Document 5bQnvZvmVRrEJoB34XYjVdNBz

AR226-2177 Dupont - Washington Works - C8 Project N O T E S from S ite Inspection C hris Arrington and D ave Roberts ' August 9 ,2 0 0 1 10:0 0 a ;m . - 3 :00 p.m . , , . ., . Met with Chris Shoop (on-site environmental) and Debbie Mulrooney (corporate modeler) C 8 is brought to the facility in "totes" and is then transferred to larger tanks T h e "tote" is a square plastic container which holds approx. 3 0 0 gallons and with dim ensions approx. 4 .5 'x 4 .5 ' x 4 .5 '. T he to te contains C 8 dissolved in w ater (20% C 8 by w eig ht). It is unloaded into the larg er container by connecting sealed couplings betw een th e tote and th e process tank. It costs them betw een $16 - $20 / gallon currently and has been increasing in cost because o f decreasing availability. T h e C 8 used to be m ade and handled as a pow der, but this w as changed as ? becam e aw are o f th e health risks associated w ith handling it in th is form . C 8 is m anufactured in M innesota by 3M . It is used by com petitors to m ake sim ilar products to Teflon in A labam a and N ew Jersey. D upont has been looking into th e possibility o f m anufacturing theT38 iFor them selves due to the decreasing supply and increased cost of th e C 8 They had looked into making it a t W ashington W orks, b ut seem ed to have ruled it out in favo r o f a location in F ayetteville, North C aro lin a. - T h e am m onium ion disassociates with the carbon chain as the C 8 is heated and dried - C 8 particles are 0.1 to 0 .3 m icrons in diam eter (possibly less on the low er en d , 6 0 % passed to bottom g lass fiber filter. T h e sm aller th e product, the m ore C 8 required in th e process rh e larg est source o f C 8 em issions is the process covered under perm it R 1 3 315D. (S tack T 6 1 F C E ) T h e products are F P 1, F P 2 , and F P 3 (pow der or sm all granular?). T h is perm it includes 3 process lines (?) w hich interchangeably `batch" produce one of the th ree products at a tim e depending on derrrand. s the original product, F P 2 results in th e sam e basic product as F P 1 , but yields 4/ WVDEP000016 \ ! EID647357 C8 n the process^Fpf ^ o u n te for'abort 2 0 ? ^ ^ ^ most permitted fo r worst case ie n rn rfl! . l 20/!>o fp ro d uctlo n - T hey are produce FP3 on one of the three lines? ^ m St<FP3 possible can't s w h e re fte C8 ddaf to ram ve w a te r and this scru b b er (tem p , m ax 2 3 0 deg. F ) u nd er norm al n n lP? dU^ 'S then run throu9 h a reco ver C8. T h e product is then (proposed to h e? \ rat, n l l u 3 " be byPassed ) to eig h teen "candle filters" and is then ven ted fh passed tro u g h a bank o f quid containing excess recovered and reverse o s m o s iiL ' 18 Separate * * * ? e c a n te rs ' filters) is C8. Som e o f th e * * * 0Ut and recover fo r recycling (facility gets a discount w hen to e j b ^ it b a c k f ^ * * * * back reassociates" with the ammonium as the'tem pe^urels'r^juce^in th^scrutoer C h * Shoop stated thal ^ scmbber has been ^ ^ a( ^ h TM ih ^ e^ ,, r ^ a 1 a e,, S ed P f " ~ e s e solution from th e reactor is r e m o v e ^ n J 3- ^ 00 ,,k e ,,th e re s t T h e Product ,herc' Mr-Sh"p* uowhiSSiliKiSiK irC8sfflinskim off the water). Therefore this p r o d u c t * the S ,ids setUe and W l (approx no*above grade ,e?Jo ab Ut 1*be,owt^ structural below^the top^f thjarge baSs of^The1c^dl3^ 6' VBntS about 5' in N a tio n stacks arranged in 2 rows of 9 Fffl k ,and'e filter bank (18 barrels with V f9 - S a v e ly should be modeled as a building) WVDEP000017 EXD647358