Document Vbnym3yxrYO3wNY75ya1bD1j

E. I. DU PONT DE NEMOURS & COMPANY KREBS PIGMENTS DEPARTMENT 256 YANDERPOOL STREET NEWARK, NEW JERSEY Serial No, Copy No. i. KN-51-40 Humerical NEWARK PLANT PIGMENT COLOR RESEARCH REPORT BEHAVIOR OF COPPER PHTHALOCYANINE PIGMENTS IN PAINT SYSTEMS 1 III. ACID PASTED CHLORINE-FREE CPC - CRYSTAL GROWTH AND FLOCCULATION OF THE ALUMINUM BENZOATE LAKE IK A LACQUER SYSTEM Period Covered AUGUST 1949 - AUGUST 1950 FILE: DATE: 223.4 10/5/51 NJ 5747 Serial Ho. KN-51-40 Copy No. 1. Numerical Copy to: #1 - Numerical Pile 2 - Research Office (223-4) 3 - library Pile (223.4J 4 - J. N. fully 5 - W. P. Spengeman 6 -A. R. Hanke 7 - Ektra 8- 9- ** * 10 - 11 - " NEWARK PLANT PIGMENT COLOR RESEARCH REPORT Progress Report BEHAVIOR OP COPPER PHTHALOCYANINE PIGMENTS IN PAINT SYSTEMS III. ACID PASTED CHLORINE-FREE CPC - CRYSTAL GROWTH AND FLOCCULATION OF THE ALUMINUM BENZOATE LAKE IN A LACQUER SYSTEM AUGUST 1949 - AUGUST 1950 CHARGE: 1101-11-20 SUBMITTED BY: APPROVED BY R. HANKS DATE SUBMITTED: IO/4/51 DATE ISSUED: IO/5/51 DUP050068655 X. jmmSSSA A previous report. KN-5Q-1S, made the observation that lose In CPC pigment strength for lacquer ana alkyd systems is due to both changes in degree of flocculation and to crystal growth. Flocculation plays such an important role that in all cases investigated except for the acid pasted chlorine-free CPC. it completely overshadows the effect of crystal growth* this present investigation wee designed to show that some separation between the effects of flocculation and crystal growth could hs had by working with the aluminum bensoate lake where the flocculation effect was slight. This would lesd to a clearer under* standing of the role of crystal growth in the loss in pigment strength* A typical acid pasted chlorine-free CPC aluminum bensoate lake was used for this experiment. XX. CQSCLUglOHg 1* Crystal growth of the aluminum bensoate lake of an acid pasted chlorine-free CPC in a lacquer system at 50*C is vary pronounced and, over s period of 2 weeks, can account for a loss in pigment strength of around 75%. The degree of flocculation of the aluminum bensoate lake is always low but sufficiently high to measure. It decreases during the early aging of the lacquer (8 weeks), after which it Increases. This follows the pattern of the other CPC pigmented paint systems reported in KK-50-18* There is some evidence that In the presence of Ti02 extender crystal growth is inhibited. III. 1 A. 1. SU mmsmsumm CPC lacquer A pulp sample of Orchem'e J4DD-1H3 Press Cake D-6624 was con verted to the aluminum be - - --- normal procedure for BL-2 dry bssis) was ussd. The A KV-36 lacquer grind was prepared according to the following outline. Site of mill 1 gal Pigment (1285-80 210 g Grinding Vehicle (KV-36-GV) 1125 g Thinning Vehicle (KV-36-TV) 525 g Steel Balls (Siae 3/l6) 9300 g Ball miU for 72 bra Add KV-36-TV 1035 g Strain off balls and .|i#e :In two 1 qt. cans* DUP050068656 2* 2. tiOg Lacquer The white pigment used wee "Ti-Pure", R-610 grade* the lacquer wee made according to the following outline. Biss of mill 2 .gal. TiOg ("Ti-Pure", R-610 grade) 420 g Grinding Vehicle (KV-36-GV) 2250 g Thinning Vehicle (KV-36-TV) Porox balls (3/16") 1050 A quantity sufficient to break the surface of the liquid Ball mill for 72 hrs. Add KV-36-TV 2070 g Strain off balls and store in one quart cans ' B. STORAGE Of MCQUB1S, PREPARATION OF EXTENSIONS AND PANELS The CPC lacquer wee stored in two one-quart cans at room temp erature for three weeks and then at 50*0 (122*F) up to a total time of 26 weeks. The TiOg lacquer was stored at all times at room temperature in one quart cans. From time to time, samples of the CPC lacquer were withdrawn and extensions made with the Ti02 lacquer. Before any lacquer was withdrawn from any of the cans they were first shaken from 10 to 30 min* in a shaking device called a "Paint Rejuvenator", manufactured by the Miracle Paint Rejuvenator Co. After the cans were opened, the contents were inspected for uniformity by manipulation with a spatula. Good uniformity was always found. All extensions were made in duplicate. To make the extensions, a 22,24 g portion of the CPC lacquer was mixed with a 100.0 g portion of the white lacquer. This gave a 10/90 toner basis extension (20/90 on lake basis). The mixing was accomplished by first stirring with a "Lightnln" mixer for 15 min. and then shaking for 15 min. on the "Rejuvenator". This shaking and mixing scheduls assured a uniform extension and assured an unchanging composition of the lacquer in the can which meant a reliable estimate of the GFC/TiOo ratio. It is true, however, that small variations in the CPC/T1Q2 ratio can be tolerated because of the spectrophctometrlc method of expressing the results. These extensions were thinned to spraying consistency (1:1) with T-3601, Primed aluminum panels 3" x 7" were sprayed in the conventional manner. An attempt was made to keep spraying conditions reasonably DUP050068657 3 constant throughout the experiment. After the sprayed panel had dried, a portion of the spatula-stirred thinned lacquer was poured over the panel, covering approximately 1/3 to 1/2 of the panel, The panel was then tipped on edge and the exeess lacquer allowed to drain Off. This is the conventional practice for making the so-called "pour out" test. This surface was then allowed to dry. C. OBTAIHXBS LIGHT m A 0.B. self-recording spectrophotometer was used. The reflec tance was recorded relative to the National Bureau of Standards Vitrolite White. All reflectance measurements were made exclusive of the specular beam. On the O.B. Instrument, this is accomplished by inserting black velvet cups in the integrating sphere* The complete curve from 400 nju to 700 nju was obtained for all panels but for calcu lations, only the reflectance at 463 mu and 616 mu was used. These reflectances were corrected to a MgO white basis using the conversion factors supplied by the Bureau of Standards. Curves Were obtained for both the sprayed portion and the poured portion of the panel. D. CONVERSION OF REFLECTANCE CURVES TO FLOCCULATION AND POSTAL.GROWTH,.DATj] 1* geasurameq^^ For the measurement of flocculation, the spectrophotomstrio curves of both the sprayed and the poured portions of the panel were used. If no flocculation existed, both surfaces would yield the same curve and the flocculation index would be aero. If crystal growth had occurred, it would not become a part of the flocculation measurement because the crystal site would be the same for both the sprayed portion and the poured portion. If a flocculation difference existed, the flocculation index - b was calculated from the following equation, the development of which is explained in KN-50-15: A b rn 30 A <f>2 *' & $1 A b * a term which increases in magnltud# when there is a relative increase in average particle size, in this case an increase in flocculation. 30 * Estimate of the constant slope of the straight line obtained by plotting as ordinate and <f>2 as abscissa for a series of extensions. a Si * ' the difference in cf?i values derived at a wave-length of 465 sax between the sprayed and poured surfaces DUP050068658 5 Af:2 the difference in values derived at a wave-length of 616nju between the same two colored surfaces* Note: The subtraction of <p values of the two colored surfaces to obtain A4> was always made poured minus <Kfc where E. is the reflectance exclusive of the specular reflectance and corrected to KgO white as the reference white* The following is a typical calculation using data derived from the speotrophotometric curves of one or the panels* Panel ObrlfrPg A1 Ufrarote. 1X1mSSEZiT~3p. Aged^ThrfleWegks % E %RtCorr. P''. Poured Sprayed 57.9 50.A 57.0 49.6 4.10 10.2 6.69 1*21 9.6 6.36 0,214 0ol99 t> +0.19 +0.015 A b - 30 A&, * A A b 30 (0,015) - 0.19 b m +0.26 Notes A large positive value indicates considerable flocculation. A low or negative value moane very little flocculation. In this case, the numerical value of +0.26 means very little flocculation. 2. Measurement... of Crystal, Growth With age of the lacquer, the appearance of the sprayed portion of the TiOo extension changed. Both flocculation and crystal growth can be responsible for this change. With an aluminum benseate lake however, the flocculation is nail and the bulk of the change can be attributed to crystal growth. Per this experiment, for the Sided of simplicity, the index derived from s comparison of the initial panel with with the aged panel will be called the "crystal growth* index* The same function is used to calculate the crystal growth index as that used to calculate the flocculation index but instead of comparing a DUP050068659 -5 sprayed with a poured surface, the surface of an aged lacquer is compared with the surface of a fresh lacquer* To distinguish the index from the flocculation index, .it will be called and the equation v/ becomes Al'*2' - |0 AfZ - 491 share A<p , is derived from the "aged" measurement minus the initial*' measurement at 465 mu, and &<Pz is derived from the aged measurement minus the initial measurement at Olornn. The following is a typical calculation from data derived from the speetrephQtometrlo curves of one of the panels* gpraved Surface ..A ..-t, W-Jft------- - A...4...6l.6.;ma . UL ll ... 5 wks. aged lacquer 58.7 Initial lacquer 57.5 51*2 50,0 4.30 12*3 10*70.269 kM 9.4 8.2 SUM. J A<pi * o.30 A<fe;o.074 A/Cm $oA<p2 - - 1.92 Kotes A large positive value indicates considerable crystal growth. In this case the numerical value of 1.92 indicates a moderate amount of crystal growth, a. Both the flocculation data and the crystal growth data are recorded in Table 1. gabk.1 CPC - A1 Bens, hake Tima (Weeks) 0 3 Lacauer Flocculation Crystal Growth ... . 0.26 ............ Index oTJl (Boom Temp* storage (from 0 to 3 weeks* i l *! v 26 -0.25 -0.20 -0,33 -0.82 -0,84 0.20 0*36 1,92 3.59 5.06 5,59 7.24 8*71 4.42 After 3 week reading i subsequent storage was at 50*C. Second fan of TiO^ used, Second Gan of OPC used* / a-:: '* ^ & '. ' DUP050068660 These data are derived tron reflectance curve# of the id extension all not the aged masatone. The eeaential erence between thie and the other 26 week panel is that In one ease the CPC aged in the abeence of the ex tender and in the other eaee it aged in the presence of the. extender* Pig. 1 plots the data of Table 1* A detailed discussion of this plot will be deferred to the section on general discussion. At intervals during the aging of the enamel, electron micro graphs were obtained of the unextended lacquer* These are Hated: below and are illustrated in Figures 2 through 5. They will ha discussed in greater detail in the next section. Pig. 2 Plate 1152 A^Bg,gc,hgjiHie,, ptMmmt 19 days at room temp. 3 * 1161 3 weeks at room temp. + 2 weeks at 50*6 4 1167 3 * at room tenp* + 3 weeks at 50*0 5 1233 3 * at row# temp. + 23 weeks at 50*0 17. .d is c u s s io n Two indices have baan defined (see section III-D) baaed on light reflectance curves. One index, the flocculation index, is bawd on tha difference between the curve appearance of a sprayed and poured lacquer surface. The seemingly complicated method of expressing the index it justified because when so expressed it is insensitive to small changes in retie of CPC to TiOg This mskss unnecessary antram pre cision in making extensions for comparison purposes. Unfoforrttuunnaattely this index is not completely insensitive to factors other than changes in floccuulation of tha CPC. It is probably somewhat sennssiittiive to changes in flloocculation of TiOo With regards' to,.fl , it is reasonabllyy inseennssiittiivvee, to channggeess' in ratio of CPC to TiOg brought about t flooding but if the flooding mechanism causes a separation of f: particles from course particles as wall as a change in pigment ratio, the magnitude of the index will be At best, it -must be rscognised that tha index measures a several factors but these self-same factors are also Ins visual interpretation of flocculation* In other words, "floe .. is fairly rigidly defined as "the formation of clusters of padpteles DUP050068661 7 separation". This "flocculation index" takaa a stop in tho dlraotion of achieving soma separation of those other phenomena hut the separation ie by no means complete. Recognising that other factors do have sosa effect on the magnitude of the index, for the sales of simplicity, it *111 still he called a "flocculation" index* The other index utilised in this report is the se~oalled "crystal growth" index* It is also derived through comparison of two colortd surfaces hut, in this case, the eolorod eurfeces ere sm initial end en aged surface* This index ie calculated in the Sene nenner es the "flocculation* index end like it, is relatively insensitive to assail changes in tho ratio of CPC to TiOg* Tho crystal growth index is ad mittedly also dependent on changes in degree of flocculation but in this experiment flocculation is low and crystal growth ie high, furthermore the act of spraying tends to give e deflocculated end rapid drying surface which will farther minimise the already small flocculation effect. Because the other factors effecting the index are mil, for the Sake of simplicity it ie justified to call this tho crystal growth B. immm. figure 1 gives a complete description of tho aging study of this lacquer system* In Interpreting the two curves it must he borne in nind that the a h values for flocculation hoar no direct relationship to tho A c values for crystal growth* Numerical values ere net to bo compared hut it ie valid to compare the directions taken by the curves. It should also he recognised that the precision in the flocculation measurement ie poor, being around 0*| A b and tho values, which eve email, are simply used to indicate a trend* nils they vdry clearly do* The precision of the crystal growth measurement ie somewhat better, being around 0.2 Ac* Referring to figure 1, it ie seen that at seem temperature aging up to three weeks, there ie very little evidence of septal growth although flocculation le falling off. Tho electron micrograph (fig. 2) also mo w s no crystal growth. At the end of three weeks, 50*0 aging wee begun end the crystal growth curve takes e sharp turn upward. Its electron micrograph (Fig* 3) now shows evidence of crystal growth. The flocculation index is still felling off. It le evident from e com* pariaon of tho two curves where one le rapidly rising end the other le falling off, that tho change in flocculation is not greatly influencing the crystal growth index* This is one of the assumptions made, end hence, for this experiment, it appears justified, this is undoubtedly due to the feet that the crystal growth is so very much greater thin the change in flocculation. Tho stoop rise in the crystal growth curve continues until shout & weeks aging at which time It tends to level off. The electron micrographs also MOW that crystal growth is quite active over that period of time. In comparing the micrographs it should ho kept in mind that fig* 2 ie at 35,000 X magnification whereas the remaining micrographs ere at 3,500 X magnification. DU P050068662 a In the interval of 5 to fl weeks it became necessary to start examining fresh cans of lacquer. The 7th week readings and all sub* sequent readings were made on tints using the second can of TiOo lacquer. For the 6th week readings, the second can of CPC lacquer was started. Fig* 1 shows the 6th week readings for both the first and second cans of CPC lacquer. From Fig. 1 it can be seen that the flocculation index was not affected by the change of cans. The two points are well within experimental error. This is not the case, however, with the crystal growth Index. The second can of CPC lacquer showed a significantly higher crystal growth index than tha first can. It in entirely postil that the two cans got off to a different crystal growth start. This can be expected when comparing crystal growth in two isolated systems. This would be particularly noticeable in the steep portion of the curve. At the termination of the experiment after 26 weeks aging, a panel was prepared of the Initial tint which had been aging at the same temperature as the masstone lacquer. This enabled a comparison of the effect of age on a CPC lacquer in the presence and absence of the TlOg extender. Figure 1 shows these points. At 26 weeks, the aged tint has about the seme flocculation index as the fresh tint. The difference in the position of tho points is too dost to the precision of the measurement to be considered significant. The crystal growth index, however, presents a different picture. The aged tint shows far less evidence of crystal growth than the aged maastone lacquer. This strongly suggests that the presence or TiOg inhibits crystal growth. There can be no doubt that the major change in this lacquer system is one of crystal growth with the ettendent loss in pigment J strength. The Ac index is not to be considered as s simple function of pigment strength, so strength in this cast had best be judged by visual inspection of the panels. Such an inspection indicates that over the period of 26 weeks, the strsngth loss has been at least 75%. Work has already been completed on a similar investigation using an acid pasted LB type and an add pasted ssmi-ehlor type of CPC in place of tho add pasted chlorine-free type used for this report. These types have also been observed to grow crystals in paint systems but to a lesser degree than the acid-pasted chlorine-free type. The results will be given In another report. ACKNOWLEDGEMENT The preparation of all lacquers as well as the preparation of all extensions and the spraying of all panels were donelby the "Pigment Evaluation Group" under the direction of Mr. B. J, Godfrey. AHHslg Atf*'' A. B. HANKS CHEMICAL DIVISION DUP050068663 Lacquer aged 19 days at room temperature Plate 1152 35,000 X DUP050068664 Figure 3* Lacquer aged 3 weeks at room temperature + 2 weeks at 50C Plate 1161 3,500 X DUP050068665 Figure 4* lacquer aged 3 weeks at room temperature + 3 weeks at 50G Plate 1167 3,500 X DUP050068666 Figure 5. Lacquer aged 3 weeks at; room temperature + 23 weeks at 50G Plate 1233 3,500 I DUP050068667