Occurrence of selenium in sulfides from some sedimentary rocks of the western United States

Coleman, Robert G., Delevaux, Maryse · 1956 · Report

Investigations of the minor-and trace-element content of sulfides associated with uranium ore deposits from sandstone-type deposits have shown that selenium

This 1956 document, Occurrence of selenium in sulfides from some sedimentary rocks of the western United States, is preserved in the Mountain Man Mining Library for research and reference. Original source: pubs.usgs.gov.

OCCU~RENCE OF SELENIUM IN SULFIDES FROM SOME SEDIMENTA~Y ~OCKS OF THE WESTEEN UNITED STATES By R. G. Coleman and Maryse Delevaux Trace .Elements Investigations Report UNITED STATES DEPAR~T OF THE INTERIOR GEOLOGICAL SURVEY

Geology and Mineralogy UNITED STATES DEPARTMENT OF THE INTERIOR GEOLOGICAL SURVEY OCCURRENCE OF SELENIUM IN SULFIDES FROM SOME SEDil1ENTARY ROCKS OF THE WESTERN UNITED STATES* By Ro G. Coleman and Maryse Delevaux November 1956 Trace Elements Investigations Report 632 This preliminary report is distributed without editorial and technical review for conformity with official standards and nomenclature. It is not for public inspection or guotationo {~This report concerns work done on behalf of the Division of Raw Materials of the Uo So Atomic Energy Commission.

Usgs - Tei-632 ·Geology .And Mi~Ogy

.No.. of copies Division of Raw Materials, Albuquerque o Divis ion o.:f Ra:w .Materials , Den.ve:r .. . ., .-. Divis ion of Raw lw.terials , Ishpeming Di,vision o.f Raw Materials, Salt Lake City Division of Ra.v .Ms.t~rials, Washington $' .

Exploration Division, Grand Junction Operations O:fflce 6 Grand Junction Operations Office ,. , ., .,. ... ·1 Technical Information ,Service Exten.sio.n, Oak Ridge ... u. S.. Bureau of Mines, Washington. (Branch of .Rare and Precioua u . ., s. Geologtea.l Survey: Geophysics Branch, Washington . -. Mineral Deposits Branch, Washington R3 L. Griggs, .Albuquerque

E. Vleissenborn,. Spokane ..

CONTENTS Pa:ge () -· 0. 0 e. 0'40,, 0 (l -· P:roeedure for the determination of selenium in sulfides o o ., -- Ge_ologie and geographic distribution of the analyzed sulfides Sulfides from the Morrison formation, Entrada sandstone, and Catskill formation Q . 0 0 0 -·." G 0 & Sulfides from sedimentary rocks of Tertiary age "Q ,... ., e - 19 Sulfides from rocks of Cretaceous age oo .,o!) 20 Sulfides from hydrothermal and igneous rocks Mechanism of substitution, selenium .for sulfur ., .,.o .. · 21 Oxidation of seleni.ferous sulfides and selenides -o-, o o o t~· .. .,. 24 Discussion e 0 . tt 0 - 0 n .. 0 0 Q 0 ... O· 0 0

· - . ILLUSTRATIONS Page Figure 1,. Locations of samples Coalified log showing typical sulfide concentration Pyrite associated with uranini te and coffinite pyrite associated with uraniniteo J~polished section 5. Roll ore body and associated sulfides, Morrison Gale:naclausthalite band and its relationship to vanadium deposit in Entrada sands tone o. o. 4:; 7. Ga.lena ... clausthali te band. Polished sex~tion ·o. o o .. e 46 Pyrite impregnating and replacing wood structure .. Lqngi tudinal polished section . ., o o o .,,. 41 9. Pyrite iinJ?regnating and replacing wood. structure. Transverse polished section

figure 10. Page Chalcocite nodule c--ontaining intergrowths of claustha.li te 11., Chalcocite and digenite containing small inclusions of clausthalite. Polished s,ection

121.. and eucairi te in central part of chalcocite nodule. Polished section o &.·

13. Euhedral pyrite cubes in barren .. silts;ton,e. Polished section ooo 52 14'. Mineralized log showing wood structur'e with the cell walls replaced by marcasite arid th~ lumens impregnated by pyrite-marcasite. Polished section 53 15. . Stability field of selenium after Delahay, Pourbaix ,' and TABLES Page Table l. Selenium content of sulfides from sedimentary rocks of Selenium content of sulfides from the Chinle formation of SeleniUJI~. content of sulfides from sedimentary rocks of 4. Selenium content of sulfides from sedimentary rocks o:f 5. Selenium content of sulfides from hydrothermal vein

OCCUMENCE OF SELENIUM IN. SULFIDES FRQ~ SOME SEDIMENTAJtY ROCKS OF THE WESTERN UNITED STATES By R. G. Coleman and Maryse Delevaux ABSTEACT Investigations of the minor= and trace-element content of sulfides assoc.iated with urariium ore deposits from sand.stone .. type deposits have shown that selenium commonly substitutes for sulfur. The Morrison formation and Entrada sandstone of Jurassic age and the Wind River formation of Eo·cene age seem to be seleniferous stratigraphic zones; sulfides deposited within these formations generally contain abnormal amounts of selenium. The selenium content of the pyrite, marcasite, andchalcocite is much greater than that reported in :previously published data. Under the prevailing temperatures and pressures of formation of the Colorado Plateau uranium deposits the maximum amount of Se substituting for S in the pyrite structu.re was found to be 3 percent by weight. Ferroselite, the iran selenide (FeSe2 ), wa,s found in two deposits on the Colorado Plateau, and it was also establishe9,. that galena (PbS) forms an isomorphous series with clausth9tlite (PbSe) in nature. During oxidation of the selenium-bearing sulfides and selenides in·_>:.;' the Colorado Plateau and Wyoming, the selenium forms pinkish cru_st.s of either monoclinic or hexagonal native selenium intergrown :with soluble sulfates, sugge.sting that under "normal" oxidizing conditions native selenium is more stable than seleni tes or selenates. The above=normal selenium content of these sulfid.es from sedimentary rocks of Mesozoic and Tertiary age is significant. The high selenium in

these sulfides i.s relate-d to per.iods of voleani'·c a:nQ. intrusive a.cti vity penecontemporaneous· with the formation of the containing sediments. INTRODUCTION Mineralogi studies of' the Colorado .Platea,u uranilllU deposits, by the u. s. Geological Survey on behalf of the D.ivisio·n of Ra.w Materials of the u. S., Atomic Energy Commission, reveal a rather 'Persistent associ ... ation of selenium with these uranium deposits.. The unique geochemicai character and restricted occurrence of selenium make it an indicator element that can be used as a tool for interpretation of'. the process.es that have given rise to these ore deposits. The first mention of the association of selenium with these ore deposits was given by Beath (2), who ha.S shown that selenium-bearing pla.nt.s are consis:tently present where the Morrison forlnation of Jurassic age crops out. Gannon (8,9,10) in more recent studies has used selenium-qearing plants as prospecting guides in the location or uranium deposits on the Colorado Plateau. Shoemaker and others (written communication), in their minor .. element study of several hundred mill pUlpS from pr.oducing uranium mines, show that selenium. ha.s 'been conce·ntrated in the o-res from sandstone-type uranium d.eposi ts The purpose of thl.S investigation was to determine the nature and amount ·ot selenium within the ore deposits and to establish its distribution with respect to the ore bodies and their host stratigraphic units. To do this the sulfide .minerals were chosen as the best possible means of d.ete,et ... ing selenium present in the ores. Goldschmidt and Hefter (19) have shown that sulfur and selenium are very closely related in ionic size and tha.t very frequently selenium substitutes for sulfur ln sulf:id.e minerals.

Pyrite is almost universally associated with the unoxidized ore bodies onvc.:th~ '-'; .: Colorado .Plateau and was chosen as the best mineral to test the distribution o-f selenium. Where pyrite v.as not found; galena 9 chalcocite, and. chalcopyrite were tested The sulfur/selenium ratio of sulfides has been used by several workers to distinguish between ore deposits of sedimentary or hydrothermal origin. Goldschmidt and Strock (20) have shown that pyrite :formed. during diagenesis of sedimentary rocks has a sjse ratio of 200,000 or more, and pyrite of" hydrothermal origin has a sjse ratio of 10,000 to 20,000,. Carstens (7) found that sedimentary pyrite from Norway contained less than 1 ppm Se as compared to 20 to 30 ppm in pyrite from hydrothermal deposits. A more recent comprehensive study :by Edwards a.nd Carlos (15) on the selenium content of Australian sulfides shows a similar genetic relationshiP They restrict the use of the S/Se rat:ip to those provinces that generally ba.ve a low selenium background and have picked pyrite or marcasite or both as the mast satisfactory ·index sulfides. and Carlos assume that when the S/Se ratio is equal to or less than 10 9000 the deposit containing . pyrite and/or marcasite :may be of hydrothermal origin Applying this concept to certain ore de_posi ts asso'ciated with pa.rti'cularly pe.rsistent stratigraphic horizons, Edw~ds and Carlos infer that these deposits. are hydrothermal and not of sedimentary origin on the basis of' the S/Se ratio of the sulfides f:ram these deposits~. This interesting relationship may be valid f ·or the .Australian deposits if it can be shown that all o:f the selenium was derived from mineralizing solutions originating at depth, but Edwards and Carlos do not indicate ¥that the B/Se ratio may have been in pre ore (diagenetic) sulfides front the host stratigraphie units. These inferences regarding origin; based on S/Se ratios must b.e used with great ·care unless one knows the · selenium

eontent of the host rock before mineralizati-on For· instance, the ur.aniu.mde_pqsi ts o.f the Colorado Plateau region of the -we.stern United States occur_in pers.istent sandstone units that a.re known to _have a.n exceptionally high background. in .selenium a.s pointed out by Beath ( 2) and Trelease _and Beath (32); therefore, one woUld e.xp;ect sulfides forming during diagenesis or later in such a host rock to contain .Se indigenous to the rock ana here it would be dif;ficult to determine the source of the selenium.-- that i.s, sedimentary or hydrothermal. METHODS OF~ MINERAL S-EPARATION To insure pure s.amples for selenium determinations, special precautions were taken using several unuSual separatory te-chniques_.. The samples -containing eulf'ide -were crushed_ fine enough to release the sulfide rrom the gangue :ma;ter_ial and then e-lutriated to remove the clay--size fraction (little or no sulfide was lost in the elutriate). The sulfide -wa.s then separated from the gangue by flotation using a .Mayeda cell (22). The advantage of the Mayeda. eell is the small quantity of material that can be usep... Bat is ... factory sepa.rati.ons were obtained on lQ ... gram samples containing less than 1 gram of sulfide The sulfide concentrates obtained from flotati.on. we-re then centrifuged in bromoform to remove the lighter gangue,. For pyrite and ·marc-asite fU.rther purification of the concentrate -was effected by treatment ·with warn-concentrated hydrochloric-acid to remove oxidation products and carbon.a.tes; when silicates remained, the concentrate was .further treated with warm: hydrofluoric aeid.. Tests have shovn that pyrite a.n.d -marca.si te are u:riharm.ed by warm hydrochloric and hydrofluori-c acid (1, 21); however 1 the other sulfides analyzed~ such as galena,. chal·coci te, and

chalcopyrite vre~e not acid treated.. The superpanne,r -was used to effect ;f'u,:rth,er purification with these sulfides. .Almost all .of -the sa;mples analyzed were at least 95 per-cent pure. X ... ra.y and spe-ctrographic determ.ina.tions on each sample were made to determine what impurities remained, if any., Quartz a:nci barite were the most common in1purities, and. it was felt that thes-e minerals would not alter the ultimate selenium determi~ nations . PROCEDURE . FOR THE DETERMINATION' OF SELENIUM IN SULFIDES The method for the determination of selenium is based on the widely u,se.d distillation of selenium tetrabromide, particularly as described by Robinson .and others (27). A l ... gram sample of the sulfide is transferred to the distillation flask and is decomposed with nitric acia. The selenium is .distilled with a solution of hydrobromic acid and bromine,. The distillate is treated. with a slow stream of sulfur dioxide to reduce the free bromine to hydrogen bromide , and hydroxylamine hydrochloride is ·~dded to :pre:cipi tate the selenium in its elemental form. · The selenium is then determined -calorimetrically or gravimetrically. For the colorimetric estimation 1 an aliquot of the distillate cont.a.tn ...

ing 3 to· 100 micrograms oi' selenium -~s us.ed,_ and estimation. is .made· YiSu&lly aga,inst a series o:f standards containing 5. 10, 20, . .30, 4-0, 50, 75, and 100 micrograms of selenium. For samples containing .more than 0 2 mg o:f se1ei1.ium, the precipitated .selenium is filtered, dried, and then weighed on a semi~micro . balance. The chemical analyses were made by the junior author.

GEO:LQGIC AND GEOG!Vq?HIC DISWIBUTION OF ·THE SVLFIDES Most of the su,lftde sample.s studied were taken from uranium and u.ranium~va.nadium ore deposits in sed.ime:ntary ;r-ocks . The .Colorado :J?lateau region vas sampled exte;nsively and nere most of the uranium deposits are within several -<:pe~sistent stratigraphic units; the .Morrison for.matiqn of Jurassic age,. Entrada sandStone of Jurassic age, Tod.ilto limesto~e of J~.assic age, and C.hinge_ formatio:t;L of Triassic age,. Outside the 9olor~d0: )?.lateau, samples :were tested from uranium ore deposits a.nd associated sedime.nts -within the Tertiary seQ.imenta.ry rocks o.f Wyoming, particularly the }lasatch and Wind River fortna.tions of Eoce.ne age. In the Black Hills area of South Dakota and Wyoming, sulfides were analyzed from the .Fall R:L ver sandstone, Fuson shale, Minnewaste limestone, and L&kota sandstone,. all of Cretac~ous age.. Other samples from nonuranium-bearing sediments and igneous rocks from the western United States were also tested to establ,ish the general selenium background. The location. of these samples is shown in .figure 1 FORM AN0 OCCURaENCE QF THE SULFI!)ES Pyrite and marca::site are the most· abundant sulfides associated with·, the uranium deposits and commonly fo.rm fine dissem.inat.ions or nodular concretions . In most of the deposits pyrite was found to be more abundant than marcasite. Wher·e· sulfides have been precipitated within carbo~us : / material, they assume the structure o.f the woody material which they replace (fig 2). The sulfides have been divided into two t~es ac.wording to their position With. resp~ct . to the ~oxidized ~~nium.,.va;na.dium ore deposits -The relat~v;~ly

abundant sullides ta;ken from Within the ore or fo:l_low±ng 'th~ outline of the ore bodies are considered as ore sulfides (ro-ck containing> 0~01 percent UsOa) a The:re is no e:vidence tel suggest that the u_raniUlll deposits in sandstone-type-d.ep,osi ts are accompanied by a surrounding zone of sulfide~ a"Way from the or.e, as is commonly :e'ound in hydrothel"lllal vein-type deposits. The ubiquitous and .scattered-d sulfides in the unm.ineral.i.zed rock show ne relationship with the ore 'bodies and are considered J,/a;rren sulfides_.,_· It is assumeQ. that ·most of tne sulfides tn the barren rock formed independent of a;nd prior to the deposition of the uranilUll=vanadiuin ores either during diagenes-is o,r later from nanu.raniferous circulating intrastratal solutions. The sulfides ·within the ore are considered to have formed from the i'luid.s transporting the metals for the uranium deposits or by reconstitution of the pre=ore sulfides by these fluids or by both (figs. 3 a.nd 4). This division is arbitrary as there is difficulty in establishing criteria to dlsttnguish early su:lfides fro.m those· sulfides produced by the late·r bearing fluids. The presence of iron sulfides has been reco~ded in many drill cores of unoxidized barren ro-cks .from the Morrison and Chinle formations on the ! Colorado ,Plate~u anq. fro.m the River format~an in Wyoming._ Shawe· (29) has shown that two types of diageneti-c changes have taken place in the sedimentary rocks the Colorado :Plateau prior to uranium lnineralization: oxidation, producing hematite and. the characteristic red, colorati();n, 2. reduction, producing iron sulfides and the characteristic green ~olorationo. Thus it seems likely that iron sulfides must have formed i:n these sediments where- ·reducUg conditions existed before ore d.e:posi tion. Another line of suggesting two distinct periods of iron sulfide

deposition has been. found in the Ni ~Co ratio of these sulfides. Hegemann (21) and Tallur.i (30) have shown that pyrites formed by .sedimentary pro.messes generally have Ni >Co and those pyrites formed by hydrothermal pro .ce.ss~s generally ~aye Co Ni. Exceptions to-this general rule are common, but the senior author has found that the barren iron sulfides used in this study usually show Ni Co and those from the ore have Ni Co 'This illustrates that the divis.ion o-f the sulfides into barren and ore groups :reflects a. change in com:posi tion of the sulfides which probably results fro.m different environments of deposition~ Undoubtedly some oi' the sulfides :are placed in the wrong group be.cause the -collection of these samples -wa.s based primarily on their position as related to the ore bodies described above; however, this classif'ica.ti.on is convenient to illustrate the partition of the selenium in sulfides with respect to the uranium deposits. lJ'he chalcopyrite , . chalcocite , and galena depos i te.d in the uranium ores are probably formed by the ore fluids , as there is no geologic evidence to prove that they .are pre -ore. Pyrite and marcasl te seem to be the only early sulfides~ The most unusual characteristic of the sulfide deposition -nth respect to the ore is the thin sulfide bands that follow the outline .of the roll ore. bodies.. The-se roll ore bodies, de,scribed by f ·ischer (16), are layered depo:si ts that cut across the sandstone bedding in sharply curvi:ng forms (figs. 5 and 6)., The sulfides within the band hav:e ·been deposited interF' sti tial to the sand grains (fig. 7) and the band is q.s·ually characterized.. 'by a monomineralic sulfide either pyrite, chalcocite, or galena.- claustha.li te.o;_ The sulfides within the ·bands usually have extremely hi~ selenium .contents , as much as 18 percent.. Sh~we (28) believes !that these

sharp depositional features were formed at a static interface between fluids of different composition and density. The coalified wood -within the deposits seems to have controlled to .a l .arge extent the -precipitation of the ore and also the sulfides. Larger coalified wood fragments and logs contain abundant sulfide that may :eom= pletely repla:ce parts of the wood preserving the :cell structure (figs. 8 and 9) The pyrite also commonly forms an aureole around the log as fine disseminations in the containing sandstone (fig~ 2). The more prevalent tabular ore ·bodies contain sulfides disseminated throughout with marked concentrations at the boundaries of the bodies and. within the bodies small carbonaceous fragments show strong localization of sulfide., Selenides are common in the ores from the .Morrison formation and Entrada sandstone and extremely rare in the ores from uranium. deposits in ro-cks of Triassic age. Clausthali te (PbSe) is the typical selenide found, and in the vanadium. deposits from the Rille and Garfield mine., Garfield County, Colo.., a persistent sulfide band contains clausthalite:galena in variable amounts; the band can be traced for several thous~d feet along the strike of the ore deposit. An exceptional concentration of cla.usthalite was .found (Lee Eicher, U. s_. Geological Survey, personal communication) in the Corvusi te mine, Montrose County, Colo. Here a band of clausthali te a quarter of an inch wide formed a semicircular band around a large log (18 in in diameter and about 10 ft. long) representing the largest concentration of selenide yet found on the Colorado Plateau. An unusual oc-currence o:r clausthal:lte is found in chalcocite nodules from the Cougar mine, San Miguel County, Colo. Clausthali te is inter grown with chalcocite ( CuzS) and digeni te ( Gu~_,xS) (figs 10 and 11) , and in some nodules the rare selenide, eucairite (CuAgSe) (fig., 12) is associated with the chalcocite

and. cla.nsthali te. Ferroseli te (5}, the iran selenld.e FeSe~~h is the only selenide so far identified in Triassic depo.ait.s ., In the Triassic it has been fo:und only in the Chinle f ·ormatio_n in the Temple .)lo:u.ntain area, where it occurs in the AEC No. 8 mine, Emery County; Utah, associated With pyrite and coalified wood-. In the Morrison forma..ti9n ot Jurassic age ferroselit.e. has been found at the Virgin No,. 3 :mine, Montrose County, Co-lo., by lt, Sha:we of the u. S Geological S1.1.rvey. The sulfides in barren .rock from the Colorauo ,Plateau are :present within mudstones anQ. s.a.ndstones, usually as nodular masses, small euhedral crystals (fig. 13), or as concretions . In mudstones the pyrite and marca.si te usually .form larger crystals and tend. to be more eu.hedral) w.here.s in sandstone these sulfides are anhedral to subhedral filling the int:er ... sti tial areas.·*'· .Marcasite and pyrite see.m to be of eq-qal abundance in these rocks . :Barren coalified wood contains abundant sulfide and soine of the sulfide in mineralized coalified. wood fragments ld thin the o:re has formed before the deposition of the uranium ore; s.ome of it may have been reconstituted by the ore fluid.s or :may have been: unchanged by these processes o£ o:re de;pos itio.n (fig. 14) Sulfides .from the Tertiary uranium deposits in Wyoming do not exhibit the· unusual distribution found in the sulfld.es· from the Colorado .flateall:, The sulfides :from these_ Tertiary deposits are present as extremely fine disseminations in sandstone and in some areas are concentrated along ba:n:ds f"O:~;d -:by se~ond'ary enrichment.. The sulfides in barren rock form s:mall concretions within. mudstones and irregular nodular masses in the sandstone: Marca.si te an.d pyrite are the only sulfides yet found asssocia,ted with these Te.rtiary d.epos.i t ·s,; selenides have not been reported..

DISTRIBUTION OF SELENIUM IN THE .SlJLFIDES In the early stages of this investigation it became apparent that the sulfides deposited in the sedimentary rocks of the Colorado )?-lateau ancl Wyoming were exceedingly high in selenium. The re:eent survey of trace elements :in sulfides by Fleis-cher (17) shows tha;t the highest selenium recorded in 11g pyrites a.nd/or marcasites from all types <lf d.e_posits 'WaS 300 ppm. Berge:felt (4) provides an example of' a selenhl.ID.-rich province in the Skellefte district, Sweden; however, even the max.imum contents he records f'or the sulfides from this district are not comparable to those determined in the -present investigation. Sulfides fro:m the -mine, Skellefte district, show the following maximum selenium contents lea·d bismuth-antimony sulfides, 6.4 percent Se; galena, 1 .. 4 percent; chalcopyrite, 0.14 percent~ a.rsenopyri te, 0 .16 percent; .and pyrite 0 .03 percent. Derriks and Va.e.s (13) have found 19 percent Se in vaesi te (NiS2) and 11 percent Se in siegeni te from unoxidiz;ed uranium ores at Shinkolobwe, but they have not analyzed the associated pyrite for selenium. In contrast to these published. data, the sulfides analy-Zed. for this investigation from. the sedimentary rocks of the western United States show the fo·llowing rnaJiCimum selenium va.lues : 11pyri te /' 5 perc.ent; marcasite, 0 65 pe;r:cent; galena cla.u.stb.EI.lite, 18 percent; and chal-cocite, 5 per-cent,. The gontroJ. on the selenium content of' the sulfides from this provin:ce seems to he stratigraphic r?-ther than a fUnction of the type of ore depos1t.,. ... that is, hydrQthermal versus sedimentary origin, as suggestedby Edwards and Carlos ( 15) for the .Australian depo:s its The sulfides from the .Morrison formation consistently contain more selenium than those from the Chinl~ fp:riQ.ation, with the exception:q. of the Temple Mountain depo.sits,

Emery County, Uta.h. pyrite and/or marcajSite .trQm both the tra.J"ren mineralized rack from the Morrison conta.in much mo·re selenium than. the average aho:wn by Flei~ :cher (l T), whereas these two types of sulflde.s frqm the Chinle deposits have much less selenium. Sulfides from. the M()n-iso.n. formation., Entrada' aand$tone, .and .C~tskiU forma.tipn Thirty ... nin.e samples of marcasit.e or pyrite or bo~h were analyz.ed from the Morrison. fo~tion (table 1) The a.ver.a.ge selenium content of the iron sulfides bQtb from barren and minerali:zed rock is 0 17 percent. The averages discussed are a.ritb:metiGal and those samples containing .Se below the sensitivity of th.e chemictW. method (3 ppm for a; 1-gram s.an1ple) a:re averaged in as zero. ;, The average selenium .content of . 20 iron sulfides front mineralized rocks is 0.20 percent a.nd for 18 su.l;fide s from ba.rren ro:ek the average selen;i.Ull'i content is o J4 percent. One pyri te-marea.si te samp1~ ,contained. 5 percent Se, but this sa.t.nple was found to COJ+t~in cla.usthalite (PbSe) a.s an .impurity.. The selenium content of sulfides from mineralized )."'OCk ls not stgni.:N.ca;ntly higher tha.n that of sulfides from barren rock. It would s.eem tha;:t .. if most of tn~ :'barren pyrite is .;preore., sele.nium was apparently av1:l.l:i.a.ble in the same relative amounts in sedim~nts of the Morr~son formation during pre-or.e sulfide fol"'ll.Ettion a.s during the period o:r uranium ndneraliza.tion. Therefore, where ·reducing conditions existed B.n.d iron sulfides have been formed in the .Morrison .formation~ either ·before o:re deposition or in the pe.:ri.od.. of uranium ore deposition, se.lenin:m. has been i;n these sulfides., although the ore fluid$ may ·ha-ve been enric.hed in ~elen.ium as they apparently .baye . tx;aye.).ed. · l~~e,ra.l~y thr'c,:tush. t:P.e . f"a.r ~efl;t distances_.

Tb,e -cbitlco:clt.e bands .and. nodules .ehara.eteristi:c tQ.e Slick .Boek di.striet, San .Mlgue1 C!)untyll Colo~, are rest·ri-cted to t.he ore bodies an:i seem to have fo.rme.d during the. uranium .minerall~align.- Some .of these sulf1d,e coneentratiQ:ns have extremely high seleni'Ulll·e.intent tor copper sulfide~. aver~in;g 1 . .,22 percent Se for 11 samples, lfith a high of 4.93 percent Se, Stuq of polished,. sections or these ·copper SUlfides re'vea.ls Sl'IJ,all inclu.sj:ons o,f .clausthalite an.d intergro'Wths o;f ela.ust~lite and ' chaleacite ... d.ige:nite (flgs. 10 and 11)_. Rarely small conceni:a:a.tion.s of eueairite (fig;. 12) are present in these copper sulfides.. These inc1uslons of se'lenides account, in part, for the exception.IJ.:y high selenium :content; of the cha:lc.o{~i te, but much of the selenlum i .s probably .substituting tor ·s:sulfur in the chalcocite structure. The persi.stent 11galen.an ·band developed in the vanadium deposits within the Entrada. sandstone is composed of ga.1ena. clausthali te in soTid. solution (fi~. 6) This band is fairly un;Lform in c.omposition but pro-- gressive varia.ti.ons in selenium along the strike of the bEUld were 11h.e average selenium. content of nine san11?les o.f galenaclaustha.lite tram. these bands is li22 pe:reen:t with .a high of 18 ·percent (table 1). In contrast, pyrite and. marcasite deposited in late fractures spatially assaeiated 'With these va.nadium ·deposits cQn.ta;in only 0. ,02 percent se_. Sandstone collected by Harry- Klemic U. s. Geological .Survey,. fr:Qlli ~lle UllPer part of the Catskill formation, of Devonian age, Carbon CQ.u.nty, . Pennsylvania, also contains a thin clausthalite 'bald (about a quarter oi' an inch thick) in juxtaposition with ura.nil.lln minerals. Further YQrk is needed to dete.rndne the relationship of the£e unique bands to the ~.an.ium.

Sulfides from the. Chinle forma.ti.on The distribution of seleniu.m in sulfides from sedimentary rocks of Jur"a.ssic age i.s no.t reflected by the sulfides :from seilil:nentary rocks cd'' Triassic age~ These sulfides, exclusive of the Te:mple .Mountain uranium depositsS~ show a much lower seleniUm content than those from the Jurassic

Fif'tyc..five s.a.mples of pyrite or marcasite or both from sedi~ ments o.f the. Chinle formation of Triassi.c age have a.n average selenium content of o _OOl5 percent.. The average .seleniutn content of 27 iron sulfictes !rom mineralized. rocks is 0 0019 percent and of 28 iron sulfides !rotn. barren rock 0.0012 percent,. These averages do not in:clude samples from the Temple Mountain deposits. .A$aln, as with the sulfides from ro.cks of Jurassic age, there is no apparent difference in the selenil.lln :monte·nt of the barren sulfides as compared to the ore sulfides This supports the premise that tb,e same relative amount o:f selenium was available for pre-ore and ore periods of sulfide deposition. The uranium ore deposits in rocks ot Triassic age usually contain more sulfide than do the deposits in ro:eks o.f Jurassic age, but the former sulfides have not mobilized to form continuous bands but have been more thoroughly disseminated throughout the deposits. The associated copper sulfides (bornite, chalc,6pyrite, and chalcocite) analyzed for selenium show the same relative amounts of selenium as the iron sulfides (table 2). In contrast to the other deposits in rocks of Triassic age, the Temple MountaJn deposits have an unusual partition· of sele·nium. The sulfide concentrates from ore in this area contain the highest concentration of selenium yet found in the Colorado .Plateau, although the sulfides from barren .rock contain .the same re1a.t,i ve amounts of selenium .as the average.

found in other sulfides froli! rocks of Triassic age,. Eight sulfide co;n ... centrates from ore containing ferroselite (FeSe2) from the Temple Moun,tain deposits average 12 percent Se, whereas four sulfide concentrates frGm barren rock .average only Q.,oo4 percent selenium. Thus at Temple Mountain the sulfide miner~,ls in ·uranium ore are strongly enriched in selenium when compared to the pre-ore sulfides, suggesting that the ore-:for.ming fluids contained abundant selenium~ Sulfides :from sedimentary rocks of· Tertiary age Sulfides associated with the uranium deposits in Wyoming sediment~ry rocks of' Tertiary age have also been tested for comparison with the ColoradQ Plateau uranium deposits... The Wind Rive,r and Wasatch formations o.f Eocene age contain the important uranium deposits of this region. The Gas Hills area in the Wind River .Basin, the J?um.pkin Buttes area, and Crooks Gap area, all in Wyoming, are the three districts sampled. Fifty samples o·f pyrj.te or marcasite or hath from these three areas have an average selenium content of o.o4 percent (table 3). The iron sulfides from mineraliZed rock have an average selenium content of 0 ,~987 percent, a marked enrichment over the iron sulfides . .from barren rock which average 0.0015 percent selenium. Secondary enrichment of uranium has occurred near the ground-water table in the Gas Rills a;ret:t; and the pyrite. from this zone-conta.ins as much as 1.5 percent Se. The solutions forming the uranium ore deposits in these sediments were con:~ sid.erably enriched in selenium as compared to the .solutions responsible for the formation of sulfides before ore deposition_. Ma.rcasi te and pyrite are ,the only sulfid~s observed in the samples collected_ from the sedimentary rocks of Tertiary 1'age from Wyoming

One sample from the ,Maybell district, Colo (table _3, no. 203) also contains significant Se as does a sample from Karnes County, Tex. (table 3, no 202). Sulfides tram rocks of Cretaceous age .Sulfides from various sedimentary rocks of Cretaceous age in the western United States h,ave been tested for selenium, but only a few samples from any one formation were analyzed (table 4-). The average selenium content of 19 samples of pyrite or marcasite or both from these rocks is · 0 0058 percent Se. Six sulfides from uranium ore average 0_.005 percent Se, and 13 sulfides from barren rock average 0 .. 006 percent Se. As these samples are so widely spaced and from several different formations, it is not possible to evaluate the relationship of the selenium content in the sulfides from barren rock as compared 'With the sulf;i.des from ore;..bearing rock. The five s~mples of pyrite and marcasite from the Mancos shale of Cretaceous age were taken from a drill core located in the Slick ·Rock district, Colo., and these average 0.014 percent selenium. Trelease and Beath (32) hay~ shown that the Mancos shale supports a variety of seleni'Ulll""' bearing plants (and it .may be that the Se is derived from the oxid.i:Zed iron sulfides). The sulfides in the .Mancos occurred in a black shale containing no uranium minerals .. Sulfides from hydrothermal and igneous rocks Various sulfide samples :from hydrothermal d.epos its and intrusive rocks from the Colorado Plateau have also been tested for seleni'Ultl to determine

El. its spatial distribution with respect to the host rock a.nd its mode or formation (table 5) Seven copper and iron sulfides fromthe laccolithic intrusions within the Colorado ,Plateau (including the La Sal, Utes, and Carrizo intrusives) average o.oo8 percent selenium. Chalcopyrite from a diorite porphyry located in the La Sal Mountains, Utah, shows the m,a.ximum concentration of 0 016 percent Se. These contents, although much lower than many of those .found in sulfides from sedimentary rocks , indicate the presence of selenium in the magmas giving rise to the intrusive rocks., Four samples of pyrite from hydrothermal base-metal vein deposits o,f this region contain as much as 0.017 percent selenium, suggesting that selenium was present in the base-metal ore-forming fluids. Further testing is necessary to establish the regional and local trends in the sulf1d.es from tne hydrothermal deposits and the igneous rocks, but it seems that the Colorado Plateau area is a selenium-rich province. MECHANISM OF SUBSTITUTION, _SELENIUM FOR SULFUR The mineragraphic stuQ.y on these ~elenium ... bearing sulfides has shown that inclusions o:r selenides within the sulfides are generally lacking., except for the clausthalite found in the chalcocites from Slick .Rock district (figs. 10 and. 11)~ Therefore, it is assumed that fluids .from which thes& sulfide.s Y~ere deposited contained selenium, and that this selenium has entered the sulfur positions in the sulfide structu.re during crystallization. The bonding in sulfides is generally considered to be a combination of covalent, metallic, and ionic.. Both the atomic a:q.d ionic radii of sulfur and selenium a.re similar.

Selenium Sulfur Ionic radius (A) Valence Rad,ius Covalent radius (A.) Goldschmidt's generalit.ation (18) on isomorphous substitution shows that where ionic radii of two elenrents are. within 15 percent or each other the possibility of substitution is good. Both the ionic and covalent radii of S and Se fall well within the )_im,it suggested by Goldschmidt Other factors that -may influenC,e the isomorphous substitution, such as coordination number, polarization, and ionic potential are also similar in sulfur and selenium. Therefore, one may conclude, on the basis oi' crystal chemistry, that diadochic substitution of sulfur by selen:Lum may be accomplished With comparative ease., The pyrite containing large amounts of ·selenium (greater than 1 percent) shows some expansion of the lattice as might be expe:cted, as selenium is ( somewhat larger than sulfur. Synthetic FeSe.2 has a structure similar to that of rammelsbergite (Gu.nnar Kullerud, personal communication) and it would seem that -- complete so-lid solution between feS2 - FeSe2 is not pos:.oo sible on tlle basis o.f crystal chemistry. The temperatures and pressu,res of formation of the u:raniwn ore d.eposi ts f'rom the Colorado Plateau have been esti~ted to be less than 138° C and 800 atmospheres (11) and from this present study it appears that 3 percent by weight of selenium is the maxi.., mum amount that may .enter the pyrite structure at these p ... T conditions where excess Se .is available. Ferroselite (FeS:e2 ) has been found associate,d with selenian pyrite; therefore where the sulfur and selenium. are present in a reducing environment with Fe, about 4 percent (molecular) can enter the pyrite (FeS2 ) structure; and if excess iron and ·selenimn are

present, ferroselite (FeSe2 ) is formed. This relationship seems to hold for the sulfide-selenide formation in the low temperature and _pressure environment o.f the western uranium deposi ts , but at highe.r temperatures and. pressures a complete FeS2 FeSe2 series with the .. pyrite structure might be possible. The ferroselite (FeSe2) from t he AEC No. 8 mine, Temple Mountain, Utah, has a ratio of Fe:Co 4:1 and here the structure is identical with that of FeSe2 and also hastite (CoSea) described by Ramdohr and Schmitt (26) Thus a complete isomorphous series between FeSe2 - CoSea seems to be possible. Galena forms a complete isomorphous series between clausthalite as shown by Earley's work (14) on synthesis or ·this series-. Most of the series was found to form under natural conditions at the .Rille mine, Colorado, where the galen.a-clausthalite band contains various proportions of Se in solid solution with S,., These varying ratios of S and Se were reflected by an increase in the unit - ceil size with an increase of Se as pointed out by Earley. Bergenfe_lt (4) has shown that selenium apparently is enriched preferentially in different sulfides from the same mines. He has listed the sulfides in decreasing amounts of contained selenium: galena, chalcqpyrite, arsenopyrite, sphalerite, pyrite, and pyrrhotite. Edwards and Carlos ( 15) show that the Australian sulfides have a completely different sequence, listed here in decreasing amounts of contained selenium: chalcopyrite, arsenopyrite, pyrite (hydrothermal) , pyrrhotite, sphalerite, and galena :·. : .. . As shown earlier, the S and Se atomic and ionic radii are similar and variations in size are probably slight from one sulfide structure to the

next. Itwould seem that the amount of selenium available, crystal structure, .and P-T conditions all control the amount of selenium found in any particular sulfide from any one mine or district. This is v~rified in this investigation "Where the main controls on the amount of Selen_ium. found in sulfides a:re crystal structure and selenium supply. OXIDATION OF SELENIFERQUS SULFIDES SELENIDES The fate of the selenium during oxidation of the seleniferous sulfides and selenides is incomplete.ly known. Seleni tes and selenates have not been /-found near oxidized selenium~bearing sulfides or selenides, but native selenium has been identified from many localities by Thompson and coworkers (31). These observations agree with the physical~chemical differences between Se and S in the geochemical weathering cycle, as pointed out by Goldschimdt and Hefter (19), who have shown that seienates and seleni tes are stable only under extremely high oxidizing conditions. Sulfates on the other hand. are stable under the normal Eh;..pH conditions found in oxidizing sulfide ore deposits The pH and Eh equilibrium diagram as presented by Delahay and others (12) (fig. 15) illustrates the possible forms of selenium that might be expected under normal surface conditions. Krauskopf (24) has shown the limi~s o.f Eh and pH that are found at the surface under normal conditions and this field has been superimposed on the diagram. The diagram shows that metallic selenium is stable over a large area of this field both in acid and alkaline solutions.. Seleni tes may form at high oxidation potentials, particularly in a_lkaline SOlutions, Whereas selena tes might be eXJ?eCted only in extremely high oxidi.zing conditions as suggested by Goldschl:p.idt

The field obse va.tions support these assumption.$ 'Which are based on ealcutations TWo. i'ol'lllS of' native se leni run have been observed~ the sta:ble hexagonal form, and an unstable·e ·mo,noclinic foxm. The native selenium commonly crystallizes near or implanted on the oxidized Sul.fide grains . Pinkish crusts composed mostly of soluble su:}..fates ·contain intergrowths of' nat!· ve selenium and these crusts form near mine portals or where mine waters h ve evaporated. Selenium has been dete.ctecl in mine and ground waters, althl ugh the form of the selenium in these waters has not been established, trt is, SeOs . HSeOs-, or SeO.[-. Limonitic concretions formed during oxidation of seleniferous sulfides contain selenium, Jd Byers ;md coworkers (6) believe that the selenium may be present in the Jmon±te as a basic ferric selenite this selenite has not bJn isolated from the selen±ferous limonite. DISCUSSION Some o! the roi ks of Mesozoic and Tertiary age of the Colorado :Plateau and Wyoming are riclil in selenium, which is strongly concentrated in the sulfides formed beft re or during ore deposition" The presence or signifies.nt quantities of selenium in hydrothermal sulfides and in sulfides formed in the intrus. · ·l. ks o.f the region shows that this area is a selenium: .. rich province. An adequate h 1 othesi.s regarding the sources of' selenium ·concentrated in the sulfides Sho1d explain the d.ifi'erenees in selenium content of' the various geologic f't.tions, variations within a single formation, and th.e a.va.ila.bili ty of sel , nium during deposition of the sulfides_.,

/ The average selenium content of igneous rocks as given by Goldschmidt (18) is 0.000009 percent Se; sedimentary rocks exclusive of United States generally contain less than 0.0001 percent Se,. Selenium is concentrated in sulfides of magmatic and volcanic origin and in native sulfur of volcanic origin. Also, selenium can be concentrated in sediments derived from seleniferous source rock, such as that re-corded in Recent sediments in the Gulf of California (25). Therefore, three possible sources of selenium should be considered: selenium derived from volcanic activity, selenium from magmatic fluids, and selenium from pre=existing seleniferous terrains. The selenium in the Morrison formation E?eems to have been derived from both volcanic emanations and reworked volcanic debris. Waters and Granger (33) have shown that volcanic materials are common in the Mo-prison, and, as the sulfides from barren and mineralized rock in the Morrison contain the s~me relative amounts of selenium, it seems likely that the selenium was introduced into the Morriso~ during its deposition by attendant volcanism. No evidenc;:e was found to support the idea that the source material for the Morrison, other than the volcanics, was seleniferous. The Chinl~ formation of Triassic age also contains volcanic debris (33), and this seems to be the source o.f the Se in these rocks, but either the primary volcanic source for this material must have contained less selenium than the primary volcanic source of the Morrison debris the over-all volume of volcanic material was much less. The Temple Mountain deposits present a, different picture as there is a strong enrichment of selenium in the sulfides from ore when compared to the barren sulfides. The source of the selenium here seems to be hydrothermal fluids (23) whose magmatic source may have been rich in selenium, or these hydr_o"Ghermal

fluids may have abstracted the selenium from~ seleniferous beds through whi-ch they passed., The source of the .selenium within the sulfides from the sediments of Tertiary age in Wyoming is apparently tuff beds. Sc;i}"\lenium~bearing tuffs in the Wind .River Basi'n, :Wyoming, have been described by :Seath and. others ( 3) and they report a :m,aximum of 187 ppm ·Se. These . tuffs are either incorporate-d in or overlie the Wind Rive,r fonnation:.. Th_e enrichment of Se in the sulfides from coarse arkosic uraniferous . Sandstone When COttl.pare'd '. to the sulfides in nonuraniferous mudstones illustrates the fact that the selenium has been i'introduced into the Wind River formati.on by-downward or laterally percolating ground water charged with selenium derived from these tuffs.. The seleniferous sulfides were conc-entrated at or near the top of the ground'"'\ water table, whereas the sulfides in barren ro.ck below the ground .. wa.ter table were protected by impervious mudstones ana· show negligible selenium. The uranium fo:llows the s.ame depositional pattern as the selenium, and it would seem to have an origin related to that of the selenium; hence, the uranium and seleniUlli are. geochemica.llY coherent in the .Wind River formatfon.

. The presence of s·eleaium in .the base-met~l hy_~otltermal c;ires f"rom the Colorado _Plateau. regiqnind.icates that deposits. prod:u.eedby-hydrothermal fluids ma.y have deri ve:d their se.lenium: £rom igneous intrusions· enriched in selenium; or these. f'luidS may ha'Ve abst,racted selenium from seleniferous beds through which they passed. The ultimate sour:ce . of the selenium Jn the sulfides fro:m the :eo lorado ;plateau and .Wyoming can be related to a magmatic . provliioe . that ha·s COIJ.taine:d exceptionally high seleniUm. during periods of volcanie and intrusive ") .

a:cti'V-1 ty in-.Mesozoic· :a.-&1. T~ftiary- ?·ti.xrie~. ' . '

BIBLIOGRAPHY 1., .Allen, E. T. 7. Crenshaw, J., L-., and Johnston, J.,, 1912 7 The mineral sulphides of iron; with · crystall.ographic study by E. s. Larsen~ Am. Jour .. Sci , 4th ser_-., v., 33, 169-236. Beath, 0. A., 1943, Toxic vegetation growing on the Salt }lash sandstone member of the Morriso:p. forma'tion~ . .Am., Jour~ Botany, v. 30, Beath, A Hagner, A. F/k, and Gilbert, Cf). s., 1946, Some rocks and soils o.:r high selenium content: Wyoming ·aeol,._ Survey Bull.-. 36,--- P- 1=23-. 4.. Bergenfelt, Sven, 1953, Om forekomsten av selen i Ske11eftefaltets sulfid:malt.er: Geol. :roren., Stockholm Forh , v,., 75, p. 327-3'59~ Buryanova, Ee; Z. and Kotnkov, A I.,, 1955, .A new minera1-ferroselite: .Akad.,~ Nau.k S .. B.S,.R. Doklady, v. 105, 812 ... 813 (in Russian),. 6~. Byers, H. G., Miller, J._ T., Williams, K. Te, and Lakin, H.-. W., 1938, Selenium oc~urrenee in certain soils in the United States, with a discussion of related topics: Third Report: u. S. Dept Agr. Te.ch_. .Bull. 6o1, p- 1=74. 7 Carstens, C }f., 1941, Zur ·Geochemie einiger norw:egisehen Kie-svorkOll'.lln.en: K,. norske vidensk,- selsk. Forh , v. 14, p. 36.-.39 .. 8& Cannon, L .. 1952, The effect of uranium-vanadium deposits on the vegetation o-f the Colorado Plateau: Jijn~ Jour. Sci., v 2_50, 735-,.. 770 9. Carmon, .H. L-~, 1953, Geobotani.cal reconnaissance near Grants 1 New MexicO.: u. $;. :Geol.-.: Survey Circ 264, 8 / Oa.nno:n, H J., .. 1954-, Botanical methods of praspectlng far ura.n.:Ll.lln: Min-. Eng._, v. 6, p .. 217-220. 11. Goleman, R .. G., 1957, Mineralogical. evidence on the temperat~e ot~ formation o:f the Colorado Plateau uranium deposits: Eeon, . (k.o:logy, 12. Delahay, P.-., ,Po.u.rba:ix, M. , Van . Rys.seller ghe , P 195~, Diagrammea d ·~qUilibre potential...;pH de quelques elements: Comptes .rendus 3ieme r~union du C.I T.-.C,.E., p ,. 15-29,_ .Berne, (1951) .. 13. Derriks, J. J, and Vaes, J. F., 1956, The Shinkolabwe uranium deposit: Current status of our geological and metallogenic knowledge: .Pro.-c Internat~. Cont. , on the :peaceful uses of atomic energy, Geneva, 1955, v. 6, .P/1105, p. 94--.128, New York, United Nations ...

14. Earley, J. w. 1950~ Des·cription and synthesis-..s of the selenide minerals: 15., Edwards, A-. E., and. Carlos, G. C., 1954, The seleni.um content o:f. some .Austr.alian sulfide deposits: .Australasian Inst_., Min. Meta~l:urgy Proc., No. 172, p. 31-64. 16,. Fischer, .R. :P., 1942, Vauad.illll'l deposits of ColoradO and Utah: u. s Geol . Survey Bu.ll 936-l' 363·394 Fleischer, .Michael, 1955, Minor elements in some su.J.flde minerals : Econ . ., Geology, 50th An.ni versary volume , 1905-19'55, pt. 2·, 970-1024. Golds.cbmidt, v., M., 1954, Geochemistry, Oxford ·univ. Press, · 730 Goldschmidt, v M., and Hefter, 0., 19·33, Zur Geochemie des Selens: Gesell. Wiss. GOttingen, Nach:r,. , .:Math. -phys. Kl.l!., Heft 2, .Fachgruppe IV, no. 36, p. 245-252. 20.:. G.o1dschlnidt, v. MJJ 1 and Strock, L:e' W., 1935, Zur Geoehemie des Selena II: Gesell Wiss. Gottingen, Nachr ., Ma.th phys., Kl., Band 1, Fachgruppe IV, no,. 11ll p. 123-142. Hegemann1 F.,, 194-3, Die geochemische Bedeutung von Kobalt und Nickel int Pyrite: Zeitschr angew. Mineral., v. 4, p. 122 .. 239. Kerr, P. F t, 1950 ~ine~legli,;:c.als tudies of uranini te and uranini tebearing deposits; JU:LJt'cl''; ~;:19-49:,:.Juiie :~->O, r950: , , 'U. · S .. : ' A:t~atLC Ener-gy , COnnnt ;~:aM{)~~,]J~·:;- :,6l::·:po 23, Collaps.e features, Temple Mountain uranimn area, Utah, Annual Report for June 30, 1954 to April 1, 1955, .Part III: u.., S. .~tomic E.nergy Collliil:: EWE-3:110 (Pt 3), 8-138 i issued by the u. S . ., .Atomi.;c Energy C-arom Te·ch rn:r., Service Extension, Ridge. 24. Krauskopf, K. B~, 1955, Sedimentary deposits of rare metals: Econ. Geo.logy, 50th .Anniversary volume, 1905-1955, pt .. 1, 411-463. 25.. Lakin, H. ltf::., and. Byers, H. G., 1941, Selenium occurrence in certain soilS in the United States, with a discussion. of related tnplcs J sixth report: u. s. Dept. ,.Agr- 'Tech. :Bull. 783, p. 1-26. 26. Ramdohr, :P., a:nd Schmitt, .M., 1955, Vier neue Kob.a.l t~e1enide vom Steinbruch 'l'rogta.l hei Iautenthal im llarz: .Neues Jahrb . Min,. 2.7,~ Robinson, ii,. O,,. Dudley, H. c , Williams, K. T~, and.. Byers, H-. G~, 1934, Determination of·_ selenium and ar.seni~ by distillation: Indus and Eng., Chemistry, Anal,., Ed,. , ,,.., 6, P. 274-276.,

28 .1~, .Snawe, D_. :ft.,, 19;56, .SigQ:ifieance of roll O':te bod:ies in -genesiS of' deposits on. the Colorado' Plate;au.: Proc Internat. Conf. on the pea.eef'ul us,es o;f atomic ·energy, Gen.eva 1 1955, v 6, 335·331;. u. s Geol. Sl.l.rvey Prof Paper 300, p.:. 239 ,2~14! 29., Sha.we, D.- .a,., 1956, 1\,lterati.on related to Colorado ,Plateau ore dep-OSits (a:ost..,): Geol,. SO'C .Am.eri·ca,: :Pr.o~am 1959 ~nua.l meeting, p., 86 30 Talluri, .4n,na., 1951, Dosatura spettrografica dell 'arsenieo ln pirlti italia.n.e·: .so.e. ~osca.m Sci., nat. ,Atti Me.m , Y:.. 5a, 3·19. 'Thompson,. Mw. E , Roach, Carl, and. :Bradd.o'ck> William, 1956, New occtP:""reno.e of native selenium: .Am.. Mineralogist, v-. 41; 156 157~ Trelease, s. F .. .,, a.n.d J3eath, o. A , 1949, Selenium; its geological occurrence and it-s biolo.gical effe:cts in relation to 'botany, :chemistry, agriculture, nutrition, and. medicineJ 292 p,_., ElJ.rll:n,e;tQn., Vt,._, The Champlatn Printers. 33· Wa.t:ers., A. c , and. Granger, II., c , 1953, Volcanic debris in uraniferous sandstones anci it.s possible bearing on the Qrigin and :preei,pitati.on , Of uranium: u. s. Geo·l. Survey Circ., 224, 1 .. 26.

~ble l content o:r sulfides rrom· sedimentary· roc:~s qf J~ssi:c age., Location ]j Mineral g) U:r.av-an district, CQu,nty, Colorad(:) 1., North Star mine 2., Virgin No. Long .Park No;.- 1 mine do,. pyrite pyrite pyrite pyrite Gypsum. Valley diatrict, 5aa .M;igue.l County, Cploradq pyrite Mon~cello distric-t, San Jm,m County, Utah 6:. :Basin No 1 mine, 7.. Cott-onwood No,- 3 mine Found claim pyrite :pyrite pyrite Thompsons district, Grandr County, Utah 9 Little Eva mine 10. Blackstone No. 6 mine pyrite=marcasite pyrite Bull Canyon district, ~ontrose County, Colorado 11.. Mineral Joe mine do. do. Mineral Joe No 2 mine 15.-. J. J. mine 16., do. do. do. · do. 21~ do. 22., do. dO;.. do-. pyrite-marcasite pyrite-marcasite pyrite.:;:rm;trcasite pyrite marcasite-pyrite pyrite ... mar~site pyrite-marcasite marcasite-pyrite marcasite marcasite~pyrite marcasite~pyrite

m.arcasite ... pyrite

:s :B B B B B B B B B B B o Q.l4 o;.oo1 Q.l9 0"0003 0.074. ' 0.,39 O.,JQ 0,200.55 01!087 o.oo3

Table 1 . --'Seleni~ eontent of sulfid.e.s from sedimentary rocks o:f Jurassic age--Continued. Locat.ion y Mineral g) ,MORRISON :FO~l'ION (cont.) Slick Rock dist.rict, San Miguel County, Colorado Grant claim King No. 4 mine Tailholt mine do. do ,. 31. Drill core 700 ft depth, u.s. V,, :Burro Canyon 32 Mucho .Grande mine Cougar mine do. do,. 36-jf: do. do. do. do. 4o.. do . pyrite pyrite pyrite (marcasite) ''chalcocite" pyrite B "pyrite-marcasite'' "chalcoci te-n chalcoci.te "chalcoci te-digeni te u 0 uchalcoci te" "ehalcoci te" "chalcocite'' chalcocite 'rchalcoci te" "chalcocite n chalcocite Grants district, Valencia. County, New Mexico 42 Woodrow Pipe mine do. 44:. 45 Poison Canyon mine pyrite marcasite-pyrite pyrite "pyrite" Henry Mountains district, Qa.rfi.eld County, Utah Walter 's cla.im 47 Ellen No -. 2 mine, North Wash do . pyrite

pyrite-marcasite Cortez district, Montezuma County, Colorado lt9 Drill core , McElmo Creek pyrite-marcasite Crook County ar~a, Wyoming 50 Shannon Oil Company claims pyrite B B B o.,66 o.o15 o.o1B 5·00 o.2 o-.42 o.ooo3 0,.000-3 l.oo 0.0003 ' o.1o o.ooo8 0.0003

; Tabl'e i. --'Selenium content of sulfides from sedimentary rocks of JW"'ass age= -don tinued Location y Mineral§./ Type 2./ ;Percent Se J:} ENTRADA SANDSTONE Rifle district, Gar£1eld. County, Colorado 51., Garfield mine · do .. 55. Rifle mine do. 59 do~ 61 do. marcasite galena-·c).austhali te galena-claustha.lite galena-clausthalite claustha1ite-galena pyrite galena-clausthal.i te clauathalite-galena marcasite clausthalite-galena clausthalite-galena. J?lacerville district, Montrose County, Colorado 62. Bear Creek mine 6.3. Fall Creek mine TODILTO LIMBSTONE galena-clausthalite pyrite Grants district, Vale.ncia County, New .Mexico 64 .F. 0.,. ~nol mine 65 Crackpot mine pyrite pyrite B B y General geographic location of samples show iii. figure 1. · o.o14 o.-41 o.oz6 2/ X-ray identify'ication on analyzed material. Quotat:;l.on marks in.dicate impure sample; :mixtures are hyphenated with dom.inant phase listed first, tnarc.asi te-pyri te; (parentheses) ind.ic.a.te trace amounts. 3/ Sulfldes divided into two types: 0 sulfides taken :from mineralized ro-ck. Tusually showing greater than o~,Ql percent U3 08 ), and B sulfld.e.s taken from barren rock. · !/ · Lower linli t of detection is 0.0003 percent Se f'or a 1-gram sa;mple; when les.s sample is; available the lower limit o.f detection is correspondingly higher.

Table 2., $eleni-qm content of sulfides from the Chinle farma.ti.oh of Triassic age .Lac a tiQn Mineral gj San Rafael district, Emery County, Utah 66 75·" so. 81·, 84·. 86~ Lucky Strike n:dne do,-., Adams Ur~nium Company mine do. do. Hidden Splendor mine do do., do. do . ., do. do. do .. Donna B. claims Flop Over mine, Temple Mountain AEC No. 8 mine, Temple .Mp:untain do. do-.. de. do. do.. North Mesa. No. 9 mine, Temple. ,·Mountain do. ma:rca.s i te E.arcasite-py:title

pyrite pyrite galena

pyrite pyrite pyrite · pyrite marcasite (pyri t.e ) pyrite d.ignite pyrite pyrite pyrite pyrite pyrite pyrite pyrite

pyrite pyrite-terroselite pyrite pyr:I.te-f'erroselite pyrite fer:roseli te pyrite ... fer.roselite pyrite pyrite

'White Canyon district, San Juan County, Uta:h Happy Jack mine ' 107. do. do. pyrite pyrite-tn.a.rcasite pyrite "cha1co.py:ri te t-t chalc-opyrite Pereent Se !J B B .B ]3 B :B B B B B B o.oo1 o.o.oo6 0.0007 o."ooo; 0,.0003 0 0003 0 . .,0005 ·0 00.3 o .. ooo4 o.oo6 · o.Q003-o.oo0-3 Q.OOQ3. o.aQQ5 Oo003· o.oo6 o.oo6 2:$,.9 1B.o 0.()02 .0;.,030 Q.,QQ03 Oe.0006

Table 2 ~eleniUJln :eo:n.tent or SUlfities from the c c nle f.ormt;i.tton of Triassic a.:ge--Continue~ li!, 1114; 112,. 130;, 132., Rappy Jack mine ·do. t1.u~ Li:za.I"d mine do. .MeCarthy=Cole·man cl~m :Fa:l'ey Ura.nlum Cotnpany mine Maybe mine do. G'ismo mine do. do,- Qkte mine do. Hideout .mine Sandy claim Notch No. 1 mine Ha;p:py Surprise .claim .Mineral gj chalcppy:r:ite cove·lli te barn:i,. te .o :pyrite B. pyrite bornite ( e:taaltop}'l" · te ) 0 cha1.c:Q:pyrite chaleopyri te ·o pyrite B pyrite :S pyrite-marcasite ,Q py:ri.te E

.B 0 pyrite pyrite B pyrite pyrite pyrite pyrite .:B pyrite B pyrite B py:rti te-ma:reasi te ]3. dist:ric·t, San Juan County, Utah I.-;4. 138· r4o. 141,,. Mi Vid,a .mine .ao. d.Q . ., do. do:. doii LEi ,Sal No,. 2 1J'line pJ'!"ite pyrite pyrite pyrite pyrite :pyrj.te pyrite pyrite pyrite _greenock! te pyrite Monument Valley distri:et, Apache County, Arizona 144. N<:>-.i, 2 nrl.ne 145,. do .. liJ-6,. :do ... pyrite pyrite pyrite :s E 0. 0003· 0 ··.0004 Q- -5 Q.Q.Q04 o·.,oclQ4 o.ooo4 o.QOQ3 O:.QOQ6 Q.Q-Ql .Q.0003 -0;,003· o OQ15 o.oo_a,; O.,QQ.Q} o ooo:; 0..-QOOB Q,.,QQJ.. .o .o· oAa .o~o.o2 o.ooo4 o.ooa5

~ple 2.: .. ~1~Ft1-vnt e~te:nt ot sldtide$ fr:GUtl t.~ .E;hi.nl"e t~rma:tioxt :cJ.f. ~1lt:$&tte q:e .. ~o.n.t:t.nuecl.- tfukn~n. mine, ~letre .s Canyon ·@ . pyrite pyrite pyrite pyrite :"~e;r:¢ea:!f ]) General ~e.ographie loe&tion. <if £1 .. · i~ntifie·ation an afUU1%ed ·Jna'te::ria ~tat ;ton:_ iJlpll:re sa.mp.le,; ll1b¢ur:es. are hy:ph~ted :with .dQ~t ;pha·se ll.$ted .firat~ J., . (J?a.renthe.S;es) 1134iea.te. trace ~ts

-- tXLfi:ie :n.t t a k

r~k T~l$l~y s~'Wing greater than Q .Ol llaQe) 1 and .:e: sulfl~s taken. .:fr bar,ren rcrek.-. · · · · · !J . lindt · Q! l$ o.QOQ3 ;per:eent .$e. to·r a .. l~qa.:m :when J.es.s a:$lrl;ple \is -!tV#ilable the/ lower limit ~tf ®teetil;)n is :eorre.~p!l:ntiJvs.lY llj,gher.

161:. 167:. 174., 175'o .178:. 119· 184,. Table 3":""""SeleniUlri e.()ntent o.f s·u,lf;i.dea from sedimentary rocks .oi' Tertiary age- Vitro .mille a.o. :do., do. .dQ:c;_ a.o. a.o. do~ do~ do. do,. do. da.-. doGe d.o-. do.- do.- d.o· do~ do., d.oo . Lu.:eky Me mine do . ., :do., ·do·. de. Sareo:pha~ Butte Ridge .No, J,. mine d.o.. ~ig Horn NG-. 18 ela:im pyrite py.rlte (marts~si te) .m.arcas ite

mareasite ... py:rite

pyrite pyrite pyrite ·

pyrite ma.rca.si te.-pyrite :pyri ) :xna;rcasite~pyri~e pJ~rite (marea.site). pyrite~r.cas1te pyrite (snare·a~ite ) pyrite pyrite pyrite naareasi.te pyri te..-marea.s ite pyrite

.marcasite .spyri te pyrite pyrite ... marcas i te pyrite

pyrite-mar:Ca$ite py:rite B B :B B B .B :B .B ]3 :B 13' :B .B B B B ]3 B B :s B B :s o .~,ooo8 .'()·.0:005 o.oo6 O ,QOQ9 0,.003 -Q$-051 Q.,.Q003 o~-ooo;

· o.lt-1 o .. lQ Q,.QCl2

o Qo.:;

Tabl e 3.-'$elenium eontent of sulfides from seAimen.t:a.:ry rocks of Tertiary a.ge ---C·qnt.inued,. _:Pumpk:J..n :Suttesdistr-ict, Campbell County, WY"Qmin_g 195.· 196-., Cha.nnel claim Van No-.. 1 ela._im C J;"ane.y :Draw do. Jeanette mine Willow Creek Bl owout mine pyrite pyrite

:pyrite :pyrite Converse County distri-et, Wyoming 197. Dead cow. ar~a 198.. .Fetterman Creek Dry Fork pyrite pyrite pyrite Crooks Gap district, Fremont County, Wyoming Hel::en May claim Sno Ball Nq:_. 1 mine JACKSON FORMATION (EOCENE) Karnes C:o:unty district, Texas pyrite pyrite :pyrite ... rnarca.site Maybell d.istri:ct, MQffat County, Colo.r.aoo Sugarloa:r Mining property T:r.pe _'j/ .;Pereent Se lJ) o.oo8 o.o:; o ,oo4 o.ono; 0:.005 1/ General geographic lo'Cation o:r· saiupres shg:wn in figure 1 §I :x ... r:ay 1dentifica.ti()n on . a.:naJ_yze;d material-- - Quotation marks indicateimpu.re sample-; mixtures are hyphenated with dominant phaSe liste-d. firs,t, i_.;e.,. , 1area$i te --pyrite; (pa:renthe se,s) indicate tra..ce amounts. 3/ Sulfides d.ivide,d into two types: 0 ;:: t~en from, mi:mer.aliZed. rock Tusually s-ho-:win,g greater tha.n Q.~,Ol percent U$0-a), and J3 sulfides taken f'r:om-barren r.o-ek_., 4/ Lower, litni t of d.ete-etion is 0 .0003-p.ere-ent Se for a l~gra.m . sa:Jn;Ple ; -when les:s sample ·is ; available the lower limit of detect lent is higher.

Table .4.~-JSeleni'Qm content of sulfides from sedimentary rocks or Mineral g! .Edgemont clistriet, ~ll Cou.n,ty, SQuth Dakota 204. Flint Hill Quadrangle, .se:e_. 23, T 8 R. 3 E-. No-e 4 mine 206,. Trail J!'ractiO:n claim 20I~ Fli·nt Hill Qua.drangle, see !r- 15; TQ . 9 S .R 4 Croak -Co:unty district, Wyoming 2o8.. .Bus,tield mine . do. 210,. Homesta.ke mine pyrite pyrite pyrite pyrite pyrite "pyr:lte't 11PY"rite" Edgemont distri:ct, Fall River County, SQUtb. Dakota Edgemo:nt N. -quadrangle, . ·: pyrite 212. DiSney uranium mine pyri.te Edgemont di·strict, .Fall River County, South Dakota Burdo:c.k Quad.r"angle, : . ; · y se.e 8, T_,.. 7 S,., R. 22 E. LAKOTA. Crook County district, l;yoming 214._ Shannon OU Caiilpa!ly c.laims 215do· 216:'-· do. T:r:MF.A.$ FOltMATION El Paso county, Colorado pyri"\ie.(marcasite) pyrite pyrite pyrite .mar.easite

.B Q J3 ]3 :B. o a.o.o7 o.oo6 o oooa 0*'"017 o.ooo6 o.ooo3,

Table 4.'"".$e.lenlum ·cQnte.nt of sUlfides tr.cam sed.:Urlentaxy r~ks of cretaoe.ous a.ge contmuetl. Drill e·ore, ,llisappointm.e:nt (depth 274 ft) (depth f"t) dO~ ( ttepth 415 8 ft) Mineral'!} pyrite :py:rite(~easite) pyrite · pyrite py:rite(:ma.reaaite) ±/ General geog:raphie lo;eation of sa.m:ples in fi~e 1 . QA~-'J28 .(#.(lQ3 o~.,ooa Q,.007 gj X...ray identi£i~ation on .a.nalY%ed ~terial~ Qu-otation ma.rk.s lniicate . impure sa1nple; m.iXtUl"es are with dominant phase listed fir.st, , ,. tna:r"e~site..;pyri te; (pal'"-enthese.s) ind.late trace a;m.Q:unts. · 2/ $ulfides divided. intQ. two types~ Q .sulfides t:aken ;trom. mineralized ro-ck T'u.sually sho:vtl:ng grea.ter than Q. 01 percent UsO;s) , and 'B sulfi.des taken. from 'ba.r:ren rock,. Jbwer limit of detection. is 0,.0003 perceAt for a. 1-.gra:m. ·sa.m;p:t;e; 'When less sample ·is; .a.:va..ilable the lower limit of det-ection is torres:pJ()l~i;dillgly higher~

TAJ:>le 5J $elenium. eonten.-t of' su.lfldes from hydrothermal v~.in dep.Qs:tts · intrua±ve· rocks. La Sal MountainS , Gr~d C.ounty-, Utah Diqr~te pQ:rphyry · 221+._ Cqntaet zo-ne !:a :Par:actor to~ tlon ~'5.. D1!lri te porphyry 226,. Diorite porphyry 227~ Iiol"nfels (-.neo:s Shale) conta.:ct ZQne 228 :Diari:te pQrplryry 229., Diorite pOrphyry eh~eo:py:rite pyrite pyrite. bornite pyrite pyrite pyrite pyrite pyrite pyrite pyrite Jl General_ geO:gt~;phic l<:)ca.tion of' .s~les sho'W!l 11'.1-flour:e 1_. o.Q16 o.OQs o.oo6 o.QJ,.5 o-oo4 'fl l,-..ray i.~:ntifieat~on. on. anal~ed Dliltterial; all samples taken fJ!Qm. non:ut~.n.ite.rQU9: r .Q;eks. 2/ Lo~r limit of detection is 0.0003 pereent Se f .or . a l·~ :yl:ten. le~s S&ntple -- a~Uable the lilrd. t .of ~teet lou . is. .cp;:r;res.~n4J.~ll higher . .- · · ·

, "' ' . .f'· . I · · · r : D 1 SOUTH j DAKOTA D

Q Casper ' t f '

.o Ex;pla.na.tion 0 Tertia;ry sedimentary roGks + Cretaceous sed.im.enter ro.cks o Jurassic sedimentary rocks A Triassic sedim.enta.ry rocks Hydrothermal and igne rocks CO.LORADO xl.' NEW' MEXICO ; · ( IJ· Flagstaff o Albuquerque P.i,gure l.-Loca.tions of' samples.

COALIFIED LOG CARBONACEOUS IMPREGNATION Figure 2.--Coalified log showing typical sulf.ide concentration. URANIUM-VANADIUM ORE CHALCOCITE ONE FOOT Figure 5.--Roll ore body and associated sulfides, Morrison for.mation. VANADIUM ORE 1 J ONE FOOT Figure 6.--Galena-clausthalite band and its relationship to vanadium

Figure ;.--Pyrite associated with uraninite and coffinite. The mottled dark-gray groundmass is an intimate mixture of uraninite and coffinite. Zoned crystals of pyrite show selective replacement by galena, uraninite, and coffinite. The light irregular patches are also pyrite and they show strong corrosion by the uranium minerals. Mi Vida mine, Monticello district, San Juan County, Utah. Polished section, plain light.

Figure 4.--Pyrite associated with uraninite. Massive pyrite on the left is replaced by calcite. Light-colored zoned crystals in the uraninite (medium gray) are second-generation cobalt-rich pyrite. A band of sphalerite is along the right-hand border of the picture. Hidden Splendor mine, San Rafael district, Utah. Polished section, plain light.

Figure 7.--Galena-clausthalite band. Galena-clausthalite . (white) interstitial to detrital quartz grains; other intergranular areas between the quartz are filled with clay minerals. Rifle mine, Rifle district, Garfield County, Colo. Polished section, plain light.

Figure 8.-~Pyrite (white) impregnating and replacing wood structure. Black interstitial areas consist .of quartz and relict carbonaceous material. . Marshbank Canyon claim, San Rafael district, Emery County, Utah. Longitudinal polished section, plain light.

Figure 9.--Pyrite (gray) impregnating and replacing wood structure. Bl~ck areas are carbonaceous material. Hideout mine, White Canyon, San Juan County, Utah. Transverse polished section, plain light. lOJJJ

Figure 10.--Chalcocite nodule containing intergrowths of clausthalite (white) and chalcocite (medium gray) surrounded by embayed quartz (dark gray). Cougar mine, Slick Rock district, San Miguel County, Colo. Polished section, plain light.

Figure 11.--Chalcocite (light gray) and digenite (medium gray) containing small inclusions of clausthalite (white). Alteration {oxidation) along irregular veins (black to dark gray) produces secondary copper vanadates and carbonates. Cougar mine, Slick Rock district, San Miguel County; Colo. Polished section, plain light. lOJJJ

Figure 12.--Clausthalite (light gray) and eucairite (dark gray) in central part of chalcocite nodule. Cougar mine, Slick Rock district, San Miguel County, Colo. Polished section, plain light.

Figure 13.--Euhedral pyrite cubes in barren siltstone underlying mineralized sandstone. Rifle mine, Garfield County, Colo. Polished section, plain light.

Figure 14.--Mineralized log showing wood structure with the cell walls replaced by marcasite and the lumens (cell cavities) impregnated by pyrite-marcasite. Second generation of pyrite (white vein), calcite (black), and galena (medium gray)· cutting and displacing the woody structure. Adams Uranium Co. mine, San Rafael district, Emery County, Utah. Polished section, plain light.

' '

I , ' I ,,

I' ' "

' ' ' :HSe03 pH ' · I' r

' Q2 ' ' Figure 15 Stability £ield of' selenium (temperature 25° C, pr:e.s-sure 1 atm, concentration o_f .Se 10 6 .M)--a:rter Delahay, Po-ur'haix, and van :Rysf?elberghe (12-)..