Document JN447KnbyROm6Y9Ej62BvQnje

TO: A.M. Matyger FROM: W.C. Smith Pedricktown 1/2B/86 G^ Near Miss Pedricktown, N.J. Plant Repetitive Accident Type Toxic Chemical Release 10A redacted SUMMARY On January 17, 1986, Vinyl Chloride was released into the Mass Resin Process Building when the manhead o-ring on Autoclave 8300 partially blew out. No injuries occurred. Board of Review A board of review was held at 0830 hours following the incident. Those attending the meeting were: A1 Matyger Stu McDonald Area Manager Mass Resin Engineer Mass Resin Lead technician Mass Resin Technician Mass Resin Technician Computer Console Technician NARRATIVE The Mass Resin area was in the process of running a routine product. At 1230 hours on 1/17/86, Autoclave 8300 (RE8300) was "set-up" for charging and became "ready" at 0253 hours. The charging technician reported that the o-ring looked good and fit snugly into the o-ring groove. He also stated that he did not have any difficulty in seating the manhead. The autoclave passed the vacuum test and at 0343 hours was charged with initial vinyl chloride under standard operating procedure without any incident. At 0453 hours, the seed charge was dropped from the prepoly. The charging technician added the rinse catalyst at 0503 and reaction temperature and pressure was achieved at 0552 hours. At 0635 hours, reported that he heard a sound similar to a gas discharge but he could not isolate the source of the noise. At 0639 hours, Console Technician, stated that the computer alarmed indicating that Autoclave 8300 had a low internal pressure of 138 psig (set point of 155 psig). NGC 16130 REDACTED put on a Scott Air Pak and proceeded to investigate the source of the low internal pressure on RE8300. Upon determining that a leak was occurring by the manhead, notified the Mass Resin building personnel, the Console Technician, and sounded the gas alarm. The autoclave was immediately put on full recovery and given full cooling water by the Console Technician. Simultaneously, while not being immediately aware of the situation (he had been breaking for the Console Technician) entered the Mass Resin Building. He sensed the presence of VCM, left the building, donned a Scott Air Pak and joined All three Mass Resin personnel, wearing Scott Air Paks, worked to minimize the extent of the leak by applying wet rags, exhaust hoses, and emergency covers to the manhead. called the Spray Dryer Technician and told him to shut down both direct fired spray dryer burners. (Standard Operating Procedure). Within a few minutes the leak was under control because RE8300 pressure had dropped. Facts Surrounding the Incident 1 Autoclave 8300 passed all precharging tests and checks. 2. The manhead o-ring was inspected and looked good prior to reinstal1 ation. 3. The manhead was installed properly. 4. safety procedures. are all experienced in the Mass Resin operation and 5. There were no injuries. 6. There was no'damage to equipment. 7. All emergency procedures were followed. Conclusion of Findings The cause of this release was equipment failure due to a failure of the autoclave's manhead o-ring. NGC 16131 Management System Investigation 1. Autoclave 8300 c-ring was inspected, as part of the standard operating procedure, prior to the Installation of the manhead, 2. All operating procedures were followed by the Mass Resin personnel. 3. All Hass Resin personnel are trained to operate the building independently of each other. 4. One of the Scott Air Paks, that was used initially, had some dust in the face mask. 5. The North door gas alarm switch was not conveniently accessible due to the presence of the leaking manhead. 6. There was good cooperation between the different manufacturing departments. 7. The loss was approximately 1,830 pounds of a combination of vinyl chloride monomer and poly vinyl chloride resin. Corrective Action 1. An inspection of all Scott Air Paks located outside will be made to insure that they are ready for use. Particular attention will be paid to the cleanliness of these units. Face masks will be placed in plastic bags. (Safety Department by 2/1S/86) 2. Investigate the need to install a gas alarm switch by the west outside door. (Bill Smith by 2/10/86) 3, Evaluate the o-ring that failed as well as other similar o-rings for any passible defects. (Bill Smith/Stuart McDonald/Kevin Boding by 2/15/86) 4. Review the incident at all shift safety meetings. (A1 Natyger) 6ak/013/re NGC 16132 e&> I / U- B ID TTmCOUKlBRJOUHNAl, SUNpfty^:JANUARY 15, JW* . * _ i'.* h. ^ "- '. * at Goodrich Plfty to 360 gallons of methyl aciyJate ana tthyi acrylate over- jrcwa &>m a storage toe* at b.?v COpdrltfc Co. Stoll ( 7 P-T. last night, author!mosaic*. tTfee epm *raa remained JnsJtt fhs plant. at the plantu -Httfc) IteUsX&nC *nd presented 5u> danger to the pu&,tfc tut a wnali amount angered ifee p^antis internal sewage plant rower St Seeter Bald. WTSflahteta from fik Dranmiaftd, Piwmt swg Par*, nixie SutHirfcftn and loufcr/Ute rwpofltfed iolhe spill. Xafc JJrtemtand Aagrtaat caftty John WWkltwoa eald o&mj 65 3)ro- J&S&tera were n Hit wen*. but mort inftttstanding fcyasa sate# PTiemi- tloo. _] UJs plant stopped its sewage flow 10 Irfetropolften sewer Dlstrtct ww> trm immediately after ih& spill and bepm diluting Its own system* See}*r said, * WJtMtadn sals 37 flrpJlg&iew waartegaif ioafes and masks spread an ateorfclngpower no ihnoJiemJqal that wag t lb& granoo, StesatdftmesfrcmthoapitJwer* dbs^ated toy light wind*. * j 'WjHtintOD eald ftTotlghteja'' beared svom *bi scene toy B p.m. Staler said two -mpteyeee were -exposed to Itm rtiemitnl and wry* atoetvod toy Sm&r&ncy Mwtftal Setyfeet technicians and company medical personnel t>al neither sof- - - ferediv\y harmful efforts. - tjfcrtmie axposui* to methyl aery. \ tale has produced InjuTlea to lung*, liver and hldaey* in fobomtoiy ant- * Itnala NGC 16133