Document pmBD4OkxY5Ee63VxK72pNQGya

Monsanto FROM. (NAME--LOCATION--PHONE) J. K. Sears - Galt 1730 DATE : SUBJECT : REFERENCE : TO : June 9, 1981 Phthalate Plasticizers as Capacitor Fluids JHO/JKS 5/18/81 V J. H. Orem B3NC CC; M. W. Farrar T3B . W. Touchette 1730 R. H./Mills T2F J. A. Cannon 1730 A. W. Morgan 1730 D. H. Paul 1730 J. T. Renshaw 1730 C. D. Bias 1730 R. J. Belle B3NB A. F. Fitzgerald B3NB Q. E. Thompson T3E Claudette Dethridge S3C W. E. Keener R3D Diane Clay 1410 R. V. Johnson 1410 John, your memo of 5/18/81 asks about the possible use of Santicizer 900 as a capacitor fluid to beat DOP at the job. We get questions similar to this several times a year from someone. Most questions we can answer immediately. Yours prompted some measurements and a closer look. I am going to answer more than you asked. You say Aerovox and Armel-Dublier combined use about 3 million pounds of DOP per year. We would like to get a large part of that business as well as that from other capacitor manufacturers. We should sample any of the industry that is willing to test a candidate with Santicizer 900 (but not Santicizer 900P) as sooon as we can. This memo should show us what to do to get back into this market, if we can at all, or it should put this question to rest neatly for a long time to come. It is concerned with five aspects or parameters of a compacitor fluid: 1) Dielectric constant, DK (should be high). 2) "Cleanability", or its ability to be cleaned up to give good resistivity; a low dissipation factor is the only indicator we have for this. 3) Stability; its ability to maintain good resistivity in use. 4) Volatility ( should be low). 5) Price. In Plasticizers we have significant data on only 1, 4 and 5 and bits, pieces and hunches on 2 and 3. A 6th parameter, flame retardance, went out with PCBs. ' . As I see it, the industry has compromised on DOP. Phosphates have high dielectric constants but poor cleanability. Benzyl phthalates have higher dielectric constants than DOP but poor stability. Adipates and tri- t. IN - lO ( REV. 8/77 ) DSW 587306 STLCOPCB4092642 - 2- mellitates are low in DK. Dialkyl phthalates win by default of others. DOP has a reasonably high D.K. with reasonably low volatility. It can be cleaned up rather well, and is reasonably stable. Its price is low. Three years ago, DOP was the best choice they had. Three years from now that may not be so. This memo is based on my assumption that a samll compromise can be made in any parameter 1, 4 or 5 (but not in 2 and 3 which are marginal now) for a significant improvement in any of the 5 parameters. By copy of this memo I am asking Bill Koemer, Quent Thompson and Caludette Dethridge to correct any errors they may find in my assumptions and my understanding of capacitor fluids. I am asking them to supply answers to questions I raise if they know them, and to let me know if they see anything to suggest the Plasticizers Division has a good chance technically to put a product into this field. Also, by copy of this memo I am suggesting questions for Diane Clay to ask the above companies. This is the very first action I see needed now, but samples need not be held up until we have the answers. Data We Have; ' Table 1 gives data we have on file which bear directly on the value of a phthalate plasticizer as a dielectric fluid for capacitors. The phthalates are arranged in order of increasing^>molecular weight as indicated by the average number of carbons in each alkyl chain. When two or more plasticizers have the same molecular weight, I have listed the most linear one first. Those that follow are increasingly branched. The data show that they are all close together in dielectric constants (DK). The differences, however, are very significant. A capacitor company would like something even slightly higher than DOP, if it can be cleaned up electrically and if the price is right. The "best" phthalate on the sheet, then, would be di(methyl isobutyl carbinyl)-phthalate, DK 5.98, except, I presume, it is too expensive. I expect it is far too volatile as well. Diisohexyl phthalate is the "next best". Does anyone use it? If not, I assume it is because it is too volatile. I believe it can be cleaned up, but I do not know. Is it possible that in DOP the ethyl group in the 2- position gives it unusual stability for capacitor use, as it gives it unusual stability compared to other branching, for plasticizer uses? If so maybe the oxo-alcohol phthalates can't win. DSW 587307 STLCOPCB4092643 -3- You will notice that for each molecular weight, increased branching increases volatility as shown, by volatile loss from PVC sheet. Therefore for low volatility linearity is desirable. Dicapryl phthalate (DCP) and DOP are the only exceptions in the table, but DCP is the ester of a secondary alcohol, octanol-2, so is not quite comparable to anything else in the table. You will also see that above Cg increased branching lowers the dielectric constant. Again linearity is good. Santicizer 900 is definitely better (slightly) than Exxon's DINP, if it can be cleaned up properly. It is better than DOP in volatility but poorer in DK. It is so close to Santicizer 711 that it raises the old question of why Santicizer 711 failed in this use. Was it low DK, price only, or poor cleanability or stability? Probalby it was a combination, and Santicizer 900 might step in at the right price. The linear and near linear dinonyl phthalates are far better than DOP in volatility but just almost as good in DK. Price would kill them unless volatility is a big enough factor to justify a high price. Can we find what price would be justified? If Santicizer 900 eventually comes on strong and we still have 7-9-11 alcohols, a linear di C9 phthalate might be the best product. Look below C8 and you see, in contrast to the above, that increased branching increases the DK. Jayflex 77 and itil's diisoheptyl phthalate have DK's of about 5.3 which is better than DOP at 5.19. Does anyone use them? Are they too volatile? Can they be cleaned up and made stable or not? Why are they not the best capacitor fluids in the world? Unfortunately we hurt DK by going to our linear C7, and the improvement in volatility in the linear is not enough. In the middle of the table the di C8 phthalates show little effect of branching on DK, as would now be expected, and our data is spotty. I can not find some data, I feel sure we have hadl Santicizer 790 should be a winner except the nearly linear Cy group does not help DK. John and Diane, we are not set up to measure DK at 100C. We could get set up, I presume, but it would require a lot of testing to be sure we were right. Perhaps Bill Koeraer's area or Quent Thompson's group are still set up to get DK at elevated temperatures. Perhaps we could sample Aerovox and persuade them to measure it. DSW 587308 STLCOPCB4092644 4 Questions in Summary; Diane, can you find from Aerovox or Armel-Dublier: 1) Can any diisoalkyl phthalte be cleaned up and made stable enough? 2) Does Armel-D use Exxon's DINP or their Jayflex 77? If not, why not? This might really tall us what we need to know. 3) How much volatility can they stand, and how much is low volatility worth? 4) How much can they compromise one property to gain a bit on another? 5) What was really wrongwith Santicizer 711 as a fluid? If any carbonee know the answers, please give me a call at 4-4259. Att /vw DSW 587309 STLCOPCB4092645 TABLE 1 PHTHALATE PLASTICIZERS FOR CAPACITOR FLUIDS: PROPERTIES FOR PREDICTION OF PERFORMANCE Plasticizer Di-n-hexyl phthalate Diisohexyl phthalate Di(methyl isobutyl carbinyl)phthalate Ave. No. Carbons Per Alcohol % Volatile Loss From PVC at 87C 1 Day 6 Days Dielectric Dissipation Constant Factor 1 KH_ 1 10 10 100 100 6 11.3 6 13.6 6 22.8 5.08 5.09 4.96 5.62 5.69 5.67 5.94 5.98 5.96 .1820 .0170 .0020 -- - - .0028 .0007 .0012 Di-n-heptyl phthalate Bisoflex 71 Di Cy (NTA) Phthalate ME 49, OR-141113 9-25-69 Di C7 Phthalate (NTA?) MIC 220 314 Jayflex 77, Exxon 1-20-81; DK 5-29-81 Diisoheptyl Phthalate Di-n-octyl Phthalate Morpel X-926 Di-nOoctyl Phthalate ME-3 . 7 .3 'r 4.96 4.93 4.94 7 4.5 21.0 5.15 5.14 5.21 7 5.1 27.6 5.12 5.11 5.08 7 6.2 5.31 5.30 5.34 7 7.1 -- 5.6 5.3 5.2 8 - 2.5 8 --- 5.04 . 4.96 4.70 .0072 .0008 .0006 .0010 .0001 .0004 .0100 .0009 .0070 - .1670 .0100 .0020 DSW 587310 i STLCOPCB4092646 Plasticizer . TABLE 1 Cont*d PHTHALATE PLASTICIZERS FOR CAPACITOR FLUDIS: PROPERTIES FOR PREDICTION OF PERFORMANCE Ave. No. Carbons Per Alcohol Volatile Loss From PVC at 87 C 1 Day 6 Days Dielectric Dissipation Constant Factor 1 KH0 10 2 1 KH_ 10 2 100 100 Dicapryl Phthalate, DCP Hatco. Di-2-ethyl hexyl Phthalate DOP Diisooctyl Phthalate 8 4.6 to 5.4 5.00 4.90 4.90 -- 8 3.6 to 20.8 5.15 .0052 4.5 5.19 .0010 4.17 .0010 8 4.3 4.90 4.90 4.90 - - Santicizer 790 p.p. Ba 1 and 2 July 20 (heptyl nonyl phthalate) Santicizer 711 Di-n-nonyl Phthalate ME-7 Di Cq Phthalate (NTA) ME-50 Santicizer 900D 5/29/81 DINP Exxon 5/29/81 ' 8 2.9 13.6 8-9 1.9 7.9 9 0.8 2.7 - 9 - 1.0 3.7 9 1.4 6.1 9 1.8 7.5 5.01 5.04 5.05 4.75 4.76 4.78 5.08 5.09 4.96 5.00 5.05 5.00 4.76 4.76 4.79 4.66 4.63 4.63 .0016 .0000 .0000 .0023 .0002 .0007 .0620 .0060 .0010 .0006 .0003 .0007 .0004 .0001 .0056 .0005 .0001 .0054 1 .3B- .. ' - ' -- DSW 587311 1 STLCOPCB4092647 Plasticizer - TABLE 1 Cont'd. PHTHALATE PLASTICIZERS FOR CAPACITOR FLUDIS: PROPERTIES FOR PREDICTION OF PERFORMANCE Ave. No. Carbons Per Alcohol Volatile Loss From PVC at 87 C 1 Day 6 Days Dielectric Dissipation Constant Factor 1 KH_ 10 2 1 KH. 10 2 100 100 Di-n-decyl phthalate ME-4 (NTA) DIDP (multi branched) 10 0.6 1.5 4.96 4.93 4.94 .0350 .0038 .0020 10 1.1 4.43 4.43 4.47 - v-- DUP (nearly linear) 11 0.6 to 0.9 1.7 -_3 4.42 4.43 4.45 - * - DSW 587312 STLCOPCB4092648