American Mineralogist, Volume 64, pages 75g-767, Ig79

Polytypism in molybdenite (I): a non-equilibrium impurity-induced phenomenon

ReruBn J. J. Npwnpnny

Department of Geologlt, Stanford lJniversityS t anford, C alifu rnia 9 4 3 0 5

Abstrsct

MolyMenite is known to occur in either of two structur€s: the @mmon 2H, polytype andthe rare 3R polytype. Most previous work has suggested that 3R forms in low sulfur fugacityenvironments and is sulfur-deficient relative to 2H,. A thorough review of the literature onmolybdenite, however, suggests that the occurrence of 3R molybdenite in nature is not r€-lated to sulfur fugacities or temperatures of fonnation, but rather to impurity content. X-raydiffraction analysis of 184 specimens of molybdenite indicates (l) there is a semi-quantitativerelationship between the mean 3R abundance and the mean impurity cont€nt of molybde-nites from a given deposit and (2) the 3R abundance is hdependent ofthe sulfidation statesof the associated sulfides. Microprobe studies indicate that there 41e 11e 5ignifisant major-ele-ment di-fferences between 3R and 2H, molyMenite in the samples studied.

A synthesis of the experimental stability studies of molybdenite polytypes suggests that 3Rgrows by a screw-dislocation mechanism and is unstable with respect to 2Hr, but does notreadily convert due to kinetic barriers. Growth by screw dislocations in nature most fre-quently occurs due to internal strains generated by high impurity contents in the growingcrystals. Rhenium, tin, titanium, bismuth, iron, and tungsten are the most common impuritiesfound in 3R molybdenite, and their role in the mineralization prooess with respect to forma-tion of 3R molybdenite is discussed for several deposit types.

Introduction

The structure of molybdenite (MoSr) is based onstacking of planar close-packed S-Mo-S layers. Al-though such layers could be stacked in a variety ofways, only two stacking sequences have been ob-served in natural and synthetic molybdenites. Be-cause both mins14fu consist of structurally identicalMoS, layers and differ only in the length of c, theyare polytypes. The most common polytype in natureis hexagonal (2Hr, spaoe group P6r/mmc), consistingof two layers per unit cell. The less common naturalpolytype is rhombohedral (3R, space group R3m),consisting of three layers per unit cell (Fig. l).

The origins and controls of polytypism are not wellunderstood, although most studies of the SiC, ZnS,and CdI, systems have indicated that the phenome-non in these systems is related to dislocation/growthprocesses and di-fferences in energies of differentstacking sequences (e.9. Trigunayat, l97l). In thecase of molybdenite much controversy exists con-cerning the origins of polytypism. Some authors sug-

gest that the two polytypes are stoichiometrically dis-similar (Clark, 1970), whereas others believe that thetrace-element contents differ (e.9. Somina, l!ff;.Still others postulate di-fferences in pressures of for-mation @adalov et al., l97l), temperatures of forma-tion (Arutyunyan et al., 1966),61sooling rate duringformation (Chukhrov et al.,1968).

Frondel and Wickman (1970) have shown that 3Rnolybdenite is present in a variety of ore depositsand is not restricted to "exotic" low/(Sr) environ-ments as suggested by Clark (1970). However,Clark's (1970) conclusion that 3R is I to 5 weightpefcent sulfur-deficient relative to 2H, has been ac-cepted by several authors (e.9. Hasan, l97l; Ctargand Scott, 1974).If, indeed,3R molybdenite is an in-dicator of low sulfidation states, its rarity in naturesuggests that molybdenite deposition occurs underrelatively high 5ulfq1 fugacity conditions. This hy-pothesis has signifigaat implications for molybdeniteore deposition, e.9., it suggests transport by high sul-fide mechanisms.

This investigation is based on a study of 184 mo-00n.3-004x/ 7 9 / U08-0758$02.00

-+ [roro] :. o

- M o

Fig. l. Structures of some MS2 polytypes projected onto (1120),modified from Takeuchi and Nowacki (1963).

lybdenite-bearing hand specimens to determine poly-type content, vein mineralogy, alteration, major- andtrace-element content, and environment of deposi-tion, in order to better understand the causes of poly-typism in molybdenite. Part I is a study of whatcauses original formation of the 3R polytype; it leadsup to a hypothesis explaining the relation betweentrace-element and 3R content. Part II examines thedetails of a particular system (Re/3R in the porphyryenvironment) and discusses the geochemical circum-stances under which 3R is preserved and/or de-stroyed by post-depositional alteration.

Review of past work

Polytypism and sulfidation states

Little information is available in the literature onthe major-element compositions of 2H' and 3R mo-lybdenite. Available analyses, however, show that nosystematic correlation exists between sulfur contentsand 3R abundances. Moreover, these data show noindication that the 2H' polytype is I to 5 percentricher in sulfur than the 3R polytype. Both polytypeshave been reported with compositions ranging from39.9Vo S (Seebach, 1926) to 4O.4Vo S (Zelikman et al.,196l) (Fig. 2). Detailed tabulations of these andother molybdenite data may be found in Newberry(1e78).

If natural polytype occurrences are a function ofsulfidation state of the environment, as has been sug-gested by several studies, then there should be somecorrelation between 3R/2H' occurrences and the sul-fidation states of the associated minerals. Literaturedata suggest, however, that there is no such correla-tion. 3R molybdenite has been found associated withsulfidation states ranging from high-native sulfur(Znamenskiy, 1969)-{hroggh intermediate--chal-copyrite + bornite + pyrite (Hasan, l97l)-to low-bismuth + pyrrhotite (Petruk, 1964). similatly,2}'

POLYTYPISM IN MOLYBDENITE 759

is found in a variety of environments. lndeed, 2H,and 3R polytypes are cofllmonly mixed within asingle specimen, as shown by single-crystal experi-ments (Drits and Chukhrov, l97l). $imilafly, gpln

size is not related to polytype abundance, contrary tothe suggestion of Wildervanck and Jellinek (1964)' asindicated by a review of the literature.

Polytype abundance and impurity content

Several studies, starting with Somina (1966)' have

suggested that 3R molybdenites characteristicallycontain high impurity contents. Rhenium is the mostcommon impurity element, ranging in concentrationfrom less than I ppm to greater than 5000 ppm. Nodetailed study of impurity contents and polytypismin the same samples has been undertaken, but thegeneral trend of the literature is that high rhenium

contents can be correlated with high 3R contents (e.9.Ayres, 1974). High contents of other trace elements,such as Ti, Bi, W, and Fe are also correlatable withthe presence of the 3R polyype (Table l). In con-

trast, 2H, polytypes are relatively pure (Chukhrov et

al.,1968\.

Synthesis studies

Published experirrental data show that (l) bothpolytypes can be synthesized (Bell and Herfert,lg57), (2) 3R converts to 2H, when heated for longperiods at temperatures in excess of 500oC (Wilder-

vank and Jellinek, 1964), (3) the time necessary forcomplete conversion increases drastically as temper-ature falls (Clark, 1970), (4) 3R converts to 2H, fasterand more completely in the pres€noe of sulfur gas(Zelikman et al., 1969), and (5) 3R converts to 2H,

only by recrystallization (Zeliknan et al.,1969).Experimental synthesis of molybdenite by reaction

of molybdenum metal with sulfur vapor or liquidgenerally results in a mixture of the 3R and 2H' poly-

types (Dukhovskoi et al., 1976). In one experimenttwo MoS, layers were formed: a2Hrlayer next to themolybdenum metal and a mixed 3R + 2Hr layer nextto the sulfur (Zelikman et al., 1976)- These results are

o 6 -

o

b 9 .e7E Ez

3 9 I 4 0 0 4 0 2 4 0 ' 4

wt. 7" S

Fig. 2. Histogram of molybdenite chemical analyses taken fromthe literature.

NEWBERRY:

3 R

c

t

NEIIIBERRY: POLYTYPISM IN MOLYBDENITE

inconsistent with the theory that 3R forms in com-paratively sulfur-poor environments. Rather, as Ze-likman et al. (1976) point out, there are differences inthe growth mechanisms of molybdenite formed at themetal- and sulfur-interfaces.

Similarly, molybdenite formed by reaction be-tween MoO, and sulfur is 2H, whereas in the pres-ence of an alkali-carbonate flux 3R forms. Again,there are differences in growth mechanisms betweenthe two experiments, rather than differences in sulfuractivities. Elwell and Neale (1971) have shown thatfluxes generally cause crystal growth by a screw-dis-location mechanism.

Finally, the addition of an impurity (in particularRe) in synthetic molybdenite affects the 3R to 2H,conversion reaction. 3R molybdenlls colhining 0.05to 37o Re converts more slowly than pure 3R molyb-denite, and 3R containing more than 3VoPie has notbeen observed to convert at all, even at temperaturesof I l0O"C (Zelikman et al.,1969\.

In summary, experimental work suggests that 3R isunstable with respect to 2H, under geologically rea-sonable P-Z conditions, but that there are kinsligbarriers to establishment of equilibrium once a givenpolytype has formed. This implies that polytype for-mation depends more on growth processes than onequilibrium processes. A survey ofthe literature alsoindicates a possible correlation between the polytypeand trace-element content, and a lack of correlationbetween 3R content and sulfur fugacity or temper-ature of formation. A lack of hard data supportingany one cause of 3R formation, a lack of a mecha-nism for explaining the formation of 3R, and a lackof data on the post-crystallization behavior of 3Rprompted this detailed study of polytypism in molyb-denite.

Table l. Impurity contents of some 3R molyMenites from thelitcrature

Locat ion Imp ur i ty Reference*

B inDenta l , Sv l tz .Kurd le Is land, USSREast S iber ia , USSRPanasque i ra , por tuga l

Ahen ia , USSR

syDtheElc

Con I'line, Yellowknife, Can.

0 . 7 " 1 w 1ca, 500 ppn Fe 227" 'Iit 3O0 ppn Nb, V 3150 ppn Fe; "sore" W 4

1 . 8 8 2 R e 5o . L 6 ' / . ' t 7 6ca. 500 ppn Fe 1

aReferences: (t) Graeser (1964)j (Z) Zwenskiu eL aZ. (lg?0).(3) soniw (1s66); (4) cL@k (.1g6s); ts) rad,rmja-oul"nuol"nuayon(1s63); (6) BeLl qd Herfeft (1gsz); (?) Bolle (is6s)

Experimental details

Five molybdenite-bearing samples representingdifferent deposit types were trimmed to 0.5-cmpieces, mounted in epoxy, polished, and carbon-coated. Qualitative wavelength scans of the five sam-ples on an ARr-Eux-sM microprobe indicated thatonly Mo, S, and Re were present in quantities greaterthan the O.05Vo limit of detection. The samples werequalitatively analyzed for Mo and Re. Overlap of theMo La and S Ka peaks prohibited analysis of sulfur;the Mo LB peak was employed for Mo analysis. puremetals were employed as standards. The followinganalytical conditions were used: output voltage, 15kV; emission current, 50 mA; sample current, 0.03mA; count time, l0 sec. Results were corrected forbackground and drift and expressed as k-ratios (ratioof sample to standard counts).

Six samples of 2H, molybdenite from the Qlims;Mine, Colorado, were cut to 0.05 x 0.5 x 0.5 cmpieces and heated to 700oC in a stream ofpure hy-drogen gas for I to 36 hours by H. Wise and P. Wen-treck of SRI International. As shown by Aoshimaand Wise (1974), this procedure results in the forma-tion of sulfur-deficient molybdenite. All sampleswere analyzed for 3R content.

The polytype content of 184 molybdenite samples Resultswas determined by X-ray diffraction techniques, fol--

Polytype abundance

Polytype abundance for molybdenite from eachdeposit is presented in Table 2, with rhenium con_tents taken from the literature. Although there is awide range ofboth 3R abundances and rhenium con_

tures of 2H' and 3R calculated by Wickman and tents within a given deposit, in general the 3R poly-Smith (1970)' type occurs in rhenium-rictr motyuaenite deposiis.

All diffractometer experiments were carried out us- Mean values of 7o3R and ppm Re are plotted in Fig-ing Ni-filtered CuKc radiation, a scan speed of /a" ure 3, which shows a tineai correlation between rhe-20/minute, and a scan range of 26 to 53o 20. Qaartz nium content and 3R abundance for several deposits.was added to all samples as an internal standard. This correlation might only indicate that rhenium

761

o

8

a

a

N

ea

N

a

* s '= J

O -

B b

N . .v N l

rJ l

. . $ l1 $ l

N Sq $H $

\ tS

0 ' o( a a

\ b

t J N

N O

8 SssB ( A

\. . 8

$ sO Bs . d

. p la sE f u*

NEITBERRY: POLYTYPISM IN MOLYBDENITE

s s F d

@ 4

o o o

d d d d N t N d d

N O dN O OI N N d

i t . l. O @ O

o d o

o o

N d O N N i N N @d H a s { o N d Nt t t t t t l

o o o € N 9 0N N O H d N

@ 9

q. . d d

> d J lo d oz a 6

. O d! . EE q d l oo . d ! o i lE d H O d t s

d g H o. o . o ) )! E O . r i! - O ! ! Ed d ! o o

E E X d d !d k 6 ! h oo o d E 6 0

z a 9 < t s i

E ? I ? ? T - 6Q O O O O O @

t N o d@ a

F@ ot r \

i l o

S - a { d N * 5 - N d d dF >H N

E !O N

. x 6x \d N o o r @ o oO N N O N N N OE r.l

H P

o a

d X5 . OO N E

. N 6 - dN t < 9 C H

< N 6 0 0< " E i 6

. 6 d ! Od . ! o d @o € . d . d t sE d ! 5 do € H d o . dt @ o t @ oo 6 i l c 6

> F 4 9 H >

6 0 d 4 dt t t t l

O N i l i l T d

4 n 9 S O i l d € d 4 h

N S 9 @ O N S O @d d i d d

d d d i N

@

s aN S

o

! . 9+ . E i oo N . . O Oo < z N dP < ^ XO . ^ O O

! d d . ! 5o u d d

t s 5 d N Oo a & F d dO Q d E Oo > d H o do ! o H d

6 d o i dI 6 6 H H Oo!o

ioI

E

d .N

g .o d d d d N

x o oa @ o N o o! N N 9 S Oo t t l

o o s

4

H

o .E !d H p. . c d t r

q q d C ^ O. d d A d : Ed i u ld d . q t s

9 9 . O O 5a = o

. . dd O € . H JA d T ! O. d d . Q dE > r o € o

i l F Ep o o a o oo E l d t st r t ! € ! o! a 6 t s p dG d o 6 d o

o < B B | l 9

@ @O O N F S@ N S d o

t i l l ld t r 4 €d o t s o N

o d@

tsq

t

{o t @

i H@ Ho dA Oo p

€ l

E - o @ o N E - - d d Nc N oo €€ pP - 'h i

d o

= lO d C s n A '

> \ N N O N N NF p o> o . z

! ^ @ dQ o O C 9 .k o ^ 0 , t q oo .d 5 i d o ts . i d

d J ! ! t d o d oo o : i = a a . o 6 z

o d o s oo 9 0 N i

t t I

9 + SN N

H N 6 O N +d

6 0 0 6 d

J S J t T

S 9 @ N N

d o n N 6d d d d d

. NN <

' r i& ! o

. 6 C O. N O O !

N < t u = 9< N

- O ^ H <. . d U O O

d E + p rE 6 ! U A O

d i ! t O . -A E H € O <

t NN 5 << !

O . A td p d !

€ P d . d6 0 d 6> . O E oo e o 'z d F o d

S q o o- @ > o E> C d F E

d i . d o oH € @ = O

E !) A ^

d o Eq ! do c o1 o -

o ^. ! N

x a vo o

o !o o

! o No d h

z q 6

d c

O H. q o

o

U Fi o6 p

.d

d

Fi

*E O5 0

o k

o!

d oo o

6

E ! ^

. d o aq p oo t r v

d o

o ^E ! No t r v

o oH O N

! q dE E d5 6 C

z @ Q

o

o

o

,IEa€

3do

o

()

cio

D

F

762 NEWBERRY: POLYTYPISM IN MOLYBDENITE

100

f eoo

concentration and 3R formation fortuitously occurunder similar conditions. However, within a group ofsimilar deposits (ag. porphyry copper deposits),those with high-rhenium molybdenites also tend toshow greater 3R abundance than those with low-rhe-nium molybdenites. There is no evidence indicatingthat the Sierrita deposit, for example, formed undersignificantly different T-J(S,) conditions than the SanManuel deposit, yet San Manuel has high rheniumcontents and high 3R abundances, whereas Sierritahas low rhenium and no 3R.

There are also no indications of a correlation be-tween low sul0dation state and 3R abundance in thesamples studied. Table 3 lists coexisting sulfide as-semblages for selected samples, covering a variety of3R abundances; the lack of correlation with sulfida-tion state is readily apparent.

C o mp o siti onal v ariatio ns

Table 4 shows the 3R content and concentration ofrhenium and molybdenum in five molybdenite sam-ples. As Table 4 indicates, there is no signifisant vari-ation in molybdenum content for the five samples,

o Porph Cu

. Porph Mo

a Cu Skarnr !V Skarn

but there is a significant difference in rhenium con-tent. These results again indicate that impurity con-tent, rather than sulfur/molybdenum stoichiometry,is the determining factor in polyype abundance.

Annealing experiments

Six samples of 2H, molybdenite were held at700oC in a sulfur-deficient atmosphere for l-36hours. None of the samples showed any stg[ of con-version to 3R. Quenching from 700oC prevented re-crystallization back to 2H,, had any 3R formed, andthe presence of a sulfur-deficient atmosphere wouldhave allowed the transition from 2H, to 3R to occurif the lowl(S,) made 2H, unstable with respect to 3R.These experiments, therefore, indicate that 3R/2}I,stability is not a function of sulfur fugacity.

In summary, the analytical results indicate that 3Rmolybdenite (l) is not uncommon in several geologicenvironments, (2) is not a product of the sulfur fuga-city of the depositional environment nor of stoichio-metry, but (3) is related to the amount of impuritypresent in the molybdenite. Finally, 3R cannot beproduced by recrystallization in a sulfur-deficient at-mosphere.

Eo.q +o

c

3"u l Pegmat i t e

20 30Mean % sR

Fig. 3. Mean 3R abundance vs. mean rhenium content for molybdenites from deposits listed in Table 2.

NEWBERRY: POLYTYPISM IN MOLYBDENITE

Table 3. Typical molybdenite mineral associations, polytype contents, deposit types, and relative sulfrdatioo states

763

Sanple ll Locatlon Assenblage 1 Deposit tyPe PolytypeSulftdation

State

26801

10884

N7604

Bn48

2056r

DDn-9-618

sM-9

ws-s-1rrh-2cx-2

Bora, NSW, Australla

Alum Gulch, Nevada

Alplne Mlne, Calif.

Brownstone MLne, Callf .

Marb le Bay , B .C.

Larap, Phil ippines

San Manuel-, Az.

S ie r r i ta , Az .

Ithica Peak' Az.

Clinax, Colo.

QgoMoTTott3

QtoMotMarct

PoOOMoOOSchTO

l"lo-Cp-Py (dtssen)

MoGOPyx48Pylcpl

PY4ott45to5

Quot'[orocntt?r,QuoMotocPgPl2.QooccooPrrrMofQ4owolf 2oMo2oPYzo

Pegnatlte

Quartz vein

TungsEen skarn

Tungsten skarn

cu-lto skarn

Cu-Fe-Mo skarn

Porphyry CopPer

?orphyry CoPPer

Porphyry CoPPer

Porphyry MolYbdenum

2rt2Ht

2\

2flt

952 3R

10% 3R

957. 3R

2Ht

2Ht

Zflt

lnternediate

low

lnLermediate

internedlate

internediate

lntenoedlate

lnterloedLate

interDedlate

Lnternediate

LhinntaL abbt:eoiations: ce=chaLcoci,te, cp=elwlcopyri.te, Ma!TMt?9ai.:e' MoanoLgbdeni'te' Mhnagnetite' Py=

py,ite, P,Fpg?otene, q=qunitz, Sch=slhieli'te, Ibur-tou,rnaline, WoLf<'toLfranLte'

2c" i,"

"rp"ogene in this satnPLe.

Discussion

Any discussion of polytypism in molybdenite mustaccount for three observed facts: (l) 3R is unstablerelative to 2H,; (2) despite this instability, 3R is com-monly formed in nature; and (3) it is kinetically dift-cult to convert from 3R to 2H'. These facts are bestdiscussed in light of the crystal chemistry of rnolyb-denite.

Stntcture-energetic reasons for polytpism

Examination of the geometry of the 3R and 2H,polytypes indicates that Mo ions in a given S-Mo-Slayer can either be (l) directly above and below Sions in immediately adjacent layers, or (2) slightlyoffset. Bonding between layers should be stronger in

case (l) where Mo ions are closest to adjacent-layer S

Table 4. Electron microprobe analyses of selected molybdenite

samplcs

ions. The 2H, polytype is of confguration (l), while

several theoretical polytypes not observed in nature

or the laboratory have configuration (2). The 3R

polytype has an intermediate geometry, in which a

liven Mo ion has an adjacent-layer S ion either di-

iectly above or directly below it, but not both' These

configurations are illustrated in Figure l.Weaker layer-layer bonding in 3R than in 2H', as

indicated by layer geometries and Vickers hard-

nesses, makes 3R less stable than 2H,. The exciton

spectra of molybdenites also indicates that the 2H'

polytype is energetically stable relative to 3R (Wilson

and Yoffe, 1969).

Polytypism and growth mechanisms

There are strong indications that occurrence of the

3R polytype is not caused by an equilibrium mecha-

nism, but rather by a growth phenomenon with a

lack of subsequent re-equilibtation. Zelikrna;t et al'

(1976\ noted that there were di-fferences in the growth

mechanisms of 3R and 2H, molybdenite, but were

unable to quantify these differences. However, by

in an alkali-carbonate flux (a flux causes growth by

addition onto screw dislocations, according to Elwell

sa[p leouber Locatlon rgo x ro2 q. * ro4

f polntsanalyzed

5059 E1y, Nevada 8

20L2L CaEnea' Mexlco 6

20550 l , tarble Bay' B.C. 8

42034 Quest4, N.M. 6

N7604 Alpine Mlne, cal i f . 5

54,9 + 0 .3 2500 + 500

55.0 + 0 .3 no t de tec ted

54.8 + 0 .3 750 + 300

54.8 + 0 .3 no t de tec ted

54.9 + 0 .3 no t de tec ted

95

0

50

0

0

*K = (counts for sample/counts for metal standard) = concedtral ion

7U NEWBERRY: POLYTYPISM IN MOLYBDENITE

Table 5. Structural and chemical parameters for MSo layer compounds

ConpoundM-S

Coordination StructureM-S Bond

r,eneth (A)Meta l

Elec tronegatlvityFormal charge

on netal

MoS2

wt2

ReS2

Tis2

SnS,

B i2s3

trlgonal prisnatic

trlgonaL prisnatlc

distorted octahedral

octahedral

octahedral

square prismat.ic

layer

J-ayer

layer

layer

layer

layeredrlbbons

2 . 4 2

z . 4 r

2 . 2 8 - 2 . 5 6

2 . 4 3a 4 a

2 . 5 6 - 3 . 0 5

1 . 8

L . 7

1 . 9

1 . 5

1 . 8

1 0

4+

4+

4+

4+

4+

Jf

xData fnom EuLLigen (i.9?6) and pauling (1960).

and Neale, 1971) is 3R, whereas molybdenite grownfrom melt without the flux is 2H, (Rode and Lebe-dev, 196l). Agarwal and Joseph (1974) used a so-dium peroxide/potassium nitrate etch to show thatscrew dislocations were at the centers of growth spi-rals in synthetic 3R molybdenite; they were unable toconfirm these results for natural 2H, molybdenite.Hulliger (1968) indicated that there were differencesin the dislocation networks of the 3R and 2H, poly-types of molybdenite.

Polytypism and impurity content

The factors which give rise to screw dislocations inmolybdenite will cause the formation of the 3R poly-type. In the absence of strong mechanical stresses,the most likely cause is compositional impurity. Innature variations in impurity content are due to fac-tors such as: (l) abundance of the impurity elements(e.9., Re, W, Sn) in the hydrothermal system, (2)availability of impurities with the correct oxidation/sulfdation states, and (3) sufficient insolubility of theimpurity elements at the time of molybdenite deposi-tion.

These factors can be illustrated for the elementsSn, Ti, Bi, W, and Re. All these elements form MS"layer sulfrdes and are of approximately the correctsize to fit into the molybdenite structure. Chemicaland structural parameters are listed in Table 5 forthese elements; the data indicate that none possessesthe perfect size, charge, and/or electronegativity tofit into molybdenite without some strain. Theamount of strain depends on the misfit of the ele-ment: a given amount of tungsten, for example,causes a good deal less strain than an equal amountof tin. Thus the level of impurity necessary to cause

sufficient strain for the formation of significant num-bers of screw dislocations also depends on the ele-ment considered-there is no one set level of impu-rities above which 3R molybdenite is formed.Whatever these levels are, however. there are envi-ronments that will give rise to a concentration ofeach impurity in molybdenite, and these are fre-quently associated with the occurrence of 3R molyb-denite.

3R molybdenite is associated with tin-bearing grei-sens at Lost River, Alaska; Altenberg, Saxony (Fron-del and Wicknan, 1970); Cinovec, Bohemia (Fron-del and Wickman, 1970); and Rooiberg, SouthAfrica (Mandarino and Gait, 1970). Semiquantita-tive spectral analysis of Cinovec molybdenite in-dicates the presence of several hundred ppm Sn(Kvacek and Trdlicka, l97O). At the Panasqueira,Portugal, tin-greisen early molybdenites are 3R, butthose which have been recrystallizsd are2Hr.

Similarly, titanium-bearing molybdenites fromhigh-titanium deposits have been reported. At MontSt. Hilaire, Quebec, 3R molybdenite is associatedwith the rare titanium minerals leucosphenite, nar-sarsukite, and ramsayite (Chao et al., 1967). Molyb-denite from a rare-metal-bearing carbonatite in EastSiberia contains 2.Wo Ti,0.037o Nb, and O.|lVo Zr(Somina, 1966).

Bismuth and bismuthinite are also associated with3R molybdenite: at Whipstick, Kingsgate, and KingIsland, Australia (Ayres, 1974); ttie Pax Mine, On-tario, Canada (Mandarino and Gait, 1970), and Mt.Pleasant, New Brunswick, Canada (Petruk, 1964). AtMount Pleasant, molybdenite containing bismuth in-clusions in greisen is 3R whereas molybdenite insphalerite veins is 2H, (Petruk, 1964).

NEI{BEKRY: POLYTYPISM IN MOLYBDENITE

Most tungsten associated with molybdenite is inthe tungstate form: wolframite and/or t"tt""1i1s. Thismolybdenite is inevitably 2H, unless there are highlevels of other impurities. Molybdenite from Binnen-tal, Switzerland, however, which is associated withseveral high-sulfur sulfosalts, contains O.lVo W and is3R (Graeser,1964). As Hsu (197'7) has shown, tung-stenite is stable only under very lowl(Or)/very high

l(Sr) conditions. The general scarcity of tungsten inmolybdenite, despite similar radii, electronegativities,etc., is due to this fact, and thus molydenite fromtungsten deposits rarely is 3R.

We have already seen that porphyry copper depos-its are commonly associated with high rhenium levelsand abundant 3R molybdenite. Due to the multi-stage nature of the porphyry copper system, however,in which temperature, pH, and other geochemicalvariables undergo significant changes with tirne, onecannot automatically assume that all such depositsare associated with abundant rhenium and 3R mo-lybdenite. The second paper in this series on poly-

typism in molybdenite will deal with the particularcharacteristics of the porphyry copper system as it re-lates to changes in the rhenium content of molybde-nite and appearance/persistence ofthe 3R polytype.

P olyt yp ism and re c ry st alli z atio n

The importance of high-temperature recrys-tallization as a mechanism for conversion from 3R to2H, molybdenite cannot be overstated. This mecha-nism is necessary because the S-Mo-S layers must berotated as well as translated to convert one polytypeto the other. The process of moving layers requiresenergy because there is significant bonding betweenlayers. Further, the strongly-repelling S ions of adja-cent layers are forced to come in close contact whenlayers are moved. Thus, the process of polytype inter-conversion requires far more energy than the simpledisplacive mechanisms characteristic of many poly-typic conversions. The difficulty of 3R to 2H' con-version is further illustrated by the results of inter-calation reactions, in which aftali-metal ions arechemically inserted between adjacent MoS' layers.Somoano et al. (1973) observed that the intercalationof Rb or Cs ions into molybdenite at 70oC does notaffect the symm€try-i.e. 3R remains 3R and 2H' re-mains 2H,-although the interlayer separation in-creases by more tftlatl0Mo.

Thus, unless energy is available for the recrys-tallization prooess, 3R will be metastably maintained.The presence of impurities in molybdenite will also

hinder recrystallization, as pointed out earlier. Re-

crystallizatiot can occut, however, and thus the pres-

ence of high levels of impurities does nlot Suoranteethe presence of 3R. What has been demonstrated inthis paper is that the general trend to recrystallizationis sufficiently hindered that a 3R/imFurity relation-ship can be established. Part II of this series dealswith the recrystallization process and how it affects3 R/impurity relationshiPs.

Conclusions

Results of polytype determinations of 184 molyb-

denite-bearing specimens, trace- and major-elementanalyses of selected samples, detailed hand-specimenexamination, and thorough literature review stronglysuggest that the formation of 3R molybdenite in na-

ture results from non-equilibrium growth prooesses'

which in general are impurity-related' No correlationexists between the 3R polytype and sulfur fugacity offormation or stoichiometry of the molybdenite asseen from depositional environments, local sulfideassemblages, and major-element analyses. Strongevidence suggests that molybdenite samples contain-

ing more than about 500 ppm impurities contain

some of the 3R PolYtYPe.A geometrical treatment of bonding in molybde-

num sulfide indicates that the 3R polytype is un-stable relative to the 2H, polytype and that there areenergetic barriers to interconversion. However, the

structure of 3R makes it the kinetically favored phase

when molybdenite crystals grow by a screw-dis-location mechanism. Such growth in nature can becaused by strains induced by high impurity contentsin the growing crystals; in particular, the elementsRe, Ti, Sn, Bi, W, and Fe occur in anomalousamounts in 3R molybdenite. As di-fferent natural hy-

drothermal systems will tend to concentrate differentimpurity elements in different ways' no general-

izations can be made about the specific environmentsof deposition of 3R molYbdenite.

Acknowledgments

Drs. G. E. Brown, M. T. Einaudi, K. B. Krauskopf' and F. W.Dickson read the manuscript and offered many helpful sugges-tions. Dr. Einaudi made the Stanford University Ore Deposits

Collection available for study. The Union Carbide Corporation,San Manuel Copper Company, Phelps Dodge Company, Ana-

conda Company, and Asarco Incorporated kindly gave permission

to visit their properties. Dr. H. Wise of SRI International donatedthe hydrogen-treated sampl€s of molybdenite. The author was

supported by a NSF graduate fellowship and Stanford UniversityTeaching Assistantships.

NEWBERRY: POLYTYPISM IN MOLYBDENITE

References

Agarwal, M. K. and B. Joseph (1974) Dislocation etching in mo-lybdenite. J. Mat. Sci.,9, 1262-126/..

Ambrus, J. and H. Soto (1974) Estudio geologico del moliMenoen Chuquicamata, Chile. Sudia Geo1.,8,45-83.

Aoshina, A. and H. Wise (1974) Hy&odesulfurization activityand electronic properties of molybdenum sulfide catalyst. J.Catol. , 34,145-151.

Arutyunyan" L. A. and E. Kh. K_hurshudyan (1966) Synthesis ofmolybdenum disulfde from sulfomolybdenate solutions at hightemperatures. Geokhimiya,f 650-658 ltransl. Geochem. Int., 6,4974861.

Ayres, D. (1974) Distribution and occurrence of some naturally-occurring polytypes of molybdenite in Australia and papuaNew Guinea. J. Geol. Soc. Aust.,2I,273-278.

Badalov, S. T., E. Kh. Khurshudyan and A. Turesebekov (1971)Polytypes of molyMenites from the Dal'nee molybdenum-cop-per-gold deposit. Dokl. Akad. Nauk llzb. SSR, 2& 4142 (norseeq extracted from Chem. Abstr. 76,88408s, 1972).

Bell, R. E. and R. E. Herfert (1952) Preparation and character-ization of a new crystalline form of MoS2. J. Am. Chem. Soc.,79,3351-3354.

Boyle, R. W. (1968) Conditions for the formation of rhombohe-dric [sic] (3R) molybdenite. Dokl. Akad Nauk SSS& 18J, 332-333 [transl. Sov. Phys.-Dok., 13, 1089-1090 (1969)].

Chao, G., D. C. Hanis, A. W. Hounslow, J. A. Mandarino and G.Perrault (1967) Minerals from the nepheline syenite, Mont St.Hilaire, Quebec. Can. Mineral., 9, lW-123.

Chukhrov, F. V., B. B. Zvyzgfu, L. P. Yermilova, S. V. Sobolevaand V.G. Khitrov (1968) Polytypes of molybdenite and their oc-curences in ores. Geol. Rud. Mestorozhd., 10, 12-20 [transl. 1nr.Geol. Rev., 12,7+86}

Clark, A. H. (1965) Rhombohedral MoS2 from the Minas da pan-asqueira, Beira Baixa, Portugal. Mineral. M ag., 3 5, 69-7 l.

----{1970) Compositional di-ferences between hexagonal andrhombohedral molybdenite. Neues Jahrb. Mineral. Monatsh.,33-38.

Craig, J. R. and S. D. Scott (19?4) Sulfide phase equilibria. In p.H. Ribbe, Ed,., Sulfide Mineraloglt, p. CSI-CSIO4. Mineral. Soc.Am. Short Course Notes l.

Drits, V. A. and F. V. Chukhrov (1971) Intergrowth of polytypemodifications in molybdenite crystals. Vop. Odnor. Neodn. Min-eral., 1971,73-76 (not se€n; extracted from Chem. Abstracts,77,106302e,1972).

Dukhovskoi, E. A., Yu. V. Makarov, N. I. Sarpov and A. A. Sitin(1976) Crystal structure of MoS2 coatings on molybdenum. Izv.Akad. Nauk SSSfi, iVeorg. Mater., 12, 1349-1351 ftransl. Inorg.Mater., 12, I I l8-1120 (1976)1.

Elwell, D. and B. W. Neale (1971) Review: mechanisms of crystalgrowth from fluxed mels. J. Mater. Sci.,6, 1499-1519.

Faramazyan, A. S. and E. Kh. K-hurshudyan (1963) Isomorphismbetween molybdenum and rhenium in molybdenite. Dokl.Akad. Nank Arm. SSR, 37, 2ll-216 (not seeq extracted fromChem. Abstr.,60, 14257b, 1964\.

Frondel, J. W. and F. E. Wickman (1970) Molybdenite polytypesin theory and occurrence: II. Some naturally-occurring poly-types of molybdeaite. Am. Mineral.,jt 1857-1825.

Graeser, S. (1964) Uber Funde der neuen rhomboedrischen MoSr-Modifrcation (Molybdenit-3R) und von Tungstenit in den Al_pen. Schweiz. Mineral. Petrog. Mitt.,44, l}l-128.

Hasan, Z. (1971) Supracrustal rocks and Mo-Cu bearing veiru inDalen. ffors& Geol. Tidsskr., 51, 287-310.

Hsu, L. C. (1977) Effects of oxygen and sulfur fugacities on rhescheelite-tungstenite and powellite-molybdenite stability rela-tions. .Econ. Geol., 7 2, 664-670.

Hulliger, F. (1968) Crystal chemistry of the chalcogenides andpnictides of the transition elements. ,Struct Bonding, 4,83-229.

-----<1976) Structural Chemisty of Layer-Type pftases. ReidelPublishing Company, Boston.

Jellinek, F. (1963) Sulfides of the transition metals of Groups IV,Y, andYI. Arkiv Kemi, 20,447-48.

Kvacek, M. and Z. Trdlicka (1970) Die Rhenium- und Selenge-halte in einigen Molybd?initen der Bdhnischen Masse. ActaUniv. Carol. GeoL, 2, 105-l I l.

Mandarino, J. A. and R. J. Gait (1970) Molybdenite polytypes intle Royal Ontario Museum. Can. Mineral., 10,723-729.

Morgan, J. W., J. F. Lovering and R. J. Ford (1968) Rhenium andnon-radiogenic osmium in Australian molybdenites and othersulfide minerals by neutron activation analysis. f. Geol. Soc.Aust., 15,189-194.

Newberry, R. J. J. (1978) Polytypism in Molybdenite: Mineralogy,Geologlt, and Geochemistry. M.S. Thesis, Stanford University,Stanford. California.

Pauling, L. (1960) The Nature of the Chemical Bond,3rd. ed. Cor-nell University Press, Ithaca, New York.

Petruk, W. (1964) Mineralogy of the Mount Pleasant tin deposit,New Brunswick. Ottawa, Dept. Min. Tech. Sumey, Tech. Bull.56.

Petrulian, N. and L. Steclaci (1964) Sur la presence du rheniumdans les molyMenites de Roumanie et sotr importan@ cornmethermometre geologique. Rev. Roum. Geol. Geophys., Geogr.,Ser. Geol.,8,83-87.

Rode, E. Ya. and B. A. Lebedev (1961) A physiochemical study ofmolybdenum trisulfide and its thermal decomposition products.Zh. Neorgan. Khimii, 6,921-936 [transl. Rzss. l. Inorg. Chem.,6,608412 (1961).

Seebach, M. (1926) MolyMiinglanz. In C. Doelter and H. Leit-meier, Eds., Handbuch der Mineralchemie, Band IV, p. 67-74.Theoder Skinkop{, Dresden.

Somina, M. Ya. (1966) Trigonal molybdenite from East Siberiancarbonatites. Dokl. Akad. .iVazl< ̂SSSR, 162, g9g_901 [transl.Dokl. Akad. Naz& SSSR, Proc. Earth Sci Sect., 167, 96-99(1e66)1.

Somoano, R. B., V. Hadek and A. Rembaum (1973) Alkali metalintercalates of molybdenum disulfide. J. Chem. phys. 5g, 697_701.

Sutulov, A. (1976) Molybdenum and Menium. University of Con-cepcion Press, Concepcion" Chile.

Takeuchi, Y. and W. Nowacki (1963) Detailed crystal structure ofrhombohedral MoS2 and systematic deduction of possible poly-types of molybdenite. Schweiz. Mineral. petrogr. Mitt., 844,105-120.

Terada, K., S. Osaki, S. Ishihara and T. Kiba (1971) Distributionof rhenium in molybdenites from Japan. Geochem I., 4, l23-t4l.

Trigunayat, C. G. (1971) Progress in study of polyrypism of crys-tals (II). Phys. Stat. Sol. (a),4,281-303.

Masov, K. A. (1966) The Geochemistry of Rare Elements. lsraelProg. Sci. Transl., Jerusalem, Israel.

Wickrnan, F. E. and D. K. Smith (19?0) Molybdenite polytypes in

NEWBERRY: POLYTYPISM IN MOLYBDENITE 767

theory and occurrence. I. Theoretical considerations of poly-

typism in molybdenite. Am. Mineral.,55, 1843-1856.Wildervanck, J. C. and F. Jellinek (1964) Crystallinity of molyb-

denum and tungsten sulfides. Z. anorg. allgem. Chem., 328' 3O9-318.

Wilson, J. A. and A. D. Yoffe (1969) The transition metal dichal-cogenides: discussion and interpretation ofthe observed optical,electrical and structural properties. Adv. Phys., 18,194-325.

Zelikman, A. N., Yu. D. Christyakov, G. V. Idenbaum and O. E.Krein (1961) Crystal structure analysis of MoS2 synthesized byvarious methods. Kristallografya, 6,389-394 [transl. Sov. Pltys-Crystallogr., 4 308-3 l2 (1961).

(1969) Structural transformation in synthetic MoS2. Kristallo'grrf ya, I 4, 7 9 5-7 99 [transl. Sov. Phys - Cryst allo gr.' I 4, 687 -69 |(1e6e)1.

(19?6) Kinetics and mechanism of the growth of molybdenumdisulfide coatings on molybdenum. Izv. Akad. Nau& SS,SR'Neorg. Maler., 12, 1665-1667 ltransl. Inorg. Mater., 12' 1367-1370 (1976).

Znamenskiy, V. S., L. M. Lebedev and G. A. Sidorenko (1969)

Molybdenite from volcanic rocks of Quaternary age on lturupIsland, Kurile Islands, USSR. Dokl. Akad. Nauk SSS& 193,487489 [transl. Doftl. Akad. Nauk SSSIR, Proc. Earth Sci. Sect.'r93, 143-146 (1969)1.

Manuscript received, July 17, 1978;

acceptedfor publication, February 26, 1979.