Document GmjZ9ZYGq8VZ4LbZQKLJ5qrmV

Monsanto W...WHKIC.TIW., yyt r. Richard -Research Center =>*'* = May 20, 1970 1 "* , ' *cccct MONTHLY SUMMARY DETAILS - APRIL, 1970 - FUNCTIONAL FLUIDS - RESEARCH" __ " ro ' :* ` R. Davis w. R. Eberhardt J. F. Herber D. R. Miller Q. E. Thompson L. R. Stark J. Quinn/W. Knapp R. H. Munch E. Keller S. Tucker W. Bannister F. S. Clark R. L. Green S. L. Reid R. W. Street J. D. Sullivan R. W. Weiss J. C. Wygant P. D, Fowlkes E. 0. Stejskal Attached is our April, 1970 . Monthly Summary with Details for Functional Fluids-Research. W. R. Richard ms I Att ,, tJ . fi .. 11261 HARTOLDMON0013210 MONTHLY SUMMARY DETAILS - APRIL 1970 FUNCTIONAL FLUIDS - RESEARCH AVIATION FLUIDS Skydrol 500B: Marketing Technical Service (JFH) Boeing is installing an AIT apparatus per ASTM D-2155. They asked us to run four fluid samples to check their apparatus. Results are: Phosphate Ester "A" Phosphate Ester "B" Engine Oil "A" Engine Oil "B" 735F 930F 765CF 755F The following fluid samples were analyzed: Fluid: PIA Utility Reservoir Texas International A/C 307 Right Systems Left Systems Vise. 100F 210 F 7.88 cs 2.60 cs 9.49 cs 3.10 cs 9.32 cs 3.04 cs Acidity 0.29 TAN 0.25 TAN * 0.34 TAN Moisture 0.45* 0.30* 0.34* Density 1.0647 1.0591 1.0593 Skydrol 500C: Flight Test (1425243-JFH) The properties of fluid samples from PAA received to date were analyzed by Jack Hinchen. The following conclusion can be drawn based on the assumption that aircraft top-up was the same for both the 500B and 500 C airplanes. The overall conclusion is that topping-up a system with 50OB does not affect fluid properties. That is no reduction in acidity is shown and the additional stabilizer is rapidly used up. In the 500C aircraft 11/2 top-ups stabilizes acidity at less than 0.40 TAN and stabilizer levels are high and remain that way. The 500B top-up shows no effect on viscosity but 500C viscosity approaches new fluid values. The 500B/500C mixtures in aircrafts after two 500C top-ups the moisture level is 0.30* and remains constant thereafter. . The 500C flight test at BOAC in Boeing 707*s is underway. An analytical method was developed based on the VI improver's differences to determine the amount of 500C in 500B used samples. The method's precision needs to be refined but ' 11262 HARTOLDMON0013211 -2- analysis of fluid from A/C ZG1 and A/C ZG2 showed 76 + 10% 500C. The ARB requires a minimum of 80/ 500C for a valid " flight test. Shaft Seals: ' (14253^3-JFH) The following cyclohexenyl olefins were synthesized, will be epoxidized and evaluated. D n c--- c n He- -c 0 C% Cy#? c> C.-0 // 1 c> uC-ftC.*/, It is interesting that- the above reaction took place .to yield I where the reaction between furan and dibutyl maleate yielded only starting material. Skydrol Reformulation (RWS-RCR) 1425043 Thermal stability tests were run on THP formulations con- taining an AIT enhancer and certain amines. The results were encouraging. THP containing 0.50 percent 2,1,3benzoselenodiazole servived 336 hours in the oxidative bomb test with a final acidity of 0.05. Metal attack was minimal and the fluid was clear and contained no precipi tates. This formulation, see Table I, had an AIT of 825 F, up 125F from THP neat. Those formulations containing 2,6-diaminopyridine and/or o-phenylenediamine in addition to the AIT enhancer were much less stable giving rise to metal attack and precipitates. - - As an alternate to THP based fluids we formulated three potentially lower cost Skydrol LD fluids using TBP/TCP and less expensive VI Improvers. These formulations sire given in the table below. Those containing Carbide's FH-25 acrylate VII, and R & H PL 10579 have better shear stability in the 10 hour probe test than Skydrol LD. Shaft seal, thermal stability, and valve tests are planned to evaluate these VI Improvers. 11263 HARTOLDMON0013212 orLi-Czc-H-i la.Ccfi -tup /. /d7 rte ic.sa'l. urns' / 3-0 7 l/['A*'/~l 7/ *> 07 r//-/5> *77 /%<>. VjAfr^dz (tS) . : '-<#> f /DO'F -S/O *f 3kb II.U J.R? ifs'ji-z /.ri'-n-i 7lp7. 7Bf 9-^7. TCP 17.01)7. PL/tfll l ZcPj' 0, n>7 p///?*o,xi7H>o &./t>7 7c/7 //.rj? rcA 7.507 Ff/ar A 50'7 0. ID? Ftttf: c. ?77. /M, 3<?/ //.i? 3.SX . -- II.rt 3.?7 t 23~t. o.9?r <5.W fl. "HP ^\\JL&ys VjLGtr - *?0 3 wv 6 -4,. /O &,. --. - -- - - '?.'?/ <?>v r.03 7.7? <?.V; l-ll' S-xf l.Tl 7.37 NtTBi 9 ! . L /0f7j /C ^ rj^r:.,/:rix . 2. Fft . , fU'fcnJ (dcjjt- SS'J !i //j bi-Ik.. ^ f'Z /?S'I f*rY'jC*~(c Aj*f\*rt'C*s frU-CzJt s.r^/Jk .L /ScL*(>TiL) *- /r-io 3- tisttsnryy oorr sryr/ifj lr r.i h 7//T.. -Oc/. m m ^ A. M HARTOLDMON0013213 -4- (1425243-RWS, RCR) Two shaft seal tests were run on formulations containing 3.4- epoxycyclohexylmethyl acetate and 4-methylcyclohexene oxide. The results of both tests. No. 54 and 55V were good; no leakage and little to no deposit formation. The formula tions were based on DBPP and contained the 500C additive package less FC-98 and Unox 221. The concentration of each epoxide was set so as to give the fluid an initial oxirane content of 0.35-0.3736. The major effort was directed at scaling up the synthesis . of these and other epoxides. 3.4- epoxycyclohexylmethylacetate. This material was initi ally prepared using m-chloroperbenzoic acid. A product of high purity (99.536 by G.C.) but mediocre yield, 51-5256, was obtained. A better procedure was needed owing to the relatively high cost of m-chloroperbenzoic acid (23.50/l00g). A procedure using peracetic acid was developed. (Peracetic acid is inexpensives #0.48/lb in bulk; #1.25/lb in gallons). This procedure is similar to that described in USP 2,838,524 and USP 2,716,123 wherein sodium carbonate and sodium acetate are used to buffer the reaction and retard formation of autoxy compounds. A 8456 yield was obtained. The product distribution was 92j6 epoxide, 8?6 olefin. This is similar to the epoxide/olefin distribution in Unox 221 (89/11). The procedure is adaptable to the other epoxidations we have planned. Methyl-3,4-epoxycyclohexylcarboxylate. This epoxide was prepared using the peracetic acid method described above. . The product consisted of 83.636 epoxide, l6.4j6 olefin. ' The synthesis is being Bcaled up-to provide sufficient material for several shaft seal tests. Two routes were used to prepare the starting material methyl-3-cyclohexenylcarboxylate. The esterification of the acid in methylene chloride was easy and direct; however, the acid is expensive (l9/l00g). The Diels-Alder synthesis was superior. The product was obtained in S8% yield from . the AlCls catalyzed reaction of butadiene and methylacrylate. This procedure follows that described in J.O.C. 30,3567(1965). Future . 1. Shaft seal tests will be run on 500C type formulations containing the following epoxides (prepared via peracetic acid a. Methyl-3,4-epoxycyclohexylcarboxylate . b. Dimethyl-3,4-epoxycyclohexyl-l,2-dicarboxylate c. 4-methylcyclohexene oxide 2. Bomb tests at 225F will be run on fluids containing each epoxide at 0.3536 oxirane in DBPP at 0.50 HaO. ' 11235 HARTOLDMON0013214 mret Z7S~e Gxwrtvp 3om& TksxvJfy Obi TKWtWLPMZPMTt FOjnuL/)ntbtS. V CjomPcsrnt/J /?%-/ 9 9 f /c 1 91 * \ 9Uf hip 1* (Mi4ty~)s'yf o.slV lt*o TSr F///AL tits, +b Metzl Armcf'- UZHGTH TAfJ 0.S0T/IN C/A^ Ho. LEz. / 7 336 o.os' -- - . V fOtL ~0> (L 6 0 H2% ~0.o7*~ d fc. Tilt wKWriH-in) h(*% j 0S% *,//3- BthpocJevfldieijclej 0. I % <a-dil YntK-tp'yncl O.S~2% /Uo% *` 33c, 'z.gjj ^zso - -0/3 i 0o -f.it ~0.0^ vi!n m 0.1$ A,/& --ObOls-Oitc*} -l.oi ~0.03 pIxr^xyifL-fj^ diam/hK* 0,S**l |tM% 1 Tiio'f.Tiifj I, Cl Umyv^t j 0. /% d, 4 *3) i . 0 (*7* )(*' ffa&lttWfJS ~ &/ M 1 to.5% *v?0 1iO.OX -6, 1 1 -? u .-.2/XZ fl.o3 0 O 1 11266 HARTOLDMON0013215 <Z7rF (DwDfiirtve Some 7esr/*jfy. EpOXlDE EVflLUfflfOp) tfJ 5D0C TYPG FLU !&. JJBP.__ Composmod TEST F/mu Mr5 To LshterN T7W O.SOTflfJ tf/ 76.37, DEft7; 1.0%. HF4t7.(3ot Atf 63/ ~Z<fo NfL SchJs); f0.o% pL'ftfsi(ib% Sift'd*)} 6.^7 ^a//om Cp/ze/de C-WffC, ot-ole h*J epoxide*} 0.fD% FP/3T-; 0.247. H2.0. =THL A}rtfcK CJ Qa* ail -0.33 Mo y Fe. fJfL * I3U-H-2. St). 107. 7,o7. HF+fz; (/#>32/-v0 /o.0V. PUfffi' ZtF% cyc/ofcwrfjc. <Wc; o.r?. Rmz; c. m% M I2ICZI-3. 7?.M7 kEft7; l.ottfF'hZ'j to. 67, W#jf7; e/cyy- cyuoH&cyinteWiL /ce77?rs, (0. 360 360 1 0.0- . *Tlf -0.02- -f.U -O.tfO 4 0X7 40.0(0 i|y\ /.23 ^32> 40.&f -3.3+ -0.11 ~3.t 40.03 4t M* to / GJ ^3 * HARTOLDMON0013216 Shaft Seal Evaluation (1425235-RLQ) Eight DBPP formulations were evaluated for shaft seal leakage and deposit. The two formulations which used diepoxide, tfnox 289, caused high leakage and deposit. The six formulations which employed other acid scavengers . did not leak although one had some light deposit. A formulation MCS 986, using Epi Res 5014 had light copper transfer from a thrust bearing cage to the races. The Seal Reports 48 to 55 are attached. (Pages 8-15) SST Hydraulics (1409035-RLG) Due to the Hydraunycoil B pump test failure by high wear and copper transfer in the AP10V/28oF (318*^ hot spot) pump, a lubrication test was devised to screen candidates before pump test. The test employs the LFW machine under the following conditions Test time Test Temp Ring Speed Load 24 hours 280F (or pump test temperature) 200 rpm 8 lbs (53 psi) with conforming Dynalloy >03HT bronze conforming blocks on hardened steel rings. The bronze alloy is the same as that used on the steel on bronze wearing surface in the AP10V pump. The forging conditions are the same since the test specimens were machined from pump cylinder barrel forgings. The weight loss results are: Test No. Fluid Wt. Loss mgs. 603HT block E9310 ring 1 Used Hydraunycoil B from pump test 36.0 5.1 2 New Hydraunycoil B 32.7 5.6 3 Skylube 450 0.8 0.2 The results for Skylube 450 were so encouraging that a 280F/AP10V pump test was started and at 210 hrs there are no operational problems. The wear correlation between the LFW wear and pump test seems to be correct when com paring the 156 hr failure of Hydraunycoil B to the Skylube 450 pump test. . If the Skylube 450 pump test is successful, the next test should at ldw temperature AP10V pump test with a Bendix valve test in parrallel with the load valve. The low temperature pump test would determine if K-salt micell formation blinds filters at the low temperature and the Bendix valve test will give a measure of the Skylube 4a . 0 erosion resistance. * * ilSSo HARTOLDMON0013217 -8SHAFT SEAL PERFORMANCE EVALUATION Fluid Formulation OR 1^02^1-5 91.72^ DSP? (A-D) 2.** HF-M12 . ^.65 Unox 289 0.02* Benzctriazole 0.25* HeO ' Summary of Results The total lealcage In 2l4 hrs la ^78 grams. deposit covers the stator seal face. A heavy resinous Visual Condition AP10V Shaft Seal "Bearium" Bronze Stator. Lealcage path from outer to Inner diameter. A heavy amber resinous deposit covers the entire seal face with a heavy concentra tion outside the sealing area on the atmospheric edge. Light wear Is under the deposit. * ft '9 7S* 0 20 40 60 80 100 120 140 160 180 200 -- Test Time (Hours'-- 1123d HARTOLDMON0013218 -9- SHAFT SEAL PERFORMANCE EVALUATION Fluid Formulation OR.l4038l-4 91.25* 3.40ib 4.85* ` 0.50* DBFP (A-D) HP-412 Unox 289 FH-132 Summary of Results 0.27* H20 The shaft seal had a total leakage of 6h4 grams in 212.0 hrs. The seal face has a white flaky deposit and no wear. Visual Condition AP10V Shaft Seal "Bearlum" Bronze Stator. Leakage path from outer to inner diameter. A moderate amount of white flaky deposit covers the center of the stator seal face. The atmospheric edge has a black sludge concentration while the fluid side edge has no deposit. No wear is seen under /.the deposit. Test No. 49 800 700* ... _ . ` .644 g at 212.0 hrs .. Tester No. C-2 Temp. 225 P Time to 120 hrs 5 irams/hr. Otter Stator thickness variation is 0.0002 in. 600* - E to ~r . , 500* 06 .. : to 400* ,, -tQJo> C4O0 _ to 300- 2004 100 i 1070 7 6 0 .Min ul.tTmiimt ~nijm ilririmrcwi.iuriM---- 0 20 40 60 80 100 120 140 160 180 200 . --Tes! Ti.rs (Hoiks'-- 11270 HARTOLDMON0013219 -10- SHAFT SEAL PERFORMANCE EVALUATION Fluid Formulation MCS 966 77.?- DBPP 7.2, HF 412 in DBPP (30? Sol.) . 10.0? PL 4857 in DBPP(30? Sol.) . 4.50? Epi Rez 5014 0.0536 FH-132 0.22? HaO Summary of Results In 213.7 hrs no leakage occurred. The edges of seal face have dark staining but no deposit is found. Visual Condition AP10V Shaft Seal "Bearium" Bronze . Stator. Leakage path from outer to inner diameter. The edges of the seal face have a dark staining while the center of the face is clean. The seal face has a . uniform smearing wear. Test Ho. 50 Tester No. C-l Temp. 225F 800 700 600 500 Time to -- 5 grams/hr. 400 . 300 Other Stator thickness vari- 9Qn ation is 0.0003 in. Bearing cage had light wear and copper trans- jqq fer to balls and races. 4. . .. . . ;* .......i I no leakage - IOYO 0 20 40 60 S3 100 120 140 160 160 200 . - - --Test Ti,7,e fHours- -- 11271 HARTOLDMON0013220 -11SHAFT SEAL PERFORMANCE EVALUATION Fluid Formulation OR 140361-1 is.yn EB?p (A-D) 7.0jb HF-412 In EBPP (5056 Sol) . 10.056 PL-4857 In DBPP (3056 Sol) 6.256 Olefin Epoxide 0.556 FH-132 0.2456 KsO Summary of Results The seal 8tator has light staining and no leakage was observed during test duration of 212.3 hrs. Visual Condition APIOV Shaft Seal "Beariuro" Bronze Stator. Leakage path from outer to Inner diameter. A light rust colored stain covers the seal face. The stator has light wear. Test Ho. 51 Tester No. C-2 p. Temp. 225*P Time to 5 grams/tir. 400 Other Stator thickness variation is 0.0003 in. 300 . 200 + T I 100 J no leakage ft 1070 V'f* 20 40 60 80 100 120 140 16: ISO 200 -- Tfcsi Time (Hours)-- 11272 HARTOLDMON0013221 -13- SHAFT SEAL PERFORMANCE EVALUATION Fluid Formulation OR l4o38l-J 80.10& DBPP 7.0* . 10.055 0.555 HF-412 (3055 Sol) PL-4857 (3055 Sol) FH-132 2.4o?5 Cyclohexene oxide 0.2455 H20 Summary of Results The leakage wa6 1.6 cc in 262.4 hrs. The seal face has staining and light wear. Visual Condition AP10V Shaft Seal "Beariura" Bronze Stator. Leakage path from outer to inner diameter. The seal face has a bright copper color surface In the center with rust colored stains on each side. The center of the seal is lightly worn. Test No. 53 Tester No. C-2 Temp. 225*F Time to -- 5 grams/hr. ` Other Stator thickness variation is 0.0002 in. 800 f] 700 . _i.. vt . j 600* u . S* I i 500 (Q o * :: 400- vVre>* J: 300- ! 200 ** l 100. - -- - ; 1.6 ce leakage in 262.4 hre e?e ' 0 ii 0 20 40 60 80 ICO 12-: 140 160 180 200 - --Test Time (Holts -- 11273 HARTOLDMONO013222 -14SHAFT SEAL PERFORMANCE EVALUATION Fluid Formulation OR 152601-1 78.7* DBPP (A-D) 7.OS HF-412 (30j6 sol) 10.OS PL-4857 3.8* 3,4-Epoxycyclohexyi methylacetate 0.5* FH-132 0.22* HaO Summary of Results The total leakage was 1.8 cc in 234.8 hrB. No deposits were present but the seal face has pitting and light wear. Visual Condition AP10V Shaft Seal "Bearium" Bronze Stator. Leakage path from outer to inner diameter. The center area of the seal face Is pitted and .lightly worn while the edgeB have light rust colored stain and little wear. Test No. 54 Tester No. C-l Temp. 225P 800 700 I i : ^ --t ' ; Wl ; 600 Ero H- 500 Time to 5 grams/hr. Other Stator thickness variation is .0.0002 in. 400 0to .r 300 -j 200 100 4- ! - --:ii ! 'i .. ! \ .4- 1*6 cc leakage In 236*4ra 1070 20 40 . 60 80 100 120 140 160 183200 --Test Time (Hours'-- . 1. 1. 274 HARTOLDMONO013223 15- SHAFT SEAL PERFORMANCE EVALUATION Fluid Formulation OR 152601-2 Bc.OJe DBPP 7.056 IH-412 10.056 PL-4857 ' 0.5 PH-152 2.556 4-methylcyclohexene oxide 0.2256 HaO ' Summary of Results -- ' The leakage was 1 drop In 200.4 hre. Light deposits and stains are on the moderately worn seal face. Visual Condition AP10V Shaft Seal "Bearium" Bronze Stator. Leakage path from outer to inner diameter. The stator race has a light resinous deposit on the atmospheric half of the dynamic seal area. A thin black stain-deposit is on the center area while rust and copper color stains are on the fluid half of the seal face. Moderate wear is present. * Test No. 55 Tester No. C-2 Temp. 225 F Time to 5 grams/hr. Other Stator thickness variation is 0.0002 in. 1070 BOO 700*1 4/) 600 h I 500 <0 CO 400 --. ;. 1 . J: _:. JI. ....... - : J j. " TTi.1 m IS.j * -7--1 1.:. m T-1T --rrrrrr -r ' . !: n 1 : I . : ... ; . ----- ----- I--------- ----- r- ;.T- : 300 . 4I . ....- I 200 --MH 100 4 0 20 40 j ; ._u. ; <t '.1i ""I drop Inj'200.4 hrs*; --^ ... I \ 60 BO 100 120 140 163 180 200 Test Time (Holts' -- tt275 HARTOLDMONO013224 * -15- 5HAFT SEAL PERFORMANCE EVALUATION Fluid Formulation OR 152601 -2 . 8C.*0 DBPP 7,0% FH-412 10.0$ PL-4857 * 0.5* FH-132 2.5$ 4-methylcyclohexene oxide 0,22% HaO Summary of Results The leakage was 1 drop in 200.4 hrs. Light deposits and stains are on the moderately worn seal race. Visual Condition APldv Shaft Seal "Bearlum" Bronze Stator. Leakage path from outer to inner diameter. The stator face has a light resinous deposit bn the atmospheric half of the dynamic seal area. A thin black stain-deposit is on the center _ area while rust and copper color stains are on the fluid half of the seal face. Moderate wear is 1 present. * Test No. 55 Tester No. c-2 Temp. 225 Time to 5 grams/hr. Other Stator thickness variation is 0.0002 in. *1070 60 80 100 120 140 160 180 200 --Test Time (Hours) -- 11276 HARTOLDMON0013225 -16- Metals In Solution (1425235-RLG) ' A 24-hr. LFW/280F/200 rpra/8 lbs load test using Mueller ' Bronze Alloy 603HT on E9310 steel was performed on, Skydrol LD, QL 30. The metal combination is the same used on most eliding bronze on steel surfaces on the AP10V pump. The bronze test block was conforming to the steel ring which allows wear with little change in geometry which is more simulative of pump wear than the 4-ball test. The weight loss results are: 0.3 mgs on the 603HT Block 0.4 mgs on the E9310 Ring ` A 72 hr test under the above conditions will be run too, if wear is accelerating and to determine if copper solubility is accelerating between 24 and 72 hrs. Skylube 450 (1432159-SLR) A Skylube 450 production run was made in the W-Bldg. facilities, totalling 2000 gal in 5 batches. Some plugging of the final cartridge filter was experienced, but it was determined that this was hot due to the fluid, but rather to rust and scale in the system. All five batches were satisfactory on foam, Pb corrosion, Ryder and 0 & C. Two additional batches, representing reworked QM-5, an earlier batch that had been double-charged with amines, gave high Mg corrosion in the 0 & C test. A lab reworked sample gave a good 0 & C test. The W-Bldg reworked batches will be further checked. Analyses (by RAL) on the latest samples of used Skylube 450 from the Northeast Utilities industrial Jets indicate continued satisfactory performance. One engine has over 620 hrs since oil drain. Stauffer Jet II is seeing considerable service in industrial engines. The lower viscosity of a sample of Stauffer's used oil compared to our sample of fresh Stauffer Jet II suggested that a VI improver might be present in their oil, but sonic shear and other tests do not indicate the presence of a VI improver. Analysis of the. base stock shows it to be a PE ester of butyric (9.5^)* hexanoic (0.5*) and heptanoic (90$) acids. Analysis of the additive package is incomplete. We have prepared complex ester salts in which chloromethylphosphonic and methylphosphonic acids replace the phosphorous acid. These salts will be evaluated in Type II formulations. We will be looking especially for possible improvement in load-carrying properties compared to the phosphorous acid complex ester salts. 11277 HARTOLDMON0013226 -17- Tentatlve Processes (1432259-SLR) - We have duplicated the literature preparation of 5-nitroindazole, a 5-AI precursor. Modifications are being studied to adapt the method to pilot plant size preparations. A . problem which may complicate self-manufacture of 5-AI is . availability of an intermediate, 2-methyl-4-nitroaniline. American Cyanamid has stopped production of this material. Central Purchasing is seeking other sources. Nyco Fluids (l**32359-SLR) We have confirmed that addition of our complex ester salt to Turbonycoil 93# while showing some improvement in 0 & C /. tests, does not bring it up to Skylube ^50 performance. Comparisons are shown in the accompanying table. - . CRD acids (1432459-SLR) . Enough Ct and Ce acids are purified and sufficient Cs acids are expected shortly from CRD to prepare a base stock sample for Ryder Bcuff test. Inadequate agitation in the 3-gal glass-lined autoclave will necessitate further modifications or purchase, of another reactor for synthesis of large enough samples of acids to make fluids for bearing rig and fatigue tests. CS7ft S41TS SA- * e> *' e. r -srs ?i y/i/vc.. -r/ fyC- /I t'-Cfc- *>.}( -jd.oz -0.7.2 // Ja.s'i, + 4 -o.&g -j.jp- //0 K-stLT ysoo 10.!b -*0.0? 10.6?. --0.!jS + 0.g/ +0.0 2. +9^1+0.01 +0.i! 11278 HARTOLDMON0013227 -18- Santomelt (1405343-JFH) Technical data Is being collected to present to the Air Force for Santomelt's use at their Tactical Air Bases. The required performance properties required are: ' Our Results Immersion corrosion Stress corrosion Magnesium corrosion pass pass report The magnesium corrosion data collected Is below. mg/in2 Distilled water (control) 0.45; 0-.39 10% Santomelt In distilled H20(pH=7) 38.4 10% " "" " (pH-10.5) 34.1 25% " "* " " ipH-7) 341 25% " "\ " (pH-10.5) 320; 343 Since the Air Force Is Interested In the 10% solution values our data Is well within their desired values. Santomelt for Commercial Use - The three problems facing Santomelt before complete acceptance are: a) Aluminum Sandwich Corrosion b) Hydrogen Embrittlement/Stress Corrosion c) Magnesium Corrosion * The seriousness of the effect of magnesium corrosion Is not completely resolved. ` - Progress was made In controlling 'the aluminum sandwich corrosion by increasing the concentration of the sodium meta silicate used. At present it is a toss-up as to include the formamide in the Santomelt formulation or not from a technical point of view. From a raw material cost picture formamide should be excluded. . A test was developed to evaluate the hydrogen embrittlement area and to ascertain the difference between stress corrosion and hydrogen embrittlement. Work in this area continues in search of a solution. . MCS 990: 75% formamlde/20# urea/5# HgO - a deicer developed by Canadian Dept, of Transport, ran into hydrogen embrittle ment problems in Air Canada's testing. Since we developed a process and agreed to custom manufacture MCS 990 for Canadian DOT, we had the hydrogen embrittlement test checked by an approved independent laboratory. Test results show the product as manufactured passes but a 1:3.dilution with distilled water fails. ^ ^ _____ - 11279 HARTOLDMON0013228 -19- DJDUSTRIAL FLUIDS (WHC, PDF, PFH, LFS, LRS) Pydraul Reformulation Pydraul 625A "* ' An additive package has been developed to improve the pump performance of 625A. The V105C pump had a total weight loss of 20.2 mg which is equivalent to the previous per formance record (Fyrquel 220-18.4 mg total) and signifi cantly better than Pydraul 625 or Fyrquel 550. A process amendment to incorporate the additives is being i8sued. Samples of the additive version of 625A and Fyrquel 550 have been sent to Alcor for deposition testing. The oil gear test is at the 450 hour mark with no apparent wear. The raw materials cost of the latest version of 625A is #1.77/gal versus #1.79 for 625. Thirty-five gals of Pydraul 625A identified as Batch OR 139046 was blended for laboratory pump testing (V105C). 49.log wt 49.50g " 0.30g " l.OOg " 0.10g " + 0.02g 20 ppm Aroclor 5442 Santicizer l4o Santolube 70A Unox 221 Benzotriazole FC-98 DCF 200, 350 cs The above formulation was tested in the V105-C pump stand with excellent results, total weight loss amounted to 20 mgs. Pydraul 540A . The V105E results on an additive version of 540A (TBP and Cu inhibitor) look excellent. A V105C test is underway. The tentative process for 540A is being written. The raw materials cost of 540A is .87/lb vs. Sf.83/lb for 540. Five gals identified as Batch OR 139054 was blended for laboratory pump testing (V105E). 11280 HARTOLDMON0013229 -20- 39.00* wt 15.10* wt 32.00# wt 11.00*.wt 2.50* wt 0.30* wt 0.10* wt Aroclor 1242 " 5442 Bright Stock 200 TCP TBP Santolube 70A Benzotriazole + 20 ppm DCF 200, 350 cs The above formulation was pump teBted in the V105E screen ing pump with excellent results, total weight loss less than 10 mgs. A second batch of Pydraul 5^0A (4o galB) was prepared (OR 139055) and is now being pump tested in the V105C pump stand, a portion of this fluid iB reserved for an Oil Gear Pump Test (scheduled for late May). Oil Gear Pump, Pydraul 625A (OR 60697) has now accumulated 500 hrs.j indications point to a passing test. Pydraul F-9A * A screening pump teBt of the Santicizer 711 version of F-9A is underway. If these results look encouraging, further properties will be gathered. Pydraul A-200A * . Testing will be completed as soon as possible on the 1242 version of A-200A. Two non PCB versions of A-200A will be evaluated. Highest priority will be given to the TCB/5^2 blend. The other will be a Arobromoclor 1232/5442 combination. - Pydraul 312A After four runB on the computer we have evolved component viscosities which blend in a linear manner. The latest printout had a measured viscosity at 100F within 2 cs of the pre dieted value. The next round should be within 1/2 cs. Once the linearization of viscosity is accomplished, the computer program can be used for. the calculation of Pydraul 135, 230, 90 and 5^0 blends. . The calculated FR level was verified on a limited scale by full scale fire testing. A blend containing Arobromoclor 1231 was equivalent to Pydraul 135 on the PM spray and hot manifold. It was better than Pydraul 312 as predicted. The composition of TCB/5442 and Arobromoclor 1232/5442 blend at various levels of FR follows. 11281 HARTOLDMONO013230 -21- Pydraul 155A It appears possible to remove PCB's by utilizing TCB/5^55. However, the sediment formation in 5^55 blends wilJL have to be stopped. Anniston is working on the problem. Analysis of the filtrate from a 5^35/TCB blend showed a mixture of mono, di, tri, and tetra chloro terphenyl with about 70# being dichloro. Anniston doubts these results and is running their own analysis. They believe the material to be slightly chlorinated quater- phenyl. The sample of 5055 sent for distillation by QET ' analyzed as 5^.2# Cl. Anniston is working on 55-57# Cl materials. Gene Arnett feels that the specific gravity changes rapidly enough below 57# Cl to enable process control to + .5# Cl. ` ) , U2S2 HARTOLDMON0013231 -22- ,C-Po`rC'-J'rO'f?ZV-).ttt*i;nJ t*\>Jrr' - *fc| M'ulZ C\`rr ' t'fbn*. Dr-S. Tp 7 ' C D? TCB 544 2. C*/ c i//IS c f' j `" ^ f A- fc.v? 1 Vt'? & jOO~i- ta\<L be.nSiipZGt Co'S't / \b HC,tui3) Densrfjj Co*.is/3 1 b - o.O -to 1 3 4O lh-ti--o.n(> 2:H- ifc-- " ' `.'S jr* f " Pj.^i > 71I 1 0 1 So.Z% .5 4.5 2 4. A 3$. 670/^ 64 '(7 /. ;6 / /.oTf 3 5.'"7 2- 5 4-S 2-1.7 32.6 6 7. 6 5.05 A / SO l.lU /.S63 4M 7.5 0.5 2 4/S 2 7. { 6 / :;.--- ' /.<?T7 fl^rr M &r.e. BrIS. T5P ; CD? ' A Bl2SZ . S4-42. 2 0:3 0^ Cr> 4-5 36.7 2 7. <2 34.7 2.S 4.5 36.7 2 /.O' 0 <> 7- _ r* 4c Sj jy---0c: C 3 / C Vl5CG> 1 OOrf f> C x L'r. ] Vii,6 IOOr`T C(i/c. Df`/:t- -`c Cf/,it/)!z & 7.0Ac - -- L/ C. 7- ^2 O <] 7 6 7.0** 1>C'.5 5/v . /. 1 n 3 * o^S A / o.v -** - /. / 6 7 D ^ `'i * ^ i. 1 20 ^ ,&44 . . ' * , 11283 HARTOLDMONO013232 -23- Pydraul 625A, AC-A, AC-A W,G, 28 . Viscosity-temperature curves, Timken loads, relative drop sizes, and Falex Wear and Load were obtained on the rev ised air compressor lubricants for a revised sales.brochure. These values follow: *- Pydraul 625A Pydraul AC-A Pydraul AC-A, Winter Grade 28 Timken O.K.Load - 30 45 24 Film Strength 11,700 15,100 8,600 Listed below are results of Falex Load and Wear . .Tests on Pydraul AC-A, Pydraul AC-A Winter Grade 28 and Pydraul 625A. Tests were run by the Chrysler Procedure. ' Falex Load Test Passing Load Pydraul AC-A Pydraul AC-A, W.G. 28 Pydraul 625A 4025 lbs 4175 lbs 4250 lbs Passing load was determined by the average of two tests* Falex Wear Test _ Wt, Loss,Pin Pydraul AC-A Pydraul AC-A, W.G. 28 Pydraul 625A 0 . 0.6 mg 1.1 mg Wt. Loss, Blocks 0.5 mg 0.9 mg 5.8 mg 11284 HARTOLDMONO013233 -24- Drops/ml at 25C Mobil DTE 24 Mobil rrE 26 Pydraul /,C Pydraul AC-A Pydraul AC, W.G. 28 Pydraul AC-A, W.G. 28 Pydraul 625 Pydraul 625A Fyrquel 550 ' Fyrquel 220 * Fyrquel 150 Fyrquel .300 ' 37 36 4l 39 39 4o 43 39 ' 37 37 37 37 The nev; formulations are completely miscible In all portions with* Fyrquel 150, 220, 300, and 550; Houghtosafe 1010, 1120, and 1130; and our nev; phosphate esters, MCS 030E, MCS 050E, and MCS 0115E. , 11285 HARTOLDMON0013234 HARTOLDMONO013235 kinematic vi'.cosnr. ccn ns foxes 18ZTT ` KINEMATIC VISCOSITY. CSNTIVTOKES HARTOLDMONO013236 saxairiiii.ia *Auso3t;iA ci.ivvnut* KKIM-I10-H -rt--H-M-*t-N-M-l IINMHMHNN HIS* In in HO H tt Id IN M Pydraul Measured Viscosities !n Centistokes . Pump manufacturers recommended operating range between 13 cs. and 1000 cs. KINEMATIC VISCOSITY. CENIISTOXES KINEMATIC VISCOSI IT. CENTISTOKES 8S3TT * if1 ' - ' ; i i i i i _' i i i i i i i i i ' i. i i i-1 i _ i ! . i . i i _ i i i i i i i i i i i i i i i i i ! i i(( -ii -;o-to-it-t-)t-;o-is it n 10 u ti to him iro ita iso no ivo 210 t*0 jeo o 100 no jo sco no too .o o uo ;o jco us jm s;s too no ;s >n ` TEMPERATURE, DEGREES FAHRENHEIT HARTOLDMONO013237 VISCOSITY, SAYCOLT UNIVERSAL SECONDS M -.* .1* II H N tt tt H M N IN III IN IN IN IN IM IN --r~ ill.. ' " " jane \ & \ . \ xnv 10000 A A XI X it:: <>:: im 1 !------! 1------- ! ____ ' I V. 1 :----r*-- 1 _ xi 1 'V 1 *V \\ ^ a\ VVIi ::c2 ui: : 11:? IM : i_____ _______ !* ! , i '\ i \L \ X, X V s, \ S, 7|NJ *-,A \ s Pis o itt o 3 JW to ji i 11 <1A1 m --------1-------- cJ* 111 X X V|;\ V \fcjVlV s-- \ VIk \|V \ * XIMX. N s> --\ \ S, > Pydraut Measured Viscosities ; In Saybolt Seconds Universal Pump manufacturers recommended operating range between 70 SSU and 4000 SSU. o>. O .* **... PYCRAUL 540 * PYDRAUL 312 PY0RAUL 230 * -0 * PYDRAUL135 PYCRAUL 90 ----------------- f''/ L/t-fA L* l_ /| l-- -/M n r- 10) I) w >L' 19 >* |fl \z w 5 u ^M i I y \SI 'N s ] \ *v V \ \ \ X k. V \ \ X PS \\ X s Nv V\ s .V X k1 X ti \ iA 1 * S \ x VsAt V b % sS >` X* . * 41 0 H* K> OJ , \ s MO ICO 7C0 7C0 IM 100 10 19 10 to so H to -- 71 -ia -i0 -to 1 a J 0 70 1 1 V1 1INIj. i1 IC 1 0 IN 1 10 MO IM HO 170 1 10 l<M) 7 M 1 0 1 N t 10 NO 7M NO 710 710 770 7C0 >10 770 )) 70 7M TEMPERATURE, DEGREES FAHRENHEIT HARTOLDMONO013238 HARTOLDMONO013239 VISCOSITY, SAYUOIT U.'IIVCRS.M. SECONDS vucocir*. cayj o it u t:iv i :sal sicorios - VISCOSITY, SAYUOLT UTlIVCnSAl SICONUS HARTOLDMONO013240 -31- German Mine Fluid A Tentative Process was written for Pydraul 28l, a revision of Pydraul '280 which provides improved low temperature performance and eliminates the use of Aroclor 1250. Fluid Recovery and Reclamation Initial work to assess recovery of Pydrauls from customer effluent streams has been started on samples from Johnson Motors, Central Diecasting, Lester Castings, and Timken Bearing Co. Qualitative results indicate large amounts of solid material present with the oily layer. Filtrations have been difficult and relatively large amounts of fluid remain with the solids. This work will continue to provide more data on quantity and quality of fluid recovered as well as suggesting methods for recovery. Turbine and Air Compressor Lubricants European Steam Turbine Control Fluid ' t Samples of Geigy Reolube TF 90, Reolube HYD 110, and tri- xylyl phosphate were screened for possible use as steam tur bine control fluids in Europe. Of these the TF 90 material was considered the best candidate at this time. Tests run to date included 350F 0 & C, beverage bottle hydrolytic stability, and a screening pump test and wick fire tests on the TF 90. MCS 979 The field trial at the AGTL Cremona station has passed 800 hours with no problems. The inner bearing on the DeLaval booster has reached 192F on several occasions and AGTL is keeping a close watch on the situation. Tom Oneson has requested us to formulate a lower viscosity fluid in antici pation of warmer weather. We would then be going back to essentially the same fluid as MCS 97^. Perhaps an AIT enhancer could be utilized to upgrade the FR of MCS 97^ but it is roughly equivalent to Pydraul 135 now. .. Water/Glycol Fluids . A Tentative Process for Santosafe W/G fluids blending using Carbide concentrate and thickener has issued. A capital project for equipment to blend these fluids at JFQ plant has been approved. . A possible relationship between lubricating properties of W/G fluids and the cloud point at higher temperatures has been checked, (polyglycol thickeners have an inverse sol ubility and come*out of- solution at higher temperatures.) '' 11232 HARTOLDMON0013241 02- HARTOLDMONO013242 HARTOLDMONO013243 -34- 11295 HARTOLDMONO013244 -55- pJU*^cL CMfapU'&u iJlju^U^d^uk. \socp UfrrA Uy fy-tfah '^00 M iqi~zooF ^/ V iU^K fly&Cdk. I50L r fio$-ZIZ lAwjdcto*-*/-*1- 5*20 ^,/5*22.fl A^4> w/6 100 /? :) W<ye^(f F/Zt / 50 505 I Z 3 > 5- & JL^t?Mu *J 0 ?e'* ,< t M.sC*l& 4 2.23 0&^a V/ttsh f4*i'fA-&r??/c> & z-o 7/ J2/<*4-*-v\. p /--3 /3 *10 -75 8 /> >2^0 t v/ 7/*p.OO-'Z/o Jo t/< Vy&ts I4-7J? `Pw4j.Vo V*^S 2b Su^^-C. 3o ~Z,? w/G 0>^-' c~- ft./ >z3o 11296 HARTOLDMONO013245 -36- Phosphate Ester Hydraulic Fluids . Pump test results on the pilot plant prepared cumylphenyl * nonylphenyl phosphate ester are listed below. '* MCS-050E (OR 142228) (NCPDPP Ba. #1)- 220 SSU which was pre pared by D. L. Kocher was pump tested In the V105C pump stand with excellent results. Total weight loss was less than 20 mgs which Is comparable to Fyrquel 220. MCS 050E (OR 139050) 220 SSU which Is Identical to the above- formulation, but using Pilot Plant Batch #2 (NCPDPP) was pre pared - 65 NCPDPP Ba. 2 3556 TPP This formulation Is presently being pump tested in "the V105C pump. Preliminary inspection indicates a reproducible test. V105C Pump Stand "J" is presently being modified for testing phosphate ester fluids. MCS 030E is scheduled to be pimp tested in this stand; start-up is scheduled week of 4-26-70. Phosphate Ester Hydraulic Concepts (1315635-KLG) Two base fluids NCPDPP and 0XDPP, one fluid with a surfactant additive NCPDPP + 0,04 FC 98, and two competitive fluid formulations Fyrquel 220 and Fyrquel 150 were investigated on the Boeing streaming potential apparatus. The results indicate that Fyrquel 150 and Fyrquel 220 are . similar in streaming potential and wall current while NCPDPP has lower streaming potential and" higher wall current. The addition of FC 98 reduces the streaming currents and potential. OXDPP has stream currents similar to the Fyrquels. The attached graphs have streaming potential and wall current data vs. volume flow rate through the apparatus. 1 1129 V HARTOLDMON0013246 K* Lm C3O*9CCAU*fCTMLMf tC 447r 7920 * I HARTOLDMONO013247 HARTOLDMONO013248 WE LOGARITHMIC 1 4T 7010 XRVCUCFtfCctt.fi 8*av( it CO.H Iff ' VOLT* & V> ItoPCOAIVCE o * HARTOLDMONOOI3249 11301 HARTOLDMONO013250 -42- Thermal Stability of Cumylphenol (P.P. SPL. 37) (H. Luebke & E.Emery) INTRODUCTION ' * ** Thermal stability data for cumylphenol were requested to provide process safety information for operation in the Pilot Plant. A maximum pot temperature of about 280C is used for the distilla tion . * SUMMARY . The thermal stability of a Pilot Plant sample of cumylphenol (Spl. #37, 3-5-70) has been determined with the Monsanto built pressure DTA apparatus. The experiment was run under a nitrogen atmosphere. Only a very minor exothermic reaction was observed above about 280-C. Gas evolution was detected at about 210C and a rate of 0.4g - millimoles/min/Kg (9cc/min/Kg, STP) was measured at the maximum proposed pot temperature of 280C. A plot of gas evolution rates vs temperature is attached. Gaseous decomposition products were identified by GC/MS as carbon dioxide, methane and 2 and 3 carbon hydrocarbon fragments. We- foresee no serious process hazards due to thermal decomposition of this compound. The rates of gas evolution are low and the slope of the curve is quite flat, indicating a small temperature coefficient. Reference - DTA #70-28 ' 11393 HARTOLDMON0013252 (``Y/uiw/se/M'l/W-il) V+tg V'-j-'AI^'3 '' oa kiM> * ; '. 'l l a i KU MtOUOIf ItlAIO M II bll>l I -. I 0(00 Of 3IWH*IUVOOTI** LoVk HARTOLDMONO013253 -44- Marketing Support Citgo FR-23 . A recent Citgo fluid similar to Pydraul 312 has been characterized as follows: i 1) 65> Aroclor 1248 ' 13# Phosphate ester; triphenyl ' 9% cresyldiphenyl 10.5^ Petroleum Oil. (Same i-iw & 'MW distribution as found in Citgo CHX-6o4) . 1.55 VI improver, (same type as found in Citg? CHX-6o4) ` 2) Citgo CHX-6o4 Citgo FR-23 Vis at 100P, cs: 70.93 TVTiis at m210 F, cs:. r7r.1^ 22 V.I. 48.4 Sp.G. at 25C 71.6 3) FR of subject sample as defined by the 'Wick Test* -(insufficient sample for normal FR determinations) indicates an improvement over Citgo CHX-6o4 which was in turn rated a definite pass by the F/K Hot Channel and H.P. Spray tests. Basically we are dealing with the same components as found in their earlier fluid, CHX-6o4. Th*e increased FR noted is undoubtedly due to less petroleum fraction as compared to the earlier fluid. (See 10/9/69 memo WRR from PFH) Both of these fluids are in the 300 SUS range; however, the VI of FR-23, is roughly 20 points higher indi eating a more potent VI improver and possibly more shear stable since it was reported that cheardown was a problem with`the CHX-6o4 fluid. We do not plan to characterize the VI improver at this tine* It is our opinion that they have incorporated more potent and possibly a more shear stable VI improver and have reduced the oil content to achieve similar viscosities. r 11305 HARTOLDMON0013254 -*5- Hamworthy Compressors/Pydraul 625 Three compressor parts from Hamworthy Compressors were tested for compatibility with Pydraul 625 and Pydraul 625A. The parts were of a perspex compound 'and the results were negative. The parts were cut in half and then partially immersed in each fluid for 168 hrs at 1586F. * . The parts did not change greatly in dimension but some dissolving did occur. This caused the fluid to become very slimy. Another part which was a gasket for a flow meter became very swollen and 80ft. These parts would not be recommended with either fluid. Recovered Pydraul 312/G.M. Bedford Recovered Pydraul 312, Findett lots CF 56 and CF 57 appear to be nearly equivalent to new Pydraul 312 as Judged by the Factory Mutual Spray fire test. Pydraul 135 Acldlty/Republlc Steel Laboratory degraded Pydraul 135 and new fluid were furnished to R. R. Klinge of Republic Steel to serve as standards for measuring fluid acidity. ror these two xpectsd as fluid 33 fcllovrad with Two characteristics are noted for degraded fluid. The initial p" crops lever and nay eventually get into the strong acid region below pi I Ji. Also, nore weak acids are being fometi and the titration curve flattens and extends to pi! 11. ) 11306 HARTOLDMON0013255 46- HARTOLDMONO013256 -47- Asslstance to Manufacturing Recovered Aroclor 1242 for Pydraul Blends "* Specifications for Aroclor 1242 returned by G.E. and ' Westinghouse for use in Pydraul blends were suggested (see following page). Steps are being taken to allow the plant analytical lab and manufacturing department to work directly with Findett on approval of recovered fluids for use in Pydraul blends. .- Pydraul ?20 Customer Complaint . - Gross separation of Pydraul 520 was reported in Japan. This appears to be caused by exposure to low temperatures. About 45# 1b In the top layer and 55 below. These layers remix readily at room temperature (75E). Fire resistance of the two layers as Judged by a pipe cleaner wick test is Birailar to the complete fluid. Interim recommendations will be to remix fluid at room temperature before using. Further .attention will be given to this problem as these fluids are reformulated to eliminate PCB's. pydraul Acidity Specification Limits Currently produced Santolube 70A has a slightly lower acidity than the old Santolube 70. This results in problems meeting minimum acidity spec limits for some pydraul blends. Each 0.1# of Santolube 70A contributes 0.05 to 0.06 TAN to acidity of finished fluid. Specification limits will be changed as needed. 11308 HARTOLDMON0013257 -48- RECOVERED AROCLOR 1242 TENTATIVE SPECS FOR USE IN PYDRAUL BLENDS Recovered Aroclor 1242 for use in any Pydraul or Santosafe Recovered Ar-oclor 1242 restricted to use In * Pydraul 312 and Santosafe 300 Series* JFQ MetP No. Color, ASTM D-15QO Appearance, visual Neutralization Number, mgKOH/gm Moisture, KF In methanol Sp. Gravity 25/25eC Viscosity, cs at 10QeF Oxlrane Oxygen GLC Analysis 1,0 max 3.0 max Clear, light yellow liquid free of sedi ment. Clear, yellow liquid free of sediment. 0.03 max ## 100 ppm, max. 0.03 max 100 ppm**ma 1.380-1.395 16.1-18.6 1.350-1.395 16.1-18.6 0.1# max. 0.1# max a) <0.2# max area low boiler peaks near new 1242 peak envelope. b) no low boiler peaks far re moved from new 1242 peak envelope. c) other peak ' typical of new 1242. same ie 8 7 11 12 # Includes Pydraul 312, 135# 790, 230, and 320; Santosafe 300, 150, and 100. ` ** Water contents as high as 180 ppm should make Pydraul with acceptable moisture and anpearance; hov?ever at this water level Aroclor 1242 alone is very cloudy. *** Procedures to be supplied later being run by Research. * Currently ' 11309 HARTOLDMONOOI3258 dielectrics -49- Biodegradable Capacitor Fluids (R. H. Munch) 'A modified Aroclor 1242 designated Aroclor 1242B has been produced by fractionating Aroclor 1242 to reduce the penta and tetra chlorinated' components to a minimum value. The .new product has only . 16% pma, 2.5% of the amount in standard Aroclor 1242 and' 16. 3% tetra, 67% of the amount in the regular material. Biodegradability data for the new product are tabulated below along with similar data for standard Aroclor 1242 for comparison. Peak 1 * Peak 2 Peak 3 Peak 4 Aroclor 1242 Amount Remaining (after 1 week; Nil 66% 64% 42% Aroclor 1242B Amount Remaining (after 1 week) Nil io% 33% 1% These data seem to indicate that the Aroclor 1242B is significantly more biodegradable than Aroclor 1242. The attached sheet gives the specification .properties for Aroclor 1242. Those which have been determined for Aroclor 1242B are given on the same sheet. These data indicate that the new fluid could be substituted for the old in capacitors with no modifications in capacitor design or processing methods required. Arrangements have been made to produce 500 to 1,000 gal. of the Aroclor 1242B. The fractionation is to be carried out in the Dayton MRC pilot plant because no suitable still exists in our St. Louis pilot plant. The adsorbent treatment will be done in the St. Louis pilot plant if the truck strike docs not interfere. The large sample of the new product is to permit use tests in capacitors by our customers. Preliminary work indicates that a product having amounts of penta chlori- ' gated biphenyl too low to detect and tetras reduced to about 1.3% can be produced by fractionating Aroclor 1132. The fractionation must be done so as to reduce the concentration of the 4,4' dichlorobiphenyl isomer to 2.0 2.3%. If this is not done crystallisation takes place. Further work will be done to determine the best procedure for making a product of this type. 11310 HARTOLDMON0013259 V/- f^n<rrr?x ./C. iu 0*0*NIC CnCMiCALt OIVISION \XXlSTON*'-AND KRUV.MR1CH. FINISHED PRODUCT . SPECIFICATION *0UCTION P. E. Heisler tbCI W. E. Ault. (CNCKikOCioirriON . ' PNOOuCT -50- joC*T. Aroclor 1242, Electrical Grade | -246 *>L&NT WGK, Ann. OAVk CKfkCIlVt mcsucT cuol. 81101 - 82612 >aiu cool u>>: s.: cs 6/17/66 1040-240-04/11/16-03/09 AFPMOVAkJ KC3LAHCH R. V/. Bucknell CUALITV COATAOL W. J; Gresham . * .* Q. xi. /0'lir. . . ., CHCUiCAU rOKUUCA . 27 . Alsor. known as Pyranol 1499 * * . * `# *4. WT. I . * .. upcnacoc* iocci. or 4-13-62 - poh *nauy;ii 3 x.5 pt. Amber Bottles Cmaract eristic . *.*'- . LIMTt ' UWD dV .* ` J. P. Praaer UNRESTRICTED INFORMATION . , WGK (R) Color (APHA) (R) * Condition - ' , 40, 'max. `.Clear . 10084 10084 ' . (R). Sp. Gr. @ 25/15,5*C, . '..A 1.381 - .1,392 K367. 10H4 (R) Acidity (mg. KOH/g.) ' 0.010, max. - 10087 ' (R) Moisture (H20) :35 ppm.; max. * - 10620 (R) Refractive Index 25*C. ' 1.6240 - 1.6260 /,fr.^7L| 10125 'R) Inorganic (Free) Chlorides . . .' 0,05 ppm., max. . . - 12741 * Pour Point ' ' /. . -14*C, or Lower -ji/V 10115 ,0> Dielectric Constant ' ' `4,70 - 4.90 ' (1000 cycles @ 100#C.) . . .1: ' Av7-3otfz.ll608 . (R) - Resistivity |500', min. (500 VDC @ 100*C, 0. 1" Gap)' . I. As OHM-CM x 100V (R) Hydrolysis Stability Test 0,5 ppm. max,' 12672 (As Chlorides) ; .. (R) Monsanto Stability Test . (As Chlorides) . , .. " 0,4 ppmi,'max, ...11503 (R) Distillation Range . ;' (ASTM D-20, Corrected) . : 10117. 10% Distilled by wt. , : 325*C., min. 90% Distilled by wt. ; 360*C., max;* ^3^ Sulfates . \m Dielectric Strength 25 *C, None ' . 35 KV, min. ' . -12169 V. 11605 Flash Point (F) . 338 - 392. - A .10123 Fire Point . . . None 10123 Corrosion Test . .` ; (6 hrs. @ 210*C. with bright I ` 10126 aluminum foil) Change in wt. of AL % . 0.0 ' A.C. Color (APHA) ' . 60, max. 100S4 A.C, Condition 7 . ** Clear . 10084 A.C. Acidity (mg, KOH/g.) 0. 010, max. , 100S7 O; A.C,' Inorganic (Free) Chlorides , * ` " Vise. 100*F..(SUS) ' 0.05 ppm., max. 82 - 92 . 12741 IOOSo i (Over). . . 4. * -Ann" isto~n~ _ .14-44-54 G-1 14-49-54 .14-42-54 14-53-54 14-34-54 014 14t47-54 T4-36-54 14-I 35-57 010 14-61-57 14-.31-54 043 14-28-54 14-28-54 14-55-57 14-44-54 G-1' ' 14-42-54 014 14-. 4-52 i \ . . 11311 HARTOLDMON0013260 I GENERAL INFORMATION Specific Keat (> 25 *C, Coefficient of Expansion (25 - 65*C) Fixed Chlorine 4 . Power Factor @ lOO'C, ' 60 cycles 1000 cycles . 0.30 0,00068 cc/*C. * 42,5 - 53,5% ' ( Typical Values) 5- 4 .. 0.30 . i SPECIAL CUSTOMER REQUIREMENTS , 4 ... General Electric . 4 Thermal Chemical Chlorides .. ; 0.65 pprn,, max,1. Westinghouse Kydrolizable Chlorine (Westinghouae Stability Test) _ ; ' ' 1,0 ppm,, max.' .- hi % ,09% 12722. T-2219 . 4 .% , I * : ` I V. )i 11312 HARTOLDMONO013261 -52- Synthesis of Capacitor Fluids (1630459-QET) * n-Butyl pentachlorophenyl ether, synthesized last.month, had an unexpectedly low power factor along with a reasonable dielectric constant. We are following up this lead by pre paring butyl ethers of various lower chlorinated phenols and by modifying the alkyl group In pentachlorophenyl ethers. ' Compounds prepared to date include n-butyl-2,4-dichlorophenyl ether n-butyl-2,4,6-trichlorophenyl ether and n-heptyl pentacKlorophenyl ether. The latter melts at B*C while n-butyl pentachlorophenyl ether melts at l6C. The relatively Figh melting points of the pentachlorophenyl ethers indicate a necessity for use of mixtures or a significant modification of the alkyl portion of the molecule to lower the melting points. . Capacitor Fluids (RWW) A second batch of dichlorodiphenyl ether (3 Kg) was prepared. Yield is considerably improved (70j6) but all too low. Air sparging after chlorination is suspected to cause polymer formation. This will be checked out on a smaller scale preparation. . Electrical properties for chlorinated or-methyl styrene dimers are attached. The dichloro derivative showed good resistivity and dielectric constant; however, the pour point is much too high. We will try blending to improve low temperature properties. Aroclor Defense (RWW) Only one experiment in the hot tube without catalyst was run. Very little isomerization of 4,4'-dichlorobiphenyl was noted. A Suggested Procedure was written for bromine recovery from HBr off-gas in Bromo Aroclor production. The process comprises scrubbing demisted HBr with agitated calcium hydroxide slurry to pH 4-5, oxidizing with chlorine and steaming out bromine. Saturated bromine with 0.053* water is withdrawn from bottom of receiver and dried with activated calcined alumina to 20 ppm water level. The aqueous supernat ant from the receiver is.recycled. Overall bromine recovery will be better than This process is preferred over oxi dation of the much more corrosive HBr. - ' Industrial Hydraulic Fluids and New Dielectric Fluids A new batch of Bromo Aroclor 1232 is prepared by bromination of Aroclor 1223. This material exceeded desired Aroclor equivalence (^1237) and will be blended in new Pydraul 312. Properties are summarized in Table II. This material will be purified and tested for electrical properties. 11313 HARTOLDMON0013262 / -53- - ac A !i. &4wjlc a, l ICO '1c, "< ^(cwf cX^Vvr * ?i' tJUUv* - C?. 0 0 . eo I2C ^ . oo / ^ roo y /(>ZO 6T*i ? /* ; D<c l.sc 4. VS 4.V . <9;" -f/r -A * " -- wO * "Xm^-Cc U mi 1131 /c?o IM c s 2i0 a. $.*3 /.iy /Vi */ ir ** ir / loy O' ^ {'m/ IZ3C *Z Vo *' 11314 HARTOLDMONO013263 -5*- Dielectric Liquids - H. Vodden, C. Coleman RUABON 760.12.712. *31 uleCliiCul . lulus C'i'b. " a visit to Uieafilf .urici. feluciauted the a tuntion in repeat .:f tneir llceucuee. Vuitasy Inure are nlruaiy .n , reduction am: *ere a.tiJ to :.e using j rodaido Id recants. iiektroalrijay .ieoi<;rad not in /Xf-auctior. =m: t.re nol expecttu to be fox irknpo . yours, . ..unuusiy Italy and .ioiicn-iy ...tin nave no connection with .-icufil. i'lJW-'t Luohuroat vue visited tc. flna out the xwu-.ii !'.i their ciucont.-nt with ...ousunto* Vneir aula rttucn vus tie uiltxuly 1J.&L wuui^htion fuctcre obtained on drun do liveries of .rocior 1243. i;j.c iuc sun- to <;. hi . hex than on coupe titivo produe ta fron iiayer and .TocLelecy and also to be hi.'hor than the value of l.^y quoted &e a typical value in our .-if. i v-l-.a of c.9 at iKJwJ, i>G;iz vus sole, to have been wltuiaed on 2'.,L,T', ,, . iroiupvt bscky ana uithough the results on tuis are not c/itclubirty tn;_. to incJcat that, our dissipation fneter is hit;!*.! than is dicirtbl*. t-r ax-eot rd. It is recommendeii that the dissipation fact-.* of futurt c<.uaJ.,-;f...<.<ite t.` c..ntoiler 8h*-uld be c;waked before being uis, a:cheo ao tbit ue .'no t-.: vbIjj ^t toia eta fwo .'.CL hklil dieloctr c test cello nave been .r^reu for oiu ^.tc.': to .-"eavy -Jlectricala y Indie. Thia exjieusts our present atnek and it i;> not proposed to have any more mode becaoee of the ch.--ne over ty v- tne mch x-ethod of ueuaure:r.unt u3ir..; tie bolabaugr- cells. - 760.12.022.44 Dlectrical . las tic glims. A further meeting -itn jdl took . lace in Lonu^n or. 1.;. `..70, it ikt a.-air euplma.aoa that aul require lonaanto oet B^cificutions for pol/ rr: .le,* film oecAiiae of tno coat of tneir development work. It -i. roe: tnia -oulr. completed qy the end of April. In the meantime we have experienced a settack in the teatirn' :f tne compatibility of the jCu expor-aental sumples. Leek of re;.r.:d .c.! iiity --fiB obtained and the cause has been traced to tne electrode cleaning ae-uorce, hut not poeitively identified. ns a temporary ex.-.'It^ity vii-- t:u =v:ent of this worky tne test is being carried out uaii^g the -C:. cell. i v* .it le . 12 ' .oouarciiy nendor. vv-e made for the purpose of sowing t?ir c.ipuciter vxndingy imprt jn.itia,, r^.u testi.*^; i.'ijuiJx.u.it. ..s r... -It of t.veir recjciaeudutionsy E'ilncupy usteX wuJ viuiteu in ox-.wr it c.r ues. n of v.indiu,j ricldne in operation. unr-i .tucmnes . . our *' &3! - a foreaouubie re , I.ruxci:.a hot.. r_i ,:iu..t.cJ ^iL. .u-: -ir.-.o'- *: : :=.. .4J J a ^|; jit . : J A ;;D t ^ * X * ,*A .1..' 'll L *< - * ** * - A * "* v l . I-f . 11315 HARTOLDMON0013264 -55- 760.12.022.d4 Alectrioal Plaatloe TilffiJ. ' ' noc<v progress has been made on teoting the sauries eubisitU'd by '.!; .nd 12 samples out of 44 ht.ve been tectea to arte. io<.r repeatability is being exi-eriencea und present efforts ore uireet'd to finu the oauce of tlao. ^lesipation factor Is bein.; measured nt $00 volt*. he-ci.usi the sonpitivity of the Lnwecd bridge is insufficient at 5>0 volte. ltolivery of the . 701 bridge is awaited. a seating with bCL took place on 17*2.70. They e^ntea aet.die of coats to equip thenaolvee for Carrying out a complete ranre of testu on their electrical films, including ooopotibility with Axoclor. They also intend to set.up a capacitor testing operation. They require from itoneanto a complete speeificatioi for electriosl fllu ao tiiat thoy can innnufacture to this-. They alu> require Monsanto to tell them the tldcknuss ana type of f:lin share thu market opvortunitl< will be most favourable , bearing in wind that they are late in tn field ana their competitors have already established' themselves in thicknesses of ic mioronj and above. They inolune to the view that the grant.st o.,-crt_nltios lie with capacitors with a single layer of metallised film b.-t im.-rjgrjntrd ratnor than dry. They are particularly lntereeted in Monsanto's developments in liquids that axe suitable for use under self-healing conditions in such capacitors. 7-0.12.022.46 ^lecellaneoim Itevelonaents. A visit to riilA disclosed that negligible progress has been t<nue on our privately-sponsored prejeot to taat VOS 717 and an Aroclor 1?.A?/V0b blend in polypropylene film capacitors. TJ* reason ,-river. whs rmlrye i:. s-tting up tiieir impregnation system. The results were promised for the end of April. A. One aspect of their generally-sponsored project covers the tfcfttiri/r of scavengero In askarele. In addition to the small sample of i yranol 2 that we have supplied# they will require a quantity of 40 - $0 gallons for larger trials. At present we are unable to supply this because it is too much to supply from the laboratory and no pilot plant is available. 760.12.022.AS. Alternatives to Aroelors. Three further modified samples have been received f.r electrical tests. e were told that attempts to refine the previous sample nave not been successful. '760.12.722.50. Electrical Fluids MTS Committee `.Vork. . A sieetlcg of "kA <YG1S4 woe attended in London, concerned with silicone fluids. This meeting was mostly A 2 day meeting of WTA in frankfurt wae attbricer. This whh solely concerned with toxicity problems and the natter of *:eci fi ertir-n.:, w.ich concerned : i-.'deleCf was not disoueaed. J#q intends to oer-1 *i >.-.r anivr ty correspondence. ~ . ' 119/0 HARTOLDMONO013265 -56- *a 1 d.* C'tJ ltUoftp .< r*r* '<:.i '..j iv.ve . j :;.iltj iiJ.ve fvjt-P ir 'n.'.irici, vut vr t: u ' li .'U'iu 3 tit!1: Uik- '.iV't ~#i: -v.. v-- ; r.j` t*. ,, .frcln-'- .' ... . . j.1.. v - f'*; * t. .lii-* r> *jf c! i. t-;-.'. T* '*1 jjfr r* Lti lit..* tX <* iftl i;.. I.-.! a u* ilco y ' 31 *!i> .l ' .1 v * ` * *. ,. Job No. 760.12.722.51. Electrical Fluids CTS. . ' The following customers were visited:- 1. ICAR, Monza, Italy: They have tested our impregnants A275 and A230 in their aluminium metallized polypropylene film capacitors and the results * are unsatisfactory. It was found that corona discharges eroded the metal coatings resulting in a reduction in capacitance during ageing tests. The mechanism of this effect is not readily apparent and further investigation is required. 2. Zanussi, Pordenore, Italy: They requested further information to enable them to change from Aroclor 1254 to 1242. They also want a sample of ^ purified Unox 206 sufficient to 'dope* 25 gallons of Aroclor 1242 for a trial. They do not use polypropylene film. 3. CGE, Milan, Italy: They prefer to continue to use Aroclor 1242 from the USA. They use P2 for motor capacitors using paper now but are looking at polypropylene film both metallized and non-metallized. They are expanding rapidly and may produce power factor correction capacitors in the future. 4. BICC, Helsby, England: Met Richardson during a visit for another purpose and he asked if I knew about a new Prodelec product called 'Metaline' with a UK of. 4*2. He also mentioned that Newport dissipation factor readings were high compared with theirs, 2 to yfo for Newport corresponding to 1 to 1.5J5 at BICC. . ' 5. LIFASA, Barcelona, Spain: Requested further information on design of a purification column and suppliers of components for a measuring system. A small manufacturer, but highly advanced technically, who uses Aroclor 1242 with Superad. 6. Bianchi, San Sebastian, Spain: Having problems with dust contamination so wish to defer change to 1242 until these are overcome. Requested details of automatic machine for injecting epoxy resin capacitor seals. Wanted further visit when they change to 1242 said to be e months hence. Advised them to run pilot trials to test sealing. Don't use polypropylene- film. . 11317 HARTOLDMON0013266 -57- 7. Fecha, Madrid, Spain: Having so much trouble with Aroclor 1242 that ' they are considering .changing back to 1254* Use Supbrad with Aroclor also. . Have made trials with polypropylene film from La Cellophane and Kalle but not using it at present. Personnel difficulties are hindering progress. 6. Cenemesaj Cordoba, Spain: Introduced Pyroclor 5 and after initial adverse reaction they appeared to accept what satisfactory experience of Yestinghouse Canada indicated that it would be acceptable. T7ill place an initial order for trial quantity and make measurements on a transformer. Bequested information on a source of supply of epoxy resin impregnated paper for transformers. BEAMA Conference on insulating materials held at Earl's Court, London during the Electrical Engineer's Exhibition was visited and a paper entitled Transformer Askarels' was read. Only interest was from Castrol! Other papers on polypropylene .film in capacitors aroused considerably more interest and many capacitor manufacturers from UK and overseas attended. Laboratory work: The following TSLR reports have been issued:- L7003-1611: Heavy Electricals, India - Assembly and calibration of 2 MCL Mklll dielectric test cells. This exhausts present stocks of components and it is not proposed to order further stocks in view of the change to the Hunch Method. ' L7001-1611: 17001-2711: L7002-1711: Insulating Sleeves and Tapes Ltd. Yestinghouse Erakd and Signal Co. Ltd. Bruce Peebles Ltd. . ' RTW and ABM visited Newport plant for 1 day and 4 days respectively in order to carry out tests using their ESI 701 bridge and compare results obteinec at Buabon. This is part of the project to attempt to reduce excessive variability now being observed at Buabon when following the Munch method of measuring dissipation factor. Job No. 760.12.022.44. Electrical plastic films. _* . The work of testing the BCL samples continues and about 30$ have been completed. In view of the difficulties in obtaining reproducible results with the Balsbaugh cells and the low rate of testing, we have temporarily reverted to using the MCL cell and cleaning procedure.. The solution method No.2 is still being followed. ) Ye have been given a suggested limit for compatibility of the film with Aroclor using this method, but the other properties of the film which are required to draft a specification for BCL are still in doubt. 11318 HARTOLDMON0013267 -58- vis** to was made in the company of Mr. J.L. Linsley-Hood of BCL who provided somesamples of 8 micron film for the 3 manufacturers. A further sample of 8 micron film for ICAR is still awaited from* BCL. ICAR did not give details of their requirements or prices paid. . There was no serious requirement for film at this time from the Spanish manufacturers visited. y- BICC have offered us at scrap value' 2 winding machines that they no longer require and this is being followed up. _ ICAR Mfetalllzed Plastic Film Capacitor Fluid (RHM) Icar has inquired about the availability of 2-butoxyethyl-2-chlorophenyl ether and 2-dichlorophenoxy methyl tetrahydrofuran, presumably for trial in their metallized plastic film capacitor project. These are fluids which Monsanto patented as dielectric fluids. Icar apparently discovered our patents while searching the patent literature looking for leads to help solve their problem. I have secured copies of our patents and read them carefully. I find that the 2-butoxyethyl-2-chlorophenyl ether is listed as having a dielectric constant of about 5.8 and a power factor of 0. 3% measured at 100C and 10K Hz. It would therefore probably have a power factor of about 60% at 100C 50 Hz. The 2(dichlorophenoxymethyl)tetra\ydrofuran was found to have a dielectric constant of 6.3 and a power factor of 1.7% at 100C and 10K Hz, or about 340% at 50 Hz. If my assumption that Icar wants to make capacitors for use at power line frequencies is correct, they would have no interest in such a material. Moreover, neither of these materials is available even in sample quantities. Because it has been our feeling that minimum gassing under electrical stress was one of the important requirements of the Icar application, we * have subjected a number of our materials to ASTM D2298-64T which is designed to determine stability of insulating oils under electrical stress. The data obtained are shown in Table I. Included in the table are data for Dialla Oil, a transformer oil, and Wcmco C, a capacitor oil. Duplicate runs were made for each sample. The table gives the initial dissipation factor and resistivity for each sample, the dissipation factor and resis tivity for each at the end of the experiment, and the average gassing co efficient in microliters per minute (ACC). 11319 HARTOLDMONO013268 -59- Thc data in Tabic I shed some light on the problem of selecting a fluid suitable for use in a metallised paper or film capacitor where the capacitpr is expected to be self clearing by burning the thin evaporated film away from the area of a short circuit. There arc two requirements.. First, minimum degradation of the dissipation factor and resistivity of the fluid. Second, minimum gas formation. If the degradation of dis sipation factor and resistivity is too great the fluid in the vicinity of the fault will become conducting and the capacitor will fail completely. Data in the table show that the resistivity of Aroclor 1242 decreased by a factor of about 6 x 10^ in this test.' This is the reason a chlorinated impregnant is not suitable. The other requirement is that the amount of permanent gas formed during the clearing process must not be enough to form a permanent gas bubble which would lead to breakdown. Thus, of the materials listed in Table 1, Dialla Oil would be the least desirable. HB40, and mono isopropyl biphenyl would be relatively desirable, witli.HB4 0 being the more desir able because its resistivity is less degraded than that of the mono iso propyl biphenyl. Of course other factors would influence the choice in practice. There may also be some question as to the validity of ASTM D2298-64T for this purpose. However, I don't know of a better test except capacitor manu facture. I believe that we have done about all that we can do to help Icar in their search for an impregnant. If we want to try to do more we should have direct contact between their technical people and ours to make certain that we have correctly understood their, problem. 11320 HARTOLDMON0013269 TABLE I STABILITY OF INSULATING OILS UNDER ELECTRICAL STRESS . BY ASTM D2298-64T . Dialla Oil (Transformer Oil) D. F. @ 100 C 60 Hz Initial Final .00054 .09936 .11136 Rs. 500V 100 C Initial Final 519 x 1012 .41 x 10 .44 x1012 ' AGC 91.9 90.3 >\ W -09- Wemco C (Typical Capacitor Oil) .00072 .01374 .03918 A-215 (Alkyl Benzene) .00012 .041* .50844 Mono Isopropyl Biphenyl .02910 . 167* . 161* HB-40 Aroclor(242 .00102 .60120 .50628 .00060 . 58** . 59** 52 x 1012 4 x 1012 58.6 1.4 x 1012 60. 2 ' $ 28,349 x 1012 .*09 x 1012 1.27 x 1012 22. 1 22. 1 0.74 x 1012 .01 x 1012 .01 x 1012 1.6 2.4 393 x 1012 .25 x 1012 .31 x 1012 1.6 1.6 . 170 x 1012 .00003 x 1012 .00008 x 1012 4' 1.6 1.6 * Measured at IK Hz ** Measured at 10K Hz 11321 HARTOLDMONO013270 -6l- AROCLOR DEFENSE (EST) . PCB, chloride, chlorine and total organic chlorine analysis of the gas samples collected during the thermal oxidation studies at the John Zink Co. established that Aroclor 1242, Inerteen PPO, and Aroclor 126o can be successfully disposed of in this manner. The preliminary data has been sent to * - CRD and a report is being Issued. Sixteen sediment and water samples have been received from WGK for PCB analysis. Their analysis will be completed by May 15. The WGK analytical laboratory has ordered the neces sary gear which will allow them to do their own PCB analysis and are scheduled to be.on stream by July 1. Along these, lines we will familiarize one of their personnel with the required techniques during the month of May. Two 7-lb Mississippi River carp have also been supplied to us by WGK and are being prepared for PCB analysis. Samples of these fish will be returned to WGK for mercury analysis. It is. also planned to examine the carp extracts by GC/Mass to see what others chemical residues can be observed. Water and sediment samples from six accounts employing Pydraul fluids have been analyzed for PCB's. The results indicate that all plants are discharging low ppb to high ppm amounts of PCB'8 to associated sewers, streams, and lakes. The companies so far monitored all Johnson Motors (Waukegan, 111.), Lester Castings (Bedford, Ohio), Central Diecasting (Chicago, 111.), Wheeling-Pittsburgh Steel (Steubenville, Ohio,), Wisconsin Steel (Chicago,Ill), and Timken Steel (Canton, Ohio). This ` information is being generated to help guide a potential recovery and recycle program evaluation by Fluids' Marketing. Muratic acid (HCl) from Monsanto Central Drumming has been analyzed by EC/GC for PCB's. Low levels of compounds having similar retention times to the lighter chlorinated biphenyls were observed. A larger sample has been extracted for GC/^flass confirmation of the suspicioned presence of PCB's. Two other products, food grade fumaric acid and saccharin are also presently being analyzed for PCB contamination. These re sults should be available early this month. If positive, no doubt, other products will have to be examined closely. GC/Mass and EC/GC analysis of silo paint scrapings and raw milk extracts submitted to us by the State of Ohio, Dept, of ' Agriculture, Foods, Dairies and Drugs Laboratory confirmed the presence of PCB's in these materials. The silo scrapings contained 2.87j6 unaltered Aroclor 1254, the raw milk extracts contained an estimated PCB level of 0.6-0.7 ppm. Although the isomer distribution of the PCB's in the milk was altered with respect to Aroclor 125*1, the isomers present were those that occur predominately in Aroclor 1254. Strikingly similar isomer alterations have been observed earlier in the electron capture chromatograms of an extract of a human fat biopsy sample, supplied to us by WARF Institute and in extracts of . ' 11322 HARTOLDMON0013271 -62- raw milk samples supplied to us by the Maryland Cooperative - Milk Producers. Both organizations had suspicioned PCB contamination; unfortunately, in both cases GC/taass con firmation could not be obtained because of the limited amount of sample available (WARF) or the low level of contamination (MCMP). - * The Bio-Test Laboratories has been contacted to arrange for * the transfer of the tissues from various feeding studies they have conducted for PCB residue analysis. A rough estimate from them indicates that they have approximately 1000 cu. ft. of freezer space that is now filled. For this reason'.it was requested that a laundry list be prepared so that selected samples could be sent. We presently have only about 30 cu. ft. of storage space in St. Louis and see no reason to duplicate their facilities. Air samples from the Appleton Coated Paper Co., Appleton, Wis. have been received and analyzed for PCB's. Low levels were detected, but the amounts present in terms of mg/m? cannot be calculated until the actual number of liters of air passed through the scrubbers is obtained from them. If their sample sizes were on the order of 100 1. or more, the calculated amounts will probably be less than the recommended "safe11 level. An informational PCB problem review of the analytical methods and the data so far generated was held for representatives . from Westinghouse. A first quarter report has been written reviewing the results of the preliminary river die away and some continuous activitated sludge degradation of Aroclor 1221, 1242, 1248, 12$4, and Arobrom 1228. In brief, the results were as follows, the only Aroclor which was degraded to any extent in the river die away experiments was Aroclor 1221. ' Isomer alterations accompanied by the disappearance of 0.3-1.0 mg per 24 hr. periods were observed with all Aroclors (compared to 20 mg of LAS Dodecene-1 which is consumed in 24 hrs) feed to the semi-continuous activitated sludge units. The actual interpretation of this data awaits the evaluation of non biodegradation routes which could account for the bulk of the Aroclor lost. This experiment is currently in progress. In the past these experiments (adsorption, volatilization, etc.) have been hampered considerably by our inability to analyze the large number of samples required to obtain enough data points to compensate for the large variance generally encoun tered in biological experiments. In the future, this should be compensated for by the additional manpower which has been assigned to this portion of the project. The Physical Chem istry Group is currently setting up the necessary equipment to carry out biodegradation studies on Aroclors and Aroclor replacement products. 11323 HARTOLDMON0013272 -63- Electron capture chromatograms of recombined laboratory * distillation cuts of Aroclor 1130 and 1142 (Aroclor 1242B) indicate that most of the "refractory" PCB isomers can be removed in this manner. Figure I compares the EC-,chromato grams of these potential substitutes with our current Aroclor 1242 production material. Figure II shows the isomer alter- *ation that cam occur via degradation of Aroclor 1242 in the semi-continuous activitated sludge units. Figure III shows the isomer alteration that occurs in a mammalian system (cow). It is important to note that the isomers present in the fractionated materials are those that appear least "refractory". The following additional reports have been written and will be Issued soon. 1. Relative Retention Times and Electron Capture Response Factors of Some Polychlorinated Biphenyl Isomers Relative to DDE. 2. A Study of the Completeness of Thermal Oxidation of Polychlorinated Biphenyls at John Zink Pollution Research. - 3. The Partition of PCB Between Hexane and Various Organic Polar Solvents. 4. Extraction of Aroclor 1254 From Painted Surfaces by Water. 5. A Study of the Open Air Combustion of Paper Containing Aroclor 1242. 6. Determination of PCB Residues in Fish from Lake Michigan. 7. Solubility of Aroclor 1242, 1248, 1254 in water. 8. A Study of Aroclor Chemical Clean-up Procedures for Use with Electron Capture, Gas Chromatography Methods. ) 11324 HARTOLDMON0013273 -64- APPLIZD SCIENCES ' Environmental Aroclor Project (Keller-Tucker-Emery-Dietrich) * It appears that disposal of PCB's can be accomplished successfully - by thermal oxidation. This conclusion was made after analysis of gas samples from John Zink Company. They were analyzed for PCB, chloride, chlorine, and total organic chlorine. Data were sent to CED and a report is being issued. Additionally, heats of combustio of the PCBfs were estimated for CED evaluations. Water and sediment samples from 6 Pydraul customers were analyzed for PCB's. It is concluded that PCB's are finding their way to sewers, streams and lakes inasmuch as amounts ranging from low ppb to high ppm were determined. EC/GC determinations for PCB's showed low levels of materials in muriatic acid from Monsanto's central drumming facility. These contaminants appear to be lower chlorinated biphenyls. Confirma tion studies are underway. * Paint scrapings from silos and raw milk extracts showed presence of PCB by GC/Mass and EC/GC analysis.' The paint scrapings contained 2.87Z Aroclor 1254. The raw milk contained an estimated 0.6-0.7 ppm PCB, probably originally Aroclor 1254 but with an altered isome distribution. * a Analytical methods and data collected to date were discussed with Westlnghouse in ah Informational meeting on the PCB problem. A preliminary report on biodegradation of.Aroclor was written. Tentatively, Aroclor 1221 degrades in river die-away experiments. Higher Aroclor products may not. .More conclusive experiments are underway. A current study involves.removal of certain fractions from Aroclor products which might be Intractable to biodegradation. Electron-capture chromatograms suggest that most refractory materia can be separated by distillation. . We are continuing to assist the WGK Plant in monitoring bromine . content of chlorine they use to make Aroclor products. Seven samples, collected in the past month, were shown by XRF to contain 250-400 ppm bromine. . We are preparing to determine the presence of dioxins in the present of other chlorinated materials. GC studies show that octachlorodibenzo--dloxin can be determined but that the tetrachloro isomer will have Interference from the components in Aroclor 1260. A better method to determine percentages of various components in undistilled Aroclor is necessary. ' A suitable GC method appears to have been developed as a consequence of studies made on 60 distilla tion cuts from 4 Aroclor samples. * irs2t> HARTOLDMONO013274 -65- SKYDROL 500 Flight' tests are currently being performed on Skydrol -500C by various airlines. Only tests Involving at least 80% Skydrol 500C will be declared valid. The other component presumably would be Skydrol . 500Bi heretofore employed. A gravimetric/precipitation method depending upon interaction with the disparate VI improvers is satisfactory for new fluid blends but not for used fluids, because of degradation of the VI improvers. We are -attempting to develop anNMR method whose success may hinge upon relatively comparable rates of the decomposition of the VI Improver. . A hard deposit from a Northwest Orient Airlines shaft seal was shown by IR to be constituted of a heavy metal, degraded polymeth acrylate VI Improver and hydrolyzed phosphate ester. We have frequently observed such a material. It is apparently derived in part from Skydrol 500B in the aircraft' hydraulic system. OS-59 * ' We determined by IR that a sample of our OS-59 returned by Maxwell Laboratories had no detectable contaminants. This tends to confirm our advice to Maxwell that OS-59, per se, ie not suitable for making capacitors. Maxwell thought their poor experience might be ascribed to contaminated fluid. ' . MCS 524 By' NMR, we showed that a commercial phenyloctadecanoic acid consisted of at least two Isomers. . TYPE III. IV CANDIDATES L-77, said to be a superior wetting agent for Teflon, aluminum and stainless steel was shown by IK and NMR to be a slloxane and that methyl groups and glycol ethers and/or methoxy groups are present. MCS 571 We received a deposit from an MCS 571 system at Dalton. IR analysis showed that 96% consisted of Inorganic salts, primarily bicarbonate as well as silicate. The balance of the sample was a carboxylate - salt but not related to any components of MCS 571. Conclusion: the fluid did not contribute to deposit formation. . 11326 HARTOLDMON0013275 -66- PYDRAUL 625 , Canadian Liquid Air, Ltd. sent us two deposits collected from an - exploded air compressor serviced with Pydraul 625. Extractions with hexane, chloroform, acetone and methanol left a highly magnetic residue. The extracts were shown by IR to contain carbonyl and carboxylate salt compounds containing phosphorus. Exact source of these products could not be determined. * SILICATE FLUIDS '. . The dielectric strength, power factor, resistivity and dielectric constant of Coolanol 25A and MCS 825A were measured to provide 'data for a p* roduct bulletin. % CUMYLPHENYL DIPHENYL PHOSPHATE * .. Timely' analytical support via GC was provided to the pilot plant production of -cumylphenol.. ^* . NASA TENSIMETER ' We received a purchase order from NASA for a recording tensimeter. This Instrument will be Identical to that which we presently use for vapor-pressure and decomposition-temperature measurements on fluids. The components and materials for construction were ordered. CAPACITOR-LIFE-TEST APPARATUS Components were ordered for an apparatus for life-testing of capacitors incorporating experimental dielectric fluids. 11327 HARTOLDMON0013276 -67- HEAT TRANSFER FLUIDS (DRM) Phenyl Ether (diphenyl oxide) * With the able cooperation of the CRD catalyst group, we this month fitted out a pilot unit for the synthesis of phenyl * ether by phenol dehydration on a thoria catalyst. The reactor is a 5/B" stainless tube 18 inches long fitted with thermowell, catalyst support screen and steady flow access leads. It is heated to a reaction temperature (ca. 425C) in a fluid bed thermostat. A metering pump supplies phenol to the reactor at preset flow rates. Reactor pressure is controlled by a pack pressure regulator on the reactor effluent line. .. Three catalyst preparations were made using three different refractory supports, namely pumice, bonded silica, and ' Celite. The thoria was formed in situ by calcining the supports at 600C after saturating with aqueous thorium nitrate. The presence of thorium oxide was confirmed by X-ray diffraction. - .At this writing an initial run at 425C with pumice-supported thoria has been completed. The first 14 ml of product was liquid at room temperature, verifying that reaction did occur. The next 30 ml of product crystallized, probably indicating low conversion of phenol. Analytical data are being collected. 11328 HARTOLDMON0013277 -68- / DEVELOPMENT PROJECTS Tractive Fluids (1331059-JCW)- New Tractants Substantial progress toward the development of new tractive fluids was made this month. A hydrogenated linear dimer of alpha-methylstyrene shows good traction, excellent oxidative stability and fair viscosity properties (Item 1, Table I). A branched octyldecalin shows good coefficient of traction, and fair viscosity and oxidation stability (Item 2). JJjlZ 350T **- & 4bil /Vi- __S_/a ____ _o_ -go U ikl . ,o?V 5.? *>.? Vir ,/7 .Ml v.'i 3,yi.? *1 OO^U- &crz'J ' .3?/ , 0 Sb is //J }}>>' /,yy*- A 7 s.y/ hi *0 (L5 vbo ?3 yj iinn tu.m 36 * S'* C 5 I y ,0 ?,s of 6 J.7 *2-5 ^r2o -z ^ v.t 32.0 Oh*-*) 3 Ay' ,/ P.o\ ! 3,ooofi/y/, i/voj coO^ X, o oc P l.) -- -- J?A f/' ; -v--U'-Ai- X * tf*K _ 1* _ . . On the other hand, a "linear" octyldecalin (Item 3) made by alkylation of naphthalene with octene-1, followed by hydro genation, had better 0 & C than the branched octyldecalin but considerably poorer traction. A dodecylcyclohexane prepared by hydrogenation of dodecylbenzene (Linear Alkylate - 215) for investigation as a blending agent has good low temperature viscosity, but poor traction. . The hydrogenated linear iv-methyl styrene dimer compares favorably with hydrogenated dlcumyl (MCS 6l4). It is almost as good in coefficient of traction, and (surprisingly) is as good in the 0 & C. Its 210F viscosity is lower, but It is better in low temperature properties, since MCS cl2* melts at about 68F. The new material has not crystallized so far. at 0F or at -4o*F. 11329 HARTOLDMON0013278 -69- Synthesls - Linear rv-methylstyrene dimer. As reported last month, the linear n-HS aimer Is obtained by allowing <v-MS to- stand at room temperature for a long time with 0.2% iodine'. After 8 days, work-up gives the linear dimer I in 70% yield, with conversion. This procedure gives a product with 2% or `less of the cyclic indane dimer II. tK, C = chL c/6 c/6 ' -y c -- c//= i - $ -t- C//, r /A. % Same polymer is formed. Hydrogenation of the linear dimer proceeded smoothly over Raney nickel to the fully saturated product. Running the iodine-catalyzed dimerization at 50C instead of room temperature did not increase the reaction rate, over a 3 day period. A possibly better catalyst for the dimerization is a POCla . HsPO* mixture. This catalyst, described in an expired Dow patent, gave 68% conversion and yield of dimer in a few hours, at about 100C. However, the content of cyclic dimer increased to about 10%. The effect of this small amount of cyclic dimer on the properties of the fluid will be determined. CRD investigated the hydrogenated cyclic dimer as a tractant. It is more viscous at low temperatures (34,000 cs at 0F). and oxidatively less stable than the linear material. Also to be determined is whether the crude o'-MS from Chocolate Bayou can be used as raw material. If so, it will lead to a very low raw material cost. Possible difficulty here may be the effect on hydrogenation of the small amounts ofjsulfur compounds in the crude rv-MS. Octyldecalins. Alkylation of naphthalene with octene-1using 7.8i AlCla and 7.8# nitromethane in heptane solvent led to the isolation of an octylnaphthalene mixture in 52/ yield. - 11330 HARTOLDMONO013279 -70- CVt>(ifiJcV/=CliL Isomers, formed by migration of the double bond along the' chain, are also formed. Hydrogenation over Raney nickel gave the 'linear" octyldecalin, item 3, Table I. This had good oxidative stability and low temperature viscosity, but rather poor traction. ' Alkylation of naphthalene with the highly branched octene . diisobutylene gave a t-octylnaphthane fraction, and also - t-butylnaphthalene, and higher t-alkyl products. This type of reorganization is a well-recognized hazard in alkylation of aromatics with poly(iso-olefins).- Using naphthalene and diisobutylene at 2:1 mole ratio, with 3.9% each AICI3 and nitromethane (on naphthalene) gave 4l% yield of t-octyl' naphthalene and 33% t-butylnaphthalene. Hydrogenation of the octylnaphthalene over nickel gave the branched octyl decalin (item 2, Table I). This is a good tractant, with feir 0 & C and viscosity properties. In another experiment, using about four times as much nitromethane, and working up the reaction 1/2 hour after olefin addition was complete, 29% t-butyl, and 55% t-octyl was obtained. However, it is suspected that the "t-octyl" fraction contains material of longer chain length; characterization of this product is not yet complete. Spectroscopic evidence indicates that an alkyldecalin is a major constituent of the Sun toric fluid. Judging from the two examples so far prepared, this type compound appears of less interest than the linear rv-MS dimer. Both have low raw material costs. The o-MS dimerization is simpler, however, and the properties of the fluid are better. MCS 998, a -^0F fluid for Lycoming, is a 70:30 blend of McS 757 with diisodecyl adipate, with 1% dioctyldiphenyl- amine as additional oxidation inhibitor. Although MCS 757 does not foam, the blend did, but foaming was controlled with 20 ppm of DC 200 silicone. Properties of this fluid are given in another section of the summary. The first two gallons of MCS 998 sent to Lycoming were lost in shipment; a second two gallons were blended and shipped. Used MCS 9612 - A sample of used MCS 9612 from Gates Rubber Co. in Denver was received. It had run for 25 hours at 200275F, and for 1 hour at 330F bulk oil temperature. The oil contained a fine black sludge, which was found to be metallic wear debris. Viscosity and composition (by VPC) of the oil were essentially unchanged; acid number was 0.-49. 11331 HARTOLDMON0013280 -71- Naphthenlc 011b - An attempt to relate differences in traction- coefficient among GMR 1204, Solnus 55, and Paratex 75, all naphthenic oils, to differences in degree of unsaturation showed no observable correlation. NMR and IR spectra were similar - there is no aromatic unsaturation. No unsaturation toward bromine was detected in the Paratex 75j the other two gave turbidities with bromine, probably due to reaction with additives. . MCS 998 (RLG) The tractive flu cl, r:cr. 99$, war. tented for ' viscosity, l'oiv.a, arid 4 Ball wear before shipment to Lyconlnn. The results are: Viscosity at -40F 100? 210 F 22,250 cs 17.40 CD 5.49'cs Foam ' Blender Foam at 200F/200cc/2 min Initial height of foam = 1/4 in. Time to break foam = 20 sec.. A STM D892-5Dr Tendency 15 ml Stability 0 sec . Four Ball Wear 200F/12C0 rpm/2 hrs/52100 on 52100 Scar diameter = 0.87 mm. The pour point data on tin. "running" teat condition:; was measured and found to be between -CoF and -55 F. Based on there results and previous 0 L C and traction testing, MCS 998 is judged to be: 1) equivalent to F50 in tra ction unde r the. si 3 p velocity -condi tions sicon1 on t; iO conin" LD 6 2) rli/;ht]y worse than I':cs 757 in foan but better than Vers;i`lube F50 5) a formula t;l on that v;1 u cold st art in th e LD 6 at -4o'`F a in ce it is fcfiU j valor1 4k.. in vi SCO :;Jty to MCS 757 alt -20 F v:hlc::h p ar.Dcd O ur cold :: tart pu:::p tent 0 2/3 -/70). Tlirc 0 rain on::; of r cr. 998 v:'.-re Me; td ':d and 0 /.alio.a; v:c-rc : 1: * | 11rd to J.y i l< 1. i. h two r. < . J:: of KCw for flu: hh:y. pu 11332 HARTOLDMON0013281 -72- Tractive Fluids The coefficient of traction on many fluids was measured by a new procedure on the twin disk test at two loads and two temperatures. The semi raw data is being reduced to traction .coefficient vs. slip velocity. Qualitatively four synthesized . fluids are equivalent to MCS 460 in traction. They are DCCDC . Hydrogenated linear dimer HLD Octyl Decalin (OD-3) H27D4 Hydrogenated still bottoms " Viscosities, 4-Ball wear, and 0 & C were run (see JCW MS). Blends are being made to meet the 0F GM Toric Fluid specifi cation of 2000 cs. * Automotive Fluids (1338459-QET) ^ Air conditioner lubricant candidates MCS 994 and 995 submitted last month to Ford performed exceptionally well in their system except under simulated idle (low rpm) conditions. The difficulty resides in the relatively low viscosity of our products. We have reformulated a high viscosity lubri cant now designated MCS 1005. This material has the following composition and properties. MCS 1005 Composition: ~33# Carbowax 200 phenyl bis(3-hydroxyphenyl) phosphate' 0.25b tolyl triazole Viscosity: at 210F 29.75 cs at 100F 775 cs . at 165F ^oO cs (extrapolated) 80 cs at l65F desired by Ford) Pour Pt. about 5-10F . Freon insolubility: better than MCS 994 + MCS 995 Santoflush - Sewage Transfer Fluid (QET) _ A request by Chrysler for a safe, low density, water immis cible, high flash point, fluid to be used on shipboard sewage systems was investigated. The best candidates appeared to be in the order DIDA > D0A > Santicizer 148 + mineral oil. Accordingly we sent Chrysler a sample of DIDA designated MCS 997. Their response was immediately favorable and they have now placed a #300 order for larger quantities of MCS 997. 11333 HARTOLDMON0013282 -73- PIONEERING (EOS) . Rub-block on Rotating Ring Friction and Wear Tester: .> Following the receipt of Capital Authorization for this pro ject, it has been placed in the hands of the Instrument Development Group who are finalizing the design. In this they are being aided by Prof. J. Georgian of Washington U-, particularly in regard to calculations of the dynamics of the tester. Completion of the device is expected by the end of June. This machine is to be used in the contract for Thermo Electron to measure the lubrication of thiophene/F-50.blends for the Rankine Cycle automobile engine. .. High-speed, High-temperature Bearing Tester; In anticipation of its use in two different projects under consideration by NASA, steps are being taken to modify the . ERDCO Universal Gear and Bearing Rig. The first step will have, to be the gear-box change to produce speeds up to . 50,000 rpm. ERDCO now estimates delivery In less than 2k weeks. We shall try to place the order for this Item as soon as possible. ERDCO has looked into the feasibility of oper ating the standard 100 mm roller bearing test head at speeds up to 50,000 rpm and has concluded that substitutes for both the Rollway test bearing and the loading/shaft-support ball bearing in the test head would have to be found to permit . operation over the rated maximum speed of 10,000 rpm. They are currently preparing an estimate of the cost of doing this. The high-speed ball bearing test head Is also being considered. MAIC has been contacted regarding the practic ability of their designing a test head which is compatible with the ERDCO test drive. This will be feasible. Further more, they will probably be able to design a test head which will accept either 8o or 100 ram ball bearings, giving us a chance to attain dN=5 x 10 without having to replace the entire test head. Who is to build the new test head is not yet decided. A visit to MAIC to finalize the specifications for the test head Is planned for May 11, when Alastair Cameron will be available. NASA-LewiB will then be visited on the 12th to see what they think of our plans. ' NASA-Lewis Contract No. NAS3-12418: A request for a no-cost time extension fo this project is being planned and will be . discussed on May 12 during the visit to NASA-Lewis. Prof. ' Cameron has contracted with his students who help man this project through June of *71. ' Type III Jet Lubes - Lubricity (FSC) New Additives - The following compounds were screened in our slow speed four ball test in MCS 524 (.1%). Those under lined show improved lubricity. 11334 HARTOLDMON0013283 -74- Ultraphos 12 P & 0 Additive #3 . phenylborlc acid trlallyl phosphate - Pltraphos 12 ' This liquid is an anionic phosphate ester surfactant manu factured by Witco Chemical Co. It gives an excellent friction curve on M50 steel. It has not been run on any other.metallurgy. This phosphate gives less deposits In an 0 & C test than MCS 293 but has a higher viscosity increase (8.3# vs. ~4). P & G Additive #3 ' This chemical Is tetraethyl perchloromethylene phosphonate * ? P? ( CaHsO) a -P -C -P - (OC 2H5 )a I Cl At 0.556, this phosphonate does not change the boundary friction of MCS 524 on M50 steel or on hard anodized aluminum. But it has an excellent curve on aluminum metal. This is the . first additive in MCS 524 which gives a curve on aluminum comparable in magnitude and general shape to Mobil Jet II. A comparison curve is shown with following graph. Phenylborlc acid This acid did not improve the boundary friction on M50, hard anodized aluminum, or aluminum metal. It is interest ing that many of our acid additives change the friction of steel or anodized aluminum but don't change the friction of aluminum metal. This metal will react Instantaneously with acids as weak as water. And lubrication theory predicts the generation of fresh metal under boundary conditions. Per haps the additives react but don't produce lubricating soaps. Perhaps the metal is so reactive it reacts with the base stock giving metal corrosion. Trlallyl Phosphate This additive did not change the boundary friction on M50, anodized aluminum, or aluminum metal. New Base Stocks . Most of the additives which we forwarded to SNECMA in MCS 524 were screened in MCS 963 base stock in the slow speed four ball. MCS 963 is 11335 HARTOLDMON0013284 S2.4J If HARTOLDMONO013285 Je+JT *z:r s* / *1 / 7 ' 3 * foAlii/uHt //'//'4 /V.' f ^n\' * #* $ J J- I ( i; r* r ./ hartldmonoo13286 -77- meta tfstfstf + . ortho *- + . meta CfotfSGf + antifoam . - . These tests were run to see what the additive response would be in a base stock chemically similar to 524 but with a different viscosity. Also, to obtain patent coverage.. The additive friction curves in MCS 963 were very similar to those in MCS 524. Thus: phenylphosphlnlc acid: . .large 4 reduction on hard ' anodized aluminum; large u reduction on M50 steel from 200-400F followed by a 4 spike; some 4 reduction on aluminum metal. perfluoroglutaric acid: worse high temp. 4 on M50 steel; large 4 reduction on hard ' anodized aluminum. Also screened were MCS 963 + monopotassium oxalate MCS 963 + bis(chloropropyl)-2-propenephosphonate and MCS 963 + A-88 + trichloroacetic acid Only the oxalate had a different pattern in the lower vis cosity base stock. It had a significant friction lowering on M50 steel in MCS 524 but only a small friction lowering in MCS 963. Other additives CP 64600 - This is Ludwig Maier's phosphosulfide. A 500F 0 & C test was run on samples of MCS 524 plus' 0.1i and 0.05^ of this sulfide. Generally, these blends gave reasonably clean tests with.fewer deposits than MCS 293. Specifically: In the presence of copper, the formulated fluid had less black deposits at both concentrations than MCS 293. Some vapor coking was observed.' In metal attack the copper corrosion was comparable to MCS 293. We increased the silver corrosion. The viscosity increase was 9% vs. for MCS 293. 11338 HARTOLDMON0013287 -78- In the absence of copper the deposits were as above. The metal attack had increased silver corrosion. The vise, increase was comparable to MCS 293. All . in all not a bad test, considering the reactivity of the additive. ' . SKF-NASA Report on MCS 2931 * The final report on the bearing tests with MCS 2931 has been written. The authors are L. A. Peacock and W. L. Rhoads of SKF. It has some discrepancies with the information given us by Bob Johnson. The report states that MCS 2931 ran two successful tests out of six with two runs giving fluid related spalling failures. Our call report says' 2931 passed five out of six tests with one spalling failure which was not fluid rfelated. . It is interesting that the lubricity failures in the report are surface initiated spalling failures, i.e. fatigue failures. This mode of failure can be initiated by water. The trichloroacetic acid additive used in MCS 2931 did contain t-1% water. We should watch water content for pos sible correlation with fatigue in any future use of this product. The authors use a dimensionless parameter h/<r* to character ize fluid performance. This is the ratio of the film thick ness, h, to the composite surface roughness. A high value implies a good film and puts the fluid in the higher EHD region (i.e. near hydrodynamic lubrication). A value below 2 implies lots of contact, i.e. a more boundary like EHD regime. The correlation of this parameter with bearing performance is not too good. EkCeajjr ` &<XZ> ACrr/t&i vwurctprniE SISSWT g TEST KS-ITS rjji: is) fCA3'J.:iD Hfcr rc 7 zvrzz :_r w oil . tLiara 3.C - .C 2. f-nru-.s:: *OL.'r.r.T:. s-:: z. ir ru'C-icsiuctf.v:. v. .:boe : uc he;:ail*.. ' oirr-rt iTHin. t. -:i" -;s:: 2Z3;::. ;;l 7. Jr.:--:?!: "Y'lc 2Z:.v:. r. r.:::K2 2S7z?.. <J.* or o1.r-V iv.rr 3.r-..r; <i.- rrf :cal cmditic* sec*. tL*?! . AW Niac?J r;vv -- ; ' 11339 HARTOLDMON0013288 -79- Of the fluids rated equal or better than MCS 2931, only the Krytox 145 AB and possibly the Dow fluorosilicone could be used in air operations. *^ Phenylphosphinlc acid - This acid has one striking friction property. That is, its u- temp profile up to about 450F* . is very similar regardless of the metallurgy. This is shown graphically on the following figure. Perhaps this - tells us something about the mechanism of lubrication by this acid. Phenyloctadecanolc acid - This material (Eastman) is a general boundary additive for MCS 524. Last month I stated that it contained no terminal phenyl group. The NMR people now say this is an error. The sample is a mixture of Gf-(CH2)iT-C02H and JJ ch3(ch2)n c --(ch2 ) _|]o2h co2h It would be interesting to see if the pure terminal acid is more effective than the mixture. Aryl Phosphate Lubricity Several additives were evaluated as steel lubricity agents for aryl phosphates. Specifically, they were tested in MCS 941 (36$ CPDPP + 37$ TPP + 27$ CDP). Our interest in this base stock has dropped since we are now making blends based on NCPDPP. However, MCS 941 is a good reference phosphate with poor steel lubricity. The screening test was the Chrysler Palex Test. This uses a hard steel (Rc 20) on a soft steel (Rb 85). The results were as follows: Additive (.1$) Falex Test Failure Load psi none 1000 . 2-propyl benzenephosphonite H-1128 ' 4250 (passes test) 3270 4250 (passes test) ^ 2700 L-77 --' 2100 CP64600 900 The results with the first two additives are not bad at all. 11340 l i HARTOLDMON0013289 \I t sX'/S pti'e-rtf/pAoipAet)!c /c,`b ,*a) HCS TT.^ O ,f % 0, ( */. 7 -, 0,1 % 4/ I> O'TS'Va - A^At'dl^SS 3 /e-e/ % 1 g 1 11341 HARTOLDMONO013290 -8i- 2-propyl benzenephosphonite 0-P-OC3H7 -------------------------------------------------- h This Is not an acid but is probably easily hydrolyzed, it should be screened in the Republic Test. . H-1128 - This compound is S153 1 (i-C3H7O)2 -P -S-CH2 -CH -S-F (0-i-C4He)a It was obtaihed from the Oil Additives Group. L-77 (Carbide) This is a silicone-ethylene oxide block polymer. It 1b advertised as a "super" wetting agent for stainless steel, teflon, aluminum, and glass. Since this polymer probably isn't reacting at the surface, the increased load carrying is probably due to increased wetting. Blends with a higher concentration will be run. Cp646oo ' This is Ludwig Maier's phosphosulfide. Pump-Flow Properties of Aryl phosphates We are also starting to look at some flow properties of arylphosphates such as cavitation probe, and streaming' potentials. A one gallon blend of NCPDPP plus 0.042PC-98 was made up for the streaming potential measurement. Water-Oil Emulsion LubricityWater-Glycol Lubricity One suggested approach to the lubrication of water-oil emulsions is to minimize the interrupted flow of two phases at the surface. My first thought to a chemical approach to minimum phase separation was to try Carbide's surfactant silicones. These are block polymers of sillcone-ethylene oxides. Thus one tail of the molecule would be soluble in ` each phase and might tie them together. Also, by varying the ethylene oxide chain, the viscosity of the water layer might be matched with the oil layer. ^ I have since found a Carbide patent for water lubricants (U.S. Patent 5,^57,175 issued 6-22-69). Water lubricants include emulsions and water-glycol blends. The feature of the invention is the use of silicone-ethylene oxide polymers in aqueous fluids. A typical structure for such a polymer is: 11342 HARTOLDMON0013291 -82- \ -J The lubricity of the block polymers is enhanced by emulsi fied polyolefins such as polymerized or-olefins. Examples are given of improved Palex loads for emulsions. The use of these block polymers may explain the excellent pump wear properties of Carbides Ucon Hydrolube 150 LT. We should evaluate these polymers in water blends of interest to us. Presumably Carbide has done considerable testing of these products. ' NASA-Lewis Research Thermal Decomposition (Tp) Runs Bill Jones asked us to rerun two Td*s in the presence of M50 steel. This was because our Td values were higher than they expected based on their experience with these fluids. He thought the metal might catalyze a lower TD. This did in fact happen with one of the samples. All of the vapor pressure and Tp curves are completed. The results have been phoned to W. Jones. He also requested copies of the curves for calculation of Arrhenius constants. Lew Fowler is preparing a graphic conversion scale so these can be read directly from the curves. This should be . ready this week. ' MCS 460 Samples One quart samples of MCS 460 were sent to Bob Johnson at NASA-Lewis and Ward Weiner at Georgia Tech. I also sent them a preliminary data sheet for this compound. Film Thickness Contract 11343 Work continued on the heat capacities and bulk modulus curves of the model compounds. Cameron recently has used HARTOLDMONO013292 -83his film thickness measurements to study freezing in the contact rone. He has expressed an interest in thermodynamic measurements of glass transitions or pour points. This might be heat capacity (via ETA or DSC) or it mlght.be density or modulus measured as a function of temp. This is still up in the air but we may make some further physical property tests on the fluids. 11344 HARTOLDMON0013293