Document YrxE06gbo3Y28qoay7yDO5egK

REACTION OF PROPELLANT II WITH WATER Reproduced by permission from SOAP & CHEMICAL SPECIALTIES Volume XU, No. 12, December 1965, page 117 By Paul Sanders "Freon" Products Division E. I. du Pont de Nemours & Company Wilmington, Delaware 19898 oo CM tH i--I D Ld 6BCON ihjn ' O M fi i N * T ; O N S OF FREON-- OR F-- WITH NUMERALS ARE DU FONT'S REGISTERED TRADEMARKS FOR ITS ELJORCCaRBON PROPELLANTS Reaction of water with propellant 11 at 130F. shown to be catalyzed by metal. Products of reaction, propellant 21, fluorocarbon 112 and acid, indicate three different reactions occur. Reaction of Propellant 11 with Water HE stabilitv of trichloromonofluoromethane (propel lant 11) in the presence of water and metals has been of con siderable interest to both the aero sol and refrigeration industries. In the aerosol industry, the instability of propellant 11 in the presence of water and metals has limited its use in aqueous based aerosol products. In the refrigeration industry, the stability of propellant 11 is of con cern because of the possibility of corrosion to the metal components in the refrigerating system and the effect of acidic reaction products on the insulation. As a result of these consid erations. the stability of combina tions of Propellant 11, metals and water has been the subject of sev eral investigations. Parmelee and Downing ('ll aged combinations of steel, aluminum, water and pro pellant 11 in both the absence and presence of air. The corrosion that was observed on steel was consid ered to be due principally to rust ing, but it was reported that hvdrolysis of the propellant ultimate ly occurred. The presence of air in creased the corrosion of the steel. Church and Maver (2) aged combinations of propellant 11 and various metals in cylindrical pres sure bombs at temperatures up to 150F. Moisture contents in the mixtures were varied from 0.001% to over 1.0% and the air content was varied up to 4.0 volume per cent. These conditions were chosen in an attempt to simulate the con ditions which might be encounter ed in industrial refrigeration sys tems. Church and Mayer postu- By Paul A. Sanders "Freon" Products Laboratory E. I. du Pont de Nemours <5 Co. Wilmington, Delaware lated three possible decomposition reactions of propellant 11 in the presence of water and metal as fol lows: (M=metal). ion as low as 20 ppm. They also were unable to detect any hydrogen halide in the vapor phase. The con centration of organic decomposi tion products in the reaction mix tures was too low for identification, but on the basis of their rate data. Church and Mayer favored reac tion 1. 2CCLF -f M-*MCl, -f CCTFCCTF (1) CCljF % FLO---CC1.TOH + HCI ,F 1/ C=0 \ Cl _ HCI (2) 2CCKF -j- 2H.0 - 2M-2CHC1T -f MCI, -f M (OH). (3) Reaction 1, which produces a metallic chloride and tetrachlorodifluoroethane tA'-iorocarbon 112), is similar to a Wurtz reaction. In Reaction 2, propellant 11 is decom posed by water to form a hvdrogen halide and an analog of phosgene. In Reaction 3, propellant 11 is re duced to dichloromonofluoromethane (propellant 21) by the ac tion of metal and water. Reaction 3 is similar to the reaction that is used for the preparation of chloro form in which carbon tetrachlor ide is reduced with iron and water. Church and Mayer determ Among aerosol loaders, com binations of propellant 11 and anhvdrous ethyl alcohol, in products such as hair sprays, were known to have caused container corrosion at various times. Initially, this was at tributed to the hydrolysis of the propellant 11 by the low concentra tions of water present in the ethyl alcohol. Later, it was shown that, under certain conditions, propel lant 11 reacted with anhydrous ethyl alcohol to produce acetalde hyde, hydrogen chloride, and dichloromonofluoromethane a s t h e major reaction products (3) . CCLF + CH3CH2OH--CH3CHO + HCI + CHCLF (4) ined die rates of decomposition of propellant 11 under various condi tions by analyzing the reaction pro ducts for chloride ion. They were unable to find any fluoride ion with a method they reported could detect concentrations of fluoride The acid formed in this reaction was responsible for the corrosion observed in the containers. Reac tion 4 was shown to be a free radical reaction and the mechanism of the reaction was elucidated. In a subse quent investigation of potential E ID 1 1 4 2 9 stabilizers for the reaction, nitro- Before proceeding to investi Water alone causes consider methane was found to be an excel gate the problem on this basis, able corrosion in aerosol tinplate lent inhibitor for the reaction (3). however, a preliminary check was containers. Combinations of water The availability of propel made to determine if propellant 21 and propellant 11 cause more cor lant 11 containing nitromethane as could be formed in the reaction of rosion than water alone as a result an inhibitor ("Freon-11" S) elimi propellant 11 with water. Mixtures of the reaction of water with pro nated corrosion problems resulting of propellant 11 and water were pellant 11. Nitromethane not only from the reaction of propellant 11 stored in tinplate containers with retards the corrosion from water with ethyl alcohol. Interest then blank caps for one week at 130F. alone but also retards the corrosion shifted to the possibility of using The propellant 11 layer was ana caused by mixtures of water and propellant 11 in combination with lyzed by gas chromatography and propellant 11. One of the functions 95% ethyl alcohol instead of anhy found to contain an appreciable of nitromethane as a conosion in drous ethyl alcohol because of the concentration of propellant 21. hibitor for water or water propel cost saving that could be achieved The fact that propellant 21 was lant 11 systems appears to be a de- by the substitution. formed in the reaction between activator for the metal. This pre In view of the previous work water and propellant 11 in metal vents the metal from acting as a on the reaction of propellant 11 containers eliminated the possibili catalyst for the reaction between with anhydrous ethyl alcohol, and ty of distinguishing between Reac water and propellant 11. the corrosion studies of Parmelee and Downing which showed that propellant 11 reacted with water alone under certain conditions, it would be expected that, in combi nations of propellant 11 and 95% ethyl alcohol, propellant 11 would react with both the ethyl alcohol and the water. In any investigation involving the reaction of propel tion 4 and Reaction 5 by analyzing the reaction mixture for propellant 21 and acid. An investigation of the reaction between propellant 11 and water alone in the presence of metals therefore was initiated, not only to clarify the mechanism, but also in the hope that this might lead to compounds or methods which might retard the reaction. Nitromethane was evaluated alone as a corrosion inhibitor and also in combination with a variety of other known corrosion inhibi tors. The combination of nitroanethane/sodium benzoate appeared to be the most effective. This combi nation was tested in three aqueous based aerosol products, a window cleaner, a starch formulation and lant 11 with 95% ethyl alcohol, it a snow formulation. The nitrome- r was important to be able to de Summary and Conclusions thane/sodium benzoate combina V termine the relative rates of reac Combinations of propellant tion was effective in retarding the tion of propellant 11 with, ethyl 11, water, and metals were stored corrosion caused by the aerosol alcohol and with water. It was con at 130F for various periods of window cleaner during storage at ceivable that the overall hydrolysis time. The reaction mixtures were 130F. but caused adverse color reactions of propellant 11 with wa analyzed and found to contain pro changes in the other two aerosol ter could be represented by the fol pellant 21, fluorocarbon 112, acid products. This indicates that the lowing equation: and, in some cases, quantities of inhibitor combination must be CCLF -f 2H:0--CO, + 3HC1 + HF (5) It was known that, for all insoluble solids assumed to be me practical purposes, Reaction 4 repre tallic salts or hydroxides. The for tested thoroughly in each formula tion before large scale use is con sidered. sented the reaction of propellant 11 with ethyl alcohol. If the assump mation of propellant 21 and fluoro carbon 112 indicated that the fol Experimental tion that Reaction 5 represented the lowing two reactions probably oc 1. Reaction of propellant 11 with overall reaction of propellant 11 curred: Water with water was valid, then, when mixtures of propellant 11 and 95 % 2CC13F + M-*MC1. -r CC1S FCCLF (1) ethyl alcohol were reacted, the rel 2CC13F + 2FLO + 2M-2CHC1 ,F -f MCI, + M (OH), (3) /-f'-V . E ID 1 1 4 3 0 ative rates of Reaction 4 and Reac tion 5 could be determined by ana lyzing the reaction mixture for pro pellant 21 and for acid. The con centration of propellant 21 in the reaction mixture would indicate The presence of acid in some of the mixtures indicated that either Reaction 2, Reaction 5 or some other acid-forming reaction took place. A. In the absence of metal catalysts In the absence of metal cata lysts, propellant 11 does not appear to react with water. Thus, in water the extent to which Reaction 4 took The decomposition of pro saturated with propellant 11 at at place, while the excess of acid (or pellant 11 by water is catalyzed by mospheric pressure, the hydrolysis halide ion) above that formed in the metal. In the absence of metal, rate was too low to measure after Reaction 4 would indicate to what propellant 11 does not appear to one year at room temperature. (4) . extent Reaction 5 occured. react with water. In the presence of steel, the hydro- Table Z. Reaction of Propellant 11 with Water in Glass with Metal One Week at 130F. Mixture No. Woter/Propellant 1 1 Ratio (Wt. %) Metal" Present 1 70/30 No 2 70/30 Yes 3 70/30 Yes Air Present Yes Yes No Concentration of Reaction Products In Organic Phase (Wt. %) pH of Propellant 21 Fluorocorbon 112 Aqueous Phase None 1.43% None 0.09% 4.3 4.9 1.43% 0.04% 4.6 Appearance of Reaction Mixture Colorless The aqueous phase was grey and there was a considerable amount of brown solid present Both layers were almost colorless but there was some grey solid present "Metsii strips from a sideseam tinplate aerosol container. lysis rate became appreciable and could be determined by analytical methods. In unpublished work car ried out at the "Freon" Products Laboratory, homogeneous solutions of propellant 11 in dioxane con taining up to 2.5% water were aged in glass for 18 months at 167 F. Judging by the fact that the chloride ion concentration did not increase during the aging period, there was no hydrolysis of the pro pellant 11 during that time. B. In the presence of metal Experiments to determine the effect of both metal and air upon the reaction between pro pellant 11 and water were carried out by sealing water /propellant 11 (70/30) mixtures in glass flasks and aging the flasks for one week at 130F. One of the samples was prepared without any added metal; the other two samples were pre pared with added strips of metal from a sideseam, tinplate aerosol container. The strips were suffici ently long so that part of the strips extended into the vapor phase. One of the flasks containing the metal strips was cooled in ice water and sealed in order to trap air in the flask. This would indicate the stability of propellant in the pres ence of water, metals and air. The other sample was cooled and evac uated to remove air before sealing. This would indicate the stability of propellant in the presence of water and metals without any air present. After the storage tests were completed, the organic phases were analyzed by gas chromatography. The results of the analyses, along with visual observations of the re action mixtures, are given in Table I. The data show that propellant 21 and fluorocarbon 112 were form ed during the metal catalyzed re action of propellant 11 with water. No other unknown reaction pro ducts were found during the gas chromatographic analysis of the mixtures. The catalytic effect of the metal strips (which included the soldered sideseam) from the tinplate container upon the re action was very pronounced. In the absence of the metal, no re action between water and propel lant 11 occurred. The effect of air was not very conclusive, judging from the present data, but there was some indication that the rate of reaction was increased bv the air present. Reaction Mixture No. 2 had more solid material present and the mix ture was more discolored than Re action Mixture No. 3 Also, the con centration of fluorocarbon 112 in Reaction Mixture No. 2 was high er than that in Reaction Mixture No. 3. The formation of fluorocar bon 112 and propellant 21 can be accounted for by Reaction 1 and Reaction 3, respectively. The mol ar concentration of propellant 21 in Reaction Mixture No. 2 was about 31 times greater than that of fluorocarbon 112. It might be as sumed from this that Reaction 3 occurred at a rate about 31 times greater than Reaction 1. However, this conclusion would also assume that neither propellant 21 nor fluorocarbon 112 themselves react ed further once they are formed. It is known that propellant 21 will react in the presence of steel and water (4) and it is likely that fluorocarbon 112 would also react. Therefore, the relative concentra tions of propellant 21 and fluoro carbon 112 in the mixture do not necessarily indicate the relative rates of Reaction 1 and Reaction 3. ' The similarity of Reaction 3 to the preparation of chloroform bv the reduction of carbon tetra chloride with iron and water has been mentioned. This was con firmed bv adding a small quantity of carbon tetrachloride to propel lant 11 and aging the mixture in the presence of water in metal containers. Chloroform was identi fied as one of the reaction pro ducts. The pH of the three aque ous phases was about the same. It is possible that no acid was form ed during the reaction. In this repect, the results would be similar to those reported by Church and Mayer. A second possibility is that acid was formed and reacted com pletely with the metal present to form metallic chlorides. In another experiment, mix tures of water and propellant. 11 were aged in drawn tinplate aerosol containers with blank caps for one week at 130F. Under these conditions, there was no sold er present in the reaction mixture. According to titration of the aque ous phase with base, there were E ID 1 1 4 3 1 twice as many mols of acid as there were mols of propellant 21 in the mixture. The presence of the acid indicated that either Reaction 2. Reaction 5, or decomposition re actions involving propellant 21 or fluorocarbon 112 occurred. Church and Mayer were unable to find any fluoride ion in the reaction mix ture, which would eliminate Reac tion 5. Neither were they able to find any hydrogen halide in the vapor phase which wouid tend to eliminate Reaction 2. However, it is a little dangerous to extrapolate their results, which were obtained with relatively low concentrations of water, to the present experi ments. II. Comparative Corrosive Effects of Water Alone and Water/ Propellant 11 Mixtures in Aerosol Tinplate Containers. The preced ing work has shown that a number of reactions occurred during the storage of mixtures of propellant 11 and water in metal containers. Considerable corrosion occurred in the container during the storage tests. Before attempting to find compounds which might retard the corrosion, however, it was neces sary to determine how much of the corrosion was caused by water alone and how much was caused by the products of the reaction be tween water and propellant 11. In order to obtain this in formation, samples with water alone and samples with mixtures of water and propellant 11 were aged in tinplate containers with blank caps for periods up to eight weeks at 130F. The corrosion ob served in the containers was judged by the rating system shown in Ta ble II. The data obtained from three different experiments are listed in Table III. These data show that water alone caused con siderable corrosion in the contain ers. Mixtures of water and propel lant 11 caused more corrosion than wateT alone, as would be expected. Therefore, any effective corrosion inhibitor for water/propellant 11 systems in metal containers would have to include an inhibitor for re Table n. Corrosion Rating System Corrosion Ratine] Description 0 0-1 l 1-2 2 2-3 3 3-4 4 4-5 5 No corrosion No corrosion, but slight darkenening either liquid or vapor phase or slight ring at interface Generalized corrosion either liquid phose or vapor phase--spotty de fining Generalized corrosion both liquid phase and vapor phase -- spotty detinning Complete detinning either liquid phase or vapor phase Complete detinning either liquid phase or vapor phase with spotty detinning in other phase Complete detinning both liquid phase and vapor phase Complete detinning either liquid phase or vapor phase with rust ing and spotty detinning in other phase Complete detinning both liquid phase and vapor phase with spotty rusting Complete rusting throughout container Container leaked tarding the corrosion of metal by water alone. If this could be achieved, the reaction between water and propellant 11 might also be retarded since the data had shown very clearly that the reac tion between water and propellant 11 was catalyzed by metal. 111. Corrosion Inhibitors for Water and WaterjPropellant Mixtures. The investigation of the reaction of water with propellant 11 had shown that the reaction was catalyzed by metal. The data had also shown that water alone caused corrosion in metal containers. The search for suitable corrosion inhib itors, therefore, was concentrated on compounds which would in hibit metals against corrosion by aqueous systems or compounds which would inhibit the corrosion of metals by halogenated com pounds. The use of nitromethane for stabilizing halogenated compounds against decomposition by metals has been revealed in patents (5) . It has also been disclosed as a stabi lizer for retarding the corrosion of metals bv aqueous antifreeze solu tions (6) . The use of nitro methane in combination with pro pylene oxide as a corrosion inhib itor for aqueous based aerosols has also been disclosed recently (7). Nitromethane was tested alone and in combination with other known corrosion inhibitors. Tests were carried out at 130F. in tinplate containers with water alone and with mixtures of water and propellant 11. The most ef fective inhibitor combination for reducing the corrosion either from water alone or from mixtures of water and propellant 11 was a ni tromethane/sodium benzoate com bination. The corrosion occurring in metal containers with various combinations of water, propellant 11, nitromethane, and sodium ben zoate is shown by the data in Table IV, using the corrosion rating sys tem given in Table II. The data in Table IV show that nitromethane was very effec tive alone as an inhibitor for re tarding the corrosion from water or from water/propellant 11 mixtures. Sodium benzoate was ineffective in retarding the corrosion from water alone but was somewhat helpful in retarding the corrosion from water/propellant 11 mixtures. The preceding experiments in.Table Corrosive Effects of Water and Water/Propellant 11 Mixtures in Tinplate Containers Run No. Composition of Water/ Propellant 11 Mixture (Wt. %) Water Propellant 1 1 Corrosion oftor Sforage at 1 30F 2 Weeks 4 Weeks 8 Weeks 1 100 0 1 70 30 2 100 0 2 70 30 3 100 0 3 70 30 2 3-4 3-4 4 2 3-4 2 3-4 3-4 3-4 3-4 4-5 3-4 5 3-4 4-5 3-4 5 E ID 1 1 4 3 2 Table IV. Effect of Nitromethane /Sodium Benzoate Combinations in Retarding Corrosion from Water /Propellant 11 was also effective for actual aque ous based products, the nitrome- Composition of Mixture (Parts by Wt.) Corrosion in Metal thane/sodium benzoate combina Water Sodium Benzoate Nitromethane Propellant 1 1 Contomers after Storage 2 Weeks 4 Weeks 8 Weeks tion was tested in three aqueous based aerosol products, a window 105.0 -- -- -- 2 3-4 3-4 cleaner, a starch formulation, and 105.0 - 0.3 -- 104.7 0.3 - - 104.7 0.3 0.3 -- 105.0 - - 45.0 105.0 - 0.3 44.7 104.7 0.3 - 45.0 T 3-4 0 3-4 1 2 1-2 3-4 0-1 4 1 4 1-2 an aqueous based snow formula 3-4 tion. The snow was formulated 0-1 with propellant 12 only and the 5 1 stabilizers were, tested for retard 4 ing the corrosion due to water a- 104.7 0.3 0.3 44.7 0-1 1 1 lone. The concentrates used for had shown that the reaction be tween water and propellant 11 was Alkyl halides are also appar ently able to react with metals to the three aqueous based products had the following compositions: catalyzed by metals. Propellant 21 and acid were among the reaction products. Since nitromethane was found to be a corrosion inhibitor for the water/propellant 11 system, produce organic radicals (9) . These can yield coupled symmet rical products. This could explain the formation of fluorocarbon 112. In this case, nitromethane could Aerosol Starch Concentrate Amaizo 513 Starch 10% Tributyi tin oxide in ''Triton" X-45 Wt. % 4.00 0.02 it should also retard the formation of propellant 21 and acid when the act either by deactivating the metal or bv reacting with the organic Glycerine Water 0.30 95.68 mixtures were stored in metal con tainers. This was demonstrated by free radicals that were produced. In some cases, metal reductions may give rise to organic radicals Aerosol Window `Triton'* X-100 Cleaner Concentrate Wt. % 0.1 aging mixtures of water with pro pellant 11 or propellant 11 containing 0.3% nitromethane ("Freon-11" S) in tinplate con (9) . Again, nitromethane could act by adsorption on the metallic surface or by reacting with the or ganic radical itself. "Emcol" 5138A Odorless Mineral Spirits Isopropyl Alcohol Water 0.1 9.0 1 B.8 72.0 tainers for 16 hours and 24 hours The mechanism by which Aerosol Snow Concentrate at 194F. The acidity of the aque ous phases was determined by ti tration and the propellant 11 layer was analyzed on the gas chroma tograph to determine the concen sodium benzoate functions as a cor rosion inhibitor apparently is not known (10). IV. Evaluation of the Ni tromethane,fSodium Benzoate In Stearic acid "Emcol" 14 Myristic acid "lucite" 46 Acrylic Resin Methylene chloride Wt. % 4.0 1.5 4.0 4.0 39.0 tration of propellant 21. The re sults are ven in Table V. The data show that nitromethane re duces the formation of both acid and propellant 21. Nitromethane may function as an inhibtor in several ways. For example, it may be adsorbed on hibitor Combination with Aque ous Based Aerosol Products. The effectiveness of nitromethane alone or nitromethane in combination with sodium benzoate in retarding the corrosion from water alone or trom water propellant 11 mixtures had been demonstrated. In order Wafer 47.5 The compositions of the aerosols and the results of the storage stability tests with the aque ous based formulations are given in Tables VI, VII, and VIII. The results indicate that the nitrome- thane/sodium benzoate combina- the tinplate and thus deactivate the to determine if the combination metal. Balezin and Chistyakov (8) evaluated various inhibitors for retarding the corrosion of car bon steel by carbon tetrachloride Table V. Stabilization of Water/Propellant 11 Reaction at 194r with Nitromethane in Metal Containers and carbon tetrachloride and water combinations. They reported that benzoic acid, one of the inhibitors Aged 16 Hours at 194F Composition of Mixture Water/Propellant 1 1 Water/"Freon-l V' S (70/30) (70/30) E ID 1 1 4 3 3 studied, functions by adsorption on the steel surface and that this ap parently is followed by a reaction of the benzoic acid with the steel. The fact that nitromethane re Normality of aqueous phase % Propellant 21 in PropiJanf 1 1 Aged 24 Hours at 194F Normality of aqueous 0.11 0.29% 0.54 0.0006 None detected o.ooi tards corrosion by water alone sug gests that it is adsorbed on the metal. phase % Propellant 21 in Propellant I 1 0.31 % None detected tion was useful in only one out of the three formulations. Definite stabilization against corrosion was observed with both the window cleaner formulation and the snow formulation. However, in the latter case, the nitromethane/sodium ben zoate combination caused the snow to become orange during storage. With the starch .formulation, the nitromethane/sodium benzoate combination did not stabilize the system against corrosion and the inhibitor combination had an ad verse effect upon the color of the formulation. These results indicate that the combination of nitromethane/ sodium benzoate can be useful but that it must be tested thoroughly. In some cases, the stabilizers ap parently react with some of the components of the aerosol. In other cases, the corrosion caused by the aqueous based formulation apparently is not due to water but to some component in the formula tion that is not inhibited by the nitromethane/sodium benzoate combination. Summary The reaction of water with propellant 11 at 130F was shown to be catalyzed by metal. The pro ducts of the reaction were propel lant 21, fluorocarbon 112 and add. These reaction products indicate that at least three different reac tions occur. Nitromethane alone or in combination with sodium benzo ate stabilizes aerosol tinplate con tainers against corrosion either by water alone or by water/propellent 11 mixtures. The data indicate that the stabilizing effect of nitrome thane is due to deactivation of the metal by nitromethane or deacti vation of the organic radicals that might be produced. The combination of nitrome thane and sodium benzoate was tested in three aqueous based aero sol formulations, a window cleaner, a starch, and a snow. The inhibi tor combination was useful only with the window cleaner. In the other two formulations, the inhibi tors had an adverse effect. References 1. Parmelee, H. M. and Downing, R. C., "Corrosion in Aqueous and Al coholic Aerosol Systems", Soap and Sanitary Chemicals 26, 114-19 (July, 1950) 2. Church, J. M. and Mayer, J. H., "Stability of Trichloro-fluoromethane in the Presence of Moisture and Certain Metals", Journal of Chemical Engineering Data, Vol. 6, No. 3, July, 1961 3. "Freon" Aerosol Report A-51, "Sta bilization of CC13F in the presence of Alcohol" (Proceedings of the CSMA, 46th mid-year meeting, May, 1960) 4. "Freon" Technical Bulletin B-2, "Properties and Applications of the `Freon' Fluorinated Hydrocarbons" ,r>. British Patent 773,186, U.S. Patents 2,567,621, 2,639,971; 2,923,749; 3,159,592 6. U. S. Patent 2,185,238 7. Glessner, A. S., "Packaging of Wa ter Base Insecticides", Soap and Chemical Specialties 40, 83-86, 112 (August, 1964) 8. Balezin, S. A. and Chistyakov, V. M., "Corrosion Inhibitors for Car bon Steel in Carbon Tetrachloride", Academy of Sciences USSR, Pro ceedings Chemical Technology, 142/ 3, 1-3 (1962) January--April 9. Walling, C., "Free Radicals in Solu tions", New York; John Wiley and Sons, 1957, p. 584 10. Technical Bulletin, "Sodium Benzo ate as a Corrosion Inhibitor", Mon santo Chemical Company Table VI. Storage Stability of Aqueous Based Aerosol Snow Formulations Concentrate Composition (Parts by Wt.) Nitro- Sodium methane Benzoate Propellant 12 Corrosion after Storage at 130F 2 Weeks 4 Weeks B Weeks 74.0 0.5 0.5 25.0 0-1 0-1 75.0 - - 25.0 2 2 1 2 The nitromethane/sodium benzoate combination caused the color of the snow to become orange after storage for 8 weeks at 130 F. Table VII. Storage Stability of Aerosol Window Cleaner Formulations Concentrate Composition (Parts by Weight) Propellant 1 2/ Corrosion after Storage Nitro Sodium Propellant 11 at 130F methane Benzoate Propellant 1 2 (50/501 2 Weeks 4 Weeks 8 Weeks 95.0 -- _ 5.0 -- 94.5 0.5 -- 5.0 -- 94.5 -- 0.5 5.0 - 94.0 0.5 0.5 5.0 -- 90.0 -- -- -- 10.0 89.5 0.5 -- -- 10.0 89.5 -- 0.5 -- 10.0 89.0 0.5 0.5 -- 10.0 3-4 3-4 3-4 0-1 2 2 0-1 0-1 1 0-1 0-1 0-1 3-4 3-4 3-4 2 3-4 3-4 2 1-2 5 0-1 0-1 1 All containers had a corrosion rating of 0*1 after storage for one year at room temperature Table VIII. Storage Stability of Aerosol Starch Formulations Composition (Parts by Weight) NitroConcentrate methane Propellant 12/ Propellant 12/ Sodium Propellant 114 Propellant 11 Benzoate (40/601 (50/50) Corrosion after Storage 1 Year at 4 Weeks Room at 130 F Temperature 95.0 -- -- 5.0 -- 94.0 0.5 0.5 5.0 - 95.0 - -- - 5.0 94.0 0.5 0.5 - 5.0 3-4 3 2 3-4 3-4 3 3-4 3-4 Concentrates with the nitromethane/sodium benzoate combination turned black during storage E ID 1 1 4 3 4 E. I. DU PONT DE NEMOURS & CO. (INC.) "FREON" PRODUCTS DIVISION WILMINGTON, DELAWARE 19898 FOR FURTHER INFORMATION Please contact one of the following district offices: 60 Glenwood Avenue East Orange, New Jersey 07017 Phone: 676-1112 (Area Code 201) 7 South Dearborn Street Chicago, Illinois 60603 Phone: 222-5000 (Area Code 312) 701 Welch Road Palo Alto, California 94304 Phone: 326-2840 (Area Code 415) 712 Main Street Suite 324 Gulf Building Houston, Texas 77002 Phone: 222-2468 (Area Code 713) 3503 North Hoyt Street El Monte, California 91734 Phone: 283-0548 (Area Code 213) or: Wilmington, Delaware 19898 Phone: 999-3620 (Area Code 302) The information contained herein is based on technical data and tests which we believe to be reliable and is intended for use by persons having technical skill, at their own discretion and risk. Since conditions of use are outside of Du Pont's control, we can assume no liability for results obtained or damages incurred through the application of the data presented. Publication of the information in this bulletin should not be understood as permission or recommendation for the use of "Freon" compounds in violation of any patents. u. S. PAT rT' Better things for better living.. .through chemistry E ID 1 1 4 3 5