72.]

ARTIFICIAL PRODUCTION OF MINERALS.

129

iron furnaces reveals new artificial formations of this nature; and the number of those combinations, previously unattained by art, is gradually being diminished.

Ebelmen (Ann. Chim. Phys. 1848 [3], xxii. 211) succeeded in producing a variety of artificially crystallized compounds, which were before only known as natural minerals, by dissolving their constituents in boracic or in phosphoric acid, or in one of their salts, and then subjecting the mixture to an intense and longsustained heat in a furnace used for baking porcelain; the acid, or other compound employed as the solvent, was thus very slowly volatilized, and various minute crystals were obtained, including spinelle, chrome iron, emerald, and corundum or ruby.* Deville and Caron (Comptes Rendus, 1858, xlvi. 764) have extended these experiments. They introduced the fluorides of certain metals into a crucible lined with charcoal, and containing a quantity of boracic anhydride supported in a small cup of carbon. The cover of the crucible was then carefully luted on, and the whole exposed for an hour or two to an intense white heat. Under these circumstances the metallic fluoride and the boracic anhydride were slowly volatilized,the vapours decomposed each other, and crystals were formed. Ferric fluoride when thus treated yielded magnetic oxide of iron in octohedral crystals; zirconic fluoride yielded dendritic crystals of zirconia; a mixture of aluminic and glucinic fluorides furnished chrysoberyl (GO,Al,O3); aluminic fluoride mixed with zincic fluoride yielded crystals of gahnite (ZnO,Al,O); and, by the use of appropriate mixtures, staurolite and other crystallized bodies previously only known as native minerals were procured. The success that has attended these investigations offers every inducement, to those who have the opportunity, to pursue this interesting subject.

* Some recent experiments by Knop on titanic anhydride crystallized from microcosmic salt have shown that the compound thus obtained, and supposed to be anatase, is a phosphotitanic anhydride, 3 TiO, PO.

Few transformations of this kind are more instructive than those which accompany the formation of staurolite, the crystallized silicate of alumina (4 AL 0,, 3 SiO). Alternate layers of silica and alumina are placed in a porcelain tube heated to redness, and through this a current of silicic fluoride is transmitted, and escapes apparently quite unaltered at the other end of the tube. The fluoride attacks the ignited alumina, removing oxygen from it. This oxygen combines with a portion of the silicon of the fluoride, producing silica which unites with a portion of the alumina, 3 SiF, + 2A1,0 ̧ = 3 SiO ̧ + 2 Al ̧F. Meantime, aluminic fluoride being volatilized, passes on and attacks the superincumbent layer of ignited silica. Alumina is reproduced, and combines with a portion of the silica, whilst silicic fluoride is again formed, again to undergo a similar series of transformations, 2 Al,F ̧ + 3 SiO, = 3 SiF ̧ + 2 Al ̧ ̧.

130

SEPARATION OF SALTS BY CRYSTALLIZATION.

[72.

The prolonged action of water at high temperatures, such as can only be obtained under pressure, often furnishes crystalline compounds which cannot otherwise be procured. Acting upon a knowledge of this fact, De Senarmont, by operating in closed vessels with water upon various compounds, at temperatures ranging between 266° and 572° (130° and 300° C.), succeeded in obtaining in a crystallized condition the principal minerals which occur in metalliferous veins, including quartz, carbonates of iron, manganese, and zinc, baric sulphate, antimonious sulphide, mispickel, and red silver ore, as well as anhydrous ferric oxide, and corundum. (Ann. Chim. Phys. 1851 [3], xxxii. 129.)

2

The prolonged action of water at moderate temperatures may often also, as Daubrée (Comptes Rendus, 1858, xlvi. 1086) has shown, cause the formation of complicated crystallized minerals. It was found by this mineralogist that the zeolites apophyllite, chabasite [(KNaCaд)2O, Al ̧O, 4SiO2,6 H2O], and harmotome [K,O, 2CaO, 3Al2O3, 12SiO2,15 H2O], have gradually been produced in the concrete laid down by the Romans around the channels of outflow, at the hot springs of Plombières.* Tetrahedrite (grey copper) has been found at Bourbonne-les-Bains, Haute Marne, where it has been produced by the corrosion of copper coins by mineral water.

It is not in all cases necessary that liquefaction should take place as a preliminary to crystallization: a solid when deposited from the gaseous state sometimes furnishes crystals: iodine, arsenious anhydride, sulphur, mercuric iodide, and camphor, offer illustrations of this mode of crystallization. Some solids change their crystalline form entirely when left in the liquid from which they are deposited: thus benzamide when separating from its solution in boiling water forms a mass of fine needles; after a time a change takes place and solid crystals begin to grow amongst the needles, and ultimately the latter are entirely transformed into large solid transparent crystals.

(73) Separation of Salts by Crystallization. The process of crystallization from solution often affords a means of separating

* This concrete rests in part on granite and in part on alluvial gravel. The mineral water flows out at a temperature ranging between 140° and 158° (60° and 70° C.). It is very dilute, containing not more than 0.3 gramme per litre of saline matters, or 21 grains per gallon. It holds in solution small quantities of silica, and salts of sodium, potassium, calcium, and aluminium. These substances thus, in almost infinitesimal quantities, penetrate the concrete by a very slow infiltration, and in the lapse of ages have altered its composition, and deposited regularly crystallized zeolitic minerals.

74.]

SUDDEN CRYSTALLIZATION: NUCLEI.

131

two salts of unequal solubility, the crystalline form of which is different, and which have no chemical action on each other: nitre is thus purified from the common salt which always occurs mixed with it. This process is very generally resorted to as a means of purifying salts from small quantities of foreign admixtures, which may be soluble in water, but which either do not crystallize, or if they crystallize, do not do so in dilute solutions. Each crystallization diminishes the quantity of adhering impurity, and after the process has been repeated three or four times, dissolving each successive crop of crystals in fresh portions of pure water, the product will in most cases be free from impurity. The crystallization of sea salt from sea water thus separates the sodic chloride from magnesic chloride, and from various other salts which are present with it in small proportions; a single crystallization gives the salt sufficiently pure for commercial purposes, though it is in this state far from being chemically free from the bodies which accompany it in the waters of the ocean. A single crystallization of many salts, however, may be made to furnish the salt very nearly chemically pure, if the solution be briskly stirred whilst the crystals are being formed. The salt is thus deposited in minute detached grains; and if these are placed to drain, and washed with a saturated solution of the pure salt, as is practised in the refining of nitre (552), the mother liquor, which retains the impurities dissolved, may be completely washed away; but if the crystals be allowed to be deposited slowly and to acquire a large volume, the mother liquor is retained between the layers of each crystal, and cannot be thoroughly displaced by the pure solution. Bodies which possess the same crystalline form, such as potassic sulphate and chromate, cannot thus be separated from each other by crystallization.

(74) Sudden Crystallization: Nuclei.-Where the forces of cohesion and adhesion are nearly balanced, as in saturated solutions, very slight causes may occasion the cohesion to preponderate; and when once this force has been set in action, its influence spreads rapidly throughout the mass. Water, for example, in a still atmosphere, may be cooled 8 or 10 degrees below the freezing point, and yet continue liquid; but the slightest vibration of the vessel causes sudden crystallization of a portion of the liquid into ice. Sometimes a similar effect is produced, as in the case of Glauber's salt, by the sudden admission of air to the solution of the salt saturated at a high temperature, and from which the air has been expelled by boiling (55).

Adhesion to a solid body may be sufficient to disturb the

132

CAUSES MODIFYING CRYSTALLINE FORM.

[74.

balance; thus, the dropping in of a similar crystal, the insertion of a thread, or of a wire, or of a piece of stick, if not sufficient to cause sudden crystallization, will generally determine the spot upon which the crystals are first formed, especially if the foreign body or nucleus be rough and irregular in its outline. For this reason threads are stretched across the vessels in which the pure solution of sugar is set aside to crystallize in the manufacture of sugar-candy; so also wooden rods are placed in solutions of cupric acetate, and copper wires are suspended in solutions of borax in order to facilitate the crystallization of the salt.

(75) Circumstances which modify Crystalline Form.-The volume of crystals is often influenced by circumstances apparently trivial. Muddy solutions generally yield the largest crystals, as is well seen in the manufacturing process for obtaining citric and tartaric acids, where the impure acid always forms the finest crystals. Occasionally the presence in the liquid of a substance which does not crystallize with the salt, yet modifies the form which the latter assumes; urea, for instance, occasions the deposition of common salt in octohedra instead of its usual form of the cube. Potassic iodide generally crystallizes in cubes, but under certain conditions, and especially when the solution is quite pure, prismatic crystals are produced. Sodic carbonate usually crystallizes in large transparent solid crystals, but when quite pure the crystals are platy and not at all compact. It was also found by Jacobsen (Pogg. Annal. 1861, exiii. 498) that sodic chlorate, which crystallizes ordinarily in cubes, could be obtained at pleasure in the hemihedral form observed by Marbach, by touching one of the regular cubic crystals of the salt with a little fat or wax, and putting it back into the mother liquor. A similar modifying influence on the form of the crystal was produced by the addition of small quantities of urea, of glycerin, or of crude hydric potassic tartrate to the mother liquor.

The investigations of Pasteur (Ann. Chim. Phys. 1857 [3], xlix. 5) have thrown an interesting light upon some of the causes which thus operate in modifying the form of crystals. The crystals which were particularly examined by him were those of acid malate or bimalate of ammonium (HNH CH2O,), and of strontic formiate. Bimalate of ammonium, when it is deposited in the cold from a pure saturated solution of the salt, crystallizes in the form shown in No. 1, fig. 46-a form derived from a right prism with a rhombic base. Sometimes,

however, the crystals exhibit the double bevel shown in No. 2. When the salt is deposited from a solution containing products of the decomposition of the bimalate by heat, it assumes a hemihedral modification, similar to one or other of those shown in fig. 47. The bimalate has a ready cleavage parallel to the sides A B

76.]

CHANGE OF VOLUME IN CRYSTALLIZING.

133

and C D (fig. 46). If a crystal of the form of 1 or 2 be broken across, as in fig. 48, and be transferred to a portion of pure saturated mother liquor, the bevel

is rapidly restored: it makes its appearance first along the edges of the cleavage plane, x Y, and the growth of the crystal is more rapid in the direction perpendicular to the plane of cleavage than it is in the direction parallel with it. If the crystal be cut at one of its angles, as at a, fig. 49, the notch becomes rapidly filled up, and when the form of the crystal is restored, its growth again becomes regular in all directions.

The general conclusion to which these observations point is that when a broken crystal is replaced in its mother liquor, it continues to increase in every direction; but that its growth is especially active upon the broken surfaces, in consequence of which the general outline of the figure is restored in a few hours.

M

FIG. 49.

P

M

If a hemihedral crystal, such as either of those shown in fig. 47, be placed in a saturated solution of the pure bimalate, the hemihedral faces quickly disappear, as the artificial injury does. On the other hand, if perfect crystals be placed in a mother liquid depositing hemihedral crystals, the hemihedral form is speedily developed upon the newly-introduced crystals, the hemihedral crystal growing most rapidly in the direction of its length, A, B, whilst the regular crystal increases most rapidly in the direction of its breadth, A, c.

In reflecting upon this last observation, it occurred to Pasteur that if he could by mechanical means compel a crystal to increase more rapidly in length than in breadth, he might compel a pure solution to deposit hemihedral crystals. He accordingly pasted stripes of tinfoil over the sides of a well-formed crystal of the bimalate, and having produced cleavage planes at the two ends parallel to A, B, he placed it in a solution of the pure salt; on the following day the bevels had reappeared along the broken faces, and each of the four solid angles of the crystal exhibited a hemihedral face. When the tinfoil was pasted along one edge only of the crystal, the hemihedral faces were developed on that side only.

(76) Change of Volume in Crystallizing.-Some change of volume usually occurs at the moment of solidification; in many instances expansion takes place. Ice, for example, at the moment of congelation, increases in volume about, and expands so forcibly as to burst the vessel in which it is contained. Instances of this occur during severe frosts in the pipes used for conveying This expansive force is so enormous that no vessels have been found sufficiently strong to resist it. Major Williams (Trans. Roy. Soc. Edin. 1790, ii. 23) made some experiments at Quebec

water.