ON LEUCOPHOENICITES I. a Norr on Form Dnwlopuonrs Peur B

1226 MINERALOGICAL NOTES

RrlBnnNcrs

PAr.Acun, C; H. BnnueN, .tNn C. IinoNonr- (1951) Dana's System oJ Mineralogl,, vol II John Wiley and Sons, New York. RtcnuoNn, W. E. (1940) Crystal chemistry of the phosphates,arsenates, and vanadates of the type AzXOr (Z). Amer. Mineral.25, Ml-479 Srnusz, H (1957)MineralogischeTabellen. Akademische Verlagsgesellschaft, Leipzig. Srerlns, L. W. (1935) Austinite, a new arsenate mineral, from Gold llill, IJta"h.Amer. Mineral 20, 172-119.

AN MINERALOGIST, VOL 52, JULY AUGUST, 1967

ON LEUCOPHOENICITES I. A Norr oN FoRMDnwlopuoNrs Peur B. Moonn, Department of the GeophysicalSciences, University oJ Chicago,Chicago, Il,linois. INrnonucrroN Leucophoenicite,3MnrSiOu.Mn(OH)2, is generallyregarded as a member of the manganesehumite group. However, unlike the monoclinic members alleghanyite, 2MnzSiOq.Mn(OH)2, and sonolite, 4Mn2SiOa 'Mn(OH)2, which seemto be normal membersof the manganesehumite series,being isotypic with chondrodite and clinohumite respectivel)-, leucophoenicitedoes not erppearto have the orthorhombic symmetry expectedfor the manganeseanaiogue of humite. The detailedstudies on leucophoenicitemorphology by Palache(1928) led him to concludethat "No interpretation of the highly peculiarassemblage of forms offeredthe slightest resemblanceto the form seriesof an1' member of the humite group to which leucophoeniciteis related chemicalll"' (p. 316). Leucophoeniciteis one of the more conspicuousand abundant of the accessoryminerals from Franklin, New Jersel'l occurring as raspberry- red to pink massesusually associatedwith green willemite and coarse granular franklinite. In most specimensit appearsas a replacementmin- eral and Palacheconsidered it a pneumatolytic product, somehowcon- nected with the pegmatite Ienseswhich intruded certain portions of the ore body'.Crystals of leucophoenicite,however, appear to be confinedto openhydrothermal veins and disptaya wide variety of developmentsand a range of colorsfrom pink to red to brown. No structure-cellstudies on leucophoenicitehave been done to mr- knowledge,and this effort was undertaken to elucidatethe speciesas a

1 'hydrotephroites' I have discoveredthat many of the from pajsbreg, Swedenare leucophoenicites, r-hich adds interest to the species. MINDRALOGICALNOTES 1227 portion of work on manganeseolivines and humitesin general.This work turned out to be full of surprises!There appear to be no lessthan four kinds of leucophoenicites,distinguishable bv their powder patterns and structure cellsand to a lesserextent phl.sicalproperties and paragenesis. Briefly stated, the various leucophoenicitescan be divided crystallo- graphically into two categorieswhich I cail rn-leucophoenicitesand c'- Ieucophoenicites.The aa-leucophoeniciteshave pseudo-orthorhombiccells with monoclinic intensity relationships,the pseudo-orthorhombiccell translationsbeing closely analogous to humite but requiring a doublingof c. All singlecrystals of pink and red leucophoenicitesexamined so far be- Iong here. The o-leucophoenicitesare of three types: The orthorhombic isotypeof humite; the orthorhombicmember closely analogous to humite but with doubled c which is representedin the abundant massiverasp- 'hydrotephroites' berry-red material from Franklin and some from Pajs- berg; and orthorhombic memberswith unusually large identity i.ransla- tions along the c axis, normal to the assumedtephroite and pyrochroite layers. Some specimensshow manifestationsof semi-randomstructure' with streaks parallel to this translation on Weissenbergphotographs. Manl' of the so-called'h)'drotephroites' and platy brown leucophoeni citesbelong here. The paper concernsthe m-leucophoenicitecrystals of Palacheand a revision and discussionof their morphology. Regarding the cr1'stal chemistry of the o-leucophoenicites,the work is as yet incomplete and shall be reported later. In particular, their chemicalvariations must be assessedwhich at present are unknown variables.

Cnvsrer, Monpnor-ocv Palache(1928, p. 315) lists all the forms and their frequenciesobtained from the 15 difierent crystalshe examined.Taking the combinationsfor each crystal alone leads to the obvious monoclinic morphology for the mineral, but if all the forms for all crystalsare plotted collectivelywith- out regard to weighting on the basis of frequency, the relationship to humite is readily apparent.Figure 1 shows a gnomonicprojecl-i'cn of all the forms recordedwith D(010)polar to fulli' display the pseudo-orthorhombic character.Palache's reciprocal cell is indicateciby the ruled lines running NW-SE. The pseudo-humitecell chosenin this paper is indi- cated by the ruled lines running NE-SW. In effect,Palache's (100) has been rotated into his (201) position, a rotation of -13"16'' For the sake of clarity his form letters are retained and for the sakeof discussionthe polesof the forms are weightedas to frequencyby the circular areas'In Tabie 1, Perlache'sS2 values are trernsformedby the addition of {13"16', thus bringing his c(001) position to 90". My cell was chosenfrom the 1228 MINDRALOGICAL NOTDS

Frc 1. Gnomonic projection of rz leucophoenicitervith D(010)polar, the forms as citecl by Palache (1928). Palache's cell is denoted by hatch marks running Nw-sE. The cell in this study has hatch marks running NE-sw Frequencies of forms are depictecl by the circular areas.Palache's form letters are retained in this studr'. structure-cellratios oI a:b:c:2.2250:1:9.3250, 0:90o0,, obtainedfrom single crl.stal rotation and Weissenbergdata for one crystal and its powderdata, yielding a: 10.78i-.05A, t:4.845 +.004 A, .:+S.tSt-.tO A, space grotp 821/d. The structure cell chosenhere emphasizesthe pseudo-orthorhombicchareLcter of nr-leucophoeniciteand is related to humite cell dimensionsby a halving of c. The non-absentreflections for hUlhaveh*l:1n with h,l:2n andlor hkl,hit:2n. The spacegroup is ir reorientedP2r,/r.71 discoveryof the c-axistranslation twice that of humite prompts further investigationon humite minerals.In llable 1, the p2and dz valubs computed therefrom are shown to be in fair aqreement MINERALOGICAL NOTES

Tasr,B1. LrucopnonNrcrrr. Fonu RnvrsroN

Moore 4n(P)t pr(P)t Or(P*13"16') 4z(calc.)2 oz(calc.)z

c 001 001 76"44', 90000' 90"00' 90"00' 90000' 6 010 010 000 000 k 211 +10 - 13 48 61 05 -00 32 00 00 60 55 J 012 018 76 4+ 4t 36 90 00 90 00 40 38 o Oll 014 76 44 23 55 90 00 90 00 23 47 :r 103 lo7 46 22 90 00 59 38 59 06 90 00 t 102 10s 37 18 90 00 50 34 50 02 90 00 e 7Ol 103 22 42 90 00 35 58 35 36 90 00 o 100 101 00 00 90 00 t3 16 13 25 90 00 m llo 212 00 00 42 56 13 16 13 25 42 44 fl IZJ rr7 46 22 41 50 59 38 59 06 4L lr j r22 115 37t8 35 28 s0 34 50 02 34 s9 I tzl 113 22 42 29 13 35 58 35 36 28 56 s 120 111 00 00 24 56 13 16 13 25 24 48 pseudo-mirror image forms : z 104 107 -73 09 90 00 -59 53 -59 06 90 00 y 103 105 -64 15 90 00 -50 59 -50 02 90 00 i ro2 103 -49 37 90 00 -36 2l -35 36 90 00 r 101 101 -27 32 90 00 -t4 16 -13 25 90 00 n rrl 2r2 -27 32 43 03 -t4 16 -13 25 42 M q r24 1r7 -73 09 42 04 -59 53 -59 06 4l lr It t2il 115 -64 15 35 44 -50 59 -50.02 34 59 u 722 113 -49 56 29 26 -36 40 -35 36 28 56 n 121 111 -27 32 25 02 -r4 16 -13 25 24 48

I Palache's calculated data. 2 Based on structure-cell ratios (this studv). with Palache'stransformed data. It wasfound in the courseof powder X- ray studies on different leucophoenicitecrystals that the ratios may vary slightly and consequently the slight differencesbetween the transformed data of Palacheand the computeddata using structure-cellratios for one crystal reflect in part the differencesin axial ratios from crystal to crystal.

Drscussrox oN THEMouocr,rNrc Cuenecrpn op M-LEUCoPI{oENICITE In Table 2, lhe frequencyof forms arelisted both for the new (hkl) and their pseudo-mirrorimage forms (hkl). The great differencesin form fre- quenciesbetween l(105) and y(10.5),l(113) and u(ll3), and s(111)and n(Ill are readily apparent. Furthermore, the forms and their pseudo- mirror imagesonll' rarely seemto have similar dimensionon the crystals (seePalache, 1935, pp. 104-105).AIso, d (Ill) appearsto be a common form, whereasg (117) is rare. This is shown in Figure 1. Thesefacts are t230 MINERALOGICAL NOTDS

T l'lt-r.,2. LoucopnonNlctre.Fnrqurncy oF Fo RMS

Leucophoeniciter Humite2

ai bi c: 2.1605i | :4.4013 liace (pseudo- Moore F recluency Paired3 Form l.requency mirror image) symbol

c 001 11 C (001) 2 b 010 11 b (010) 3 n h 410 lr2 (410) 2 I 018 2 ez (014) 1 o 014 () e4 (o12) 2 r(z) to7 .5(4) 4 t\t) 105 4(7) 0 7,1 (20s) 1 e(i) 103 7(10) 7 t2 (20s) 2 a(r) 101 8(12) 6 X3 (2or) 3 m(.p) 212 2(2) 1 n) (2r1) 2 d(s') 117 6(2) 1 j(.h) 115 1(2) 0 t(u) I lJ 7(6) 1 s(n) 111 s(6) 2

I Fifteen crystals. 2 Four crystals. Not appearing on leucophoenicite: a(100)(1), m(210)(1), or(430)(1), e,(015)(2),4(or3)(2), ei(011)(4),a(2r3)(2), rr(4.1.10)(1), rz4l8)(2),r{41.6)(4),ra(4r4) (2), r5(ar\Q). 3 Number of crystals where both a form and its pseudo-mirror image appear. perhaps the most likely reasonswhy Palache could not recognizethe relationshipto humite. Table 2 also tabulates forms and their frequenciesfor humite. This tabulation is based on only four crystals, the ones depicted in Dana (1898,pp. 534-536),the celi therein requiring a simple doubling of.a to obtain the ratios 2.t605: 1 :4.4013, consistent with the structure-cell ratios of 2.158:1:4.400(Taylor and West, 1929).The indicesbased on theseratios require a doubling of hlor the forms listed in Dana and note that when comparingz-leucophoenicite with humite in Table 2, a mental doubling of I for humite is required. The similarity between m-leucophoeniciteand humite is obvious,though the common m-leucophoenicite forms (113) and (111) are missing in humite and the common humite forms (011) and (416) are missingin ar-leucophoenicite.

MonpuorocrcAr ANALysrsol M-LEUCopHoENrcrrE I'he apparentlypeculiar form developmentof rz-leucophoeniciteseems to be remarkably consistentwith the Bravais Law and the Donnay- MINDRALOGICAL NOTDS l23l

Tesro 3. Moxpnor,ocrcAr-ANervsrs or Lrucopnonxrcrto

Violations Symbol Frequency Violations Symbol Frequency

001 206 M(8) f oo2 a tt2 a 003 015 004 L(11) ll3 M(7) o 100 207 c 101 a 1,0,10 a toz 014 005 a 0,0,11 c 103 016 2 104 115 s(2) I 006 d 208 c 105 a 017 a 007 c 1,0, 11 106 a 116 008 s o,0, 12 c ro7 018 s(2) J 200 2W a 201 117 M(4) 202 L(10) 300 a 203 210 2 a 009 c 301 a 108 2tl a 204 212 s(2) b 010 302 a 011 a l,0, t2 012 u 213 a 205 c 303 a 013 a 019 109 a 0,0, 13 J 0,0, 10 a 118 014 M(e) 2,0,10 M(s) 110 tll M(s)

Code: L large M medium S Small a hhl hll*2n b 0k0h+2n c h0ll+2n d |fitr1,:2n,h*2,6,10 . . . e lfiIl:4n,h14,8,12.. . J hoo,ou,,h,l,+4n g already accounted for ? forms expectedto be present, but not observed.

Harker Principle. Roughly stated, the forms listed here though seemingly unusual, are a consequenceof the unusual cell shapeand the spacegroup derived from the orientation chosenin this paper. In Table 3, the inter- planar spacingsfor m-leucophoeniciteare tabulated in decreasingvalues. 1232 MIN]],RALOGICAL NOTES

Furthermore, the indicesin conflict with the glide conditions,the screw axis, and the B-face centeringare codedfrom a to f dependingon their extinction violation. Effectively, all coded indices should lose their rank and since the higher orders are also listed, they can be consideredelimi- nated from the table. Of the 63 symbolslisted, only 14 should appearas face forms, their relative degreeof frequency roughly correspondingto their position in the table. For this argument, it is assumedthat all fifteen of Palache'scrystals are nc-leucophoenicitewith space group 821/d. As the cell is orthogonal,the pseudo-mirrorimage forms neednot be con- sideredindependently: the frequencyof a form and its pseudo-mirrorimage is taken as the averageof the two. Of the list of spacings arranged as to decreasingreticular density, all but two of the fourteen observedforms-the pseudo-mirrorimage forms excluded-can be found: b(010)and ft(410)are missing.For the former, the screw axis necessitatesa higher order (020) and as for the latter, it may be considereda minor form on the basisof the observeddata. Even more noteworthy is that the arrangementof forms and their prominence almost faithfully obey the Bravais Law and the Donnay-Harker Prin- ciple. Within this series,only three forms are missing on the observed m- leucophoenicitecrystals, which are expectedto be present: (012), (016), and (210), but even then, the latter two are fairly well down the list. It appearsthat the form development for ra-leucophoeniciteis not so pecu- liar after all ! The really interesting problem which begs explanation is the monoclinic (though markedly pseudo-orthorhombic) symmetry of. m-leucophoeniciteand the existenceof leucophoenicite'polytypes', results which hopefully shall be reported in a future paper. Informative in this respect would be studies on the hypothetical manganesenorbergite, MnzSiO+ 'Mn(OH)2, but as yet no such compoundhas been discovered.

Acxnowr,nocunNrs This work was undertaken while I was an NSF Postdoctoral Fellolr,' at the Swedish Museum of Natural History. The museum and the Institute of Inorganic Chemistry, Uni- versity of Stockholm generously offered facilities and specimens. Professor Clifford Frondel loaned me Palache's precious crystals used in his morphological studies.

RrlnnrNcns

DeNe, E. S. (1898) The SyslemoJ Mineralogy, Sirth Edition. John Wiley & Sons,New York, 535 536. Par,acHn, C. (1928) Mineralogical notes on Franklin and Sterling Hill, New lersey. Amer. -329 M iner aI . 13, 297 , especially 3 15 . -- (1935) The Minerals oJ Fronklin and Sterling HilI, SussexCounty, New lersel'. U S Geol,.Su,yv. ProJ. Pap.l80, 104t-105. Tevron, W. H. aro J. Wrsr (1929)The structure of norbergite. Z. Kristall,ogr.7!,46l-474.