Document MJV6dLgyq82OOLzG8pVyYgoyM

"M onastral" P lie s - Technical Corresp 'LITHOSOL" PAST VIOLET 4RNL PASTE E. I. DU PONT DE NEMOURS & COMPANY ORGANIC CHEMICALS DEPARTMENT CHAMBERS WORKS DATE OP ISSUE; MAY 4, 1964 COMPANY CONFIDENTIAL C.W.P.D. REPORT NO. 77 uLITHOSOLn FAST VIOLET 4RNL PASTE ^~f?LITHOSOLlt PAST VIOLET 4RN PASTE^^ Prepared by; Jfr TT iij Jw q?txT\' S. P. Whearty Supervisor; H H. Hully tP. ABSTRACT The results obtained in a series of investigations undertaken for the purpose of improving these colors In quality and yield are summarized. DUP050467824 N33028.01 TABLE OP CONTENTS Introduction Results Part I - "Lithosol" Past Violet 4RNL Paste Introduction Experimental Part II - "Lithosol" Past Violet 4RN Paste Introduction Experimental References Appendix 1 Page 1 1 3 3 44 44 DUP050467825 "l it h o s o l " p a s t v io l e t 4r n l p a s t e "LITHOSOL" PAST VIOLET 4RN PASTE Introduction The overall purpose of this investigation was to Improve these colors In quality and yield. This report is presented In the form of an outline of the work undertake^ and a summary of the findings. Recommendations, when appropriate, are made as each phase of the study is discussed. Results The objectives of the study were only partially attained. A number of important process improvements were made and several promising leads for future investigation were uncovered. However, the process conditions necessary for the production of "Lithosol" Past Violet 4r n Paste consistently equal to standard in shade still remain, to be established. , The following is a list of the important accomplishments of the study: 1. A process using filter cake in place of dried, ground isodibenzanthrone was developed for the chlorination step in the production of "Ponsol" Brilliant Violet 4RN Crude Wet. Savings are realized through the elimination of the drying and grinding procedures for the intermediate (see Part I, Section 4). 2. Material failing to meet standard in the solvent bleed test for "Lithosol'* Past Violet 4-RNL Paste can be improved satisfactorily by rebleaching or by a purification consisting of a filtration from the reaction medium at the chlorination step (see Part I Section 7 and Part I Section 15a). 5. The optimum :chlorine content of the crude color and the optimum conditions -for bleaching "Lithosol" Past Violet 4RN Paste Crude Acid were established. As the study progressed there 'was..-some evidence that the bleach conditions used for the crude color could be. modified or the step eliminated entirely (see Part I Sections 11 an<jU15). DUP050467826 4. Water and iron salts as contaminants in the chlor ination medium are detrimental to color quality* Water in rela tively large amounts acts as a chlorination inhibitor (see Part II Sections 5a and 3h). 5. An increase in the amount of dispersant used In the regular', milling during standardization of "Llthosol" Past Violet 4RNL Paste resulted in a marked improvement in hue and brilliance of final color (see Part I Sections 10 and 19 but also note Part II Sections 4 and 6Aa). , 6. "Lithosol" Past Violet 4RNL Paste can be standard ized directly from selected charges of "Ponsol" Brilliant Violet 4RN Crude Bleached. Also, color prepared from selected crudes by the viscous milling technique used In the "Lithosol" Past Violet 4r n Paste process is satisfactory for the 4RNL type. Es tablishment of either one of these procedures, which would be de pendent on the production of uniformly high quality crudes, would result In substantial savings through the elimination of drying, acid-pasting and bleaching steps (see Part I Sections 15j 16, 22 and 24). 7. A statistical study indicated a serious lack of re producibility in the testing procedure used for evaluation of "Lithosol" Past Violet 4RNL Paste quality (see Part I Sections 17 and 23). 8. A viscous milling procedure resulting in improved quality and considerable savings was developed for "Lithosol" Past Violet 4RU Paste (see part II Section 4). 9. Impairment of color quality of ,fLithosol,r Past Violet 4RN Paste, as indicated by a shift in shade from red to blue in paper dyeings may occur with certain physical treatments including very prolonged viscous milling and acid pasting (see Part II Section 6a ). In work of a preliminary nature a digestion procedure was developed which reversed this shift in shade. Correlations were shown between shade on paper and the appearance of diffraction rings produced in electron microscope studies of the color (see Part II Section 6b ). 2- - DUP050467827 PART I "LITHOSOL" PAST VIOLET 4RNL PASTE Introduction This color is marketed for the pigment trade and is used in paints, varnishes and enamels. Late in 1956 it was reported that material then "being produced was unsatisfactory because of weakness and dullness in lithovarnish rub-out testing. At this time the color was made according to the following procedures: 1. Suspending dry isodibenzanthrone in o-dichlorobenzene and chlorinating to the dichloro derivative with sulfuryl chloride in the presence of iodine as a catalyst. * 2. Removing the reaction solvent by steam distillation and isolating the crude color from the steam still slurry by filtration. 3. Purifying the crude color by bleaching with hypochlorite. 4, Effecting a further purification by acid-pasting the dried crude and bleaching again with hypochlorite. Some time prior to this period a change was made in the procedure for the preparation of isodibenzanthrone. This consisted of the substitution of methyl alcohol for ethyi alcohol in the alcoholic potassium hydroxide fusion melt and was made for reasons of economy and convenience (Ref. No. 1). See Appendix 1 for an outline of the steps involved in the current production of this color starting with benzanthrone. Experimental To correct the above mentioned deficiencies in the final, color and generally to improve the processes involved, the follow- . ing studies were undertaken: -3- DUP050467828 1. Reevaluation of methyl alcohol as a substitute for ethyl alcohol in the process for isodibenzanthrone crude. 2. Purification of the isodibenzanthrone crude by vatting. Purification of the isodibenzanthrone crude by acid crystallization. . 4. Rebleaching and improving the crude color bleaching procedure. 5. Using anionic, cationic and nonionic surfactants in the acid-pasting drowning water. 6. Evaluation of variations in the chlorine content of the crude color. 7. Using the isodibenzanthrone as filter cake and dehydrating the o-dichlorobenzene suspension prior to chlorination. While tii'ese studies were in progress' a new analytical procedure, consisting of a spectrophotometrie analysis, was der-. veloped for isodibenzanthrone (Orchem Method 52-5-11-1). With it isodibenzanthrone content (purity) and dibenzanthrone as the principal impurity can be determined. Although this procedure is an improvement over the older method (acid pasting recovery) it can account for only 70 to 80$ of the material present and is thus of questionable merit as a research tool. The results obtained in these studies are summarized as follows: 1. Replacement of methyl alcohol with ethyl alcohol in the fusion melt resulted-in some instances in improved isodi- benzanthrone and in final color. Although the findings were not conclusive the ethyl alcohol procedure was reestablished as standard practice (Ref. No. 2). ' It is recommended that the return to the use of methyl alcohol be considered, particularly- if an-improved analyt ical method for isodibenzanthrone'is developed. -4- DUP050467829 2. Isodibenzanfchrone purified by yatting tended to give improved color by lithovarnish rub-Outtesting (CWNB 1696 pages , . 30-55)* Plant trial runs demonstrated that this procedure is operable and that the reslurry;rand-, oxidation steps are unnecessary. (Ref, No. 3). However, these changes were not adopted because of . inconclusive results in the final color testing. 3. Isodibenzanthrone, purified by crystallization from sulfuric acid, failed to give improved color. . Filtration diffi culties were experienced especially when the acidity was adjusted to strengths much below 93*0$. At acidities higher than 93-0$, the yield loss was excessive (CWNB 1696 pages 18-21, 24-29, 34-36, 39-40, 47-51). 4. It was.demonstrated that in the chlorination step the isodibenzanthrone can be charged as filter cake instead of being first dried and ground. The cake is suspended in the re action medium and dehydrated before chlorination. This change was adopted as standard practice (Ref. No, 4),. 5. In some laboratory runs it was shown that bleaching of the crude color could be eliminated (CWNB 1696 pages 99-113). Because of testing uncertainties this change was not given a plant trial. This point should be investigated further. 6: The use of cationic, anionic and nonionic surfac tants in the acid-pasting drowning water had no effect on color, quality (CWNB 1655 page 178). The conclusions drawn from the above phase of the investigation were based on lithovarnish rub-out testing which was subsequently proved to be unreliable. Although certain process fundamentals had been es tablished at this point, lack of uniformity in final color . quality persisted. The investigation was continued as follows: -5- DUP050467830 Solvent Bleed 7. In August, 1957 it was reported that several lots failed to meet the solvent bleed specification. This test had been suspended for several years but was reinstated at a cus tomer^ request. It was shown that the solvent bleed deficiency could be corrected by rebleaching the color. This treatment also improved shade in the lithovarnish rub-out test and in a newly established jiffy paint grind testing procedure (CWNB 1758 pages 54-55* 100-101), This method of improving the color was demonstrated again, early in 1958 (CWNB l8l4 pa^es 15-16, 102-105, 129-131, 178-179). Acid Pasting 8. At this time .a study was . started to determine the optimum acid-pasting conditions for the production of uniformly satisfactory color. Many sets of conditions were tried and a number of additives were used in the drowning water. The results, based on lithovarnish rub-out and jiffy paint grind testing, were inconclusive (CWNB 1758 pages 137-158, 148-149, 151-152, 157, 159-164, 177-178; CWNB 1765 pages 194-195; CWNB l8l4 pages 3, 8, 17-19, 23-25, 59-62, 77-81, 95-96). o-Dichlorobenzene vs. Nitrobenzene as the Chlorinating Medium 9. In the chlorination step, nitrobenzene would be preferred to o-dichlorobenzene as the reaction medium because of occasional difficulties with the control chlorine analysis due to incomplete removal of the solvent. In addition, nitrobenzene Is the cheaper solvent. In an evaluation study it was shown that with nitrobenzene the color quality was not significantly different from the control, using o-dichlorobenzene. This was trua for both vatted and unvatted isodibenzanthrone. It was concluded, therefore, that although there are certain advantages for nitrobenzene they are outweighed by the hazardous nature of this material and that o-dichlorobenzene should be retained as the chlorinating medium (CWNB 1765 pages 168-169, 171-172, 194-195). (See also Bart II, Experimental Section No. 5k.) 6- - DUP050467831 Milling with a Manton-Gaulin Mill vs, HF Milling 10.. In a milling study it was shown that material test ing blue and dull by the llthovarnish rub-out procedure after passage through a Manton-Gaulin mill will be red and bright after HF milling. The redness is accentuated when the amount- of dis persant, Compound 8, Is doubled (CvJNB l8l4 pages 26-27) . Optimum Chlorine Content of'Crude Color , 11. In a study in which unvatted isodibenzanthrone was under- and overchlorinated, it was shown that the optimum final chlorine content of the color lies between 14.0 and 15.0$ (the theoretical value for the dichloro derivative is 15.5$). Overchlorination tends to give darker masstones, redder shades and heavier solvent bleeds; underchlorination gives bluer shades. It was observed that the chlorine content will drop, as much as one percentage point during the bleaching and acid-pasting procedures. It is, therefore, necessary to use 14.5 to 15.5$ as the control chlorine specification (CWNB l8l4 pages .44-4.7,).' This value has been established as standard practice (Ref. No. 5). Vatted vs. Unvatted Isodibenzanthrone 12. Samples of vatted and unvatted isodibenzanthrone were converted to color by the regular processes of chlorination, steam distillation, acid-pasting and bleaching. It was concluded from the results obtained in llthovarnish rub-out testing that vatting is not necessary (CWNB 1814 pages.52-58) Bleaching Studies 15(a). A composite of isodibenzanthrone charges was chlorinated by the regular procedure to the crude color. This was filtered from the solvent and washed with fresh solvent as a means of purification and then steam-distilled. -One portion of the purified color was acid-pasted and bleached; another portion was bleached first and then acid-pasted. All final colors were very good in llthovarnish rub-out and solvent bleed testing and In the jiffy paint grind test. The procedure of acid-pasting first and then bleaching gave slightly better results. ''Llthosol'* Fast Violet 4RNL is now made by acid-pasting crude and then bleach ing (Ref. No. 6). It should be noted that the/crude color was -7- DUP050467832 not given the usual bleach treatment, which suggests that this step could be omitted if the color is to be produced as "Lithosol" Fast Violet 4RNL Paste. However, the laboratory crude was solvent-filtered, whereas the plant crude is not. This study ...also showed that the purification by filtration from the solvent is an excellent means of obtaining a color which will satisfy the solvent bleed specification (CWNB 1814 pages 20-22, 112-119, 172-176). 13(b). In a later study, involving tile paste crude color, it was shown that one-quarter to one-half part of bleach is the optimum. In this case the color had already received the regular bleach treatment at the crude stage. The hypochlorite (outside) bleach has slightly less dulling effect than chloride (inside) bleach (CWNB 1952 pages 55-5^, 70-80). Principally be cause of convenience, one-half part of inside bleach has been es tablished as standard practice (Ref. No. 6). Salvage of Off-quality Material 14. Lots 86 and 88 of uLithosolM Fast Violet 4RNL Paste were extremely blue in lithovarnish rub-out and in jiffy paint grind testing. Laboratory runs indicated that this could be corrected by an alkali-boil treatment (CVJNB 1876 pages 187-189; CWNB 1$10 page 25). Lot 86 was upgraded satisfactorily by this procedure! Lot. 88 failed to respond and had to be disposed of as the "Ponsol" type for the textile trade (Ref, Nos, 7 and 8). Elimination of the Paste Crude Acid Pasting and Bleaching Steps 15. Lots 90 and 92 of "Lithosol" Fast Violet 4RNL Paste were made by milling directly "Ponsol" Brilliant Violet 4RN Crude Bleached, thus eliminating the usual acid-pasting and final bleaching steps (CWNB 1910 pages 124-125). Lot 90 was satis factory in lithovarnish rub-out testing; Lot 92 was not. The shortened procedure, therefore, had to be abandoned pending fur ther investigation. It has attractive possibilities. 8- - DUP050467833 Conversion of ''Lithosol" Past Violet 4RN to "Lithosol" Past Violet 4RNL 16. ' The redness, brightness and strength brought out in the viscous milling procedure used in the.''Lithosol'1 Past Violet 4RN Paste process can be retained for "Lithosol" Past Violet. 4RNL Paste. The large amount of dispersant present in the 4r n type is first destroyed by an acid boil, the color isolated > by filtration and then redispersed by the normal procedure for : the 4RNL type (CWNB 1910 pages 4?, 60-62, 80)5. It might be possible to destroy the excess dispers ing agent with bleach and then standardize as the 4RNL type. It is recommended that this procedure be tried. TP 10-24 Testing Procedure 17. At this point in the investigation, a new procedure for evaluating final color was established^ It was the Pigments-,. Department's full paint grind test designated as TP 10-24. It replaced the lithovarnish rub-out and the Jiffy paint grind pro cedures, both of which were found to be unreliable. It was learned later that the new test was also unreliable, possibly be cause the method details differed significantly from the procedure used at Pigments Department (see Section 25 of this study). Effect of Triton X-100 18. Using the TP 10-24 testing procedure it was shown that the addition of 1 to ~yfa Triton X-100, a nonionic dispersing agent, would bring out redness and brightness in color previously judged blue and dull (CWNB 1910 pages 162-165, 167, 172$ CWNB 1952 pages 107-108). However, the use of this additive was pro hibited by the Pigments Department (Ref. No. 9). Increase in Dispersant 19. A fourfold Increase in the amount of Compound 8 used as the dispersant resulted in marked improvement in redness and brightness of final color (CWNB 1918 page 153). -9- DUP050467834 Lot 94 of "Lithosol" Past Violet 4RNL Paste was pre pared with this change in procedure and was judged satisfactory in TF 10-24 testing. This demonstrated that the additional amount of dispersing agent is necessary and can be tolerated in the final paste. This change has been adopted as standard practice (Ref. No. 10)(see also Part II Section 4). Additional Studies 20(a). In a renewal of the acid-pasting study by other workers (c. E. Carr and F.. C.- Eberly) many variations in the technique were tried. The results.were generally inconclusive. Samples testing satisfactorily in the lithovarnish rub-out pro cedure were generally unsatisfactory in the TP 10-24 paint grind testing. Some of the acid-pasting conditions led to material which jelled in the paint grind system (CWNB 1918 pages 112, 115, 118, 150-152, 154, 140-141, 159, 166-170, 172-17?, 182-185, 185-189; CWNB 1905 pages 61-65, 65, 68, 76, 78, 87-88, 95). 20(b). Color made from crudes extracted with boiling acetic acid and crudes prepared from purified isodibenzanthrone (solvent-extracted and acid-crystallized) gave unsatisfactory re sults `in TP 10-24 testing (CWNB 1932 pages 25-27; CWNB 1918 pages 108, 114, 116, 120-121, 124, I5O-I5I, 135, 137, 139, 142-145, 149, 157, 161, 165). 20(c)-. Vatting of the; crude color followed by oxida tion and bleaching was not beneficial (CWNB 1918 pages 147, 172-173). 20(d). Milling with a dispersant after acid-pasting and before bleaching brightened the color (CWNB 1932 pages 72-73, 76-78). This change in process has not been given a plant trial. Quality'of Competitive products 21. In May, 1959/ seven competitive products were tested against our standard using the TP 10-24 procedure. The results are listed in the following table (Ref. No. 11): 10 - Parts 100 105 100 102.5 100 110 975 Material "Lithosol" Past Violet 4RNL Paste Lot 83 Cibanone Violet 4r Double Paste Caleosol Violet 4RP Single Paste Indanthrene Brilliant Violet 4R Infra Paste Sandothrene Violet N4R Double Paste Sandothrene Printing Violet N4R Double Paste Carbanthrene Brilliant Violet 4r Paste Calcoloid Violet 4RD Single Paste Shade Standard 2 Blue, 2 dull 1 Blue, 5 dull 3 Blue, 1 dull 1 Blue, 2 dull 4 Red, 1 dull Jelled 2 Blue These results show that these competitive products do not approach the "Lithosol1* offering in shade' and brilliance. Elimination of the Acid-Pasting and Pinal Bleaching Steps ' 22. Earlier in the study {Section 16) it was pointed out that the redness, brightness and strength brought out in the viscous milling in the "Lithosol11 Past Violet 4RN Paste procedure will be retained after the dispersant is destroyed by an acid boil and the color redispersed. It was later shown that many labor atory samples and plant charges made by the regular viscous milling of "Ponsol11 Brilliant Violet 4RN Crude Bleached were excellent in TP 10-24 and Fade-Ometer lightfastness testing and entirely satisfactory as "Lithosol1' Past Violet 4RNL Paste (CWMB 1952 pages 81-82, 163, 174-176; CWNB 1949 pages 4-5, 23, 25* 95* 102) (se.e also Part II* Section 4). This change should be considered for plant adop tion because of the considerable savings possible through the elimination of the drying, acid-pasting and bleaching steps now normal for this color. - 11 - DUP050467836 Reproducibility of the TP 10-24 Testing Procedure 23. The testing inconsistencies experienced with the lithovarnish rub-out and jiffy paint grind procedures seemed to continue after the establishment of the Pigments Departments full paint grind test, TF 10-24. To verify this observation, the following statistical study of the method was carried out: Six samples from each of six different lots, in cluding the standard lot, were prepared and coded for testing. Half of these samples were laked at the l,Ponsolrt Laboratory by two individuals; the remainder were laked at Technical Laboratory. All laked samples were tested at Technical Laboratory by the TP 10-24 procedure set up by this group. This testing included both the 50:50 and the 5:95 tintings with a white base. The statistical findings proved that significant shade and strength discrimination existed among the testing per sonnel and indicated serious lack of reproducibility in the TP 10-24 testing procedure (Ref. No. 12). It was recommended that the necessary adjustments to the procedure be made and the statistical study be repeated. "Lithosol'1 Past Violet 4RN Double Paste 24. ; In response to a request by the Pigments Department for a cheaper color in the form of a double paste. Lot 1 of nLithosol" Past Violet 4RNL Double Paste was prepared from selec ted charges of 11 Ponsol" Brilliant Violet 4RN Crude Bleached which were O.K. for paper dyeing. The press cakes were put through a PolXows and Bate mill in the presence of the normal amount of dispersing agent and a small amount of carboxymethyl cellulose added as a thickener. The lot was entirely satisfactory in TP 10-24 and solvent bleed testing against the standard-Lot 83 of "Lithosol" Past Violet 4RNL Paste and was produced at an estimated savings of 8o4/pound (Ref. Nos. 13 and 14). - 12 - DU P050467837 The shortened procedure was similar to that used for Lot 90 of regular production., reported in Section 15 of this study. It is recommended that more consideration be given to this method of producing the 4RNL type. It will be necessary to use selected charges of "Ponsol" Brilliant Violet 4r n Crude Bleached until process controls are developed and applied so that uniform high-quality material is produced at the crude stage. This phase of the overall problem is discussed In Part II of this report. - 13 - DUP050467838 PART II "LITHOSOL" PAST VIOLET 4RN PASTE Introduction This color is marketed as a pigment and is used to dye paper. For many years quality has been inconsistent from lot to lot principally because of uncontrollable shade differences. A study to correct this deficiency has been carried out simultane ously with that reported in' Part I for the 4RI& type. Many of the changes studied in Part I were.evaluated by standard paper dyeings but were generally ineffective in producing a color of uniform quality. See Appendix 1 for an outline of the steps involved in the current production of this color starting with benzanthrone. i Experimental The following additional studies for product and process improvements were undertaken:' 1. Acid-Pasting The following acid-pasting procedures were tried. None was successful in producing uniformly satisfactory color. In general, shades were much too blue compared with standard. (a) Use of 101.5$ sulfuric acid to dissolve the crude color (CWNB 1696 pages 167-169). (b) Turbulent drowning (CWNB 1696 pages 199-200). (c) Acid swelling - slurry in 85$ sulfuric acid (CWNB 1758 pages 15-14, 45-47). (d) Use of Glaubers Salt in the acid (CWNB 1758 pages 15-16, 45-47, 151-152). (e) Drown to 56$ acidity (CWNB 1758 pages 17, 45-47). (f) Drip-drown in ice 'and- water (CWNB 1758 pages 47, 148-149). - 14 - DUP050467839 (g) Drown in boiling sodium hydroxide solution (CWNB IT38 pages 50-51, 151-152), 50-51). (h) Drip-drown in boiling water (CWNB 1738 pages (i) Drown in boiling saturated sodium sulfate solu tion (CWNB 1738 pages 65-66, 149), (j) Drip-drown in a boiling solution of sodium hydroxide and sodium hypochlorite (CWNB 1738 pages 65-66,*151-152). (k) Drip-drown in boiling water then bleach with sulfur dioxide (CWNB 1738 page 76). (l) Drip-drown in different amounts of boiling water (CWNB 1738 pages 144-145). (m) ' Drip-drown in water at 85 to 90C. containing llAlkanolH HC (CWNB 1738 pages 148-149). (n) Use of naphthalene in the acid (CWNB 1738 pages 148-149). (o) Drip-drown'in a boiling solution of sodium car bonate (CWNB 1738 pages 151-152). (p) Drip-drown at 20 to 40C. using a special funnel (CWNB 1738 pages 159-160). (q) Drip-drown at 20 to 40C. and at 85 to 90C. with a special funnel and using nitrobenzene in the acid (CWNB 1738 pages 159-160). . 2, Vatting The following vatting procedures were tried. None was successful and again the color shades were much too blue com pared with standard. (a) Vatting during HP milling with the milling con tinued after the vat was destroyed by slow oxidation (CWNB 1738 pages 62, 68-69). - 15 - DUP050467840 (b) Passage through a Manton-Gaulin mill in the presence of triethanolamine, then vatting and oxidation with "Sitolu oxidizing agent and the milling continued (CWNB 1738 pages 63-64). 158). (c) Vatting, oxidizing and bleaching (CWNB 1738 page 3. Chlorination Because it is possible that the cause for persistent blue shades is at least partly chemical, the chlorination step to pro duce the crude color was investigated as follows: (a) Water as a Contaminant One factor suspected of adversely affecting the re action is the presence of a small amount of water. Sulfuryl chloride, the chlorinating agent, will react with water to form sulfuric acid and hydrochloric acid. The former might direct the chlorine into undesirable isomer positions in the isodibenzanthrone molecule. Such by-product formation could be responsible for blue shades of final color. Water could be present because of the in complete dehydration of the o-dichlorobenzene-isodibenzanthrone slurry. In one run the slurry was dehydrated thoroughly and then contaminated with 2$ water prior to chlorination. Gross underchlorination (6$ vs. the standard of 14 to 15$ chlorine) re sulted even after repeated additions of sulfuryl chloride (CWNB 1970 pages 94-96). For a smaller amount of water (0.4$) chlorination was complete but the final color was- judged 3 blue 4 dull against standard; a control was judged 3 blue, 4 bright against standard (CWNB 1970 pages 155-158). These results indicate that water is detrimental and in relatively large amounts will act as a reaction inhibitor. It is recommended that this factor be studied fur ther and that plant equipment be inspected frequently for sources of water. - 16 - DUP050467841 (b) Sulfuric Acid as a Contaminant Thoroughly dehydrated o-dichlorobenzene-isodibenzanthrone slurry was contaminated with 6$ of 98$ sulfuric acid just prior to chlorination'. Even with double the process amount of sulfuryl chloride and an extended reaction period, the chlor ination was negligible... (1.2^rchiorine)(CWNB 1970 page 120). This result indicates that under the conditions im posed, sulfuric acid in the amount used will inhibit the reaction. . (c) Hydrochloric Acid as a Scavenger Because of incomplete filter cake washing the isodibenzanthrone often contains sodium hydroxide as a contaminant. The alkali will react with sulfuryl chloride with the reaction products possibly acting as inhibitors. Prior to dehydration* a reaction slurry which tested strongly alkaline was acidified with hydrochloric acid, dehydrated and then chlorinated according to .process conditions. The results were inconclusive. No improvement over a control sample was realized; both samples were bluer than stsuadard (CWNB 1970 pages 140-143). (d) Thionyl Chloride as a Scavenger Thionyl chloride reacts with water to form hydrogen chloride and.sulfur dioxide and, unlike sulfuryl chloride, leaves no sulfuric acid residue. It should, therefore, serve as a water scavenger. Thoroughly dehydrated reaction slurry was containin' ated-with 0,4$ water. Sufficient thionyl chloid.de to destroy the water was added and the chlorination carried out according to process The results were inconclusive. No improvement over a control sample was realized; both samples and the control were judged 3 blue against standard (CWNB 1970 pages 140-143). The procedure was given a plant trial with inconclusive results (Ref. No, 15). - 17 - It is recommended that the use of thionyl chloride , as a water scavenger be considered seriously for inclusion In the regular procedure as a precautionary measure. (e) Effect of alpha-Picoline alpha-Picoline appears occasionally as a contaminant in the..plant recovered o-diehlorobenzene. Addition of 5*0$ of this material to fresh o-dichlorobenzene used In the regular chlorination had no detrimental effect on final color (CWNB( 1970 pages i4o-l45'). (f) Fresh vs. Recovered o-Dlchlorobenzene Regular chlorinations were run in fresh and recovered o-dichlorobenzene. Slightly better color was obtained.,when the former was used, indicating the possibility of miscellaneous detri mental contaminants in the recovered o-dichlorobenzene (CWNB 1970 pages 152-152; CWNB 1994 pages 6-8). (g) Monoohlorobenzene as the Reaction Medium Inferior color was obtained when monochlorobenzene was used in place of o-dichlorobenzene as the chlorination medium (CWNB.1970 pages 152-154).;. . .. (h) Effect of Iron During the chlorination, contamination with iron salts could occur from the action of sulfuryl chloride vapors and hydrochloric acid on the upper parts of the reaction system. To determine whether such contamination is detrimental, a run was made using o-dichlorobenzene containing 0.7$ ferrous chloride (peCl2'4H20). Pinal color was inferior In shade to a control sample indicating that kettle corrosion is a definite factor in product et'erioration and must be avoided (CWNB 1970 pages 155-158). It is recommended that the chlorination equipment be inspected at "intervals to eliminate any possible sources of iron contamination. (i) Catalysts - The regular chlorination catalyst is iodine. The substitution of potassium iodide for Iodine resulted in no - 18 - DUP050467843 improvement in color quality. In this study it was shown that satisfactory chlorination can be accomplished.without a catalyst particularly when a high reaction temperature (100 to 105C. vs. the process temperature of 65 to T0C.) is used (CWNB 1994 pages 96-98, 112-113). Indiffereiit results were obtained in plant trial runs using potassium iodide as the catalyst (Ref. No. 16). It is recommended that higher temperature noncatalytic chlorinations be tried In plant rims.' (j) Chlorination with Thionyl Chloride An unsuccessful attempt was made to chlorinate with thionyl chloride In place of sulfuryl chloride. The re action was carried out at reflux (125 to 130C.) over a period of six hours. The chlorine content of the product was only 11.6# as compared with the control level of 15 to 16#. A final sample for dye testing was not prepared (CWNB 1994 page 114). As thionyl chloride is an excellent water scavenger, it is recommended that this investigation be continued. ,(lc) Nitrobenzene vs. o-Dichlorobenzene as the"Chlorinating Medium The following table summarizes the results ob- . tained In a study to compare nitrobenzene with o-diehlorobenzene as the chlorinating medium using vatted and unvatted"Isodibenzanthrone. Regular chlorination and bleaching procedures were used and the final samples were prepared by the viscous milling technique (CWNB 1765 pages 168-169, 171-172, 186-193)(see also Part- I, Experimental Section No. 4). - - 19 - DUP050467S44 Type of Isodibenzanthrone Vatted Vatted Unvatted Unvatted Reaction Medium 0-Dichlorobenzene Nitrobenzene o-Dichlorobenzene Nitrobenzene . paper Dyeing vs. Standard 2-1 Blue, 2-1 bright Off-shade, brighter 3 Blue, 3 bright 2 Blue +20 -10 -10 -10 These results indicate that o-diehlorobenzene is ' preferred, reaction medium and that purification of the isodibenzanthrone by vatting is not necessary and is actually somewhat detrimental. (l) Isodibenzanthrone Quality To demonstrate that the quality of the final color is dependent to some degree on the quality of the starting isodi benzanthrone;, two selected charges were chlorinated. The regular chlorination and bleaching procedures were used and the final color paste prepared by the viscous-milling technique. In addition, the chlorination mass was filtered and the cake washed with fresh solvent (CWNB 1910 pages 141-143). The results are summarized in the following table: Isodibenzanthrone______ Charge' No, Purity Dibenz. 86 83.3$ 7.0$ 87 75.3 4.5 Paper Dyeing vs. Standard 5-4 Bright -10 1 Blue +20 These results indicate that in this study superior color was made from the better quality isodibenzanthrone. '. " It is recommended that an Improved procedure for determining Isodibenzanthrone quality be developed by the Analytical Methods Improvement Group and a satisfactory Quality Control Limit be established. - 20 - DUP050467845 4. Viscous Milling A study showed that substantial improvements, partic ularly in intensity and brilliance, result when the final paste is prepared by viscous-milling the bleached crude color (CWNB 1696 pages 170-171, 178-183, 194-195, 197-198; CWNB 1758 pages 25-24, 75, 77). Pour plant trial charges of "Ponsol" Brilliant Violet 4r n Crude Bleached VM were made by this(technique (Ref, No. 17). This material was standardized to thi!*ee lots of ,,Lithosol'f Fast Violet 4RN Paste with an average solids content of 7.7# as compared with the standard solids of 11.7$. This pro cedure resulted in a 20$ reduction in mill cost which, on the basis of an annual production of 60,000 lb., represented a gross annual savings of approximately $17,000. Because of this excellent performance, the viscous milling procedure was established as standard practice (Ref. No. 18). In the plant procedure as first established, the dispersant consisted of 0.28 part of Blancol (based on the 100$ weight of the crude color). A study showed that this amount usually could be reduced drastically without impairing quality or yield. The laboratory millings were done with the process amount of sand (5 to 6 parts per part of crude color) with a . 5-hour milling period after attaining maximum viscosity. The results obtained are summarized in the follow ing table (CWNB 1952 pages 174-176; CWNB 19^9 pages 4-5, 25). Parts of Blancol 0.33 0.17 0.11 0.08 0.06 0.03 Particle Size Index ' 13.5 15.6 15.5 15.0 16.9 17.1 Paper Dyeing vs. Standard 4 Red 4 Red 2 Red 4-3 Bright 4-3 Bright 4 Bright 4 Bright 1:1 1:1 -10 +15 -10 +5 - 21 - DUP050467846 As a result of this study successful plant trial runs were made using 0.075 part of Blancol. The use of 0.075-0.15 part of Blancol was recommended for adoption as standard practice (Ref. No. 19). One reason for reducing the amount of dispersant to a minimum is the possibility of using the viscous milling technique for the production of the. 4RNL type. As pointed out in Part I, Section 22, "Lithosol" Past Violet 4r n L Paste made-by this proced ure is satisfactory in TP 10-24 and lightfastness testing (CWNB 1952 pages 81-82, 163, 174-176$ CWNB 1949 pages 4-5, 23, 25, 95, 102)(see also Part I, Sections 16 and 19)- 5. Effect of Milling Temperature' In a short study to determine the effect" of milling tem perature on color shade, a slight advantage was shown for 70 to 80C. over 30 to 40C. Two-hour visbous millings using 0.15 part of Blancol,were made. A summary of the..results'obtained is given in^-the .following table (CWNB 1994 page 86) : Milling Temperature 30-40C . -. 70-80C. ' ' Particle Size Index . " i.12.6 '12.4. Paper Dyeing vs. Standard 4 Blue. ' 4 Bright -10' 4 Red 4 Bright -20 '6. Physical Factors Affecting Shade The attainment ofStandard shade for`pigment colors is dependent, in part at least, bn such factorb as crystal form, particle size, particle distribution and optical -phenomena in volving light - reflection, refraction, diffusion and scattering. In the case of the subject color, persistent blueness against standard has been the principal cause for rejection. Color on the red side of standard is acceptable. Therefore, avoidance of physical treatment which would cause a shift from red to blue Is necessary and application of physical treatment which would accomplish the reverse is desirable. - 22 - DUP050467847 A. Degradation of Color Quality In the following study, certain physical manipula tions which will- degrade the color were demonstrated. (a) Prolonged Viscous Milling A composite of four plant charges of ?,Ponsol" Brilliant Violet 4RN Crude Bleached was selected for a series of laboratory millings. Plant milling of this material produced "Lithosol11-Past Violet 4r n Paste testing 3 fed, 4 bright -20 against standard. Each laboratory milling consisted of 60 g. 100$, of crude color, 20 g. 100$ of Compound 8 and 325 g. of fine sand. Milling periods were from 30 minutes to 16 hours after maximum viscosity was reached (CTOJB 1910 pages 58-59). following table: The results obtained are summarized in the Milling Period, ______Hours______ 1/2 1 2 4 8 16 16 Particle Size Index 17.6.. 17.0 19.9 10.7 13.0 12.2 14.1 Paper Dyeing vs. Standard 2 Red 2 Red ; 3 Red 3 Red 4 Blue 2 Blue 4 Blue 4 Bright 4 Bright 4 Bright 4/3 Bright 4/3 Bright 4 Bright 4/3 Bright -15 -10 -1; 1. -5 -10. 1:1 -5* *Used 54 g. 100$ Compound 8. These data indicate a definite shift from red to blue as milling is prolonged. In a 16-hour milling using a large excess of Compound 8 (54 g. 100$ vs. 20 g.), the color was judged 4 blue 4/3 bright -5 against standard, showing that an increase in dis persing agent will tend to prevent the shift in shj.de. - 23 - DUP050467848 Although these data also show that there Is no correlation between color shade and particle size index, consider- able plant experience does seem to indicate that there is such a correlation. In a similar study using a coarser sand (Key stone Sand No. 22) prolonged milling failed to bring about a shade shift (CWNB 1910 pages 122-123). following table; The results obtained are summarized in the Milling Period, ______Hours______ 1 2 4 6 8 Particle Size Index 19.1 18.217.1 17.0 16.2 Paper Dyeing vs . Standard 3 Red 3 Red 3 Red 3 Red 3 Red +20 +20 +25 -30 -30 This type of sand was recommended for plant trials, but because of filtration difficulties with it when milling other colors, it was decided to cancel the study. In view of added experience with this technique it is suggested that this de cision be reconsidered. Using the coarser sand, a laboratory evaluation milling procedure using 20 g. 100^ of crude color instead of 60 g. was developed. This procedure is useful when small samples only . are available for testing (CWNB 1910 pages 151-152). (b) Acid Pasting ... . Portions of plant Charge 4-6 of.''Pohsol.", Brilliant Violet 4RN Crude Bleached were: 1. Viscous-milled for control. . 2. Acid-pasted, drip-drowned in Ice and water and viscous-milled. , 3. Acid-pasted, drip-drowned in boiling water and viscous-milled (CWNB 2207 pages 41-42, 67-68). - 24 - DUP050467849 following table: The results obtained are summarized in the Conditions - Control Acid-pas te (cold) Acid paste (hot) Particle Size Index 11.5 17.7 19.5 Paper Dyeing vs. Standard 5 Red 4 Bright 1 Blue Off-shade 1 Blue Off-shade These data indicate that acid-pasting is detri mental by causing a definite shift in shade from red to blue. (c) Vatting A portion of plant Charge 3-525 Qf "Ponsol" Brilliant Violet 4RN Crude Bleached was viscous-milled for control. Another portion was vatted, oxidized by blowing-with air and viscous-milled (CWNB 1994 pages 87-88, 148-149). table: The results obtained are shown in the following Conditions Control Vatted Particle Size Index 17.6 33.7 ,' Paper Dyeing vs. Standard 4 Red 4 Bright -20 1 Blue 4/3 Dull 1:1 These data indicate that vatting is detrimental by causing a definite shift in shade from red to blue. (d) Solvent Digestion Portions of plant Charges 3-523 .and 3-525 of "Ponsol11 Brilliant Violet 4RN Crude Bleached were viscous-milled for controls. Other portions were dehydrated in o-dichloro benzene and held at 140 to 145C. for six hours. The solvent was removed by steam distillation and also by filtration and washing with alcohol. All filter cakes were then viscous-milled (CWNB 1994 pages 83-84, 87-89, 109, 148-149). ing table: The results obtained are shown in the follow - 25 - DUP050467850 Crude Color Chg. 5-523 5-523 Chg. 3-525 3-525 3-525 Particle Size Treatment Index Paper Dyeing vs. Standard Control 13.8 ODCB digest 16.4 Steam distillation 4 Bed 4 Bright -20 1 Blue 4 Bright +25 Control 17.6 ODCB digest 14.8 Steam distillation 4 Red 4 Bright -20 1 Blue 1:1 ODCB digest Filtration Alcohol wash 13.9 1 Blilie 4 Bright 1:1 These data indicate that prolonged digestion in o-dichlorobenzene is detrimental by causing a definite shift in shade from red to blue. B, Upgrading of Off-quality Color The above results indicate that degradation of the color to unacceptable blue shade can be readily accomplished by a number of procedures involving physical change. A procedure which would bring about a reversal of this effect is desirable.' The following study, which is only preliminary in nature, shows that this may be possible. Portions of plant Charge 4-6 of "Ponsol" Brilliant Violet 4RN Crude Bleached were: 1. Viscous-milled for control. 2. 'Acid-pasted and drip-drowned in boiling : waterv A part of the acid-pasted color was suspended in a 50$ solution of o-dichlorobenzene in ethyl alcohol and digested at 60 to 70C. for 16 hours. After isolation by filtration and washing with alcohol the cake was viscous-milled by the regular procedure (CWNB 2207 pages 41-42, 67-68). table: The results obtained are summarized in the following - 26 - DUP050467851 Treatment Control Acid-pasting Alcohol-ODCB digestion of acid-pasted color Particle Size Index 11.5 19.5 13.7 Paper Dyeing vs. Standard 3 Red 4 Bright +25 1 Blue Off-shade 5 Bright +20 3 Yellow 3 Bright -10 Although the shade of the digested sample was judged yellow, it was more red than blue and indicated' that a definite shift in the desired direction had taken place. This was the first time that red color had been degraded to blue and then re stored to red. An attempt to apply this technique to color orig inally on the blue side of standard was unsuccessful (CWMB 2207 page 66). That some physical change takes place during the digestion procedure is shown by the appearance of diffraction rings produced in electron microscope studies. Material testing red on paper produces sharply defined rings j that testing blue on paper produces a diffused pattern. It is strongly recommended that this lead be ex ploited. If the proper physical treatment of the final color caii be established, then process alterations and improvements starting with the production of dibenzanthronyl. sulfide can be studied with the assurance that the results will not be jeopardized by a last minute change which will degrade the color. - 27 - DUP050467852 REFERENCES 1. Experimental Change in Process S.N. 56-62 for Isodlbenzanthrone Crude S, dated February 29, 1956 and the corresponding Results Sheet, approved May 2, 1956. 2. Chambers Works Process for the Manufacture of Isodibenzanthrone Crude S Wet, dated August 2, i960. * 3. Test Authorizations TA-C-57-152 dated April 4, 1957 and TA-C-57-152, Part II, dated May 29, 1957 and the corresponding Test Conclusion, dated November 10, 1959. 4. Test Authorization TA-C-57-261 dated May 31, 1957 and the corresponding Test Conclusion, dated August 29, 1958* 5. Chambers Works Technical Standard 770-638 520 dated October 27, 1959, "Ponsol" Brilliant Violet 4RN Crude Wet. 6. Chambers Works Technical Standard 770-455 940, nLithosol" Fast ' Violet 4RN Paste.Crude Acid, dated March 28, i960. Chambers Works Technical Standard 770-455 920,. "Lithosol" Fast Violet 4RN Paste Crude, dated January 29, I960. 7. Test Authorization TA-C-58-509 dated July 2,. 1958 and the corresponding Test Conclusion, dated June 8, 1959. 8. Test Authorization TA-C-58-969 dated December 12, 1959 and the corresponding Test Conclusion, dated November 6, 1959. 9. Letter dated May 7, 1959 from Mr. H. Amoul, Pigments Department, Newark, N.J. to Mr. R. L. Burnett, Orchem, Wilmington, Del. 10. Test Authorization TA-C-59-491 dated June 8, 1959 and the corresponding Test Conclusion, dated August 5, 1959* 11. Letter dated May 28, 1959 from T. G. Webber, Technical Labor atory to Mr. R. B. Morton, Pigments Department, Newark, N.J. 12. Chambers Works Chemical Control Department Monthly Summary, page 57, November, 1959. - 28 - DUP050467853 13. Test Authorization TA-C-61-766 dated October 5,1961 and the corresponding Test Conclusion, dated May 23, 1962. 14. Letter dated October 23, 1961 from A. . Bilancioni, Technical Laboratory, to Mr, R. L. Burnett, Dyes and Chemicals Division. 15. Test Authorization TA-C-60-684 dated September 12, i960 and the corresponding Test Conclusion, dated June 14, 1961. 16. Test Authorization TA-C-61-188 dated March 10, 1961 and the corresponding Test Conclusion, dated April '19, 1961. 17.. Trial Manufacture Request TMR-C-57-81 dated August 14, 1957 and Part II dated August 28, 1957, and the corresponding Test Conclusion,dated February 19, 1958. 18. Interim Technical Standard, ''Lithosol" Fast Violet 4RN Paste, authorized March 7, 1958. 19. Test Authorization TA-C-59-981 dated December 22, 1959 and the corresponding Test Conclusion, dated February 7, 1961. : /ves - 29 - DUP050467854 APPENDIX DUP050467855 APPENDIX 1 The following chart shows the steps involved in the current production of "Lithosol" Past Violet 4RN Paste and "Lithosol" Past Violet 4RNL Paste. Benzanthrone Pure j. (bromination? in nitrobenzene) Bromobenzanthrone Pure Wet " (fusion under pressure in ir aqueous Na^S-S mix) Dibenzanthronyl Sulfide condensation in a CoHp-OH-KOH v melt Isodibenzanthrone Crude S Wet (chlorination in o-dichlorov benzene)' ' ''Ponsol" Brilliant Violet 4RN Crude Wet (hypochlorite bleaching) v "Ponsol" Brilliant Violet 4RN Crude Bleached "Ponsol" Brilliant Violet 4RN Crude Bleached VM "Ponsol" Brilliant Violet 4RN Crude Bleached Dry (standard!zation) v (acid pasting) V DUP050467856 "Lithosol" Past Violet 4RN Paste "Lithosol" Past Violet 4RN Paste Crude Acid (hypochlorite " bleaching) V* "Lithosol" Past Violet 4RN Paste Crude ( (standardization) v "Lithosol" Past Violet 4RNL Paste For the textile trade, "Ponsol" Brilliant Violet 4RN Crude Wet is standardized to "ponsol" Brilliant Violet 4RN Paste. it DUF050467857 DISTRIBUTION S. N. Boyd Jackson Laboratory J. P. Froning) In E. w. Parr ) Turn Chambers works A. V. Willett, Jr. ) H. I. Stryker C. L. Richardson ) ) Turn E. G. Wiest ) Process Dept. Central File) Chambers Works Process Dept. Central File nPonsoln-"Monastral"-nLatyl" Group File (2) Vital Records DUP050467858