NOTICE: “BEST AVAILABLE COPY”
PORTIONS OF THE FOLLOWING DOCUMENT ARE ILLEGIBLE
The Administrative Record Staff S REPORT IS REPRODUCED ON LIGHT AND HEAT SENSITIVE PAPER
I I I I D 8 Im 8 I I I 8 I I I I 1 I I A report prepared for the Deputment of Energy / Rocky Flats Office and managed by the Los Alam Technology Office Rocky Rats Plant Golden, Colorado
PYROPHORIC POTENTIAL OF FINELY DIVIDED PLUTONIUM METAL IN SOIL AT THE 903 DRUM STORAGE SITE, PLANT, ROCKY FLATS GOLDEN, COLORADO
David L Clark
and Nuclear Chemistry Division haAL.mos National Laboratory
June 1991 Table of Contents
Executive Summary ...... Pnmary Factors...... Additional Mitigating Factors...... Conclusions...... 1.0 Introduction ...... 2.0 Terminology ...... DNm 3.0 Hidtory of Rocky Flab Piants 903 Storage Site. - 3.1 Uw of 903 Drum !%rage Site ...... 3 3.2 I#t.cn...... 3
3.3 capping d 903 Site>,I...... 4 3.4 soil chvrteriutior\...... 4
4.0 Factors Affedng the Fymphoricity of Plutonium ...... 3 4.1 AutoigniW and Watjop\ B&avior ...... 6 42 Partick s& d Activity ...... 7 4.3 Effect d Mo&ture".-""...... " ...... *...... 9 4.4 EfieddHjTdrQp..-...... *...... * ...... 9 4.5 Purity, Cotn- and Cmamtration of Plutonium in Soils. 13 4.6 Amormt of Rotmsve Stdace Oxide ...... i .1 4.7 i(inetic and Thermodynamic Considerations ...... 6-
5.0 GmdwioruDNm on tho Pyrophoric Potentul of Plutozuum Metal .- at th Xk3 Storage Site ...... - 6.0 Achowkdgments...... -
7.0 Glossary ...... 13 7.1 pyrophau, and Pyrophoric...... 19 7.2 I@tion, Piloted Ignrtion, and Autoignition ...... 19 73 IpWa Thmparture ...... 19 7.4 and Spontaneous Combustion ...... :3 8.0 m....."...... 2'3
11 1 - 1 The Department of Energy’s DOE’S Rodcy Flats Plant, near Denver < o.( rldt,, ! is cunently addressing the question of plutonium pyrophoricity in regard +u environmental restoration activities at the 903 hrnStorage Site Plutonium I metal has been hwnto ignite spontaneously, and the possibility of A ylutor,iirn fire is a concern. 1 In the eariy 196(k, oils containing p~utoniumleaked from dru s*ored a! !?‘..e I 903 Drum Storage Site; the estimated release was about 5,000 gal. The distnboQonut
U plutonium in soil under the 9a3 Pad is lrhmpmms, making it difficult to estimate exact cuncuttratiam A study performed a few yean after the site was I from capped in 1969 found both rsopU (in cancentrations ranging 427 pCi / g to rOA5 I pCi/g) and (in -tias ranging from 282 pCi / g to 2,273 p~i/R, m the soli beneotfi the cap. 1 Thb repat examhe8 the pyrophoric potential of finely divided plutonium at the 903 Site by Surmnuiring literature data ud Uuiynng the probabdiry and ertmr 1 pyrgrhanc to which plutmium is in its present state in the 903 Site sod. I P-F1ctob* I CJculrdoCu show that putidcr of &st- plutoruum have a cntd ptiderfirrd-y Smthat is, only partrdes smaller than this sue I wo\Jd karpebd to rtwm pymptumc betuvior on exposure to air. Oils were sreved befoa I phdin thc dnrmr, SO plutonium partides that entered the wtl were no more than 100 pm in dhneter. Dilution of these particles with large amount3 of soil duces rheir wrophoriaty because heat produced in oxidation processes is absorbed by the surrounding rmtenal. Additionally, oxidation of plutonium by gueous rlpntr much as oxygen or water vapor generates an adherent
111 oxide film on the surface of the plutonium particles. TIUSoxide surface furt.Ser reduces the potential of pyrophoriaty. Thus, the combinanon ot :actors-the requirement for a very small particle, the presence of the protective oxide 5u1:4tce, and the dilution of putides in soil-should render the plutonium :n the %?7 Pad soil highly inert with respect to pyrophoriaty.
Additional Midgating F#ton During the in- g years, however, other mecharusms have further rduaed the podbibility ad pyrophoricity of plutonium m the wnb. It is probable that reactions between the plutonium and aqueous groundwater mixtures of chlonnated solvents known to be pmnt in the sail beme& the 903 Pad have rendered the plutcmium IYWI-PyrDPhOCK* UndCI thc canditioru presented. The plutcmlum oxidatjosh Idn+tia showr that dw &time d a 1- plutomum particle LS only 8 4 days in liquui water, utd 4 minutes in a carbon tetrachloridemethmd muture
Gmdusimw
It is conducted from th discusions and data presented that the plutoruum :n is is them Pad sdl inert with nqr+ct to pymphmaty. This amdusron supported by the fad that IY) pyrophiaty was found during the November 1968 remedlai action at the RX3 Ikrnn Sbqe Site, which involved suaping plutonium-
am- ad Irmrlh area The reactions ot plutonium meul or oxlde - in plrrsa wltb thr wyanmsive organic mixtures the soil under the 903 Pad* are stpcrd to k lwtmdt0111and complex. Thus, without exact data on the form chemical and speciation of plutonium under these conditions, it is impossible to predict with absolute accuracy the ultimate fate and transport of plutonium con taminants in the 903 sic mil. Prophoric Pottntid of Finely Divided Plutonium Metal
Soil at the 903 Drum Storage Site h * at Rocky Flab Plant, Golden, Colorado
BY David L Clark
TkDOE'S R0dE)r Flats Plant, near Denver, Cobrado, 19 addresing the quesa d pbWm -ty at the 903 Dnun Storage Site Because plutonirw mctJ hr betn known to ignite spontaneouslv, concern about the appclpriate.1-i Ihe pyroptmidty o€plutonium u possb&ty of a plumurn fire LS a prMr, Seriour mtba kame d worker, and envirmtal health and safe T?tb irrurCb time!ly-the DOE b pluvungremodial activlty at the 905 DNm WhM Strmge Site, has kcn contaminated with plutonium 5 From October 1958 to January 1%7, drums ccmtlining waste macRine cutturg oil that was radbctidy amtarnhtd were stoted at the 903 Site.', As a result d leakage of about 5,OWgl. ot wdf that occuned from dnuns while they were stored at the rgpppdnudl %ita 86 8 (5.3 Q) of plutonium was released into soils st the 903 Drum Sqdirr'3 rrphrk kr OOMahwnt pad was applied at the site in oaober 1969. Several years after the asphalt amer was applied, a study of the soil beneath the 903 Pad indicated that the soil wucontaminated with both plutonium and amenaum is The purpcme od this report to examine the pyrophoric potential of finels divided plutonium at the 903 Site. The report sumarlzes literature td:a on !dC:\C. influencing pyrophoriaty and, using this information, analyzes the grobaoLrts drd extent to which plutonium is pyrophoric in the sol1 at the 903 Drum %rage 5;~
'Ibc fdhfhg informrtim, which was drawn from historical records, aerd
p-phy, ud in&n&w~with plant personnel, is an abridged version of data presented in the FdPhase II RCRA Fadlity InvestigationlFeasbdity Study Work pim.5 From ocdober 1958 to January 1967, the 903 Site was used to tore drum\ contiuning waste machine cutting oil that was radioactively contaminated 3 Litg3t (it the drums were 55 gal. in size, but some 3O-g;l.l. drums were also present, and not a.i of the drums were full.6 About 75% of the drums at the storage site were contaminated with plutonium; the maprity of the remaining drums contained ur ani urn. Mort ot tk plutdrncontaining dnuru held wate lathe coolant consistirg d a high-moleculu-waght, straight-bin hydrocarbon oil (Shell Vitrea) and carbon retrrChbriQindiVaWppXWXU. Other site drums contarned lrqwd waste, such as hydraulic oils, vicuum pump oils, trichlorosthyiene, tetrachloroethylene, silicon db, acetsme, still bottom, e&. The drum ammts were ongutally Lndiated on dwoutridc, but -and lac% of other supporhng records has made detaikdidcntiAatbrrdiffieJt
Shipmattddrums bo them Sile ended in January 1%7, when dnun removaldsolrtrl#gh 6yJunc1%8,alldntmrhdkcnranoved. Fneberg desuhs moreeutensivdy the chrondogy of the 903 Site deanup' w 32 -IUlu(lrddffr0ar *drums was first noted edy during the site's use, and UtLA in 1999,ahdambuwaa to the oil to reduce the conosicm rate of the steel drum& Durbrg huuiling routllu &tnn operations in 1964, drum leakage was again
noted 8t the 933 SIe. The umtmts of leaking drums were transferred to new drums, and the artire u11 wufenced to restrict access.8 Infonruticm provided by
individuals nho WLII involved in site operations indicates that about 420 drums leaked b mme of these, rpproximrtely 50 leaked their entire contents 8 It IS
3 I
I estimated dut 5,000 gal. of waste oil leaked from drums onto the ground at the 903 Site. Biwd on oil samples taken from drums, the average plutonium concentratinn 1 x CI) was 4.54 10-3 g/L (280 pCi/L). Thus, approumately 86 g (5 3 of plutonium &'a5 I released into soils at the 903 Drum Storage site.'
I 3.3 cappingof903Site In November 1%8, grading began at the 903 Site In preparation for an a5phd:t i cap to be pWover the contaminated area. This work induded moving from I ccmtaxninated sad around the fenced area to imide the fenced area 7 Of primary sipticane is the fundunenu fact that IU) plutonium py~ophonatywas I found during this ranedid ~ction.7 In May 1%9,33 drums d rdhetively cuntuniruted soils were removed fill I h.oan thearea- Tm,layersddcu\ material were placed overthesitedunng the I late SullUllcI d 1969. The asphalt cuntlinnwnt pad was applied in ocbober 1969 TIe apu h 1 approocinudy8an thick and coven a rectangular area about 370 x395 h is fill (146,150ftq. Ih, cap underhn' by 15 cm of loose grave1 and 8 an of duty kbnury dditiaul I In 1970, coarse road bmaterial was applied to sods east and south of the asphalt pad beurw of low-level plutonium contamination in the I surrounding Wil.7 I 34 sofibrrtrbairr$ioa L rfta I scrrnl ymuw the upNt cover was applied, Navratil, Thompson, and
Kod\arexururwd* SOiL knrth the 903 Pad to determine the level and distribution I of plutor\ium and ammidwn.9 This study indicated that the soil beneath the 903 I Pad wucontaminated with both %(h 427 pCi/g to 20,455 pCi/g) and 241Am 8 (from 282 pCi/g to 2,273 @/@. The contamination appeared to be restricted to 4 shallow depw blow the original ground surface.9 It was estimated that tne d$FrLdit pad covered 18,000 tons of contaminated soil. In addition, data trom the Phaw [I
RCRA Facility Investigation Remedial Investigation Study Work Plad I Round One, 1989, Volume 111 Appendices) showed plutonium contamination in w.dice:s for stations at Operable Unit 2. Phase I Remedial Investigations of envrronmental media showed :hat volatile organic contamination exists in soils, surface water, and ground water .n areas adjacent to the 903 Pad, as wdas under the Pad itseIf 5 Groundwawr ramr;ic?s with were found to cmtain carbon tetraddon& concentrattons ranging from several hundred to several thousud micrograms liter. Wells were found to contain tetrachkoethylene m corupntration, ranging from a few pg/L to 528m Pg/b and ykne in amcentrations up to l2,O pg/L Samples contained - compounds, sigrufbndy lower axumtraths d other voiatile orguuc including vinyl chbrib, 1,l-dlchloroethykru, 1,ldkhbxhne, chloroform, di-, di-, and aetaw Groundwata samples from the 903 Pad ate also contained radiimuclicb, although individual correntraticms were bebw 1 Kilt 5 These data tha~suppart the rewuring that the plutonium in the 903 Pad site weil confined to thesoil beneath th Pad and its immediate s- 8-
Ubarsurr patrfnfng to the tem pyrophoricity ranges from mathematical developmt d Qwnblutbn theory12 to dixwsions regarding unanticipated fires or explodiau that have occur& in the agriculture, fuels, and metallurgy industries. Early literature pmsatts an hirtoricll record of the state of knowledge at the time of npotted writing, such as many Occurrences of spontaneous combustion of zuconium or UTMium metal scraps. At least two review arhclrs on pvropCor,( .:'/ are avaihble.13~14 Many utides have been published on the pvrophonc:tv (1: urconium and rircalioy powders154 and metal powders in grmeral42,C 'n
4.1 Autoignition and Oxidation Behavior
When an element experiences combustion, or catches tire, It is znderqoing oxidation. chemidy, d elements not m thmr highest stable oxidation state ~n d.r can burn; conseq~ently,all metals that form 5table oudes can oxidire in ar tinder appropriate conditions. Many metals oudua 50 slowly that the heat generated during oxidation dissipates and igxutkm temperature u never reached, such as uon papa Unda canditiocrr, however, sane metids oluciize rapsdly m the presence of airamoi.trrru,- ruffida\t heat to reach theu igrution temperature and undago autuignit&m.
Ckmdy8 autaigniticn can OCCUI whenever a reachon or reactions are Str~glY- 'c, have negative free energies of reaction at the igruaon temperature, and have a greater rate d heat produchon than rate of heat transfer away froan the repctjQI ullc In such instances, the surface Oxidahon becomes vigorous enough to make the rawtion self-sustauung. fine For powden, the oxidation behavm 1s similar to that of larger spec,mer.s Unierr the nmaoal d heat relased by the owdation process is too Slow. In thus case
the &spmtum d tfu powder rises, leading to an increase m the rate of oxxiation '3 For #u&axwm, this autoatalpc process usually culmnates in the
powder i& igtith bamperature and catching fire. C.RSdmitthrrwi+wtd the parameters that induce the pyrophonat), ot
various xnatcrirls and points Out that they are numerous and often interrelated l3 The variabLes sd\mitt dtes as affecting pyrophoriaty include the following: 1. putw,sizeandactivity, I
I 2 moisture content, 3. hydrogen content, I 4. stress, I 5. purity and composition, 6. amount of surface oxide, and I 7. massofmaterial. The mre immtof these factors will be Myreviewed in light of cur~mi 1 literature, and their potcntsll impact on the pymphoriaty of piutoruum at the 993 I Drum Storage Site will be disawed. I u Pard&SiZeaadktivity "Activity" refers to the popa\rrity of a powder to react with its mvlronmont; I it dererrmncr the type- rrtcdrartionr7 Theactivity ot a pvckparMe 1 detemhs the rateofmrtairl transport by bulk and surfacediffuuon, the rate of adsorpth and drsoaptiac and the ne of oxidation mctims with the I environment.lf A major frbor that stroctefy influarar activity is the amount of findy surfact area available fa cknkal activity. Because divided partxks have J ios u greater surface area avW dvmial activity, thy are perally mor0 eb susceptible to Wrophoricty than the same quantity of metal in bulk, or massive, I form wdbe49 Indrrd for reactive metals such as titanium, zirmnium, I uranium, and @ition temperatures are markedly dependent upon putidtsiaa~~ I In dkudam d prtid+ dte characteristics, a useful quantity is the total 1 surface area of one grun of powder, called the speafic surface.47 This vdue is usually expresd in d/g.Because any reaction of partides will start on the sun'ace, is I the specific surfrcr an excellent indicator of the conditions under which the reaction and of the rate 1 will start of cwction. For a fine pwder, the large surface I area relati- to the mrss (or volume) of material available for absorption ut '.-e ?PJ' produced during oxidation permits a rapid nse in particle temperature $\'hen !?a! temperature reaches the ignition temperature, autoignition can occur Evans, Borland, and Mudon determined that pyrophoric powders haie a crihcal surface area-above which they will undergo autoipinon- of about 6 m21g.a Applying this criterion to the factors controlling the combusmn ut zirconium powders, Anderson and Belz discovered that a zircun;um particle diametu of 10 )un repmmts appKndrmteIy the threshold between hazardous and no- powder, Using 8 0.01- to 0.04-J spark igruhon energy as a cnten0n.3~
Other Llytljll aimdisplay thdrown critical particle sd,wh& IS mque to each s- s- Faexampie, mctrb such as inm, coppe!r, nidref (001- to o.o3$lrn ptlde size), tunpten (1- partide a),or sodium (1Oym particle size) spontaneously igni!einllirwithdirttrrt#ncr 01 li@ hpad.45 is putLde Fnnnthis~it clear that size must be considered in the evllua- d dw wropbarldtyd piufoniurn parades dcpoated in the ooils under tho903Pd Ihrdl wassieved witha 1Oo-pm screen before It was placed in drurrs therdrre, at them cmtm WtageSitt; an upper size hitof 100~can be for putida assumed plulor\irrm in the d.50 Assuming a density for Gstabrlrzed plutonid1 of 15.92 g/cd ud pfdyspherical particles, a 100-p partxle ot will rpdAc plutonium have a surface of approximately 0.004 m*/g, whereas a 1-bm would ~uranrUmprbd, hvt a spmfic surface of about 0.4 m2/g. Ibr rpdAc mrll surfc~for plutonium are due to the assumptqn of pafactill #p&&cd potidrr ud the fact that plutonium metal is among the densest of all eiunentrs l-bvemr, mabl putides will not be perfect spheres; cort5equantly, tky am expectd to show much higher speclfic surfaces than those cd&taiforsphaialparticles. The dticd putide size has been experimentally determined for zirconium,~5
8 and this rnforaution can be used to calculate a conechon factor to obtain more reliabie critical puticle sitcr for uranium and plutonium If one assumes a de~srt; for vrconium metal51 of 6.49 g/m3 and a spherical particle. urconium particle speafic surfaces are calculated to be 0.009 m2/g (100.~partide) and 0 9 rn2, y ;-,m partide). Assumiq a critical surface area49 of 6 m2/g,one can calculate the crincal m. particle size fix a dKlonium sphere to be on the order of 0 2 In canpanson, putrd+ Anderson and found the critical size for zirconium to tw 10 pm. n'is provides a mversion ham of 50 to obtain mae reliable estmata of the actual critical part& bites. Using this oonvenion flctor and a cnhd surface area of 6 m2/g leads fo an estimated critical putrck size of 5 WI for Gstatnlwd plutonium
4.4 EffedofHydmgen is The effect of hydroga\ ofdm interrelated with the effect of moisture in
9 causing haeased pympbriaty. For example, reaction of plutonium with water generates hydrogen gas,*% which, under the appropnate condihons, can adsorb through cracks in the metal to form hydrides49 Hydrogen gas liberated by reaction of an active metal with liquid water may also be ignited by the heat of reacton nis represents the prinapd fire hazard of alkali metals such as sodium or potassium For plutonium, however, oxidation by liquid water converts plutonium metal into a monoxide monohydride, Hpuo, which does not pose a pyrophoric threat 56
4.5 Purity, Cornparition, and Gmentration of Plutonium in Soils In someases, impurities an cauae metal powders to become more pyrop~~in othm, they CUI decrease the pyrophaiaty of powders- NaK alby, for ewsnple, is monwlaphaic IJ\d hrr a knmr mdtingpoint than the we eicmonts.~ Lilrcwir+, thewrophoriaty of sinnedanents ma- when they are
alloyed with such ekntnts U tin, lud,OrgoM.~ While Qxllminind isnitiorr tanpentum of pure and alloyed plumurn, fouad schnizldn and Firdwr that low-purity samples have the hughest 1gNhOn temperatures and the higfwrt-puity samples have the lowest igxutmn temperatures-606f Th mmcqtMity of piutonium to oxidation can therefore be reduced by llloyine it with othr metals, such as aluminum or gallium, which has thecffrctdinuadngttw~ticmtemperature. Inrklitlar_inaauLni.l such as sand can be mixed with pyrophoric An powders ta damp Br, ud expbhns. inert material reduces the combustlbihty ParQr of a hrruv it akxbheat. Normally the amount of inert matenal
necessq to prevent a fire 01 expWon is considerably higher than the cancontraticm that would be tokrated for a metal powder. Durins work on zirconium and &alloy pymphoridty, Cooper observed that 90% or more inert material dddamp the acpIoriaru of dust clouds of drconium.45 This obviouslv
10 bv has important rrmifications for reducing pyrophoricity of piutonium d;,:.on with soil under the 903 Pad.
The preceding discussion shows that the distnbution ot plutonium An WJ,:~.j a cruaal factor in determining how much heat per unit area will tw produced !x is oxidation reactions, whether there enough soil to asorb the heat, and therefore whether autoignition can occur. The distnbution of plutonium at the 903 Pad is inhomogeneous, and the plutonium wll occur on surfaces of large aggregate Xs shown by Navratil et PI., the contamination range for plutonium in soil ranges fram 427 pCi/g to 20,455 pCi/g of d.9 Thu latter ]unit represents a plutonium mncentntim of approxinutdy 033 pg Pu/g of sod. Although these simple conversions estimate the plutonium soil concartntiOn, they do not provide infornution concerning the chemical form of plutonium in roil. Thus, other ficton must be amsidered
is Pluta\ium a very reacthe rmtd that ondizes readily when exposed to oxygen a waba vlpa. A number of studies have besn wnducted to characterize the nridlth ot plu&miuxn.- 67-76 Most of these studies were undertaken to e* compare the rwrtivity d pIutonium and plutonium alloys and to OWsome quantitative oxidadan data on th, effects of moisture Charactemation data (obtiirud x-ray dUfmdh, x-ray photoektron spectroscopy, eiectron drfracuon s-, s-, .nd mctrtbenphy) show that water vapor reactions wth unallo!d plutonium podua a diverse mixture of products, depending on temperature These ozcida~products include Puq,PqOs and a monoxlde monohvdnde HPuO.77
Thc oxidation cd a metal is preceded by dissociative adsorption, with the hlgh heat of akwptbn falling as further oxygen IS adsorbed onto the mitial laver 42dAt hsstage, many medm form protective oxide films; the growth of these films is governed by a parabolic rate law in which the rate of oxlde growtn IS IRVYTWLL proportional to the film thickness and can fall to zefo at some limiting tiim thickness.39N This limiting thickness typically lies in the 10- to 3O-A range '' b3 The kinetics of the process are dependent on the diffusion of an ion or efecncln through lattice defects. An adherent oxi& diffusion layer forms on the surface of plutonium dcxg gaseous oxidation,*m thi, layer is criticll in preventing the pyrophoncltv ot plutonium metal exposed to air. It then can be seen that 03 pg of piutoruum metal will have a much greater pyrophoric potentid than WIU 03 pg of plutonium dionde. €bweva,in- * with liquid water, the inaormng water mokuk s drcmicrl able to diffuse through the oxide kyer. in rertiolu with liquid water, 'Jur plutonium metal anbe quantitatively converted into the monrmde monohvdnde. HPu0.S Thus, for plutonium metal in d that is in cmtact wth liquid groundwater, fornutiar ot a prpteaive oxide layer is not significant
4.7 ~andT&naodyurnlccOnrid~ti~
PluWum metal is a highly reactive substance that CM react wth praacaLv every eknwnt eLLlcpt wbk gases. A dotailed summary of amdernd oudat:on Can be found in revhaby cdmaure5.6.66
Mart have a free energy of formation that is on the order oi whifc kilocrlab pa mdrr, actinide metal oxides have a free energy of formation ondrcaQId~dldloalaresper mde. As a consequence, the actmide metals are thamodynunicdly unstable in air, and in principle, one would expect complete cOnveniOn of actinide metill into the oxide that is most stable m au. in practice, this dcm not OCCUI for the actinides because of the kinetics of the oxldation process in air. In liquid water, however, the reaction kinetics are very different from
12 those d the gwus reactions, and a particle of plutonium immersed in !quid v.a:t-r will undago complete corrosion.S5$6
The kinetics of the reaction of plutoruum with liquid ptidw-5 are round :o differ substantidly from those observed for gaseous reactions nepr;rnark difference appears to be that in 4uid water and carbon tetrachloride reachunr. +!-e is corrosive liquid able bo diffuse through the oxide coating, with an end resuit t>r mrnpkte oxidatba There have been several reports on the conwon of piutoniqim caused by liquid water in synthetic seawater3 and sdt solutms TIW reactrons ut u- and bplutonlun\ (29% Gal in water and I-M ult soluboru were studrod by theproductjon 04 hydmgen gas formed by a se~uonceof hydrolvw reacthas Inamtrast~OrcwtbruplutoniummetalandgaseouJoxygen, whMfamapotcdiw~layer,repcttonrwirh liquid water are extremely Uquid cwlodoc to plu&miunr. In water rmctkms, plutonium metal was found to mvert qwntitativdy to a AM biwlt powder, which was identified as plutonium momxi& marohydridr ~HPuO)by means of thennogravimotnc analyru, x-rav dihction, and x-ray -. Other hydroips products were i~ucmrr aimidenmai* in hydrolysis mxlia; these included a second hydride (Hgu.rOo)ud1~dlailk(PUZOLPUIO\ZI~bPU\OO\bPU120LZ,and Pu@). Kinetic studia cm the liqurd water readon with plutonium reveal that the reaction is atdyosd by If lubmagcrd in seawater at ET,a I-pthick sheet of pluba\i\lm myfJ Wip be amverted to HPuO powder in approximately 16 days & B#B$ d &e ckrmodynunic and kinetic dnving forces for oxidation ot plu~trLpMllf#lrtoamduckthat after 23 to 33 years of exposure to cissokd oxy- water, and habgauted organics in the immediate environment, the plutonium bpodteci in the 9Q3 Drum Storage Site hilo oxidized into a non- Pm- form. For the axrosion of ktrbilized plutonium in ordinary tap water, a diffusion-
13 I
I controlled corrosion reaction predominates with a conosion rate of 0 26.1 mg Pul(cm2 hr) for the initial stages of reaction.55 During this i:>.miprocess, water I must diffuse through an oxide product layer before reacnng with plutomum rea, 1 After a podof time, the corrosion layer cracks and flake, and the reacnon rate is accelerated to 0.52 mg Pu/(ad hd.55 i These important Linttic data make it possible to calculate the lifehme or lOo.(lm plutonium pMide in ordinary tap water As5uming a density51 for I &stabilized plutonium metai of 15.92 g/cd and a sphencal particle, a 1oO-prn
plutonium particle weighs 6.67 x 165 g and has a surface area of I 26 x 10-3 an2 At i the initial corrosion rate of 0.261 m8 Pu/(cd ht) for the diffusmn-cmtrolled lipd 1 water reaction, a 1- plutonium partick will be ComplQtCiy consumed in 200.9 hours, a 8-37 &ys, in mnnal tap water. A larger putde surf- area A I greatly haease therateofamosh by lquid water. If the aaud surface area of a 1oo.rUn plutonium partide is So times greater than that of a perfect sphere, the puade I calculated lifctinw of the pIutonium u reduced to only 4 hours. Therefore, 200 houtscur be usai as an upper limit for the hfetime datW+im, &subdued plutonium partde in adtnuy tap water. I The aquams d plutonium also has been shown to be catalyzed by fm salt, and th rate bstabilized plutonium corrosion in a I l.O-MCaCl2rd~timi~inocuhddramatically to 15.85 mg Pu/(an2 hr1.55 For piuooaiurn, I a-phase t& same -ion conditions produce a corrosion rate of 31 mg h&dh?).s Ibru, naturally occurring salts in Rocky Rats groundwater are I expectd M &te th, amdon rate of plutonium in contact with groundwater under the 903 Pd. *
I Dab fran -water sampling at the 903 Pad revealed concentrations of carbon I 1 tetrachkide and other chlohatcd orgurja.5 Crisler has reported the facile oxidation and dissolution of plutorniwn metal by a carbon tetrachloride-methanol mixture.n The heat of
I 14 I be 7 readon for this oxidation p'ocess was found to -205 kcai, ,nok %::h a cormwlf' *t- --- 13 mg Pu/(cm2 min) at 25"t. At Crislefs corrosion rate of 13 mg I'u/ I cm2 m.~ -xw? spherical 10QIrm plutonium particle wil have a lifetime of only 4 minutes in A tetrachloridemethano1 mixture.
Because of the simikr reactivity of methand and water, it IS highlv probdo1e that plutonium d also be oxidized by carbon tetrachlande-water mixtures rn 5011 or by other liquid water mixtures with chlorinated organics in the sod Several recent artides discuss the dissolution d actinide met& with halogenated orgamcr Karraker has reported thc diuolution of "Qpnuuum and plutomum metal wth 1,Z-diiodoethanein THF solutioru, producing solvated actinide iodides fp Cluk and Zwick have found tiut simple )ulosnu will axnpkely dissolve amide metab porU thorium, uranium, neptunium, and plutonium in O~MCdonor mlvmts, produdng 8dvent adducts Thx4(L)r,Ux3m4 (X = &, n, Npl3CL)& and PuI3(Lf& where L is a wide vuioty of 0rg.mk dauw dvents.~~~in addition, Sue? tfd haveshownttutthir~. b acdtrated by the use of prmc solvents such as aia3hols and water.= As indicated by the above reports and the known mivebehavaor of water and hlogenatui otgurio towud plutonium, it u likely that aqumus groundwater mixtures of chldxutd sdvents will be highly Cotrosive tow& plutoruum u\ solis atthe9(13DntmStumgeSlk
Calculations show that &stabilized plutonium particles w11 have cnbcal particle sizor of appmcimately S p;partides smaller than this size would be I
I expected to show pyrophdc behavior on exposure to air As noted, howesrr 'TP
oils were sieved before they were placed in the drums, so plutonium particles in the I p. soil could range in size up to 100 As discussed in Sectlon 4 5, the disnibunon ut I plutonium in soils under the 903 Pad is inhomogeneous, making kt d:rticult to U estimate exact concentratiam. It is also known, however, that dilution of these particles With lap amounts of soil will reduce the pyrophoricity bv absorption 01 1 heat produd in oxidation proasses. It is ab0 well errbbluhcd' that oxidation of plutonium by gaseous agents such I as oxygen or water vapor gene!rates an adherent oude film on the surface of the plutonium particles. This oxide surface further rduces the potenaal of I pymphoricity. Additionally, other mechanum (reactJans between the plutonium and aqueous poundwater xnix!um of chloriruted solvents known to be present ut I the soil knuh the 903 Pad) will prevent the Wrophonaty of plutoruum in the w. I llKsellquldraraa\r* are the ultinute bamer to pyrophonaty d the plutonium dcpoutEd in du soil. The kinetio of the sudatron of plutoruum show I that the lifotim+ d a 1- plutw\ium partide is only 8.4 days in Iiqwd water, and 4 minutes in a carbon betr;lchbndc- methanol muture Tholefon, I from the above data and dixussioru, it can be conduded that !he I plutonium in th 903 Pad roil is inert with respect to pymphonaty because oi the low spedfic rrtrfrata plutIlnium combrned* with a protective oxide surface and
dilution d pIsccddl h #u Additionally, no plutonium pyrophcxiaty was found In I W the NOVU&U 1%8 action at the 903 Drum Storage Site wheflsoil was I saapediMorrtuaa-7 1 U 1 16 I I
1j
M The author is grateful to Dr. John Haschke and Dr John W t4;ard !rum toS the Nuclear Materials Technology Division at Alamos Sational Labora!w. :<\r helpful discussions and reprints of pemnent references To provide a common basis for discussing pyrophormtv, &fini:ions : I technical terms used in this report follow. Many of the detinitions were taxrrn verbatim from the Fite Protection Handbook,'o which 1s a compreher.sive ant authoritative text on this subpa.
7.1 PympbortrrandPyropborir
The word "pyrophorus" is derived froa\ the Greek word for hrebeanng i1 En@& Dicaonuy Tho Oxford defina pyrophonu as any substance capabk
(apecUlly in a findy divided state) of taking fire spontaneously on exposure to au.11 is The adjective "pyrophoric" ddined as having the property of taLung fire on exporun to air." A distinction is often made between those dements and flune compounb that burst into on exposure to air because of theu intrinsic drm\ial behavior and W that do so only because of thew special method ot pnpsntior\ (such as finely divickd powders). In this report, the term "p)?ophonc' is used in icrkodcrt w,thu is, it applies to rUge masses of pure metals, or alloys, cubides, hydn&s, etc, rather than being restricted to fineiv divided rneta,s
72 IsaitiOa, PhhdI@h, and Autoignition*o lgmitbb ir the poccor of initiating combustion. The Oxford Enghsh as heating to the point of cornbushon or chermcai
-11 fr further defined depending on its cawthat resulting from the mull introduction of some external flame or spark is called "piloted ignihon', that Ocrurring spontaneody, or without the assistance of an external pilot source, is called "autoignition."
18 73 Ignitba hnperattutl0
The "ignition temperature" of a substanco IS the minimum temperature :t must attain to ignite. Usually, the piloted igrution temperature ot a substa~ce$3 considerably lower than its autoignition temperature
7.4 Combustfon and Spontaneous Combustion10 Combustion is an exothermic, self-sustaining reaction involving a iuei an3
M oxjdant 'Iha proau i8 usually associated with the O%AdatIon01 a fuel bY atmospheric oxygen. The orfad En@ Dictimdry defines 'cornbunon- as the action Q pcess of burning, whereas "spontaneouscombustion' u defined as the within act at takiq fire, ottnlming through oonditiais produced the substurce itseIf.11 1. D. E. Patterson, "The Rocky Flats Fire," Fin \ourmi, 64, 5-7, 15 (Jdnuarv 1970)
2. U. S. Atomic Energy Commission, Serious Accidents, Issue >JO I~u,5m~. Plutonium Fire Leads to %pr Property Damage Loss" (November 27,1957,
3. U. S. Atomic Energy Commission, Serious Accidents, Issue Lo 258, Cattizg Wheel Residues in plutonium Waste Cause Expl-ion" (December 17, 1%;)
4. u. S. Atomic Energy Commislrion, Senma Accr&nts, issue No 306, 'Fire- RdCy Flats Plant-May 11, 1969" mcembe? 1,1969).
c 3. FdPhase II RCRA Facility Investigrltiar Remedul. hvestigatmn/Feuibiltry Study WorL; Plan, tlocLy Flats Plant, 903 Pad, Mound, and East Trenches Area (q#lblcUmt Na 3, U. S. -t of Energy (Apnl1990).
6. K. W. e-, Manaudum to L M. Joshel, lhw Chemnf Company, Rocky FkB avidat (AUguSt 19,1970).
Momaudrnn 7. K. J. Rkh~& to E A. Putzor, Dow chemrcal Company, Rocky Flaa Diviria\ (Apnl 1% 1970).
8.
Thompww\, 9. J. D. Navratjl, C. H. R L Wen,Waste Management of Actin& Contunhted Soil," Ibchwdl International, lnternal Report CRD79-016 IS, 1999.
IO.
11. '131~She O+d English Dictionary On Historical Principles, 3rd Ed. wth CorreaionS, oxfad University Press, Oxford (1987). I
I 12. G. E. Zixna, "Pyrophoridty of Uranium in Reactor Environrr,ents. ' Kcport !f 62442, Hanford Atomic Products Operation, FLchlanJ, LL'a*h;ng'on L Jmtiar'. 22 I 1960). C. 13. R Schmitt, "pyrophoric Materials-A Lterature Review, I fire Jn.i I Fhmmbility, 2, 157 (April 1971).
1 14.
I 'An 15. W. R mlidh&mr, Evaluation of the Zirconium FLuard,' HW-44989, i Hurford Atomic Products Operatim, Richland, Washngton (1956). 16. N. U L&t& 8. J. Julkn, and M. J. Steindler, 'Management of Waste Cladding Argonne I Hulls, Part E- 'ty and Gnnplchon." ANL-8139, Nahonal Arlparn, I Labor-, Illinob (1979. 17. I. Hutm;3inn, J. Nagy, and N. Jambsen, 'Explosive Characterstla of Titamam I Zirconium, Thorium. Uranium and Their Hydrides," R I 4835, L'.S Bureau of Mines, WuhingtDI\, 0.C (19Sl). 1 18. F. E. Hdt and S. G. Wb"A Codc of Practice for the Handling cf Zirconium its and AUoys," ICR-TN-$578, United Kingdom Atomc Enew Authority I Great Britain (3957).
19. W. W. a''Zbwnirrm, 2kdoy, and Hafmum-Saie Practlce Guide tor I Disposal, swvWsbrEng-8.- ' g and Suap WAPD-?.V-:' Aloodc I (Rc9p)E Pomr bboratory, Pittsburgh, Perasvlvania (1960) "An 20. 8. I(ullar, N. LA@ and M. Wdler, Assessment of Zirconium I -ty and -tiow for Handling Waste Hulls," ANL-77-03, I Natiarul L&xatoty, Argonne, Illinois (1977). A 21. M. J.oob#c Cooper, and J. Nagy, "Explosibility of Metal Powders," R I. 6576, u USBureau d Mines, Washington, D.C. (1964). 1 21 i I 22 s. Co+ ud N. J. Wen,'The Measurement of the Ignition Temperature ot Zirmnium and zltcrloy Powder," AEEW-M-J84J,Technology Dsvrsron AEE Winfrith, Dorchester, Dorset (1981).
23. J. K. Damn, V. C. Baugh, and J. F. White, "Observations un the Edrlv Stages clr Oxidation of Zirconium and Zircaloy-2," Efectrumechr n~alTechnology 4( 3- 4):137 (1966).
A 24. E. Culbransen and K. F. Andrew, "Oxidation of Zlrcomum and Zirconium Alloys in Water VapAtmaspheres Containing Trace Amounts of Oxvgen at
3750 and Z"t8" Ektrochcmud Tahnology 404) (1966)
25. W.8. Hurir, Wthand Safety in Powder MetaJ Handhg,mNYO4869, L'S .. AtomicEnezgy- ,New Yak Operations (1931).
26. E. W. Kmw, "How On Company Avoidr Ziramium Pyro~horiatyHazard.' Nuclcorria 15(8k72 (August 1937).
Expbdw\ 029, n. USAEC, "Arr ud HUu& of Meullic ZiKoni~m,'CEH-23, US. Atomic Energy CanmisQicm, Washington D.C. (June 20, 1956).
28. Argonne Nlltiorul Labotatoty, "Metal Oxidation and Ignition,' ANL-RCV-SL- (FebruKv 2280, Parts I and II, Argoruw National Laboratory, Arpnne, Illbas 19sI, - Apnl1957). ?
29. R 8. smith, "PppkxWty: technical Mystery Under Vigorous Attack.- Nm1k28-33 -kr 1956).
30. F. E. utthum and F. M. Church, "Reactions of Titanium with Water and Aqucour Sdutbnr,' AECU-3675, U.S. Atomic Energy Commission, Menlo Park, cllifanu (Much 1958).
31. E. L Stout, "Safety Consikauons for Handling Plutonium, Uranium, Thorium, the Alkali Metals, Zirconium, Titanium, Magnesium, and Calcium, LA-2147,La Alrmor National Laboratory, LOJ Alamoa, New Mexico (- (- 4,1951).
22 32. J. A. HeridceJ, et d., "Zirconium Hazards Research Progress Rcp~rt#I, 4; C L 3642, U.S. Atomic Energy Commission, Pittsburgh, Penn\vlvania (August 1957 i
33. J. A Mckes and P. A. Richardson, "Zirconium Hazards KLvearch Progress Report U2," AECU-3682, US. Atomic Energy Comrmssion, Pittsburgh, Pennsylvania (February 1958).
34. R 8. La\& "MumReports Regarding Acadents, Storage and Hand!ing, CEH-231)87,Wstin%o\w Electric Corporation, Pittsburgh, Pennsylvania (August 1956).
35. M. L. Smith, "A Survey of the Pyrcqhmaty Hazards of Ruptured Fuel Eksnenb: HW4013, Hu\ford Atomic Products Operation, Richland, Washingm (Janwy ts, 1957).
36. J. H. =tal, Spar- Ignitiorr of Highly Irradiated Fud," TID-2508 (Deccmkr 195s).
37. A. J. Bend)a, "Fast Transient Heating and Explosion of Metals Under Stagnant Energy AECU-3623, USAm Commission, Columbia L'NVWSI~, New York (kkuuy 1958).
38. L. S. Nelr;olc "Explodut d Bunring Zirconium Droplets Caused bv Sitrogen, Science l4&1S!M (Juri l8,196!9.
39. H. C. IRb L H Bek, "Facton Controlling the Combustion of ZiranErPrn hw&S/ 1. EWoch. Lx.,100,240 (1953).
40. N8tbnut RaWctbn Assodation, "Zirconium," Nattonal Fire Coda 3 (1982).
41. I. Hartmm, J. Nagy, and H. Brown, "Inflammability and Explosibility of Metal Powdar,' RJ. 3722, U.S. Burcru 04 Mines, Washington, D.C. (October 1943).
23 42 J. P. Evuu, W. Bodand, and P. C.Mardon, "Pyrophotiaty of F,ne >fetal Powders," PdnMetallurgy 1, 17, (1976).
43. G.R Longhunt, R M. Neilson, Jr., and L. J. Porter, ''Tritium Storage Metai-W Pyrophoridty Measurements," CONF-871007,Proceedings of the 12th Symposium a\ Fusion Engineenng, EEE Service Center, Pixataway, S J Kktok,1967).
44. G. R Longhunt, "Pyrophoridty of Tritium Storage Bed Materials,' Fusion Technology 14,750 (1966).
46. R 8. Smith, Ihr Industrial Atom, The Fm Pqer&s of Metallic t'raxuum,' Rcpat Tnr&oI?, Thr U. S Abamic Energy C-, Washington, 0. C. (April 1% 19s).
47. A. R Pot-, Edi-, Hondbodr of Metal Pdm, Reinhold Publishmg Corp I New York, N. Y. (1966).
48. W. D. Wiudnrar, Uranium Metallurgy, Vol I, Uranium Process Mttdfurgy, Vol 11, Utrnium Cwmsh ad Alloys, Wiky Intencience, New York (1962).
49. W. M. hhdk,J. P. Bhdedge, and C.G. Libowitz, Metal Hydndcr, Academc Pmah New Yak (1968).
50. 1. D&rris, paronrl annmunication.
51. R C Wlllt, CKC PfandW of Chcmzstry and Physics, 64th Ed., ERC Press, aevdmd#ohb (1984).
52 F. A. cottar\ and G. Willrinson, Advanced fnorgonrc Chmutq, Fifth Ed., Wiley Intendexice, New York (1968).
24 53. W.H. Dennis, Mrtrllvrgy of the Nun-Ferrous Metols, Sir Isaac i'itman 6r Sons, Ltd., London, p. 574 (1954).
9. 8. Lustman and F. Kene, Jr., Thc Mdallurgy of Zirconium, Sational hbciuar Energy Series WI-4, McGraw Hill, New York (1955)
55. A. E. Hod-, XU, J. J. Reynolds, J. M. Haschlre, "Corrosion of Dvlu Plutunrum (by]Rocky Flats Tap Water," Rackwell International, Rocky Flats Plant, &port m-m1 (September 10,1979).
9. M. Md). Baker, L N. b,S.Orman, Trans. Faraday Sor. 62,2513 (1966).
59. J. J. Kat& G. T. Seabar& and L R Mans, E&., 77w Chistry of the Actinde Efmmts,2nd Ed., 2 Vda, Chapman and Hall, New York (1966).
60. J.G.sd\nizidrr and D. F. Fe,"Plutonium-ignition Studies," Argonne Ndoaal Ubardq0Chartid Engineering Division Summary Report, ANL6723 Mpd. May, June, 1963).
I 61. J. G sdrdr'lhr .nb D. F. Fie8"Plutonium-ignition Studies," Argonne Nathal-, Chanid Engineering Division Summary Report, ANt.66uI (oaokr,bb-8 December, 1962).
62 J. C.schnizldn u\d D. F. I.ldrer, "Isothermal Oxidation of Plutonium," Argom NatkmaI Laboratory, Chemical Engineeriq Division, Aln4687 Bmuuy, Ftbnruy, 1963).
25 I
63. L. E. Rurscl, "?'he Sintering of Pyrophoric Powders, Powdcr ,S?etdJurAy 10, I 239 (1967).
C I 64. J. L Swanson, L A. Bray, H. E. Kpmo, J. L. Ryan, L Matsuzah, S G Pitran, and J. H. Haberman, 'Uboratory Studies of Shear/Leach Processing of Zircalov I Clad Metallic Uranium Reactor Fuel," PNL-5708,Paafic Sorthwest Laborator);, Richl~~I,Washington (Ikember 1985). I 65. I I 66.
I 67. I I 68.
I 69. J. L. StrlrCbrLc, Vxidath of Plutonium-lwt.$ Gallium in Nitiogenl5S"Tc oxyglm 1 --"j. Nwd. MtCI. 148,115 (1987). 1 70. J. L SUebke, L A Lmb, "rhr!Kinetics and Oxygen Pressure Dependence oi 1 th T- -tion of Pu-lwt.%Gil," 1. las CO~WI.iUd. 136, 349 (I=#&
I n. 0. J. Ed., Phbaivnr Handbodc A Gvde to the Technology, 2 vols., (Am. I Gcx&n IRd Bmd, New York Nud. Soc. Reprint, 1980). 72 L. R crirlclr, "Investigation of the Carbon TetrachlorideMethanol-Plutonium I bhtd SySt8!n," Dow QILnriul Co., Rocky Flats Division, Report RFP-2181 (IkCt!!nb I 6,1974). 73. J. G. sdrnizldn and D. F. FfKher, j. EIcctruchm. SOC. 114,s(1967).
I 26 I I
I 74.
1 75. 1. L. Stakebake, 1. Lrss Cam. Met. 123, 185 (1985).
I 76. J. W.Lindsay, K. Terada, and M. A. Thompson, /. Elecfruchnn Soc 119, 73 I (WZ). n. J. M. Haschkeurd J. L.St&ebake,”Kinetiaand Roductsof the Pu + Hfl 1 Reaction at High Temperatures,“ Rockwell IntmutiOruJ, Report KFP-4392 A. U 78. E. Hodges, KU, and J. M. Haschke, “Cmmsion d Delta PIutunium by Synthetic Sea Wata,’ WwdlIntenuthd, Report Rm-2919, Rocky Flats I pknt (ocbokr1979). 79.
I Bott, 80. D. L Clark,A P. %ttdbqerOSG. ud R N.Vrtis, “LawuBuc Adducts of Uranium Triiodidrc: A New Class of Syndmidly UdRecunon for u Trivaknt Uranium ChatWqf Iwg. Uum. 28,3771 (1989).
I 81. 8. D. Zwidt, t. R Av-, and A. P. Sattebqer, persod communication. N. I 82 S. Sua, J. G. Watkin, and W.G. Van Der Sluys, persand commurucation I I I I I I I I’