accelerated exposure, definite advantages have appeared in favor of National Lead Company's Pigment E, bariun potassium chromate, over zihe chromate in both Kronik and 52P18 with and without a top coat, particularly in the 52P18 formulation. In the salt spray cabinet the single .coat of 52P18 has rusted very badly in the case where zinc chromate was used but only slightly where Pigment E was used; Here also, although the single coat panel with regular Kronik has begun to rust, the comparable one with Pigment E is in good sjiapo. In salt water immersion, water immersion, and humidity cabinet tests the results were not so definite. In some cases the blistering was worse with those paints where Pigment E had been substituted for the zinc chrom ate and in others just the opposite. The advantage, if any, favprs Pig ment E. $177 - Pigment Studies ' j : Work is in process to correlate opacity relationships!with film ' permeability and binder quantities. It i3 hoped that some evidence will be developed to enable the formulator to anticipate film characteristics, as well as.hiding effects in and around the high dry hiding zone. 414 Dry It will be necessary to substitute-a blend of 414/511 dry for 412 dry in some Rex 471 SKP Gloss IRhite production. Assuming 414 dry with a 90/l0 zinc/lead composition and 511 dry at 50/50 zinc/lead, each 100 pounds of 412 was replaced with 37.5 pounds 414 dry plus 62.6 pounds' 511 dry. Using normal 511 dry and various samp]es of 414 dry in* this pro portion it was found that the color of the 414 must be 45 or better to make acceptable Rex 471. For tints and Rex 450 Underconter a 414 pigment with a 40 color or better is acceptable. , ] "i i 414 dry for blending with 511 dry to make Rex 471 should, comply with all other specifications on 414 dry as outlined in the November, 1946 report, pages 7 and 8. A further check is being made, of the composition of recent production of 414 dry to establish the exact lead/zinc ratio. 0007-SWP-044978 0007-SWP-000114349 -4- jf!87 - Physical Properties of Pigments In the continuation of the permeability work of the Inst several bo nths, a number of other factors besides the effect of particle size and average molecular size of the vehicle ffere studied for their effect on the change of the CPVC (critical pignent-volune-concentration), Batches were made up at 25$, 30$, 35$, 40$ and 45$ PVG of 1403 TiOe in alkali refined linseed oil (44), (44) plus (2182) linseed fatty acids at 1$ of the pigment volume, (44) + (2182) at 2$ of the pigment vol ume and (44) + (2182) at 5$ of the pigment volume, respectively.' They were ground to about 6 Hegnan grind in the l/2 pint ball mills described pre viously, drawn down on standard permeability peper,' and permeabilities determined for each batch. The results indicate that u definite shift of the CPVC toward higher PVC is achieved with increasing percentage of (2182) to about 2$ of pigment volume, above which point, a further addition of the. fatty acid seems to have no appreciable effect on the critical point. Data obtained for the adsorption of (2182) in naphtha'solution by 1403 Ti02 indicated that linseed' fatty acid corresponding to about 2$ of the pigment volume was adsorbed from solution. These would coincide with the break in the CPVC versus per cent (2182) curve, the CPVC being obtained from the permeability-gloss data outlined below. 0$ (2182) 1$ (2182) 2$ (2182) 5$ (2182) PVC Perm. Gloss Perm. Gloss Perm. Gloss Peri. Gloss 25$ 31 89 23 91 46 93 38 92 30$ 38 87 ' 40 ' 89 50 81 30' 82 35$ 30 12 42 32 59 59 55; 39 40$ 184 10 49 12 59 25 56 23 45$ 357 10 272 11 88 20 65 23 Yfork in this direction is being continued with the study of the effect of various quantities of bodied linseed (48) in raw linseed oil (2) on the shift of CPVC, 1403 Ii02 being used as the pigment. To substantiate the hypothesis that the degree of dispersion of pigments in vehicles affects the critical pigment volume concentration, four hutches.of Kromik Primer with PVCrs of 35$, 59.5$ (which is standard for Rex 16174), 45$ and 50$ were cade up and milled simultaneously in four laboratory boll mills, samples being extracted at 1 hour, 2 hours, 4 hours, 9 hours, 24 hours, and 48 hours. These dispersion periods corre spond to a Hegnan grind of 0, 1, 2, 3, 4 and 6 respectively. The samples so obtained were drawn down and the permeability and gloss characteristics determined on the dried films, with results a3 shown in the table below. 0007-SWP-044979 0007-SWP-000114350 0007-SWP-000114351 jf!87 - Coni; -6- The results indicate that the degree of .dispersion aq deter mined on the Hegman gauge strongly effects the CPVC, moving thfs value toward higher PVC with increasing fineness of grind. The general trend is probably most clearly shown on the' 5C( PVC data which reveals that although tie 46-houx grind shows no 'increase in permeability, the poorer grinds show substantial increases at'this PVC. This trend is already X well under way at 45% PVC and seems to start only shortly above the PVC of the regular formulation. ! Similar work is being continued on 1403 Ti02 in (2). ' The rusting and blistering ^characteristics are being detert mined for" the Kromik Primers tested pbove for permeability, but the `. data is not yet available, i ^182 - Study of Improved Mixing and Grinding Practices . Special studies were made of the grinding properties of the ' vehicle components of SYJP. Differences in consistency and flay of SAP f had been observed where different lots of (2) raw linseed oil had been fused. Two samples of (2) were obtained from the Oil Mill, one fresh from :'the presses and the second a well aged sample, Characteristic^ were as follows: Fresh A|ed (2) Heated Oil Foots Content (S-W Method) 0.4* Trace Iodine Number (Wij s one hour) Acid Value Color (Gardner, 1933) 183.4 .2.3 10 - 11 183.0 . j 2.4 10-11 Appearance Cloudy Clear Laboratory hutches of SWP Gloss White made with each of the ^vabove oils were ground, mixed and tested under.identical conditions. /.'Brush-puts of each showed no differences with respect to flow, ^brushing, .tody or In general appearance, Wetting and grinding tests were run with both the frejsh and Well aged linseed oils as single oils in the mill pastes. Both! samples showed identical wetting characteristics, producing mill pastes'; of equal consistencies at equal additions of oil. -The mill pastes had to be made at lower consistencies than usual, as the pastes were thixotropic and gave some trouble on the apron of the roller mill. Also, if less oil was used there was a tendency to run dry on the rolls. The paste pre pared from the aged (2) milled somewhat faster than that made with the Very raw oil, which is perhaps contrary to expectations considering the usual beneficial effects of lecithin, etc. 0007-SWP-044981 0007-SWP-000114352 : -14-- .#148 - Examination of White Lead Panels 7186-9 F were examined after 30 months in Florida, testing S-TT Rex 17 Soft Paste White Lead against Dutch Boy and Eagle-Picher Dutch Soft Pastes. The guide was two coats of Rex 450 and one coat of Rex 4^1. The white leads are all 3-coat work according to the directions for reduc ing Rex 17 as per label. Rex lT^is slightly best of the white leads, though the characteristic alligator checks are beginning to show many microscopic cracks and some visible ones. Dutch Boy is essentially the same as Rex 17 except a little more dirty, most of the dirt being mildew. The Eagle-Picher lead-in-oil is by far the poorest in appearance, due to dirt, but is only slightly more advanced in cracking than Rex 17 or Dutch Boy. The guides show no failure and in general look better than the tests. t\ #112 - Fire Retardant Paints Introductory tests have been made on the testing procedure for fire retarding coatings. Work has been resumed on the formulation and testing of this type of product. Exposures 7354-9 C were examined after 40 months at Chicago, ver tical south, on cedar siding primed with Rex 450 Undercoater, Borax is tested in CH 3265 - lire Retardant House Paint. The other materials tested are not in the same formula but in Rex 471 - SWP Gloss White of 7-21-43, which serves also ns guide. The special materials ares lead borate from .Kentucky Color Company, and zinc borates from Kentucky Color, New Jersey Zinc and Pemco. These are put in at 260 pounds (2556) for asbestine, leav ing 60 pounds of asbestine, except in the case of lead borate, which is used at 300 pounds with 186 pounds of asbestine. There is little or no choice among the several borates as regards durability or appearance. There '.3 no failure except slight erosion. A special Pemco s&jpple, #FP-?2, . shows micro-checking and slightly more erosion, and also has a yellow cast. The regular CH 3265 is dirtiest of the group and shows a surface wrinkling or.roughness. The panels will be fire tested to determine if fire retard.ancehas been maintained during exposure. #163 - One Coat House Paint Panels 11471 - 11504 were exposed as two coat work on new wood south and north vertical at Chicago, and repaints south only, covering nil SWP and One Coat House Paint house tests painted in 1946. #12 - Exterior Primers As in the case of SWP top coats, suggested alternate formulas were pent to Cleveland to afford possible substitution for magnesium silicate. Alternate extender combinations include 2/1 2203 Carbola Talc/ 2201 Metro-Nite and 1/1 2201 Metro-Nite/1989 Microije Talc. !: 0007-SWP-044982 0007-SWP-000114353 JUum #163 - Pretreatment of Metal Surfaces for Painting An additional number of aluminum1 siding panels have been obtained to supplement earlier tests. The panels'.will be prepared for pointing-as follows: No cleaning. Solvent wipe with mineral spirits. Solvent spray cleaned using.50/50 xylene/mineral spirits. H3FO4 cleaned, using Deoxidine No. 412. Vinyl butyral Resin Wash Primer VIP-1 (see Project #105). After this surface preparation, various primers will be applied, including oleoresinous zinc chromate primers OJ '39 K at 35$ PVC and 540 K at 42$ PVC, Kromik Primer and SWP 450 Undercoater. After an inter mediate coat of 450 Undercoater, SWP Gloss White will be used as the top coat for all the panels. ! #162 - Studies in Flat Wall Paint Formulation Flat Wall Paints With the prospects of having a limited amount of Ti03 and some Dehydrol available for flat wall paint production this spring, 'we have been working on an alternate formula for Rex 810 based on pure Ti02 and inert. - Experimental formulas have been made using 200 pounds 1403 dry, 200 pounds 2188 dry and 200 pounds 566 dry, as well as 160 pounds 2173, 218 pounds 566 and 218 pounds 2188 dry. The experimental flats were too low and too uneven in sheen compared with the formula on Rex 810 dated 3-21-44 (based on titanium calcium pigment). Work will be continued on this problem. . -- ...................... ............ i . ... : . t) Duringthe month we obtained a sample of Soya gel 70 from Archer- Ehniels-Midland Company for test in flat wall paint formulations. Soyagel is offered initially only at 70$ non-volatile in mineral spirits and has the following average analytical constants: Acid No. 3-5 Color (Gardner) 10 -11 Viscosity Specific Gravity .913 (7.60 lbs. per gal.) N.V.M. 70 + 1$ 0007-SWP-044983 0007-SWP-000114354 -16- #152 - Confc. Soyagel 70 was compared with Linogel - (2596) in the Paintercraft One Coat Flat Wall Paint #10 (OJ 692 K) for drying, brushing, flow and uni formity of sheen and tint. The present vehicle "s used in the Paintercraft Flat #10 is 'composed of 50 gallons $880)E and 9 gallons (2396). The experi mental flat made with 30 gallons (5680) E and H-l/4 gallons Soyagel 70 was found to be similar to the regular in all respects except for body, which was slightly on the low side. Soyagel could be used as a replacement for Linogel in the above formulation, but some adjustment would be required to increase the body by the reduction of thinner or by making some revision in the pigmentation. Flat-Rite Undercoater \ Mention has been made in the December, 1946, and January, 1947, Reports of Investigations of a study being made of the effects of additions of certain surface active agents, as well as aluminum and calcium stearates, on the suspension properties of Rex 50 Flat-Rite Ondercoater, The penetro meter is used to check settling, as described in the January report. The various batches are now three months old, having stood undisturbed for two months,prior to the current inspection. : The penetrometer data shows that all the batches except the excessively bodied 1* AM Ac 118.5 - C batch were the same in character from the top down lto within 3/8 inch above the bottom of the package. This is equal to the point 92 on the penetrometer scale, so that the tine consumed between 92 and 100 gives some indication of the firmness bf settling. ! Agent 1826 Dry Amount Used 3// Penetrometer Scale 6# 20 O 0 30 20.2 23.8 40 22.6 19.3 50 20.2 19.3 60 19.0. 18.9 70 19.3 19.2 80 19.2 19.0 86 12.2 12.0 92 11.7 11.7 96 12.6 11.3 98 8.0 100 36.8 16.0 547 Dry Acto 450 3# 1* 1/2* 0 26.0 19,5 19.5 19.0 19.4 19.4 12.0 12.5 14.7 12.2 20.0 0 19.8 21.8 22.5 19.0 19.7 19.3 12.4 13.0 9.0 15.8 17.0 0 20.7 21.0 21.4 19.0 19.5 20.0 13,5 17.7 22.6 14.1 19.6 Armeen C 1% 1/2* 0 20.0 22.2 20.0 18.7 19.7 18.8 12.2 11.2 10.0 6 20.8 21.5 20.0 19,0 19.4 19.0 12.0 11.4 11.7 21.9 50.6 AM Ac 118.5-C 1*iI 1/2* 00 22.4 19.4 '...7 22.0 23.6 21.2 23.9 19.0 20.5 19.2 19.1 19.5 ip.3 11.7 11.8 11.8 ^.0 16.5 15.9 lj.2 32.8 92 to 100 49.4 . 35.3 46.9 41.8 56.3 31.9 42.3 20.2 65.2 Posted values - time in seconds required to penetrate from one scale position to the next. 0007-SWP-044984 k S 0007-SWP-000114355 l.. ^ All the batches containing \% of the surface active agents show better suspension than the control, which carries three pounds of 1826 dry calcium stearate. Those with l/%% surface active agents settle harder than the control, suggesting that inadequate amounts may have an adverse effect on suspension. Three pounds of aluminum and calcium stearates are about equal, with six pounds of calcium stearate showing superior suspension. The 1% AM Ac 118.5-C batch shows superior suspension but is unsatisfactory due to excessive body. The. Vehicle Research Department have submitted the following varnishes for evaluation in Semi-Lustre tints: CVR 770 - #2 Tall oil, wood oil, soybean oil, Velsicol maleic pentaerythritol varnish. CVR 741 Velsicol-Pentalyn M-mnleic treated soya varnish. CVR 787 Velslcol-mnleic soya-P.E. tall oil varnish. CVR 761-2 Tall oil-soya-C.W.O.-maleic-Velsicol varnish. The saaple of CVR 761-2 was reported in February as being CK for sheen but too low in viscosity for Semi-Lu' tre tints. Additional work indicated that a satisfactory viscosity can be obtained by the use of 7 gallons CV 4526 end 2 gallons (1273) for the 9 gallons of (48) norm ally used in Rex 201. We have found CVR 761-2 to be somewhat reactive as it has shown afterbodying tendencies in the above fonnulati&n. 4i Semi-Lustre Ivory made with CVR 770-2 and CVR 787 v/as too low in sheen compared with the present standard. The Velsicol-Pentalyn M-naleic treated soya varnish - CVR 741 produced a higher sheen than CVR 770-2 and CVR 787 but it is also quite reactive with the lithopone-titaniun calcium pigmentation. While is being continued on this problem. See next page for chart. 0007-SWP-044985 0007-SWP-000114356 L~JMi #162 - Cont. Batch Vehicle -18Pigment Viscosity - #5 Fresh 24 Hrs. Aged Sheen 0J 654 K 33 gals. CVR 761 #2 7 n CV 4526 2 " (1273) 0J 684 K 33 gals. CVR 761-2 7 " CV 4526 2 " (48) 0J 683 K 33 gals. CVR 770-2 7 " CV 4526 . 2 (48) 0J 685 X * M 0J 688 X 33 gals'. CVR 787 7 " CV 4526 2 " (48) 0J 689 X it 192# 461 L 475 2132 20 2188 it n 192# 461 L 475# 2132 n 192# 461 L 475# 2132 20# 2188 22.8" 28.2" 26" 20,8" 26" 34" 25" 41.6"' 31" 28.2" 27.4" 46" 60" OK (14 days) 4t).2" Low (14 days) 33,4" Low (4 <}ays) 29.2" Low (4 days) 23,6" Low (3 days) 32.8" Low (3 days) 0J 710 X 34 gals. CVR 741 192# 461 L 7 CV 4526 475# 2132 2 ". Linseed (48) 19.4" 37,4" SI. (3 days) lower than stock (Hex 201, C 377) The factory has experienced low sheen on Seni-Lustre fftjite Bade with (3848) E. The pignentation was 75 pounds 2188 and 575 pounds 2132 dry and the vehicle as Alternate A foroula dated 8-22-45. A nunber of labora tory hatches of Rex 210, based on this fomuln were Bade to check on fact- ory production. In the laboratory batch- variations were nade in the drier, anti-skinning agent and the type of (48) used. Minor changes of this type produced no increase in sheen and it was decided that' the pig mentation would have to be revised to correct the' condition. :The foroula has been referred to Cleveland for revision by Mr. Einbecker. Cleveland, however, has reported that their production of (3848)E give high sheen in Semi-Lustre so a sample of their (3843) is being obtained;for further testB. - 0007-SWP-044986 0007-SWP-000114357 -19- ' J162 - Cont. Ennmeloid A serie3 of (S109)-(4100) type Rex 710 emmels has been pre pared to observe tile effects of Soligen rind Ferro calcium naphthenates on drier retention. The calcium naphthenates were present to the extent of 0.2^ calcium based on the vehicle solids. After seven months', eging at normal room temperatures, the mat erial containing Soligen calcium nnphthenate shows substantially improved drier retention while that containing the Ferro product shows no advantage. Both naphthenates induced excessive bodying and retarded settling com pletely. Hie control batch had settled appreciably during this period, but was readily dispersed by stirring. A summary of experimental data is tabled below: Control Soligen Ca. Ferro Ca. Fresh Enamel (Viscosity #3 (Dry-tc-touch 66 sec. 68 min. 79.2 sec. . 56 min. 86 sec. 80 min. Aged 1 wee; Dry-to-touch .60 min. 39 min. 70 min. Aged 7 no. (Viscosity #3 (Dry-to-touch 81 sec. 120 min. 10.8 sec. 95 min. 110 sec. 130 min. Ho work was done during the month on the following investigation: #160 - Painting of Bricks. 0007-SWP-044987 0007-SWP-000114358 "3 > ' -'P- ; fiflOS - Cont. 4' .. : Panels 10452 F - 10461 F jrere returned from Florida for inspec tion, after 12 months' exposure at 4p south, These are 6* x 12" cold rolled steel panels, primed, with Bejjc 16174, card #19 and top coated by Spraying with Me talastic Green and ptLack, formulated at P.V.C. varia tions from If# to 304 by adjusting the quantity of inert in the pigmenta tion. In both the Green and the Blhck the 104 P.V.C, is best in general appearance and durability. Both chalking and flatting increase rapidly with increase in P.V.C. 1 #167 - Physical Properties of Pigments .1 ^ 3. . Sufficient data on the rusting and blistering characteristics of the Kremik Primer series, the permeability curves for which were re-?, ported in the March report, have been obtained to substantiate the close relationship between CPVC, degree of dispersion and actual performance' of the paint films. With increasing fineness of grind the CPVC has shifted substantially toward higher PVC giving, for instance, a nonrusting panel at 4 H and 6 H grinds for 454 PVC, whereas the 3 H grind and lower show substantial failure of the paint film at the same FVC. Ihe typical picture of the results of the 34 salt spray test is shown in the accosponying data. vIhe humidity tests for blistering show the same characteristic trend that would be expected from both the permeability and salt spray data, i.e., those panels showing low permea bility and absence of rust show the greatest amount of blistering, while those showing bad rusting show the least amount of blistering. ! 'V *... ' . . Kromik Primers .r Ruistinc in 34 Sait Sorav at 95 F. -MOO Hours " v - - O ^ v*>.- -tT .V i-i-wt . v> v Cj 'r- Tine of .Grind , No i 1 Hr. . 2 Hrs. . 4.Hrs. o9.,Hrs. .-24 Hrs. j - 0 ~ ,C i - - 'll Fineness ,6-H..f O-H, :1-H c-' : 2--H , : 3-Hr 4-H : -f :4fl Hys 6^H 354 PVC r- (C) : (c) r : (s); A-.-. A 'A V39.54 p v c j -**v S n S ; (s) . A . A - ::.S 454 PVC ' VB VB ' ^ B' B (B) ' A A 504 PVC ' ,v b . . VB VB . VB VB VB ' B It is interesting to note in the above data that Ihe no-grind, 1 hr. and 2 hr. grinds in the 354 PVC and 39,54 PVC series, as well he the 4 hr. grind in the 39.54 FVC show a pin point type of rusting at those places where agglomerates of undispersed pigment occur. These did not increase substantially In extent or else from 100 hours 19 to 500 hour exposure, ifcile those panels which rusted in the 454 and 504 PVC series showed bod progressive failure as the exposure period increased. 0007-SWP-044988 0007-SWP-000114359 -9- | #187 - Cpnt. In Taint, Oil and Chemical Review" of Septenber 5, 1946 Mr, H. 2ohn describes a method for the rapid evaluation of the degree of rusting by neasuring the E.M.F. of the rusting panel against a colonel half cell in ordinary tap water. A similar series of E.M.F. determina tions was run on the above mentioned Kromik Primer panels exposed to salt splay, but the data obtained showed no relationship between the amount or degree of rusting and the potential developed as measured against a stand ard colonel cell. .f ' Hie dispersing ability of (2), (2) + (48) 1:1, (48) and (3109) on. 1403 TiOjg was determined by grinding in ball mills under carefully Con trolled conditions. The results indicate, that the dispersing effect rjaih closely parallel to the GPVG obtained in the same pigment-vehicle system. The dispersibility of (2) on 14&3 was definitely the poorest, it alsp' shows the lowest GPVC. . The varnish (3109) shows the highest dispersibil ity as well as the highest relative CPVC. The (2) + (48) 1:1 and straight (48) showed the sane degree of dispersing .ability and they also show nearly the seme relative CPVG to the pigment system involved, being slightly below ' (3109) and considerably above (2). Further work is being carried out in an .attempt to simplify the process of dispersion by substituting a mill of the Premier type for the ball mills, so that operating conditions are nqre easily controllable. It is hoped that a system of dispersibility factors can be worked out for various pigment and vehicle systems that can be cor related .to CPVC. To shorten and simplify the permeability. nethod for determining the CPVC of paint films, a new static method has been devised, which has the advantage of being very much' faster and not requiring a delicate ' analytical balance. The method consists essentially of determining the electrical resistivity as measured by a Kohlrausch type A-C conductivity bridge, of an electrolyte waked paint film. -. The paint to be tested is drawn down on the standard permeability draw-down paper as in the moisture method. After drying, -it is placed into a conductivity cell which is very much like the moisturo-permeability cups used so far, with the bottom cut out. The resistivity is measured directly on the A-C bridge. Whereas the moisture-permeability method requires olnost 8 hours for a series of-re sults, the some data can be obtained in less than an horn* by the new method. A substantial number of checks run between the two methods showed that the CPVC values obtained checked within about + p.5. Where the resistivity bridge is not available, a colorimetric method can be used. The same cells ore used, but instead of a salt solu tion electrolyte, Fed3 and K^Fe^N) 6 are placed on opposite sides pf the film. With a porous film, where the PVC lies above the GPVC, the typical Prussian Blue precipitate will color the film, whereas below the CPVC no color reaction results. This method iB not quite as delicate as the re sistivity method. 0007-SWP-044989 0007-SWP-000114360 #177 - Pigment Studies The 71. H. Loomis Company has supplied a special sample of Mag nesium Silicate #151-B which they would be able to supply in larger quantities. To test the material, Rex 471 batches were made substituting the regular magnesium silicates normally used in the formula. The hatphes \were ground on the stone mill. Results are given below: Paint No'. RB 1887 Me Silicate 20 F Viscositv #3 22.4" Grind 3H Gloss 49 Plow Fair Grind Time Mill Set | 65 Initial- 1888 20 MSK 17.2 3-1/4 H 47 II 40* 11 1889 Loomis 151-A 1890 Loomis 151-B 24.6 20,6 3 H 43 3-1/4 H 44 It 40f Tightened 11 60 lightened further. 525 mesh residues on the above magnesium silicate samples axe as follows: 20 F 4.8* 20 MSC 0.6* Loomis 151-A 1,4*. Loomis 151-B ...3,1* The color of ell the points produced' is satisfactory. . -. Lamar chalking teBts on various Ti02 pigmentB have been repeated in the Weather-Gaeter to establish more definitely the characteristics of 1307 dry relative to other established types. . In this series we again found 1307 dry to be similar to 1403 dry. Comparing various pigments at the point that "considerable".chalking occurred we find the following: Days Exposure to Reach Considerable Chalk 1307 Dry 5 1403 5 2173 " 25 2174 " 51 Ti Pure LW Not chalking after 31 days exposure. 0007-SWP-044990 49- ^'jjh'g - Exterior Primers In view of limited availability of 1552 dry formulas have been i set up and submitted to Cleveland In which replacements are used in 450 Dndercoater. Diealite L-10 has been approved as an equivalent for Celite 110. From work to date, it appears that 2015 dry - Dicalite"'L may be used when 1552 dry is not available, with 2077, Celite #281, to be ; used only as on emergency alternate, 1552 dry is the standard and pre ferred material, but factories are urged to watch shipments to see tijat screen residues are kept within specification limits. Panels 9234-7 CY, FY were examined after exposure for 21 months at Chioago,and 18 months in Florida vertical 60Uth. Hiese are pitch and knot sealing tests on low grade yellow pine, using shellac substitutes .under h'e SWP two coat system. All sealed sections show more peeling :than when 450-471 la applied directly on the wood. Orange Shellac is [somewhat "better than Zinnser's Zinlac, a zein-resin coating, or CAC 5-21, [Vehicle Besearch's match for Zinlac. -#162 - Studies in Flat Wall Paint Formulation in : - - Flat Wall Prints . A request was received from Or. E. 0. Witman, at Ohio State : [University, for approximately 20 gallons of Flat-Tone White containing F,5 DDT based on the total weight of paint. A formulation of this type Jjwaa developed and portions shipped to Ohio State University} Montreal, [Canada; Bound Brook, New Jersey; Oakland, California; and Orlando, Florida, An experimental Metropolitan Flat White - BL -321 A based oq lithopone pigmentation was received from.Newark for inspection. ' Tfye ^sample was cheeked for flow, drying, uniformity of tint and washability 3over glue size. It was found to be satisfactory for brushing and drying `characteristics but poor in flow and uniformity of sheen. The uniformity > of tint and washability over glue size were rated fair; f. Some tine was Bpent on the reformulation of Mnrtin-Senovir's ?#2652 Hu-Hu b Flat White to improve the odorB during its drying period, |Voriations were made on the standard formula using Shell's Special 1 ^Mineral Spirits #8296 and Dispcrsol #8235 as thinners of inproved odor. !* Other changes include the substitution of (2396) for (6063) and (5026) ' -^for (1767). In our opinion these changes have improved the drying odorp. f>Tfork continues. Work was also continued on the developnent of a flat well paint based on pure H03 and inert. To date we have tested 1989, 566, 1817, 2188, 2209, and various combinations of each inert in conjunction with . 1403 and 2173 dry. Die experimental flats were too low and uneven in sheen and/or too thin in consistency. Our present work indicates that a combination of pure Ti02, lithopone and inert would be a more desirable pigment combination to use until 2132 dry becomes more available. 0007-SWP-044991 0007-SWP-000114362 -ZQ- Seml-GlosB~ Paints The Vehicle Research Department submitted a washer of phenolicVelsicol tall oil P.E. tung oil soya maleic varnishes for evaluation in Semi-Lustre. Most of these varnishes produced paints of too low viscosity, but the principal difficulty was low gloss. The Velsicol resin was sus pected, so an additional liquid (CVR 833) was supplied based on toll oil, ester gum, maleic, P.E., tung and soybean oils. This varnish gave a higher sheen Saai-Lustre, still somewhat below stock paints. CVR S08 (tall oil-castor-P.E) produced a satisfactory Bheen. (IS This varnish was used in conjunction with (48) since higher bodied oils i I gave a tendency to reduce gloss similar to the Velsicol varnishes. A ftjo. r fU?.- plant batch of of Semi-Lustre CVR 808 is being made, and will be used in a factory batch based on Cleveland's suggested pigmentation (461 L, 1307, Ks-r.' 2188 and 566 dries). '<ijU br JjSSHSlS. ,,-r * lafi-cr ..: Sir .' - Sonnies of Devoe's Mirrolcc White and Jfade Green Enamel were obtained from Devoe and Reynolds Company, Inc. during March for inspec tion^ The two samples were evaluated for general performance with kccEnaneloid White based on the alkyd formulation. Enamel Evaluation Devoe'a Mirrolae White and Jade Green vs. Enaaeloid Wiite top aolur Kr 3 lifter r?- Devoe Mirrolae White 70Q1-08-LM4Z ` Devoe Mirrolae Jade Green Enamaloid White* };GK 981* 1 (Vehicle (3109) and'(4100) Rfype vehicle .-. 72.5* modified alkyd vnrnish-- 64.7* modified alkyd varnish Alkyd varnishes yjt, per Gallon T A 9.95 9.61 9.65 9.70 iscosity at 75 F. 68.4" ; 55.2" 60* ^'Set-to-touch SO min. 57 min. 1 hr, 15 min. Condition ft after 4 hours Slight tack Very tacky Slight tack + * 6 hours Very alight tack Definite tack V.V. slight tack * 24 hours Mo tack Slight tack No tack (Cont.) 0007-SWP-044992 0007-SWP-000114363 -8- yiB7 - physical Properties of Pigments Ihe term "dispersability factor" has been used in this work to inscribe the relative efficiency of a given vehicle to disperse a given pigment. In further work on this topic a departure from the method of |fSn mill grinding was tried with success. The method consists of a JjErect determination of agglomerate size diameters in microns as well as j-fie distribution curve of the agglomerated pigment particles dispersed ^various vehicles. The preliminary runs were confined primarily to the raluation of the dispersing ability of (2) and (48) and a 50/50 combinaioh.of these two oils on 1403 Ti02 and 20 F magnesium silicate. In con dition of the data obtained from the ball mill method it was found it (48) has a very much superior dispersability over (2) on 1403, while Latively little difference is found between these two vehicles on 20 F. Ihe sedimentation curves were obtained by making up 3? sus pensions of the paints previously prepared for CPVC determinations, in ~phtha (63), distributing the well agitated suspensions in ten Gardner- ldbyiscosity tubes and drawing off the upper portion to a previously termined level by. means of a pipette after specific periods of elapsed j V., The specific gravity of this material was then determined in a dard 5 cc weighing bottle and the amount of solids sedimented calcu lated. The particle size is calculated by Stokes' Law according to the Suation: 18 nh (D2 - DjJ gt equivalent diameter of particle; n = viscosity of suspending : height of liquid column, D2 and Dj. = density of solid particles iquid respectively, and t s time in seconds. The particle (agglomer- gff,Bize distribution curve is obtained by the conventional tangential- jtercept method. . Some of the typical data obtained for 1403 in (48) and (2) is below. (See next page.) 0007-SWP-044993 0007-SWP-000114364 Li i ^ i7 - Coot., ,9- 1403 Ti02 in i - I*s') Hay Linseed. Oil .jSSiiz--e----R-ainese in77w^- Below 2 2-4 4-6 6-8 0-10 10-12 12 - 14 14 - 16 16 - 18 18-20 20 - 22 22 - 24 24 - 26 26-28 . 28 - 30 Above 30 %. 2 4 10 15 15 10 10 9 7 4 4 2 1-1/2 1 1 4-1/2 1403 Ti02 f48l Bodied Linseed Oil ySize Ranee in Beloif \Z .2 - .4 .4 - .6 .6 - .8 .8 - 1.0 1.0 - 1.2 1.2 - 1.4 1.4 - 1.6 1.6 - 1.8 1.8 - 2.0 2.0 - 2.2 Above 2.2 __ 0.1 5 15 20 20 15 B 7 4 2 2 2 20 F Me. Silicate in f2l Size Ranee in . -L. Below 3 . 3-6 6-9 9-12 12 - 15 15 - 18 : 18-21 21 - 24 : 24 - 27 27-30 30-33 S3 - 35 36 - 39 39-42 42 - 45 45 - 48 48 - 51 51-54 54 - 57 57-60 Above 60 2 5 12 , 10 8 .6 -5 5 4 4 5 5 2-1/2 2-1/2 2 2. 1-1/2 l-l/2 1 1 20 20 F Me. Silicate in f48l Size Ranee in ' ,c< d , Below .3 . 3-6 6-9 9-12 12-15 . 15-18 18-21 21-24 24 - 27 27 - 30 30-33 33 - 36 36-39 39 - 42 42 - 45 45 - 48 48-51 51-54 54-57 57-60 Above .60 .4 8 8 9 7 .6 5 5 4 4 3 3 2-1/2 2-1/2 2 2 1-1/2 1-1/2 1 1 20 0Q07-SWP-044994 0007-SWP-000114365 -10- gjjhB7 - Cont. From the above data, the diameters of the (48) dispersions are pseen to vary from that of the single particles to about 10 times this jV value, while (2) dispersions vary from about 10 to 100 times the diameter Kof the average single particles of 1403. The aggregate volume of the agglomerates for (43) is from about 1 to 1000, while that for (2) varies I'from about 1,000 to 1,000,000 times the volume of a single particle. It is interesting to note .that commensurable agglomerate size ^distribution curves were obtained in each instance whether the original .'ground paste was chosen with a PVC above or below the CPVC. The use of larger settling containers so that larger volumes : of samples are available for specific gravity determinations and the use (of a thermostat will probably give considerably more accurate results. (#177 - Pigment Studies Rex 210 Pigment Opacity Studies A series of Semi-Lustre Whites was compared for practical hiding tty brush-out over gray-white Morest charts. These paint3 were also comJpared statistically according to the method set forth in Bulletin RC-81. "(it was found that the hiding power progression was in order statistically land practically, but that the magnitude of the differences was not as [predicted statistically . This comparison is shown below: Pigmentation Theoretical it*^' Paint No. No. Lhs. Opacity Gloss RB 1904 - 2132 575 545 35 RB 1898 2132 475 366 + * 46 a-." 461 L 192 RB 1900 2173 . 200 372 " 10 te-v RB 1905 1307 250 365 " 39 f'v-1 RB 1899 1307 200 316 ' 20 RB 1901 1403 200 316 " 10 . RB 1906 1307 150 461 L 360 288 -6 * " 3B ^Theoretical opacities bearing the + should be weighted higher due to the extension effect of combining low and high opacity pigments. 0007-SWP-044995 0007-SWP-000114366 -11- Ito - Cont. The above paints are arranged in order as regards actual Ifand theoretical opacity. It is outstanding that the difference be tween RB 1904 and RB 1898 appears small in view of the wide differ ence in the theoretical values. On the other hand, the opacity of ?paint RB 1906 is even poorer than we would have expected. Gloss dif- ferences undoubtedly affect the visual ratings of these paints. 414 Dry - Leaded Zinc j|E.> In an effort to evaluate the texture and dispensability of 414 dry, two paints containing 412 dry and 414 dry, respectively, were tmilled in steel ball mills side by side with the following results: Sample Heavy Paste Mix Grind Rating 10 Minute Milling 90 Minute Milling BA*:; 412 Dry 2H 3-1/2 H 5H 414 Dry 1H 3H It was also found that definite discoloration due to abra sion of .the steel has occurred with the 414 dry, whereas the 412 dry laid not discolor. Another characteristic which was found to be associated with 414 dry in comparison to 412 dry was the degree of oil retention ' the pigment. It was found that when paints made with 412 and 414 [were applied to primed paper the 412 dry paint dried glossy but the paint was completely flat when dry. Inspection of the paper back showed that the oil "bled-" away from the pigment into the paper. jj.i Filter paper absorption tests served to illustrate this JSjjaracteristic even more conclusively. Working with 18 dry - BCWL, fltjky " BSWL, 1707 dry - Zinc Oxide, 412 dry, and 414 dry it was found at .this characteristic appears to be associated with chemical comipsition since 1707 dry - zinc oxide loses its oil more readily than 14, 414 more than 412, etc., with the straight lead pigments retaining heir oil very closely. All of the paints tested carried an excess of gil over the critical PVC requirements. ' Dead Cvanamid . Dr. Grummitt obtained for us a sample of lead cyanamid from |re G. Hochst plant in Germany through Mr. Robert Frye, of the De- sf*of Co,I,merce* As a yellow pigment it was compared with repre sentative samples of zinc chromate and a light lead chromate pigment. 'oh 1WS fo'Jnci that the mass tone and shade tended to be similar to zinc th0^6 - 12054 dry, except that the color was somewhat greenish and brightness was down. The tinctorial strength was about one-half the zinc chromate and the let-down emphasized the mass .tone de- ^ciency evident in the straight color rub-up. Oil absorption and texture -^naracteristics of the pigment are comparable with the reference materials. 0007-S W P -044996 0007-SWP-0001 Panels 4800-5 C were examined after five years' exposure at 'Chicago south vertical. These are $ests of Bex 496 SWP Ivory over 450 tfodercoater 1940 formulas with specials in which half the extender bulk is introduced as, Cellto #281. There seems to be a little more tendency ' toward erosion with the higher amounts of Celite #281, partly offset by ^better clean-up. This test may have some significance in connection with the replacement of 1552 dry ini 450 Dndercoater. 4178 - Studies of Dark Colored House Paints . -. ' ' "5 ; T, Panel exhibits and our analysis of results on series 10436- 43 C were sent to Cleveland to show the effects of the use of cuprous oxide and a semi-alkyd vehicle in four SWP dark colors, Hillcrest Green, Willow Green, Woodland Brown and Manor Brown. There is a very definite (improvement in the browns, in color and gloss retention and in mildew control. Willow Green shows some color change but is clean and bright. There is too much color change in the Hillcrest Green to continue the pse of cirrous oxide, and we have advised Cleveland to eliminate it from such greens until further work can be done. .. - 162 - Studies in Plat Wall Paint Formulation "S'ipiV-' Semi-Gloss Paints During the month of April, the Vehicle Research Department sub mitted a number of phenolic-Velsicol-tall oil P.E.-tung oil-soya-maleic varnishes for evaluation in Semi-Lustre. Draw-downs (.003") made during :the first part of April on the experimental batches of Semi-Lustre formu lated with the submitted varnishes have become brittle. The varnishes under .test were: CVR 784, 792 , 794 , 800, 801, 802 and 80S. The regular .Semi-iLusfcre Ivory ((Alternate C formula based on (2855)G) and the Semilustre White formula based on (5848)E were also somewhat brittle,A, more' elastic film woO.d be preferred for this product. .iti V The Vehicle Research Department has also submitted two tall oll-soya-tung-raaleic P.E. ester gum varnishes CVR 333 and C751 838 for evaluation in Semi-Lustre, Both produced paints having toe j sheen uaable..t .VWUi . . : - : V .' : (Jl'i) . -i.iWork was continued on the formulation of Semi-Lustre 03iag the .pil--castor-P.E., varn3h CVR 808, leading to two formulae, one con- taining Zinc oxide and the other containing Calcium Nuodex for improved Spying. The Vehicle Research Department has provided three modifications f:CVR 808 under the designation CV 7036, 7037 and 7038. CV 7036 is the Settle cook using (2325), 7037 the small kettle cook starting with AU3) and 7036 the large kettle cook starting with (2325). CV 7036 gave Semi-Lttatre too high in body and 7057 slightly low viscosities. . CV 7038 Bave satisfactory results except in color which was dark due to the (4075) 0007-SWP-044997 -21jfl62 - Cont. Semi-Gloss Paints - Cont. It was decided to proceed with a large kettle cook of CV 7038, [ making 1500 gallons. This was done with entirely satisfactory results f/i'and the factory was asked to proceed with the manufacture of one or two * batches of Semi-Lustre on the proposed formula carrying zinc oxide, CTV 4746 and CTV 4748, j?.E. esterified Neofat 19 varnishes, . were submitted by the Varnish-Resin Laboratory for evaluation as Semi- Lustre vehicles. They are not usable for this purpose as both produced ^ very low sheen paints and CTV 4746 gave very thin bodied Semi-Lustre. * The varnishes were turned oyer to the Chicago Technical Service Department to test in railway paints. Devoe's Mirrolac White and Jade Green were compared with i,\ Enameloid (alkyd formula) for embrittlement on glass after aging for a j period of eight weeks. At the end of this period ho difference could i&be detected in the elasticity of the various films. All three enamels ' ^produced a good ribbon when scratched with a spatula. No work done on the following projects during the month: *#148 - Ex-ami nation of White Lead '#160 - Painting of Bricks Iftss. - Pretreatment of Metal Surfaces for Painting 0007-SWP-044998 ti 0007-SWP-000114369 I #105 - Cont. ) -2 - ; A study of the comparativejresults of humidity and immersion accelerated testing has been undertaken in cooperation with a group in Subcommittee VII, Committee D-l, A.SiT.M. The premise is that if results are identical, then humidity testing?can be dropped as immersion testing is much simpler. To date no appreciable results have developed in either method of testing after about three feeks of exposure. I #187 - Physical Properties of Pigments A series of paints of increasing PVC's has been prepared to study the effect of moisture on the properties of pigment-vehicle com binations, 1405 Ti02 was U3ed in (2j and (48), with moisture contents varying from 0$ (these pigments and .vehicles were dried) to 5$ water based on the weight of the pigment. ' The CPVC (critical pigment concen tration) was determined in each case by the usual technique employed during thiB investigation; drawing the milled paint3 down on the stand ard permeability paper, drying, and determining the conductivity of the electrolyte soaked film on an AC resistivity bridge. The gloss charac teristics and dispersibility of the vehicle-pigment systems were also determined. The results obtained are shown in the following curves and tables. Figure I shows the effect of increasing water content in mat erially lowering the CPVC of the 1405 - (2) combination. Curve A rep resents the values obtained for samples made with dried pigments and vehicles; B, the "normal1' pigment and vehicles as commercially obtained; C, with a water content of 0.38$ of pigment volume; D, 0.63$ and E, F, and G, 1.1$, 2.1$ and 5.1$ of pigment volume respectively. The very sharp drop in CPVC in the initial low water content samples is of extreme interest. While the dri^d pigment-vehicle system shows a CPVC of about 58$, the presence of only little more than 0.1$ of moisture drops this value down to about 52$; The presence of addi tional moisture reduces the values of the CPVC still further - 0.38$to 28$ CPVC, 0.63$ to 26$ CPVC and .1.1$ to 22$CPVC. Beyond this point, more water introduced into the system seems to have no further disad vantageous effect on the CPVC. The values resulting for 2$ and 5$ moisture are substantially the same as that for 1$. A plot of moisture content against critical pigment volume concentration is shown in Figure in. ---- It is interesting to note that the dried sample of 1405 after exposure to. normal atmospheric conditions over night reabsorbed substan tially the same quantity of moisture it originally contained, namely about 0.1$ . 0007-SWP-044999 0007-SWP-000114370 6000 PVC 1, -5-.Permeability of film in .Relation to Pigment Volume Concentration with Various Moisture Contents Systems 1403/(2) pigure X Conductivity in A x 10 "4 ---------- 5*- 0007-SWP-000114371 6000 0007-SWP-000114372 Variations of Critical Pigment Volume Concentration with Moisture Content Figure III; ooov-swp-oasooa 0007-SWP-000114373 0007-SWP-000114374 Fspr -7- Hiat other factors, such;as gloss and ease of dispersion are strongly affected by the presence of even small quantities of moisture is shown in Table I. The glos3 of the film originally containing 0.1$ moisture falls off sharply from the values obtained in the case of the film with the previously dried constituents, and with increasing mois ture content a further loss in gloss is encountered. The dispersive effect of the vehicle on the pigmeiit also decreases materially with increasing moisture content as can be seen in the data of Table I. A substitution of (48) for (2) shows the same general trend, as is shown in Figure II and curvejB of Figure III, although the initial slope of this curve indicates thatjvery small quantities of moisture do not affect the CPVC as strongly as in the case of (2). Table II repre sents the values obtained for gloss and dispersibility determinations of 1403 in (48) with various moisture contents. Similar determinations.on the effect of moisture on the charac teristics of paint films'are being determined for various other combina tions of pigments and vehicles. A paper was presented before the Research Meeting on June 10, at Dayton, Ohio, on "Considerations of Figment Surface Areas." #177 - Figment Studies In the report for May, oil retention properties of 414 dry were compared with various other white lead and 2inc pigments, with the conclu sion that the chemical conposition was the controlling factor. ' Zinc oxide appeared to lose its oil readily, whereas the lead pigments retained theirs closely. ...... '`7~ Further work has now been done which illustrates that chemical composition is not the only controlling factor. Blends of pigments have been made to equal 412 dry composition. It was found that none of these blends had the oil holding property associated with 412 dry. Oil penetration tests on filter paper show the following in millimeters penetration: ; v 412 dry 414 " 414-9 dry blend 414-511 blend 1707 dry 1707-9 dry blend 7 iron 10 iron 15 mm 17 mm 16 93 13 sun 0007-SWP-045004 0007-SWP-000114375 -6- Introductory work was done to establish the characteristics of the pulp lead being supplied by the National Lead Company and to investi gate what changes would be required before the product could be incorporated readily in current formulations. This study is being extended by the Cleve land Technical Service Department and a committee from the Auxiliaries De partment. Their reports represent the latest i'^ormation on this topic. Work continues on the extension of the data presented in Research Bulletin RC 81 which discussed the opacity of white pigments. It has been found that the Ti02 pigments shift to the high dry hiding zone above the range of 45$ HPVC. #182 - Study of Improved Mixing and Grinding Practices Triton 773, a 20$ aqueous dispersion of an aryl polyether sulfate supplied by the Rohm and Haas Chemical Company was tested for its effect on the dispersibility of anatase H03 A-24, 2147 dry, in (3109), a soya alkyd type vehicle. The material was used on the basis of 1$ of the weight of the pigment. It was compared with an agent-free control paste and one containing 1$ of Acto 450 - GW (2343), all made up to 100 units Plastograph consistency. Dispersions were effected at a fixed setting of the 11" 3-roll Lehmann mill. - Triton 773 showed a slight effect upon the vehicle requirements of a paste of 100 units Plastograph consistency. It increased the mill output substantially but failed to measure up to the performance of Acto 450. See experimental data below: Paste Volume Grind Gals, per Hr. $ Gain 33.9 gals. 6H 19.1 - 29.1 V 6 H 27.6 44.4$ 32.2 " 6H . 4.8 29.8$ A sample of Wallace Tiernan's Detergent S.P. was evaluated as to consistency effect and grind aid efficiency. 1$ of this material, based on the weight of the pigment, caused an increase in paste consistencies as measured by the Stormer viscosimeter. This indicated a lack of efficiency as a wetting agent in the (3109) type of Enameloid White - Rex 710 paste system. However, it suggested the possibility that this material could have a favorable effect upon mill outputs by increasing the "tack" of the pastes on the rolls. Mill rate tests show a very slight advantage in mill output through the use of this agent, but little in comparison with that from Acto 450. See experimental data below: 0007-SWP-045005 0007-SWP-000114376 .WlWnW- -18- A house test at 10446 Trumbull Avenue, Chicago, started in June, 1946 was examined. On the north side, CH 4697 One Coat House Paint with 50/50 1403/2173 titanium looked .'essentially equal in whiteness and general appearance to a straight 1403 formula. Chalk was rated "less than considerable," dirt at "considerable." The south side was similarly alike as regards 1403 versus 1403/2173, but the advantage of less shading was with the 50/50. The best test area, on the west showed the straight 1403 clearly cleaner and whiter, and looking very good considering the rather serious dirt known to have been present at 6 months (midwinter). A total of 66 gallons of one coat house paints whites and tints were sent to Cleveland for house tests there. #178 - Studies of Dark Colored House Paints During the past few years a number of revisions have been made in barn paint vehicles, based on various oil substitutes, such as Petropon #2138, esterified tall oil. Brown Resin #90, etc. To check the effect on durability, barn paints were prepared for exposure, representing the last five regular revisions for Rex 54. The paints will be exposed on cedar and yellow pine at Chicago and Coffeyville, #86 - Leaded Zinc Pigments (Reinstated) /shc. Panels 11656-61 Cs, CN, Cof. and F were prepared for exposure r-north vertical Chicago, and south vertical at Chicago, Coffeyville and Florida. The purpose Of the test is to evaluate 414 dry for durability //characteristics and whether the variations in composition influence the "co-fumed" durability performance. Hex 450 Un^ercoater, Rex 471 Gloss ' White and Rex 496 Ivory were all prepared using standard 412 dry, 414 dry. 39? minimum standard, and 414 diy AURX. 398 average standard. tl-f-'v Panels 5581-5595 C, CD were examined after five years' exposure iiat Chicago south vertical. The paints were exposed on cedar and Douglas ' _ :fir siding over Rex 450 and also self-primed. The paints were prepared //during the titanium shortage of 1941 and carry 455 1405 Ji02 in the pre-war / .vehicle at '3055 : PVC. Because of the generally high zinc oxide pigmentation, 5'/.7% mica was used .to reduce checking and cracking. Tne paints includes i(a) Dow BCWL content, 16%, .7% mica. 1. Using 414 W leaded zinc - 49.5^ ZnO, 8.755 BS7IL, 1556 20 L. 2. Using 1526 leaded zinc - 3556 ZnO, 23.5% BSWL, 15% 20 L. 3. Using 412 leaded zinc - 37.8^ ZnO, 20.4% 35STL, 15% 20 L. 0007-SWP-045006 0007-SWP-000114377 -19- #86 - Cont. (b) No BCWL, 7? mica. 4. Using 414 W leaded zinc r 59.8? ZnO, 10.4? BSWL, 19? 20 L. 5. Usii^J 1526 leaded zinc -41.7? ZnO, 27.9? BSWL, 19? 20 L. 6. Using 412 leaded zinc - j>2 ZnO, 24.4? BSWL, 19? 20 L. i (c) Specials 7. 2 coats lead-in-oil, 8. Sears 0. W. over Rex 450. 9. Rex 471, stock, over Rex 450, Durability runs close to normal pre-war Rex 471 for the (a) and (b) paints, which is good performance for paints with such high zinc oxide content, All are too dirty to be acceptable, as chalking is inadequate. It'is also worth noting that the "No BCWL" paints are definitely inferior used self-primed colored to their use over Rex 450, illustrating the beneficial effects of lead soaps in priming. Lead-in-oil is dirtiest and shows more cracking and flaking on Douglas fir than any of the other paints, but shows good durability on cedar. #12 - Exterior Primers In line with Cleveland's request, a check was made on a number of - laboratory and plant batches of Rex 450 SUP Undercoater for after-body tendencies. Cleveland had reported that a number of production batches of Rex 450 Undercoater were showing considerable after-bodying in the package. However, the results of our investigation did not show any general increase. . Ihere were a few laboratory batches made with 414 dry that were quite high in body but this was not evident with factory batches made with 414 dry. It may be that certain lots of 414 dry are more reactive than others, hut the Auxiliaries Research Laboratory reports that there are only minor dif ferences in the reactivity characteristics of 414 and 412 dry in house Paint formulations. 450 Undercoaters will be' made using 412, 414, 511, 2295, 2059, 108 and XX 505 dry to check on the after-body tendencies of these pigments having varying reactivity characteristics. In connection with the recent trend to use more soya oil in Paints to replace the higher cost linseed, a series cf house paint under coaters has been formulated using various percentages of soya, castor and tung oils. The oils used include (3316), (2325), (2381), (1273) and C7R-828, a bodied tung oil. Top coats of Rex 471 made with all linseed, ^d also using the maximum amount of soya will be used in exposures. 0007-SWP-045007 0007-SWP-000114378 -9- jfiS7 - Physical Property5 of Pigments in continuation of the study on the effect of moisture on rCPVC batches of 1405 Ti02 in (3109) as well es 20 F magnesium silicate ''-in (2) (48) and (5109) were made with moisture contents of 0% added is" commercially available pigments and vehicles), l/2, 1% and 2% ^water added and the CFVC determined. Figure I shows graphically the ^results obtained. Curves A and B are again included, although previously reported, to show thoir relative positions to the other values. As can ibe seen from the figure, the use of (5109) as a vehicle with 1403 pigment ^materially reduces tho deleterious effect of moisture on the CPVC as com'pared to (2) and (40), a reduction of only about 5% of CPVC resulting from the addition of Z% water to the batches. 20 F on the other hand, acts entirely differently from 140S in 'its'response to CPVC change with moisture content. As curves D, E and F fdlearly show, the reduction of CPVC with increased moisture is almost negligible. I0 study this apparently entirely different behavior of these two pigments with respect to water and the various vehicles employed, it seemed desirable to obtain a quantitative value for the different "wetta/ilitiefl" of the liquids and solids involved. Bertell and co-workers developed the Bartel'- cell to determine the contact angle ami adhesion tension between liquids and powdered Solids, such as pignuts. In this cell, capillar/ action - a measure of tie adhesion tension and contact angle - is balanced b.- the static pres sure of a gas. The capillaries are pores produced try ecmpreusing the cwder under hydraulic pressure. The Bartell cell is rather complicated 'ncTrequires considerable technique for its manipulation. WK A new method has been developed for the determination of wettaUity utilizing simple apparatus and technique. A dynamic flow method ^employed, balancing the rate of viscous flow into a plug of compressed (paent against the capillary attraction causing imbibition. The appara tus, as shown schematically in Figure II, consists of a cylinder A of Jjhstantial thickness into which a plug of pigment B is conpressed by a .draulic press. The upper end of the cylinder is threaded to receive jf;cap C, inserting a rubber washer for airtight fit. A thin piece of _ tubing leads from this lid to a piece of rubber tubing D, the "ar end 0f which is attached to the bottom of a burette. F is a vent ^Permit the escape of the air when the.liquid is first introduced into ^ Pigment plug, and is closed during the testing period. .r 0007-SWP-045008 0007-SWP-0001 -10Figure I Variations in Critical Figment Volume Concentration with Moisture Content Percent Moisture ------0007-SWP-045009 0007-SWP-000114380 0007-SWP-000114381 -12- The procedure for obtaining the adhesion tension of a liquid W:,ioT a given powder consists of compressing a plug of the pigment to be tested at from 1,000 to 5,000 lbs./in.2, depending.on the nature of the pigment, assembling the apparatus as indicated in Figure II and placing iLthe liquid to be used into the burette. The burette is now raised until |?the liquid flows into the cell above the porous plug. Vent F is kept |open until this point, at which time it is closed. The time is now noted Von a stop watch, and the volume of liquid imbibed by the plug recorded |at various periods of elapsed time. By plotting the volume of liquid taken up by the plug against S|'the square root of the elapsed time, a straight line, results, the slope Wof which, squared, and multiplied by the viscosity of the liquid being Trun at the temperature of the experiment is a direct measure of the |adhesion tension according to the equation: . v: 2 A k3 fcos & Jwhere V is the volume imbibed in time t; 7^ is the viscosity of the liquid; |A is a constant' determined by the pigment; B is the instrument constant f:and Yce5 is .the adhesion tension. To minimize pressure losses due to hydrostatic head, the level |0f the liquid in the burette should be kept fairly closely to the level the pigment plug.. 03y: .. For the benefit of those interested in the theory of the appara tus j the following short derivation of the principles involved is givens Box . ; ........... ; - The driving force causing imbibition of a liquid into a porous iPlug is caused by the capillary action of the thin channels and is equal W?1 a 7rr /"co.J i w^ere r is the radius of the individual pore3 Fend.' c o s 0 is the adhesion tension between liquid and solid. The Retarding force slowing the flow of the liquid into the plug is due to , ^;the viscous resistance of the liquid and is equal to: X = 2 TX 77 [_ Inhere ^ ia the coefficient of viscosity of the liquid, is the . Elshear velocity and L is the length of the capillary. At equilibrium these ^forces must be equal and fy - . By a series of integrations and sub- Vftitutions it can be shown that the volume, V, of liquid imbibed in time, is equal to: i yea e xt J" " 7 f- N is the total number of capillaries in the plug, B is the average f , ius f the capillaries and the remainder of the designations conform V) those above. 0007-SWP-045011 1 0007-SWP-000114382 p.tl-' s'. pft7 - Cont. -15- This equation can be reduced to the values already mentioned by substituting the pigment constant A and instrument constant B for A more explicit form of the derivation will be presented in a research bulletin, which will be distributed at a later date. The results obtained for the,adhesion tension of water and various organic liquids including solutions of (2), (48) and (3109) in (63) against 1403 and 20 F gave values which substantiated the results to be expected from the moisture - CPVC experiments. The adhesion tension of water against 1403 was extremely high, being about 130 dynes, while all the values for the organic liquids run are considerably lower, running between 45 and 100 dynes. For 20 F, on the other hand, the adhesion , tension against water is only about 35 dynes, and below any of the values for the organic liquids tested, which run from 55 to 75 dynes. Other com binations of pigments and liquids are in the process of being tested. A Bartell Cell is under construction to test the compatibility of the two methods of obtaining "wettability." #182 - Study of Improved Mixing and Grinding Practices The Vehicle Research Department has submitted several samples of tall oil sludge extracts for evaluation as wetting agents. ' Mill rate tests on Eoameloid showed unexplainable inconsistencies and irregulari ties in the performance of . the various controls used in the several series of tests. Work on the tall oil sludge samples was suspended until the cause of such irregularities could be-traced down. A study of the mixing technique disclosed that in two control batches the 1B26 dry calcium stearate had been introduced at different points in the mixing schedule. TShen the calcium stearate was added early the vehicle requirements for a 100 Plastograph units paste consistency were considerably greater and the mill output was less than when the 1826 dry was introduced into the paste with the last of the pigment. Compare the data on batches CC 122 and CC 124 given below. Batches containing CW (2343) Ac to 450 at lj6 of the weight of the pigment showed the same consistency effects resulting from differences in the point of addition of the calcium stearate. This suggests the possi bility that an efficient mixing technique can provide the same degree of wetting obtainable with an effective wetting agent. See CC 124 and 126. 0007-SWP-045012 0007-SWP-000114383 -5- - Pretreatment of Metal Surfaces for Painting Two proprietary .products for treating rusted steel surfaces prior to painting were examined and placed on exposure tests. They are called Rustrem and Corodex, Rustrem is a bituminous-like liquid that fel is brushed or wiped on a rusted surface, allowed to dry, and then painted j|r over. The sample under observation required a week to become tack-free gri and caused very bed cracking.of both\ Kromik and Red Lead primers. It also bled to the surface of the primers and any subsequent top coats, K"'' causing very poor drying. It is a very unsatisfactory base for oleofe* resinous or oil paints. Corodex is some type of a liquid acid pickling H solution. For best results the rusted surfaces should be immersed in H the Corodex but repeated wipings can be used successfully. In either j method water rinsing is necessary. Corodex actively attacks rust, even in the cold, and has an effective inhibiting agent so that the steel IK. surface is not appreciably etched. Kith oleoresinous or oil paints no ip undesirable characteristics were noted when applied over Corodex treated Bp- steel but with baked vinyl finishes a dark, spotted discoloration occurs. Bp Exposure tests will be reported when completed. I!?. #187 - Physical Properties of Pigments The critical pigment volume concentration of a paint has been l|y,recognized as the transitional point above or below which substantial S'- differences in the appearance and behavior of a paint film will be en!&_ countered. It is that point at which just sufficient free binder is present to completely fill the voids left between the pigment particles g.. incorporated in the film. Among other things, such qualities as gloss, ^Trusting, .blistering and hiding are intrinsically.connected with CPVC, gl and differing characteristics can be imparted to a paint containing the fv, same constituents by formulating above or below it. , Bp"' The CPVC has been determined by making up batches of increasing |g|pigment volume concentration, usually in 5$ increments, covering the Erregioq in which the CPVC is expected. For normal purposes, PVC's of 5$, I|s 30%, 35$, 40%, 45$, 50$, 55$ and 60$ fully cover, the range. These batches gte-are then ground to the desired dispersion, drawn down"on "permeability H paper," dried for a sufficient period to ensure a representative film, H and the degree of porosity of the various films determined by the moisture K;permeability or the conductivity methods. The whole procedure requires ||, considerable manipulation and is very time consuming, requiring from 'several days to several weeks, depending on the rapidity with which a ^ completely dried film of the paint in question can be obtained. To eliminate the disadvantages of the above cumbersome procedure> a new method has been devised in which the CPVC can be determined on a single sanple of paint in the wet state. 0007-SWP-045013 0007-SWP-000114384 6- - jfl87 - Cont. The method consists of making up a sample of the paint, the CP VC of which is to he determined, at a pigment volume concentration of about 30$, grinding to the proper dispersion, v/ith or without the addition of thinner at this point. The ground paint is then diluted to a total pigment volume of about 10$ by the addition of further thinner, placed in the CPVC cell, all excess liquid filtered off and the volume of the remaining paste cake determined. The ratio of the true volume of pigment in the paste to the apparent volume of the cake has been found to give substantially the value of the CPVC as determined by the draw-down method. The CPVC cell consists, as the figure illustrates, of a standard ground glass joint, to the top portion of the male section "A" of which a perforated brass plate "B" has been firmly cemented. The female section is con stricted shortly above the ground glass portion to a narrow opening around which a calibration mark "D" has been placed. A piece of filter paper, cut to the proper diameter is placed on top of the brass plate to act as the filter medium. CPVC Cell 0007-SWP-045014 0007-SWP-000114385 -4- M05_ Cont. f,- In the A.S.T.M. study of comparative results of humidity and /immersion accelerated testing reported in the Monthly Report for June, *11947, no definite or even tentative conclusions can be made,' No blister ing occurred on any of the humidity panels nor on one set of the immersion /panels. Only the one remaining set of immersion panels blistered. Our /thought is that the apparatus used far humidity testing did not produce /& relative humidity of IOC# and thus 'failed to delicate standard humidity cabinets. This first test was run at' 100 F. A second test is being ^conducted at 110 F. - Pretreatment of Metsl Surfaces for Painting A meeting was held to discuss the painting of aluminum clapboard aiding, both production line finishing for prefabricated houses and when already on houses before painting. The minutes of the meeting have been Stencilled and have been distributed to those receiving copies of this report. 87 -Physical Properties of Pigments // Work with the new CPVC cell as described in last month's report /continued. The critical pigment volume concentration of a number of Sgmmercial paint samples were run and the relationship between the actual 'pjjjjjula pigment volume concentration and the CP70 determined. A certain amount of difficulty was encountered in the filtra^ti-df some samples, as the material seemed to run around the edges of ^filter paper rather than through it, carrying some cf the solid (Mfccleis_ along into the filtrate. Various /vetting .techniques tried completely eliminate this, so we have resorted tc a cell with a plate. This seems to eliminate the difficulty coupletely. -^ration of about ?0 samples through the same fritted plate without p^equent acid wash docs not materially decrease tire filtration rate, , anticipated plugging-of the filter pores with small pigment ides'does not seem to take place. Further work is being conducted 7 - find to increase the general applicability of the method,. \ v A centrifuge method for the same purpose was tried, but as was cipated, packing volumes of lower values were obtained than by use RvCPVC cell or. the draw-down method. if: an attempt to determine the reason for tile decrease of CPVC *f2v TS?nce of incfeased moisture as reported for the case of 1403 * v' find (3109) in a previous monthly report, the adsorption fthe eris^CE f the above binders on 1403 end 20 F were determined presence of an increasing amount of water. 0007-SWP-045015 -5- )jhS7 - Confc, ' The method consists of.. "grinding" 20 grams of the pigment with '50 cc of a 5% solution of the vehicle in (63) in a glass container, after '0=, 0.5%, 1% and 2% of water (by-volume of pigment) have been added to the various samples* The dispersion was carried out for 3 hours by rolling with steel balls, after which thH pigments were centrifuged out and a 10 cc Sample of the supernatant liquid`carefully dried for 3 hours at 110 C. The difference in weight of,the vehicle remaining to that of the original solution, represents -the weight of vehicle adsorbed on the surface of the pigment. The accompanying chart 'shows the results obtained. Piement 1403 1403 1403 20 F 20 F 20 F Vehicle ! Per Cent Moisture ISail 0.6* Mi (2) (4a) (3109) (2) (48) (3109) '.005 ' .012 -,025 ' .007 .010 .017 .005 .012 .025 .006 .010 .017 .006 .011 .029 .006 .010 .017 mi .006 .011 .023 .006 .010 ;016 The figures represent the grans of vehicle adsorbed per gram of `igment. They are contrary to the differences which might have been expected from the CFVC data as to vehicle adsorbed with increased percent ages of moisture. - If water molecules would displace oil molecules adsorbed .oh:the pigment surfaces, large differences might be expected. As it is, the differences between the vehicle adsorbed at low moisture content and high moisture content are small, scarcely greater than the experimental- .error*' It is interesting to note that these differences are in the' same der; of .magnitude for bqth 1405 and 20 FI '1403 showed a very large ^ecrease of OPVC with increased moisture'content,'while in the case of 0,F this change was negligible*. " v' '. : v-y :-V.*r:>V-. ' *' . : ?.r-<4- In contrast to the above, it can be shown that (48) molecules i|ist completely displace previously adsorbed (2) molecules from the, surface of 1403 and 20 F. 25 cc of 5% solution of (2) in (63) were ' -Jhced in contact with 20 g.of 1405 and 20 F..respectively, and "ground" -Hir^hburs in the previously described method. Then'25 cc of 5% solu~ ,4%of' (40) in (63) were added and the samples subjected to an additional ^hour dispersion. Subsequent drying and weighing showed that the amount adsorbed was almost quantitatively the same as that absorbed from straight (48) solution. - Work on both the CPVC - moisture relation and adsorption studies he continued. d: 0007-SWP-045016 JL -14- }M ff fig - Studies, in Flat Wall Paint Formulation Semi-Lustre Two plant batches of CV 7041 (25 gallon CYfO-Dehydrol-Tall Oil-P.E. Cornish) wore made and tested in Semi-Lustre Cream. The characteristics Sjrere: Plant Batch CV 7041 No. 1 No. 2 Eody T- U W Color 16-17 18 Acid Value 4.0 4.9 Batch No. 1 was slightly low in body and No. 2 appeared somewhat in Semi-Lustre after three days. A factory batch of Semi-Lustre iCreom will be made with batch No. 2 to check the sheen and body character istics of this varnish. apt...;'' a'tch Varnish Viscosity - M Drying Photovolt Fresh O'Nite 7 Days Set-to-touch O'Nite Sheen 93.L CV 7038 94 L CV 7041 &- \ CV 7041-#2 fr* Enamels 23.8" 20.2" 26.6" 26.2" 26.6" 55" 24.4" 25" 55" - 42" (3 days) 62" ' OK ; OK -OK 58 63 60 2 ; ! i . A plant batch of (2848)E based on the kettle dehydration of (113) as compared with standard (2848)E in Rex 710 Enemeloid White. The enamel appeared to be standard in brushing, drying, etc., with flow slightly igfaer with the special (2848)E. (Cont.) 0007-SWP-045017 0007-SWP-000114388 -15- 1 >. jfi1 t r t Vfccosity #3 Fresh Over Might Drying Set-to-touch Over night Water Spot Test for 24 hours 1$ Soap Test for 5 hours Brushing OJ 191 L (2B48)E 43" 48" 1 hr. 30 min. Good SI, whitening Good Good 0J 192 L (2848) E Sp. 35.2" 38" 1 hr, 25 min. Good SI. whitening SI. + whitening Good Slightly high Gieck tests on Lowe Brothers' enamel LH 501 were made,after .two months' age. The conclusions at two months were identical with those |P ot the one week stage. Enamel LH 501 is an excellent product compared ' with other brushing, air drying enamels but, as would be expected, does '..not compare in hardness with baked appliance finishes. No work was done during the month on the following investigations: #178 - Studies of Dark Colored House Painty f- #177 - Pigment Studies .7* #?7 - Mildew Inhibitors , g \?8_ - Leaded Zinc Pigments #148 - Examination of VJhite Lead 0007-SWP-045018 0007-SWP-000114389 fi.87 - Physical Properties of Pigments '` The recognition, of tha physical meaning of CPVC as being .that point at which just sufficient bitter is present to completely fill the yoids left between the`particles of a pigraent/binder system, -leads to a necessary change in the concepts of pigment/binder relationships. Until the present, the binder in a paint system has been classified into two ^general types, namely the'."bound'1; oil and "free" oil. The bound oil has generally been considered to consist of an envelope of liquid molecules '-firmly adsorbed to the surface of the. pigment particles and may physically be identified with the pigment. These molecules of binder are removed -from the surface of>the pigment only through considerable effort and can consequently be regarded as being truly "bound" to the pigment by strong physico-chemical, forces,1 The quantity of such binder, firmly adsorbed to the solid particles can be obtained from an adsorption from solution deter- .mication and varies between low values of less than 1% to well above 10* ; of'pigment volume, depending:upon the type of vehicle employed in the . System, Some typical values are; 1403 in (2), 2.5*; 1403 in (48), .4.8564 1403 in (3109), 10.3*; 20 F in (2), 2*; 20 Fin (48), 3,1*.and 20 F. in ' ; (3109), 5% of oil by pigment volume adsorbed. The free, binder has been* considered to be all the rest of ;the oil present in the paint system at the formulated-PVC. It is that :qil which is. easily'removable from- the system in the wet state by such 6impie- methods-as filtration and can.consequently be regarded as being, 'truly ."free." : ; -- ! y '. . ; ...With the concept of CPVC this "free" portion of the binder must be subdivided into two major.portions namely, I, that portion which is required to ..fill the interstices between the pigment particles when ' ^ttiese pro-at .their.densest^.degree of. packing commensurate'with .their, dispersion, and- II,- that portion of the binder which is in excess of' -this amount tot the formulated PVC. The first of-this oil can be regarded as being..^interstitial!,'.free binder, .end the-second ns "excess" free 'binder, : . > .' : .' .* *' \ * ' . These reintionships' can probably be most .adequately repre sented by Figure I. 0007-SW-045019 p S'? ~ - Cont. 0% Pigment 100/6 Binder Volumetric % Pigment-*^-Volumetric % Binder 10C# Pigment 0% Binder 'Figure I If a permeability Vs- PVC curve is determined for n given Ipigment/binder system, the -results Con be represented by a typical graph fas e.bpve ', The sharp break-in rttte permeability represents the .CPVC, or Hhnt percentage of pigment "to- oil tdiere just sufficient binder is present j give a continuous solid/liquid phase after all thinner hns volatilised,, tie PVC of the paint Is determined by the "formulation, while the bound oil |;can b evaluated by an adsorption determination, Mathematically, the ssyste^i can be defined by the simple relation: Total volume = PVC + excess free oil + interstitial .. free oil + bound oil and: excess free oil -f interstitial free .oil - total free oil Where agglomerated systems are encountered it is necessary to : subdivide the interstitial free binder into two further groins, namely: 1, intra-agglomerate oil and 2, extra-agglomerate oil, The first repre sents that binder which is present inside the pores of the individual agglomerates, while the second Is that whieh fills the spaces between them, or: interstitial free oil = intra-agglomerate oil + extra-agglomerate oil. 0007-SWP-045020 0007-SWP-000114391 > $r $1B7 - Pont. 8- - An evaluation of these two .terms can be made with the assump,-tion. that the particleB participating in an agglomerate ere- at their . densest degree of packing. Consequently, in. an isolated agglomerate l. the some pigment/vehicle ratio, would, exist as is obtained by- a CPVC .. determination on a monp-disperse-.system of .the same pigment. The arithmetical difference between the CPVC of the mono-disperse system and that -:of the agglomerated. system, then represents- the extra-agglomerate oil, . -..end. the difference between this and:the interstitial-oil is the intra agglomerate oil.-, -These relations ate represented in graphical form in Figure XI, where the remainder of the values^conform to those of Figure I. X exira-agglomera te J - -oil ^ injnra-agglomerate oil ' h09 aco, I *rf Je sj s O Mw>> /4 "i I- -=* I Figure II 0007-SWP-045021 0007-SWP-000114392 t -9- ft87 - Coni. From Figure II, it can-easily- be .perceived that; the : eactrcagglomerate oil repf-esfents the difference:in efficiency .of packing be tween an agglomerated and a:completely.dispersed.pi^ne'nt .system. For the latter, the intra-agglotoerate oil simultaneously represents the interstitial free binderi A consideration' <St thefie factors immediately-raises the ques tion, "Which portion Of. the vehicle is responsible for the vri.de diver-r, gence of pigment packing pehavicr?1' Ety,:the -.use of various vehicles and agents the same'pigment system can be induced to give high or low values of CPVC. -. .- , The answer can readily be fpuiid through the use of the CPVC. , cell. 1 It is possible to coaqsletely substitute. .the free binderof- a paint system by a solvent and still retain the. same packing .characteristics as characterize the original paint> The interstitial binder and excess binder consequently must be considered, to ploy.-brily a. minor, role.in the establish-^ ment of.'.the CPVC, while (he bound .oil; is the main factor, in determining: this value. "'."V ' '. ' In; -....... =- :.Theyehicle.substitutionswerecarried-out-.hy,;dlluting,.the:- original-paint'.Itlr'by volume with naphtha or.Otherhiidtahle.Bdlyehtrcnd. ceptrifuging: out'the: pigment*-.The. supernatahtliquid'ii- decanted :and replaced with, the same volume of Solvent, ' ihg pigmeht'is' then redispersed into,' the solvent by rigorous- shaking^ '.and.>epietiti6as of; this whole. pro-: cedure made until ..only tiaces. of .bind^ycah'.tie .detected idythe solution!. A;,C? VC determination :by means ..of :.the!,-CPyC. Cell . is -.:theh. barji.ed out on - i this'dispersiori, '.while another . iavmaie ohythe^brigiriai phirit. .The re-..- sults._ obtained:by .these two methods ''chiebfcljweli ;withih;.Cxperinjehtal.;error for-systems exhibiting both ^highly'.agglomerated, as .well as highly dig- . persed states. ....... : That the,adsorbed or bound oil hot-removed from the surface -.of pigment particles by simple, dilution .of this .type has been poirited . out ..by ff. ' Ewing in "Symposium on'New Methods for-Particle >lze i)eterminhiioa .- in the Subsieve Range" page 108'-',(1941), "and has been further corroborated in the course of this' investigation^ .J An attempt w c b made with .the cooperation bf .the OilBesearch Laboratory to separate 20, F: magnesium ,-ailicpte into ^various .particle size fractions by use of a long elutriatioh .coli4iiuu . :Diifor.tunately,, the results obtained.were not very encouraging .because of1 the inability .to fluidize the material. The pigment packed, very badiyalong the sides.of the elutriator and even the use of pre-heate<i air had ho .advantageous effect in breaking up this packing. Further work in this direction has been halted for the time being. 0007-SWP-045022 I 0007-SWP-000114393 ffi.62 - Studies in Flat Wall Paint Formulation Semi-Lustre As stated in the September report, the first factory hatch of CV 7041 produced Semi-Lustre which Was too low in body, while the second hatch was too reactive. A third batch was prepared under the supervision ;of the Vehicle Research Department to determine whether a slightly lower . bodied varnish would be more stable. Batch No. 3 had a body of U to V and an acid value of 6. However,<when used in Semi-Lustre Cream it was 'as reactive as Batch No. 2. No further work is contemplated. Porch and Floor Enamel In the August report reference was made to a number of "A11 resin "varnishes submitted by the Vehicle Research Department for evaluation in Porch and Floor Jinamels. Enamels prepared with these varnishes have been evaluated for embrittlement and scrub resistance in comparison with ether resin and alkyd formulations as well as regular Porch and Floor Enamel based on (6173) and (2841)G,' ' -T. Considerable difficulty has always been encountered in evalu ating enamels for scrub resistance. As indicated, by the data below, the regular Porch and Floor Enamel (OJ 152 L) made with laboratory prepared jrarnishes showed a definite break in the film after having been scrubbed for six hours, whereas an identical formulation made with factory var nishes (oj211 L) lasted only 3-1/4 hours. However, the results should "somewhat indicative of the comparative abrasion resistance of the raisbes tested^/ (See next page for chart.) 'f'"- ' Panels 12009-17 C and F were prepared to expose these Porch /Floor Enamels. The panels- are 24'U yellow pine and will be gpospd 45 south at Chicago and Miami; Panels 7602-31 C and F were examined after exposure 45 south or 12 months in Florida and 40 months at Chicago, and a final report Men.' .The panels were Dougles Fir. The paints tested were Porch ;Deck grays of 1942, covering various types of varnishes with two es'of pigmentation: (l). leaded zinc end (2) Titanox RC. As stated previous reports the latter paints were far superior in durability hPPearance to those made with leaded zinc. Selection of superior shes is made difficult because of pitch exudations. Possibly m. (3111), (2948) N and (2946) EA would be given preference, while (485Q)H and (429)A seem inferior'. 0007-SWP-045023 -26- 1162 - Cont. Abrasion and Embrittlement Tests on Porch and Floor Enamels After One Month (0.003M Films on Glass) Batch Varnish Used Elasticity Hours Scrubbed Before Film Broke :QJ 152 L (6175) - CVR 911 and (2841)G Laboratory varnishes ' Good 6 hours 0J 153 L CVR 894 - 18 gal. (48) "A,f Resin Varnish rt 4 hours 45 minutes OJ 154 L CVR 896 - 18 gal. (44) "A" Resin Varnish M 2 hours li'oj 155 h CVR 903 - 15 gal. (3017) "A" Resin Varnish It 1 hour 20 minutes K.OJ 156 L CVR 906 - 15 gal.. Kettle Dehydrated Castor "A" Resin Varnish gToj .157 L v: CVR 910 - 15 gal. Solvent Process Castor "A" Resin Varnish ;0J 1B4 L Alkyd Varnishes (4103) and (4100) II 8 hours 15 minutes W 2 hours 45.minutes 5 hours 45 minutes |0J 210 L RB 7019 - Ether Resin Varnish ?DJ 211 L (6173) - Factory Production ' (2841) G " 27 hours 50 minutes 5 hours 15 minutes 0007-SWP-045024 0007-SWP-000114395 -9- ^197 - Physical Properties of Pigments. The study of the effect of moisture on CPVC has been continued, 1403 Ti02 was exposed to atmospheres of 20$, 50%, 75% and 100$ relative humidities at 85 F. for 24 hours. Under these conditions the percentages of moisture retained in the pigments a?e Relative Humidity 0$ (dried sample) 20$ 50$ 75$ 100$ 0$ 0,06$ 0,12$ 0.17$ 0,34$ The relative humidity was established by the use of solutions of potassium fluoride, calcium nitrate, end sodium acetate in water in the presence of excess salts for 2C$, 60$ and 75$ R,H, respectively, while the 100$ R.H, was obtained in the presence of water alone. The pigments treated as above were made up to a PVC of 30$ in (2) raw linseed oil, ground over night on ball mills and the CPVC deter mined by means of the CPVC cell. The results obtained check well with those resulting from the addition of water to the batches as set forth in previous reports, and are represented in the following table. Relative Humidity CPVC of 1403 in (2) 0$ (dried) 20$ 50$ 75$ 100$ 37$ 32$ .51$ 29$ 25% It is obvious from the above that 1403 Ti02 exposed to differing atmospheric conditions can result in substantial differences in the CPVC and consequently can cause erratic behavior of paint batches of this pig ment in (2). That this behavior is not due, as might be suspected, to differences of adsorption of (2) on the pigment, can bo shown by selective adsorption from solution determinations. The 1403 exposed to the various conditions of relative humidity all showed substantially the same amount of adsorption of raw linseed oil as the dried sample, indicating that the Cause for this phenomenon is probably not to be found directly at the pigment/liquid interface, A paper was written and graphs and slides prepared for pre sentation at the A,A.A.S, meeting in Chicago, December 26, 1947, on the subject "Some Aspects of Critical Pigment Volume Relationships." 000V-SWP-045025 0007-SWP-000114396 -22Leaded zincs 1526 dry end 414 W are compered in exposures 7380-7 F, exposed 36 months at Florida, vertical south, cedar beveled siding primed with Rex 450* The paints used are SWP bodied oil tint bases of 7-21-43, Cream and Slate. ; There is no failure greater than microscopic, hence the 414 W is stainding up fully as well as 1526 so far. 414 W is represented by both a routine sample from Coffeyville, lot #251, and a special "Canadian fume composite'' supplied by G. E* Prewitt as representative of production for the period July 7 - Sept ember 13, 1945. There is no difference between the two samples to date. Past results show a slight disadvantage in appearance due to slow chalk ing of 414 W, but this effect is no longer noticeable. 0007-SWP-045026 0007-SWP-000114397 *-25#177 - Pigment Studies Panels 12078 - 12120 CS, CN and F have been prepared to study the relationship between, the reactivity of .special lead pigments and the film durability obtained* It has been suggested that due to the rela tively higher PbO contents of Be sic Silicate White Lead and Tri-base (highly basic lead sulphate) these materials should be formulated into paints using smaller increments. In panels 12078 - 12120 a variety of lead pigments are studied at various lead content levels from 125 down to 15 in five steps. Exposures are at Chicago north and south, vertical; Coffeyville south, vertical; and Florida south, vertical. 0007-SWP-045027 0007-SWP-000114398 -25#162 - Studies in Flat Wall Paint Formulation oer Bound in 400 Bound drums. f.o.b. New York. tsvei, wnexi xnuuuucBu i/a on cquuj. wot u u o j .s> - ,. detect any improvement over (3082) from the standpoint of mas ing yp cal paint odors. The Metro-Site Company have suggested tho use of Metro-Nite ' 2201 dry for lowering cost and. conserving hiding pigment in flat wall i point formulations. . To check these claims, flat wall paints have been made by replacing the present inerts (2188 dry and 20 t) on an equal f volume basis with 2201 dry. In a second variation 60 pounds oi 2132 ary jj,. and all inert was replaced with 2201 dry. The paints were compared with regular flat-Tone for brushing, i flow, drying, washability and hiding, as well as uniformity of tint and sheen. The results .of the tests indicated that 2201 dry is too low in consistency compared with the present inerts to obtain proper bo jr and 'suspension properties in the finished paint. Paint made with 2201 ary : 1 were less uniform in tint and sheen compared with the standard formula tion. bhere 2201 dry was used for hiding pigment the pa~nt was lower in hiding, as would.be expected. There seem to bo no advantages and' some disadvantages in the use of this extender in Harwell paints. . Semi--Lustra B: of rv Chicle Research Department has prepared a factory batch 1 variation *a a Sei-I>ustre vehicle. This varnish is a te (4C1ZK) ^7 170 in whioh half of the linseod had been replaced with t faetorv ?U,aids cnd esterified with polyhydrfo alcohol. The *' aCt0ry batch of 7043 had the following chaznctoristics: Body - J Color - 16 A. V. - 9.6 0007-SWP-045028 0007-SWP-000114399 -26- Seml-Lustre - Cont. In line with Cleveland's request, a formulation was developed for Semi-Lustre tints and copies of the proposed formula OJ 568 L has been sent to Cleveland, The new formulation appears to be satisfactory for brushing, flow, sheen and drying characteristics. However, the raw material cost is approximately 25^ per gallon higher than the present production and we do not feel that ttfis increase should be absorbed if it is possible to dispose of the (40^5) overstock in some other way, A small plant batch of Semi-Lustre Cream will be made with the new varnish CV 7043 end samples of the new production will be sent to Cleveland for their inspection. Enameloid The Vehicle Research Department has submitted three 25 gallon P.E.-Rosin-C.W.O. varnishes prepared by the solvent process and a 25 galT Ion Peritdyn A - Chnawood Varnish - CVR 975, made by an open kettle cook-l ing procedure. < f!4l) t (346) CVR 973 CVR 974 CVR 975 CVR 976 4 22.5 - - . 41 Coffey-Lukas Solvent Process 22,5. " " " " 45 Open kettle - regular 45 Coffey-Lukas Solvent Process The varnishes were compared with (2848)E in the Enameloid Cream formula dated 5-21-47. The enamels were compared for general performance with the regular (2848)E formulation. The solvent processed varnishes compared quite favorably for drying with the so-called regular CVR 975 which was made in the open kettle from Fentalyn A and chinawood oil. However, it was found that a 25 gallon Dehydrol-rosin varnish such &3 (2848)b produces a higher gloss and a tougher film after drying over night. All four experimental var nishes when pigmented showed some tendency to bloom, which apparently is due in part to the kerosene content of CVR 973 and 974. The four varnishes were also used in a Semi-Lustre formulation but in all cases were found to be too low in sheen. The Vehicle Research Department is now planning to make a 25 gal lon P.E.-Rosin-Dehydrol-C.W.O. varnish by the solvent process for use in Enameloid and Semi-Lustre formulations. The use of part Dehydrol may improve the gloss characteristics of this varnish. 0007-SWP-045029 0007-SWP-000114400 -&7- #162 - Cont. Porch and Floor Enamels A sample of ether resin, varnish - RB 7019, submitted by the Resin Research Department, was used in a formulation for Porch and Floor Enamel. The enamel based on this liquid had excellent scrub resistance and elasticity after one month. The ether resin enamelj however, was very short in flow, and any improvement that can be made in the flow characteristics of RB:7019 would be advantageous,' No work was done during the month on the following projects: #77 - Mildew Inhibitors #160 - Painting of Bricks #183 - Pretreatment of Metal Surfaces for Painting #148 - Examination of White Lead Discontinued 0007-SWP-045030