Article No : a07_281

Cobalt and Cobalt Compounds

JOHN DALLAS DONALDSON, Brunel University, Uxbridge, United Kingdom
DETMAR BEYERSMANN, University of Bremen, Bremen, Germany

1. Introduction . . . . . . . . . . . . . . . . . . . . . . . 429 6.5. Cobalt-containing High-Strength Steels . . 447

2. Physical Properties . . . . . . . . . . . . . . . . . . 431 6.6. Electrodeposited Alloys . . . . . . . . . . . . . . . 448
3. Occurrence . . . . . . . . . . . . . . . . . . . . . . . . 433 6.7. Alloys with Special Properties . . . . . . . . . . 448
4. Production . . . . . . . . . . . . . . . . . . . . . . . . . 435 7. Cobalt Compounds . . . . . . . . . . . . . . . . . . 449
4.1. Concentration of the Ores . . . . . . . . . . . . 435 7.1. Chemical Properties of Cobalt. . . . . . . . . . 449
4.2. Extraction . . . . . . . . . . . . . . . . . . . . . . . . . 436 7.2. Commercially Important Cobalt
4.2.1. Leaching of Cobalt Ores and Concentrates . . 436 Compounds . . . . . . . . . . . . . . . . . . . . . . . . 452
4.2.2. Separation of Cobalt from other Metal Ions in 7.3. Industrial Applications of Cobalt
Leach Solutions . . . . . . . . . . . . . . . . . . . . . 439 Compounds . . . . . . . . . . . . . . . . . . . . . . . . 455
4.2.3. Electrowinning . . . . . . . . . . . . . . . . . . . . . . 440 7.3.1. Glasses, Ceramics, and Refractories . . . . . . 455
5. Cobalt Powders . . . . . . . . . . . . . . . . . . . . . 440 7.3.2. Driers, Paints, Varnishes, and Dressings . . . 456
5.1. Production . . . . . . . . . . . . . . . . . . . . . . . . . 440 7.3.3. Catalysts . . . . . . . . . . . . . . . . . . . . . . . . . . 457
5.2. Uses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 441 7.3.4. Electroplating . . . . . . . . . . . . . . . . . . . . . . . 458
5.2.1. Cemented Carbides . . . . . . . . . . . . . . . . . . . 441 7.3.5. Electronics, Solid-State Devices, and Batteries 459
5.2.2. Cobalt Powders in Powder Metallurgy and 7.3.6. Agriculture, Nutrition, and Medicine . . . . . . 460
Cobalt-containing Metal . . . . . . . . . . . . . . . 441 8. Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . 461
6. Cobalt Alloys . . . . . . . . . . . . . . . . . . . . . . 442 9. Economic Aspects . . . . . . . . . . . . . . . . . . . 461
6.1. General . . . . . . . . . . . . . . . . . . . . . . . . . . . 442 10. Toxicology . . . . . . . . . . . . . . . . . . . . . . . . . 461
6.2. High Temperature Alloys (Superalloys) . . 442 References . . . . . . . . . . . . . . . . . . . . . . . . . 464
6.3. Magnetic Alloys . . . . . . . . . . . . . . . . . . . . 445
6.4. Hard Metal Alloys . . . . . . . . . . . . . . . . . . 447

1. Introduction [1–4] The g-rays emitted in the decay of 60Co have

energies of 1.17 and 1.33 MeV, and these taken
Cobalt, Co [7440-48-4], is a metallic element. with the 5.3-year half-life of the isotope provide a
Along with iron and nickel, it is one of the first- widely used source of radioactivity for use in
row transition-metal members of group VIII of food sterilization, radiography, and radiotherapy
the periodic table (old IUPAC recommenda- as an external source. The isotope is also used in
tion; according to the new IUPAC recommen- chemical and metallurgical analysis and in bio-
dation, cobalt belongs to group 9). Its electronic logical studies as a radioactive tracer. The 57Co
configuration is [Ar] 3 d7 4 s2, giving it an isotope decays by electron capture to give 57Fe,
atomic number of 27. The relative atomic mass the most widely used isotope in g-resonance
of cobalt is 58.9332. Only one of its isotopes, (M€ossbauer) spectroscopy [5].
59 Pure metallic cobalt has few applications, but
Co, is stable and occurs naturally; the other
12 known isotopes are radioactive and have its use as an alloying element and as a source of
the following mass numbers (half-lifes in pa- chemicals makes it a strategically important
rentheses): 54 (0.2 s), 55 (18.2 h), 56 (80 d), 57 metal. End uses of cobalt-containing alloys
(270 d), 58 (9 h), 58 (72 d), 60 (10.1 months), include superalloys for aircraft engines, magnet-
60 (5.3 years), 61 (99 months), 62 ic alloys for powerful permanent magnets, hard
(1.6 months), 62 (13.9 months), and 64 metal alloys for cutting-tool materials, cemen-
(< 28 s). ted carbides, wear-resistant alloys, corrosion-

2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

DOI: 10.1002/14356007.a07_281.pub2
430 Cobalt and Cobalt Compounds Vol. 9

resistant alloys, and electrodeposited alloys to G. BRANDT, a Swedish scientist, first isolated
provide wear and corrosion-resistant metal coat- cobalt metal as an impure sample by reducing
ings. Cobalt chemicals, among their many appli- an ore; in 1780 T. O. BERGMAN showed that the
cations, are used as pigments in the glass, cera- metal was in fact an element. The use of cobalt as
mics, and paint industries; as catalysts in the a metal dates from 1907, when E. HAYNES pat-
petroleum industry; as paint driers; and as trace ented a series of cobalt – chromium alloys
metal additives for agricultural and medical use. named Stellites that were the forerunners of
About 36 % of the worldwide annual production modern superalloys. In 1930 it was shown that
of cobalt is converted to chemicals, whereas addition of cobalt to certain alloys of iron, nickel,
high-temperature and magnetic alloys account and aluminum enhanced their properties as per-
for 41 % and 14 % of the consumption, respec- manent magnets.
tively. Detail on all aspects of the science and From the 16th to the 19th century, the world’s
technology of cobalt in general reviews [1–4] can supply of cobalt – mainly as smalt and zaffre –
be supplemented by reference to published ab- was produced in Norway, Sweden, Hungary, and
stracts [6], [7]. Saxony. Outside of Europe cobalt-containing
ores were worked in Burma from 1651 and in
History. Although very little cobalt metal New Caledonia from 1864. Major developments
was used until the 20th century, its ores have been in cobalt recovery in the 20th century have
used for thousands of years as blue coloring stemmed from the discovery of new ore bodies
agents for glass and pottery. Blue glazed pottery and improvements in the methods of winning
found in Egyptian tombs and dated at ca. 2600 the metal from ores. Cobalt-containing copper
B.C. has been found to contain cobalt, as have ores were discovered in Zaire in 1914, and
Persian glass beads dating from 2250 B.C. The extraction of cobalt from the ores began in
Portland vase in the British Museum provides 1924, whereas recovery of cobalt from pyrite
evidence that cobalt pigments were used by roasting residues was started in Germany in
Greek glassworkers around the beginning of the 1926. Mining of cobalt – nickel – gold – silver
Christian era. Cobalt-containing materials were ores began in Morocco in 1932, and production
also used to impart blue coloration to Chinese of cobalt from copper – cobalt ores began in
pottery during the Tang (600 – 900 A.D.) and Zambia in 1933. The Outokumpu copper –
Ming (1350 – 1650 A.D.) dynasties and to cobalt – zinc deposit was discovered in 1913
Venetian glass produced in the first part of the and has been exploited on a large scale since
15th century. The brilliant blue pigment used for 1928. In 1940 the International Nickel Co.
these purposes can be produced by fusing an ore (INCO) introduced a method of recovering co-
containing cobalt oxide with potash and silica to balt from nickel ores, and from 1952 to 1955 the
produce a vitreous material called smalt, which is production of cobalt by the Canadian compa-
powdered to produce the pigment. The secret of nies, Falconbridge Nickel Mines and the Sher-
making the pigment was apparently lost during ritt Gordon Mining Co., began. Developments
the Middle Ages and rediscovered during the include the commissioning in 1975 of a cobalt
15th century. LEONARDO DA VINCI was one of the refining plant based on a combination of pres-
first artists to use the rediscovered blue pigment sure leaching, solvent extraction, and electro-
when painting ‘‘The Madonna of the Rocks.’’ winning by the Sumitomo Metal Mining Co. in
In the 16th century, P. WEIDENHAMMER produced Japan, and centering of interest on Laterite ores,
a blue pigment, which he called zaffre, from particularly in Australia.
the silver – cobalt – bismuth – nickel – arsenate Worldwide production of cobalt in 1997 was
ores found in Saxony. The next development in just over 28 000 t, but this is expected to increase
the use of cobalt-containing materials was the to 34 000 t/a.
discovery, early in the 18th century, that solu- The production of cobalt is usually subsidiary
tions containing bismuth and cobalt could be to that of other metals, such as copper and nickel,
used as sympathetic inks. The invisible writing and its output cannot be significantly increased
obtained by using the inks became green when without a corresponding increase in the markets
heated, with the color change apparently due to of the primary metals. Another important factor
the presence of cobalt. It was not until 1735 that in the economics of the supply of cobalt is the fact
Vol. 9 Cobalt and Cobalt Compounds 431

that a large proportion of the world’s ore comes Table 1. Properties of the allotropic forms of cobalt
from one country, Zaire, and is therefore depen- Hexagonal e-form
dent on the political stability of that country. Cell dimensions
Cobalt is regarded as a strategic metal and is (ambient temperature) a ¼ 0.250 71 nm
stockpiled in some countries. c/a ¼ 1.6233
(417
C) a ¼ 0.2541 nm
The name cobalt is derived from the Greek c/a ¼ 1.631
word kóbaloV or the German word kobold for Density (20
C) 8.832 g/cm3
goblin or evil spirit. The term kobold was applied Co – Co interatomic distances 0.294 – 0.251 nm
by miners in the Harz mountains of Germany to Stacking fault energy
(20
C) 31107 J/cm2
an ore that gave no metal when smelted and that (370
C) 20.5107 J/cm2
also produced highly toxic fumes of arsenic
oxide. Cubic a-form
Cell dimension
(ambient temperature) a ¼ 0.35441 nm
(520
C) a ¼ 0.35688 nm
2. Physical Properties [1–4] (1398
C) a ¼ 0.36214 nm
Density
The physical and mechanical properties of cobalt (ambient temperature) 8.80 g/cm3
(1495
C) 8.18 g/cm3
metal are susceptible to variations because of Co – Co interatomic distance 0.251 nm
variations in the structures of metal samples Stacking fault energy
arising from a slow cubic to hexagonal phase (500
C) 13.5107 J/cm2
transformation. The data quoted in this sec- (710
C) 18.5107 J/cm2
tion are for the highest purity metal available;
details of the effects of impurities on the proper-
ties of cobalt are to be found in a review by
W. BETTERIDGE [8].
Cobalt exists in two allotropic modifications, Thermal Properties. The melting point of
a close-packed hexagonal e-form stable below cobalt is 1495  2
C, and its boiling point at
ca. 400
C and a face-centered cubic a-form normal pressure is 2800  50
C. The vapor
stable at high temperature. pressure of cobalt varies from less than 105 Pa
The transformation temperature is 421.5
C. (at 1250
C) to greater than 105 Pa (at 3200
C).
The free energy change associated with the The heats of fusion and vaporization of cobalt are
transformation is low, being ca. 500 J/mol for 17.2 and 425 kJ/mol, respectively. The thermal
the e ! a change and ca. 360 J/mol for the conductivity (l) of the metal falls steadily with
a ! e change; this accounts for the slowness increasing temperature; there is no marked
of the transformation. The mechanism of the change in conductivity at the transformation
phase change is martensitic and involves dislo- temperature, but the data show a minimum in
cation movements on the octahedral planes of the the region of the Curie temperature at 1121
C.
cubic lattice. Grain size affects the stability of the The values of l at 0, 100, and 800
C are 102, 85,
two allotropes, with finer grain size favoring the and 50 W m1, respectively. The heat capacity
cubic high-temperature form. For this reason the of cobalt rises steadily with increasing tempera-
fine grain sizes found in cobalt powders, sponge, ture. There is only a slight inflection in the cp vs. T
thin films, and fibers are responsible for the curve at the transformation temperature, but a
retention of essentially cubic structures down to large change does occur at the magnetic trans-
ambient temperature. The cubic form is also formation (see Magnetic Properties). The ther-
favored by the presence of a few percent of iron mal expansion coefficient of hexagonal cobalt
in the lattice. Table 1 shows the cell parameters depends on the orientation of the crystals and, for
and other properties of the allotropes. The density example, is 14.62  106 K1 along the h0001i
of the metal decreases with increasing tempera- and 10.96  106 K1 along the h1120i direc-
ture and shows an anomalous fall of ca. 0.15 % at tion (at 20
C). The average value for polycrys-
the transformation temperature. The liquid den- talline cobalt is 12.14  106 K1. The curve of
sity just above the melting point is 7.73 g/cm3, the coefficient versus temperature shows a sharp
and at 2000
C it is 7.26 g/cm3. change at the transformation temperature
432 Cobalt and Cobalt Compounds Vol. 9

because of the volume expansion of ca. 0.36 % range of 600 – 1000
C. Both elongation and
associated with the phase change from hexagonal tensile strength values show changes at the phase
to cubic crystals. Cubic cobalt has a higher mean transformation.
thermal expansion coefficient with values of The activation energy for creep is higher for
(14.2, 15.7, 16.0, and 16.8)  106 K1 at hexagonal cobalt than for the cubic form; the
200, 400, 600, and 750
C, respectively. values near the phase change temperature are 540
and 190 kJ/mol, respectively. The hexagonal
Mechanical Properties. The mechanical crystals have lower values of friction and wear
properties of cobalt are critically dependent on than the cubic crystals. The friction coefficient of
the purity of the metal and its thermal history. For cobalt rises steeply at temperatures above 300
C
well-annealed samples of high-purity cobalt, the to values of ca. 1.6 at 400
C. The low coefficient
Vickers hardness at normal temperature has va- associated with the hexagonal form is due to easy
lues between 140 and 160 N/mm2. The Vickers slip on the basal plane accompanied by the
hardness of annealed cobalt decreases with formation of a surface layer with the basal planes
increasing temperature to values of under oriented parallel to the surface. In actual applica-
50 N/mm2 at 750
C with no marked change at tions, however, the frictional and wear behavior
the phase transformation. Microhardness studies of the metal will be determined by surface oxide
on zone-refined single crystals of the metal show films.
variations with crystal orientation and give
values between 81 and 250 N/mm2. Electrode- Magnetic Properties. Pure cobalt has the
posited cobalt has high hardness (270 – 310 N/ highest known Curie temperature. Hexagonal
mm2), presumably because of its fine grain size. cobalt is ferromagnetic at all temperatures, but
Measurements of elastic properties on poly- the cubic form becomes paramagnetic at 1121 
crystalline bars of high-purity sintered metal 3
C. The magnetic properties of single crystals
annealed at 1000
C give a Young’s modulus of show marked anisotropy. For hexagonal crystals
211  103 MN/m and a shear modulus of at ambient temperature, the c-axis is the direction
82  103 MN/m at room temperature. The mod- of easiest magnetization, but with increasing
uli decrease with increasing temperature, with temperature magnetization becomes easier in a
the shear modulus showing an inflection and the direction perpendicular to the c-axis; however, at
Young’s modulus a hysteresis at the phase ca. 250
C the metal is magnetically isotropic. At
change. For single crystals the moduli do vary even higher temperature, the c-axis becomes the
with crystal direction, as shown in Table 2. most difficult to magnetize. Near the phase
Values of Poisson’s ratio for cobalt are found change temperature, the direction of easiest mag-
to be in the range of 0.29 – 0.32. netization of the cubic form is along the h111i
Vacuum-melted, hot-worked cobalt annealed cube diagonal. Differences between this and
at 800 – 1000
C has tensile strengths of 800 – other directions diminish with increasing tem-
875 MN/m2 and an elongation of 15 – 30 %. peratures until, at 1000
C, the cubic form is
Air-melted samples are much less ductile. Cobalt magnetically isotropic.
has a maximum ductility at 500
C, and this is the The magnetic properties of polycrystalline
best working temperature for the metal, although samples of cobalt depend on the purity of the
the vacuum-degassed metal free from lead, sul- metal and its thermal history; this applies partic-
fur, and zinc, can be hot-rolled in the temperature ularly to the development of preferred orientation.
Magnetic anisotropy can, for example, be devel-
oped by slowly cooling the metal over the phase
Table 2. Elastic moduli at 20
C for cobalt along different crystal
directions
transformation temperature range in a strong
magnetic field. Typical values for magnetic prop-
Crystal direction Young’s modulus, Sheer modulus, erty parameters are listed in Table 3. The mag-
103 MN/m2 103 MN/m2 netorestrictive behavior of cobalt is also strongly
1120 174 62.2 anisotropic and reaches a maximum value for
1010 175 62.2 hexagonal crystals of Dl/l ¼  27106 in a
1012 169 74.1 magnetic field of 560 kA/m along the h1120i
0001 213 62.4
direction. The volume magnetorestriction shows
Vol. 9 Cobalt and Cobalt Compounds 433

Table 3. Magnetic properties of cobalt Table 5. Optical emission lines of cobalt used in atomic emission
analysis


Curie temperature 1121 C
Maximum permeability 3.1104 H/m Wavelength, Relative intensity
Coercive force 707 A/m
Remanence 0.49 T nm Arc source Spark source
Saturation magnetization 1.79 T
352.9813 1000 30
346.5800 2000 25
a maximum contraction of 106 in a field of 345.3505 3000 200
480 kA/m and a magnetorestrictive expansion 340.5120 2000 150
251.9822 40 200
when the fields are above 760 kA/m. 238.8918 10 35
237.8622 25 50
Electrical Properties. The electrical resis- 236.3787 25 50
tivity and the temperature coefficient of the 230.7857 25 50
228.6156 40 300
resistivity of high-purity (99.999 %) polycrystal-
line cobalt are given in Table 4. The cubic form
has a lower resistivity than the hexagonal modi- lines in the range of 974 – 201 nm. The most
fication whenever the forms coexist. The resis- important lines for analytical purposes are listed
tivity of single crystals of cobalt shows a marked in Table 5. The characteristic K a1, K a2, and
dependence on orientation. K b1 X-radiation from a cobalt target occur at
0.1789, 0.1793, and 0.1621 nm, respectively; the
Other Properties. The self-diffusion of co- K absorption edge is at 0.1608 nm.
balt in the temperature range between 1057 and
1306
C follows an Arrhenius-type law, the
diffusion coefficient being given by the formula 3. Occurrence [1], [2], [4], [9]
2.2 (exp  295/RT). The rate of diffusion is very
dependent on grain size, being about 10 times Cobalt occurs in nature in a widespread but
faster for powder-metal samples than for bulk dispersed form in trace quantities in many rocks,
samples prepared by melting. soils, and plants. It is also found in sea water and
The optical reflectance of polished cobalt at in manganese-rich marine nodules. The cobalt
ambient temperature and near normal incidence content of the earth’s crust is about 20 mg/kg,
increases with wavelength from 60 % (at 1 mm) whereas its concentration in sea water has been
to 97 % (at 12 mm). The reflectance is much less reported as being 0.1 – 1 parts in 109. The con-
at high angles of incidence. The optical emission centration of the element in marine nodules is
spectrum of cobalt is complex and has over 330 usually 0.1 – 1 %. The largest concentrations of
cobalt are found in mafic and ultramafic igneous
Table 4. Electrical resistivity and temperature coefficient of resistivity rocks; the concentration of the element and the
for cobalt nickel – cobalt ratio decreases from ultramafic to
acidic rocks, as shown in Table 6. The nickel –
Temperature, Resistivity, Temperature
coefficient,
cobalt ratio changes because cobalt enters the


C 108 W m 1011 W m/K lattice of early crystallizing magnesium silicates
less readily than nickel. Sedimentary rocks
 100 2.65 –
 50 3.84 –
contain varying amounts of cobalt, with average
0 5.25 30
50 6.81 34 Table 6. Average cobalt content of igneous rocks
100 8.52 39
200 12.80 50 Rock type Cobalt content, Ni – Co
300 18.38 60 mg/kg ratio
400 24.72 71
500 30.65 75 Ultramafic 270 7
600 39.1 87 Gabbro 51 2.6
800 57.9 105 Basalt 41 2.5
1000 78.5 100 Diabase 31 2.3
1200 92.7 41 Intermediate igneous 14 1.9
1400 100.0 34 Felsic 5 1.1
434 Cobalt and Cobalt Compounds Vol. 9

values of 4, 6, and 40 mg/kg being reported for include the copper-containing carrollite, which is
sandstone, carbonate rocks, and clays or shales, one of the main sources of cobalt in Zaire,
respectively. During the formation of metamor- linnaeite found in Zaire, Zambia, and the United
phic rocks, very little movement or concentration States, and cattierite, which is also present in ores
of cobalt took place; hence, the levels of cobalt from Zaire. The arsenide ores include smaltite,
found in metamorphic rocks depend essentially which is found in the silver – copper ores from
on the amount of the element in the original Cobalt, Ontario and in Morocco, and skutteru-
igneous or sedimentary source. Those formed dite, the main cobalt mineral in Canadian and
from ultramafic or mafic sources contain an aver- Moroccan deposits. The sulfoarsenide, cobaltite,
age cobalt content of ca. 100 mg/kg, whereas is found in ore bodies in Zaire, Canada, and the
gneissic granites and metasedimentary rocks con- United States. The oxide mineral heterogenite is
tain an average of 16 and 8 mg/kg, respectively. a hydrated metal oxide containing varying
Under oxidizing conditions, cobalt shows a amounts of cobalt and copper and is one of the
strong tendency to concentrate with manganese main cobalt-bearing components of the Zaire
oxides. In the weathering process of mafic and deposits. The hydrated manganese – cobalt
ultramafic rocks to form laterites, nickel tended mineral, asbolite, is the source of most of the
to be leached downward with magnesia and cobalt in ores from New Caledonia.
silica, whereas cobalt with manganese oxides
was residually enriched near the surface of the Ore Deposits. Cobalt is produced mainly as
deposit. The concentration of cobalt and other a byproduct of the mining and processing of the
metals in marine nodules has been attributed to ores of other metals, particularly those of copper,
the strong ion-exchange properties of the submi- nickel, and silver, but also those of gold, lead, and
croscopic particles of colloidal manganese diox- zinc.
ide from which the nodules were formed. The deposits of cobalt can be classified under
Cobalt is a major constituent of about 70 the following headings:
minerals [10] and is a minor or trace constituent
of several hundred more, particularly those con- 1. Hypogene deposits associated with mafic
taining nickel, iron, and manganese. The miner- intrusive igneous rocks. The massive and dis-
als that have been mined or concentrated for their seminated iron – nickel – copper sulfides
cobalt content and those that are relatively high in containing cobalt are important examples of
cobalt are listed in Table 7. The sulfide minerals this type of deposit. They include the ore from
the Sudbury district of Ontario, which has an
average cobalt content of 0.07 %. The suite of
Table 7. Cobalt minerals
ore minerals consists of pyrrhotite, pentlandite,
Mineral CAS Cobalt pyrite, marcasite, cobaltite, and gersdorffite in
reg. no. content, % veins stringers and disseminated grains within
Cattierite, CoS2 (pure) [12017-06-0] 47.8 an igneous host rock of Precambrian age.
Linnaeite, Co3S4 (pure) [1308-08-3] 58.0 2. Contact metamorphic deposits associated
Siegenite, (Co,Ni)3S4 [12174-56-0] 20.4 – 26.0 with mafic rocks. Deposits of magnetite, chal-
Carrollite, (Co2Cu)S4 [12285-42-6] 35.2 – 36.0
Cobaltite, (Co,Fe)AsS [1303-15-7] 26.0 – 32.4
copyrite, and cobalt-containing pyrite,
Safflorite, (Co,Fe)As2 [12044-43-8] 13.0 – 18.6 formed by contact metamorphism of carbon-
Smaltite, (Ca, Ni)As2 [12006-41-6] ca. 21 ate rock by sills and dikes of diabase, gave the
Glaucodot, (Co,Fe)AsS [12198-14-0] 12.0 – 31.6 ore deposits at Cornwall and Morgantown in
Skutterudite, (Co,Fe)As3 [12196-91-7] 10.9 – 20.9
Heterogenite, CoO(OH) (pure) [12323-83-0] 64.1
the United States.
Asbolite [12413-71-7] 0.5 – 5.0 3. Lateritic Deposits. The weathering of perido-
Erythrite (CoNi)3(AsO4)2  8 H2O [149-32-6] 18.7 – 26.3 tite and serpentine generally gives laterite that
Gersdorffite, (Ni,Co)AsS [12255-11-7] (low) is rich in iron, nickel, cobalt, and chromium.
Pyrrhotite, (Fe,Ni,Co)x1Sx [1310-50-5] up to 1.00
Pentlandite, (Fe,Ni,Co)9S8 [53809-86-2] up to 1.50
Commercially valuable deposits contain 40 –
Pyrite, (Fe,Ni,Co)S2 [1309-36-0] up to 13.00 50 % iron, 1 – 2 % nickel, and 0.01 – 0.1 %
Sphalerite, Zn(Co)S [12169-28-7] up to 0.30 cobalt. Major lateritic deposits occur in Cuba,
Arsenopyrite, Fe(Co)AsS [1303-18-0] up to 0.38 New Caledonia, Australia, the United States,
Manganese oxide minerals 0.10 – 1.0
and Russia.
Vol. 9 Cobalt and Cobalt Compounds 435

4. Massive sulfide deposits in metamorphic bodies are laterites with only small quantities of
rocks, largely of volcanic sedimentary origin. nickel and cobalt that are costly to refine. Zaire
These deposits consist mainly of pyrite and produces more than half of the world’s cobalt
pyrrhotite and are mined in the United States. because its ores can be treated more profitably.
5. Hydrothermal deposits are subdivided into In addition to the land-based reserves listed in
two classes: vein deposits and replacement Table 8, almost 6109 t of copper is available in
deposits. Some of these deposits are the only marine nodules if it can be recovered and worked
ones that have been mined specifically as economically.
sources of cobalt. In Canada, veins that con-
tain as much as 10 % cobalt occur in the
Cobalt-Gowganda region of Ontario, whereas 4. Production
veins in the Bou-Azzer area of Morocco
contain an average of 1.2 % cobalt. Hydro- 4.1. Concentration of the Ores [1], [2],
thermal replacement deposits containing [9]
0.5 % cobalt as gersdorffite are found in
Burma and at Outokumpu in Finland; a cop- The first stages of the production of cobalt from
per-rich sulfide deposit contains 0.2 % cobalt its ores involve the separation of cobalt-bearing
mainly in linnaeite. minerals from the gangue and from minerals
6. Strata-bound deposits: the copper – cobalt containing other desirable metals but not cobalt.
deposits of Zaire and Zambia are of this type. The concentrates obtained by applying physical
They occur in folded shale and dolomite and separation methods such as gravity separation or
contain a number of minerals, including froth flotation to ores can increase the cobalt
chalcopyrite, bornite, chalcosite, linnaeite, content to 10 – 15 % from cobalt-rich ores.
and carroleite. The ore bed is 6 – 24 m thick However, in general, these processes only
and currently provides the world’s major increase the level of cobalt from 0.1 – 0.6 % in
source of cobalt. ores to a few percent. The concentration proce-
7. Deposits formed as chemical precipitates usu- dures for the main types of ores are described in
ally contain chemically precipitated cobalt in this section.
associated marine manganese nodules.
Arsenide Ores. Cobalt arsenide minerals
Reserves. As generally accepted, there are have specific gravities of 6.5 – 7.2 g/cm3 and
(3 – 5)  106 t of minable cobalt reserves and can be separated at relatively large grain size by
a further (4 – 5)  106 t of potential reserves. gravity methods. Concentrates containing 13 –
Table 8 lists the reported reserves in a number of 14 % cobalt have been produced by these meth-
countries. However, the minable reserves do not ods from Moroccan ores at Bou-Azzer contain-
have the same degree of profitability. Cuba has ing 2 % cobalt.
minable resources of cobalt, but provides only
5 % of the world production because its ore Sulfoarsenide Ores. The main mineral in
this type of ore, cobaltite, can be effectively
separated by froth flotation with xanthate at pH
Table 8. Cobalt reserves
4 – 5, although iron sulfide minerals are also
Country Reserves, t Country Reserves, t floated under similar conditions and must be
separated from the cobalt concentrate.
Australia 295 000 New Caledonia 385 000
Brazil >9 000 New Guinea 18 000
Burma 16 000 Philippines 159 000 Sulfide Ores. The chalococite and carrollite
Canada 250 000 Puerto Rico 68 000 fractions of the ores of the Shaba province in
Columbia 22 500 Solomon Islands 22 500
Zaire are floated together at pH 9.2 by using lime,
Cuba 1 048 500 former Soviet Union 181 500
Dominican Uganda 8 000 xanthates as collectors, and triethoxybutane as the
Republic 89 000 United States >764 000 frothing agent to give a concentrate containing
Finland 22 500 Venezuela 60 000 43 – 53 % copper and 0.5 – 3 % cobalt. The
Guatemala 45 500 Zaire 1 920 000
copper – cobalt sulfide ores of Zambia are first
Japan 2 500 Zambia 370 000
treated to float the copper minerals selectively at
436 Cobalt and Cobalt Compounds Vol. 9

pH 10.5 – 11; at this stage the presence of sodium (3) reduction of cobalt ions to metal. Electro-
cyanide depresses the cobalt minerals. These are metallurgical processes are used in the electro-
subsequently activated by sulfuric acid and winning of the metal from leach solutions and in
floated at pH 8.5 – 9 with sodium isopropyl refining the cobalt that has been extracted
xanthate; the concentrates contain up to 3.5 % by hydrometallurgical or pyrometallurgical
cobalt. The flotation process at the Outokumpu methods.
Keretti Mill first separates the copper minerals
with xanthates at pH 11 – 11.5 in the presence of
lime and sodium cyanide. The cobalt minerals 4.2.1. Leaching of Cobalt Ores and
are then floated at pH 9 to recover 65 – 75 % of Concentrates
the cobalt from an ore containing 0.23 – 0.31 %
cobalt; the concentrate contains 0.7 – 0.82 % Cobalt can be leached from its ores and concen-
cobalt. The nickel – copper – cobalt sulfide ores trates in both acidic and alkaline media. Process-
of Ontario are first treated to separate the sulfide es involving acid-sulfate leaching include (1)
minerals from gangue, and the cobalt minerals are treatment of oxide ore concentrates with sulfuric
then contained in the nickel sulfide concentrate. acid containing a reducing agent, (2) extraction
of cobalt sulfate from sulfide ores following an
Oxide Ores. Two methods have been used oxidizing roast, (3) extraction following a sulfa-
on a commercial scale to float oxide ores. The tizing roast, and (4) extraction by pressure leach-
palm oil process collects the concentrate in a 3 : 1 ing with sulfuric acid, often in the presence of
mixture of hydrolyzed palm oil and gas oil emul- oxygen; the Chemico process for the treatment of
sified in carbonated hot water; this process can arsenic-containing ores is, for example, based on
give concentrates containing 23 – 25 % copper an acid pressure leach. Acid leaching processes
and 2 – 3 % cobalt, representing a cobalt recov- based on hydrochloric acid extraction and chlor-
ery of 50 – 70 % from the ore. The sulfidization idation roasting have also been used.
process is used for mixed oxide – sulfide ores to Processes that involve the extraction of cobalt
float sulfide ores first and to concentrate oxide into alkaline solutions depend on the formation
minerals subsequently in a xanthate flotation. of the hexamminocobalt(II) species [Co
(NH3)6]2þ, which remains in solution for long
periods because of its kinetic stability, even
4.2. Extraction [1], [2], [9] though it has low thermodynamic stability.
Ammoniacal leaching has been successfully ap-
Cobalt can be extracted from concentrates and plied to ores after a reducing roast, and the
occasionally directly from the ore itself by hy- Sherritt Gordon process depends on ammonia
drometallurgical, pyrometallurgical, and electro- solution pressure leaching.
metallurgical processes. Although most methods
of extraction are based on hydrometallurgy, Cobalt from Copper – Cobalt Concen-
cobalt concentrates, mattes, and alloys have been trates. The copper – cobalt ores of Zaire and
reduced to metal by pyrometallurgical methods. Zambia are treated by a sulfatizing roast in a
Most arsenic-free cobalt concentrates can, fluidized-bed furnace to convert copper and co-
for example, be mixed with lime and coal and balt sulfides into soluble oxides and iron into
melted in a reducing atmosphere to give cobalt – insoluble hematite. The calcine is subsequently
copper – iron alloys. Two types of alloys are leached with sulfuric acid from the spent copper
obtained, a white alloy containing ca. 40 % cobalt recovery electrolyte. Oxide concentrates are in-
that is processed for cobalt and a red alloy troduced at this leaching step to maintain the acid
containing ca. 4 % cobalt that is processed for balance in the circuit. Iron and aluminum are
copper. Recovery of copper from the red alloy removed from the leach solution, and copper is
produces a cobalt-rich slag that can be recycled electrowon on copper cathodes. A part of the
with the ores. The hydrometallurgical processes spent electrolyte enters the cobalt recovery cir-
involve (1) leaching of concentrates to give a cuit and is purified by removal of iron, copper,
cobalt-containing solution, (2) separation of co- nickel, and zinc prior to precipitation of cobalt as
balt from the other metal ions in solution, and its hydroxide. In the final stages, this cobalt
Vol. 9 Cobalt and Cobalt Compounds 437

hydroxide is redissolved and the metal is refined diagram of the process used at the G
en
erale des
by electrolysis and by degassing to remove any Carrieres et des Mines (G
ecamines) plant in
traces of hydrogen. Figure 1 shows the flow Luilu.

Cobalt from Nickel Sulfide Concentrates.

Nickel sulfide concentrates can be treated by
either roasting or flash smelting to give a matte
from which nickel and cobalt can be recovered
hydrometallurgically, or they may be treated
directly by an ammonia solution pressure leach.
In the Sherritt Gordon process used at Fort
Saskatchewan in Canada, which is illustrated as
a flow diagram in Figure 2, a feed of matte and
sulfide concentrate, containing about 10 % nic-
kel, 2 % copper, 0.4 % cobalt, 33 % iron, and
30 % sulfur, is pressure leached in autoclaves at
83
C and 0.7 MPa (7 bar) with an ammoniacal
medium. This converts most of the sulfide via
thiosulfate and polythionate to sulfate and solu-
bilizes nickel, copper, and cobalt as ammine
complexes. Most of the copper in the leach
solution is precipitated as copper sulfides when
the solution is distilled to recycle ammonia. The
solution is treated with hydrogen sulfide to re-
move any residual copper, and any sulfide or
sulfamate remaining at this stage is converted to
sulfate by a pressure oxidation hydrolysis reac-
tion at 6.5 MPa (65 bar). The solution then en-
ters reduction autoclaves, where nickel powder is
precipitated by reduction with hydrogen at
3.6 MPa (36 bar). The remaining solution,
which contains 1 – 1.5 g/L of both cobalt and
nickel, is treated in an autoclave with hydrogen
sulfide to precipitate cobalt and nickel sulfides,
which are subsequently leached with sulfuric acid
at 140
C and 6.4 MPa (64 bar). Iron is removed
from this leach liquid at pH 2.5 – 3.0 by addition
of ammonia solution. The cobalt(II) in solution is
oxidized in air to give a very soluble cobalt(III)
pentammine complex. Nickel is subsequently
precipitated from the solution in two steps as a
nickel ammonium sulfate. Cobalt powder is then
added along with sulfuric acid to reduce the cobalt
(III) pentammine complex to cobalt(II), which is
finally reduced to cobalt powder by treatment
with hydrogen at 120
C and 4.6 MPa (46 bar).
Cobalt-containing mattes and mixed sulfide
concentrates have also been treated by pressure
leaching with sulfuric acid, for example, at Amax
at Port Nickel in the United States, MRR in South
Africa, Nippon Mining and Sumitomo Metals
Figure 1. Flow diagram of the G
ecamines process Mining Co. in Japan, and Outokumpu Oy in
438 Cobalt and Cobalt Compounds Vol. 9

Figure 3. Flow diagram of the Outokumpu process

medium used by Soci
et
e le Nickel is chlorine in

iron(III) chloride. At the INCO Copper Cliff
plant nickel – copper mattes are cooled slowly
and crushed prior to magnetic separation, which
Figure 2. Flow diagram of the Sherritt Gordon process isolates a nickel – copper – cobalt alloy. The
alloy is smelted in a top-blown rotary converter,
Finland. The flow diagram for the Outokumpu desulfurized by oxygen lancing, and granulated.
Kukkola plant (Fig. 3) is used to illustrate this Nickel is volatilized from the granules as
type of cobalt recovery process. Two companies, [Ni(CO)4] by the INCO carbonyl process to give
Falconbridge in Canada and Norway and Soci
et
e a residue from which cobalt can be leached; the
le Nickel in France, leach nickel mattes and solution obtained allows cobalt to be electrowon
sulfide concentrates in chloride media at ambient after some purification steps.
pressure. Falconbridge use a hydrochloric acid
leach at 70
C for about 15 % of their feed and a Cobalt from Laterite Ores. There has been
direct chlorine leach for the remainder. The leach an increasing interest in the recovery of cobalt
Vol. 9 Cobalt and Cobalt Compounds 439

from laterite ores. At least 150  106 t of laterite the cementation to metal is also utilized. For
reserves containing up to 15 % of cobalt have examples, see Section 4.2.1. The relative solu-
been identified in Australia alone. It is predicted bilities of cobalt(III) and nickel(II) hydroxides in
that this will lead to the production of about the presence and absence of complexing agents
10 000 t/a of cobalt in Australia and about and of cobalt(II) and nickel(II) carbonates in
3000 t/a from both Papua New Guinea and New solutions containing NH3 and CO2 have also
Caledonia. been used to separate these elements. Fractional
The typical laterite ore process involves a crystallization of chloride-containing media can
pressure sulfuric acid leach of slurried ore at be used to concentrate cobalt in solution and
about 250
C followed by neutralization and nickel in solid nickel chloride dihydrate.
clarification of the leach solution to give a preg-
nant liquor containing dissolved cobalt and nick- Physical Methods of Separation. Many
el at a pH of about 4.4. Extraction efficiencies for physical methods have been used to separate
cobalt and nickel are claimed to be grater than cobalt from other elements, but the three most
90 %. Cobalt is separated from the pregnant important are electrolysis (described in Section
liquor by solvent extraction prior to precipitation 4.2.3.), solvent extraction, and ion exchange.
as sulfide. The cobalt sulfide concentrate is then In most solvent extraction processes, cobalt is
refined, redissolved, and the metal extracted by extracted with a tertiary amine from chloride
electrowinning. solutions, although cobalt and nickel can be
separated from each other with phosphate or
Cobalt from Arsenide Ores. Arsenic-con- carboxylic acid reagents from ammoniacal sul-
taining concentrates are roasted in a fluidized bed fate solutions. In the sulfuric acid leach process
at 600 – 700
C to remove 60 – 70 % of the of Nippon Mining at Hitachi, solvent extraction
arsenic present as arsenic(III) oxide. The roasted is used to extract zinc from the leach solution
ores can be treated with hydrochloric acid and with diethylhexyl phosphonic acid (DEHPA) at
chlorine or with sulfuric acid to give a leach pH 2 – 3 prior to solvent extraction of cobalt with
solution that can be purified by hydrometallurgi- an alkylphosphonic acid. The Falconbridge and
cal methods and from which cobalt can be recov- Soci
et
e le Nickel processes both involve solvent
ered by electrolysis or by carbonate precipitation. extraction of cobalt from chloride solution with
triisooctylamine (TIOA). The Sumitomo Metals
Mining Co. uses a solvent extraction process to
4.2.2. Separation of Cobalt from other coextract cobalt and nickel with versatic acid.
Metal Ions in Leach Solutions [9] The extract is then stripped with hydrochloric
acid, and cobalt is reextracted from the chloride
A major part of cobalt recovery from leach solution with tri-n-octylamine (TNOA) to give
solutions is concerned with its separation from an extract from which cobalt is stripped and
other elements. Because of the similarities in electrowon. The flow diagram for this process
chemical behavior of cobalt and nickel, the sep- is shown in Figure 4, and the solvent extraction
aration of these two elements has been studied process can be expressed in terms of the follow-
extensively and is well known. This section deals ing chemical equations:
with chemical methods of separation of cobalt Extraction:
from nickel and with physicochemical methods 2 RHðorgÞþNiðCoÞSO4 ðaqÞþ2 NH4 OH!
of separation of cobalt from other elements.
R2 NiðCoÞðorgÞþðNH4 Þ2 SO4 ðaqÞþH2 O

Chemical Methods of Separation. The Stripping:

chemical methods of separating cobalt from
R2 NiðCoÞðorgÞþ2 HCl!2 RHðorgÞþNiðCoÞCl2 ðaqÞ
other metals use the different solubilities and
kinetic or thermodynamic stabilities of their
compounds. The most important compounds that Extraction:
have been used in separation are hydroxides,
NiðCoÞCl2 ðaqÞþ2 R13 NHClðorgÞ!
carbonates, ammine complexes, double ammo-
nium sulfates, sulfides, chlorides, and carbonyls; ðR13 NHÞ2 CoCl4 ðorgÞþNiCl2 ðaqÞ
440 Cobalt and Cobalt Compounds Vol. 9

Figure 4. Flow diagram of the Sumitomo process

Stripping: acid produced at the anode. Current densities of

5 A/dm2 are used with current consumptions of
ca. 6.5 kW h per kilogram of cobalt. Electrolyti-
cally refined cobalt is usually purer than 99.5 %.
where RH is versatic acid in kerosene and R31N is
tri-n-octylamine in xylene; (org) and (aq) are
organic and aqueous phases, respectively. 5. Cobalt Powders [4], [11], [12]
5.1. Production
4.2.3. Electrowinning [9]
Cobalt powder can be produced by a number of
Cobalt can be extracted and obtained as high- methods, but those of industrial importance in-
purity metal by electrolysis of sulfate or chloride volve the reduction of oxides, the pyrolysis of
solutions. Solutions of pH 1 – 4 are necessary for carboxylates, and the reduction of cobalt ions in
the electrowinning of cobalt, and the pH is often aqueous solution with hydrogen under pressure.
maintained by suspending cobalt hydroxide or Very pure cobalt powder is prepared by the
cobalt carbonate in the electrolyte to consume the decomposition of cobalt carbonyls.
Vol. 9 Cobalt and Cobalt Compounds 441

Reduction of Oxides. Gray cobalt(II) oxide tured as follows: pure cobalt powder is added to
(CoO) or black cobalt(II)/cobalt(III) oxide tungsten carbide (particle size ca. 1 mm), and the
(Co3O4) is reduced to metal powder with carbon mixture is ball-milled for a long period by using
monoxide or hydrogen. The reactions occur hard metal balls. The mixed powders are then
under conditions well below the melting point consolidated by cold pressing, by hot pressing at
of the oxides or the metal. The purity of the ca. 1300 – 1400
C, by cold extension, or by slip
powder obtained is 99.5 % with a particle size casting to produce small artifacts, large artifacts,
of ca. 4 mm, although the density and the particle constant-section rods, and complex-shaped arti-
size of the final product depend on the reduction cles, respectively. The consolidated parts are
conditions and on the particle size of the parent sintered in a reducing atmosphere at 1300 –
oxide. The finely powdered metal is stored at 1600
C. The sintering process involves the
very low temperature. formation of a liquid phase; the cobalt liquefies
at ca. 1320
C and dissolves tungsten and car-
Pyrolysis of Carboxylates. The thermal bon from the carbide. When the mixture is
decomposition of such cobalt carboxylates as cooled, most of the tungsten and carbon repre-
formate or oxalate in a controlled reducing or cipitates, but the binder phase is much stronger
neutral atmosphere produces a high-purity (about than pure cobalt because enough tungsten and
99.9 %), light, malleable cobalt powder with a carbon are present to stabilize the metal in its
particle size of ca. 1 mm that is particularly cubic form. During the sintering process, the
suitable for the manufacture of cemented car- consolidated powders shrink by 20 %. The bind-
bides. The particle size, form, and porosity of the ing mechanism depends on (1) the liquid metal
powder grains can be changed by altering the phase to wet the carbide particles so that surface
pyrolysis conditions. tension forces help densify the sintered phases
and (2) on the ability of the liquid to dissolve and
Reduction of Cobalt Ions in Solution. retain traces of carbide impurities.
Purified leach solutions containing cobalt pen- The properties of cemented carbides are con-
tammine complex ions can be treated in auto- trolled by the amount of cobalt present and the
claves with hydrogen under pressure and at high particle size of the carbide used. The hardness of
temperature to give an irregular chainlike form of the cemented carbides increases with decreasing
the powder that is suitable for consolidation by particle size and with increasing cobalt content.
direct strip rolling. The impact strength of cemented carbides is
directly proportional to the cobalt content; it rises
from less than 1 kg m for a sample with 6 %
5.2. Uses cobalt to ca. 2.5 kg m for a sample with 25 %
cobalt in tungsten carbide with a particle size of
5.2.1. Cemented Carbides [4], [13] 1.4 – 3.1 mm. The properties of cemented tung-
sten carbides can be improved by the addition of
One of the most important uses of metallic cobalt other carbides such as those of niobium, tanta-
is as a bonding agent in cemented carbides that are lum, or titanium.
used extensively as cutting tools for metals, rocks, Cobalt powders are also used as binders for
and other high-strength materials (! Hard Ma- the diamond tool industry [14].
terials). No suitable substitute has been found for
cobalt as a cementing agent for carbides. The
important properties of cobalt for this use are (1) 5.2.2. Cobalt Powders in Powder
high melting point, (2) high-temperature strength, Metallurgy and Cobalt-containing
(3) the ability to form a liquid phase with WC at Metal [13]
1275
C, (4) the ability to dissolve WC by the
formation of an eutectic containing 10 % of WC at Cobalt powders have been used in the formation
1275 – 1350
C, and (5) the ability to reprecipi- of a number of alloy phases including maraging
tate WC from solution as the mixture cools. The steel by hot extension of prealloyed powders,
most commonly used cemented carbide, tungsten cobalt-based superalloys by consolidation of
carbide containing 2 – 30 % cobalt, is manufac- powders involving liquid phase sintering,
442 Cobalt and Cobalt Compounds Vol. 9

fine-particle magnetic alloys, and bearing mate- Alloys with high-temperature stable hexagonal
rials impregnated with low-friction substances lattices (temperature in parentheses) are cobalt –
such as graphite, lead, nylon, and molybdenum chromium, 90 : 10 (800
C); cobalt – molybde-
disulfide. num, 90 : 10 (950
C); cobalt – osmium, 60 : 40
(1250
C); cobalt – platinum, 80 : 20 (650
C);
cobalt – rhenium, 80 : 20 (750 – 950
C);
6. Cobalt Alloys [1–4], [9] cobalt –ruthenium, 70 : 30 (1100 – 1400
C);
cobalt – tungsten, 80 : 20 (950
C).
6.1. General

About 64 % of the cobalt is consumed in alloy 6.2. High Temperature Alloys

compositions. The most important cobalt alloys (Superalloys) [13], [15]
are discussed in this article under the following
headings: Cobalt-based superalloys and cobalt-containing
nickel superalloys are used in aircraft engines as
. high-temperature alloys turbine vanes and blades and in sheet form for
. magnetic alloys exhaust case assemblies and burner liners.
. hard metal alloys Cobalt-based superalloys are complex combi-
. cobalt-containing high-strength steels nations of elements intended to accomplish one
. electrodeposited alloys or more specific purposes. Normally the matrix
. alloys with special properties is face-centered cubic. Commercial alloys are
commonly hardened by carbide formation and
Solute metals in cobalt can be classified in by solid-solution strengthening. The properties
terms of their effect on the phase transition (cubic responsible for the industrial importance of
to hexagonal) of the host metal. Those elements carbide-strengthened alloys are high creep
that raise the transition temperature are classified strength at high temperature, high thermal
as restricted-field components, whereas those fatigue resistance, good oxidation resistance,
that lower the temperature are enlarged-field excellent hot corrosion resistance, an air melt-
components. Some alloying elements must be ing capability, an air or argon remelting capa-
classified as combined components because they bility, good welding properties, and ease of
both raise the e ! a transition temperature and casting. The effects of added elements on the
lower the a ! e transition temperature. Table 9 properties of superalloys are listed in Table 10.
lists some alloying elements according to their A number of elements are detrimental to the
classification. The phase diagrams of a number of alloy properties. Sulfur, for example, affects the
binary and ternary cobalt alloy systems have ductility and must be kept at a concentration of
been reported [1], [3]. According to the phase less than 0.006 %. Other elements that have been
diagrams, a number of binary alloys have stable reported to show deleterious effects are bismuth,
hexagonal structures at high temperature and lead, antimony, thallium, selenium, cadmium,
these may be of practical importance where the mercury, tin, and silver. The cobalt-based alloys
hexagonal structure is required, for example, to consist essentially of a cobalt – chromium matrix
maintain low friction and wear characteristics. with ca. 20 % chromium, which provides major
resistance to high-temperature corrosion. Up to
Table 9. Classification of alloying elements in cobalt-rich binary 30 % of the cobalt can be replaced with nickel,
alloys and up to 15 % tungsten may be added as well as
smaller levels of other beneficial metals. The
Enlarged-field Restricted-field Combined
carbon content lies in the range of 0.25 – 1 %
Al Nb As Re Au for casting alloys and 0.05 – 0.4 % for hot
B Ni Cr Rh Be
C Sn Ge Ru Ga
worked alloys. Table 11 lists the compositions
Cu Ti Ir Sb Pb of several cobalt-based high-temperature alloys,
Fe Zr Mo Si Pd indicating the alloys that are normally cast; Table
Mn Os Ta V 12 shows the stress rupture strengths for some
Pt W
cobalt-based superalloys.
Vol. 9 Cobalt and Cobalt Compounds 443

Table 10. Effects of several elements in cobalt superalloys

Element Effect*

Chromium improves oxidation and hot-corrosion resistance; produces strengthening by formation of M7C3
and M23C6 carbides
Molbydenum and Tungsten solid-solution strengtheners; produce strengthening by formation of intermetallic compound Co3M;
formation of M6C carbide
Tantalum and solid-solution strengtheners; produce strengthening by formation of intermetallic
Niobium compound Co3M and MC carbide; formation of M6C carbide
Aluminum improves oxidation resistance; formation of intermetallic compound CoAl
Titanium produces strengthening by formation of MC carbide and intermetallic compound Co3Ti; with
sufficient nickel produces strengthening by formation of intermetallic compound Ni3Ti
Nickel stabilizes the face-centered cubic form of matrix; produces strengthening by formation of intermetallic
compound Ni3Ti; improves forgeability
Boron and Zirconium produce strengthening by effect on grain boundaries and by precipitate formation; zirconium produces
strengthening by formation of MC carbide
Carbon produces strengthening by formation of MC, M7C3, M23C6, and possibly M6C carbides
Yttrium and Lanthanum increase oxidation resistance
*
M ¼ metal.

The presence of carbides in the grain bound- pinning action and (2) the formation of a grain
aries of superalloys is important in determining boundary skeletal network that can support load
their mechanical properties. The grain bound- as well as hinder sliding. Both mechanisms also
aries are strengthened by carbides in one of two strengthen by limiting grain-boundary migration.
ways: (1) their precipitation as discrete particles Solid-solution strengthening is principally due to
in the grain boundaries hinders sliding by a contents of chromium, tungsten, niobium, and

Table 11. Composition of superalloys

Element

Trade name C Cr Ni Mo W Nb Ta Ti B Zr Fe Other

*
(C) HS-31 (X-40) 0.5 25.5 10.5 – 7.5 – – – 0.01 – 2 –
(C) HS-21 (Mod.
Vitallium) 0.25 27 3 5 – – – – – – 1 –
S-816 0.38 20 20 4 4 4 – – – – 4 –
HA-25 (L-605) 0.1 20 10 – 15 – – – – – – –
(C) HE-1049 0.4 26 10 – 15 – – – 0.4 – 3 –
UMCo-50 0.1 28 – – – – – – – – 21 –
(C) ML-1700 0.2 25 – – 15 – – – 0.4 – – –
J-1570 0.2 20 28 – 7 – – 4 – – 2 –
J-1650 0.2 19 27 – 12 – 2 3.8 0.02 – – –
(C) WI-52 0.45 21 1 – 11 2 – – – – 2 –
(C) HA-151 0.5 20 – – 12.7 – – – 0.05 – – –
(C) MAR-M 302 0.85 21.5 – – 10 – 9 – 0.005 0.2 – –
(C) MAR-M 322 1.0 21.5 – – 9 – 4.5 0.75 – 2.25 – –
(C) X-45 0.25 25.5 10.5 – 7.5 – – – 0.01 – 2 –
NASA-CoWRe 0.4 3 – – 25 – – 1 – 1 – 2 RE**
(C) AiResist 13 0.45 21 1 – 11 2 – – – – 2.5 3.5 Al; 0.1 Y
(C) MAR-M 509 0.6 24 10 – 7 – 3.5 0.2 – 0.5 – –
UMCo-51 0.3 28 – – – 2.1 – – – – 19 –
AiResist 213 0.18 19 – – 4.7 – 6.5 – – 0.15 – 3.5 Al; 0.1 Y
MAR-M 918 0.05 20 20 – – – 7.5 – – 0.1 – –
HA-188 0.1 22 22 – 14 – – – – – 1.5 0.08 La
(C) AiResist 215 0.35 19 – – 4.5 – 7.5 – – 0.13 – 4.3 Al; 0.17 Y
CM-7 0.1 20 15 – 15 – – 1.3 – – – 0.5 Al
(C) FSX-414 0.35 29.5 10.5 – 7 – – – 0.01 – 2 –
*
C ¼ cast alloy.
**
RE ¼ rare earth.
444 Cobalt and Cobalt Compounds Vol. 9

Table 12. Characteristic rupture strengths of selected cobalt- and nickel-based superalloys

Characteristic rupture strength, MN/m2

815
C 870
C 980
C 1095
C

Trade name 100 h 1000 h 100 h 1000 h 100 h 1000 h 100 h 1000 h

(C)* HS-21 (Mod. Vitallium) 152 98 115 91 65 48 – –

(C) HS-31 (X-40) 179 138 134 103 76 55 – –
(C) X-45 131 103 96 69 48 31 – –
(C) FSX-414 152 117 110 83 55 34 21 –
(C) WI-52 (HA-152) – – 172 152 96 76 34 –
(C) HA-151 255 228 186 165 96 79 – –
S-816 172 124 107 69 – – – –
(C) S-816 þ 1 % B 301 223 206 145 100 54 – –
J-1570 228 165 158 110 – – – –
J-1650 317 228 220 145 90 – – –
(C) MAR-M 302 276 207 207 158 110 76 41 28
(C) MAR-M 322 276 193 228 158 138 103 69 55
(C) MAR-M 509 269 228 200 138 117 90 55 38
HA-25 (L-605) 165 117 107 72 48 26 – –
HA-188 154 110 105 70 41 25 15 –
MAR-M 918 207 138 110 76 41 22 17 –
Nimonic 80 A 193 116 – – – – – –
Nimonic 90 239 154 139 77 – – – –
Nimonic 105 324 234 139 134 68 32 – –
Nimonic 115 402 309 301 201 121 70 – –
Udimet 700 400 296 290 200 110 52 – –
(C) B-1900 503 379 386 255 179 106 – –
(C) MAR-M 200 524 413 400 290 186 131 76 45
Hastelloy X 107 69 69 48 36 21 – –
Rene 41 310 200 193 117 69 – – –
*
C ¼ cast alloy.

tantalum, not to the precipitation of carbides. ence of cobalt arises from several effects,
Table 13 lists the strengthening mechanisms in namely:
cobalt-based superalloys.
The important properties of the alloys for 1. Its relatively minor contribution to solid-
high-temperature use are time-dependent creep, solution hardening.
stress rupture, or fatigue strength at the operat- 2. Its influence on the solubility of the matrix for
ing temperature. Properties such as tensile titanium and aluminum. At temperatures be-
strength, hardness, and modulus are not impor- low 1080
C, the cobalt-containing alloys
tant for high-temperature operation, but do have have a lower solubility for these metals than
an effect on manufacturing operations involved the cobalt-free alloys and, therefore, increase
in the production of parts. The oxidation of the amount of Ni3(TiAl) precipitated at a given
uncoated cobalt-based superalloys can be quite level of alloying; this is a major factor in giving
severe, and protective coatings are required increased high-temperature strength.
when the alloys are used in hot sections of 3. Its effect on the formation of carbides in the
turbines in jet engines. grain boundary. The presence of cobalt in-
creases the solubility of carbon in the matrix
Cobalt-Containing Nickel-Based Alloys. and modifies the carbides formed in the grain
Most highly stressed components in high-tem- boundary.
perature engineering use are nickel-based super- 4. It improves the hot workability.
alloys strengthened by the precipitation of the
intermetallic compound Ni3(TiAl). Addition of The presence of cobalt in nickel-based alloys
ca. 20 % cobalt improves the high-temperature has a pronounced effect in decreasing the stack-
properties of the nickel-based alloys. The influ- ing fault energy, which results in a decrease in the
Vol. 9 Cobalt and Cobalt Compounds 445

Table 13. Strengthening mechanisms in cobalt-based superalloys

Mechanism Comments

Carbide formation effective as strengthening agents but overaging a problem; optimum distribution not
(MC, M6C, M23C6, M7C3) always achieved

Solid-solution hardening effective at temperatures above ca. 1095
C, but not potent below this temperature
(Ta, W, Mo, Nb)

Intermetallics overaging above  870
C; possible deterioration of ductility and oxidation resistance due
(Co3W, Co3Mo, Co3Ti, Co3Ta, s-, m-, to s, m and Laves; appreciable hardening can be achieved below 815
C
or Laves phases)

Grain structure (directional, single, fine) directional solidification yields better thermal fatigue and ductility, but higher
cost because of special processing; fine-grained more ductile, possible better
fatigue but poorer creep

Dispersoids (ThO2, other oxides, carbides) expensive, technique very critical, no outstanding alloy yet developed

Minor elements (B, Zr) fairly well explored, effective; excessive boron must be avoided due to low ductility
and incipient melting

Eutectics (intermetallics) requires further investigation, offers promise of high-temperature stability

and good strength

Face-centered cubic to hexagonal close field requires much more investigation before assessment can be made, question of stability
packing
transformation and dislocation structure

Fiber reinforcement not enough information developed as yet

Combinations of the above adds to expense so that properties developed must be improved enough for
economic justification

creep rate of nickel or nickel – chromium Nickel – Iron – Aluminum – Cobalt Alloys
phases. (Alnico Alloys). Nickel – iron – aluminum
alloys (see also ! Magnetic Materials) have
a high coercive force, but are very brittle and
6.3. Magnetic Alloys [1], [2], [4], [16], must be cast into their final shape. Addition of
[17] cobalt to the alloys greatly improves their
magnetic properties. The gains are progressive
Cobalt is used in the manufacture of three main with increasing cobalt content up to 20 – 25 %
types of permanent magnet materials: quench- cobalt. Typical compositions of cobalt-containing
hardened carbon steels, nickel – iron – alumi- Alnico alloys are given in Table 14 along with
num – cobalt (Alnico) [12605-54-8] alloys, and their magnetic properties. The magnetic proper-
rare-earth cobalt alloys (! Magnetic Materials). ties of these alloys, particularly remanence, can be
improved by applying a strong external field to
Quench-Hardened Carbon Steels. The them as they cool from the solution treatment
addition of cobalt to quench-hardened steels temperature. The resulting magnetic properties
increases the coercive force and the BHmax. The are anisotropic, being higher in the direction
improved properties are mainly due to magne- parallel to the field than perpendicular to it. The
torestrictive strain effects that increase the coer- properties of anisotropic magnets can be further
cive force together with a small increase in improved by careful control of thermal gradients
saturation magnetization. Currently, magnet during cooling, which results in the growth of
steels are used less, but they still have advantages columnar grains with the h100i crystal axes lying
in situations where the magnetic alloy must be along the direction of growth. When these
machined and worked. columnar alloys are subjected to a magnetic field
446 Cobalt and Cobalt Compounds Vol. 9

Table 14. Composition and magnetic properties of Alnico alloys

Composition, % Method of Remanence, Coercive force, BHmax,

manufacture T kA/m kJ/m3
Co Ni Al Cu Ti Nb

3–5 21 – 28 11 – 13 2 – 4 0 – 1 – Cast 0.55 – 0.75 36 – 56 11 – 12

12 – 14 16 – 20 9 – 11 3 – 6 0 – 1 – Cast 0.65 – 0.80 40 – 50 13 – 14
17 – 20 18 – 21 8 – 10 2 – 4 4 – 8 – Cast 0.6 – 0.7 60 – 72 14 – 16
23 – 25 12 – 15 7.8 – 8.5 2 – 4 0 – 0.5 – field treated 1.2 – 1.3 46 – 52 40 – 44
32 – 36 14 – 16 7–8 4 4 – 6 – field treated 0.0 – 0.9 110 – 140 40 – 45
24 – 25 13 – 15 7.8 – 8.5 2 –4 – 0–1 columnar 1.3 – 1.4 56 – 62 56 – 64
32 – 36 14 – 16 7–8 4 4 –6 0–1 columnar 1.0 – 1.3 110 – 140 60 – 74

during cooling, the magnetic properties are again protective atmosphere, before it is compacted to
markedly improved. shape in an applied field. Two dominant groups
The permanent magnet properties of Alnico of rare-earth – cobalt magnets are in commer-
alloys depend on the separation of the original cial production. One group is based on single-
body-centered cubic lattice into two isomor- phase SmCo5 [12017-68-4] and the other on
phous phases: (1) a nickel – aluminum-rich precipitation-hardened alloys near the compo-
nonmagnetic matrix and (2) magnetic iron – sition Sm(Co, Cu, Fe)7.5. A large amount of
cobalt-rich precipitated particles that are epitaxi- the samarium in SmCo5 can be substituted by
ally orientated in the matrix and elongated along neodymium, praseodymium, or the natural mix-
one of the h100i directions. ture of rare-earth metals (mischmetals); this
The Alnico alloys can be prepared by casting either increases the saturation polarization or
in shell molds or by powder metallurgy. The use decreases the cost of the alloy components.
of powder metallurgy depends on a knowledge of Cobalt is also the major component of zero
the domain theory of ferromagnetism. In the magnetorestrictive amorphous alloys. These are
process the particle sizes of the alloy powders metallic glasses with the lowest technically real-
used are reduced so that they approximate the ized coercivities. They have no directionality in
size of single magnetic domains. The fine parti- terms of their magnetic properties and are rela-
cles (0.02 – 0.05 mm) of the iron – cobalt alloys tively hard and strong materials giving rise to low
used can be prepared by thermal reduction of hysteresis eddy current losses. Their properties
salts such as formates or oxalates or by controlled favor their use in applications such as magnetic
electrolysis of mixed iron – cobalt salt solutions. shielding, magnetic heads, amplifiers, pulse
transformers, and HV/HW transformers.
Rare-Earth – Cobalt Alloys. The rare- The uses of the magnets include focusing
earth – cobalt alloys often have the composition magnets in traveling wave tubes, magnetic bear-
MCo5, where M is a rare-earth metal, and contain ings in ultrahigh-speed centrifugal separators and
60 – 65 % cobalt (see also ! Magnetic Materi- inertia wheels, and actuators, motors, and gen-
als). The anisotropy that gives high coercivity in erators of various sizes from watches to 100-
these alloys arises from a hexagonal lattice that horsepower motors, including brushless motors.
has the direction of easy magnetization along the The magnetic properties of these alloys depend
hexagonal axis. Single domain particles com- critically on their manufacturing history, but are
pacted in an external magnetic field retain characterized by an almost linear demagnetiza-
the high coercivity of the individual particles. tion curve between the remanence and coercive
Exceptionally high values of coercive force have force. Typical magnetic properties of some of the
been obtained in these alloys – the values for MCo5 magnets are listed in Table 15. The Sm
alloys based on samarium or praseodymium are (Co, Cu, Fe)x phase owes its coercive force to
the highest and can be as high as 650 kA/m. pinning of domain walls on coherent precipitates
The rare-earth – cobalt magnets are normal- embedded inside the grains either of a M2Co17
ly manufactured by melting the components in a phase embedded in an MCo5 matrix if x < 7 or
protective atmosphere. The cooled melt is then discrete MCo5 particles embedded in M2Co17 for
crushed and milled to a fine powder, again in a x > 7.8. The presence of copper is essential for
Vol. 9 Cobalt and Cobalt Compounds 447

Table 15. Properties of MCo5 magnets 5 – 20 % iron, and 0.1 – 2 % carbon in a cobalt
Magnet Remanence, Coercive force, BHmax,
base. The microstructure of the Stellite alloys
T kA/m kJ/m3 consists of hard particles of Cr7C3 dispersed in a
cobalt-rich solid-solution matrix. The adhesion
SmCo5 0.85 600 140
Sm0.5Pr0.5Co5 0.95 600 176
and cavitation-erosion wear characteristics that
(Mischmetal)Co5 0.65 300 64 have been attributed to the matrix alloy are
PrCo5 0.94 660 150 influenced by a phase change from a strain-
induced cubic form to a hexagonal form in the
matrix alloy. The toughness and abrasion resis-
tance of the alloys are, however, due mainly to
this type of hardening, but it decreases the satu- the volume fraction of carbides and their
ration polarization. morphology.

Cemented Carbides. The precipitation and

6.4. Hard Metal Alloys [4] applications of cemented carbides as hard alloys
are discussed in Section 5.2.1.
Because of their inherent high strength, corrosion
resistance, and ability to retain hardness at
elevated temperature, cobalt-based alloys have 6.5. Cobalt-containing High-Strength
been used extensively for over 50 years in wear- Steels [18]
related engineering applications.
Cobalt is not one of the elements commonly
Cobalt – Chromium Alloys (Stellite Al- added to steels because its direct effects on the
loys). The Stellite alloys were developed properties of steel are usually less than those
around 1900 and over 20 compositions are now achieved by other alloying elements. Cobalt is,
commercially available. They can be divided into however, an important component of steel when
two classes: (1) alloys of cobalt, chromium, high strength is a requirement, and it is a unique
tungsten, and carbon and (2) alloys of cobalt, alloying element in favoring the austenitic struc-
chromium, tungsten, molybdenum, nickel, iron, ture at the ferrite – austenite boundary and in
and carbon. For special hard-facing applications, raising the martensitic transformation tempe-
these alloys can also be modified by the addition rature.
of small percentages of silicon and boron. The addition of a few percent of cobalt to low-
The traditional alloys of cobalt, chromium, alloy steels that contain strong carbide-forming
tungsten, and carbon contain ca. 30 % chromium elements, such as molybdenum or tungsten, in-
with varying amounts of tungsten (4 – 17 %) and creases the tempering resistance of the steel.
carbon (1 – 3.2 %). They are used primarily in Solid-solution hardening of martensitic steels by
lubricated or high-temperature wear applications. the addition of cobalt is particularly noticeable in
The composition of a typical tough-grade hard chromium stainless steels. High chromium fer-
alloy is 30 % chromium, 4 % tungsten, and 1 % ritic and martensitic steels containing 6 – 7 %
carbon, which has a Vickers hardness of 375 N/ cobalt are used for high-temperature applications
mm2 at room temperature and of 100 N/mm2 at up to 550
C. Cobalt is a particularly useful
900
C. A typical medium-grade hard alloy con- alloying element in controlled-transformation
tains 30 % chromium, 9 % tungsten, and 1.5 % steels in which initial cooling of the steel gives
carbon and has hardness values of 470 N/mm2 at austenite, which can be cold worked and subse-
room temperature and of 150 N/mm2 at 900
C. quently transformed into martensite by temper-
A typical hard-grade hard alloy contains 32 % ing or refrigeration. A typical steel of this type
chromium, 18 % tungsten, 2.5 % carbon and has contains 6 % cobalt.
hardness values of 600 N/mm2 at room tempera-
ture and of 200 N/mm2 at 900
C. Maraging Steels are ultrahigh-strength al-
The Co – Cr – W – Mo – Ni – Fe – C loys and contain cobalt, nickel, and molybdenum
alloys contain 25 – 32 % chromium, 0 – 19 % with 5 – 20 % cobalt. Cobalt plays an indirect
tungsten, 3.22 % nickel, 3 – 10 % molybdenum, role in the formation of intermetallic compounds
448 Cobalt and Cobalt Compounds Vol. 9

containing minor alloy components such as tita- Dental Alloys. Cobalt – chromium alloys
nium, which accelerates and enhances the hard- have been used in dentistry since the mid-1930s
ening process. Cobalt also plays a prominent role (! Dental Materials). The most widely used
in the short-range ordering of the solid solution alloy is Vitallium [12629-02-6], which is a
that leads to additional hardening. The advan- cobalt-based alloy containing 30 % chromium,
tages of maraging steels are (1) good tensile 5 % molybdenum, and 0.5 % each of carbon and
strength and toughness that is maintained up to silicon. The alloys are used for cast denture bases,
350
C and (2) their weldability either in the complex partial dentures, and some types of
annealed or aged conditions without preheat or bridgework. The properties of the alloys that
postheat stress-relieving treatments. make them suitable for dental applications are
Cobalt (5 – 12 %) can be added to improve their good castability, resistance to tarnish, com-
the hot hardness of high-speed tool steels. patibility with mouth tissues, high strength and
stiffness, and low density. A modified alloy is
used for porcelain coatings for crown restora-
6.6. Electrodeposited Alloys [9] tions; it enables the crown to be fused to metal
bridges. Alloys containing 12 % chromium, 9 %
Protective coatings of electroplated nickel – co-
nickel, 8 % titanium, and 1 % molybdenum have
balt alloys have been used since the mid-1930s.
expansion characteristics that match those of
The 25 % cobalt alloy deposited from sulfate
dental porcelain.
electrolytes is about three times harder than elec-
troplated nickel and has about the same hardness
Surgical Implants. Cobalt – chromium
as pure cobalt. The Vickers hardnesses of the
alloys similar to Vitallium are used for surgical
electrodeposits vary with cobalt content, and for
implants and for bone replacement and repair
deposits from sulfamate – chloride solutions,
because they are very resistant to corrosion in
they rise from ca. 200 N/mm2 for nickel plate
body fluids. The best known prosthesis is hip
to ca. 540 N/mm2 for a deposit containing 40 %
joint replacement, but the alloys are also used for
of cobalt. The cobalt – nickel alloy deposits
plates, screws, and nails for fracture fixations.
also have improved corrosion resistance in
Because the dynamic loadings on bone replace-
many environments and have been used in the
ment implants amount to several million cycles
production of mirrors, decorative coatings, and
artificial joints must also have high fatigue
for electroforming. Electrodeposits of cobalt –
strength.
tungsten alloys are of interest because of their
hardness at high temperature. Hardness values
Low-Expansion Alloys. Cobalt is used as
in the range of 400 – 600 N/mm2 are obtained
an alloying element in low-expansion iron –
for deposits with 30 % tungsten; this increases to
nickel alloys of the Invar type. The cobalt content
500 –700 N/mm2 after heat treatment at 300 –
can be up to 25 % and the alloys are used for
350
C and remains as high as 650 N/mm2 at
sealing metals to borosilicate glasses. These
600
C. These electrodeposited alloys have
alloys have a face-centered cubic solid solution
been used to improve the wear resistance of hot
structure that is ferromagnetic within the low-
forging dies.
expansion temperature range. The low-expan-
The magnetic properties of a number of elec-
sion behavior is probably caused by a contraction
troplated alloys have attracted attention for use in
that is due to magnetorestriction canceling out
recording systems and computer applications.
the normal thermal expansion of the alloy. Low-
The alloys with suitable magnetic properties
expansion behavior can also be found in the
include cobalt – iron, cobalt – nickel, cobalt –
cobalt –chromium – iron alloy system and the
platinum, and cobalt – phosphorus.
composition with 54 % cobalt, 9 % chromium,
and 37 % iron has an expansion coefficient close
6.7. Alloys with Special Properties to zero over the temperature range from  60 to
þ 60
C.
A number of cobalt-containing alloys have
special applications; they are described in this Spring Alloys. The alloy Elgiloy is typical
section. of the alloys used in springs. It contains 40 %
Vol. 9 Cobalt and Cobalt Compounds 449

cobalt, 20 % chromium, 15 % nickel, 7 % molyb- carbon, phosphorus, and sulfur on heating.

denum, 2 % manganese, 0.15 % carbon, 0.04 % The reaction with sulfur is influenced by the
beryllium, and the balance being iron. After formation of a low melting eutectic (877
C)
special treatment, this alloy has a Vickers hard- between the metal and the Co4S3 phase; the
ness of 650 N/mm2, an elastic limit of ca. reaction between cobalt and sulfur is rapid
160 MN/m2, and a tensile strength of above this temperature. Below 877
C, a protec-
2300 MN/m2. tive layer of sulfide scale is formed. In an atmo-
Alloys with temperature-independent elastic sphere of hydrogen sulfide, cobalt also forms a
moduli can be used as hair springs for watches scale of sulfide, but in air containing sulfur
and chronometers and for spring balances. dioxide, a mixed oxide – sulfide scale is formed.
Although many of the alloys used for this purpose In bulk form, cobalt is resistant to many mild
are based on nickel, a series of cobalt-based corrosive agents, but it is more readily attacked
alloys containing 50 – 60 % cobalt, 25 – 35 % when it is finely divided. Table 16 gives the
iron, and 15 – 18 % chromium have suitable corrosion rates of cobalt in some aqueous media.
properties and have been used for this purpose. Cobalt is strongly attacked by concentrated nitric
acid at ambient temperature. The metal dissolves
Magnetic Recording Alloys. Much work slowly in dilute mineral acids, the Co2þ/Co
has been carried out on the development of thin potential being  0.277 V.
films of cobalt-containing alloys for use as The solubility of oxygen in cobalt is 0.006,
magnetic recording materials [16]. The alloy 0.013, 0.125, and 0.4 wt % (at 600, 1200, 1510,
films can be manufactured by vapor deposition, and 1700
C, respectively). Nitrogen is only
sputtering, ion-plating, and calendering. Several slightly soluble (40 mg/kg) in cobalt at its melt-
alloy compositions have been described for use ing point; the solubility rises to 60 – 70 mg/kg at
in recording applications; they include cobalt – 1750
C. The solubility of hydrogen in cobalt
nickel, copper – nickel – zinc – phosphorus, increases with temperature from 1 mg/kg at
cobalt – iron – magnesium – carbon, and co- 700
C to ca. 8 mg/kg at the melting point. The
balt – molybdenum – zirconium – niobium. solubility of hydrogen in the liquid metal in-
Microcrystalline or amorphous films of the alloys creases to ca. 20 mg/kg at 1500
C and to ca.
are deposited on polymers and alloys such as 27 mg/kg at 1750
C.
Co – Ni (80 : 20), and Co – Mo – Zr – Nb The main oxidation states of cobalt are Co2þ
(81 : 7 : 10 : 2) have been used. Video recording [22541-53-3] and Co3þ [22541-63-5]. In acid
tapes with smooth surfaces have also been solution and in the absence of complexing agents,
produced by depositing a ferromagnetic layer of Co2þ is the stable oxidation state, with oxidation
g-Fe2O3– Co powder. The magnetic properties to Co3þ being difficult.
of the films are important in recording applica-
tions. The films should have a nearly square B/H ½CoðH2 OÞ6 3þ þe !½CoðH2 OÞ6 2þ DE
¼ 1:84 V
hysteresis loop and no magnetorestriction.
The oxidation can be achieved by electrolysis
or ozone, but the Co3þ ion is very unstable and is
7. Cobalt Compounds [9], [20]
Table 16. Corrosion of cobalt in aqueous media at 25
C
7.1. Chemical Properties of Cobalt
Corrosive medium Rate of cobalt
Cobalt is much less reactive than iron. It is stable corrosion,
mg dm2 d1
to atmospheric oxygen unless heated. When
heated, it is first oxidized to Co3O4 and then, Distilled water 1.1
above 900
C, to CoO, which is also the product 5 vol% Ammonia 5.3
10 vol% Sodium hydroxide 5.6
of the reaction between the red-hot metal and Conc. phosphoric acid 7.4
steam. The activation energy of the oxidation 50 % Phosphoric acid (aqueous) 65.1
above 900
C has been calculated as 155 – 5 vol% Acetic acid 12.5
160 kJ/mol. The metal does not combine directly 5 vol% Sulfuric acid 56.8
50 % Hydrofluoric acid (aqueous) 178.6
with hydrogen or nitrogen, but it combines with
450 Cobalt and Cobalt Compounds Vol. 9

rapidly reduced to Co2þ, even at room tempera- Cobalt(II) Compounds. Cobalt combines
ture, with evolution of oxygen from the water. with oxygen to form cobalt(II) oxide [1307-96-
The solution chemistry of cobalt in acid solution 6], CoO, which is stable above 900
C. This
in the absence of complexing agents is dominated oxide has the sodium chloride structure and is
by Co2þ. The most common species present is antiferromagnetic at ordinary temperature.
[Co(H2O)6]2þ, although the ions [CoX3], When cobalt(II) oxide is heated at 400 – 500
C
[CoX4]2, and [CoX6]4 (X ¼ halide) are also in an atmosphere of oxygen, the cobalt(II,III)
found in solutions of hydrogen halides. Cobalt is oxide, Co2O3 [1308-06-1], is formed. This
often removed from solution as its sulfide, and it mixed-valence oxide has the spinel structure with
is interesting to note that although the precipi- cobalt(II) in tetrahedral sites and cobalt(III) in
tates of cobalt sulfide obtained by using H2S are octahedral sites. Cobalt hydroxide [1307-86-4],
not readily soluble in dilute acids, those obtained Co(OH)2, is a product of the hydrolysis of solu-
by using Na2S or (NH4)2S are soluble. All sulfide tions containing Co2þ ions. The hydroxide is
precipitates become less soluble as they age in amphoteric, dissolving both in alkali, to give
the atmosphere to form Co(OH)S. The cobalt blue solutions containing the [Co(OH)4]2 ion,
dihalides (except fluoride) are also readily solu- and in acids. Cobalt(II) forms an extensive range
ble in some organic solvents such as alcohol, of simple and hydrated salts with all of the
acetone, and methyl acetate. common anions including acetate, bromide, car-
In alkali, Co2þ is more readily oxidized to bonate, chloride, fluoride, nitrate, perchlorate,
3þ
Co and sulfate. Many of the hydrated salts and their
solutions contain the pink octahedral [Co
CoOðOHÞ ðsÞþH2 Oþe !CoðOHÞ2 ðsÞþOH DE
¼ 0:17 V
(H2O)6]2þ ion. Although complexes of cobalt(II)
are generally unstable to oxidation, a number of
In the presence of complexing agents, oxida- octahedral species are formed, including (1) a
2þ
tion is very easy in any solution because Co3þ has series of [Co(N – N)3] complexes with neutral
a particularly high affinity for complex forma- bidentate donor ligands, such as ethylenediamine
tion. With N-donor ligands the redox reaction in and bipyridyl, and (2) the acetylacetonates [Co
the presence of ammonia is (acac)2]  2 H2O and [Co(acac)2]. Tetrahedral
3þ 2þ
complexes [CoX4]2 are usually formed with


½CoðNH3 Þ6  þe !½CoðNH3 Þ6  DE ¼ 0:1 V monodentate anionic ligands, such as chloride,
bromide, iodide, thiocyanate, azide, and hydro-
Oxidation of Co solutions containing com- xyl. Tetrahedral [CoX2L2] complexes are formed
2þ

plexing agents can be achieved with air or hy- by a combination of two such ligands with two
drogen peroxide. Thus, the solution chemistry of neutral ligands2þ(L). Addition of cyanide to a
cobalt in the presence of complexing agents is solution of Co produces3a dark green color
dominated by the complex chemistry of Co3þ. attributed to the [Co(CN)5] ion.
The sensitivity of the reduction potential of the The colors produced by cobalt(II) complexes
Co3þ/Co2þ couple to different ligands whose in aqueous media have been used to distinguish
presence renders Co2þ unstable to air oxidation between octahedral and tetrahedral coordination
is shown by the data in Table 17. in the complexes. In general, octahedral species
are pink to violet, whereas tetrahedral species are


Table 17. DE for some Co /Co couples in acid solution
3þ 2þ blue. This is not an infallible distinction, but it
does provide a useful guide that can be improved
Redox reaction DE
by analyzing the electronic spectra of the
complexes.
[Co(C2O4)3]3 þ e
[Co(C2O4)3]4 0.57
[Co(EDTA)] þ e
[Co(EDTA)]2 0.37 Cobalt(III) Compounds. In addition to the
[Co(bipy)3]3þ þ e
[Co(bipy)3]2þ 0.31
[Co(en)3]3þ þ e
[Co(en)3]2þ 0.18
mixed-valence oxide Co2O3, impure forms of the
[Co(CN)6]3 þ H2O þ e
 0.8 unstable cobalt(III) oxide, Co2O3, have been
[Co(CN)5(H2O)]3 þ CN prepared. Very few simple cobalt(III) salts are
EDTA ¼ ethylenediaminetetraacetic acid; bipy ¼ bipyridyl; en ¼ known. The blue sulfate, Co2(SO4)3  18 H2O
ethylene diamine. [13478-09-6], which contains the [Co(H2O)6]3þ
Vol. 9 Cobalt and Cobalt Compounds 451

species, can be obtained by the electrolytic oxi- In addition to the two most stable oxidation
dation of cobalt(II) in 4M sulfuric acid solution. states, 2þ and 3þ, cobalt also forms compounds
It is stable when dry, but decomposes in the in the 1, 0, 1þ, and 4þ oxidation states. There
presence of moisture. The alums MCoIII(SO4)2  are only a few reported examples of CoIV [20499-
2 H2O (M ¼ K, Rb, Cs, NH4) are also known, 79-0] compounds, mainly fluoro complexes,
and a hydrated fluoride 2 CoF3  7 H2O has been [CoF6]2, and mixed-metal oxides (the purity of
reported. compounds in this oxidation state is
The chemistry of cobalt(III) is dominated by questionable).
complex formation. Cobalt(III) complexes are The formation of cobalt compounds in
kinetically inert, and for this reason, indirect oxidation states lower than 2þ requires the
methods of syntheses are used to obtain them. presence of stabilizing p-acceptor ligands. The
Usually the ligand is added to a solution of Co2þ, 1þ state [16610-75-6] is not common for
which is then oxidized with some convenient cobalt, and most of the known examples are
oxidant, often in the presence of a catalyst such pentacoordinated complexes of the type [Co
as activated charcoal. A wide range of cobaltam- (NCR)5]þ (NCR ¼ organic nitrile). Cobalt
mines have been prepared and studied; the spe- forms a wide range of complexes in which its
cies identified in solution or as solid derivatives formal oxidation state is 0 or 1 – [16727-18-7].
include [Co(NH3)6]3þ, [Co(NH3)5H2O]3þ, [Co Many of these contain ligands such as CO,
(NH3)5X]2þ (X ¼ Cl, Br, NO 2, NO 3), and cis- CN, NOþ, and RNC, but other ligands such
and trans-[Co(NH3)4X2]þ (X ¼ Cl, NO 2). as tertiary phosphines also stabilize lower oxi-
In addition to these simple cobaltammines, a dation states to give compounds like [Co(P
number of polynuclear species containing bridg- (CH3)3)4]þ, which is prepared by reducing an

ing groups such as NH 2
2 , NH , NO2, OH , and ethereal solution of cobalt(II) chloride with
O2 have been prepared. Polynuclear cobaltam- sodium or magnesium amalgam in the presence
mines that have been identified include: of trimethylphosphane.

Cobalt Carbonyls [19], [20]. The carbonyl

complexes of cobalt are important because of
their uses as hydroformylation catalysts (see
Section 7.3.3). Because cobalt has an odd number
of valence electrons, it can only satisfy the
eighteen-electron rule in its carbonyls if Co –
Co bonds are formed. For this reason, the princi-
pal binary carbonyls of the element are octacar-
bonyldicobalt [15226-74-1], [Co2(CO)8], dode-
cacarbonyl-tetracobalt [19212-11-4], [Co4(-
CO)12], and hexadecacarbonylhexacobalt
[12182-17-1], [Co6(CO)16].
Octacarbonyldicobalt is prepared by heating
Complexes of cobalt(III) with O-donor the metal to 250 – 300
C at 200 – 300 bar of
ligands are generally less stable than those with carbon monoxide or by heating cobalt carbonate
N-donor species, although [Co(acac)3] [14024- under similar conditions in the presence of hy-
48-7] and [Co(C2O4)3]3 are known. Apart from drogen. It is an air-sensitive orange-red solid with
the O2-bridged compounds mentioned above, the a melting point of 51
C. The compound can be
octahedral fluoro complexes [CoF6]3 and reduced with sodium amalgam in benzene to give
[CoF3(H2O)3] are the only known high-spin co- the tetrahedral monomeric ion [Co(CO)4], acid-
balt(III) complexes, the paramagnetic moment of ification of which leads to tetracarbonylhydrido-
[CoF6]3 at room temperature being about cobalt [16842-03-8], [HCo(CO)4]. This hydride
5.8 mB. All other cobalt(III) complexes are is a yellow liquid that forms a colorless vapor. It
low-spin and diamagnetic, including the hexa- melts at  26
C and decomposes above this
cyanocobaltates (III), [Co(CN)6]3, and the hex- temperature to H2 and [Co2(CO)8]. It is partly
anitritocobaltates(III), [Co(NO2)6]3. soluble in water to give a strong acid solution
452 Cobalt and Cobalt Compounds Vol. 9

Figure 5. Structures of carbonyl complexes of cobalt

A) [Co2(CO)8] (solid state); B) [Co2(CO)8] (solution); C) [Co4(CO)12]; D) [Co8C(CO)18]

containing [Co(CO)4] ions. The hydrogen atom The structure of [Co2(CO)8] in the solid state
in [HCo(CO)4] is bound directly to the cobalt, is different from that in solution. The solid-state
giving a Co – H infrared stretching frequency of structure shown in Figure 5 A involves two
ca. 1934 cm1. The [Co(CO)4] ion is reoxi- bridging carbonyl groups and can be best ratio-
dized to [Co2(CO)8] by carbon tetrachloride, and nalized in terms of the formation of a bent Co –
further reaction with this reagent leads to a triply Co bond. In solution, however, this structure is in
bridged chloromethynyl derivative, from which a equilibrium with the form shown in Figure 5 B,
carbidocarbonyl compound, disodium carbido- in which the dimer is held together by a Co – Co
pentadecacarbonylhexacobaltate, Na2[Co6C bond. The structure of [Co4(CO)12] is shown in
(CO)15], can be obtained: Figure 5 C. The structure of the carbidocarbonyl
(Fig. 5 D) shows the presence of a carbon atom
at the center of a distorted square antiprismatic
cobalt cluster.
In addition to the carbonyls, cobalt also forms
complexes with N2 and NOþ, which are isoelec-
tronic with CO. Examples of this type of complex
are [CoH(PR3)3(N2)] and [Co(CO)3NO], where
R is an alkyl group.

7.2. Commercially Important Cobalt

Other carbidocarbonyls are obtained by simi- Compounds [9], [16], [21], [22]
lar routes, including [Co6C(CO)4], [Co8C
(CO)18]2, and [Co13(C)2(CO)24H]4. Dodeca- Table 18 lists the known applications of cobalt
carbonyltetracobalt is obtained as a green black compounds. The most important commercially
solid by heating [Co2(CO)8] in an inert atmo- available compounds are the oxides, as well as
sphere at 50
C. hydroxide, chloride, sulfate, nitrate, phosphate,
Vol. 9 Cobalt and Cobalt Compounds 453

Table 18. Industrial uses of cobalt compounds

Compound CAS reg. no. Formula Uses

Acetate(III) [917-69-1] Co(CH3COO)3 catalyst

Acetate(II) [71-48-7] Co(CH3COO)2  4 H2O driers for lacquers and varnishes, sympathetic inks,
catalysts, pigment for oil cloth, mineral supplement,
anodizing, stabilizer for malt beverages
Acetylacetonate [14024-48-7] Co(C5H7O2)3 vapor plating of cobalt
Aluminate [13820-62-7] CoAl2O4 pigment, catalysts, grain refining
Ammonium sulfate [13586-38-4] CoSO4(NH4)2SO4  6 H2O catalysts, plating solutions
Arsenate [24719-19-5] Co3(AsO4)2  8 H2O pigment for paint, glass, and porcelain
Bromide (II) [7789-43-7] CoBr2 catalyst, hydrometers
Carbonate [513-79-1] CoCO3 pigment, ceramics, feed supplements, catalyst
Carbonate (basic) [7542-09-8] 2 CoCO3  Co(OH)2  H2O chemicals
Carbonyl [15226-74-1] Co2(CO)8 catalyst
Chloride [7791-13-1] CoCl2  6 H2O chemicals, sympathetic inks, hydrometers, plating baths,
metal refining, pigment, catalyst
Chromate [24613-38-5] CoCrO4 pigment
Citrate [18727-04-3] Co3(C6H5O7)2  2 H2O therapeutic agents, vitamin preparations
Ferrate [12052-28-7] CoFe2O4 catalyst, pigment
Fluoride [10026-17-2] CoF2 fluorinating agents
[10026-18-3] CoF3
Fluoride [13817-37-3] CoF2  4 H2O catalyst
Fluorosilicate [15415-49-3] CoSiF6  6 H2O ceramics
Formate [6424-20-0] Co(CHO2)2  2 H2O catalyst
Hydroxide [1307-86-4] Co(OH)2 paints, chemicals, catalysts, printing inks
Iodide [15238-00-3] CoI2 moisture indicator
Linoleate [14666-96-7] Co(C18H31O2)2 paint and varnish drier
Manganate [12139-69-4] CoMn2O4 catalyst, electrocatalyst
Naphthenate Co(C11H10O2)2 catalyst, paint and varnish drier
Nitrate [10026-22-9] Co(NO3)2  6 H2O pigments, chemicals, ceramics, feed supplements, catalyst
2-Ethylhexanoate [136-52-7] Co(C8H15O2)2 paint and varnish drier
Oleate [14666-94-5] Co(C18H33O2)2 paint and varnish drier
Oxalate [814-89-1] CoC2O4 catalysts, cobalt powders
Oxide [1307-96-6] CoO chemicals, catalysts, pigments
Oxide [1308-06-1] Co3O4 enamels, semiconductors
Oxides (mixed metal) pigments
Dilanthanum tetroxide [39449-41-7] La2CoO4 catalyst, anode
Tricobalt tetralan-
thanum decaoxide [60241-06-7] La4Co3O10 catalyst
Lithium oxide [12190-79-3] LiCoO2 battery electrode
Sodium oxide [37216-69-6] NaCoO2 battery electrode
Dicobalt manganese
tetroxide [12139-92-3] MnCo2O4 catalyst
Dicobalt nickel tetroxide [12017-35-5] NiCo2O4 catalyst, anode
Lanthanum trioxide [12016-86-3] LaCoO3 oxygen, electrode
Phosphate [10294-50-5] Co2(PO4)2  8 H2O glazes, enamels, pigments, steel pretreatment
Potassium nitrite [17120-39-7] K3Co(NO2)6  1.5 H2O pigment
Resinate Co(C44H62O4)2 paint and varnish drier, catalyst
Succinate [3267-76-3] Co(C4H4O4)  4 H2O therapeutic agents, vitamin preparations
Sulfamate Co(NH2SO3)  3 H2O plating baths
Sulfate [10026-24-1] CoSO4  x H2O chemicals, ceramics, pigments
Sulfide [1317-42-6] CoS catalysts
Tungstate [12640-47-0] CoWO4 drier for paints and varnishes

carbonate, acetate, oxalate, and other carboxylic able: gray cobalt(II) oxide, containing 75 – 78 %
acid derivatives. cobalt, and black cobalt(II) dicobalt(III) oxide,
containing 70 – 74 % cobalt.
Cobalt Oxides. Two main types of cobalt Cobalt(II) oxide, CoO (78.66 % Co), is usual-
oxide distinguishable by their colors are avail- ly prepared by the controlled oxidation of the
454 Cobalt and Cobalt Compounds Vol. 9

metal above 900
C followed by cooling in a hexahydrate in a moist atmosphere. The fact that
protective atmosphere to prevent partial oxida- the anhydrous chloride is blue and the hexahy-
tion to Co3O4. Cobalt(II) oxide has a cubic unit drate red is used as a humidity indicator in silica
cell with a ¼ 0.425 nm. It is insoluble in water, gel desiccants. The chloride is also used in the
ammonium hydroxide, and alcohol, but dissolves electroplating, ceramics, glass, chemical, agri-
in strong acids in the cold and in weak acids on cultural, and pharmaceutical industries.
heating. Cobalt oxide (CoO) absorbs a large
amount of oxygen at room temperature. Cobalt Cobalt(II) Sulfate. Cobalt(II) sulfate hexa-
(II) dicobalt(III) tetroxide, Co3O4 (73.44 % Co), hydrate, CoSO4  6 H2O (20.98 % Co), is a
can be prepared by the controlled oxidation of brownish-red crystalline compound with a
cobalt metal or CoO or by the thermal decompo- monoclinic unit cell (a ¼ 1.545 nm, b ¼
sition of cobalt(II) salts at temperatures below 1.308 nm, c ¼ 2.004 nm, b ¼ 104.7
). It is pre-
900
C. The cubic spinel lattice has a ¼ 0.807 pared by concentrating a sulfuric acid solution of
nm; the solid material readily absorbs oxygen at cobalt oxide or carbonate. It is an efflorescent
room temperature, but never transforms into substance and loses one molecule of water when
cobalt(III) trioxide, Co2O3. The mixed-valence exposed to dry air or when heated gently. The
oxide is insoluble in water and only slightly hexahydrate obtained loses water in two stages to
soluble in acids. give the monohydrate at 100
C and the anhy-
drous sulfate above 250
C. The monohydrate,
Cobalt(II) Hydroxide. Co(OH)2 (63.43 % CoSO4  H2O (34.08 % Co), has been manufac-
Co) is prepared commercially as a pink solid by tured and sold commercially. Cobalt sulfates are
precipitation from a cobalt(II) salt solution with very soluble in water and are generally more
sodium hydroxide. It has a hexagonal crystal stable than cobalt(II) chlorides or nitrates; for
structure with a ¼ 0.317 nm and c/a ¼ 1.46. It this reason they have been widely used as sources
is insoluble in water and alkaline solutions, but of cobalt(II) in solution for the manufacture of
dissolves readily in most inorganic and organic chemicals and for the electroplating industries.
acids; for this reason, it is commonly used as a The sulfates are also used in the ceramics, lino-
starting material in the synthesis of cobalt che- leum, and agricultural industries.
micals. Co(OH)2 decomposes thermally by loss
of water, starting at 150
C, to give the anhydrous Cobalt(II) Nitrate. Cobalt(II) nitrate hexa-
oxide at 300
C. Care must be taken to store the hydrate, Co(NO3)2  6 H2O (20.26 % Co), is a
hydroxide in the absence of air because slow air red-brown crystalline compound with a mono-
oxidation leads to a product that is poorly soluble clinic unit cell (a ¼ 1.509 nm, b ¼ 0.612 nm,
in weak acids. c ¼ 1.269 nm, b ¼ 119
). It is prepared by con-
centrating a nitric acid solution of cobalt oxide or
Cobalt(II) Chloride. Cobalt(II) chloride carbonate. The hexahydrate loses water rapidly
hexahydrate, CoCl2  6 H2O (24.79 % Co), is a at 55
C to give the trihydrate; the monohydrate
dark-red deliquescent crystalline compound with can also be prepared. The nitrates are very solu-
a monoclinic unit cell (a ¼ 0.886 nm, b ¼ ble in water, alcohols, and acetone. Cobalt nitrate
0.707 nm, c ¼ 1.312 nm, b ¼ 97
170 ). It is pre- is an important source of high-purity cobalt for
pared by concentrating a hydrochloric acid solu- use in the electronics and related industries, and
tion of cobalt oxide or carbonate. The chloride is the compound has also found uses in the chemical
very soluble in water, alcohols, and a number of and ceramics industries.
other organic solvents. The solubility of the
chloride in aqueous media does, however, de- Cobalt(II) Phosphate. Cobalt(II) phos-
crease with increasing hydrochloric acid content. phate octahydrate [10294-50-5], Co3(PO4)2  8
The hexahydrate dehydrates thermally in three H2O (34.63 % Co), is obtained as a purple floc-
stages, giving the dihydrate at 50
C, the mono- culent precipitate when an alkali-metal phos-
hydrate at 90
C, and the anhydrous chloride at phate is added to the solution of a cobalt(II) salt.
130 – 140
C. The phosphate is soluble in inorganic acids and
The anhydrous chloride and the lower hy- particularly in phosphoric acid, but is insoluble in
drates are very hygroscopic and transform to the water or alkaline solutions. It is used in the paints
Vol. 9 Cobalt and Cobalt Compounds 455

and ceramics industries and is a component of formate, acetylacetonate, and citrate have been
some steel phosphating formulations. produced for use in a variety of applications
including catalysts, drying agents, metal film
Cobalt(II) Carbonate. Although pure production, and medical uses.
CoCO3 has a cobalt content of 49.57 % and a
pseudohexagonal unit cell, the material available
commercially is a mauve basic carbonate, 7.3. Industrial Applications of Cobalt
[CoCO3]x [Co(OH)2]y  z H2O, of indeterminate Compounds [4], [9], [16], [21–24]
composition and a cobalt content of 45 – 47 %.
The basic carbonate is insoluble in water, but The more important applications of cobalt che-
dissolves easily in most inorganic and organic micals listed in Table 10 can be considered under
acids. For this reason it is often used as a starting the following headings : glasses, ceramics, and
material for the manufacture of other chemicals. refractories; driers, paints, varnishes, and dres-
The basic carbonate is also used in the ceramics sings; catalysts; electroplating; electronics and
and agricultural industries. solid-state devices; and agricultural, nutritional,
and medical uses.
Cobalt(II) Acetate. Cobalt(II) acetate tetra-
hydrate, Co(CH3CO2)2  4 H2O (23.68 % Co), is
a deliquescent mauve-pink crystalline compound 7.3.1. Glasses, Ceramics, and Refractories
with a monoclinic unit cell (a ¼ 0.847 nm, b ¼ [9], [23]
1.190 nm, c ¼ 0.482 nm, b ¼ 94.3
). It is pre-
pared by concentrating solutions of cobalt hy- The addition of cobalt oxides to provide a blue
droxide or carbonate in acetic acid. The tetra- pigment for glass, ceramics, and enamels has
hydrate is soluble in water, alcohols, inorganic, been used for many centuries (! Ceramic Col-
and organic acids including acetic acid. The orants). The level of cobalt in the final product
compound loses its water of crystallization at ca. depends on the color intensity required, but is
140
C. It is used in the manufacture of drying typically 0.4 – 0.5 % for colored ceramic bodies,
agents for inks and varnishes, catalysts, and 0.5 % for the blue glass used in welder’s goggles,
pigments, as well as in the anodizing and agri- and a few mg/kg in camera lens glass. A range of
cultural industries. cobalt-containing mixed-metal oxides and cobalt
tripotassium hexanitrite (a yellow pigment) is
Cobalt(II) Oxalate. Cobalt(II) oxalate di- also available for use as pigments. An indication
hydrate, Co(C2O4)  2 H2O (32.23 % Co), is ob- of the colors obtained from mixed-metal oxide
tained as a pink precipitate when oxalic acid or an pigments is given in Table 19. The shades of the
alkali-metal oxalate is added to the solution of a colors also depend on the exact composition of
cobalt(II) salt. The dihydrate loses water when it the oxide and on the method of preparing the
is heated gently in air, and at 200
C it decom- pigment. The compositions of some ceramic
poses to cobalt(II) oxide. The oxalate is insoluble pigments are given in Table 20.
in water, slightly soluble in acids, but soluble in The pigments are normally prepared by mix-
solutions containing ammonia or an ammonium ing the ingredients as oxides or as readily
salt. The main use of cobalt oxalate is as a starting
material for the preparation of cobalt metal
powders. Table 19. Colors of cobalt-containing pigments

Metals in Color
Cobalt(II) Carboxylates. The cobalt salts mixed-metal oxide
of carboxylic acids can be made by the direct
Co Al blue
reaction of cobalt powder, oxide, or hydroxide Co P violet
with the organic acid or by precipitation reac- Co Zn green
tions involving the addition of the sodium salt of Co Sn Si light blue
the acid to an aqueous solution of a cobalt salt, Co Cr Al turquoise
Co Mg pink
such as the sulfate. Cobalt(II) resinate, oleate, Co Fe brown
linoleate, soyate, naphthenate, ethylhexanoate,
456 Cobalt and Cobalt Compounds Vol. 9

Table 20. Compositions of cobalt-containing pigments of a mixture of aluminum oxide and two cobalt
Color Content of Content of other components
coats, one of which is high-firing and the other
Co3O4, % (% in parentheses) low-firing. The purpose of the latter is to seal off
oxygen at low temperatures to protect the base
Mazarine blue 68.0 SiO2 (12), cornish stone
(16), CaCO3 (4) metal from oxidation from the start of the firing.
Willow blue 33.3 CaCO3 (50), SiO2 (16.7) These coatings are claimed to withstand com-
Dark-blue 44.6 Al2O3 (55.4) bustion products and corrosive vapors up to
Matt blue 20.0 Al2O3 (60), ZnO (20) 750
C and have been used to protect flue gas
Blue-green 41.8 Al2O3 (39), Cr2O3 (19.2)
Black 20.6 Fe2O3 (41.1), Cr2O3
and gas turbine exhaust pipes.
(32.4), MnO2 (5.9) The metal – glass coatings used to protect
plain carbon steels typically contain 80 % pow-
dered chromium and 20 % of a borosilicate glass
decomposable salts and then calcining the mix- containing 5 % cobalt oxide. Superalloys have
ture at 1100 – 1300
C prior to milling the prod- also been coated with a ceramic consisting of
uct to obtain the pigment as a fine powder. In high-firing cobalt ground coats with high alumi-
ceramic applications, the pigments can be added num oxide content or with boron nitride contain-
to the ceramic base materials to give a body color ing a lithium cobalt oxide binding agent.
or, after being mixed with suitable fluxes, applied
as an underglaze (as on Delft china), or overglaze
decoration. The final shade after the ceramic has 7.3.2. Driers, Paints, Varnishes, and
been fired may be modified by reaction between Dressings [9]
the pigment and the clay base.
Cobalt pigments are used for decolorizing A number of cobalt pigments have been used in
both glass and pottery articles that contain iron oil paintings, including Cobalt blue smalt
oxide, which would give a yellow coloration to [1345-16-0] (a silicate), Thenard’s blue
the products in the absence of a decolorizer. The [13820-62-7] (an aluminate), Cerulean blue
yellow color is masked by the complementary [68187-05-3] (a stannate), Cobalt violet
blue color of the cobalt pigment added. The [82196-89-2] (a phosphate), and Aureolin
levels of cobalt required for the compensation yellow [13782-01-9] (potassium cobalt(III)
of iron oxide colorations in typical pottery bod- nitrite). Some of the pigments have also been
ies, enamels, and glasses are 0.003 – 0.02 %, used for inks, for printing on fabrics and paper,
0.002 – 0.01 %, and 1 – 2 mg/kg, respectively. and for coloring plastics.
Cobalt compounds have also been used to re- Cobalt acetate is used for coloring the oxide
move color that can arise in the polymerization of layers in anodized aluminum to give shades
poly(ethylene terephthalate) used in clear plastic varying from bronze to black. The anodized
bottle manufacture. metal is immersed in a 20 % solution of cobalt
The color of the pigments and of the com- acetate and then either in a solution of ammoni-
pounds used for decolorizing arise from the um sulfide to produce cobalt sulfide (black) or in
optical absorption spectra of cobalt atoms modi- a solution of potassium permanganate (bronze).
fied by lattice effects in the pigments and in the The different depths of bronze shades are
final products. achieved by different numbers of alternate im-
In the vitreous enameling of steel sheet, a mersions in the cobalt acetate and permanganate
small amount of cobalt oxide (0.15 – 1.0 %) is solutions. In addition to its use as a coloring
included in the ground coat mixture of feldspar, agent, cobalt acetate is also used to improve the
sand, borax, and soda ash to improve the adhe- lightfastness of anodized aluminum dyed with
sion of the enamel to the steel. As generally organic pigments.
accepted, the cobalt does not contribute directly The cobalt(II) salts of a number of organic
to the adhesion process, but it is involved in acids have been used as drying agents for unsat-
creating suitable conditions for the development urated oils in paints, varnishes, and inks. The
of good adhesion. salts used for this purpose include the oleate,
Ceramic coatings for low-carbon and low- ethylhexanoate, naphthenate, soyate, linoleate,
alloy steels for high-temperature use can consist resinate, and tallate. These cobalt salts are either
Vol. 9 Cobalt and Cobalt Compounds 457

soluble in the unsaturated oils or they react with hydroformylation. The catalysts used in the
them to form soluble compounds that then act hydropurification reactions including hydrode-
both as oxidation accelerators of the oil and as sulfurization consist of cobalt and molybdenum
polymerization catalysts, forming a film with oxides supported on an inert material such as
increased stability, resistance, and flexibility. alumina and contain 2.5 – 3.7 % cobalt. The
The drying action is due entirely to the redox homogeneous catalysts used in the liquid-phase
behavior of the cobalt species present; the car- oxidation processes are soluble cobalt(II) carbox-
boxylate moieties simply act as oleophilic groups ylates such as acetate, naphthenoate, and oleate.
to solubilize the cobalt compounds in the oils. Relatively large concentrations of cobalt (0.1 –
The amounts of cobalt added to the oils as drying 1 %) are used as cobalt salts in the catalysis of
agents are in the range 0.01 – 0.6 %; the salts can hydroformylation reactions in which [HCo(CO)4]
be added as solutions in organic solvents or as is the active material. The amounts of cobalt
dispersions of ultrafine powders (! Drying Oils catalysts used in hydrodesulfurization, oxidation,
and Related Products). Cobalt-containing drying and hydroformylation have been estimated at
agents are also used in low-temperature curing 800 – 1800, 200, and 800 t/a, respectively.
processes for silicone resins. The cobalt species In hydrotreating, the cobalt – molybdenum
promote drying at the surface of the film in catalysts function in the sulfide form and catalyze
contact with atmospheric oxygen. Its action is two reactions, hydrogenation and hydrogenoly-
so fast that a hard polymeric film can be formed sis of carbon – heteroatom bonds. Two theories
that prevents deep drying; for this reason cobalt is have been used to explain the enhanced catalytic
often used in conjunction with other metals such activity obtained by having both cobalt and
as zinc and calcium that do not have a drying molybdenum oxides in the catalysts: the pseu-
action but slow down and control the effects of dointercalation model and the remote control
cobalt. model. The first model assumes an association
between molybdenum sulfide (MoS2) and cobalt
to produce active sites for the catalysis. The
7.3.3. Catalysts [24] remote control model, on the other hand, assumes
that active centers are created in the MoS2 lattice
Cobalt compounds are versatile catalysts. by a mobile species of hydrogen produced at a
Cobalt-containing materials have been used to cobalt sulfide component. The cobalt in the cata-
catalyze the following types of reactions: hy- lyst thus controls the nature and the number of
drogenation; dehydrogenation; hydrogenolysis catalytically active sites formed.
including hydrodenitrification and hydrodesul- The types of oxidations catalyzed by cobalt
furization; hydrotreatment of petroleum salt solutions include p-xylene to terephthalic
products; selective oxidation; ammonoxida- acid, cyclohexane to adipic acid, and hydrocar-
tion; oxidation; hydroformylation; polymeri- bons or acetaldehyde to acetic acid. All of these
zation; selective decomposition; and ammonia liquid-phase oxidations follow classical radical
synthesis. In addition to these more general chain mechanisms. The following reaction
reactions, there are a number of reactions for mechanism for cyclohexane oxidation shows the
which cobalt chemicals are among the best role of cobalt in decomposing the hydroperox-
catalysts. These include their uses as driers, ides formed in the reaction and thus avoiding
described in Section 7.3.2., their use in the unwanted side reactions:
oxidation of ammonia to nitric acid, and the
syntheses of fluorocarbons. Initiation:
The effectiveness of cobalt as a catalyst is
related to the ease with which the element forms C6 H12 þR!C6 H11 þRH
complexes and particularly to the large variety of
ligands in these complexes. Chain reaction:
The main industrial processes using cobalt
catalysts are the removal of sulfur from various C6 H11 þO2 !C6 H11 OO
petroleum-based feedstocks (hydrodesulfuriza-
tion), selective liquid-phase oxidation, and C6 H11 OOþC6 H12 !C6 H11 OOHþC6 H11 
458 Cobalt and Cobalt Compounds Vol. 9

Hydroperoxide decomposition: catalyses, including the oxidations of propene,

butane, butenes, and methanol. New systems
containing cobalt can also be used to catalyze
the conversion of thiols to disulfides in gasoline
sweetening.
Cobalt compounds have considerable poten-
tial in homogeneous catalytic systems and are
used with copper in the oxidation of toluene to
benzoic acid; good selectivities for the conver-
In hydroformylation and hydroesterification, sion of methanol to ethanol have been achieved
oxo reactions such as those listed below occur by using cobalt cluster catalysts.
(! Carbonylation; ! Oxo Synthesis): Cobalt fluoride can be used as a catalyst in the
fluorination of hydrocarbons. These reactions
CH3 OHþCO!CH3 COOH utilize the cobalt fluoride as a fluorine carrier
and take place in two steps – oxidation of CoF2
RCH ¼ CH2 þCOþH2 !RCH2 CH2 CHO to CoF3 at 150 – 200
C followed by fluorination
of the hydrocarbon at 250 – 350
C. Both stages
RCH ¼ CH2 þCOþ2H2 !RCH2 CH2 CH2 OH of the reaction are exothermic, and careful con-
trol of the reaction is needed to obtain optimum
0
RCH ¼ CH2 þCOþR OH!RCH2 CH2 COOR
0
yields.
Cobalt has been a component of the catalysts
The cobalt salts added to the reaction mixture used in the Fischer – Tropsch synthesis of hy-
are converted into [HCo(CO)4], which catalyzes drocarbons from synthesis gas. The mixed co-
the oxo reaction by the following types of mech- balt –thorium oxide – magnesium oxide – kie-
anism: selguhr catalyst (100 : 5 : 8 : 200), for example, is
generally prepared by boiling solutions of sodi-
½HCoðCOÞ4 þRCH ¼ CH2 !½RCH2 CH2 CoðCOÞ4  um carbonate containing cobalt and thorium
nitrates to precipitate the metals as carbonates
½RCH2 CH2 CoðCOÞ4 þCO!½RCH2 CH2 COCoðCOÞ4  prior to adding magnesium oxide and kieselguhr.
The mixture of metal carbonates and support
½RCH2 CH2 COCoðCOÞ4 þH2 ! materials is then filtered off, washed, dried,
RCH2 CH2 COHþ½HCoðCOÞ4  milled, and reduced in a hydrogen atmosphere
to produce the Fischer – Tropsch catalyst.
In addition to the main types of reactions Raney cobalt can be prepared by leaching a
catalyzed by cobalt species, cobalt compounds finely powdered aluminum – cobalt alloy con-
have also been used as catalysts for automobile taining 40 – 50 % cobalt with sodium hydroxide
exhaust gas purification, for the manufacture of at 15 – 20
C, which dissolves the aluminum to
nitric acid, for heterogeneous oxidation reac- leave a very active porous cobalt residue. The
tions, and for the oxidation of toluene to benzoic properties of Raney cobalt are similar to those of
acid. Cobalt(II) dicobalt(III) tetroxide is one of Raney nickel, and the materials can be used to
the most effective catalysts for the oxidation of catalyze the same types of reaction.
carbon monoxide and also has a high activity in
oxidizing nitrogen monoxide; these properties
have led to the use of Co3O4 and the LaCoO3 7.3.4. Electroplating [9]
(perovskite phase) as components of exhaust gas
purification catalysts. The oxide Co3O4 is also a Cobalt is readily electrodeposited from a number
highly selective catalyst in the oxidation of am- of electrolyte solutions. Pure plated layers of
monia to nitric acid, although it loses activity due cobalt are of little commercial value, but cobalt
to sintering after repeated use and due to poison- is an important component of a number of alloy
ing by sulfur dioxide. electrodeposits. The electrochemical equivalent
Cobalt is often cited in the patent literature as of cobalt is 1.099 g A1 h1 and, with cathode
an additive in nearly all heterogeneous oxidation efficiencies of 90 – 100 % for most electrolytes,
Vol. 9 Cobalt and Cobalt Compounds 459

Table 21. Compositions of cobalt-containing plating baths

Electrolyte composition Content, Metal

g/L deposited
Metal salts Content, g/L Additives

CoSO4  7 H2O 332 H3BO3 30 Co

CoCl2  6 H2O 300 H3BO3 30 Co
CoSO4(NH4)2SO4  6 H2O 200 H3BO3 25 Co
Co(SO3NH)2  4 H2O 450 HCoNH2 30* Co
Co(BF4)2 116 – 154 H3BO3 15 Co
NiSO4  7 H2O 240 H3BO3 30 Co – Ni
NiCl2  6 H2O 45 NaHCO2 35 Co – Ni
CoSO4  7 H2O 3 – 15
CoSO4  7 H2O 150 NaCl 28 Co – W
NaWO3 15 – 50 H3BO3 40 Co – W
Na heptanoate 100 Co – W
*
mL/L.

the deposition rates are ca. 1 g A1 h1. Solu- when sodium molybdate replaces the sodium tung-
tions of cobalt sulfate, chloride, sulfamate, or state in the electrolyte. A number of other cobalt-
fluoroborate have been used as electrolytes. The containing alloys can be obtained as electrodepo-
compositions of typical plating baths are listed in sits, including cobalt – iron, cobalt – platinum,
Table 21. To obtain a smooth metal deposit at a cobalt – gold, and cobalt – phosphorus.
cathode efficiency of 90 – 100 % and a current Electrodeposited cobalt and cobalt alloys
density of 5 A/dm2, the bath should have a high have been used as matrices for composite
cobalt ion concentration (65 – 100 g/L) and wear-resistant coatings in systems that codeposit
should be operated at a temperature of 25 – suspended particles of materials such as alumina
50
C and a pH of 4 – 5. and silicon carbide with the plating metal.
Additions of cobalt sulfate to nickel plating
baths have been used to produce bright nickel
plates consisting of nickel – cobalt alloys. The 7.3.5. Electronics, Solid-State Devices, and
proportion of cobalt in the electrolyte controls the Batteries [16], [22]
composition of the alloy deposited. Since the
deposition potential of cobalt is less than that of Since 1990 there has been increasing interest
nickel, the cobalt – nickel ratio is higher in the in the use of cobalt-containing materials in
deposit than in the electrolyte. A typical bath the electronics industry. The most important
composition is given in Table 21. If this bath is applications are magnetic recording and batte-
operated at 55 – 60
C, pH 4 – 4.3, and with a ries.
current density of 2.5 – 4 A/dm2, an alloy de- Cobalt-containing magnetic recording media
posit containing 18 % cobalt is produced. The are of two types, (1) oxide or metal particles and
cobalt content of the bath must be maintained by (2) metal films. Cobalt-modified iron oxide par-
the addition of the cobalt salt when nickel anodes ticles are presently the predominant material in
are used. Cobalt sulfamate is used as an alterna- video tapes, while cobalt and particularly cobalt-
tive to cobalt sulfate in some chloride – sulfate chromium alloys are used in perpendicular re-
nickel plating baths. cording applications. In digital storage of optical
Cobalt – tungsten alloys with tungsten con- data and bubble memories, high recording den-
tents of up to 10 % can be deposited from baths sities are required, and cobalt alloys such as
containing cobalt sulfate and sodium tungstate Gd – Co or Gd – Fe – Co are used in these
(typical electrolyte composition given in Ta- magneto-optic devices.
ble 21). A temperature of 70 – 90
C and a pH The three major types of high energy density
of 7 – 9 are used to obtain smooth crack-free battery use substantial amounts of cobalt, and the
cobalt –tungsten deposits. Cobalt – molybdenum expected consumption of the metal for this
alloys can be obtained under similar conditions application is about 3000 t/a. All lithium ion
460 Cobalt and Cobalt Compounds Vol. 9

batteries use p-type semiconductor LiCoO2 as Table 22. Properties of cobalt oxide coated solar panels
the positive electrode active material to operate
Coating method Substrate Absorbance, % Emittance, %
in the range around 4 V that is unique to this type
of battery. The purpose of the lithium cobalt Plated Cu, Ni 93 28
oxide phase is to achieve charging and dischar- Plated Ni-plated steel 90 7
Plated/thermal
ging by the reversible exchange of Li between an decomposition Ag-coated steel > 90 20
LiC6 electrode and the Liþ ions in the oxide Plated Ni 95 –
lattice. The electrochemical reactions in Ni – Cd Plated Cu-coated steel 96 13
storage batteries involve the reduction of Ni(III) Oxidized paint Al 91 30
Paint Ni alloy 85 10
to Ni(II) and the oxidation of another metal Spray pyrolysis Al, galvanized Fe 91, 92 13, 12
species during discharge with the reverse reac- Spray pyrolysis Stainless steel 93 14
tions occurring during charging. When Co is
used as the other metal in the cell the discharge
reaction is:
Coþ2 NiOðOHÞþ2 H2 O ! CoðOHÞ2 þ2 NiðOHÞ2
paints. Table 22 gives absorbance and emittance
data for cobalt oxide coated solar panels prepared
The cobalt in these cells is part of a porous in a variety of ways on different substrates.
positive electrode formed by impregnating Ni Cobalt(II) oxide has been used in thermistors
powder on a perforated metal foil with co-pre- to improve both the resistivity and the tempera-
cipitated nickel and cobalt hydroxides. Cobalt is ture coefficient of the resistivity of the device. A
an important component of both electrodes of number of mixed-metal cobalt oxides have been
nickel/metal hydride rechargeable batteries. In used in devices, including NiCoO4 and La2CoO4
the hydride electrode, cobalt reduces the volume as anodic materials in water electrolysis,
change during the hydrogenation – dehydroge- NaCoO2 as cathodes in sodium batteries, and
nation cycles and is an important component of CoMoO4 as a fuel cell electrode.
the protective surface layer formed on the elec-
trode. Both functions improve the electrode sta-
bility by minimizing corrosion and decrepitation 7.3.6. Agriculture, Nutrition, and
and are essential factors in determining the cycle Medicine [21]
life of the cells. In the positive electrode the
presence of cobalt helps to reduce the effects of Cobalt chemicals are used to correct cobalt defi-
swelling on the sintered nickel electrodes. The ciencies in soils and in animals. Soil treatments
advantages of using cobalt along with nickel in usually involve top dressings containing cobalt
the Ni – Cd and Ni hydride batteries stem from sulfates, whereas treatment of ruminant animals
three factors: The similarities between the crystal involves the use of either salt licks containing ca.
structures of cobalt and nickel compounds, the 0.1 % of cobalt as sulfate, or concentrated feeds,
greater ease of oxidation of Co(II) compared with or pellets of cobalt oxide bound in an inert
Ni(II), and the higher electrical conductivities of material such as china clay.
cobalt oxide phases. The medicinal uses of cobalt are dominated by
Black cobalt oxide, Co3O4, is a selective the use of vitamin B12. The vitamin is obtained
coating material for high-temperature solar col- from the mother liquors of the microbial forma-
lectors. The cobalt oxide coatings are superior to tion of antibiotics such as streptomycin, aureo-
the black chrome coatings often used. The func- mycin, and terramycin after the removal of the
tion of cobalt oxide as a coating intended to antibiotic. Vitamin B12 and cobalt treatments are
operate up to 1000
C is to concentrate the solar used as remedies in certain types of anemia
radiation on the collector surface by a factor of up (! Vitamins, 6. B Vitamins). Cobalt salts,
to 2000. For this purpose, a high solar absorbance administered intravenously, have been used as
with low infrared emittance is required. Cobalt an antidote in cyanide poisoning.
oxide coatings can be prepared by plating, The mechanisms of the agricultural and
pyrolysis of cobalt salts on heated substrates, medicinal uses of cobalt chemicals are discussed
vacuum deposition, and by use of Co3O4-based in Chapter 10 of this article.
Vol. 9 Cobalt and Cobalt Compounds 461

Table 23. Annual production statistics for refined cobalt for major countries and producers (excluding production in CIS for internal
consumption)

Country Producer 1992 1993 1994 1995 1996 1997

USA OMG 2100 2200 3000 3610 4160 5000

Zambia ZMCC 4610 4211 2639 2934 4799 3949
Canada Falconbridge 2300 2414 2923 2804 3099 3417
Zaire/Congo Cecamines 5023 829 2439 3422 3540 2808
Bahamas ICCI 688 1218 1820 1730 2070 2250
Canada INCO 1465 1410 1130 1362 1544 1500
Australia QNI 657
Japan Sumitomo 105 190 161 222 228 263
Belgium 1000 1000 1000 1200 1200 1200
China 933 800 1076 1200 1200
RSA 234 243 246 190 292 294
Brazil 88 158 165 166 193 266
Morocco 80 220
France 121 144 141 161 174 159
India 110
CIS exports 1199 1960 2070 2787 1654 3200
Total 18933 16910 18534 21664 24233 26493

8. Analysis [25], [26] to about 33 % by 1998. The statistics for cobalt

production by the major producer companies and
Cobalt can be determined gravimetrically as an- by other countries are given in Table 23 for the
thranilate, as Co2O3 or Co after precipitation of the period 1992 – 1997. These supply figures can be
1-nitroso-2-naphthol complex and subsequent de- compared with the current demand for the metal
composition, and as a pyridine thiocyanate. Volu- (excluding demand in the CIS) of about 29 500 t.
metric methods include the use of 8-hydroxyqui- Estimated demand by end use for 1998 was:
noline, ethylenediamine tetraacetic acid (EDTA), superalloys (24.3 %), hardfacing and other alloys
and iron(III) cyanide solutions as titrants. In the (6.9 %), magnets (8.5 %), hardmetals (15.2 %),
case of complexometric methods, cobalt is often catalysts (8 %), pigments (11.5 %), batteries
back-titrated with magnesium or zinc solutions (9.5 %)), tires and driers (11 %), other chemicals
after an excess of EDTA has been added. The (5.1 %). Some forecasts suggest that the cobalt
metal can also be determined by electrodeposition consumption could double in the period up to
on perforated platinum electrodes. 2012 and that the price sensitivity to any such
Trace levels of cobalt in most materials can be increase will depend on the speed with which
determined by a number of instrumental techni- new sources of cobalt from laterite ores are
ques, including X-ray fluorescence, activation developed. The marginal cost of production has
analysis, atomic absorption, atomic emission, and been estimated at about $ 6 per pound. The cobalt
polarography. The reagents that can be used to price (for 99.8 % metal) has fluctuated from a low
measure cobalt by colorimetry include dimethyl- of about $ 7.8 per pound in 1990 to peaks of
glyoxime, dithizone, 8-hydroxyquinoline, nitroso- between $ 30 and $ 35 in 1992 and 1996 to
naphthols, diethyldithiocarbamate, and tripyridyl. $ 15 – 16 per pound in 1999.

9. Economic Aspects [4], [27] 10. Toxicology

Although in the third quarter of the 20th century, Distribution in the Environment. Data on
cobalt production was very dependent upon the the distribution of cobalt in the environment have
situation in African countries, this dependence been reviewed [29], [30]. Cobalt concentrations
has been eliminated with the increase in produc- in ambient air are exceedingly low; in remote
tion in the rest of the world. In the period 1970 – areas atmospheric concentrations are below
1979, over 70 % of available production came 1 ng/m3, in urban environments in the range of
from Zambia and Zaire, but this had been reduced 1 – 10 ng/m3, and in the vicinity of cobalt
462 Cobalt and Cobalt Compounds Vol. 9

processing industries above 10 ng/m3. The cobalt requirement of ruminants is 0.13 mg/kg feed d.m.
content of soils is in the range of 5 – 70 mg/kg; [32]. Additions of 2 – 4 mg/d to the feed of sheep
the average is estimated to be 8 mg/kg. In fresh and 20 mg/d to the feed of cattle are tolerated,
water, cobalt concentrations are generally low, whereas larger doses are toxic [29]. Vitamin B12
between 0.1 and 10 mg/L [29], but considerably deficiency in humans causes pernicious anemia.
higher in the vicinity of metal mining and proces- This is a very rare disease and has been observed
sing industries. Cobalt is mobilized into surface only with strict vegetarians after decades because
waters by weathering of rocks and transported of the long half-life of the vitamin in the body. The
into the oceans, where it is mainly coprecipitated symptoms of the disease are neurological and
with iron as hydroxide. Hence, the cobalt con- mental defects and low erythrocyte counts.
centration in ocean waters is very low (ca. 2 ng/L)
and cobalt is enriched in deep-sea sediments. Resorption, Distribution, and Elimination.
Radioactive cobalt isotopes, especially 60Co, After oral uptake, the absorption of cobalt occurs
have been released into the environment from in the small intestine by the iron uptake system.
nuclear power plants, radioactive waste dump- In the blood, cobalt ions are preferentially bound
ing, and nuclear accidents (Chernobyl). Because to serum albumin and distributed to the organs
health effects of 60Co are predominantly attrib- [29]. The average concentrations in human
utable to radiation and not chemical properties of organs are 0.09 mg/L in whole blood, 0.25 mg/L
cobalt, radioactive cobalt is not treated here and in urine, and 42.3 mg/kg in liver. The liver is
readers are referred to other sources, e.g., the U.S. the main location of vitamin B12; it contains
Toxicological Profile for Cobalt [30]. about 50 – 90 % of the total human vitamin B12
content [29].
Nutrition. Levels of cobalt in drinking After inhalation by exposed workers of cobalt
water usually are very low (0.1 – 5 mg/L) and metal dust or sparingly soluble cobalt com-
do not significantly contribute to the human pounds, these are taken up by lung macrophages,
supply of cobalt [30], [31]. The human dietary solubilized, and resorbed from the lung into the
intake of cobalt varies depending on the type of blood and excreted via the urine. Elevated inter-
food [30]. Relatively high concentrations of nal concentrations of cobalt have been found in
about 1 mg/kg dry matter (d.m.) are found in workers of the cobalt and hard metal industries
cocoa, tea, and coffee. Green vegetables and [30], [31]. Inorganic cobalt is excreted slowly via
unprocessed cereals have cobalt contents ranging the kidney with half-lives up to 90 d. Vitamin
between 10 and 600 mg/kg d.m., and meat and B12 is retained even longer with a half-life of
fish contain only 1 – 50 mg/kg d.m. The total about 400 d [29].
human intake of cobalt is estimated to be about
10 – 30 mg/d [29]. The cobalt content of nutri- Toxic Effects
ents has gained much attention because of the Oral Toxicity. Oral toxicity of cobalt com-
role of cobalt as the central constituent of vitamin pounds has been observed rarely. After adminis-
B12 (cobalamin). This vitamin is the only utiliz- tration of high doses of cobalt chloride (100 mg/d)
able form of cobalt, whereas inorganic cobalt is as a remedy for certain forms of anemia, adverse
of no use for mammalian health. Vitamin B12 side effects on the alimentary tract were observed
occurs almost entirely in food of animal origin, [30], [31]. Severe outbreaks of cardiomyopathy
and it constitutes only a small fraction of the were recorded among heavy consumers of beer
total cobalt in the human diet. The recom- fortified with cobalt as an foam-controlling agent
mended human intake for vitamin B12 is 2.5 – [30], [31]. Also, in experimental animals, the
5 mg/d, corresponding to only 0.1 – 0.2 mg Co/d acute oral toxicity of cobalt is moderate. The oral
[29], [30]. LD50 values for various inorganic cobalt salts
ranged from 150 to 500 mg/kg body weight (b.w.)
Essentiality and Toxicity for the anhydrous compounds, corresponding to
Essentiality. Inorganic cobalt is essential for 1.5 – 3 mmol Co/kg b.w. [30], [31].
the nutrition of ruminants, which require cobalt
for the bacterial biosynthesis of vitamin B12 in Skin Sensitization. Repeated dermal expo-
their rumen (first stomach). The normative cobalt sure of metal workers to cobalt and/or cobalt
Vol. 9 Cobalt and Cobalt Compounds 463

compounds has resulted in a contact dermatitis toxic but that cobalt enhances the genotoxicity of
manifesting as eczema and erythema [30]. other agents because it inhibits the repair of
damaged DNA [34]. Furthermore, the genotoxi-
Pulmonary Effects. The major concern in city of cobalt-containing hard metals is several-
cobalt toxicology are pulmonary diseases caused fold higher than the genotoxicity of the cobalt
by inhalation of cobalt and cobalt compounds, as component alone. This enhancing effect is attrib-
occurs in the production and processing of cobalt uted to a catalytic role of tungsten carbide in the
and cobalt compounds and in the production of formation of reactive oxygen species [30], [33].
hard metals composed of cobalt and tungsten
carbide. The symptoms include decreased pul- Reproductive Toxicity. There are no data
monary function, asthma, inflammation, and about the reproductive toxicity of cobalt in
ultimately lung carcinoma [30], [31], [33]. For humans. In experiments with male mice, high
the nonmalignant pulmonary effects, the U.S. oral doses of CoCl2 caused reduced sperm count,
Agency for Toxic Substances and Disease Reg- testes weight, and fertility [30], [34]. After mat-
istry has estimated a LOAEL (lowest observed ing of treated males with untreated females,
adverse effect level) of 0.0151 mg Co/m3 and a embryonic losses were registered. Also with
NOAEL (no observed adverse effect level) of male rats, oral administration of CoCl2 caused
0.0053 mg Co/m3 [30]. degeneration and necrosis of the testes and re-
duced spermatogenesis [30], [34].
Carcinogenicity. Few epidemiological
studies on cohorts of cobalt-exposed workers Environmental Toxicity. There are little
have been performed. They showed some evi- data about the environmental toxicity of cobalt
dence for lung carcinogenicity of cobalt and and cobalt compounds in microorganisms,
cobalt compounds [31], [34]. However, no firm plants, and animals except farm animals. Acci-
conclusions about the human carcinogenicity of dental feeding of several grams of cobalt to sheep
cobalt and cobalt compounds can be drawn be- have caused fatalities [29]. Extrapolating from
cause of small numbers, limited statistical confi- experimental data, it is probable that the levels of
dence, and co-exposure to other metals. Studies cobalt encountered in the environment do not
on workers employed in the production of hard exert significant toxicity on environmental spe-
metals revealed a significant correlation between cies. Of course, this conclusion does not apply to
exposure and incidence of lung cancer [33]. In biological effects of effluents and waste dumping
experimental inhalation studies with both rats sites of metal industries.
and mice, cobaltous sulfate evoked the formation
of lung tumors [35]. Regulations and Guidelines. The qualita-
The epidemiological and animal data are sup- tive classification of cobalt and cobalt com-
ported by mechanistic studies in vitro which pounds with respect to carcinogenicity is
reveal that cobalt compounds are weakly geno- summarized in Table 24. Whereas previously

Table 24. Classification of cobalt and cobalt compounds as carcinogens [30], [31], [33], [34]

Agency Substances Classification

IARC 1990 [31] cobalt and cobalt compounds 2B (possibly carcinogenic to humans)
EU 2004 cobalt dichloride 2 (considered to be carcinogenic for humans)
cobalt sulfate 2 (considered to be carcinogenic for humans)
DFG 2001 [34] cobalt and cobalt compounds 2 (considered to be carcinogenic for humans)
DFG 2004 [33] hard metal, composed of tungsten 1 (causes cancer in humans)
carbide and cobalt
Germany 2004 cobalt (elemental) 3 (possibly carcinogenic to humans)
soluble cobaltous salts 2 (considered to be carcinogenic for humans)
cobalt oxide and cobalt sulfide 3 (possibly carcinogenic to humans)
ACGIH 2003 cobalt, elemental and inorganic compounds A3 (confirmed animal carcinogen with unknown
relevance to humans)
464 Cobalt and Cobalt Compounds Vol. 9

Table 25. Synopsis of limit values for cobalt and cobalt compounds (as elemental cobalt)

Description Limit values Source

3 a
Occupational limit values for Cobalt and inorganic 0.5 mg/m TEL Germany 2004
compounds
3 b
0.1 mg/m TEL Germany 2004
0.1 mg/m3 PELc OSHA 2003
0.05 mg/m3 RELd NIOSH 2001
0.02 mg/m3 TLVe ACGIH 2000
Occupational biological exposure equivalent for 25 mg/L in blooda Germany 2004
cobalt and its compounds 300 mg/L in urinea
5 mg/L in bloodb Germany 2004
60 mg/L in urineb
Occupational biological exposure index for cobalt 15 mg/L in urine ACGIH 2000
and its compounds
1 mg/L in blood, both at end ACGIH 2000
of shift at end of work week
Food. Cobaltous salts and their derivatives Prohibited from use in human food FDA 2000
a
Technical exposure limit, production of cobalt powder, catalysts, hard metals, magnets.
b
Technical exposure limit, all other processes.
c
Permissible exposure limit.
d
Recommended exposure limit.
e
Threshold limit value.

some regulatory agencies have classified cobalt Development Institute, UK, 1985 – 1994 and Cobalt
and its compounds as a whole group, recent News, Cobalt Development Institute, UK 1995 –
classifications tend to classify cobalt com- 1999.
8 W. Betteridge, Prog. Mater. Sci. 24 (1979) 2.
pounds individually according to the degree of
9 J. D. Donaldson, S. M. Grimes, S. J. Clark: Cobalt in
evidence for carcinogenicity to humans. Ta- Chemicals, The Cobalt Development Institute, Guildford
ble 25 summarizes quantitative toxicity and 1986.
limit values issued by various agencies. This 10 R. W. Andrews: Cobalt: Overseas Geol. Surveys, Mineral.
table reveals considerable differences in quan- Resources Div., H. M. Stationary Office, London 1962.
titative risk estimation between agencies of 11 H. Hagon, Metall (Berlin) 29 (1975) no. 11, 1157 – 1158.
different countries. Obviously, there is a lack 12 H. Hagon, Met. Powder Rep. 31 (1976) no. 1.
of data with respect to quantitative effect – 13 J. D. Donaldson, S. J. Clark: ‘‘Cobalt in Superalloys,’’ The
Monograph Series Cobalt Development Institute, London
exposure relations and a need for harmonization
1985.
of regulations in the future. 14 S. Rimlinger: Proc. The Cobalt Conference, The Cobalt
Development Institute, Guildford 1999.
15 C. T. Sims et al.: Superalloys, vol. II, Wiley, New York
1987.
16 J. D. Donaldson, S. J. Clark, S. M. Grimes: Cobalt in
References Electronic Technology, The Cobalt Development Insti-
tute, Guildford 1998.
1 Cobalt Monograph, Centre d’Information du Cobalt, 17 M. S. Pierre, Cobalt News 1999, no. 2, 3.
Brussels 1960. 18 A. Magnee, J. M. Drapier, J. Dumont, D. Coutsouradis, L.
2 R. S. Young: ‘‘Cobalt,’’ ACS Monogr. Ser. 149, New York Habraken: Cobalt-Containing High-Strength Steels, Cen-
1960. tre D’Information du Cobalt, Brussels 1974.
3 W. Betteridge: Cobalt and its Alloys, Ellis Horwood, 19 Gmelin, vol. 58.
Chichester 1982. 20 B. F. G. Johnson, R. E. Benfield in G. L. Geoffroy (ed.):
4 ‘‘Cobalt Facts’’, Cobalt Development Institute, Guildford Topics in Inorganic Stereochemistry, Wiley, New York
UK — a regularly updated document. 1981, pp. 253 – 335.
5 N. N. Greenwood, T. C. Gibb: M€ ossbauer Spectroscopy, 21 J. D. Donaldson, S. J. Clark, S. M. Grimes: Cobalt in
Chapman Hall, London 1971. Medicine, Agriculture and the Environment, The Cobalt
6 Cobalt and Cobalt Abstracts, Centre d’Information du Development Institute, Guildford 1986.
Cobalt, Brussels 1958 – 1975. 22 J. D. Donaldson, S. J. Clark, S. M. Grimes: Cobalt in
7 S. M. Grimes (ed.): ‘‘Cobalt Abstracts,’’ ATB Metall. Batteries, The Cobalt Development Institute, Guildford
24 (1984) Cobalt Abstracts in ATB Metall., Cobalt 2000.
Vol. 9 Cobalt and Cobalt Compounds 465

23 J. D. Donaldson, S. J. Clark, S. M. Grimes, Z. Hussain, By-products; Some other Halogenated Compounds;

Cobalt News 1999, no. 4, 11. Cobalt and Cobalt Compounds), International Agency
24 J. D. Donaldson, S. J. Clark, S. M. Grimes: Cobalt in for Research on Cancer, Geneva 1991, pp. 363 – 472.
Catalysts, The Cobalt Development Institute, Guildford 32 M. Anke: ‘‘Essential and Toxic Effects of Macro-, Trace
1990. and Ultratrace Elements in the Nutrition of Animals’’ in E.
25 C. Tombu, Cobalt 1963, no. 20, 103 – 110; no. 21, Merian, M. Anke, M. Ihnat, M. Stoeppler (eds.): Elements
185 – 189. and Their Compounds in the Environment, Wiley-VCH,
26 M. Pinta: Modern Methods for Trace Element Weinheim, Germany 2004, pp. 305 – 341.
Analysis, Publ. Ann. Arbor Science, Ann Arbor, 33 Deutsche Forschungsgemeinschaft: ‘‘Hartmetall’’ in
Michigan, 1978. Deutsche Forschungsgemeinschaft (ed.): Gesund-
27 ‘‘Cobalt Statistics,’’ in Cobalt News 1999, nos. 1, 2. heitssch€adliche Arbeitsstoffe. Toxikologisch-arbeitsme-
28 Cobalt and the Environment, The Cobalt Development dizinische Begr€undungen von MAK-Werten, 39. Liefer-
Institute, Guildford 1994. ung, Wiley-VCH, Weinheim 2004.
29 G. Schrauzer: ‘‘Cobalt’’, in E. Merian, M. Anke, M. Ihnat, 34 Deutsche Forschungsgemeinschaft: ‘‘Cobalt und Cobalt-
M. Stoeppler (eds.): Elements and Their Compounds in verbindungen’’ in Deutsche Forschungsgemeinschaft
the Environment, Wiley-VCH, Weinheim, Germany (ed.): Gesundheitssch€ adliche Arbeitsstoffe. Toxikolo-
2004, pp. 825 – 839. gisch-arbeitsmedizinische Begr€ undungen von MAK-Wer-
30 Agency for Toxic Substances and Disease Registry: ten, 33. Lieferung, Wiley-VCH, Weinheim, Germany
Toxicological Profile for Cobalt, U.S. Department of 2001.
Health and Human Services, Washington 2004. (www. 35 NTP (National Toxicology Program, USA): ‘‘Toxicology
atsdr.cdc.gov.). and Carcinogenesis Studies of Cobalt Sulfate Heptahy-
31 IARC: ‘‘Cobalt and Cobalt Compounds’’;, in IARC drate in F344/ Rats and B6C3F1 Mice (Inhalation Stud-
(ed.) Monographs on the Carcinogenic Risk to Humans, ies)’’, NTP Technical Report 471, Research Triangle Park
vol. 52 (Chlorinated Drinking Water; Chlorination 1998.