Document gEQXVJbQ58qaK32KVzB4rdm5e

E. I. DU PONT DE NEMOURS & COMPANY KREBS PIGMENTS DEPARTMENT 256 YANDERPOOl STREET NEWARK, NEW JERSEY CONFIDENTIAL Serial No, Copy No. 1. Numerical Pile NEWARK PLANT O0 PIGMENT COLOR RESEARCH REPORT \ Pinal Itsport i ULTRA~IOLET TRAPS;,ilSGIDN p? DISPERSED HS~D50U5 OXIDES Period Covered HAT 18?0 to AUGUST 1950 T NQs NJ 8747 FILE: DATE: 210*2 itaroSi 15; 1951 Copy tot #1 * Buaerleal Silo 2 * Research Office 3 - Library Filo 4 * V* Clmlupaki t " 1.1 Pciltoii/yr Be. Oallaw^ 11 S: i* stmie/PJii 8 - D. B, KilliaVA. Siegel Serial .So* KJr-Sl~I3 Copy So* 1, Numerical Pile miss p l a h t PIOSENT COLON SESEAECE REPORT Pinal Report ^gM^YIOLET SS3T I960 to ACCOST 1930 CBABGEs 1101*11-958 SUBMITTED B3T* APPROVED m l DATE S *1iV.tvmtn*t Jan. 51# 1951 DATE ISSUED* March 13, 1951 DUP050068544 mmomGTxmt ^ The suggestion that hydrous nickel oxides improve paint film durability by absorbing harmful ultra-violet light prompted this investigation, Thie report eovere the observations of a wi* limlnary nature that vere made on the ultra-violet light transmission of various hydrous oxides* So claim is made that the ^transmission curves Obtained hate a high precision or that the regions of lee light transmissions found are due entirely to light absorption. Liriit scattering may also bs a part of tbs explanation for lorn light transmission* SB spite of the uncertainty introduced by not being able to distinguish bstwsen light absorption and ligit scattering, the data ere still useful in that they establish limiting values for the ultra-violet absorption of tbs pigments. Thus, if consider able ultra-violet light smattering did exist, the oxidss would be even lees effective than the data show them to be* ftgamHY a h p ammimi 1* Light transmission data were obtained on the following eater dispersions of hydrous oxides* iron oxide (XXI) Chromium oxide (III) Titanium oxide (IV) Hotel oxide (XX) Aluminum oxide (XXX) 8, For all oxldee except hydrous aluminum oxide, a marked deerease in transmission was noted as the wave-length decreased, the transmission passing through a low minimum* All these strong minima, however, were well on the short wave-length aide of the lower limit of ultra-violet light in sea-level sunlight. 8* Following is a list of the hydrous oxidss in dsoreaslng ____ of effective removal of TJ.V. light* Iron (XXX), Chromlua(XXX), Titanium (IV), Hotel (II), Alumlnum (XIl) It is known that from4 the point of viow of enhancing film durability on hydrous titanium oxide are poor* Hydrous nioksl oxids is far better tiiaif hydrous titanium or hydrous aluminum oxide* Frew above list it would appear that althougi the difference between iron aivmtvnvm wiigMt have a partial 0jpl8&&tlwi in H.v* absorption, the unusually good behavior of hydrous nickel oxide is due to oauaes other than TMT. absorption. A* general ATuiftHmaw Model D* V* nsetposhotosittp us used for --tv tr&nsmlaeicn measurements. Aqueous suspensions wars plaeed in 1 centimeter lightpath quarts cells. DUP050068545 g Ha most s u m the dispersed hydrous oxides were obtained by vapid mixing of a dilute water solution of tbo natal chloride with a alight axaaos of dilute sodium hydroxide. light transmission data nave Obtained Immediately thereafter before the hydrous oxides had a Shanes to eoagulate to the point there it seriously altered the light transadeaion value* in the ease of hydrous nickel oxide* an aqueous suspension of a plant lot U-2-P, lot 74910) naa alao run* ' In addition to this, a Med film prepared from thia aeon lot naa ***nwesmm. '^Aaspeinls* fAcvmVewmae |rAtm^f **1* semawa^^*Bi* tfw| a # aweuf ^ssw ^mwsj**.w* Thera naa enough difference between 'the ourves however* to auggeat that the degree of diapersion does alter the oum somewhat* which amla4a &m> <imAwtawon rtf flttftOhllSff (ffilw a Wrtiifl*i aMBtitfttlVA significance to tha data* The curves are shown in Figure 1* A solution of HiClg.eEoO naa preparad (0*8 gram/100 milliliters)* The C.F. salt was assumed to be 100$ H101g.e%0. A oauatio solution was prepared by dissolvli** 0*816 grama sodium hydroxide in 100 milliliters of boiled distilled water* ibis was analysed by titration with standard eulfurlo sold whereupon the concentration was calculated as 0*499 grams sodium hydroxide par 100 milliliters* A 10*9 milliliter portion of the niokel ohloride solution diluted to 100 milliliters and a 4*8 milliltfir portion of tin sodium hydroxide solution was diluted to 100 milliliters* The solu tions were rapidly poured together into a third beaker with agitation* Transmission data were obtained immediately thereafter and the curve is shown in Figure 1* of reagents was ehosen so that the hydrous niokel oxide concentration* assuming a formula of was 0.01 graw/100 milliliters in a m% excess of caustic* Hi'(G~ H),M ulatlan of the hydrous niokel oxide* light transmission m were also Obtained m the supernatant liquid to shew that the Observed decrease in transmission mm due to Idie hydrous oxide and not to the dissolved salt* These measurements show considerably m light transmission and in no way resembles the curve of the hydrous nickel oxide* TransmisaIon measurements were also obtained of a *1 rtV*i chloride solution of *w* aouivalant to iwntetam nickel oxide dispersion* 'The llght transmlseion measuremsnte of the solutions were all very dose to 100gt all of which shows that the transmission curve obtained of the hydrous niokel oxide dispersed ia the eolution was due to tho hydrous oxide and not due to niokel salt or anything alas 3m solution* 2* A plant batch of hydrous nickel oxide (A-2-P* lot 74919) wae washed with water to the point Where the pressoake began to peptize* A portion of this peptlsed suspension analysed 5*1# total DUP050068546 solids and was further diluted by dispersing a drop ( 0,05 milliliters) in approximately 3$ milliliters of _ gave a well dispersed suspension with very little tendency to settle and of Idie right concentration for transmission measurements* A light transmission our** was obtained and found to be similar to the emm obtained from the hydrous oxide directly formed. Figaro 1 . shows the ourve after recalculation to a mors comparable comesntratl basis for comparison with -the curve of the directly formed hydrous oxide* To make mis calculation it was assumed that the optical da city C-Xcg T) was proportional to someemtratism* 3* Ths seme peptized pulp of hydrous nickel oxide wee used for this work as was used for the experiment described in the prettes section* The original dispersion of approximately S*l solids wsi diluted with water (29.4 grama to 100 grams) to make a slurry of ftp* proxismtely l.S$ solids* A piece of clear flat quarts was dipped in and out of the slurry* one side was wiped clean and the film on fcht other side was allowed to air dry. A light transmission curve Obtained of the dried film adhering to the quartz slide* Tl was rscaloulated to a concentration basis comparable to the hydrous nickel oxide ourves fir comparison purposes using the described in section 2, and is illustrated in Figure 1* It can be seen that all curves of are essentially the same*regardless of the hydrous oxide or Obtaining the curve* ffcer however, which may well have to do with the the oxide* oxide of M>* . ,i differences, of dispersion of A stock solution of C*P* GrCls*lG BfeO .(0*5 grWlOQ millllitero) was prepared* The celt was assumed to be loo Ck*01**10 H20# a 13*1 milliliter portion of chromium solution was diluted to 100 nHn* liters* A 4*3 milliliter portion of the sodium hydroxide described in B*4|: was diluted to 100 miHilitera. The sal rapidly poured together with agitation into a third beaker end trans- IS*a light tranaaiesion curve Is illustrated in Figure' 2* 0* A stock solution of titanic chloride was prepared by adding 3 -milliliters (3*2 grams) of anhydrous TlCla to 12 grass crushed ill end * milliliter* coiwwitrated hydrochlwio acid, the volnmi aftw the iee had melted was. 15 milliliters* This solution was diluted in two steps** A 10 milliliters portion was diluted to 100 milliliter** giving a TiClg concentration of 3*5 gpma/100 milliliters* This solu tion was further diluted, 10 milliliters to Iters* to give DUP050068547 * 4 - a final 5*1014 concentration of 0*55 gro./lOO milliliters* -w solution remained clear during the time it was used tut with am* night standing, it became cloudy* A 9.4 milliliter portion of the TlCla solution urns diluted tolOO milliliters* A 7*0 milliliter portion of the sodium hydroxide solution described in B*1 was diluted to 100 milliliters. ffae solutions were rapidly poured together fete a third beaker end transmission nsasursnouts imps Ob** talned immediately thereafter* ' The concentration of the solutions was so ehoaen that the final solution had a eouoentratien of 0.01 gram/100 milliliter TifUf^# The light transmission curve is shown in Figure 8. A stock solution of ferric chloride was prepared by Resolving ouo Fe01 in water to give a concentration of 0$ graa/lOO ailliliters* A 10*85 mwiilllliliter portion of the FOCI* solution was diluted to 000 milliliters and 4*5 milliliters of the sene eodlua hydroxide solution described in B*-l -wan dilated to 800 mi111liters* He sola* felons were rapidly poured together into a third beaker and agitated. Transmission measurements were obtained immediately thereafter* the concentration*of the solutions were so chosen that the final solution had a oonoentration of 0*005 graas/100 milUliters Fe(OH).. His is only one-half as strong as previous concentrations because it was Ob* served that at concentrations of 0*01 gram/100 milliliters, the hydrous ferric oxide flocculated too rapidly to permit the desired measure* 'light transmission) ourve is &o*n in Figure t* A stool solution of aluminum chloride was prepared by dlsaplv* ing anhydrous Aids in water to give a concentration of 0*49 gram/100 milliliters. A 7.4 milliliter portion of the AlClg solution was diluted to 800 milliliters and a 7*8 milliliter portion of the sans sodium hydroxide solution described in B*1 was diluted to 800 mi111* liters. Both solutions were rapidly poured together into a third beaker with agitation* Transmission measurements were obt* ` immediately thereafter# He final o<BsoentrafeIon was 0*008 milliliters VA1(GH)$. A stronger concentration caused He ..oxide to flobeulate too rapidly* He .Jagptftransmission ourve le Hewn in figure 8* Admittedly the concentration of the dispersed hydroxide is only an approximate figure, hence the calculation of the absorption coefficient ie only approximate* but it earn bo ueod as a of the effectiveness of ultra-violet removal " with the hydrous 'oxide* It Is likewise true to toll through a measurement of transmitted being removed by absorption or by scattering* It was assumed that the light wae absorbed* If scattering features to a significant extent* ana effectiveness of the bvdrous t removing ultra-violet h w Ims DUP050068548 is even less than the calculated figure* hence the calculated figure an be considered ae a sort of limiting value, the true value being cue making the oxide an aim less efficient remover of ultra-violet light, it can he seen from Figure 5* that below 500 millimicron* there ie no ultra-violet light in eea level sunlight. By referring to Figure 2, it can be soon that the region of strong absorption for. the hydrous oxides ia on the abort wave aid of the noiah below which there ie no ultra-violet light in sunlight, Above 500 milli- nlcrons the intensity of ultra-violet flight in sunlight inoroasOs rapidly and for the aake of' making comparisons between the oxldea studied, a wave lmgth of 560 millimicrons was ohosen ae being wuffi* ciently representative of the ultra-violet light in sunlight, A typical dried pigmented paint film contains about IS# pigment solids. Using reasonable values for the specific j the hydrous oxide, it becomes possible to calculate the film required to cut the incident light intensity at 560 oillimioroas wave length to anv oredetexvilned value h mMs , use of the Beer-Lambrt lay* A table showing the film thiokneaa required to out the incident light intensity to 10# of its incident value will give sens idea of hew muoh protection the hydrous oxides studied will give from (die photochemical effect of ultra-violet light. Following is a typical calculation showing the method used. Caleolation for Brapoua KiakalOitlfla The Beer-Lambert law is usually expressed by the following set of equations. D * otk T * Transmittance measurements obtained from the curve. It ie $ f $ 100, *o * **<****> intensity, I * Transmitted light intensity* D * Optical desnlty defined by the equation -log ^ d . o ** Pigment concentration (grams/100 millllitera) t Optical light path (centimeters) k at Proportlonality eoaltaht usually called absorption coefficient} It is the optical density par unit concontration per unit light path, of Radiation Energy*, lat ed*j DUP050068549 * d the sjpeetrophotometrio * e log f * l> hydratu w4 wtw>i * log 0.96 0.0177 * D & D .: m kw 1.77 If tho pas is to to reduced to 10$ of its incident intensity .In the paint fils, J/l lOAOO * 0.1 log 0.1 m 1 m (the required optical density) Assume a specific gravity for hydrous nickel cuddo of 4.1. lot the pigment concentration (g/100 ml) in a 15 volume % dried fils o. 0 * IS * 4.1 * 61.5 e/100 nl. Let t fils thickness required for on optical density of 1. * ' S? " b u s 1! t ;w * * 01088 s" $o convert to mils - 0.01088 cm x 503.7 4,3 sdt* This distance represents the fils thickness necessary to decrease the light Intensity to X0 of Its incident value. The following table lists the calculated fils thicknesses for the oxides studied. Xros (in) ________ ,111) Titanium (17) Hlckel (II) <m) Specific aenttL. 3.46 - a.S 3.6 4.1 6TOHoail6OlaoikOuMKlaSiatscgsd.I/TfHOiJlUisl*t 1 0.05 0.8 8. 4. v. much 1arger Even though the fils thickness values are adslttedly only actions the difference between the oxides is large enough to bach some quantitative signlfioanoe to the data. It is seen that according to the table, the hydrous oxides of aloalnus and nickel should be much worse in improving film etsbillty than Iron. It is DUP050068550 true that alualtma is much versa, and TJ#V# absorption may be* at least la part, responsible but tbs very good behavior of niokel as compared be titanium sad alumlmaa leads one bo believe that the unusually good behavior of hydrous nickel aside is due bo causes other than TJ#V# absorption* this ork is reseeded ia HB*13S6~10* ARHllg DUP050068551 lOOimiTT w 20 40 60 80 W 20 40 60 80 WAVELENGTH IN MILLIMICRONS DUP050068553 DUP050068554