Equilibrium studies in the Co-Cr-C system: Solubility of Cobalt in M23C6 and M3C2

Bachelor Thesis

Stockholm, Sweden 2011

Equilibrium studies in

the Co-Cr-C system

Solubility of Cobalt in M

C

and M

C

KARIN EKSTRÖM

SHANAR KORDBACHE

1

2 2011

Karin Ekström Shanar Kordbache

Bachelor Thesis MMK 2011:x CMATD3

Equilibrium Studies in the Co-Cr-C System

Karin Ekström Shanar kordbache

Approved

2011-05-05

Examiner

Anders Eliasson

Supervisor

Malin Selleby

Commissioner

Department of Materials Science and Engineering

Supervisor at Sandvik Susanne Norgren

Abstract

This project concerns cemented carbides and the carbides that are formed when adding chromium. Cemented carbides are composites, often consisting of tungsten-carbide particles embedded in a cobalt-rich matrix, and are because of their extreme hardness used in for example cutting tools and drills. Chromium is sometimes added when making cemented carbides in order to lower the melting point, reduce grain growth and/or increase corrosion resistance. When adding chromium there is a risk of forming unwanted carbides such as M

C

, M

C

and M

C

. It is therefore of great interest to know the stability of these carbides.

The purpose of this work was to investigate the solubility of Co in M

C

and M

C

by equilibrium studies.

The aim was to produce samples equilibrated in the three-phase regions between liquid‐

M

C

‐M

C

and M

C

‐M

C

‐graphite, to study the solubility of Co in M

C

and M

C

respectively.

Initial studies were performed at Sandvik Mining and Construction (SMC) to determine the

compositions of the samples to be produced and temperatures for the heat treatments.

3 The alloys were heat‐treated at 1450⁰C for three days and were thereafter investigated with LOM and XRD. The work was carried out in collaboration with Sandvik Mining and Construction (SMC).

It was difficult to analyze the results with XRD since the intensity peaks in the diffractograms are close or overlapping for the M

C

and M

C

carbides. The solubility of Co in M

C

could not be investigated accurately.

The M

C

-M

C

‐graphite sample did not reach equilibrium in the three days of heat treatment.

The conclusions that can be drawn from this project are that further work, using longer

annealing times, has to be done in order to get more knowledge about the Co solubility in

M

C

and M

C

carbides.

4

5

Acknowledgments

We would like to thank our supervisors Susanne Norgren at Sandvik and Associated Professor Malin Selleby at Department of Materials Science and Engineering, KTH.

Special thanks to Andreas Markström (Thermo-Calc Software AB) and Bartek Kaplan

(Sandvik Tooling) who helped us with the experimental part and guided us through our

project.

6

7

Nomenclature

Designations

Symbol Description

C Celsius

K Kelvin

M Co, Cr

Abbreviations

Co Cobalt

Cr Chromium C Carbon

LOM Light Optical Microscopy

XRD X-ray Diffraction

SEM Scanning electron microscope EDS Electronic Data Systems

SMC Sandvik Mining and Construction WC Tungsten carbide

PEG Polyethylene glycol

8

ABSTRACT ... 2

ACKNOWLEDGMENTS ... 5

NOMENCLATURE ... 7

1 INTRODUCTION ... 9

1.1 Aim ... 10

1.2 Literature survey ... 11

1.3 Background ... 13

1.3.1 Manufacturing Cemented Carbides ... 13

1.3.2 Milling ... 13

1.3.3 Pressing ... 13

1.3.4 Sintering ... 14

1.3.5 Characterization methods ... 14

1.3.5.1 Light Optical Microscope (LOM) ... 14

1.3.5.2 XRD ... 14

2 EXPERIMENTAL PROCEDURE ... 15

2.1 Solubility of cobalt in M

C

... 15

2.1.1 Sample preparation ... 15

2.1.2 Sample analysis ... 16

2.1.3 Results ... 17

2.1.3.1 LOM ... 17

2.1.3.2 XRD ... 18

2.1.4 Discussion ... 18

2.2 Solubility of cobalt in M3C2 ... 19

2.2.1 Sample preparation ... 19

2.2.2 Results ... 19

2.2.3 Discussion ... 19

3 CONCLUSIONS AND FUTURE WORK ... 20

4 REFERENCES ... 21

9

1 Introduction

Cemented carbides are composite materials, composed of hard particles of a refractory carbide ceramic such as tungsten carbide (WC) or titanium carbide (TiC), embedded in a metal matrix such as cobalt or nickel. The first cemented carbide was a WC-Co-alloy. The hard, brittle tungsten particles were cemented with a soft, ductile cobalt-rich binder.

Cemented carbides are manufactured from powdered raw materials by liquid phase sintering.

Cemented carbides are used for their hardness, wear resistance and temperature stability. The main applications for cemented carbides are cutting or drilling tools and other wear parts, where the material is expected to have high hardness as well as some toughness.

The hard carbide particles provide the hardness needed for cutting or drilling but, being extremely brittle, are not in themselves capable of withstanding the stresses during such operations. The toughness comes from the inclusion of carbide particles in the ductile metal matrix, which isolates the particles from one another and prevents particle to particle crack propagation. Both particulate and matrix phases are to some extent refractory, to withstand the high temperatures generated by the cutting action on hard materials

.

Chromium is often added to cemented carbides to inhibit grain growth during sintering and to increase the corrosion resistance along with mechanical strength of the binder by solution hardening

. However, when adding Cr there is a risk of forming unwanted carbides such as M

C

, M

C

and M

C

. It is therefore of great interest to know the stability of these carbides.

In previous work

, the solubility of cobalt in M

C

has been studied and it was found that the solubility of Co in the M

C

carbide was much higher than the values assessed by thermodynamic calculations.

In this project equilibrium studies was made to investigate the solubility of Co in the M

C

and M

C

carbides at temperatures higher than previous studies. In figure 1, the compositions

of the samples produced in this work are presented. In order to find the Co solubility in M

C

and M

C

, samples with compositions between the points in the phase diagram should be

produced.

10

Figure 1- The ternary system C-Co-Cr, calculated at 1723 K, with the desired compositions for the samples in this work ^iv.

1.1 Aim

The aim of this work was to increase the knowledge on the ternary Co-Cr-C system. The aim was to produce samples in equilibrium between liquid- M

C

- M

C

for the investigation of the solubility of cobalt in M

C

.

For the study of the solubility of cobalt in M

C

the compositions of the samples were targeted to be within the three phase equilibrium between M

C

-M

C

-graphite.

Initial studies were performed at Sandvik Mining and Construction (SMC) to determine the

sample compositions and temperatures for the heat treatment. The alloys were to be heat

treated and thereafter investigated with LOM and XRD.

11

1.2 Literature survey

To gain the information needed to do this project a literature survey was made. The studied articles concern the Co-Cr-C system.

A Thermodynamic Evaluation of the Co-Cr and the Co-Cr-C Systems

Alexandra Kusoffsky and Bo Jansson

CALPHAD vol. 21, number 3, (1997), p 321-333

The thermodynamic properties of the systems Co-Cr and Co-Cr-C was analyzed with the CALPHAD technique using a computerized optimization procedure. The Gibbs Energy of the stable phases was evaluated applying the compound energy model. A good agreement with a set of selected consistent experimental phase diagram and thermo chemical data was obtained.

Investigation of (Co, Cr)

C

-fcc-Graphite Equilibrium in the Temperature Interval 1373 to 1473K

Thérèse Sterneland, Andreas Markström, Susanne Norgren, Ragnhild E. Aune and Seshadri Seetharaman, Metallurgical and Materials Transactions A, vol. 37, (2006,) number 10, p. 3023-3028

The purpose of this article was to investigate the solubility of cobalt in M

C

as well as the solubility of Cr in the Co rich fcc-phase, since earlier studies have indicated higher solubility than the ones calculated. Heat treatments of appropriate mixtures of Cr

C

and Co were performed at 1373, 1423 and 1473 K. The samples were then quenched in liquid nitrogen and examined by SEM and XRD.

From the results, the compositional regions of the three-phase triangle M

C

+FCC+graphite were determined. The results showed that the Co solubility in the Cr

C

within the experimental temperature interval was higher than previous investigations performed at higher temperatures.

Experimental and Thermodynamic Evaluation of the Co-Cr-C system

Andreas Markström, Susanne Norgren, Karin Frisk, Bo Jansson, Thérèse Sterneland

Int.J.Mat.res 97(2006)9

12 In this study the solubility of cobalt in M

C

in equilibrium with liquid was investigated and a thermodynamic description of the system was made. Heat treatments were performed at 1523 and 1723 K. The samples were analyzed with EDS, WDS and SEM and based on the experimental results the Co-Cr-C system was analyzed with the CALPHAD technique.

The results of this study showed a higher solubility of cobalt in M

C

than previous investigations.

Equilibrium study of chromium containing cemented carbides

Solubility of chromium in tungsten carbide and η-phase

Bonnie Lindahl, Master of Science Thesis, Stockholm, Sweden, 2010

In this study the W-C-Co-Cr quaternary system at low carbon contents was studied. The aim was to find the four phase equilibrium between WC, Co-binder phase; η-phase (M

C) and a fourth unknown phase, and study the solubility of chromium in WC and establish the fourth phase. The first step was choosing a number of alloys which compositions were calculated with thermodynamic calculations and then analyzed by LOM, SEM, EDS, WDS and XRD.

The result showed that no four phase equilibrium could be found. The solubility of chromium was proved to be much higher in the η-phase than previous investigations indicated.

Thermodynamic Modeling of Carbides in Multicomponent Systems

Andreas Markström , Licentiate thesis 2009, ISRN KTH/MSE- -09/04- - SE+THERM/AVH SE-100 44 Stockholm

ISBN 978-91-7415-237-1

This thesis was about thermodynamic modeling of carbides in multicomponent systems with focus on systems interesting for cemented carbide production.

A re-assessment of the Co-W-C system was presented and new experimental results on the maximum solubility of Co in the M

C

were defined along with a new thermodynamic description of the Co-Cr-C system which accurately describes the temperature interval 1373- 1723 K.

Also, experimental work on the C-Co-Ti-V-W-Zr system was done in order to determine the

extension of the miscibility gaps in TiC-ZrC and VC-ZrC into the (TiC or VC)-ZrC-WC

13 system. The new experimental information was used to assess the thermodynamic description for the TiC-ZrC system.

1.3 Background

Cemented carbides consist of a group of very hard and wear resistant materials that are used in cutting tools, drilling parts etc. The high hardness comes from a high fraction of carbides, typically 90 % or higher. The carbide particles, which have sizes from 10 µm to 100 nm, are bonded together by a metallic aggregate, typically cobalt.

Cemented carbides are fabricated by powder metallurgy. Casting a melt with the right

composition would require very high temperatures in order to get a homogenised melt and the solidification process would not give the required microstructure

.

1.3.1 Manufacturing Cemented Carbides

There are different ways to fabricate powders. One method is Comminution of Solid Materials. In this process the chosen material is crushed and separated from other materials and then mixed with carbon. The mix is then sent into a furnace. The new material from the furnace is crushed again and the wanted material is sorted out.

Another process to produce powder of metal is using electrolysis, which is really expensive.

In the Atomization Process the metal is first melted and then separated into small droplets.

The droplets will then rapidly be cooled down with either water or gas (argon or nitrogen)

.

1.3.2 Milling

When the desired raw materials have been chosen, the first stage is milling, in order to get a powder with a homogenous composition and to reduce the particles to the desired size. The simplest ways of milling are ball milling and wet milling. Ball milling consists of a cylindrical jar filled with solid cemented carbide milling bodies, which are added together with the powder. When the cylinder rotates, the milling bodies will fall down and crush the powdered material

. Ethanol is added during the milling in order to protect the powder from oxidation together with polyethylene glycol (PEG), which acts as a binder when the powder is later compacted providing sufficient strength for the compacted body. Before the powder can be used it will be spray dried with warm nitrogen which gives spherical granules and prevents further agglomeration

.

1.3.3 Pressing

The powder, with particles with sizes around 1 µm or less, is hard to work with because it will

form agglomerates. To facilitate the procedure a controlled agglomeration will be performed

which gives spherical granules. Each granule is glued together by an additive. The powder

will then be easier to handle and can be pressed into a mold. If the powder does not keep the

14 shape after the pressing the procedure has to be redone which costs money and gives a poor result

.

1.3.4 Sintering

Sintering of a compacted powder body most often results in a dense and pore-free material, depending on the sintering conditions. The first step is to remove the PEG-additive from the compacted body, so-called debinding. The PEG-additive is burnt off at low temperatures, and the actual sintering is performed at 1350-1500 C. At that temperature the cobalt melts. During sintering the porosity is eliminated through different diffusion processes which lead to a linear shrinkage of about 18%

. The driving force in sintering is the reduction of surface energy when the granules are agglomerated. There are two ways of sintering. One of them is called Solid Phase Sintering, where particles are bound together through necking at elevated temperatures below the melting temperature of the component with the lowest melting point.

Liquid Phase Sintering implies that part of the material is melted. In the case of cemented carbides the Co-rich binder phase melts. Because part of the material has melted, a more extensive densification is possible. When sintering cemented carbides, both solid phase sintering and liquid phase sintering are used

.

1.3.5 Characterization methods

1.3.5.1 Light Optical Microscope (LOM)

LOM is a type of microscope which uses visible light and a system of lenses to magnify images. The image from a light optical microscope can be captured by normal light-sensitive cameras to generate a micrograph. LOM is the easiest characterization method and is used to investigate different microstructures. A LOM is demonstrated in figure 2.

Figure 2 - image of a light optical microscope.

1.3.5.2 XRD

X-ray diffractometry (XRD) based on the elastic scattering of X-rays from electron clouds of

the individual atoms in the system. By observing the scattered intensity of an X-ray beam

hitting a sample as a function of incident and scattered angle, polarization, and wavelength or

15 energy a material can be characterized based on thickness, present phases and their crystallographic structure, chemical composition and physical properties (for example residual stresses).

2 Experimental procedure

2.1 Solubility of cobalt in M

C

2.1.1 Sample preparation

According to observations from previous studies, 1410 C is the minimum heat treatment temperature needed to be able to study the equilibrium in liquid+M

C

+M

C

. The result from earlier experiment showed that this alloy contains the solid phase (fcc) instead of M

C

. In previous work at 1500⁰C and 3 hours heat treatment, equilibrium has been reached in liquid+M

C

+M

C

, but the proportion of melt to solid material has been high.

On the basis of these results, the samples in this project were heat treated at 1450⁰C for 3 days. The compositions of the samples were determined by previous studies and are presented together with the sintering times in table 1.

The samples were prepared by weighing Cr

C

, Co and Cr powders to get the right composition. The raw materials were mixed by vibration in order to get a homogenous powder.

Table 1- Shows the compositions and sintering times and temperatures for the liq. +M23C6+M7C3 samples.

The powder was not milled in this project, since it could easily have been contaminated in the milling process.

The powder was pressed into pellets in a PTC (Powder Testing Center), which is a small scale press for laboratory experiments. No additives were added since there was no demand on the samples keeping their shape during sintering.

The heat treatments were performed in a STA 409 CD furnace at 1450⁰C for three days with

stationary argon. With flowing argon the samples could have oxidized even with the very low

oxygen level in the argon. The samples were then cooled by filling the furnace with cold gas.

16 The cooling rate was set on 50 degrees per minute. The cooling procedure is demonstrated in table 2.

Table 2- Cooling rate after sintering

The sintered samples were hot-mounted in Bakelite to make them easier to analyse with LOM. The samples were then ground with abrasive paper and polished with a diamond slurry in order to get smooth samples with as few scratches as possible.

Before analysing the samples, they were etched with Murakami’s etching reagent, which consists of 10g potassium ferricyanide, 10g sodium hydroxide and 100 ml water.

2.1.2 Sample analysis

The prepared samples, for detemening the solubility of chromium in M

C

were analyzed with LOM and XRD.

LOM was used to investigate the microstructure in the samples.

The XRD analysis was carried out to investigate which phases that had formed after the heat

treatments. The XRD model that was used was a PanAnalytical X'Pert Pro Multipurpose

Diffractometer.

17

2.1.3 Results

2.1.3.1 LOM

In figure 3-8 the microstructure, gained from LOM, of the liq+ M

C

+M

C

samples are presented. The black phase could be M

C

, the dark gray M

C

and the liquid has solidified in a eutectic reaction of fcc (light gray), M

C

and M

C

. According to the LOM images Sample AM400 is the only sample in the equilibrium with the melt.

AM400

Figure 3-AM400-etching line Figure 4-AM400

AM401

Figure 5-AM401 Figure 6-AM401- etching line

18 AM402

Figure 7-AM402 Figure 8-AM402 etching line

2.1.3.2 XRD

The samples were difficult to analyze in the XRD. However it seemed that most samples contained large amounts of M

C

but since the intensity peaks for M

C

are close to or overlap those for M

C

it is difficult to distinguish them from each other. From looking at the XRD results it looked like some Cr-rich oxide had formed and that some of the original raw material (Cr

C

remained in the samples.

2.1.4 Discussion

It was hard to analyze the results from the experiments since the XRD results were so limited.

To get better results that were easier to interpret, the samples should probably have been crushed and then mixed with a standard material. A well-defined standard material would be a good reference when comparing the results with structures in the database.

The LOM images were hard to analyze without accurate XRD. However when looking at figure 3-8 it is reasonable to assume that the AM400 sample is the only one that surely reached equilibrium with liquid.

The sample with the highest Co content should contain the most liquid. This means that

sample AM402 should have the highest amount of liquid phase (see table 1). Therefore it is

possible that samples AM400 and AM402 were mixed up during the sample preparations or

the heat treatments.

19

2.2 Solubility of cobalt in M3C2

2.2.1 Sample preparation

The heat treatments of these samples have been too short to reach equilibrium between M

C

+M

C

+graphite (3 hours at 1500⁰C). In this work longer heat treatments were performed. The samples would probably need a heat treatment at least for a week but in this work the samples were heat treated together with the liq+M

C

+M

C

samples, for three days at 1450⁰C. The compositions of the samples were determined by previous studies and are presented together with the sintering times in table 2.

The samples were prepared (mixed, pressed and sintered) in the same way as the liquid+M

C

+M

C

samples.

Table 3; Shows the compositions and sintering times for the M3C2+M7C3+graphite samples

2.2.2 Results

Only one of the samples was sintered, sample10AM500, because of problems with the furnace. However it was clear that the 10AM500 sample (M

C

+M

C

+graphite) had not reached equilibrium. The sample was impossible to polish and prepare for further analysis due to incomplete sintering.

2.2.3 Discussion

It was clear that the M

C

+M

C

+graphite sample that was sintered did not reach equilibrium.

To reach equilibrium in the samples the time of the heat treatment would probably have to be

doubled.

20

3 Conclusions and future work

The conclusions made from this work are first and foremost that the liquid+M

C

+M

C

samples should be further analyzed in order to investigate the solubility of Co in the M

C

carbide. In future work the samples should be analyzed with SEM to get clearer images than the LOM images and new XRD tests should be made.

For the solubility of Co in the M

C

carbide it is clear that three days heat treatment is not

enough to reach equilibrium between M

C

+M

C

+graphite. In future work the heat

treatments should probably be about a week.

21

4 References

i William D. Callister, Jr, Materials Science and Engineering An Introduction, 7th edition, USA, 2007

ii A. Kusoffsky and B. Jansson, A Thermodynamic Evaluation of the Co-Cr and the Co-Cr-C Systems, Calphad, vol. 21, (1997), p. 321-333

iii T. Sterneland, A. Markström, S. Norgren, R. Aune and S. Seetharaman, Investigation of (Cr;Co)7C3-fcc- Graphite Equilibrium in the Temperature Interval 1373 to 1473K, Metallurgical and Materials Transactions A, vol. 37,(2006) number 10, p. 3023-3028

iv A. Markström, S. Norgren, K. Frisk, B. Jansson and T. Sterneland, Experimental and thermodynamic evaluation of the Co-Cr-C system, Int.J.Mat, vol 97, (2006), number 9, p. 1243-1250

v M. Hillert, J. Ågren and A. Borgenstam, Mikro och nanostrukturer i materialdesign, Institutionen för Materialvetenskap, Kungliga Tekniska Högskolan, 2005

vi Randall M. German, Powder Metallurgy & Particulate Materials Processing, 2005

vii Bonnie Lindalh, Master of science thesis, 2010

viii Björn Uhrenius, Pulvermetallurgi, Stockholm, 2000

22