MINDRALOGICAL NOTES r4) /

(Busz, 1893),we recommendedthat kamarezitebe removedfrom the list of acceptedmineral species;this recommendationhas beenaccepted by a majority vote of the Commissionon New Minerals and Mineral Names, I.M.A. (Chairman of the Commission,written comm., 1964).

AcrNowlnoGMENTS The authors are grateful to Prof. Clifford Frondel of Harvard Univer- sity, Dr. Brian Masonl of The American Museum of Natural History, Dr. George Switzer of the U. S. National Museum, and Dr. Helmut Winkler2 of the University of Marburg, who made available to us speci- mens labeled "kamarezite." Appreciation is also expressedto the follow- ing: Prof. Neuhaus, for his assistancein attempting to locate the type specimenof kamarezite; Dr. Max Hey, for his helpful comments and suggestionsrelating to possibleinterpretations of discrepanciesbetween the analysesof kamareziteand brochantite; Dr. Alfred A. Moss, for the zinc analysison the British Museum'sspecimen of kamarezite;Mr. Owen Ramsburg,private mineral collectorof Washington,D. C., for the loan of his specimenof brochantite; and Harry J. Rose,Jr., our colleagueat the U. S. GeologicalSurvey, for x-ray fluorescenceanalyses.

RtrnnrNcBs

Busz, K. (1893) ["Kamarezit," ein neues Mineral.l Sitzbn. nieilerrhein. Ges. Bonn, 5O, 83-84. --- (1895) Mittheilungen iiber Caledonit, Kamarezit, Breithauptit und Magnetkies. Neues lalrb. Mineral l, 7ll-727. Fon-n, W. E. (1910) On some remarkable twins of atacamite and some other copper minerals from Collahurasi, Tarapaca, Chtle. Am. lour. Sci.30, 16-24. Lrw, A (1824) On the new mineral substance (brochantite). Ann. Phi.t.8,439-441. Per.ecnr, C. (1939) Brochantite. Am. Mineral.24, 463481. --, H. Bnnn.tN eNo C. FnoNou (1951) Dana's System oJ Mineralogy. Tth ed , Vol. II. John Wiley & Sons, Nern York.

r Present address: U S. National Museum, Washington, D.C. 2 Present address: Gdttingen Univeritiit, G6ttingen, Germany.

THE AMERICAN MINERALOGIST, VOL. 50, SEPTEMBER, 1965

MELANTERITE-ROZENITEEQUILIBRIUM Enxosr G. Enr.nns AND DAVrD V. Srrr-ns, Dept. oJ Mineralogy, The Ohio State (Jniaers,ity,Colwmbus, Ohio.

INtnoouctrox Samplesof alteredpyrite werecollected from an abandonedmine dump in the upper reachesof Sandy Run in Brown Township, Vinton County, Ohio, as part of a stlldy dealing with the oxidation of iron sulfides.The 1458 MINERALOGICALNOTES is from the Middle Kittanning (No. 6) Coal, which comprisespart of the Allegheny Formation of the PennsylvanianSystem. The pyrite sampleswere encrustedwith either massesof pale green fibers, extremely fine white powder, or a mixture of both. The fibrous material was determinedoptically to be melanterite, FeSO+'7HrO.The remaining material, as determined by standard r-ray powder diffrac- tometer techniques, proved to be mixtures of melanterite, , FeSOr.4HzO, and (Mg,Fe)Alz(SO4)4'22HrO.The r-ray spacings and intensities of melanterite and rozenite agreed almost com- ptetely with those given in a comprehensivestudl' by Jambor and Traili (1963).Magnesium halotrichite, obtained from an adjacent localitl', has recently been describedby Brant and Foster (1959).Rozenite, which is monoclinic, has been ciearly distinguished from triclinic siderotil (Fe, Cu)SO+'sHrOby Jambor and Traill. Their nomenclaturehas been approvedby the Commissionof Mineral Names,IMA. X-ray diffraction samples were prepared by grinding with an agate pestleand mortar. The fine powder was mixed to a slurry with acetone.A few drops of the siurry was removedwith an eye dropper, and allowedto evaporate on a glassslide. The sampleswere run on a Norelco high angle diffractometerusing Cu radiation with a Ni filter at 35 KV and 15 MA. Severalof these smear mounts were re-examinedby r-ray techniques severaldays after being collected,and it was discoveredthat the relative proportionsof the melanteriteand rozeniteshowed considerable change; the proportion of rozenitehad increased;a few da.vslater it was found that the proportion of melanteritehad increased.Midgley (1962), during a discussionof an occurrenceof rozenite (referredto as siderotil) stated that melanteriteis the stablehydrate at room temperatures.Jambor and Traill (1963)state that, "Contrarv to the statementof Midgley (1962,p. 408),the tetrahydrate (and not melanterite)is the stablephase at normal temperaturesand humidities." In view of the apparent nonstability of both melanterite and rozeniLein our samples,it was decidedto examine the equilibrium reiationsbetween these minerals under controlledcondi- tions. The pentahydrate of iron sulfate, siderotil, was synthesizedby Eckel (1933) from cuprian melanterite (pisanite) in an environment with less than 38/6 relative humidity at 22" C. Jambor and Traill (1963) who experimentallyexamined the stability of this mineral, concludedthat the pentahydrate will onll' form in an environment in which at least I.2/6 copper is present. Thus it was not expectedto enter into the present study.

ExpBnrlmNt.q.r,PnocBounp A device was constructedwhich allowed examination of a powdered sampie by r-ray diffraction at fixed relative humiditl' values. Air, ob- MINERALOGICAL NOTES 1459 tained from a standard line, was brought to a tee connection.From the tee, the air could be brought through either a two-foot tube filled with either Drierite or Linde Molecular Sievefor drying purposes,or bubbled through two 1000 cc cylinders of water for achieving maximum satura- tion. The relative humidity in the system was controlled by the propor- tion of air allowed to flow through the welting or drying sidesof the sys- tem. The wet and dry air flows were recombinedin a large bell jar. Wet and dry bulb thermometerswere placed within the jar to measurethe relative humidity. A rotating fan blade within the jar aided in the mixing of the wet and dry air, as weII as encouragingevaporation on the wet bulb thermometer. The accuracy of the humidity readings was checkedagainst a Serdexhumidity gauge,as well as completelysaturated air, and found to be * 2/6 above approximately 60/6 reiative humidity, grading down to + 5/6 at the minimum obtainable humidity ol 36/6. The 3616 mini- mum value appeared to be causedby the constant presenceof moisture from the wet bulb. A few runs weremade using almost completelydry air (which had beenrun through the Molecular Sieve)without the presence of the wet bulb. The air at known relative humidity was then brought to a Norelcohigh angler-ray diffractometerla pair of holeswere drilled in the centerof the flat side of the radiation shield, and the air was constrained to flow through one of these,directly over the smear slide sample.The narrow r-ray entrance port was covered with Scotch tape, which caused no noticeabledecrease in peak intensities.The air was allowed to escape through the secondhole. The samplesexamined by this techniquewere Fischer C. P. iron sul- fate, which initialiy consistedof FeSO+.7HrOas determined by r-ray diffraction.

ExpBnrlrBNrAL RESULTS

The FeSO+'7HrO,which was the starting compound used in this ex- periment, was completely converted to rozenite when subjectedto hu- midities, lessthan 70 8070.The convertedphaseswere in ever)'casetoo fine-grainedfor observationwith the polarizingmicroscope. The powderdiffraction pattern of rozenitepresented is virtuall,videnti- cal to the data presentedbv Jambor and Traill (1963)on rozenitefrom Manitoba. The stability- fields as seenin Figure 1 are a function not only of humidity but also of temperature. As would be expected, the field of stability of rozenite increasesat the expenseof melanterite with increas- ing temperature. Conversionof melanterite to rozenite,or rozenite to melanterite was usually accomplished within an hour, although in a few casesthree to 1460 MINERALOGICAL NOTES

oo Rozenife l. 20 / I Melonferite o,o,. /f phous

30 40 50 60 70 80 90 loo PERCENTRELATIVE HUMIDITY

Frc. 1. Stability fields of melanterite and rozenite at room temperatures as a function of relative humidity. The phases indicated have in all cases formed from conversion of other phases of difierent hydration states. The open circles indicate rozenite, filled circles melanterite, and crosses amorphous material.

four hours wererequired; this did not seemto be a function of the state of the particular sample, as duplicate runs were often performed on the samesample. Above approximately 95/6 relative humiditv, no diffraction pattern was obtained,as a result of deliquescenceof the melanterite. An attempt u'asmade to determinethe equiiibrium boundary between rozeniteand the mono-hydrateszomolnokite, by running completelydry air over a rozenite sample.A run of approximately eight hours did not causean) detectableconversion. DrscussroN The non appearanceof the pentahydrate siderotil in this study agrees with the conclusionof Jambor and Traill (1963) that at least a small amount of copperis necessaryfor its synthesis.It has beenclearly shown by reversiblereactions that either melanteriteand rozeniteare stable at room temperaturesas a function of the relative humidity. The diversestatements of Midgle.v,and Jambor and Traill on the sta- bility of melanterite and rozenite under "normal" conditions can now be viewed in proper perspective.Rozenite has been demonstratedto be the stable phase for humiditv conditions less than 70-8070, whereas melanteriteis stableat higher values.Midgley's work was donein Great Britain, which is known for its high relative humidity. Henceunder nor- mal British conditionsmelanterite is probably the stablephase for much MINERALOGICAL NOTES 146l of the year. Jambor and Traiil conductedtheir investigationsin Ottawa, Ontario, a mid-continentalregion of moderatehumidit-v, in which roze- nite is probably the stablephase for much of the year. Henceboth setsof investigatorsreached correct conclusionsfor their particuiar localities. The problem is to a great extent causedby the normal reluctanceof melanteriteto dehydrate under conditionsof low humidity. This metas- table persistenceis probably due to large grain size and lack of appre- ciable air movement over the sample.

AcrNowtorcMENTS The authors wish to thank the National Institute of Health for finan- cial support of this project under grant WP-00340, as well as Michael Fleischer of the U. S. Geological Survey for information on the proper nomenclature of the minerais examined.

RnlnnnNcrs

Bn,lxt, R. A. exo W. R. Fosron (1959) Magnesium halotrichite from Vinton Countyr Ohto. Ohio Jour. Sci,.59.187-192. Ecr

THE AMERICAN MINERALOGIST, VOL. 50, SEPTEMBER, 1965

SULFATEMINERALS: THEIR ORIGIN IN THE CENTRAL KENTUCKY KARSTI E. R. Ponr., CaaeResearch Found.ation, Yellow Springs, Ohio AND Wrr-r-r.q.uB. Wurro, Materials ResearchLaborotory anil Departmentof Geochemistryand, Mineralogy, The P enns ylv ania S tate (Jnia er s'ity, U n'h ers'ity Park, Pennsylaania.

Sulfate minerals, especially gypsum, are present in large quantities in most of the extensive limestone caves of the Mississippian plateaus and outliers of Centrai Kentucky. Many descriptions of these deposits have been published(Locke, 1842;Welier,1927; Huff, 1940);however,the origin of the depositshas remained obscure.

1 Contribution No. 8 of the Cave Research Foundation and Contribution No. 63-78 of the College of Mineral Industries, The Pennsylvania State University.