Page images

covered. Mendelejeff predicted the properties of the three elements; he stated the atomic weight, spec. grav., general physical properties, and the formulae and chemical characters of the chief compounds, of each element. The descriptions given by Mendelejeff of the elements in question, several years before these elements were discovered, might almost be adopted now as descriptions of germanium, scandium, and gallium, so exactly in nearly every particular have they been realised.

There were two gaps in Group III., in Series 4 and 5, respectively. The differences between the values of the atomic weights of the elements in Series 2 and 4, in the various groups beginning with Group I., and of course omitting Group III., are 32, 31, 36, 37, 36, 36.5; hence, it was argued, in Group III. the difference will probably be about 33. The differences between the values of the atomic weights of the elements in Series 3 and 4 are 16, 16, 20, 20, 20, 19.5 ; hence, in Group III, the difference will probably be about 17. Boron, 11, occupies the position III.-1; now 11 +33 = 44. Aluminium, 27, occupies the position III.—3; now 27 + 17 = 44. Therefore, it was concluded that the atomic weight of the element which is to occupy the position III.—4 would be about 44. Similar reasoning led to the value 69 for the atomic weight of the element in III.–5.

The elements in Group III., when Mendelejeff's prediction was made, shewed a gradation of properties from the non-metallic boron to the distinctly metallic thallium ; boron was succeeded by the metal aluminium ; the elements of the group did not fall very distinctly into two families. One of the unknown elements would find a place in Series 4 succeeding the positive metals potassium and calcium, and followed by the elements titanium, vanadium, chromium, and manganese, all of which are metals but several shew decidedly negative functions: the other unknown element would find a place in Series 5, following the decidedly metallic elements copper and zinc, and followed by the metal-like non-metal arsenic, which is again followed by the non-metals selenion and bromine. The relations of the unknown element in III.- 4 to aluminium should, it was argued, be fairly similar to those of titanium to silicon, or of vanadium to phosphorus; the unknown element would probably less closely resemble aluminium than calcium resembles magnesium, or potassium resembles sodium ; but it would more closely resemble aluminium than vanadium resembles phosphorus, or chromium resembles sulphur; because when members of Series 3 and 4 are compared it is found that the resemblance is most marked in the lower members of the series.

The unknown element to be placed in III.-—-4 would shew analogies with boron ; therefore although it must be similar to aluminium, it probably would not form an alum. But the unknown element to be placed in III.— 5 must resemble aluminium more distinctly than the other unknown element does; therefore it probably would form an alum.

As calcium, which occupies in Group II. a position similar to that to be occupied by one of the unknown elements in Group III., is distinctly more positive than the first member of its own family (beryllium), but is very similar to the other members of its own family (strontium and barium), so probably would the unknown element to be placed in III.—4 closely resemble the succeeding even-series members of its group (yttrium, lanthanum, ytterbium).

The elements coming in Series 3, 5, and 7, are unlike each other in Group I., are similar but not very closely and intimately related in Group II., are very similar in Group V., and yet more similar in Groups VI. and VII.; therefore it was concluded that the unknown element in III.-5 would be distinctly similar to, but yet would shew differences from, both aluminium and indium.

Reasoning such as this guided Mendelejeff when he tabulated the properties of the elements scandium and gallium in Group III., and the element germanium in Group IV., while yet these elements were unknown.



The elements of Group VIII. are divided into three 483 sections. These elements are not placed in any of the ordinary series; but they find their places, one section, iron, nickel, and cobalt, between Series 4 and 5; another section, rhodium, ruthenium, and palladium, between Series 6 and 7 ; and the third section, iridium, osmium, and platinum, between Series 10 and 11.

It is probable that Group VIII. will some day be completed by the discovery of three elements to come between Series 8 and 9.

Copper, which is the first element of Series 5; silver, which is the first element of Series 7, and gold, which is the first element of Series 11, that is, the three elements which in order of atomic weights immediately succeed the respective sections of Group VIII., are sometimes placed in Group VIII.


(Section 1. Fe= 55.9 Ni= 58.6 Co= 59 Group VIII,

Section 2. Rh=104 Ru=1044 Pd=106.2

(Section 3. Ir=1925 Os=193 (?) Pt=194:3
Section 1.

COBALT. Atomic weights


59 The molecular weights of these elements are unknown. Sp. grs. (approx.)


8.6 Atom. weights


6.8 spec. graus. Sp. heats 114 •108

107 Melting points 1500-1600°


1400°-1500° (approx.)


Section 1.

Appearance, and Greyish-white; lus- White; lustrous; mal-
general physical trous; crystalline ; mal- leable; ductile; tena-
characters. leable; ductile; fair cious; hard; slightly

conductor of electricity; magnetic.
hard; magnetic.
Iron obtained by elec-
trolysis of FeCl, Aq is
said to be silver-white

and very soft.
Occurrence, and Found native but not in Metal is found in ine-
preparation. large quantities; oxides, teorites.

sulphides, carbonates, Chief ore contains ar-
&c. occur in enormous senide of Ni; sulphides,
quantities, and very silicates, &c. also occur,
widely distributed. not widely distributed,
Prepared by reducing but in considerable
Fe2O3 by Cat very high quantities.
temperatures; or by re- Prepared by reducing
ducing Fe,03 or FeCl2 Nio by C or H.
by H;

or by electrolysis

of FeCl2Aq.
General chemical Oxidised, chiefly to Slowly oxidised in
properties. Fe3O4, by strongly heat- moist air; oxidised by
ing in oxygen.

strongly heating in
Slowly oxidised by oxygen.
exposure to ordinary Dissolved by most
moist air.

Combines directly with Decomposes steam at
CI, Br, and I, also with S. red-heat.
Dissolved by most
Forms compounds re-
sembling alloys with C
and Si.
Decomposes steam at

Closely resembles







General formulae and characters of compounds. Nickel and cobalt very closely resemble each other in their chemical properties. The salts of cobalt are generally pink when hydrated and blue when anhydrous; the hydrated salts of nickel are usually green, and the anhydrous salts, yellow. Cobaltic chloride Co,Cl, form a large series of compounds with ammonia, e.g.

Co,C1,10NH,. 2H,0; Co,C1.10NH,; Co, Cl..12NH, These compounds resemble the chromium-ammonia compounds; corresponding nickel compounds are not known.

The cyanides of iron and cobalt form compounds with potassium cyanide of the forms K.M(CN)., and K,M(CN), (M= Fe or Co); the acids of which these compounds are salts, viz. H,M(CN)6 and H, M(CN). have been obtained. Nickel cyanide does not form a corresponding salt; the compound K Ni(CN), is known.

The only compound of the three metals the vapour density of which has been determined is Fe Clo; the valency of the atom of iron cannot be decisively determined from the composition of this molecule; the atom is probably tetravalent.

[ocr errors]



[ocr errors]





The formulae for the compounds of the three metals, with
the exception of the haloid coinpounds, are the simplest by
which their compositions can be expressed.

Oxides. MO, M,0%, M,0z: hydrates of all are known.
Sulphides. MS, MS.
Haloid compounds. M,X,, M. X..
Salts. Mx, M 3X ; X = SO, 2NO, PO,, &c.

,, ; Oxides. Ferrous oxide Feo® is difficult to prepare free 486 from ferric oxide, as it combines very rapidly with oxygen. The hydrated oxide FeO.H,0 is obtained by pptg. ferrous sulphate dissolved in air-free water with potash in absence of oxygen.

Nickelous oxide Nio, and cobaltous oxide Coo, are obtained by ppg. solutions of the corresponding salts by alkalis and heating the ppts. out of contact with air. These oxides combine with oxygen when carefully heated in air, forming the oxides M.O, which at a higher temperature are decomposed to MO and oxygen.

The protoxides, MO, dissolve in acids forming salts MX.

The oxides M,0, are formed by heating the oxides MO in air; Fe,0, is also obtained by adding an alkali to a hot mixture of ferrous and ferric sulphates (or other salts) in the ratio FeSO,: Fe (SO.)g: Ferroso-ferric oxide, FeO, interacts with acids to form both ferrous and ferric salts; e.g.

Fe, 0+4H_SO, Aq = FeSO Aq + Fe2(SO4)3Aq + 4H,0. The corresponding oxides of nickel and cobalt form nickelous salts only, and evolve oxygen, or chlorine if hydrochloric acid is used.

Ferric oxide Fe,O, is obtained by adding an alkali to a solution of a ferric salt, e.g. to Fe (SO.), Aq, and drying and heating the hydrated oxide, Fe,0,.3H,O, so obtained. This oxide interacts with acids to form ferric salts.

Nickelic oxide Ni,Og, and cobaltic oxide Co, O., are obtained by oxidising solutions of nickel or cobalt salts in presence of an alkali; e.g. by passing chlorine into potash containing Ni0.«HO or Coo..H. in suspension. These oxides dissolve in acids to form salts MX and evolve oxygen, or chlorine if hydrochloric acid is used; they are decomposed to MO and oxygen when heated in air.

When ferric oxide is heated with potash and a little bromine, or when very finely divided iron is heated with potassium nitrate, and the product is poured into water, a reddish







« PreviousContinue »