Document X8nka2yx4JgdZ4yd4JMbBmZMR

A.M. OAUDIN r uiMKuuTtt twwf cmhini a*. MfltG 002270 PLAINTIFF'S EXHIBIT WRG-387M OMMbtr 2, 1999 <*638876 Hr. John B. Myui, Vie* President Zonelit* roniny Libby, mamtm* Hr Hr. tymi At year nqwit I Mte i brief visit to Ubby to study year Billing operation*. Ihl* visit, la turn, suggested that a day at the Coapeny's tea*arch Laboratory la ivansten, 111., would be profitable. la the foliating pages will be found ay dandons regarding the properties of wrmlcullta, the ere, the actual aialng aad Billing operations, as eell a* son* suggest ion* for alternative a*ana for exploiting and concentrating ore*. This report is concluded with son* specific Tsroaasaditions (p. 14). VUUOCXIT1 The Libby vondcullt* 1* a ainerai of the alee faaily. Lib* ether alee niner*l such as auscevit* and biotit* it occurs in sheet* of great lateral extent and ainut* thickness. The sheets are coapesed of tightly knit greips of etaos of silicon, alualma, oxygen, iron and saywiiia, arranged In a pattern that l* non evil understood. The connection between sheets Is diff erent in vermiculit* on the on* hand and ordinary nice* on the other. It is on this difference that your Conpany depend* for it* existence. In auscevit* and blotit* the sheets are negatively charged and this is cozens a ted by positively charged potassiiu atan*. These ions ere anhydrous. The spacing of the sheets is such that the structure repeats Itself approx imately every 10.2 Angstroa units (one Angstroa utit Is approximately one two hundred and fifty millionths of an inch). Of this spacing a very minor pert only is due to tho potasslvm, which seems to fit almost exactly Into recesses in the atoms of the sheets. in "true" vermicullte the sheets ere likewise negatively charged, and VK'. coapensated by rositively <"hargad calciua and segneslia ions. These 201)97*0 004211 23638877 loot in half at mmrout at tha equivalent potassiums of othar micas, and considerably smaller In tlxai ana night thartfora a^act tha spacing of ahaatt to bo latt than 10A. Actually, hovaver, It It about 14 to ISA, that it about 4 to SA aidar than in auacovita or blotlta. Ihla inertat* Lu lattice spacing it accompanied by vstar noloeulat ahich populate tha tpaca between tha thoatt. I than tha vim of laller* that tha magnesium and calcln loot in warnicuiita an not anhydrout, but hydratad, far ainpU [b| N|0)] **, that# hydratad coapltatt being auch larger than tha arhydreut potaaaln lent (figure 1). It it nail knoan to phytical chanittt that tha tnallar tha ion and tha higher itt charge, tha greater tha probability af lta being hydratad. Since Ng** it about 1.2 A latt in dianetar than K*, and tinea attar it about 2.? A in dianetar, [mj(M|0)] H dwuld be about 4.2 A greater in dianetar than K*. The tana it alto true of hydratad calrtia ion (Ca(tt|0)H eheuld exceed K* by about 4.9 A). Ihut, nglate* nent of awhydrort potaatiun by a polyhydratad cation theuld increata tha tpaclng in nica by about 4 to SA. Naturally if the tpaclng between aheeta it lncreeaed by 4 to SA, it beconet auch larger than voter noleculet, dtlch then can neve freely in the lnterplanar apacet. Ihlt thould provide opportunity for edtarptlan of eater tc tha atone of either or both of the adjacent layert. hrthanore the tpacat night be vide enough to permit the entrance and adaorption on adjacent layert of additional lent, eapeclally frov telutiont of high Ionic atrength. Ihlt it in accord with the findings of Doctors Ziegler and Hayet at tvanston. Oi heating to a relatively lov temperature the adsorbed interlayer ater beccvas gradually desorbed and It is evltted as steam. At a nore elevated le^eratwe the complex hydrated cations suddenly become dlss>cietedt this provides steam, causing the well-xnown expansion of vermlcullte. * G. F. Walker, Vermlcullte end some related nixed-layer vinerale, in "X-ray identification and structure of the clay minerals". Chapter Vtl, pp. 199-223, Mineralogical Society of Grtat Britain Monograph (19M). 004212 T Sweat* Laytf X Silicot* Loy*r X SWCOI* LOy*r 1 SWeot* Layer Mognttwm ion %Vh6ounT m>ffr O'9o**` WWtr Fm. li CWlgtimtion of ll interlayer niit in mwirvlhr (4ii0UiiMliri. from WidtH.* Ini Fully hydrated. <4 Alnul half hninlnl iflft trawnl of the uhsaad water. If the au water moleeiile? in the original hydration del of mag* amhim (are at, only four are wow in artital nmtart with the inn. in Folly dehy drated with magnemima in hole? ia the atlimie layer*. *01557** 004213 03638879 On tit* kMlt of thtft picture th* only dlfforonco between blotlto and vom'.cullto 1 that Ntawn anhydrous potasslim on th# on* hand and hydratad cations plus froo ueter on th# other. In aany parts of th# nine th#r# occur rather spectacular pieces of ore spangled ulth coarse aica crystals. The operators refer to this material as blotlte, ehlle the mere disseminated and ertably stuff is thought to contain the wermlculito. I am not convinced that aueh a classification is correct. If you had tao distinct minerals, blotlte on the one hand and vemicullto on the other, the systematic assay for potassium diieh is present in one and absent in the ether should be diagnostic. As I uas saying to you in Libby this can bo dene readily in a routine fashion by a flame photometer or on the basis of the gams radiation of potassium. It seems rather likely to me that veralculite arose from blotlte by replacement, but no evidence has been offered to thou that the replacement Is complete in some places and has not occtmed in ethers. 1 shall therefore assume this replacement to have occured in a random or statistical fashion on a micro scale, and at the same tlae someehat more attentively In certain macroscopic areas adjoining cracks and crevices, and less estentlvely nhere the crystals mere initially coarser and sounder, or uhere thick "books" of elemental crystals occur In place of thin "panphlets". In this view there is no sharp demarcation betuven blotlte and veralculite, but rather all possible gradations. It is because the Libby mineral seems to have retained same potetaitas that Gruwr described it as hydroblotlte, thereby focussing attention on the Insertion of eater rather then on the replacenent of potastlra by a hydrated cation. It mi^it be uorthehlle to correlate the useful properties of veralculite llh Its potassiua content. It eight also be useful to sees to eitend the repleceuent made by nature by reacting veralculite with aqueous solutions e.g. cf calciia chloride or of aegneslua sulfate. This reaction could be studied not only In the cold, but also at elevated temperatures Including temperature? obtainable only In autoclaves. My though*, is that If the "bi.'V.'e" mineral can be reduced In potasulua content the yield of the a.r* c.'uld be Increased and a product of higher quality aade. Furthermore, >- a study prove fruitful tailor-made veralculite? for various . . ul.i be envisaged, and the mine production aignt include a larger 201**7*3 004214 ft -5- pryorUoi of the moro valuable eoarso flak*. Tho aapiwloa of vendcullte on heating, together with reduction of the lattice dlaonslon nernel to the alca shoot* with soao warping or crink ling in the shoots, 1* easy to rationalise as a consequence of the randoaneaa of replaconoot of whydnua potaaaiua by hydrated calclta and magnesium. At low teapotafa the adsorbed water la oaitted. This reduras locally the thickaoaa of the iatarlayor space) then dissociation of the aqueoaplea loan, sfeich oecara at higher taaperaturoa croataa water aolecules free to aove la the vapor state, and with no place to go* This ateea is planed ia soar laylayer apace by the potasslta rivets* ddeb saroud it mile now bonds are forwad between the enhydroua cations sad oaygen atoaa in the meitfibaring layers. Ivontually this trapped ateaa escapes, but by that tlae the vendcullte is permanently defamed to the emended atau. fhis suggests that it would bo of value to probe into the scale of the randoaness of replaceaant of *>lch 1 spoke, teas work la already being dene In that direction by electron alcroecopy. the developaeat of additional techniques would be welcoae, for simple the esaainetion by electron nler^> scopy of replicas of emended veraicullte both parallel and at rlpt angles to the basal plane. Ooctor Noyes mentioned to ae that swelling of the veraicullte could bo obtained by seeking the alneral In the unexpended fora first in strong saline solutions, than in water. For solutions of lithium, sodlm, *Itself. calc lies and magnesias (all with hydrated cations) there Is an Increase in voltme of ip to five or acre fold. At the seas tlae the veraicullte takes an e silvery ar golden color reminiscent of the color it acquires after thermal expension. It is not known whether the "smelling" is due to an exfoliation or to o true Interlayer emailing or to both. I have suggested emerlments to prokw Into this aspect of the swelling process. 1 also believe that the emailing property can be used as s basis for on ore-concentration process. Bsworandm !to. I attached to this report outlines the concept of this process. It has been reported In the technical literature that vormlcullte can be expanded net only thermally, but also by soax log it in a concentrated sjIjUxi of hydrogen peroxide. As this Intrigued me I had Dr. Hayes make i <!-*:. demonstration of this phenomenon. It then secured to me that some *01557*4 004215 I 03638881 aerifications should be colored to a* to find out another this procoat can bo aodo useful. A suggestion to this and ia of farad In asaorandta Mo. 2 attached to thia raport. THE ORE Tha ora conalata of tna aora or laaa blotitic varmicuiita OaacrLbad j at length la the preceding aaction alsed with an aabaatoa alnaral, pyrosono i and partiana of igneous rocs, aodarataly coaraa in cryatailisation and of a coopoaition approaching syenite. Tha voraicullte occura In coaraaly crystallised aaaaaa with flakes up to ana insn or aora acroaa. Soar of thata aaaaaa 9 to tha aiaa a/ a daaa or largor appaar vary high grada indaad. Thay ara fragile, breaking raadlly at grain botndary aacapt ahan frosan aftar becoalng aatar-loggad. Tha wralcullte also occura In finer lntargroath alth tha aabaatoa and tha pyroaano. Tha fiaaaa than aaaa to ranga froo l/l6 Inch up to 1/2 Inch acroaa, and ara thinner than In tha othar variety. Tha aabaatoa alneral does not aaaa to be of thv clnd that haa proved ao useful and profitable elseuhere. It ia pale blulah gray In color, fragile and brittle* In general It occura aa aora or laaa felted cluetore character* 1 latlcally l/i to 3/m inch acroaa. Ip to no* it haa been an imltigated | nulsancei aaaa attention aay perhaps be given to find out Aether it can be I -aade into a useful byproduct. Tha pyroaene breaks readily In sand-slxo particles often of apple-green color ehen vleued trader eater. The syenite breaks In blocay fashion readily regaining as placet of roc rathor than as pure alneral grains. Occasionally it gives rise te tree quartz and feldspar. There is also a little vein quarts in the ore occasionally containing ainor quantities of nore or les acidised sulfides, 1 e.g. peril*. \ In specific gravity the alneral* offer very little range. The veral- culite It the lightest (2.) to 2.7), the asbestos Is but very slightly 1 neaviei to 2.95), the syenite quartz and feldspar are all In the range j 2.7 also. The pyroien# is slightly heavier being generally heavier , \ *1 .. "S The range offered by the various ainerals and their overlap eases l r ' jra.ity separation sc difficult that it could fairly be described as *m5?45 004216 fut.ll*. 03636882 Fortwately th* minerals dlffar in both tlx* *nd shape. Veraicullte It both th* coarsest and flattest. This help* greatly. Mixtures of wral- eulit* and ether einerals else can be nade to behave differently ehen crushed in nechines such as rolls In *ilch centre! Is exercised in regard to th* near est approedi of the crushing surfaces. These phononsna are utillxed in th* earlier or dry nill and in th* never or net mill. Up to non no use ha* been nade at th* Libby nill of difference* in ettability betnssn veraicullte and gangue. Difference* in uettablllty in an environment controlled by chenical reagents is st th* basis of both flotation snd aggleaeratien. Theta are currently the nest widely -used ere treataent processes. They have been so successful because of their extra* ordinary flexibility in reducing the specific gravity of one or another ainersl in alaest any ere at stiff*, flotation is suitable for particle* finer than 0.01 to 0.02 inch and It therefore net useful for the beneficietlen of veraicullte. Put aggleaeretlon dileh can be used for particle* qp to l/t inch in else, or even seaeshet coarser* could profitably play a large part in your Libby operations. 1 will rtvert to this topic later la th* report. Other possibilities shlch do net seen to hove been explored sufficiently ere electrostatic separation and dry aognetic separation. Out no thee* pro cesses asy require oxtroerdlnorlly dry solids they net only aoy be tee expensive, but else objectionable in view of th* ulterior processing of the wreiculUe. Fiffther Inquiry 1* desirable but it should be given lev priority. finally sea* attention night be given to th* thought of reducing by swelling th* specific grovity of veraicullte to ss to facilitate Its gravity separation. This is th* topic presented In one of th* atnorahda ettechod to this report. THE CURRENT OPERATIONS Figure 2 Is o block dlegraa of the current operations. This Is very hloh.> condensed froa an actual flowsheet and designed to bring out the wmicutil* discards. First there It the large tonneg* discarded at th* nine. Th* co^osltlon of discard is vxicnovn. Mhen completed a stapling project noe In court* *0i 004217 p*tnn< put *6tuuo; otvpojdd* !*) ue^ido 6ui!lW >w3 je utxfttia taou Z ItfiOIi Wfl cu> A ^ ItuefitMippibi >9 A SS4 '*! 4. M! P* *P A <06 Atp/l 00 ; (IM9) 09tJ9UttU*0 1 % fOUlIPW JO 0UIMWU9 I1M M|* J* uonttd odoyt put nil Ajq C889G9C0 A *51 tm aim *" nna 09 fcnim 09 II* r *P/1 ost C* M O) nMPWNN ir Atp/1 057 (A Jj^) (> poo A9ta>* m T*?/! 0061 (A *5C) I "t: ' {mtott>T3 t*JV 09UJ 4q ptvtos m.otfi * W toP/l 005 ,, i-- - t ------ . OMp 9% tpui y/i ,9HTd AQ#owx9 % UQT*d-- Nff | <AA SOI 09 09 .0.005 -t*-1f *16 ,* fc 4q nm au; ui inRwr'i mo 6un*og I T uiH 004218 I --v *-- s. unrealtitle to expact too ouch fro* It bocouoo of the oiti-- difficulty of iMpUof truck leads, in *ltk the or* portlclo* rang* in tiro fro* several foot to dust sod *ud. Fosslbly this largo voltan of rock nould contain on tho order of 10 per cent vomicullt*. In other *ords poshes Mo thirds as auch voxaiculita (500l/day) Is discorded hor* os Is asrkotod (7)0Vdy). Secondly thoro Is tho aodost tonnage discorded st the transfer point. Ihls *as oboorood to be extroasly high on one occasion often hii grade (C) are *as being transferred, end to be aoderstoly high Sas lo trade or* (ft) one transferred. My lapression (Mich nsy be faulty If freaen ore *as abundant on tho occasion of ay t*o Inspections of the transfer point) Is that tho discord at this point Is net auch loner In grade than tho food accepted by the ailli I suppose It aey contain 30* Vn, or 190 tone of uemicuilte per day. Tinting the 1790 tons of tho various Bill discards et 3B$ ft suggests e veralcullto loss in tailings of 390 tons par day. In son the oara&eullta distribution *ould so** to b# of the follonlng ordert In concontrstos In rock oasts In coarse discard In mil tailings and mil loses* 790 900 tonei'day 190 390 per day 03636884 Total loses* 1,000 ton*/day Sines the concentrate* or* only about 851 pure the net vernlcuUte recovery roca-fac* to shipaants is of the order of 6)0 tons out of 1690 or about 40 per cont. Adnlttdly these estimates are crude. Soar iaprovonent In our inform nation Is cortelnly obtainable for nost of the alll losses, but nine-lest information will be difficultly obtainable. Furthermore this would be rather ecademc with the pro tently used concentration processes. The concentration process outlined by Figure 2 is extraordinary in th*. there is practically no breaking, crushing, or grinding of or*. This procedure ** adopted to prevent the degradation in site of the fragile 20199748 004219 10- vermleullte. Qt the other hend since tht separation processes ere not very positive auch ruining arousd of Interned late products is necessary end appre ciates degradation does occur, even if it has tone unmeeswed. this is suggested by the fact that Jig concentrate particles are frequently ragged at the edges Utile the particles of the sane else in the final product are often rounded. One of the objections to the present floesheet is that locked particles of ehlch the constituents are aoderately tightly bonded to each other ronsln locked, hence lost In tellings, or e source of off-quality concentrates. Another of the objections is that the nest effective gravity concentration nethod for coarse particles (heavy nsdla separation} la net In actual plant use. 2t nay be that the Jigs actually used art the bast typs af Jig, utilising advantageously tho fist shape of the veralculite. tut it oeene aero then likely to no that a careful nlll conperlson of heavy nadla ispsrstlsn with Jigs aould shea the foroor to be definitely superior. The efterlaanta non In course should certainly be continued. ft. J. . Calkins has node s pilot-mill conperlson betee on Jigging and heavy-nodla separation, using a ssall cone-shaped separator. He found a clear advantage far the heavy nodIs operation. The larger-scale pilot non oacaAmc located in the sill, Utich uses e dnn Inateed of a cone as separating vessel dees net shoe as such advantage, the thick books af veralculite, especially, being lest in the tailing. I watched that plant in operation, ft. Calkins believes its ^satisfactory performance nay be due te the lee range In gravity of the suepension in the dnn (between batten and tap) at conpared elth the range obtained In the cone. Possibly thet is pertly the reeson for the peer current results, although I suspect other factors nay also be Involved, e.g. the fact that In the drta the heavies" are drain through a layer of 'lights* possibly entrapping son* cf the letter. Also, the ere eav have been different, with larger proportion of the thick-book veralculite Uten the dna plant oat used. Cooparisen of the drta with the cone mo* an order) Is clearly desirable. The cooparlsan should be node so as te eliaenate the ore and the nedltas density as variables. I suggested to Mr. Catkins to test the series use of heavy-aedla drras. Ue made a short inconclusive run. The matter Is worthy of additional study. 20l^7O> 004220 n Z have not liked the dry Bill* This aay be aerely in expression of ay sen dlelLte of the duetinea* of dry allle. In this cate, however, conditions seen so bee that I aould rate the operation a teaptatlon to lta operators to stay away froa the aachines entrusted to thea. It nay even be a serious health haxard. The dust losses and heaping plies of concentrate spill are witness of the inability to practice good housekeeping without excessive cost, --or aeybe even a&th excessive cost. Finally, the aultlpllcity of units justifies the desire to sees a siller and aare direct processing plant. Aunmnvi coeumuTiie mass In Pigive 3 I have sketched a flowsheet rfilch I believe capable of et*- stantlally increasing the production of veralcullte froa the taaaoge of ore that Is nee processed, k auaber of points raised by this proposal require study, these are the object of the next section In this report. Ihis flowsheet Involves the following features! 03638886 1)lesser discard to the alas duap, 2) insertion of a crushing plant, thus eiiaoneting the present discard at the transfer point, 3) all-wet concentrating alll with no segregation of ere types, 4) discard of eliaee by screen-site control rather than by close- lflcatlon, 3) heavy asdla separation of the coarse sices (say about b - atah), b) aggloaeratlon concentration of the 6 to 33-aoah ore, possibly in two sire cuts, 7) wet site-shape cleaning of the heavy-wedla concentrate, 8) net roll crushing of the aiddllng froa a Ire-ahape separation, 9) dry site grading for market sites. It is hoped that awen a flowsheet should be able to produce about 1100 tons of net veralcullte (In about 1200 tons of concentrate) froa the seat 8000 tons of ore instead of about 630 tons of net veralcullte (in about 7)0 ton* of concentrate). These figure* at present are estimates based on opinion rather than fact, and although I believe then to be real liable, they oust of necessity be checked by auch experlaental work yet to be done. l^e nine features of the proposed flow sheet will now be examined in a :tile oore detail. 004221 / 1200 t/dar at 90-95* V FlCUtS ) Altamatlva FlonAMt for Libby Ora 1^v 1 \V> 004222 13- JIlatjfjaB Much but not oil that non 900* ovor tho wine diap la eoarao 0363881 syenite. Csrtalnly syenite should continue to 90 there, but neither lo-- 9rade ore nor availed hlotlto ore* should be routed there. The *blotlteB, especially, dwuld be uatched as 1 fear auch very hlgfygrade ore Is Inadvert ently so labeled In the field. The coanentt In tne first section of this report are appropriate In this respect. Cruahlna Plant The crushing plant required uould consist of a j-- crusher of large gape, followed by a gyratory or cons. Careful arising of the capital and operating cost of such a plant against the loss la the present transfer-point discard and labor to roaovo the discard Is clearly In order, and It nay bo that ay present hunch that the crushing plant uould pay for Itself Is urong. All-uet ulll eith no are toereoatlon This Is Intended to slapllfy the present aultlplldty of ores and plants. It speaks for Itself. dll-- discard because of the flaky character of the veraicullte and of Its lesser gravity, classifier control of the discard site Is aanlfestly the urong way of going about this operating step. Use of fine uet screens on large scale is unusual. It nay require cooperation with an equip--nt --nufecturer for so-- dowelapneat uorfc for a suitable screen operating eith the screening surface --dor eater. Heavy--dla seecrayon of coarse alias This topic has already been discussed in the preceding section. It --y be pertinent to add that there is no reason --y two or --re heavy --dla separating cells carnet be used in series in the aa-- fashion that several flotation cells or Jigs are used In series to gradually deplete the ore pulp of the --sired constituent. Alternatively tne heavy---dla separations at different gravities --y be helpful in producing a final tailing, a concentrate and a elddllng. A--to--ration concentration of lntereedlate sizes Expert--nu -- at Libby several years ago show that extraordinarily 90-- results are realltable. The reagent cost has been thought prohibitively high, but 1 -- not believe that the subject has been appropriately explored. 1 believe experl--nta with an anionic collector such as a fatty acid (oleic acid) --st not bo owltted. date expert--nts sees to have been aade with cationic reagents In rather < > ..sort--nt, but without probing Into the relationships of ulneral, and aqueous solution. I believe considerable reagent econouy is 004223 -14- pcsslble am riMt has been obtained to date, mough in feet to Justify i different conclusion in ngard to um of oggloaorstion at Llidy. heavy --dla concentrate The hoavy oodla concentrate would consist of ouch fm verulcullte, sou* free asbestos, tone locaed partlclos of wral- cullte, asbestos and pyrossno, and occasional quarts, pyroaono or syonlto partlclos. SIso-stupa saparatlon la oator should wort out about as uall as la air, and so should roll crushing. Ho groat difficulty Is thorsfora anti- dpatod in adapting thasa stops at this staga. If tha nothod of ana11lag tha verulcullte by osnotlc prsssiua does actually anil each particle rather than exfoliate thlct boobs Into thin penphlets, a second haavy>nodl(a saparatlon at lor gravity nay do a batter Jab than sl*^shape separation. In that case sealling foiland by a second loner gravity hoavy nodi a separation uould taka the place of ols^ ehape separation. In that case also, roll crushing could perhaps bo replaced by red-sdllgrlading or also tha heavies free tha second hoavy oodla separation could go to uaste directly. at roll crushliin There Is no problen hare. Drv sire oridInn for partst Thera is no problen hare either. The unusual properties of vermlcullte, etilch to data are only partly mderstood suggest tha desirability of nsny researches, so oany In fact that sons choice wet be node as to out should be done first. Aaong the ere lnportant areas of study, 1 mid suggest the following! 1) Increase both absolutely and relatively the effort that Is presently being aade to understand the alneral veralcullte, end In that connection iaprove the dissemination of the fundeaentel knowledge so acquired aaong these charged lth the nore util Iter ten phases of your wors. 2) Iaprove the anoaledge presently available of alne and mill losses. 3) ti>e a thorough evaluation of heav^aedla separation in ail its ranificetiont including pre-swelled ore. 4) Hake a thorough evaluation of agglomeration especially directed at a reduction In reagent cost. ;) Study under-water screening as applied to Libby ore. t) Study the artificial veralculltlzatlon of the acre blotitlc varieties 004224 15- f Mm Ubby riaml with a vin to loproving tho quality of tho food to expanding plants* 7) Further study ths blotlu aspect of tho art, ss by systsaatlc petssaiw analysis* Vary truly yours, sssv's; hi M* Gsudln I *15?54 004225