August 31, 1893] 



NATURE 



427 



was found in a collection of minerals where it ha J attracted n° 

 attention until Mr. Spang obtained the specimen and brought 

 it to the notice of Mr. Penfield ; the present specimen has 

 probably remained unnoticed in the Cornish collection at Caer- 

 hays for a large number of years. 



The local collection from which the American specimen was 

 obtained belonged to a man living near Tombstone, Arizona, 

 who had gathered together his minerals within a radius of about 

 two hundred miles, so that although the exact locality and 

 mode of occurrence are unknown, it is almost impossible that 

 this specimen can be also Cornish. 



From the typical character and appearance of the associated 

 clinoclase and lirocanite the British Museum specimen (although 

 no label or history is attached to it) can be pronounced to be 

 without the least doubt from the St. Day district, near Redruth, 

 in Gomwall. 



The American specimen is described as "a rounded mass of 

 impure cuprite which was mostly covered with hexagonal crystals 

 of spangoiite, associated with a few crystals of azurite and some 

 slender prism itic crystals of a copper mineral containing chlorine, 

 probably atacamite " ; it therefore differs considerably from the 

 Cornish specimen as regards the associated minerals. 



The only apparent difference between the spangoiite on the 

 two specimens is in the habit of the crystals, which in the 

 American mineral are short prisms with bevilled edges and a 

 large base, quite unlike the acute Cornish pyramids in a^pect. 

 , The pyramid angle found by Penfield is 53° 11 A', and the specific 

 'gravity 3 '141. Penfield further made some interesting observa- 

 tions concerning the etched figures of spangoiite; he describes 

 and figures ceitain beautiful triangular markings produced upon 

 I the basal plane by the solvent action of very dilute acids. We 

 jhave found that precisely the same characteristic figures are 

 .engraved upon a cleavage flake of the Cornish mineral when it 

 jis immersed for a few minutes in dilute acid. 

 y The American crystals attain considerable dimensions ; the 

 , largest had a length of 5i mm. and a breadth of 8 mm , and 

 by sacrificing half the specimen Penfield was able to obtain more 

 |than 3 grams (!) of pure material for analysis. The Cornish 

 crys-als are not more than 24 mm. in length and \ mm. in 

 toreadlh, and it will be difficult to obtain sufficient material for 

 |i complete analysis, unless other specimens can be found. This 

 i.s unfortunate, for the composition is so peculiar that, although 

 jPenfield's analysis is without doubt perfectly reliable, it would 

 nave been interesting to confirm his formula f-om a new locality. 

 ]rhe preliminary examination serves, however, not only to 

 jtstablish the identity of the mineral, but also to prove the m^st 

 mportant point — that the aluminium and chlorine are essential 

 ': 'onstiluents. 

 ; ' The formula deduced by Penfield is 



CueAlClSO,„.9HjO. 



'1 which, as he remarks, the aluminium is just suflficient to satisfy 

 I'le quantivalence of the total acids, thus : — 



, I (AlCI)S04.6Cu(HO).j.3H20. 



I The mineral is therefore closely related to connellite, a very 

 : |ire sulphate and chloride of copper also found in the St. Day 

 > lisirict, which, moreover, it somewhat resembles in appearance, 



laving the same black colour when viewed by reflected light 

 ; |one. The colour by tran<mitted light, together with the per- 



ict basal cleavage are, however, suflicient to distinguish 

 ' iiangolite from all kn )wn minerals ; fjrlher, the basal plane is 

 ; ! common on spangoiite as it is rare on connellite. 



It is to be hoped that search will be made among old col- 

 : 'ciions and upon copper ores from .St. Day for further specimens 



! this interesting mineral. H. A. Miers. 



I 



DESULPHURISA TION OF IRON. 



. I E elimination of sulphur from iron and the chemical re- 

 actions, whereby sulphur, in the presence of powerful basic 



' rjali, is removed from crude iron, hai recently attracted 

 leraVile attention. There are many reasons for this ; pure ores 

 become comparatively scarce, and to some extent the same 

 lie said of the fuel or coke used in the process of smelting, 

 irven if this be not strictly applicable in all districts where 

 iianufacture of iron is pursued, yetit cannot be gainsaid that 

 ^ive competition, with concurrent low prices, have had an 

 ince in rendering the strictest economy in the ma>ufacture 



VO. 1244, VOL. 48] 



absolutely necessary, and thus in a measure preventing the free 

 use of pure high-priced materials. 



I may even go further and assert that under favourable condi- 

 tions, that is, as regards general manufacturing expenses, local- 

 isation of plant, &c., the cost of pure good materials, 

 unquestionably suitable for smelting purposes, may become 

 quite prohibitive. In numerous instances manufacturers have 

 therefore been compelled to use cheaper fuel and ores falling 

 within the margin of economic working. At this point, how- 

 ever, other fresh difficulties have to be combated ; for when the 

 problem of the production of iron and steel at a reasonable rate 

 has been solved, it is too often found that the metal thus manu- 

 factured fails to meet demanded requirements. It is often the 

 case that when iron thus produced is converted into steel, a want 

 of uniformity in quality can be distinctly traced throughout the 

 manufactured product. Though the steel can hardly be termed 

 bad, nevertheless, as a general rule, it compares unfavourably 

 with the metal smelted from purer ores nith good fuel or 

 coke. 



The causes tending to the production of this inferior metal or 

 steel are well known, and may be summed up in a few words, 

 (i) The use of inferior coke in the blast furnace is at once a 

 cause of deterioration, for the heat is less intense, and this tends 

 to the production of a low grade iron. 



(2) it is evident that the use of inferior cheaper ore causes 

 a further deterioration in quality, whilst any attempt to remedy 

 this by lightening the furnace burden of ore — in other words, 

 using a greater quantity of coke — is, in many instances, counter- 

 balanced by the inevitable additional impurities charged, i.e. 

 sulphur and phosphorus, and other additional incombustible 

 matter or ash. 



It follows as a matter of course that the blast furnace can only 

 work in this direction within a very narrow limit, either plus or 

 minus attempts to limit the quantity of coke used resulting, as 

 before said, in the production of low grade iron. On the other 

 hand, an increased quantity of fuel with the use of inferior 

 ore increases the total amount of impurities. 



The working limit on either side is soon reached, and any 

 further attempts at improvement either way became simply use- 

 less. Certainly, very highly heated air or blast might to some 

 extent obviate some of the difficuliies, but as in modern practice 

 this is already thoroughly carried out, the employment of a 

 higher temperature of blast would appear to be practically im- 

 possible, and it is very likely that the attempted use of abnor- 

 mally heated blast or air would entail other serious practical 

 difficulties. 



This is the common experience of those engaged in the manu- 

 facture of iron and steel, more especially in blast furnace 

 smelling operations, showing that under the unfavourable con- 

 ditions before mentioned, it is practically impossible to produce 

 a high-class iron containing the minimum percentage of sulphur 

 and phosphorus together, with the requisite quanlitiis of silicon 

 and graphite necessary to ensure the production of good steel. 



Thanks to the pa'n;'.aking investigations of Mr. Stead, we can 

 now form a tolerably clear idea of the reactions involved in the 

 elimination of sulphur, both in the blast furnace and by other or 

 secondary processes. These may be broadly summed up in his 

 statement that sulphide of iron is dissolved out of the metal in 

 the first instance by free or loosely attached lime, in a highly re- 

 ducing atmosphere at a high temperature, as by the .Saniter 

 process, where lime dissolved in calcium chloride is used ; and 

 in the blast furnace by the excess of lime in solution in the slag, 

 or even a mixture of ordinary blast furnace slag and lime, the 

 latter being capable of eliminating sulphur from iron, and may 

 be substituted for Saniler's mixture. The results, however, so 

 far as can be ascertained, are somewhat irregular with either of 

 these methods. F'inally, there can be little doubt, as suggested by 

 Mr. Stead, that if lime alone is brought into intimate contact with 

 molten iron by suitable mechanical appliances, neither calcium 

 chloride nor slag is needed, these having little or no direct 

 chemical action on the metal, l.ut merely forming vehicles for the 

 transn.ission and mixing of the liine with the iron, and conse- 

 quent washing out of solution of iron sulphide, followed by the 

 subsequent conversiion into calcium sulphide and iron oxide. 



"The reactions in this process are, however, exceedingly com- 

 plex, and there are changes which occur of which we know 

 little or nothing. It is, however, my opinion that the sulphide 

 of iron is dissolved out of the metal in the first instance by the 

 free or loosely-attacheil dissolved lime ; but I do not care at 

 present, without more extended investigations, to hazard an 



