Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                

Recovery of Nano-Sized Cobalt Powder From Cemented Carbide Scrap

Download as pdf or txt
Download as pdf or txt
You are on page 1of 8

Turkish J. Eng. Env. Sci.

29 (2005) , 343 350.


ITAK

c TUB

Recovery of Nano-Sized Cobalt Powder from Cemented Carbide


Scrap

Sebahattin GURMEN

Istanbul
Technical University, Metallurgical & Materials Engineering Department,

Istanbul-TURKEY
e-mail: gurmen@itu.edu.tr

Received 17.05.2005

Abstract
A process is developed to recover nano-sized cobalt from cemented carbide scrap (WC-Co) by nitric
acid leaching and finally obtaining a nano-sized spherical pure cobalt powder by ultrasonic spray pyrolysis
(USP). The optimization of the leaching and preparation of cobalt powders by USP are conducted within the
framework of an environment-friendly and economical process, enabling reduced waste streams. According
to the results obtained, it is possible to dissolve cobalt with high efficiencies (91.5%) via HNO3 leaching
of cemented carbide scrap. The optimum leaching conditions were 90 m, 25 C, 2 h, 0.5 M HNO3 , 900
rpm, and 1/10 solid/liquid ratio. The leach residue (WC-TiC) has a BET specific surface area of 1.08
m2 /g and finds has a wide range of applications in key technologies, varying from mechanical to chemical
industries. After leaching, nano-sized cobalt powder (300-700 nm) can be produced from Co(NO3 )2 by
ultrasonic spray pyrolysis with hydrogen reduction. The nano-sized powders can be used for the production
of new nano-structured thermoelectric materials, nanomagnets, nanosensors and nano-structured tungsten
carbide-cobalt powders.
Key words: Recovery, Hydrometallurgy, Ultrasonic spray pyrolysis, Nano-sized cobalt powder

Introduction
Nano-sized metallic powders with high purity offer
new physical and mechanical properties. The growing interest in the specific properties of nanocrystalline materials is supported by a strong tendency
towards the miniaturization of high-technology materials. Magnetic metal nanoparticles (such as Co,
Fe and Ni) have applications such as high-density
magnetic data storage, magnetic toners in xerography, magnetic inks and ferrofluids (Xicheng et al.,
2003). The increasing demand for metals worldwide has required intensive studies for the extraction of metals from low-grade ores and/or secondary
resources. Cobalt can be extracted from secondary
resources like chemicals (for rechargeable batteries),
super alloys, cemented carbides, and magnets. Most
scrap materials are richer in cobalt than even ore con-

centrates. Cemented carbides, often called Hard


Metals, are a range of very hard, refractory, wearresistant materials produced by powder metallurgy
techniques (i.e. sintering). These materials have
found a wide range of applications as cutting tool
inserts (about 50-60%), dies, nozzles and valve components, etc., where severe wear and high temperature properties are required. The majority of cemented carbides are based on WC and Co (Kang
et al., 2000; Ullmann, 1963; Enqvist and Uhrenius,
2003; Andren, 2001; Lin et al., 1996; Brookes, 1998).
When the tools and components are scrapped, they
are collected with the purpose of recovering the tungsten and cobalt in the light of their strategic importance. The $ 2 billion worldwide cemented carbide
industry generates large quantities of scrap due to
parts rejected at various stages of production and
worn out cutting tools. Numerous recycling pro343


GURMEN

cost of the apparatus and raw materials (Messing,


1993; Kim et al., 2003; Tsai, 2004). Spherical nanosized cobalt powders were obtained by the USP of
aqueous cobalt nitrate solutions followed by thermal
decomposition of the aerosols generated in hydrogen
atmosphere.
In this study, a novel process was developed for
the recycling of cobalt from cemented scrap carbides.
Finally, cobalt is recovered in the form of spherical nano-sized cobalt powder (by USP), which is the
most important product today, for the production
of new nano-structured thermoelectric materials and
nano-structured tungsten carbide cobalt powders.

cesses have been developed that can be classified into


different categories. The Zinc melt and the ColdStream processes are also popular methods to recycle cemented carbide (Ullmann, 1963; Lin et al.,
1996; Brookes, 1998; ASM Handbook, 1998; Lin et
al., 1995). Regarding the Zinc melt process, the
binding metal (cobalt) reacts with high purity zinc
(99.995%) during the cemented carbide recovery operation. The energy consumption is therefore high;
about 4000 to 6000 kWh is required for 1 t of tungsten carbide scrap. The cycle time of the complete
zinc process is 24 h. The Cold-Stream process is
a mechanical method where cemented carbide scrap
is transported by a high velocity air stream (>1000
km/h) to hit a cemented tungsten carbide plate. Nitric acid is used to dissolve the binding medium (anodic dissolution of cobalt) at a current density of
2-10 A kg1 of the scrap (an electrochemical process) (Lin et al., 1995; Lin et al., 1996; Latha et al.,
1989). Cobalt leaching from cemented carbide scrap
is a hydrometallurgical method for cobalt recovery
and offers the potential of environment-friendly management. Selective leaching is a potential alternative
in view of its lower energy consumption and environmental impact.
Ultrasonic spray pyrolysis (USP) is a useful tool
for the large or small scale production of particles
with controlled particle size and has been used to
prepare nano-sized powders of metallic and intermetallic compounds, and ceramic materials. USP
has advantages such as simplicity, safety and low

Experimental
Materials
Experiments were carried out with powderized cemented carbide scrap. The chemical composition
and X-ray diffraction analysis of the cemented tungsten carbide scrap powder are given in Table 1 and
Figure 1.

Table 1. Chemical composition (wt%) of the cemented


tungsten carbide scrap powder used in the experiments (particle size -90 m).

W
75.86

Co
8.14

C
6.07

Ti
4.18

20000
WC
TiC

18000

20-1047 WC
05-0727 Co
73-0472 TiC
06-0696 Fe

WC

16000

Fe
3.40
100

80

14000
12000

60

Count

Co
WC

10000

TiC

8000
6000

Co
Fe

4000

40

Co
Fe

TiC

WC

WC

WC

Co
20

TiC

2000
0

0
10

20

30

40

50

60

70

80

Figure 1. X-ray diffraction analysis of cemented carbide scrap powder (-90 m).

344

Nb
1.35

Ni
1.00


GURMEN

By the X-ray diffraction analysis, it was found


that the structure of cemented carbide scrap is WC,
TiC, and Co.
Experimental procedure
Cemented tungsten carbide scrap was ground in a
vibrating mill for 15 min. Screening of the ground
material was conducted in a Retzsch brand sieve.
Leaching experiments were carried out in a 3-neck
glass reactor with 0.25 l capacity, heated by a thermostatically controlled heater (Heraeus). The mixture was stirred from the top with a teflon-plated
propeller, which was hooked onto a mixer (Ika-Werk
RW-20) with adjustable speed and stirring direction.
A spiral condenser was also utilized to return evaporation losses to the system. The ground cemented
carbide scrap powder was added to the agitated nitric acid solution at the required temperature. Samples of the solution were taken at 30 min intervals.
All the samples were filtered and analyzed by ICP
(Spectroflame Modula EOP). The carbon content of
the cemented carbide scrap was determined using
a Leco-CS 400 analyzer. The surface area of WCTiC (mixed powder) was measured by BET equipment (Flow Sorb II 2300). An X-ray diffractometer (Siemens D 5000) and a scanning electron microscope (Zeiss DSM 982 Gemini) were utilized for
the characterization of the material and product.
All the chemicals used in the experiments were analytical grade (Merck). Generally, the experiments
were conducted under the following conditions: a
solid/liquid ratio of 1:10, 20 g sample and 90 m
particle size. A purified leach solution from cobalt

extraction experiments from cemented tungsten carbide scrap using nitric acid was used as the starting
material for the USP experiments. The final concentration of cobalt was 0.08 Mol Co/l. A spray
pyrolysis system was used in this study, consisting
of an ultrasonic atomizer, a quartz tube reactor, a
furnace and collection units. Very fine droplets of
the aerosol were obtained in an ultrasonic atomizer
Pyrosol 7901 (R.B.I. with a frequency of 800 KHz).
The aerosol was transported with reducing gas into
the quartz tube (0.7 m length and 0.02 m diameter)
placed in the furnace (Str
ohlein, Germany) with a
temperature control 1 C. Hydrogen atmosphere
was used for cobalt production. Nitrogen with a flow
rate of 1 l/min was used for the air removal prior to
the reduction process. Under spray pyrolysis conditions in hydrogen atmosphere and at a flow rate
of 1 l/min, the dynamic reduction took place in the
quartz tube reactor. Figure 2 shows the schematics
of the apparatus.
Results and /iscussion
Leaching
Effect of leaching temperature: The temperature effect was examined in the range 25-70 C under the
standard conditions of -90 m particle size, 0.5 M
HNO3 , 2 h, 900 rpm, and 1/10 solid/liquid ratio.
Figure 3 shows the effect of temperature on the dissolution of cobalt in the HNO3 and H2 O system.
Cobalt solubility decreased from 91.5% to 78.2% between 25 and 70 C.

Flowmeter

Stirrer
Condenser

Quartz Tube

Thermometer

Furnace
Co-Solution

N2

H2

Solution
Cooling
Water

Leaching

Thermostat

Outlet

Collection
Chamber

Ultrasonic Atomizer

Figure 2. Schematics of the experimental apparatus for the synthesis of cobalt nanoparticles.

345


GURMEN

leach residue to determine the re-utilization possibilities in the Hard Metal Industry (or Powder Metallurgy), revealed that it has a density of 11.9 g/cm3 ,
a BET value of 1.08 m2 /g and a grain size of -90 m.
Effect of reaction time: The highest cobalt efficiency (91.5%) was obtained after a leaching period of 2 h (Figure 5). Iron, on the other hand, dissolves within the first hour of leaching with 90% efficiency and shows a dissolution trend similar to that
of cobalt for the rest of the leaching. This shows that
the reaction period and rate depend on the liberation rate of the particles in ground cemented carbide
scrap. In contrast to the fact that cobalt gives a very
fast dissolution reaction, nickel shows a slow reaction
with HNO3 and acts as the rate-determining factor
on total efficiency.
In practice, 120 min of leaching was optimum.
Extended leaching periods (over 120 min) seem to increase the efficiency for cobalt, iron, and nickel only
insignificantly.
Effect of initial acid concentration: Figure 6 displays the effect of acid concentration on the dissolution of cobalt at 25 C. Various experiments were
carried out by changing the HNO3 concentration
from 0.5 M to 1.5 M. At 0.5 M HNO3 concentration,
91.5% cobalt extraction was obtained in 120 min,
whereas at 1.0-1.5 M HNO3 concentration cobalt extraction of approximately 90.5-89.9% was achieved
within 120 min, as shown in Figure 6. Increasing the
acid concentration from 0.5 M to 1.5 M caused the
dissolution efficiency of iron to decrease from 94.5%

100

Extraction (%)

80

60
40
Co
20

Fe
Ni

0
25

40

55

70

Temperature (C)

Figure 3. Effect of temperature on the dissolution of


cobalt and associated elements [-90 m, 2 h,
0.5 M HNO3 , 900 rpm, and S/L: 1/10].

The results shown in Figure 3 indicate that rising temperature has no advantageous effect on the
leaching rate, and cobalt should be extracted at
room temperature. The leach residue was washed
with distilled water. X-ray diffraction analysis of
the leach residue (for the optimum leaching conditions) is given in Figure 4. While the main structure shows typical WC and TiC characteristics, some
cobalt peaks observed in the X-ray analysis indicate
that additional grinding might be necessary in order to liberate all the cobalt present in the scrap.
The characterization studies, conducted on WC-TiC

100
WC
TiC

30000

25-1047 WC
05-0727 Co
73-0472 TiC

WC

80

25000
Co

Count

20000

60
WC

Co

15000

40
WC

10000
WC

TiC

WC
20

WC
TiC

5000
Co

WC TiC
0

0
10

20

30

40

50

60

70

80

Figure 4. X-ray diffraction analysis of the leach residue (the optimum leaching conditions).

346


GURMEN

creasing stirring speed are given in Figure 7. The


highest extraction appeared to be at 900 rpm.
100
80
Extraction (%)

to 90.0%. On the other hand, nickel extraction increased with increasing HNO3 concentration. The
viscosity of the solution increased as a result of high
acid concentration. This leads to a reduction of the
diffusion rate of the ions. The reaction rate remained
constant for higher acid concentrations, indicating
stabilized iron dissolution. The main determining
factor for the rate of leaching, as mentioned earlier,
is the continuing reaction of cobalt particles that are
entrapped in WC particles.

60
40
Co
20

100

Fe
Ni

0
700

Extraction (%)

80
60
40

900

Figure 7. The effect of stirring speed on the dissolution


of cobalt and associated elements [-90 m, 25

C, 2 h, 0.5 M HNO3 , and S/L: 1/10].

Co
Fe

T= 25 C

20

Ni
0
0

30

60

90

120

150

180

Reaction Time (min)

Figure 5. Effect of leaching time on the dissolution of


cobalt and associated elements [-90 m, 25 C,
0.5 M HNO3 , 900 rpm, and S/L: 1/10].
100

Extraction (%)

800
Stirring rate (rpm)

80
60
40

It is known that higher stirring rates often result in an increased dissolution rate. In diffusioncontrolled processes, the rate of dissolution can be
expressed as a function of stirring speed. In this case,
the process is diffusion controlled and the boundary
layer thickness decreases with increased speed of stirring.
Effect of particle size: The effect of particle size
was studied using 2 ground size fractions of -90 m
and
-250 + 90 m at 25 C, 2 h, 0.5 M HNO3 , 900
rpm, and 1/10 solid/liquid ratio. Smaller particle
sizes result in a more efficient leaching of cobalt. As
seen from Table 2, leaching efficiencies are lower than
those obtained with a particle size of 90 m, which
shows the effect of particle size.

Co
20
0
0.5

Fe
Ni
1
Acid Concentration (M)

1.5

Figure 6. The effect of acid concentration on the dissolution of cobalt and associated elements [-90 m,
25 C, 2 h, 900 rpm, and S/L: 1/10].

Effect of stirring speed: The effect of the stirring speed was studied under the conditions of -90
m particle size, 25 C, 0.5 M HNO3 , 2 h, and 1/10
solid/liquid ratio. The results show that the leaching
rate of the cobalt increases quickly when the stirring
speed is amplified from 700 to 900 rpm. Extraction
percentages of Co, Fe, and Ni in relation to the in-

Table 2. Effect of particle size on the dissolution of


cobalt and associated elements [25 C, 2 h, 0.5
M HNO3, 900 rpm and S/L: 1/10].

Particle size
-90 mm
-250 + 90 mm

Co %
91.5
68.5

Fe %
94.5
49.1

Ni %
50.7
54.1

Effect of solid/liquid ratio: Figure 8 gives the


cobalt recovery during leaching, as a function of the
S/L ratio. According to these experimental cobalt
extraction increases with decreasing S/L ratio and
at ratios higher than 1:10 there is no significant increase in cobalt recovery.
347

Co-Extraction (%)

GURMEN

100
80

ions in aqueous solution with hydrogen under pressure. USP is a powerful tool for preparing powders
and films, because of the easy control of the target
composition, the excellent availability of the precursors, and the high quality of the products. In the
preparation of a powder by USP, a solution is atomized into a hot reactor, where the aerosol droplets
undergo drying, droplet shrinkage, solute precipitation, thermolysis, and sintering to form the final particles. Metals, metal oxides, and non-oxides can be
readily produced by spray pyrolysis. Figure 9 shows
the X-ray diffraction pattern of the USP cobalt powder. The X-ray diffraction analysis of the powder,
produced at 800 C from Co(NO3 )2 solution in H2
atmosphere by USP, indicated the formation of pure
cobalt powder.
A typical SEM micrograph of the cobalt powder
obtained from cobalt solution (5 g/l) under dynamic
conditions at 800 C is shown in Figure 10. As seen,
the particle size of the powder varies between 388.0
and 490.7 nm, although larger particles (about 700
nm) are also observed.
Ideally, spherical non-agglomerated cobalt powders are produced by USP. This powder could be
suitable for Powder Metallurgy and Hard Metal Industry applications. In this study, a novel process
was developed for the recycling of fine, spherical
and nano-sized cobalt powder from cemented carbide
scrap by leaching and USP.

S/L:1/10

60
40
20

S/L:1/20

0
0

30

60
t (min)

90

120

Figure 8. The effect of S/L ratio on the dissolution of


cobalt [-90 m, 25 C, 0.5 M HNO3 , and 900
rpm].

Synthesis of nano-sized spherical cobalt powder by USP


The composition of the homogenized leach solution
for the USP experiments (analyzed by ICP Spectroflame Modula EOP) is shown in Table 3, which
indicates that Fe and Ni are two 2 impurities. The
cobalt is the major component to be recovered.
Table 3. Chemical composition of leach solution.

Initial solution
mg/l

Co
5000

Fe
10

Ni
8

Cobalt powder can be produced by a number of


methods, but those of industrial importance involve
the thermal hydrogen reduction of oxides, the pyrolysis of carboxylates, and the reduction of cobalt

100
15-0806 Co

3000

80

Count

2800

2600

60

2400

40

2200
20
2000
0
30

40

50

60

70

80

Figure 9. X-ray diffraction analysis of the USP cobalt powder [reduction periods: 2 h, flow rate of H2 : 1 l/min].

348


GURMEN

Figure 10. SEM micrograph of the cobalt powder obtained under dynamic conditions at 800 C
[2 h, 1 l/min H2 ].

2 h of leaching, 0.5 M HNO3 , 900 rpm, and S/L


ratio of 1/10. Cobalt can be more efficiently dissolved when fractions of 90 m particle size are
used. The particle size of the cemented carbide scrap
can easily be decreased below 90 m through classical grinding processes. The highest leaching efficiency was attained at 25 C and temperature does
not seem to affect the leaching efficiency of cobalt.
Stirring speed, however, has a determining role in
leaching. The obtained leach residue (WC-TiC) has
a density of 11.8 g/cm3 , BET specific surface area
of 1.08 m2 /g and grain size of -90 m. Nano-sized
cobalt powder (300-700 nm) can be produced from
Co(NO3 )2 by ultrasonic spray pyrolysis with hydrogen reduction. A total of 2 h reduction periods are
required for cobalt powder. Hydrogen gas plays a
significant role in a pyrolysis reactor as carrier and
reducing gas. No liquid waste (neither acidic nor alkaline) is generated in this process, since Co(NO3 )2
solution is recycled back to the USP step.

Conclusions

Acknowledgments

The recycling of cemented carbide scrap is becoming


increasingly attractive. For this purpose, the suggested method, combining, leaching and USP, represents an alternative to the classical recycling techniques. The optimum leaching conditions -90 m, 25

C,

The author would like to thank the Alexander


von Humboldt Foundation for its research fellowship at IME Process Metallurgy and Metal Recycling, RWTH Aachen University, Germany, where
this work was performed in the context of a process
development for recycling hard metal scrap.

References
Andren, O.H., Microstructure Development during
Sintering and Heat-treatment of Cemented Carbides
and Cermets, Materials Chemistry and Physics, 67,
209-213, 2001.
ASM Handbook, Powder Metal Technologies and
Applications, Production of Refractory Metal Powders, 7, 192-201, 1998.
Brookes, K., Hardmetals and Other Hard Materials, International Carbide Data, East Barnet 1998.
Enqvist, H. and Uhrenius, B., Determination of the
Average Grain Size of Cemented Carbides, International Journal of Refractory Metals & Hard Materials, 21, 31-35, 2003.
Kang, Y.K., Roemer, G.J. and Ghosh, D., Microstructural Characterization of Cemented Carbide
Samples by Image Analysis Techniques, Powder
Technology, 108, 130-136, 2000.
Kim, H.J., Babushok, I.V., Germer, A.T., Mulholland, W.G., and Ehrman, H.S., Cosolvent-Assisted

Spray Pyrolysis for the Generation of Metal Particles, Journal of Materials Research, 18, 1614-1622,
2003.
Latha, M.T. and Venkatachalam, S., Electrolytic
Recovery of Tungsten and Cobalt from Tungsten
Carbide Scrap, Hydrometallurgy, 22, 353-361,
1989.
Lin, C.J., Lin, Y.J., and Jou, P.S., Selective Dissolution of the Cobalt Binder from Scraps of Cemented
Tungsten Carbide in Acids Containing Additives,
Hydrometallurgy, 43, 47-61, 1996.
Lin, J.C., Lin, J.Y. and Lee, S.L., Process for
Recovering Tungsten Carbide from Cemented Carbide Scraps by Selective Electrolysis, US Patent,
5.384.016, 1995
Messing, L.G., Zhang, C.S. and Jayanthi, V.G.,
Ceramic Powder Synthesis by Spray Pyrolysis,
Journal of the American Ceramic Society, (76),
2707, 1993.

349


GURMEN

Tsai, C.S., Song, L.Y., Tsai, S.C., Yang, C.C., Chiu,


Y.W. and Lin, M.H., Ultrasonic Spray Pyrolysis
for Nanoparticles Synthesis, Journal of Materials
Science, 39, 3647-3657, 2004.
Ullmanns Encyclopedia of Industrial Chemistry,
A12, Urban & Schwarzenberg M
unchen-Berlin, 603614, 1963.

350

Xicheng, M., Yuanhua, C., Ning, L., Qing, A.,


Shenli, L., Fengzhao, L. and Shulin, W., Microstructural Features of Co-Filled Carbon Nanotubes, Materials Letter 57, 2879-2884, 2003.

You might also like