Document NeqN70oxEgKYzQGKBvrzeEN7D

l: vff vi '' 1 > V $' 4*- ** . -s * mm m EXPLORATORY AROCLC _ GLYCIDYLATION OF HYDROLYZED AROCLORS il' .J:* :'' Y-- * . ^k. - ' ; | - I : *' * ' 5 RD-58-Int. Report No. II83 (jL) October 9, 1958 R&E RESEARCH By: J. Dazzi Dayton, Ohio i i '. ; <*'' .1 - i *-:? h-'* ; ... . . (1 Chemists: J. Dazzi and D. B. Sharpv>l"'iji^i1-1:-r' ... . " " a.Hi vvtit !t 1,. ,. . .... x'.^ oJi*? ; i ... > ` 7 'i. ` f . , ' S T R I BUTI 0N It:. Organic Chemicals Division Tal JO IT Lum (b) M. 7. Throdahl ic) F. B. Zienty i. i . _ . . -? i Researcr. ar.o Engineering Division TH1 J. K. Craver/D. T. Mowry " W. Gluesenkamp/ F. C. Meyer Milton Kosmin - H. X. Nasor. , | D. B. Sharp EExxttrraa i^. .. .' j?; II :s H? ; l' :' i* if*f i-f ' St. Louis St. Louis .St. Louis 111 : : ' * Dayton .. Dayton | *_ Dayton i el; >St. Louis 1 .Dayton Extra .. mJt 5 I 1| `i I. DSW 621592 STLCOPCB4095623 MI* i h. 5 ' A & * ... / .. .11 1&T..S &. s. '* <V- & 4* t S*^ :tr ^ -f * J*' * " II. ' SUMMARY III. CONCLUSIONS IV. RECOMMENDATIONS I i Iid DATA AND DISCUSSION General Discussion of Olycldylatlon a. Glycldylatlon of the model compound, 4,4'-bl-(2,6- dichlorophenol) b. Glycldylatlon of hydrolyzed 8 Aroclors 1270 and 1271 c. Glycldylatlon of hydrolyzed Aroclor 1268 d. Glycldylatlon of partly di- 10 hydrolyzed Aroclors 1262 and 1248 e. Glycldylatlon of hydrolyzed 10 Aroclor 5460 f. Olycldylatlon of tetrachloro- 10 blsphenol A g. Glycldylatlon of dihydroxy- 10 diphenyl sulfone Evaluation Data 11 a. Evaluation of glycidyl ethers 11 other than In epoxy resins b. Curing studies using dihydroxy 11 Aroclor 1268 if ; - ||H it ii ii-41 ii< CPERIMENTAL* S ECTK)N Apparatus tearia-ls * |. v H DSW 621593 STLCOPCB4095624 m S J t<" . did 01 yc _t Aroclor),; f: S cv r if hydrolyzed 13 18 Qlycidylafeton bf partly dl- 22 hydrol3^|dfcb^Ioj> .1268 .j ,, d. G1 yc 1 dylation bf hydrolyzed ' 28 Aroclor 1262 and 1248 e. Dlglycldyl ether of tetrachloro- 29 blsphenol A - f. Glycidylatlon of hydrolyzed 31 Aroclor 5460 g. Attempted reaction between 31 Aroclor 1268 and glycerol h. Glycidylatlon of dihydroxy 31 diphenyl sulfone 1. Di-hydrolyzed aroclors as curing 32 agents for bls-glycidyl ethers VII. PATENT STATUS VIII. REFERENCES IX. 'OST ESTIMATES X. RECOMMENDED PROcfJs^ ' ' * XI. MATERIAL SPECIFICATIONS XII. ANALYTICAL PROCEDURES T|!f H , j ti| r -,'j + yi -| XIII. TOXICITY AND HAZARDS * * i3 ! i fI i t , , XIV^ iACKNOVLEDGMENT S J t mm i fm * |40| DSW 621594 STLCOPCB4095625 P tJ H P Ofsl This report is a continuation of Hydrolysis of Aroclors and Similar Compounds by J. Dazsi (R-i ). The mein market visualized for hydrolyzed Aroclors is as intermediates for preparations of the corresponding glycldyl ethers which, on curing/form epoxy resins. This is one of the fastest growing markets in the polymer'field (R-l, -2). The purpose of this project was to prepare glycldyl ethers from five hydrolyzed Aroclors, s model compound, and two additional Monsanto diphenolic development products, tetrachlorcbisphenol A and dihydrcxydiphenyi sulfone. The potential value of these g-yoidy. et.-.er evaluaticr. ;r; Chemicals D i . i cnemical. as epcxy Intermediates was shewn be ;r local ir.eir cats : - r I CL A ~ cc,, entanc by the Organic - * n j 1- - - - j - . A r ;r 1269 as a tew development Reported ex;er rents were March 1=== and 1 -7 :n a part time basis between *r if l y , f. in/o/nwioT * Hf f 4 - H i V "***1 - st cw | ? |H r f ; DSW 621595 STLCOPCB4095626 Pi * * ^ ,P % i4 If : 1' t $4 v, * v Hydrolyzed Aro.^ors 126-2, 1268/ 1270. 1271 and 5*6v,- t,,t, :u1t (2 ,6-dichlorophenol), tetrachloroblsphenol-A, and dihydroxydiphenyl sulfone isom^afceac.t rapidly and quantitatively with .edtdWss : eplchlorbhydrin|(ECH). This forms "the correspond ing chldronydrin glyceryl ether which on treatment with excess caustic forms the corresponding glyc idyl cth >rs in high conversions and yields. Polymeric ` ther content is ic-w, but careful ccntr-'.l cf reaction conditiens for the deny arc hiorination step is re viired. Glycidyl ethers,of very high (9O-97.5* of theory) di-epoxy content, lcon-mereial Epon`828 is cJ't -poxy) were prepared from di--hhyyddrroollyyz:ed 'roc lore 12 68, 127h, I2~l , : t rachi or: h * srh* - -nv 1 1 'cpl:encl) . * :ca in , and the model ' "odel cooc -arch progre- - L ? 1 d2 Za , rV 'rretation :or d i-hydroly n:d 11 *75 w 0r/ dre e n . />5r? `winio w^ r. * - y - ^ J j -T> 3 - * ' * ' ,4 _ i r 9 . tetraohl .thiVl ci. I fone . Ci i y - ^ 5- *' -j \ v ^ ' hi1 o r p 'i ipnenol A (R- 3 g.ycicylaticr. - --j were wrrkecr ' t'p.9 model OTD evaluation f compounds 0} hydrolysed -dihydr cxy.rcclor 126 was Pilot Plant 1 .L-'; 1 10 Ci . k 's tetracr.lor ~4 : f -d that. "itenr. and r- f glycidyl 'th-rs ieri :ed from di-hyc -lysed Aro:lc: 1268 art 1270, and .. w. or.- -'del dtphenol ar ir.: rsely related C ';r 1 ^ * ->>-^-* -rro^ro dn-*-v?.4 - -~ ' - -v;raq-c- * -v . n - 7o-r t:.-- first two cot- -tf** 4`h~ -r-.-:d-. eii6l cmC ofloercutlhee it was reversed! high -apex content product nad a hi gn tro 13 ' C . T.^.*-meltlogether- are desirable f-r f'c-orol a t if A e h s--- r a ^ <7 icw as | * eii g -y mist: -`p! I16-4; |he|b ; dj - jar: t Arcc-lt-: ,'Z.edi Ar yxyr iXo.rot erpl:r :> 1 | 3.. fc: ^ ,*|alsliyTf p* a'4 te- nt5 c 0-v e- E1 :i liccn. v'r ducts DSW 621596 STLCOPCB4095627 g 0 r-c ,, ... '*. * i Glyoidylatici. cf hydrolyzed Aroclors, tetrachlcr^oblsphenol A, 4,4 '-hi- (2 ,6-d lchl orophenol), and d ill yd rqxyd iphenylg fulfone went 1 smoothly. This is a two step synthesis;cflrst, 'the epoxy group of * epichlorohydrin (ECH) reacts to form the l-chloro-2-hydroxy-3-glyceryl ether; this, in the second step dehydrochlcrination, re-fcrms an epoxy ring giving the glycidyl ether. The latter process step is critical and required development of special reaction conditions for each compound to get products with high epoxy and low hydrolyzable * Cl content. ,; Glycidyl ethers derived from d1-hydrolyzed Aroclor 1270 and 1268 show valuatle properties as intermediates for epoxy resins. The St. Louis Filet Plant process for the glycidylation of tetrachlorc 'bi'spheno'l ` is the test method for dihydroxy Aroclor 1268; this diphenol an Organic Chemicals Division chemical. hr.ers 1e: ived from, partly dihydrolyzed Aroclors appear o be cf little cal .e for resins, per se, but may have application n surface -cat *. nzr- (Ref. 3) . . - Dihydrolyzvd A: :c;-r 1268 car. be used as a curing agent. It reacts faster .-it:, dig lye idyl Arcelor 126? than with Epon 828. Cur work with ii. ? rrdel compound, 4,4'-bi-(2,6-dichlorcphencl), gave valuable data which greatly facilitated solutions of problems___ encountered in the early*work with the hy&rolyze<3 Aroclors. Syntheses and evaluat i :.-.r f ~cdel c:rr.p;ur.ds are most helpful and time saving for exploratory res arch in r.ew fields as well as for analytical studies. vt- Ll* i* DSW 621597 STLCOPCB4095628 o'! vife - e t a ft "si - PI Ellrf Hj f -? \ . i ' 1 H. _____ ' . ; ll tsf? fpptf We recommend Investigation cf the glycidylation usirig mohe ' efficient stirring and higher reaction temperatures^ oosslbly by using chlorobenzene or dimethyl sulfoxide'b'^sblv%n&`. r ' 't* Further applied resear: 12bS and 1270 is justi: diglyc1dyl ethers derived from Arocler Glycidyi ethers derives from Arcelor 5^60, new compositions, are recommended for development first if Monsanto decides to enter tht epoxy resin field with a oreduct having patent protection. These epoxies may he of additicr.nl importance because each molecule has --c-re two glyc idyl groups an' the Trade has expressed inter or-1 Ar = t. ' j \ o W * *-* - promoted as flame-proofing, heat^aratio to HET anhydride. Di- __:reer.ed as curing agents for all new on glycIdyl ethers see Ref oenyl sulfcr.es should be dio oew compositions. . f- I Hi 1*4iit-4{tt4iit!*iM i f- |H| l \ DSW 621598 STLCOPCB4095629 fe ? # A' A t * 1 0 1l',5 ' >.-''t ,3" l.r General Discussion of Glycidylation features of glycidylation are given first, and more specific data are reported separately under the subdivisions. Glycidylation of hydrolyzed Aroclors 1271, 1270, 1268, 1262, 1248, and 5460, 4, 4'-bi-(2,6-dlchlorophenol), tetrachloroblsphenol-A, p,p'- and o,p'-dlhydroxydlphenyl sulfone takes place on contacting the phenolic compound with excess epichlorohydrin (ECH) in the presence of alkali. This is carried out stepwise. Initially, the phenolic group adds to the oxirane oxygen to form a l-chloro-2hydroxy-3-glyceryl ether, as 3hown in Equation 1. Alkali in less than stoichiometric amounts catalyzes this reaction. The excess ECH, which serves also as a good solvent, is distilled off and the chlorohydrin ether is stirred with excess alkali. This dehydrochlorinates the chlorohydrin completely, closing the ring and re forming the epoxy ring system, as shown in Equation 2. The chemistry of glycidylation of dihydrolyzed Aroclor 1268 is as follows: --OH + 2CCH2ch"Xch2 NaOH (catalytic) ^ -cl6.5/ > t CH2CHCH2-Q-- ( Cl 0 H / och2chch2 b Cl H + 2NaOH -> - 5I i 1j H2CflCH20 >---- 0CH2(^HCH2 + 2NaCl + 2H20 > I I . r : T ifi5y:i * i ffaroughput this ^eporfJLetiher formation Is referreditoias;Equat^ort 1 . _ ;| 'i pfd; pqhy(!rochl6rin^ti<^pi rere fifed ?to as Equation" 2. fThe Aether* : f | ; formation is quantitative|and rapid in the presence of:excesscECH i arid. filkall..Th4 niast fcpnveniientiprQcedure found,is to|dissolve, the i , I DSW 621599 STLCOPCB4095630 With epoxsf r^c^f^.ciii&^WM ^fj^fr^ffll^K^t^fj^^llfeci^dHtfiifl^aicritlBP 1 cah^S* removedfp ^ iution Witfi fiL^splveri^lsuch as^benzene, toluene, chlorobenzene f ^tdfoxkhe, whereupon NaCl can be filtered off. Epoxy and hydrolyz- '. ^Paflelplvvalues indicate that partial epoxy ring closure occurs at ^pconditions used |i|i;;Equation 1. The NaCl formed can be removed by y1 Mwatei^washing-bei^re or? after removal of excess ECH. -"> v $ ` ' ' '" ' The complete dehydrochlorinatlon of the chloro ether to the glycidyl compounds shown in equation 2 is tricky. Specific reaction conditions had to be developed for each phenolic compound investigated; these were time-consuming studies. The recommended process consists of dissolving the chloro ether in a solvent such as benzene, toluene, chlorobenzene, diethylene glycol diethyl ether, or dioxane, than excess caustic is added as a 30-50$ aqueous solution. This solution mixture is stirred at temperatures between 8l-110C for a period of time according to the rate of dehydrochlorinatlon characteristic of the compound used. The model molecule was the slowest to de- hydrochlorinate and required up to 50 hr reflux. Dihydrolyzed Aroclor 1268 and 1270 reacted much faster and this second step was completed- in 8 hr or less. In some of the cases the water was removed during the dehydrochlorinatlon step. This did not change the rate of reaction. The final product can be isolated f.-om NaCl formed and excess NaOH used by washing with water or by filtration. Final product is always heated with stirring at reduced pressure to remove solvent. Finally, the material was always kept for 15-20 minutes at bath temperatures between 100-120C. Partly hydrolyzed Aroclor glycidyl ethers, prepared during the early part of this investigation, were partly polymerized ted above; 100?C.; *These were first ground to a fine powder and ; - then washed In a Waring Blendor. Solutions in benzene, or chlorobenzene of the glycidyl ether of 4'-bi-(2,6-dichlorophenol), the model compound, were Drecipitated by hexane addition. ` |;J|l glycl(dyl?a|i0h experiments of hydrolyzed Aroclors, poly- richer ^formation took-place. ' This undesirable reaction could not ;U. controlled, especially in larger size batches. We do not know .?' the^cojidiitlons wnich; cause this reaction. The model compound L|ar| conditions ^always gave much less polyether. ; -No ex- f | "*as|foiufd fori this behavior * * ' ' 1' - - *aS iyc- iylaflon 'of -the model Ic:bommpnboujrffd;: ....,, (2,6-dlchrorophenol) DSW 621600 STLCOPCB4095631 Sue'; ffal _ *-aeerinagMMp^<y,f;iitefei?-mThfe6ry;rjgl^g' 105* excels; wait: added; and w|s refluxefl- for ^6 aht'ftolglv^ (6^5-5^ of. tjifoip^ glycidyl ^ewierv This: product (wa| :^r.era^lfK)r 14 mbre; hr fntd^xane with excess caustic, and gavecthe |P|^flSi5 8l .2$ conversion, based on diphenol charged. Dl-eooxy content was Q0.7$ of theory. * f? - j ' Dioxane^being a solvent for the ether and watertwfe added to the ft ! - I . chlorobenzene solution after It had refluxed for2l hrs. This mixture was heated for 6 more hr without Increasing the epoxy content (Expt 2.). Again, the ether had to be refluxed for 16 more hr In dioxane to Isolate a product analyzing 97.556 of theory dieccxy and 0.1$ chloride. "` Gradual addition of KCH and a reflux for only 10 hr gave a lower epoxy content (41.7$ of theory). Or. heating this in dioxane with core NaOH, a high epoxy compound (96.9$ of theory) in 90$ conversion was obtained (Expt. 7). Expts. 4-1 for a r e p r u ^ model compound r c 1 vether root *- -- -cu 1 e; led .re: ,Expt 1-7) for the preparation of the .er.high in epoxy and low in chloride quecus solution of the disodium salt of the excess ECH. On refluxing for 2 hr only a =0$ Lyoidyl ether occurred (Expts. 5 and 8). Reheat r ccr.d it ions given in Expt. 5 formed pclyether . icmplete dehydrcchlorir.at i an took pla :e when the reaction was carr^fleCgut-in absence of a solvent using aqueous Nl0H*f^0$) at (Expts' 6-7). Epoxy content was 72$ of theory with 1.58$ Cl*. Experiments r)-~- r.ta i r. re action cc recommended ur : (Expt. 1-7) 170cC . trie mciei cerncun: data for Expts. 1-10 show that high err .er.C parallels :. :gh acini cf finaal cprrqodduuCct,. * PPrractically mlvethcr. formation was noted in this se ries?4f| fxplr|lm9nts . It was quite unexpected to note that the dehydrochl<|rit|alon step was much slower for the model mol ecu le than fcr^othej|. dilbh^ncls tried. r-ai -bf % a ^iilK *$44 i lI;t H i :f;fthjit if wfl-lfbe fery difficult rfehydro-f ' ' [onj cf the model mol ear .e ?tc : a one s'tep Steric is* a ppssi'alei explanat#16r4 for. this lunij ili&ii * f i ..t i _ IS I DSW 621601 STLCOPCB4095632 raluesf* / 3tage|br|our in^ea$igatl6m|)u^4M[ |f-V|int'erpi?etatiori?-1j^ltlSU8^^frequireiaientsfled^ialA:ofpre^.f^Ign^ fmelting glycidylptthers ll^stead of -products whfch-werphf^er|?f_ |in epoxy content jbut [had: lower melting points. These [two factoS[ ^delayed solutionvbf'early^difficulties. . *' T' ; The earliest glycidylatlon experiment was made by adding the ^ f * disodium salt of hydrolyzed Aroclor 1271 to the calculated ' amount of ECH at 65C, a product low (42.8$) in epoxy content (42.8$) was obtained. A polyether type product was prepared using less than the stoichiometric amount of ECH (Expt. 12). Isolated product is high melting (220-30C ) and low in epoxy (17/6) and Cl" contents. Using an excess of ECH increased the epoxy content Immediately (Expt. 13). sf The next successful step was to use excess ECH and only part of the alkali needed, and then to distill off unreacted ECH (Expt. 14). The liquid residue wus diluted with a solvent such a3 chlorobenzene, KOH (40$ excess) was added, and the mixture was refluxed 24 hr. Quantitative conversion to a low melting f60-64C ) glycidyl ether, very high in diepoxy content (96.2$) was obtained. This was the first Aroclor glycidyl ether sample which cured in a manner similar to the model compound. This behavior Is attributed to complete di-functionality In the hydrolyzed Aroclor 1271 used. - Reaction conditions developed for the model molecule (see Expt. 1-3) were successfully applied to dl-hydrolyzed Aroclor 1270 using dioxane as a solvent. With shorter reaction ... time, dl-epoxy content was somewhat lower (87.57) and Cl- was higher (0.90$). These reaction conditions were repeated in a six-fold scale up (Expt. 16). The latter gave completely dehydrochlorinated but partly polymerized product. The soluble lower melting fraction analyzed 66.47$ epoxy while the Insoluble material melts as high as 275-80C and contains only 14.8$ * of theory epoxy. ' ii * * v .Since the scaled-up Expt.|l6 failed to duplicate Expt. 15, Hi 45 ' attempts were made to* reduce the heating period and increase ; dehydrochlorlnation temperature. At 0.20 using diethylene > . 3 glycol diethyl ether as solvent, the Cl~ content was reduced ..to 1.48$ in 3 hrt reheating for 8 more hr j at 120C reduced.. !the Cl~| to 0.1$i^t the diepoxy content ..decr-eased ito ,75$ .of i liy j^hebry^luef to pol|rIth|>| j^qrmatlc rt (Expt * if)| f surface - f il jpoatingi evaluation-groupfwas espeecciiaallllyy InntteeWreastt.epd i1nn tt-h.hilsn * i "aateriai, f A llatarrgge^ji|sslijz|ee ^%^^tchh t, ((550Q22 g ^;) |iwwaass|made|a^ Expjb^L8i .cat* - ' - * ` '' ~ * * `* DSW 621602 STLCOPCB4095633 of partly dlliySrOfy'zec |prl^po;P^jefailria Explfi^i^-2lii^ |Re4ctlon.- was ? nofc Jcomplete :' .-.Sherefpphef profluet twalu-pf ` value. The next series < .. lpts\ p2y2)Was madd'after^ll%cfihg experience and analytical ' 'r&'ouif situdies on the-mollelt compound. For process studies ' partly di-hydrolyzed Aroclor 1268 prepared in the St. Louis pilot plant was used, until di-hydrolyzed material was made available by0the Payton^methanol process.4 4The first large size (Expt. 22) gave quantitative' conversion and 88.5# pure diglycidyl ether. The only variation In the next Expt. (23) was to add the NaOH gradually to the dioxane solution; this caused incomplete epoxidation (80); retreatment with NaOH increased thi3 to 90.256 of theory. Attempts to force reaction 1 and 2 into one step by using theory or excess NaOH failed to Increase the epoxy content (Expts. 2*t and 27) Starting with the sodium salt of hydrolyzed Aroclor 1269 and excess ECH and then heating, as suggested by a consultant (Dr. Leonard), was no* successful; seme polymer was formed (Expt. 25). Another variation was to start with sodium salt of hydrolyzed Aroclor 1268 olus excess NaOH and then add to excess ECH at 95C., epoxy found was only 57.of theory (Expt. 23). 2. Di-hydrolyzed Aroclor 1268 This product, prepared by the methanolic-aikall, pressure hydrolysis was used as starting material for Expts. 29-35 and converted to the glycidyl ether using the best experimental detailsrcollected in previous runs. Solvents used were dioxane v* v , and ber.^en^, the latter being preferred (Expt. 29 and 30). Data ! ' nonn f.lflpesspe Trnuinnss*'tw(perri>e aadlsaon igr*ivveinn in Reference 13. The Organic Division recommended a pilot plant process which they developed for glycidylation of tetrachloroblsphenol-A. This was first duplicated in Expt. 31, giving. In 92-5% conversion, a diglycidyl Aroclor 1263 analyzing 89.556 of theory epoxy and very loWf Cl" i(0.1?6) y This rur. was duplicated on a four- and six-fold scatLe-up (Exptf. 32 and 35). ? The final product is quantitatively obtained by filtration of the reaction mixture, thus removing fJaCl ana NaOH' as solids. A clay ,treatment of the melted resin should fe^irate St races of impunities;, product in Expt. 35 cor;SLn.fciQCF5|mil.lleguivalerit NaOH/g reslri. ?I r Miilli* ? 11 it -1? i f <o>*C|x|iie|w'e naq^ of redpetiprj |.n tjune'forfdehydrochlorination felilretclifeinainl goal waS tp fsupply jhighf epoxy materials to '' ' '* ' ' " ' in wc' " '' m DSW 621603 STLCOPCB4095634 S ' -if% t '1 262 and 1248 the. airlll sfy^t ^C^glycid^l^li^^f^^^^^these r' polyzefi Al^lprs|wre lnvest|Lgatecli^ii%'^'e^y prelimin- t -f ifel#- fcs. 56139} olyclaylation of hydrolyzed Aroclor llpireparet in theijsti Louis pilot plant was then assigned to igfleld. here Dr. S. Gebura developed a good process using dloxane as solvent (R-5).;. Valuable Information promts wo^c^ere jthenk i4a|sd:|Ln our work (R-6). fr.5om- ' D*;r. Gebura's " e. Glycidylation of hydrolyzed Aroclor 5460 New glycldyl ethers were prepared from hydrolyzed Aroclor 9460, a mixture of di- and trl-hydroxypolychloroterphenyl. The material from Expt. kj> contains some polyether but it is low in Cl~ and cures in a manner similar to the diglycidyl ethers. Minor variation of the glycldylaticn procedure, l.e., replacing dioxane by benzene or chlorobenzene, should reduce the dehydrochlorination cycle and consequently reduce polyether content and mp of final product. In addition to patentability, this product may be partly tri- -- functional. The trade is mo3t anxious to get a trl-glycidyl eth.e: to be used in epoxy resir.s. Hydrolysis of a completely chlorinate terphenyl (C-joCl-,^) might yield trihydroxy polychloroterpher.y 1 for use in such Sr. application. f. Glycidylation of tetrachlorobisphenol A The first glycidylation of tetrachlorobisphenol A inside Monsanto was carried out in Expts. 39-41 at.the request; of the Organic Divlsidn. Products were then curedibytJ-. Herblg (R-3). Glycidylation went femcfoth'ly to obtain in the second attempt (Expt. 40) a product high in epoxy (88^ of theory) and very low in Cl~ content (0.12$). Conversion was quantitative, Expt. --1 Is a duo? ication, of recommended pilot riant procedure giver, ir H-4). ' t ` g. , Glycldyiaciorit of dihydroxydipher.yl Isulfonej. s *. Since Monsanto ha3 a good process but no market for and o,p!-d|hydeoxydiphenyl sulfone it was logical) to inve; Lgate the glycidylation of{ these -.two, compou. nds; he:co f ethera are not *"J -- in. Datenjts .i till SiAJfthfs . .. . in;Ch--e*r'aic.a..l.|.oA.r.$b....sd..tc.-r.fa.c.fc.l.rt.:s..!.ijbfj lipt1 atentable, noifurthel i-dyl ` * * Just|.4ie f fei tcidj hese known <copoinid.s j DSW 621604 STLCOPCB4095635 ^ f ' 3f#Si > ' i; - l A^sampleloffglycid^r #%her of v ; .f t_ #1262, a viscous liquid, was also submitjEffll 'for .'evaluation*It! 'plasticizer and stabilizer for polyvlnyx%hlbride?fomulat|j|,,. This product, evaluated as Lot No. P-1087, was compatible Withf|? polyvinyl chloride at 40$ plasticizer concentration. It had a'- f poor flex of + 20C, but volatility was low (0.4$) as was kerosene extraction (0.1$) (R-7). Costwise, P-,1087 13 out of the plastic | clzer price range. Nevertheless, this compound might be considered in combination with another plasticizer (to reduce the flex) for specific applications requiring a very low kerosene extraction combined with fire retarding properties. Contrasting to P-1087, the glycidyl ether3 of blspher.cl A and tetrachlorobisphenol A are both incompatible with polyvinyl chloride. < As a stabilizer for polyvinyl chloride, P-1C87 is comparable, as expected, to Paraplex G-62 (epoxidized soybean oil); the latter Is the commercial stabilizer most used for PVC (R-7). Diglycidyl Aroclor 1268 wa3 submitted fc ' biological toxicant screening as CP 161J7-J. Of Incomplete cv.- luution data received, r.c bio logical activity has been uncovered. b. Curing studies using dlhydroxy Aroclor 1268 The value of a diglycidyl ether is its ability to cure to a polymeric material. This was shown at an early stage of our investigation when diglycidyl Arcelor 1268 (N3P 5l608>4) was heated with maleic anhydride ir. a test tube to yield a polymer melting above 730C. Systematic curing studies on the glycidyl ethers described In this report were made by J. Herblg and later . by J. K. Craver and J. Scihwenderman; results are reported in R&E, i Da: final report 1115 (R_3). - s `-*'Hr***f Di-hydrolyzed Aroclor 1268 acts also as a curing agent for It3 glycidyl ether. This behavior contrasts to that cf bisphenol A ar.d tetrach.lorc.bispher.ol-A, whi ?h do not react with their -c rresconbir. - glycidyl ether3. Data or. r> * iyzed Aroclcr 1263 as a c :r.g agent for it3 glycidyl ether and Epon 828, ' ' - v ^ - J i are g_ iven .a The former glycidyl ether t reacts faster than the latter us shown by the N.E. values found. A possible explanation for this reactj ivity of the hydrolyzed Aroclor is Its high acidity (pl- .5). It is possible that;the . * . t curing properties found for di-hydrolyzed Aroclor ;1258 arid Jhomdlbgs fofceri i substantial! ifiarfcetf for;; thif (c|mp9\|nd|-.i/4i|ch | I I i compete with hexachiorocybloper.tadler.e-malelc anhydride fir: 1 *: f f 7 i %,i i lariduct > (HET.). The latter?.gives of f. HClfgas ^on |heatlr.g: wrfigh is i i -*>. DSW 621605 STLCOPCB4095636 STLCOPCB4095637 >r x,vt *> *+ ^ *" : '><`rf \ v iX t ; i ", i -mk 'UfP l1* i<A .V. i . ' uf -a' ! ' (' " -v : ' itus A11 *eSdpgM.mencaag|&pe conducted ;in 's^anda^ftJaper, ?Pyretgisss|, i, * equipment; Th^lther formation was;,carried put in 3-necked roundbottom flasks equipped with dropping.funnel, condenser and thermometer. Excess ECH was distilled off at reduced pressure with good stirring using a 10" Vigreux column. The flask was heated by a hotfwater bath while the "`receiver was kept in ice water. Tehydrohaloger.ation was carried out in 4-necked, round-bottom flasks equipped with dropping funnel, 3tirrer, thermometer and bean and Stark trap carrying a reflux condenser. The flask was heated by means of a heating mantle. Low boilers of the final resins were always removed at 1-2 mm under good stirring. Where necessary, heating by means of an oil bath eliminates local over renting. Materials Used lame of the hydrolyzed Aroclors used were prepared by the Author;.___ thei'r preparation is reparted under Final Report No. 1182. or. Hydrolysis of Aroclors ar.d similar chloro compounds. St. Louis Pilot Plant partly di-hydrolyzed Aroclor, and Dayton - Methanolic-prccess di-hydrclyzed Aroclor were also employed. The partly di-hydrolyzed Aroclor was from the St. Louis allot plant batch run in ethylene glycol. Tlhydroxydiphenylsulfor.e, the p,p*- and the o,p'-i3omers were obtained from the Organic Chemicals pivision t P .A - i s- ; ' i < f & * Tetrachiorobi3pheno1 A, 3t. Louis pilot plant material. upon 328, Shell's trade name for bisglyoidyl ether of bisphenol A: epoxy content 3.23j. : -.lorcr.ydrin. Shell, technical grade, 95^ v* i . f I - f A ^ I ;f t* 'then cr.emlcals user were 3tockroom items. 'Experimental! Details ;i- 1 ;| I V ... * * --(3J lpr|l|>y r* oOppdhl Qefpl QX,/l),l j|, jfc U ^ ft f F ~~ f tm f f t fffTfTtfll ffp I ! * This dri |)| ctohnetaminfsrc; jethl|em?odj,4i%cftllxceidyt'iUi 4.ttfL- oiXi o< f* f i JVm*I L.5bi-S 1, :; k 41. '.........................' DSW 621607 STLCOPCB4095638 iW'M' . v M KOff; H 40$ ia.quec st sqLUtilStt noted. IptftferfSolnliii , -_,_ IsoJatlqn|tfes:aieut.ral tq ph after 46`minuses viW...i.s ^ pq'a<. &a.4klkalirie p. ap. eV. e33 f epichlorohydrin was v'distille$T'oft while stirring at re '.pressure# at 100C bath temperature; 1029% epichlorohydrin was collected In an ice cooled receiver. The residue was filtered front' solid KC1, washed with chlorobenzene, and diluted with more chlorobenzene (total 1150g). It was refluxed while' KdH (1.67 moles, 235'nil aqueous solution, 10$ excess) was added. After four hours, reflux water was drawn off the Dean and Stark trap (total 60 ml ). Refluxing was continued for 36 more hr giving 60 more ml of H2O. The reaction mixture, consisting of two layers, was diluted with 600 ml of hot water and washed with 3 x 400 ml of hot water using a preheated separatory funnel. The organic layer was dried by azeotroping benzene-water mixture. The glycldyl ether was pre cipitated by adding the solution to 2.8 1. of hexane. The cooled mixture was filtered, giving 360 g. Anal. Found ($): Epoxy, 4.55, 4.67; Cl , 4.63: Total Cl, 35.67, 35.77; C, 47.51; H, 3.13. Eased on this epoxy value only 63.5$ dlglycldyl ether formed. Analysis of the combined wash waters indicate a low NaOH content, equal to 8.65 ml N. The partly reacted material (350 g) was dissolved in 1200 ml of iiaxar.e, 0.75 mole NaOH was added as 30$ aqueous solution, and the system was refluxed for 14 hr. The two layers were separated ar.d filtered. The organic solution was divided into equal vol umes and added_ to,2 1. of hexane, each, giving 28l g ether, mp It2 and 1850c.' 'Concentration 6f*the*solvent gave1 3T " ether II mp l32-l86C. ., Inal. Found ($) I eppxy; 6.57, 6.60 equal to 89-5$ of theory; Cl" 0.17; II eooxy, 6.69, 6.75 equal to 91.5$ of theory; Cl" 0.34. Conversion 8l.8$ based on bipheh'ol charged i iSxpt. 2 ,347316, 347324 P |i . , ' This eJpfifiment -is dToSeftbf'q gv|plicpti;or of Exot., 1 usj f, moli ; Phis i3tP: I# added- KOH CO* * , ' 54li#Ui4 limdl i\ if I < i i i t I DSW 621608 STLCOPCB4095639 W&tt&r.* i * 'J > ceac Pi fded, and? C'OT ng, there was obtained 4104.^5^oof fp'&rold'ptircM^ |mtapj I ,73 and 4. frtihiwash water equal to *21-. 4 ml N: " The partly reacted ether (408 g) was diluted with 1400 ml of dioxane. To this was added 0.9 mole of NaOH as a 30% aqueous solution and the system was refluxed for l6 hr. 200 ml more- i dloxar.e was added, the slurry- was filtered and washed. The - ? filtrate solution (2410 ml) was divided into equal parts and each added to 2 1. of hexane, cooled and filtered to isolate 328 g, mp l33C(sharp), containing 7.03 and 7.262 epoxy equal to 97.5j of theory'; Cl" was 0.102- Concentration of hexane solution gave 32 g, mp l8'-c: Cl" was 1.21, 0.32. Conversion was 82.5$. Exot. 34732;, ""3871 same ar Ln Expt. 2. The '-prd r*- r : ~ - ! -ix wn ! gradually, instead of at the ' a total of 10 hr while the solution was hept The material was washed, and unreacted KOH. Precipitation of 1830 g of solution in 'r.ex.r.r.-- gave 44i g, mp ll6-17C, analyzing 3. 0-0 _u-.J 3-1 -;7 epc xy, and " .61 and 7.32 hydrolyzable Cl". Part of this product (if3 g) was dissolved in dioxane (1500 ml), ICO ?.s added as'302 aqueous solution, and the system was lor ; hr. Then Jj ml MaOH was added (total 1.273 moles). ;fhux r 10 Isolation by standard procedure gave 357 g, mp 185-IS7 C analyzing t I 7.10, 7.122 epoxy (96.82 of theory). Cl-, SO.25, 0.262- t-t ? Concentration of hexane s~lut'. or. gave 22 g, mp l8cC, Cl" 0.192- rriQ ** TM i- *". 1 ,o vig r- .O ^jcnt ?xt six experiments the recommended he glycidylatier. of >i, 4'-bi-(2,6- -"^; 14 The; biphenol (O'.3 mole, 171 g) failed to dissolve at reflux | i 1 S I f temperature ln six moles (555 g) of epichlorohydrin) therefcfre || i ^43 nfdes^ary to add 100, g pf dioxane jbg achieve j solytdgifj || i f *1 P 3a 'added'. 0.12 mole (2o.6 n&) hi '%>$ ifceSdsf JlaOH MS Iff th-a system was _uxed f: feisti] oxi7 excess in ir.dJdiqj 3. e due ed| prje s sure 1266 16 8-t >i DSW 621609 STLCOPCB4095640 . ,... . hr .. ______ . . ______- _____, ,, 5 g product, mp 809C .(not.sharp),' was ISc$.| ig6i90f7.00#- Cl" ; This 'Indicates 'only ^ . ,v., _Lycidyl? ether. To complete the dehydroha] _ . 4^|)|;^-'Ki3L15 mole) was dissolved in 330 g of dioxane, 0.126 mole (10# excess] of NaOH was added (2.5 N aqueous) and reflux con- tinuedvfor 4.5 hr; the mixture became neutral. Total isolated material; was 47 g, mp 150-15SC, analysis 0.6# Cl-, epoxyv3-92#. Theilatter value is meaningless since the product failedto dissolve completely in the analytical reagent. Expt. 6 NBP 341283-5 In order to keep the biphenol in solution it was necessary to dissolve it (0.2 mole, 64.8 g) in 600 g (6.5 moles) of ECH. Then 10.8 ml of 30# NaOH was added and the system was refluxed for 1.25 hr. The isolated material melts over a wide range ca 100C. It was treated with 50 ml of 30# aqueous NaOH for 6.5 hr at 12cSC bath temperature, The mixture t irned solid after 5 hr stirring. The reaction product was O * ~ id to a fine powder suspended in water a washed free of NaCl, to give 81. 5 g, :30 145-7C. Epoxy, 4.90, 4.oi; ci-. 2.87; C,, 48.0' U11. '*tO'7i :; residue 1.07. These values reveal that >nly partial ipnv.c rvohalogenation oocarre: Expt. 7 34123?, 3412 39 A solution containing biphenol (0.6 mole, 194.4 g) and ECH (l6i*2*i$oles, 1600 g) was refluxed while NaOH (0.25 mole, 32.4 ml aqueous solution) was added at once; no exothermic reaction occurred. After refluxing for 2.5 hr, excess ECH (1320 g) 'was distilled off giving 409 g of crude material. It was heated in the same container up to ].4oC pot temperature and then 150 ml ~f 30# NaOH was added; the system was stirred for 10 hr at l?0c bath tempera' :re (pot temperature 11' H* f f | The 'reaction mixture was worked up t o give 2o7 g of ether, mp 15^C, ^analyzing 5*25 and 5-23# epoxy, equal to 72# of theory. Hydrolyzable Cl was I.5S#. A total of 0.114 mole of NaOH I? I iiil ichaage<|.Jremained unreacted as shown by analysis of the wash iifilPfmttro w tif ** f IBiff furl' ' ? t> i* if DSW 621610 STLCOPCB4095641 r m i` jn was; Irt= ir |2 hr t ~3i "temper1 ^ teacfion ' took pIsKs^liranec ftaixtlire' was! refrftfr^BlEo&'3 more hr/! 'Ae^Vatei^ 0.196 mole NaClt | i*n organic layer was Cheated to>; removehexeess ' ECH. To this residue was added 30 ml of 30$ NaOH over a 4.5 hr period while the solution was kept slightly alkaline. Product was Isolated by* s;tankard procedure to collect 90. g, mp ,120C. t Anal. Found '(*): C, 46.20; H, 3.59; residue 0.12; epoxy, 2.99, 3-00; Cl- 6.49. These data Indicate only 41$ dehydrochlorination to diglycldyl ether. ,, Expt. 9 341297 347301 ": A mixture consisting of bipher.ol (0.25 mole, 8l g) and excess ECH, (4.4 moles, 502 g) was refluxed, and KOH (0.1 mole, 18.7 ml of 30$ solution) was added at once. Heating was continued for 1.5 more hr. After filtration, excess ECH was distilled off. The residue was dissolved in 220.ml of chlorobenzene, and 93-5 nil of 30$ KOH (0.5 mole) was added over 0.5 hr. The turbid mixture was refl'txed- for 14 hr at 9S-105C, and then diluted with hot water, with good layer separation. This chlorobenzene solution of the ether was precipitated hot from 500 ml of hexane to collect 100 g of a nice white material analyzing 4.39 and 4.40$ epoxy (60$ of theory) and 5-25$ Cl-. This material is more soluble than samples prepared in previous experiments; this could be attributed to less cross linkage. The material (90 g) was ? , dissolved in 540 g of dioxar.e; to this wa3 added at once 0.152 . .. ( rmole NaOH as 30$*aquebus solution. The3system*'wasi5refliuc##`fb'r* ~ 12 hr, while the solution turned neutral. Total isolated material was 64 . mp l80C, analyzing 0.4i$ Cl- and 6.71 and 6.71$ epoxy, equal to 31-5$ of theory. Conversion was 30.3$ based on combined 3tens. c 341299, -3Q0 1 " i k ' f I I j > ' ; I I1 ; I r size batch was used in this expt. using jnethod establis in the previous run. | Biphenol (0.3 mole, -260 g) %as dissolved i 1200 ml - ECH, reflijxed,( and 0.32 mole', or KOH30& solution:wasr 1 added a once! I After (20 jmorq mi!n|ites|t5o| , sfjiutfdr L* I off|at| gl-residui Ik r ilfin; reflux fas: p< )f!EHHsome of ft? e^.ap_ c _ |re I | The 387 g of White fc] >f ici v DSW 621611 STLCOPCB4095642  r. >* w*: <v "&> '*>' -* \r, **.. was 40.5 ................. i at " ' """'" but oii cdallrig to this required more ECH.f< itfasKtfater was eqv |btbd using lOfO fcl' ^started to pfecipijl b. Qlycidylation of hy^raiyz'ea Aroclor 1270 Expt. 11 ' 510446-7 '%* f^ ' * ' ' M ' # v f- -- A solution of hydrolysed Aroclor 1271 (0.25 mole, 115.5 g; Anal.(^): C, 31.58; H, 1.12;_C1, 61.86) was dissolved in 200 g of 10% NaOH and heated at 6-5C while 0.5 mole (46.3 s) of ECH was added dropwise over a 1-hr period at 65Jf. The mixture turned solid and was worked up by washing the powder by means of a Waring Elendor to isolate 130 g rap 120-125C. Anal. Found (v?) : C, 36.58; H, 2.2: Cl. 49.50; epoxy, 2.31, 'o .ues were obtained much later, :al method was developed. :urate anal inversion in this run was 91 Ret 316058-60 jxro rimer.t wuc . n ,-5 :ordins ' 00Oni , example recoramer :he preparatl of a high melting partly polymerized bispho:.~l A glycidyl ether. Hydrolysed Aroclor 1270 (0.7 >mole-J 442 g (HE? 310445) was'stirred in 1.61 moles of 10# NaOHtaiad(-I.25 moles, **- ft was added over 85 minutes while the temoerntar-'- 'was gradually increased from 42 to 10QC. ?After 15 ml ECH were added all phenol was in solution. The formed ether precipitated after JO minutes when a total of 59 ml (of ECH was adder. The material gradually solidified. It was ideated for 1 nor hr. The hard material was ground to a powder,, and washed by Blendor to is,olate 445-i1i sg, ^;;mi>,*;^!^f42pB.0o*-C, equal <- of u V.'a-~--' * ,^o` . e rsi,on. Anal. Found {%) : C, |H| L.71; res, 3-24 (NaQl )j |epQxfp[iiiBtL|l,p2 ((equal to 17.lj6|Qrf-theo.ry) : ^ ^ ' ' Mf $1 STLCOPCB4095643 *4r *3^. - 5 : c > j i. , , -5. .* S * * * .# 'V % A \' S ..- ..^fWiS'iiraWTf r excess (0.6 mole) ECH at 40-45C oveif a I , it ^erl^d^flt^Was-kept for 0&Ki$ &t 4^-50%. - Proluct f` isolated weighed 52 g, mp 65-82C; epoxy content 3.52 3.24. A *erxm after 36 days gave 3.03 and 3.06jg epoxy. This experiment shows that excess eplchlorohydrln Increases the epoxy content of final product. `" Expt. 13 NPB 328979-80 ' The purpose of this experiment was to increase the epoxy content- I by using an excess of ECH. Hydrolyzed Aroclor 1270 (0.15 mole, 6? g, (NBP 310^45) was dissolved in ECH (0.90 mole, t^.^g, mole ratio 1:6) and stirred at 44C while NaOH (0.3 mole, 30^ I aqueous solution) was added over 2.15 hr; no exothermic reaction occurred and all caustic added was used up. Excess ECH was distilled off and final product purified by dissolving in ethyl methyl Icetor.e to collect 56 g, mp 42-46C. I :nd (?): f, ?-.r'7: H, 2.09; total Cl, " 5O.5I: epoxy, 2.00, 1.99 (eq. :o 3t I theory). Correctness of the latter va; questionable assuming the I material is free of solvent, indicates high epoxy content. such a low mp a 1'ways B Exr r1' A mixture containing hydrolyzed Aroclor 1271 (0.25 mole, 115.= j (NBP 334601-2^ analyzing. 7.48 and 1.12% OH) was dissolved in " I ,2.5 moles of r.CH and refluxed while 0.1 mole of K0H (as 30^ solution) was added5at once; no exothermic reaction occurred. The mixture turned neutral after 20 minutes heating, the excess ECH was distilled off, and the residue (liquid at 100C) was I diluted with 300 g of chlorobenzene and refluxed while 0.44 mole :f K0H (40;t solution) w r a 3 hr period. After 24 h: at 100uc/l mm gave es re3id. ,i';4 g, mp 60-r4cC, of a yellow resin analyzing 0.32, 0.33# Cl~ ' 55-5/f of theory). C: similar to the model molecule: t tI DSW 621613 nf % ^ f fS STLCOPCB4095644 . f# l 3 *s: % ifc Pound' (jj): Ci"', 0.87;* epoxa 90.556 of theory. (equal^ to Expt. 16 ` 353862-64 ' fi ; *!- -.- - - . . -.-3 = This experiment represents a six-fold scale-up of the previous run. Hydrolyzed Aroclor 1270 (1.5 mole, 66l g from combined samples 328989 and 353853) and six moles of ECH (555 g) were heated to 85C, then 0.6 mole NaOH as 30^ solution was added and system was refluxed for 3 hr. Then 163 g of ECH and 71 g of water was distilled off, giving 1100 g of residue. It was diluted with dioxane, filtered, (36 g of NaCl) and the filtrate was diluted with more dioxane to a total wt. of 2400 g. To this refluxing solution was added theory + 20" excess of NaOH (2.88 moles, as 30^ solution) over six hr. The mixture was refluxed for 12 hr, and filtered. The filtrate was again re fluxed for 8 hr while 0.6 moles of NaOH was added gradually, keeping the pH at 8. Layer separation of the reaction mixture was poor, therefore 1 liter of chlorobenzene was added. The ' solution was refluxed jverr.ight in the presence of 0.3 mole of NaOH as 3solution. Next day the liquid and solid organic layer were worked up separately. The liquiq fraction (i) gav< 627 g of a light yellow brittle resin, mp 5 C. Anal. Found (): H|# ; Cl , 0.039, 0.096; epoxy, 3-79, 3.86^ (equal to 66.4;g of theory); C, 37-53; H, 1.8356; no residue; Cl, *49.05, *+9-19; 0H> 7-57, .41. . . Nr conclusions can be drawn from these OH values since the resir. failed to dissolve completely in the reagent. The solid fraction II represents, after refining, 193 g of a yellow powder, mp 280C, try heat stable and does not change cclo; *Anal*. Found (): Cl~, 0.31, 0.35; epoxy, 0.79, 0.S6. Expt? 17 . 353894-6 . , i \ i . < .. * . . nel scaled>up, experiment,was not successful, .additional ftoas$ i|e^.|nee(did 4 tln|.this huh 6.96 ino|e| |*P3 fgj fNfcP I f Hydrolysed 'Aroclor 127P was dissolved in 4 moles? ! t 95,C , while,,0.4 mole of, NaOH (53-'3 ml ii ,' f1 STLCOPCB4095645 |d|then heated _ as Witft ho t wa oilehtf63&fg of resJ - 'tail unreacted Kd(piBa|iJ^TO8fgQuall to .077 g| Hoping tfo ge?** fister any more tqM^^^^aSvydro'SlfLorlnatlon at higher tempera-T ture, 1200 ml of diefcnnriene ^giycol diethyl ether was used as 1 ` ' &- solvent for the 652 g^eslduei. ; This' mixture was stirred at " 120C while 1.5 molesJ^aOIL, (5jO# solution was added over 5 hr, and the mixture .k~ ~i innfir> m~ --* &' was separated from 1 , , _ __________ __ solvent at 100C/l mmf total recovery of solvent was 1198 g, nfj-' 1.-J093/ of diethylene glycol diethyl ether. The residue (691 g) was washed with hot water until free of sodium chloride, and then heated to 120C/l-2 mm to collect 574 g of yellow brown resin, mp 58C, still, containing 1.47, 1.49< of Cl". The resin was retreated bytdissolving it in 1200 ml of recovered diethylene glycol diethyl ether, then 0.5 mole of NaOH was added as 50^ solution, and the system was stirred at Or* t or 8 hr. The material was isolated as described above iect 5J2 g of resir. analyzing 0.11, 0. 10% Ci" and 4.l4, 4.] epoxy (equal to 75$ of theory), The latter value ind! formation of polyether. ;es oartial ucra ao 553898-900 Trw evaluation groups requested additional quantities of material similar in epoxy content to the product prepared in Expt. 17. Hydrolyzed Arcelor 1270 (1.06 moles, 437 g (NBP 353889 ar.d 3t3l$-) wtis dissolved in. 4.4 moles of ECH. To this was added l. o' mole of NaCH (50^ solution )and the system was refluxed for 4 hr. The excess ECH and .JJaCl were removed and the'residue (6Cl g) diluted with Jaustic (1.25 moleSf .nil of. diethylene glycol diethyl ether. % t # 4 solution) was added ar.d the 3vstem for 5 hr at 120oC, then filtered. Total unreacted as equal to 20 ml II. The salt free resin solution was i e ,,re ted 'with 0.62 mole of NaOH'solution and heated for 5 more 120-1252 A small sample jwas worked up separately and to be free of Cl". The whole batch was worked up as zed -.0 Expt. 17, |tosgiv4 511 g,| mp 56-8cC, ..analyzing ,C1 ",l. Or 0.1l; epoxy 4. 5'|.,5. 4.|iS5f. fThis is a good duplication and'' the values ^requested by the evaluation group. rrom .ne wasn water was' . (ann additional Q9g8. -- of resin, Ii 1 mp 15:-39C. Total uni" iketfed falloali round in wash water was. % L equal to 24.3 ml N* J Cc ibfhBdJcIrfversIoh wasf 99.5/6. . I I imh M * *' i t f -f !' * * * ': rtH I !: : "' Ii,t! - 14 DSW 621615 STLCOPCB4095646 : KXpt .Iff Crude ,-zed-Arcelor 1< of Aroclor with 4.47 moles ofdCOHl The crude hydrolyzed mixture- (620tj KaOH (12), heated to 65C and 2 hr. The- mixture, turned 3oll,d,,.^ Isolated. '' \'- villi'! .. Jby- refltutlngf'^ mole si * )(> g ethylene glycol -for 8 hr is dissolved In 2 moles of of- ECH was added over .1.25 lof;55^ 92^..Was * s.'8 Anal. Found C, 37.94; H, 2.70; residue, 2.25 (NaCl); total Cl, 42.27, 42.42; epoxy, 2.06, 2.06 (analyzed later). Conversion was quantitative. ~i Exrt, 20 323178, 323110-1, 323336 , d 7:. -rder to get a tctt-ru recipitation of the rooir., glycidylati or. ..no carried out ir. a are -i ol flask equipped with a high speed rer but the resin f j 11 e i to precipitate ir: 1 he desired fine f w. A mixture consisting of 0.5 mole (208 g) of crude hydrolyze-; ''.or 1268 (321023, :nd heated dc it 'e-5y , ,T. Katon) was .d is while 1.05 mdes solved "f ECH in cne mole was added wise. K > exotherm*. ; reaction occurred. A Waring Elendor was 'or the tedious i sdatior. of the resin to c ollcct 256 g as a V-1 7 OwV;--ewMhitne= p:--:.v-.er mo 100-102C. Aral un; (*): C, 33-79; H, 2.83; residue, 1.59; Cl, 35-63. s experiment was repeated twice ainder islmllar conditions,, giving^, r.tical partially dehydrbchlorii!:Jt^5CF'efherff.Which are of no value epoxy resins (see HEP 323110-323111). 21 323112, 323132-3 3 experiment the rye: > - ~ if" J- - ` .ruder 12- f C\ C (321 r' 2oO x ssolved ir. 1 mod:; of 10^ Na 9: (300^ excess )v of ECH 3tirreqf over iigh|sp%edi'- it li6p5-7m<3iinO^Cu,, t!(:es TmvthUo,is,, 44 e was heated for or.- more hr ?0C; The separated material viscous. It;, wuS washed in a Walking;Vendor.; The product : :ined liquid.j.It/was not analyzed ^jrfee at (that time (th? epoxy i Llue3- were ir.acb\irjkfe.4 mt |7f. Pte 3IerC i. faOi tchlcrohydri n 1'" ii DSW 621616 STLCOPCB4095647 7WbS3Be& dihydrolyzed AK>|lo#Ml3|;1(analya:e1^ff'^^^SJ:l jpgiVerit InV.;used for Table^d ^^H^ared ih this prSrdijfoMet|ts*. study. ; th# -< S&.^ =fdi? 4Ek ofciUitaSStfcp-iio&L<JESbt plants W '; I' `T> ; * dt $$& sir S ^ i i ! :.J A mixture containing 440 g (equal to 2 equivalents) of partly hydrolyzei Aroclor 1268 and 4 moles ECH (100$ excess) was stirred at 75C while 0.4 mqle NaOH (30`S) was Jadded|^)yei,|20iminutes.The mixture turned'neutral after 1 hn. 'Ex6%%s %CH and water '"' -.-r - were distilled off. The residue (669 g) was diluted with dioxane and then filtered. To the filtrate (1421 g) was added 1.92 moles (20$ excess) of was 32$ NaOH and reflux was continued washed with he water to collect for 4 hr. mp 40-c _o. ?.v;n resin. Anal . Found ($): C, 31.13; H, 2.51; no residue; Cl", 0.53; 0.66$; er:xy. 5-12, 5-13* (equal to 88.5$ of the'.-'-}. juar.t: 'ted NaOH found wa3 equal to fo ca *h is led up t ior.3 ar.d moles, VJ re a 20$ $ 5 ex c r. z V or t 3. ? 33) Expt. The v?a s 22 or.l xa r.e solut i on (1 Q^: Q . a1- the i - r. medium was k tr> t> a t a p H :i " . - ; -rat_or. v:u: ,er. the mix ture ur r.ed neut ral It this print was added ( IaOH scluti on, an reflux continued for an additional 6 hr. Isolated fnaterial (848 g) still ctr.-rir.s 2.07, 2.0?$ Cl"; epoxy was 4.47, 4.49$. 'or oor.pl te dehydrohalogenatior; 820 g of this material was .iso live d ir. 1700 ml of dioxar.e, ar.i C.7 mole of NaOH as ?0$ u.luti or. ..-as added at or.ee. After a 6-hr reflux, 7^3 S of re si O o -v _ C* ^.diVoio i", 0.3,4, 0.3b; epoxy 5-23, if ; -1 oo 90.2$ of the or; Conversion -was 95-5$ for this ;p ehy Irochlorir.at i or;. rJCOt 560248-9: - ; \ t i r. ft. .ft ft o' Lnv DSW 621617 STLCOPCB4095648 imi 1 h.'i iWf -y- T,- /- le dlat|l||^^resldi|e'' litas $<UlA$d. with 130 g of _id 0.5 mdm|o>|50^ aqiieou3 NaMlfeS added, and refluxfmi frtaintalne<f|P6ifT^hr-fpot em^e^ftire 110C); 23 ml of watei was distilled bff. A small sample-was collected after|5 ; hr reflux and analyzed 3.82$ epoxy. The final material analyzed 3483$ epoxy. Some resin on the walls of the flask * (insolublelin boiling toluene) contained 2.33$ epoxy and . : 0.68$ Cl~. This is a polymer, mp 270C. The isolation of the ether was complicated due to emulsion formation. Some of the ether polymerized during the heating. Expt. 25 360240-1 Consultant Dr. N. J. Leonard suggested reaction of hydrolyzed Aroclor 1268 (0.6 mole 264 g) excess ECH (2.4 moles) and 1.2 moles NaOH (30$) at bOC followed by a reflux period. After 1 hr the mixture was neutral. ECH and water were distilled off. The residue was in this case partly rubbery and failed to dissolve completely in 450 ml of benzene. The mixture was refluxed for 4 hr. Analysis shows 3-80$ epoxy content. Sample analyzed after 3 more hr reflux gave an epoxy content of 3.72$ while additional 12 hr reflux failed to change the epoxy content (3.76$ found). All epoxy values are based on benzene soluble material; as stated above not all material (polymer) is benzene soluble. This suggested variation does not look promising. Expt. 26 367452 it _ , A mixture Consisting of 0.75 mole hydrolyzed Aroclor (330 g) and four moTes ?ECH wa3 heated to' 80C, then 1.45 moles NaOH 50$ (by mistake, wanted to add 0.66 mole) was added at 80C over 5 minute time. The mixture was refluxed for 2 hr while it turned neutral. Then it was washed, distilled to remove excess ECH and water, and the residue wa3 diluted with benzer.f dried and filtered. Removal of the solvent gave 393 residue, , mp 56-39C,! analyzing 3.99$ epoxy and 1.70, 1.75$ ti .< Total Cl- found ?Lh wish water equal to 1.28 moles of NaCl. *A ~ i portion (370 g) of the residue was dissolved in 400 ml of benzene, theory + 5*0$ excess NaOH (21.6 ml 50$ solution) was system ?was refluxed at 8lC; 22 ml of watertwas j The 3'olution was filtered (17 g of partly! | ' - ' ]6lkds|)|to|giy4 666* g pf filtrate. j jsoiatiop, * rl^dfasl followsfIn:$ne?case 42 g of solution?was? 1-2 mm pressure to collect 23-l.g * ~ ........ ' ' 0. ' * 11 * ' '''IH 111 STLCOPCB4095649 torr Iconditions used reheatin ffurther epoxy formation, * Expt. 2* 7 5,674, 55- 6 ? I''" * !: ?r' vf f; f *7 *7a content without f: ff ff . # t f' t< If 1 The use of more caustic was investigated in this experiment. Hydrolyzed Aroclor 1268 (0.7 mole, J08 g) was dissolved in 4.2 moles of ECH and NaOH (1.68 moles, 20# excess as 50# solution) was added over 5 minutes at 80C.Some heat was given off. tf This mixture was then refluxed for 1.5 hr, 100 ml of water was added and reflux continued for 1.5 nor hr. A 100 ml portion of benzene was added and solution was washed (good layer separation). Total unreacted NaOH found in 2157 g wash water was equal to 0.53 g NaOH. The ECH and water were distilled off. The residue was solvent purified to remove all NaCl and then heated at l40c/l-2 mm for 15 minutes to collect 572 g resin containing 1.52, 1.72# Cl- and 4.18, 4.18# epoxy. This material was dissolved in 400 ml of benzene and 0.27 mole NaOH as 50# solution was added. This mixture was heated for 2.5 hr while removing water. Isolation, exactly as given in the first part of this experiment, gave 549 g of resin, mp 96-8C, analyzing epoxy, 5.86, 5.88#; Cl-, 0.26#. Longer heating gave again dehydrohalogenation and polymerization but epoxy content was not increased. Expt. 28 567457-8 Another variation tried was to dissolve the hydrolyzed Aroclor 1268 (0.5 mole, 220 g) in 20# NaOH (240 ml using a 20# excess) and add this solution over a 40 minute period to 5 moles of boiling ECH. The mixture turned neutral after 1-hr heating. More NaOH (0.09 mole) was added In two portions while reflux was continued for 2 more hr. The resin was Isolated and heated at 100C/l-2 mm for 0.5 hr to Isolate 284 g, analyzing 4.4l# Cl- and 5.26, 5-30# epoxy (equal to 57.7# of theory). Analysis of the wash water shows that some of the caustic charged was also used up to hydrolyze ECH a3 a a own by the J..33 moles of NaCl found In the wash waters. j ;j f f S - (1) Glycldyl ethers derived from di-hydrolyzed Aroclor 12btS. MW. 416.7 " * . ; - - : : | f | . 1 ; * ? * ;Expt, 29 , 367460-6l| * " * ' * 1 m i 5 f t-I * till? i t i m * All the experiments; described In this fsectior were made from di-hydrolyzed ,Aroclbc 1268^prepared by. jc ic jpslng i ( re 1 DSW 621619 11 ?:TT! 1 STLCOPCB4095650 S- ^ 'f rk ? il&&3M ifr&ss i-.-,,? #! I:#* t rf I ii $ i II t i|K ^enltfianatlbr 1(614, ^95f2 |4r^#^imsiaiiutidwfth|2|j8 g IlSi - NaOH t50$f was added.! Reflux MdnWttl refiin was purified by1 Standard . . , naif resj as j kept at l40C/l mm- for 15 minutes to glvef'1'28 g, Imp , . analyzing 4.97, 5.02# epoxy (equal to 81.9$ of theor^)^ Cl" content was 1.09$; conversion waB 97.8$. ' -' Expt. 30 ?67465-65 The best procedure found for glycldylatlon of partly hydrolyzed Aroclor 1268 (Expt. 25) was used In this experiment. Di- hydrolyzed Aroclor 1268 (0.25 moles 105-5 NBP. 555855) was dissolved In 2 raole3 of ECH, heated to 90C and 0.1 mole NaOH was added at once. After 20 minutes 0.1 mole additional NaOH as 40$ solution was added and system was refluxed for 1.5 hr. Excess ECH was removed by distillation. The residue was diluted with 220 g of dloxane and refluxed while 0.4 mole NaOH (40$) was added over 5 hr, and reflux was continued for 8 more hr. Filtration gave 17 g of NaCl. The filtrate was distilled and the residue was dissolved in benzene, filtered,__ _ and solvent was removed. The residue was kept at l40/l-2 mm for 15 minutes to give 121 g, mp 52C, analyzing Cl-, 0.90, 0.93$; epoxy, 5.51, 5.52# (equal to 87.3$ of theory). Conversion ' was 92.8%. Reheating of 110 g of this resin with 0.125 mole of NaOH (50$) in 250 g of benzene for 1 hr at reflux gave, lifter purification, 102 g, mp 50C, analyzing Cl-, 0.17, 0.20$, and 5-35, 5-40$ epoxy. These data reveal that further heating reduced the Cl- content without increasing epoxy content. Expt. 51, NBP 367466-8 I i f #*4 > This experiment is an attempt to adapt the St. Louis process for the glycldylatlon of tetrachlorobisphenol A to our di- hydrolyzed Aroclors. Di-hydrolyzed Aroclor 1268 (0.25 mole, 103 g. .7. P. LeBlanc :s Expts. No. 12, l4, 15) was dissolved in 1.5 mole of ECH. At 9Q0C, 0.25 mole NaOH (50$) was added over 12 minutes, then reflux was continued for 5Q mqre minutes.! The ECH was distilled off at 100C/l-2 mm' ttlen u. 5 mole4 NaOH a was added as 50$ solution over 20 minutes.. The system wa3 . ?. f stirred for 5-5 hr at 100C. Next.day, 150 ml of benzene was ' added, the mixture was heated to 80C and(filtered! |4| " analyzing 8 g NaOH; calculated NaCl is 29 *25 g-pllil 7l|, dlstlllefd |ff 44 If wasf then Tceptf at# l40?/-l for 20 minutes tq give 32i 5?fj mp '49--50C, analyzing Cl-, 0.10,7 0|1Q$ and, edoxyy| 4(ec(jza3|; Itq 9.45$ pf theory). * ^srfyessjLQnlwa-31''''* DSW 621620 STLCOPCB4095651 li af $ J -t f} I led ^roelli| iSS8 f(l (dfln g; N.E;1?0iBP|3iiM) was! 1 mole of NaOH walw1aiaiQ)a^p50^ olutl^i. f Exothermic reacti^^atarted5 after *7 miutes,l when ; -5 : '45 ml|oaustlc had been addedjf^he remaining NaOH was addddjover , the next 10 minutes. The mixture was refluxed for 46 morel'; minutes then distilled giving 183 g ECH and 40 ml' H2O. The^residue was stirred at lOOoc/1-2 mm for 20 minutes. 4 Theft* ^ |resldu#was diluted with 340 g of benzene, refluxed (ahd: 2 moles of NaOEt (50#) was added over a 10 minute period. Reflux was continued while collecting 150 ml of water during the next 90 minutes. Total reflux time was 5-5 hr. Since this reaction mixture filtered slowly it was diluted with 400 ml of benzene and filtered (l8l g NaCI- resin mixture). The filtrate was concentrated, the residue was kept 10 min. at 1006C/l mm, then 20 min at 140C at 1-2 mm, to give 442 g, mp 49~50C, analyzing, 0.07, 0.09# Cland 5.33, 5.47# epoxy (equal to 89-5# of theory). This 3y3tem seems to eliminate completely the base since 1 g of sample is neutralized by only 0.013 milliequivalent of acid. Resin found in l8l g of salt mixture was 27 g. Overall conversion was 89.7#. Expt. 33 NBP 367471-2 _* 1 The purpose of this experiment wa3 to modify the general directions given in the two previous runs by eliminating a part of the - NaCl before distilling off the ECH. The di-hydrolyzed Aroclor 1268 (0.5 mole, NBP 365971) was dissolved in 3 moles of ECH, and 1 mole 50# NaOH solution was added over 2 minutes. The mixture was refluxed for 40 minutes then 350 ml of water was added. The organic layer was separated from the neutral aqueous layer. Excess ECH was distilled off; benzene (2.2 moles 170 g) was added then NaOH (0,5 mole) as 50# solution was added and* the system was refluxed for 4 hr (5-5 in previous expt.), azeotroping off the water. The reaction mixture was then diluted with 200 ml of benzene and filtered (very fast) to give 255 g of resin, mp 47-8C Isolated by the procedure given in the previous 2 runs. Conversion 97.6# Analysis gave Cl-, 0.l8, 0.21# and , epoxy, 5.14, 5.20# (equal to 85.1# of theory). There is no. advantage in this modification. | | :> i ' I - i Expt. 34 ; NBP 367489 * I In this experiment more* ECH-.was used in an effort;, to increase the ` s< epoxy, borvtent i The., final resin solution, was . contacted twitli water tcf a'fire^*t-llj l^n&fsaf# are fvemofedf 4 1f sitting of O. 5 mole? iydrolyzed Aroclor 1268 (NBP|36597lr i '* arid1 iou.es |trpate<liwithl.0.5lniole. of NaOH addec oyer ' ^ ** *---* ni&utf - f4 I! DSW 621621 STLCOPCB4095652 ('*6res Bettisne* rxlteredf (125? g ani rgfL.Ii& aol^e!wlS^n^4taf|l2(fe nmlilj)t.ywaasa t-h-o-r-o-u-ghly-A Jjcashed 800 mlfoi^&terf^ithd .ragln^.soiateid by standard |ro$edure to collect 221 g, mp 4?~C.7 ' - , ?# i f Anal. Pound (*): total Cl, 42.94, 43-16; Cl~, 0.11, 0.12; % j| --i epoxy, 5-47, 5-^5 (equal to 91.2# of k'f ? j ; theory. Total water-insoluble material ; " found in the salt residue was 16.3 g. Combined conversion wa3 91.5#*r. Expt. 35, NBP 367^91-95 The Duroose of this run was to demonstrate glycldylation n a larger^lze batch. Di-hydrolyzed Aroclor 1268 (1.5 mle%6</ 8 6 was stirred in 6 moles of ECH at 90C then 1.5 moles of 50# StoH was added over 35 minutes. After 40 minutes the solution 2!s neutral It was diluted at 58C with 700 ml of water good laver separation. After removing excess ECH, benzene (500 g) was added and reflux was continued at 87C while NaOH (1.5 mole as c0<y solution) was added at once. It was refluxed for 7 hr whlle_ tS? water was collected at the rate of distillation. It was refluxed for 8 more hr. The hot solution was decanted from salt, and three benzene wash-decantations of the salt were carried out (1.0, 0.8 and 0.8 1. benzene). The benzene solution was filtered and washed (emulsion) then dried to give 617 OB of resin mp 46-47C of excellent purity, analyzing Cl , 0.08, 0.0^ , aacnnoadnveeeppr<osJAxio\y, n, * 49, 5.54#. y, y jyr is 79-5#, not The yield is 90.5* of theory, and co__u__n_tin__g_____r_e4s-hin^.^ oinaUtheml vfatultr^mixture The i. leehpcoxy. and Cl." co-ntent check very well with the values of the Ffvfp*ijevir'oiouus experiment;! t*"" * r- ' 1 * - f- ? d. Glycldylation of hydrolyzed Aroclor 1262 and 1248 Expt. 3b 316063-4 i Partly hydrolyzed cruae, Aroclor 12o2 (1.33 moles, 382 g, Ne, I 3<58.6> was dissolvedfinf8l0 g of H20 containing 2 24 moles oi l,aMldanewd ^oTvterw^alOs Sh^enaitneudteast.i 4T0h*Ce , tethmepnerEactHure(x.w7a5s minoclreesa,selbd-*- ggra) dwuaaslly tl 101C, and kept for 1.25 hr at 102-104OC. In the same time 1: i a % __ i____ Tl-i 1*2 Jdil' 14 fen ,402 446#;] n- residue? total Cl; 43.14# .... i i. t. -......................... DSW 621622 STLCOPCB4095653 . resftf '"by using more EGfi^ih'eiice reducing . f' ?' ' *' Hsrdl^l^rted A roclor 1262^(0^3 Snolevf^06-%r(l $ ': dissolved in 0/6 sole fNaOH (^.5>N)f #To ttM * "^'Vas; added at 70Ci 0.6 mole of ECH over 1 hr; heatlx riued' for 1 hr at 70c. The solid reaction product ^waS^pui*ijpi#d in the Waring Blendor to collect 120 g, mp 95-10CPC. ^ * Anal. Found ($): Ci 42.19; H, 2.65; ;0#|43.58,#43.76 (by old unmodified Clvmethod). Olycidylatlon of hydrolyzed Aroclor 1248 Expt. 38 323114 One mole of partly hydrolyzed Aroclor 1248 (251 g, J| Katon, 321037-1) was dissolved in 2 moles of 10$ NaOH while 2 moles of ECH were added at 70C over 102 minutes. Reaction was slightly exothermic and gave, after longer heating (1.25 hr) a light viscous resin. It was not worked up since starting phenol was mostly monofunctional, therefore product is of no value as an epoxy resin. e. Diglycidyl ether of tetrachloroblsphenol A The purpose of this series of experiments was to supply material for evaluating glycidyl ethers made from Monsanto tetrachloroblsphenol A. Expt. 39 334642-i A solution containing tetrachloroblsphenol/A. .^(0.3 |nole, ; 109.5 g, mp 134.1-134.9C) and ECH (1.2 moles. 111 g) was stirred at 80C and 0.11 mole NaOH was added as 30$ solution. After a 1-hr reflux unreaCted ECH was distilled off. .The residue was diluted with dloxane, filtered, and the filtrate was-diluted with dloxane to a total of 500 g. This solution was,, treated with O.^o mole NaOH (30$ solution, theory + 10$ excess) and refluxed for 45 analyzing 3.54, minutes.j 3-57% Cl'-, Total and isolated 4.37, 4.4< le^ethpbxyw-ra3r if1 e to 6656 diglycidyl ether content). Expt.440 334645-6 t- I Ut Slftlioii conia^ningftJtf*:! _ ,r #.5is)i* ,;ECH (4 moles/ 370' g,f 100$ excess) Wasf stl nole NaOH was added (as p0$ .solution ovei IWC and :es?. i A: J .. It imp eraj it DSW621623 STLCOPCB4095654 Of filtrate"It sfaS^eated to raflux and lTTb mb] , , ,, Excess) of 30 jNa05solution was added,' reflux was fcontinued 4 hr.| The two iaye)r8 were separated and the dioxane)was dla-t tilled off at reduced pressure (100C pot temperature) Ibogivt 478 g residue (quantitative conversion), ng5 1.5882. Solvent purification (by J. Herbig) increased the mp to 95C. \|* H Anal. C21H204 (crude material) Calcd. Pound C 52.6 52.95 H 4.18 4.34 epoxy oxygen 6.66 5.87, 5 theory) hydrolyzable Cl 0.0 0.12, 0 Expt. 4l 360245 __ This experiment was performed much later to become familiar with the St. Louis pilot plant process recommended by R. C. Cass (R 4) . for the glycidylatlon of tetrachlorobisphenol A. It was desired to adapt the process to the glycidylation of our hydrolyzed Aroclors. Tetrachlorobisphenol A (0.55 mole, 202 g) was dissolved in 3-3 moles of ECH (mole ratio 1:6) then 1.1 moles of NaOH was added as a 50^ solution at 80C over 55 minutes, some heat was given off. Mixture started to reflux in 12 minutes, after ,33 ml of NaOH solution were added. The mixture was heated for * 10 more minutes) then distilled at 20 ram to give 139 g distillate containing 105 g ECH, (bath temperature to 150C). The crude residue was diluted with 190 g of benzene, heated to 80C, and 0.55 mole NaOH as 5056 solution was added; reflux was continued for 6 hr. A small sample was Isolated and analyzed 5.87# of epoxy. Original reaction material was diluted with water and washed ,3 times to neutrality. Solution was dried using azeotrooic ; 1 > 'distillation, *2.5 g charcoal and 2.5 g of filter-aid wab a'ddedf ~ I and filtered. The filtrate was distilled and finally kept at| -150C (bath temperature) and 20 mm to collect 238 g, analyzing 10.6l, 0.66# Cl- and 6.09, 6.1556 epoxy (equal to 91.25/6 theory)), fconverslon waa 91.5#* This confirms data reported by St J. Louis ff H * fMi ifliflffI STLCOPCB4095655 f#I mli*'t*s. ^i3wv4fc- 56 ( e|i^^^^^'eontalning 46f22# Cl f ,, _ ^ 5? IljppflsJ d^a^ld|lnf ECH . (1.8 moles, , 1t3'iaEfeld.6fa&l#of added over 3.5 hr. r Pot teraperatlare was' Inc rea 3 ed gradually from 42 to 95C, and the system Was stirred for 6 hr. Material was worked up to Isolate a resinofxs product (l21 g)< ' vr : Mi f . > Anal. Pound (#): C, 44.i2; H, 3.02; no residue; total Cl, 42.82; Cl", 7.28, 7.31. . This high Cl" content indicates Incomplete dehydrohalogenation. This synthesis was repeated later when we had collected more experience on glycidylation. V- ^ ' ^ . J; ' Expt. 43 353865-6 * A mixture of hydrolyzed Aroclor 5460 (0.2 mole, 196 g, same batch as previous run) was dissolved In 1.6 moles ECH at 85C, then 0.2 mole NaOH (30#) was added, followed by reflux for 1.5 hr. The excess ECH was distilled off, the residue was diluted with dioxane, filtered, and the filtrate (350 g) was refluxed. To this was added NaOH (0.8 mole, 30#) over six hr and reflux was continued overnight. The solution remained alkaline. After removing the NaCl and solvent the resin was dried at 115C/l-2 mm pressure, to give 228 g of a yellow-amber resin, mp 85C, analyzing 0.04, 0.17# Cl- and 3.60, 3.61 epoxy. The latter value shows that some material has polymerized. Most likely this was caused by the relatively long heating period. This product possesses good properties in coating;application (R-3). ^ All - * f 3*4# k - - g. ^Attempted reaction between Aroclor 1268 ar.d glycerol & . - Expt. 44 f NBP 288842-3 : , Since Aroclors react smoothly with mono hydroxy alcohols an unsuccessful attempt was made to react Aroclor 1268 with glycerol in thej.Bresence of NaOH. 1 |}o Jether was formed. It was planned to dehydrate" the' glyceryl ether ^to tfte ^corresponding glycidyl i 4 ii - it 1 t i tj-iyffLycidylatlon of .dihydroxy diphenyl sulfor.e . :t fli'H* ` M------. Bi. l. uy * ^ - - -- " ' - - - f ' ' ef, 41-iiihydroxydiphenyll si Lfone ; (l. 5, moles,i 372.5 g) hoi ko " * twbjteri- L ;Ejthap<jiy(200 il 3] : was `' ~ DSW 621625 STLCOPCB4095656 3r,,.' f rbre this -urn ..___ ^ Jfehe' sfcdn^n^;'|li^'lthj|the' cal-' valfie ht !$#f f The product Is' partly"polymerized. , ,, .- \ * * ' - 4' - -- f * h % -* 46 323103-5 ` $*? foext step was to start with o,p'-dihydroxydiphenyl sulfone, mower cost Isomer. The sulfone (1 mole, 250 g) was dissolved lfi.moles of NaOH (10#) and to this solution was added 2 moles of ECH at 75C over a 77 minute period. Some exothermic reaction took place. Material precipitated In one piece. It is not homogenous and was not further worked up. The next experiment was made using a high speed stirrer and a beaker container. The sulfone (0.25 mole) wa3 again dissolved in 10# NaOH (200 ml) stirred at 70C while ECH (0.55 mole, 10# excess, 51 g) was added over 35 minutes. Isolation of material gave 70.1 g resin, mp 115-120C. Anal. Found (#): C, 57.64; H, 5.03; S, 9.42 ___ No epoxy analysis was available at that time. 1. Dl-hydrolyzed aroclors as curing agents for bls-glycidyl ethers " ** This work was carried out by J. Schwendeman and J. K. Craver. The purpose of these few experiments wa3 to investigate dlhydrolyzed Aroclor 1268 as a curing agent :for its glycldyl , ether as well as for Epon 828. It was stated by?. Dave .Cummings t % t f ^^^^(f:fEp6rv 82: and`glycidyl^ ether of tetr^hfoTObisphenol A do not cure with their corresponding bisphenols. Samples of fipon ;828 and diglycldyl ethers of Aroclor 1268 were stirred with dlhydrolyzed Aroclor 1268 in a 500 ml flask at 180C under N2Samples were withdrawn after 1 and 2 hr period, respectively. Data on'these experiments are summarized in Table I. rreliininary |:ut!lng studies1 of glycidyl Aroclor 1268 using dlethylenetriamlne, 3uccinic anhydride or maleic anhydride were made ^by|the ^author using test tube size experiments. In one t ; instance .when glycidyl, ether prepared from partly hydrolyzed i]||l268|ai^imaleicianhydrlde wefe; heat4d|injtest tube and gllOjQ. eyesight,|a. feleaffi ^|s$:n4ne}ftingiabp4e i ^|df Me NBF|. 31^083p); f | Yf "fll f f\ ' \ t |u v . JL & i. # ... *' V l 4: S 9 tr. *' F #. WrT: T i f DSW 621626 STLCOPCB4095657 Glycidyi'kticmfoiftSiptienollc maierich*Hasfbeei? gbl patents. 1 As stated In the remarks given under:rei_____ preliminary study of the patent problems associated 1fit| and selling epoxy resins fomed from hydrolyzed Aroclors| eplchlorohydrln was prepared by Mr. P. C. Wellington:of|i and given In the Appendix Section. ^ 3 > thg Pi- Four patent disclosures were submitted on glycidyl ethe'rsv^ their curing and uses on work given In this report. %* ' 0 ?' ' D-I869 Dazzl New halogenated epoxy compounds and their uses, dated ^-1-55. .^ D-2J20 Dazzl Glycidyl ether derived from Arociors as<p*5lf stabilizers for polyvinyl chloride, dated 12j-^1-56. D-2J21 Dazzi Glycidyl ethers derived from Arociors as plasticizers for polyvinyl chloride. a D-2^77 Dazzi-Herbig Hydrolyzed Arociors as curing agents for glycidyl ethers. Disclosure D-1870 (Dazzi) New Glycidyl ethers derived from lihydroxydlphenyl sulfor.e will be abandoned since It was found ir. the patent literature. . j, . 5. I; I t -1 * r ; I t* i iiMlN uv {VMif DSW 621627 STLCOPCB4095658 IL'-sf Oi^l^iatlori' o^ ^henolS.'C^c^rapdundi,' eapedillEf^lblaphlr^kAJ f ^JpfpSi aa<p^pbher reaction! oilf^^|oxirari4. ring (riui*i|g)' iajfeported. I m in ;se^eral hundred:.!!. S.fpiterita,' arid very few' aummariea.f Most pertirient data are found in a good summary "Chemistry"off Poly " Epoxides", by R. Wegler, Angew. Chem. Vol 67, page 582-592, ' ' 1955. (209 references). Trade literature from Shell . i teaches how to cure their glycidyl ethers derived from bisphenol '-Ai $ Few details were published on epoxy resins back in 1955 when this " ? work was started. Now more Is being published in form of symposia papers presented in U.S. and England. 1. Dazzi-New Epoxy compounds derived from Aroclor, Memo to the Idea Review Committee, dated ll-l6-5;i. 2. Dazzl, "The Aroclors and other HaLogenated Aromatics as Chemical Intermediates". Memo tc Dr. C. A. Hochwalt by J. Dazzi dated I-I6-56. Parr I (part II and III a3 continuation). . 3- J. H. Schwendeman. J. K. Craven, J. Dazzl, J. R. LeBlanc and J. A. Herbig. RD-58 Final Report No. 1115 "Synthesis and Application studies on hydrolyzed Aroclors and Derivatives". Cass, R.C., St. Louis, Research Report No. P-762, "Suggested Procedure for the Preparation of Diglycidyl Ethers of Tetrachlorobisphenol A". Dazzi, J., to S. E. Gebura. Preparation of glycidyl ethers ; derived from hydrolyzed. Aroclors 1262, 1268, 1270, .memo iU dated 9-6-55- ' '' ! ' '* ' *` ! .6 Gebura, S.E., Preparation bf high epoxy oxygen resins from hydrolyzed Aroclor 1262, memo to M. F. Drumm, Springfield, CC to J. Dazzi. Touchette, N.W., St... Louis, Evaluation of glycidyl ether of . | hydrolyzed Aroclor'lfe6;^ as polyvinyl chloride plasticizer and f f5 stabilizer, memo to J. Dazzi dated S-l^-po. V-o .' - . :-... - ;- -- - ) \> .# ' $f I t Wellington,,F. C.,iEpoxy Resins Utilizing hydrolyzed Aroclor, See Appendix.;; I 4 f i * . a ' ; * * '* * *f si M > *- * MII ll ?. i DSW 621628 *11 7" 'If ?: STLCOPCB4095659 STLCOPCB4095660 STLCOPCB4095661 pa'H cHjfSj .* . . f % ; # - r v * . ' Ii ff^llPtMSf rW'- x ' \ ft k " No; material specifications were set on any of the products described. Specification for hydrolyzable chlorine in glycidyl Others . derived from,bisphenol A (Epons)lls 0.25. >Low : hydrolyzable Cl content is very important in curing epoxy resins. , \ #14 r | f * sf f ' * ii HI ill ill f !iI imi ks $ f I fi DSW 621631 STLCOPCB4095662 4 ;i r J fV w*<? ft II: Will 1 l 'tRjfi (t f I ;:;`i -I I * 6- ' y - 1 s; determined by the E-2-55 procedure hydrolyzablfe chlorine, the C-3B-56 method was used. Combustion Analyses for carbon and hydrogen were run at 800C. Total chlonLfiA was determined by the C-35-55 and H-8-57 methods. The latter was developed specifically for Aroclors and Is superior to previous methods. For details see Final Report on hydrolysis of Aroclors by J. Dazzi, No. 1182. i - . jjl ff fr |i i t i- * I*H f if M -I t iif. i J If i i S :.& I lv| .1 ifilliil DSW 621632 STLCOPCB4095663 jk& i m* ai frS * I ijg | *^ !f. ; *' i ',' ^ * # ;u No toxicity data have^been obtained on products report;ed^her#.A Special care should be exercized in handling eplchlorohydrin,f a toxic material. Longer exposure to Aroclors, octachloro- i naphthalene and their derivatives can cause skin irritation, which was experienced by the author. if t Hf H *s I I i-l H-fMt 4 *; - !I DSW 621633 STLCOPCB4095664 OT .$?, St. Mr". J. A. Herblg's curing studies on the early glyeidyljether ^ fsamples and his very valuable^suggestions are gratefully;acknow- * I' ledged. Ve thank the Plastic Division, especially Dr. S. ' Qebura, Springfield, for the process study on glycidylation of hydrolyzed Aroclor 1262. Most helpful in this project were the evaluations, suggestions and team work given by Messrs. J. K. Craver and J. L. Schwendeman from the surface coating group. We thank Mr. J. R. LeBlanc for developing a process for making completely di-hydrolyzed Aroclor 1268. Messrs. F. R. Short and H. W. Schwartz of the Analytical Group developed a procedure for epoxy determination for Aroclor derivatives. Their contributions are gratefully acknowledged. We thank Mr. J. R. Darby's group, St. Louis for screening samples. f nf i-l f * it |l|| j? f ? : 11 DSW 621634 STLCOPCB4095665 STLCOPCB4095666 STLCOPCB4095667 Fxto*i $iN*OHSA; f: ICOMP/ Hr. H. J. Erase Dayton, Ohio March 20, 1956 S. M. Evans' _ . _ H. K. Nason - St.4 Lojlis ) . J. H. Lura - St. LottiV : E. W. Gluesenkainp - bay ton * M. Kosmln - Dayton W. R. Amon Jr. - St. Louis J. M. Butler - Dayton J. Dazzi - Dayton D. B. Sharp - Dayton R. L. Kelly - Springfield EPOXY RESINS UTILIZING HYDR0LY2ED""AR0Ci;0E3 This is a summary of a preliminary study of the patent problems associated with making and selling epoxy resins formed from hydrolyzed Aroclors and epichlorohydrin. Inf ringement Castar. - 'J. S. 2,321*, ^83 would be Infringed if the subject epoxy resins were condensed with a polybasic carboxylic acid anhy dride (such as maleic anhydride). However, this patent expires - tuy 01 .990. Greenlee (Devo'e & Raynolcs) - U. 3. 2,698,315 might be Inf ringed- depending upon whether the (i.e.. . ori.nated) dipner.ols obtained by hydrolyzing Aroclors. The file history of this patent has been ordered In an attempt to clarify the Interpretation of this patent. No study has been made of the validity of either the Castan or the Greenlee patents. There are a large number of other patents directed to epoxy resins modified or mixed with other re3ins or cured according to so-, 'iflc techniques. Many of these could be infringed if the particular modification or curing technique of the patent were to be utilized with the hydrolyzed Aroclor epoxy resin. ;entability f The method of * hydroll'zlfig Aroclors in ethylene glycol with sodium hydroxide Is unpatentable over Smith (Monsanto) U. S. 2,^9.088 . ff f' DSW 621637 STLCOPCB4095668 Rentable over?' s<n |The|ch3 ihah|$lj| r- )leci unexpectedly superior properties when compared with the octa- ichloro or lesser chlorinatedfdihydroxyaiphenyl compounds. At the present time we know of no such superior properties. The use of hydrolyzed Aroclors, as a general class, in epoxy resins is unpatentable over Moss (Celanese) - U. S. 2,319,876, Carpenter et al (Courtalds) - U. S. 2,602,075, or Hatcher (Allied Chemical & Dye) - U. S. 2,695,276, all of which disclose the use of ^,U1-dihydroxydlphenyl in epoxy resln3 and state that the dihydroxydlphenyl compounds can be substituted with halo radicals (expressly including chloro radicals). It is possible that some specific halogenated dihydroxydlphenyl com pound will be found to have unexpectedly superior properties which will make that compound patentable over the broad class of chlorinated dihydroxydlphenyl. At the present time we know of no such properties. There is alsc a possibility for patent protection relative to particular types of Aroclor hydrolysis products - e.g., chlori nated trihydroxydiphenyl or trihydrox;. terpher.yl. However, these products have not yet been evaluated. __ 'lor ar t does leave a fair degr of latitude for making oa` ;able i er.tions with respect t odified (e.g., esteri- resin3 blends (e.g., with ure melamine or phenolic >) or f or special curing techni s utilizing epoxy resins upon h ydrolyzed Aroclors. Thu s, it is not at all un- i that Monsanto could still obt aiu a significant patent .on to back up a manufacturing ar.i sales effort with ;t to h ydrolyzed Aroclor epoxy resir.s. It should be emphasized, ;r, tha t the inventions necessa for such, a patent position oot yet been made. . Efforts to air. these patents would :ly hav e to be preceded (or at 31 accompanied) by a fairly Le appl ications or formulation earch program. If such a : posit ic-n is to be achieved, t hydrolyzed Aroclors should not Leased or disclosed to the trad r.til our position has been L * shed. iI H It U tut 4* '. Wellington M IMf Lli DSW 621638 STLCOPCB4095669 tm n- : | fm % i t APPENDIX - C IE S tM I. Curing of IHgiyciaylnsthers With Dihydrolyzed ArocTor 1268. II. Analysis of Partly Hydrolyzed Arocior 1263, St. Louis Pilot Plant. Ill. Calculated Analysis of Diglycidy'. ethers Derived From Hvdrolvzed Aroclors. n t i* mi illiiiil r ii1 STLCOPCB4095670 STLCOPCB4095671 STLCOPCB4095672 STLCOPCB4095673