• No results found

Behavior of cutting tool coating material Ti1-xAlxN at high pressure and high temperature

N/A
N/A
Protected

Academic year: 2021

Share "Behavior of cutting tool coating material Ti1-xAlxN at high pressure and high temperature"

Copied!
47
0
0

Loading.... (view fulltext now)

Full text

(1)

Department of Physics, Chemistry and Biology

Bachelor’s thesis

Behavior of cutting tool coating material Ti

1-x

Al

x

N at high

pressure and high temperature

David Dilner

LITH-IFM-G-EX--09/2209—SE

Department of Physics, Chemistry and Biology Linköping University, SE-581 83 Linköping, Sweden

(2)
(3)

Department of Physics, Chemistry and Biology

Behavior of cutting tool coating material Ti

1-x

Al

x

N at high

pressure and high temperature

Carried out within the Nanostructured Materials Group at IFM, LiU In collaboration with the R&D department at SECO Tools AB

David Dilner

12 September, 2009

Supervisors

Prof. Magnus Odén

Dr. Mats Johansson

Examiner

Prof. Magnus Odén

(4)
(5)

Datum Date 12 September, 2009 Avdelning, institution Division, Department Nanostructured Materials

Department of Physics, Chemistry and Biology Linköping University

URL för elektronisk version

Rapporttyp Report category Licentiatavhandling Examensarbete C-uppsats D-uppsats Övrig rapport _____________ Språk Language Svenska/Swedish Engelska/English ________________ ISBN ISRN: LITH-IFM-G-EX--09/2209—SE _________________________________________________________________ Serietitel och serienummer ISSN

Title of series, numbering ______________________________

Titel

Title

Behavior of cutting tool coating material Ti1-xAlxN at high pressure and high temperature

Författare

Author David Dilner

Sammanfattning

Abstract

The high pressure and high temperature (HPHT) behavior of Ti1-xAlxN coatings on cutting tool inserts have been of interest for this diploma work. A literature study of HPHT techniques as well as measurement methods has been done. A diamond anvil cell (DAC) would be a good device to achieve high pressure and high temperature conditions on small samples. Another way to obtain these conditions would be a cutting test, which has been performed on a Ti1-xAlxN coated cutting tool insert with x = 0.67. Also a cubic press could be used to apply HPHT on a Ti1-xAlxN sample or a large volume press on a whole cutting tool insert. To measure hardness on thin coatings a nanoindentor could be used, which have been done on heat-treated Ti0.33Al0.67N and TiN samples. X-ray diffraction (XRD) is a suitable method to measure phase composition of a sample and was performed on the cutting tested insert as well as on an untreated reference insert. Three ways to continue this project have been outlined all starting with more comprehensive cutting tests.

Nyckelord

Keyword

(6)
(7)

Acknowledgements

I would like to thank my main supervisor and examiner Magnus Odén for giving me this opportunity and also for all his help and support.

Moreover I would like to thank Mats Johansson my supervisor at Seco Tools AB and Axel Knutsson at the Nanostructured Materials group for all their help during this

diploma work. I would also like to thank Mats for his help with XRD and depositions and Axel for his help with nanoindentation.

Also I would like to thank Lina Rogström and all other kind people at the Nanostructured Materials group for their help and interesting discussions.

Last but not least I would like to thank my friends and family for their support and understanding while I was doing this work.

(8)

Sammanfattning

Under detta examensarbete har fokus legat på fasförändringar hos ytbeläggningar av Ti 1-xAlxN vid högt tryck och hög temperatur (HPHT). En litteraturstudie om HPHT samt analystekniker har genomförts under arbetet. Diamantstäd (DAC) förfaller vara en lämplig metod för att uppnå HPHT på små prov. Ett annat sätt att uppnå dessa förhållanden är ett skärprov med ett Ti1-xAlxN-belagt skärverktyg, något som också genomförts på Ti0.33Al0.67N. Även en kubisk press skulle kunna användas på Ti1-xAlx N-prov alternativt även en storskalig press på ett helt Ti1-xAlxN-belagt skär. För att mäta hårdheten på tunna beläggningar används lämpligen en nanoindentor, detta utfördes på värmebehandlade Ti0.33Al0.67N- och TiN-prover. Röntgendiffraktion (XRD) kan användas för att bestämma fassammansättningen hos ett kristallint materialprov. XRD användes för att bestämma fassammansättningen på ett skärtestat skär samt ett obehandlat referensprov. Tre möjliga vägar att fortsätta detta projekt är skissade och alla dessa startar med mer ingående skärprover tillsammans med XRD undersökningar.

(9)

Abstract

The high pressure and high temperature (HPHT) behavior of Ti1-xAlxN coatings on cutting tool inserts have been of interest for this diploma work. A literature study of HPHT techniques as well as measurement methods has been done. A diamond anvil cell (DAC) would be a good device to achieve high pressure and high temperature conditions on small samples. Another way to obtain these conditions would be a cutting test, which has been performed on a Ti1-xAlxN coated cutting tool insert with x = 0.67. Also a cubic press could be used to apply HPHT on a Ti1-xAlxN sample or a large volume press on a whole cutting tool insert. To measure hardness on thin coatings a nanoindentor could be used, which have been done on heat-treated Ti0.33Al0.67N and TiN samples. X-ray diffraction (XRD) is a suitable method to measure phase composition of a sample and was performed on the cutting tested insert as well as on an untreated reference insert. Three ways to continue this project have been outlined all starting with more comprehensive cutting tests.

(10)

Abbreviations

DAC – Diamond Anvil Cell FIB – Focused Ion Beam HP – High pressure

HPHT – High Pressure and High Temperature PTM – Pressure Transmitting Media

PVD – Physical Vapor Deposition SEM – Scanning Electron Microscopy STP - Standard Temperature and Pressure XRD – X-Ray Diffraction

Comment about units

When possible SI-units are used.

Notations

• c-TiAlN is written [rocksalt]-TiAlN • c-TiN is written [rocksalt]-TiN • c-AlN is written [rocksalt]-AlN • h-AlN is written [wurtzite]-AlN

• Ti1-xAlxN is the notation used for the general [Ti, Al]N coating system and could thus contain the phases [rocksalt]-TiAlN, [rocksalt]-TiN, [rocksalt]-AlN and [wurtzite]-AlN

(11)

Content

1 INTRODUCTION...1

2 BACKGROUND ...2

2.1 COATING MATERIAL TI1-XALXN ...2

2.2 PHASE TRANSFORMATIONS IN GENERAL...4

3 MEASUREMENT METHODS ...5

3.1 X-RAY DIFFRACTION (XRD)...5

3.2 NANOINDENTATION...5

3.3 RESISTANCE MEASUREMENT...7

4 HIGH PRESSURE HIGH TEMPERATURE (HPHT) TECHNIQUES...8

4.1 DIAMOND ANVIL CELL...8

4.1.1 Pressure Generation...9

4.1.2 Heating Device ...10

4.1.3 Pressure Transmitting Media (PMT)...11

4.1.4 Sample Preparation...12

4.1.5 Pressure and Temperature Determining ...12

4.1.6 In situ Measurements...14

4.1.7 Classification of DACs ...14

4.2 HIGH TEMPERATURE NANOINDENTATION...16

4.3 LARGE VOLUME APPARATUS...16

4.4 CUTTING TEST...18

5 RESULTS ...19

5.1 NANOINDENTATION OF TIN AND TI1-XALXN...19

5.2 CUTTING TEST...19

6 DISCUSSION AND CONCLUSION...22

7 REFERENCES...26

8 PICTURE SOURCES...31

9 APPENDIX...32

9.1 APPENDIX A–HPHT TECHNIQUES SUITABLE FOR TI1-XALXN ...32

9.2 APPENDIX B–DAC FOR USING ON TI1-XALXNSAMPLES...33

9.2.1 Purchasing DACs ...34

(12)

1 Introduction

Hard protective coatings on tools for metal cutting applications were introduced in the late 1960s and can greatly enhance the tool life1. The cutting tool inserts themselves are typically made by hard ceramic materials1, often carbides. A common coating system used is Ti1-xAlxN, often used as a hard and wear resistant coating on tungsten carbide (WC:Co) cutting tool inserts. The overall quest with such tools is to enhance the tool life by, e.g., improving its temperature resistance allowing for higher cutting speed data and thus faster production and less cost.

The actual metal cutting operation is in fact a very harsh operation during which, and mainly due to the friction between the cutting insert and the work piece material, the temperatures of the insert locally at the cutting edge rises over 1200 K. At the same time a high pressure is induced at the contact area. Such working environment certainly affects the tool performance, often in a bad way, e.g., by coating oxidation, etc. that severely degenerating the tool life. Today the knowledge of the wear mechanisms and coating structure evolution at the tools cutting edge is very limited. It is thus important to address this knowledge gap and propose solutions for the development of next generation cutting tools.

This diploma work, performed at Nanostructured Material Group at Linköping University and in cooperation with research department at Seco Tools AB, is a first approach towards this goal. The project aims to investigate the possibilities to study the relation of cubic structured phases ([rocksalt]-TiN, [rocksalt]-AlN and [rocksalt]-TiAlN) with respect to the less favorable hexagonal [wurtzite]-AlN phase in Ti1-xAlxN coated cutting tool inserts as a function of hot static pressure control, i.e., at conditions similar to that during machining and to suggest solutions for the same. This approach has previously not been applied to the Ti1-xAlxN system.

(13)

2 Background

2.1 Coating Material Ti

1-x

Al

x

N

TiN is well established as a coating material on cutting tools2-5 and classified as the first generation of hard coatings3. TiN has a cubic crystal structure called rocksalt-structure. The hardness of TiN is usually around 25 GPa2-5, however, hardness values between 30 and 40 GPa have been reported5, 6. The shortcoming of TiN is due to its corrosive behavior at temperatures above about 800 K3, 7. In effect, oxidation of TiN causes a TiO2 layer at the surface of the TiN7 which greatly reduces the films hardness8.

Figure 1 Quasi-binary phase diagram for TiN and AlN

Introducing Al in TiN, i.e., towards a Ti1-xAlxN composition enhances the oxidation resistance of the coating8, 9. There is, however, a solubility problem with Al in TiN. This can be seen observed in the qausi-binary phase diagram of TiN and AlN in Figure 1. The maximum solubility of AlN in TiN is reached at about 2700 K and is only about 5 %10, 11. This issue is, however, less of a problem for coatings grown by physical vapor deposition (PVD) methods where the growth temperature can be kept quite low10. Typical PVD methods for industry includes magnetron sputtering deposition and reactive cathodic arc vapor deposition11 for which the growth temperatures of Ti1-xAlxN is about 800 K9, 12, 13. This temperature is high enough to maintain the plasma but low enough to hinder

(14)

diffusion during depositions allowing the aluminium atoms to be trapped in the cubic [rocksalt]-TiN lattice10, 11. This cubic [rocksalt]-TiAlN phase is not represented in the phase diagram above since this is not a thermodynamically stable phase, i.e., it is said to be metastable11. In fact, PVD techniques, especially cathodic arc evaporation allows an Al-content in the TiN lattice phase far beyond the above mentioned 5 %8, 10, 11 and Ti 1-xAlxN coatings with x as high as 0.66, containing only the [rocksalt]-TiAlN phase, have been produced14. Typically, Ti1-xAlxN coating is reported to have a higher hardness than TiN8, 14.

Figure 2 Hardness for Ti0,34Al0.66N and TiN as a function of annealing temperature

What happens if Ti1-xAlxN coatings are subject to elevated temperatures?

In contrast to the oxidation effect of TiN and a decrease in hardness, Ti1-xAlxN coating improve its hardness up to about 1250 K and decreases above 1250 K14, see Figure 2. The mechanisms behind this behavior are believed to follow the decomposition and/or phase separation of Ti1-xAlxN. Data suggests that the initial decomposition, i.e., up to about 1250 K is of a spinodal type rather than the usual nucleation-and-growth12. The practical difference is that a periodical composition fluctuation forms instead of growing nuclei of the precipitated phase11. The single cubic phase [rocksalt]-TiAlN decomposes to cubic [rocksalt]-TiN and metastable cubic [rocksalt]-AlN15. During this process, the coating hardness improves due to coherency strain between the as-formed cubic Al rich and Ti rich areas, respectively hindering crack propagation in the coating. This phenomenon is commonly referred to as the age-hardening effect in analogy with age-hardening in metallic alloys16. The second transition occurs above 1250 K where the metastable [rocksalt]-AlN transforms into stable hexagonal [wurtzite]-AlN reducing the coating hardness12, 14.

(15)

2.2 Phase Transformations in General

Two phases can coexist under certain p-T conditions given by the Clausius-Clapeyron equation17: (2.1) v T L dT dp Δ =

Where L is the latent heat of transition, p the pressure, T the temperature and Δv is the

volume change for the two phases.

The most common phase transformation in solids is nucleation and that occurs when ΔG

for the phase transformation is negative16: (2.2) ΔG= rGv +4 rGs

3

π

Where r is the radius of the nucleus, ΔGv is the volume free energy and ΔGs is the surface

free energy.

When two, or more, substances are mixed the entropy rises, this is since a mixture is more chaotic then the pure substance. The entropy of mixing SM is given by18:

(2.3) SM =−Nk

(

xlnx+(1−x)ln(1−x)

)

Where N is the total number of atoms, k is Boltzmann’s constant, x is phase content A and (1-x) is phase content B.

As mentioned earlier [rocksalt]-TiAlN is a metastable phase and will hence decompose to [rocksalt]-TiN and metastable [rocksalt]-AlN if enough energy is added. However, this transformation does not follow the nucleation and growth mechanism but rather a spinodal decomposition mechanism, as mentioned above.

(16)

3 Measurement methods

3.1 X-ray Diffraction (XRD)

For measuring the phase composition of the material X-ray diffraction (XRD) is good analysis method. The working principle behind the XRD is to measure the distance between lattice planes, this works because the wavelength of the X-rays is in the same order as the interplanar spacing. The principle of diffraction is that the waves, in this case X-rays, are reflected by lattice points, in this case atoms, and at certain wavelengths the reflected beams reinforce each other. The interplanar spacing is calculated using Bragg’s diffraction law19: (3.1) d n 2 sinθ = λ

Where λ is the wavelength, n is an integer, d is the interplanar spacing and θ is the half angle between the incident beam and diffracted beam. XRD could be either angle-dispersive or energy-angle-dispersive20. In angle-dispersive XRD, which is the most common, the sample is rotated and the diffraction peaks at certain angles are obtained16, 20. In energy dispersive XRD the sample is still and the energy, which is inversely proportional to the wavelength, is varied and diffraction peaks for certain wavelengths are obtained20. For a powder, or polycrystalline material, the X-rays is reflected by those crystals whose planes that have their normal at an angle θ from both the X-ray source and the X-ray detector16. XRD is widely used in materials research, there are several examples when Ti1-xAlxN have been studied using XRD9, 21 and as will be discussed further down XRD is used for in situ studies at HPHT20.

3.2 Nanoindentation

Nanoindentation uses a small sharp diamond tip to penetrate into the material in order to measure hardness, elastic modulus and other mechanical properties22. A rule-of-thumb is that the intender tip should not penetrate more then 10 % of the films thickness11, 23, this means that percussions have to be taken during nanoindentation and also that ordinary hardness tests are unsuitable for thin films. Nanoindentation is an important tool to investigate the mechanical behavior of hard coatings. This technique has been performed

(17)

several times on Ti1-xAlxN3, 14, 24, 25. The hardness can be calculated using equations 3.2 and 3.323: (3.2) A P H = max (3.3) A=24.5h2 +δ

Where Pmax is the maximum load, h is the total displacement and δ is correction factors

for the tip shape. Figure 3 shows a nanoindentation load-displacement curve. Figure 2 in section 2.1 shows hardness versus annealing temperature curve for Ti1-xAlxN is presented and in that figure corrections have been made for residual stresses14.

Figure 3 Load displacement curve for Nanoindentation

Hardness is not a straight forward property of the material, instead the hardness depends on both how the material is processed as well as under what conditions the hardness test is performed11. Thus, hardness data is not universal for a specific material system. Therefore, both the hardness data of the investigated material as well as that of reference materials should be obtain under as equal conditions as possible. This is why new hardness data for Ti1-xAlxN as well as for TiN is measured during this project as well as for comparison with TiN/Ti1-xAlxN-multilayers, like those done by Knutsson et al26. The hardness data is collected for annealed samples, like in Hörling’s study14, using a UMIS nanoindentor equipped with a Berkovich diamond tip. The coated substrates was sliced, molded into bakelite, polished in steps down to 1 µm using liquid diamond in the final step as well as de-magnetized and ethanol-cleaned before mounting. An average hardness was calculated for about 30 indents for each sample, to avoid effects of materials defects. Fitting and calculation was performed using the IBIS software, utilizing the evaluation method developed by Oliver and Pharr.

(18)

3.3 Resistance Measurement

The conductivity depends on how easy the free electrons can move in a material. For the electrons in a metal there is no distinction between the conduction band and the valance band which offers good conductivity. For semiconductors and isolators there is a band gap between these two bands that need to be overcome for the semiconductor/isolator to conduct electricity. Generally, at higher temperature, lattice vibrations make it harder for the electron to move freely16, and hence increasing the resistance. In opposite, increasing the pressure reduce the restistance27. Moreover, the conductivity of alloys is much lower then it is for a pure metal16. Different phase in an alloy usually have different conductivity and this property can be used to detect phase transformations. A good example of this is the conductivity difference between [rocksalt]-TiAlN and the combination of [rocksalt]-TiN and [wurtzite]-AlN, used by Hörling to detect phase transformations14, see Figure for analogy with electric circuits. Also polycrystalline materials have less conductivity then single crystal materials because the grain boundaries act as obstacles for the electrons. If temperature and/or pressure is varied the conductivity will very however if the temperature and/or pressure change causes a phase transformation the conductivity will change more rapidly in that particular pressure-temperature regime then it would if no phase transformation occurred. This is basically how resistance can be use to detect phase transformation and if the resistance for a particular phase is known it could be conclude what kind of phase transformation that have occurred. Resistance measurements have been used for detecting phase transformations in Ti1-xAlxN12 and in HPHT research using a diamond anvil cell28 as well as using a cubic press29.

(19)

4 High Pressure High Temperature (HPHT) techniques

Most everyday phenomenon, like rain or plant growth, occurs quite near standard temperature and pressure (STP). However there are several cases where the temperature and pressure, separately or together, greatly exceed what is to be considered normal. These cases could be natural like in the interior of the earth. Such conditions could also be created by man, for instance in a jet engine where temperature is high, under a skate where the pressure is high or during a cutting application where both the temperature and pressure are high. The conditions of high pressure and high temperature (HPHT) are of scientific interest both for fundamental understanding and for industrial applications. For Ti1-xAlxN coatings on cutting tools, we are interested in the HPHT behavior of the coating due to its similarities with the environment during metal cutting.

4.1 Diamond Anvil Cell

Diamond anvil cells (DAC), schematically illustrated in Figure 5, saw daylight in the late 1950s20, 30. The DAC is a device which can generate very high pressures in a small sample20. Pressures up to 500 GPa, which is far beyond what is interesting for this study, have been obtained with this technique20. The DAC basically consists of two opposing diamond, a pressure generating device and a metal gasket to be put between the diamonds20. A hole, smaller then the diamond culets, is drilled in the gasket and the sample is placed into this hole together with a pressure transmitting media20, 31. The length across samples should be around 50 - 100 µm20, 32. The DAC technique is probably the easiest method to obtain a high pressure on a sample33. Laser-heated DAC’s are reported with lower investment and operational costs then other HPHT techniques32. It could be mentioned that laser-heating is more complicated to use then other heating methods.

(20)

Figure 5 Principal sketch of a diamond anvil cell

4.1.1 Pressure Generation

The pressure on the sample is often applied through a pressure transmitting media placed in the DAC chamber, which is in the gasket hole and between the diamond culets. The pressure that can be generated is greatly depending on the gasket hole and culet size. The maximum obtainable pressure is inversely related to the hole and culet size. However, since there is a lot uncertainty about the deformation of the gasket, geometrical methods does not give accurate estimation of the pressure inside the DAC31. Usually spectroscopic methods are used on certain reference materials, often ruby34. The diamond culets are typically larger then the gasket hole which is commonly a few hundred micrometers20. The maximum attainable pressure can be estimated by the following formula31:

(21)

(4.1) d D h C P − ≈ 2 max

Where is h is the gasket thickness, D the culet diameter, d the gasket hole diameter and C is a proportionality factor.

The force applied on the anvils can be generated in different ways, with a screw mechanism, a spring-lever mechanism or a gas-membrane mechanism31. The advantage with the gas-membrane mechanism is that is more easy to operate and can be controlled remotely.

4.1.2 Heating Device

Many high pressure experiments performed in DACs is done at room temperature, which makes the ability to heat the sample redundant. For materials studies concerning cutting tools, however, the ability to heat the sample is crucial trying to mimic a true cutting scenario. This is also the case in mineralogy as well as in the study of materials for applications in engines, energy systems and other high temperature operations. The DAC technique offers several options for heating the sample20, 28, 31, 35. The DAC could be used up to about 900 K without special percussions, but higher temperatures require special attention to the selection of engineering material36. Also, the applied HPHT conditions may cause the diamond to transform back into graphite34. When heating the cell externally this is a problem at temperature above 900 K when the cell is heated in air and above 1600 K in an inert environment20. Internal resistance heating could generate sample temperatures above 2000 K and laser heating temperatures above 6000 K20. Laser heating makes it possible to obtain extreme conditions similar to those in the interior of the earth. Laser-heated DACs have been used for example when studying the resistance properties of olivine mineral (Fe0.125,Mg0.875)2SiO4 at 35 GPa and 3450 K28, in a Raman scattering study of NaBi(WO4)2 at 12.2 GPa and 1658 K37 and in a X-ray diffraction study of alumina at HPHT conditions up to 135 GPa and 2350 K38. When using laser-heated diamond anvil cells, graphite powder can be added to absorb the heat from the laser39.

(22)

The use of an internal resistance heating, as an alternative to laser heating, have been reported to reach temperatures up to 2000 K and more20. Internal heating means just heating the sample without affecting the diamonds20. However this method is, as the laser heating, complicated, since the sample chamber is quite small. Gasket heating is something in between internal and external heating. The UK based company easyLab has gasket heaters that can be used to heat the sample up to 1250 K40.

Even though laser heating and internal resistance heating offer higher maximum temperatures then gasket heating they seem to be more complicated to use, and since they are not marked in any found DAC-supplier, they would most certainly be more expensive to purchase.

4.1.3 Pressure Transmitting Media (PMT)

The pressure induced by the diamond anvils must be transmitted to the sample in order to obtain the wanted conditions and for that purpose, a pressure transmitting media (PTM) is used. The pressure transmitting media works best as a fluid20, which, on the other hand, means that a different pressure media will be suitable for different pressure and temperature ranges. For example, a methanol/ethanol mixture is a suitable pressure-transmitting media below 10 GPa at room temperature 20. At higher pressures, this media starts to solidify. Nobel gases can keep their hydrostaticity up to 25 GPa and helium to at least 50 GPa20. As an example, a mixture of water/methanol/ethanol was used in the HP (room temperature) experiments of AlN. This pressure transmitting media kept its hydrostatic pressure up to 14 GPa41. Similar, HP (room temperature) analysis of a MAX phase material (Cr0.5V0.5)2GeC, was preformed up to 49.5 GPa with a methanol/ethanol pressure transmitting media42. There are, however, other examples where a pressure transmitting media is not used, e.g., in a study over several polycrystalline ceramic materials at HASYLAB28, 43. Cryogenically loaded inert gases could be used as pressure transmitting media, for example N2 have been used when studying TiN33, other inert gases that could used is Ne, He and Ar32, 44. N2 which becomes a solid at 2.3 GPa (at room temperature) have the advantage of giving a high N2 chemical potential which makes nitrides more chemically stable, even at high temperatures32. Note, however, that

(23)

an advanced gas-loading system is needed to use gases as PTM44. Also alkali halides such as NaCl could be used as PMT32, 36, as in a Raman scattering study of NaBi(WO4)2 in a laser-heated DAC37.

4.1.4 Sample Preparation

As mention earlier the sample is usually around 50 – 100 µm across, but they have to be even smaller if pressure in the hundred GPa range is to be generated20. On the other way around they can be larger if lower pressures are to be obtained. The sample is usually smaller in one direction, normally about 10 – 20 µm32. The general idea is to do the experiments on Ti1-xAlxN with as little interaction with the surrounding as possible. In effect, this means that the sample (coating) should be separated from the substrate. A way to obtain a Ti1-xAlxN powder is to grow it on a steel plate, or other metal having a low acid resistance, and then to dissolve the steel plate in a strong acid. One suggestion would be to use hydrochloric acid which dissolves the iron as ferrous chloride, while the Ti 1-xAlxN stays unaffected. However, according to easyLab, the sample needs to be in one piece45 and hence the powder needs to be pressed into a pellet which could affect the microstructure. Another way to prepare the sample is with a method called focus ion beam (FIB) cutting out a sample in one peace.

4.1.5 Pressure and Temperature Determining

As mentioned in the pressure generation section, florescence methods are typically used for calibrating the as-obtained pressure in DACs. For example, a small ruby grain is placed in the DAC chamber together with the sample and the PMT34. The change of the R1,2 doublet shift Δλ in ruby is measured by Raman spectroscopy31 In the pressure range 20-30 GPa, the pressure vs. doublet shift is almost linear and the pressured is obtained by equation 4.231, 34:

(4.2) p= 742. ×Δλ

Where Δλ is measured in nanometers and p is obtained in GPa. Ruby have been calibrated, using gold as reference material, up to 200 GPa34, which resulted in the calibration curve presented in Figure 6.

(24)

Figure 6 Ruby calibration curve, doublet change as a function of pressure

Even though ruby florescence offers a well established and reliable method for pressure calibration, it has the disadvantage of being temperature dependent44 which in turn limits its use in HPHT research. The florescence of SrB4O7:Sm2+ would be more suitable for calibrations at high temperature since it is, more or less, unaffected of changes in temperature31. In this case, the pressure is calculated from equation 4.331:

(4.3) p= 923. ×Δλ

An alternative method to florescence is to use fixed-point scale involving freezing temperatures31. This method, however, would limit the calibration opportunities and thus the understanding where phase-transitions occur.

Moreover, the temperature also needs to be measured during a HPHT experiment. This is usually done by spectroradiometry meaning that the temperature is calculated from the black body radiation curve46. In this way, the temperature T is obtained from the Planck radiation function 4.446: (4.4) 1 ) exp( ) ( I λ 2 5 1 − ⋅ ⋅ = − T C C λ λ ε

W Where C1 = 3.7418·10-16 W/m2 and C2 = 1.4388·10-2 m·K are constants37, I(λ) is the

intensity, λ the wavelength and ε the emission factor which is 1 for an ideal black body and less the 1 for a grey body. In order to achieve an accurate value of the temperature, the intensity should be measured at several wavelengths46.

(25)

4.1.6 In situ Measurements

Some phases only exist in a certain pressure and temperature domain thus the ability to make measurements in situ is valuable for study of HPHT properties. For metastable materials, however, such as cubic Ti1-xAlxN, the HPHT induced phase transitions towards phases with a lower free energy would not be reversed when pressure and temperature is normalized. However the DAC offers several options for in situ measurements and in situ measurements may give more exact knowledge of a materials HPHT behavior. Thus it is worth mention something about in situ measurements possibilities in DACs.

Diamonds are transparent to electromagnetic radiation with wavelengths below 0.095 nm (hard x-ray region)20as well as wavelengths in the near-UV to far-IR region32 and hence offers a wide range possibilities for in situ measurements in DACs. This makes the DAC suitable for different kinds of analytical methods involving electromagnetic radiation like x-ray diffraction and spectroscopy. In fact, many studies involving DAC has been performed using synchrotron radiation for x-ray diffraction20, 47. As mentioned, resistance measurement can also be used to detect phase transitions in DACs. Resistance measurements with temperature as variable have been done by Hörling et al on Ti 1-xAlxN12. Also, the use of resistance measurements with the temperature as variable and at high pressures (31 GPa and 35 GPa) have been performed on (Fe0.125Mg0.875)2SiO4 up to 3450 K in a laser-heated DAC35.

4.1.7 Classification of DACs

There are a lot of different DACs and some classification would be of help when choosing a DAC. On way to classify DACs is presented in a review article titled “High-pressure crystallography”31. This classification gives a good starting point for using DAC’s. However, note that this classification is focusing on the use of DAC’s combined with x-ray diffraction studies. A modified version of this classification is presented in Table 1.

(26)

Table 1 classification of DACs

Classification

criteria

Description of

classification criteria

Methods

Pressure generation The way the pressure is generated is important for the usefulness the DAC.

Spring-lever, screw, gas-membrane

Heating High temperature as well as high pressures is needed in many material studies, including this

Non-heated, external heating, laser-heating and internal resistance heating In situ measurement options Sometimes material

behavior only occurs in a particular p-T range and thereby must be examined in situ.

XRD, Resistance, Raman

Other things that may differ in different DACs include construction material, e.g., steel, Inconel or tungsten20, 31, and size. DACs as small as 22 mm been manufactured20. DACs can be bought “on-demand” but could also be custom-built according to your own specification. It seems likely though that custom-built DAC are made when needed at research institutions but could possibly be ordered from DAC companies.

Another type of cell is the gem anvil cell which is exactly the same as a DAC but replacing the diamond anvil with a much cheaper version made of zirconium or sapphire20. The main reason for choosing a zirconia or sapphire anvil is due to lower thermal conductivity and higher thermal stability20, 48. Using anvils other than diamonds, however, greatly limits the measurement possibilities due to differences in transparency for zirconium and sapphire20.

(27)

4.2 High Temperature Nanoindentation

As described above, during nanoindentation a large pressure is applied by a sharp diamond tip on a coated system to evaluate coating hardness. Typically, the tip load is chosen such that the tip only penetrates the coating at shallow depth and hence not affect or being affected by the substrate. Under such conditions, phase transformation of the coating may occur beneath the indenter22. Thus, this should be considered as a possibility also for Ti1-xAlxN. Also, a semiconducting/metallic transition has been detected using in situ resistance measurement in combination with nanoindentation22. This technique could be considered as a possibility for Ti1-xAlxN, especially at elevated temperature. In fact, nanoindentation on Ti1-xAlxN has been performed at temperatures up to 773 K25. There are other studies of nanoindentation at elevated temperatures (up to 673 K) of other material systems49-51. However, no reports of nanoindentation at temperatures in the range needed for this study, i.e., up to 1000 K have been found. The reason for this could be the problems induced by the diamond-to-graphite transitions.

4.3 Large Volume Apparatus

Today there exist several large volume apparatus for generating high pressure. One of the most common types is the cubic press, which have been used in several recent studies29, 52-55. Pressures up to around 5-6 GPa was applied29, 29, 52, 53, 55 at high temperatures (more than 1400 K)52-55. The cubic press has six anvils opposing anvils that exert a uniform load to the sample chamber, se Figure 7.

Figure 7 Principal sketch of a cubic press

In one example, metastable cubic MgxZn1-xO with x = 0.50 have been produced under HPHT conditions, using a cubic press52. Ti

2AlC has been studied at high pressure (HP) at

(28)

room temperature combined with resistance measurements29, in a similar way to the high temperature resistance measurements done by Hörling et al12, see Figures 8 and 9 for comparison. In this case, approximately 9-10 mg of the material was and placed in an h-BN capsule, and pressures up to 6 GPa was applied. The same material was also studied in the same type of cubic pressing apparatus and elevated temperature at conditions of up to 1673 K and 5 GPa55. Post characterizations were made by XRD and SEM55.

Figure 9 Resistance in Ti0,34Al0.66N as a function of temperature Figure 8 Resistance in Ti2AlC as a function of pressure

Another large volume apparatus is the piston-cylinder apparatus, which was developed before the DAC20, . This technique has recently been used for HPHT synthesis of MAX compound (Cr0.5V0.5)2GeC42. Other large volume techniques that have been used is the

(29)

toroid type apparatus which have been used in some cases56-59 and the split cell apparatus which also have been used recently60-63.

One of the greatest advantages about these large volume apparatus is that they can take much larger samples then the diamond anvil cell20. However, it is usually easier to do is situ measurements in a DAC then in a large volume apparatus20.

A different approach would be to place the whole cutting tool insert into a pressing apparatus and apply pressure and heat on the film/substrate system. This would not require any time consuming sample preparation and would be more close to the real cutting environment. The film would be subject to pressure gradients, however this would be most apparent at the edges of the cutting tool insert and since the film is thin these effect would probably not have an impact on the middle of the cutting tool insert.

4.4 Cutting Test

The most straight forward HPHT technique for use on cutting tools materials is machining. In a cutting application, the cutting edge is subject to high mechanically loads as well as high temperatures that depend on the cutting parameters as well as the work piece material selected. The applied force can be calculated and the temperature can be measured for example by using black body radiation measurement. However, the temperature and pressure cannot be varied independently since high cutting speed generates both high temperature and high pressure. From a strictly scientific point of view, this testing does not give the exact explanation how the material behaves at HPHT due to that many other processes are active, e.g., corrosion etc. Nonetheless, it should be a good tool for a fast and relatively easy screening method for a wide range of applications, e.g., for an initial evaluation of a new coating system during metal cutting. Cutting tests have been performed before on Ti1-xAlxN, e.g. by Hörling to test the tool life time on inserts with different x14.

(30)

5 Results

5.1 Nanoindentation of TiN and Ti

1-x

Al

x

N

The present results, Figure 10, of hardness data of annealed TiN and Ti0.33Al0.67N reveal good fit with the same of Hörling et al14. These hardness values are, however, about 3 GPa lower which partly could be explained by the fact that no correction for residual stress was made in this work. Another explanation is that these tests were made under somewhat different conditions. This justifies the efforts of doing a new hardness study.

0 5 10 15 20 25 30 35 40 673 773 873 973 1073 1174 1223 1273 1323 1373 Annealing temperature (K) H a rdne s s ( G P a ) TiN TiAlN

Figure 10 Hardness of TiN and Ti1-xAlxN as a function of annealing temperature

5.2 Cutting test

Initial studies of the concept using continuous turning, see Figure 11, to generate HPHT conditions for Ti1-xAlxN coated cutting inserts. XRD measurements were performed for phase characterization. The tedious work of correlating cutting data to temperature/pressure (T/P) has previously been done by Seco Tools AB in other projects. However, due to the relatively short amount of time available for this diploma work we decided to focus the efforts on the possibility for phase characterization by XRD during machining. The basic idea was to see if it is possible to detect phase changes in cutting

(31)

zone, i.e., the part of the insert in contact with the work piece material using a narrow incident X-ray beam.

Figure 11 Continuous turning operation (left) and cutting tool inserts (right)

Figure 12 shows to the left a schematic view of the cutting zone including typical wear phenomena occurring and to the right a Ti0.33Al0.67N coated insert after about 5 min of continuous turning. To verify if the concept, this insert was measured by XRD before and after turning.

A

B

C

Figure 12 Cutting tool inserts: (left) schematics of the cutting zone with A: crater wear, B: edge wear and C: notch wear and (right) a Ti0.33Al0.67N coated insert after about 5 min of continuous turning (vc

= 230 m/min, fz=0.15, ap=1).

Figure 13 shows the diffractograms of the as-deposited and machined Ti0.33Al0.67N coated insert. The small changes of the diffracted intensities at the (111) peak positions of TiN, TiAlN and AlN indicates that TiAlN has begun to decompose into [rocksalt]-TiN and [rocksalt]-AlN during turning. For comparison, Figure 14 shows the diffractograms of a Ti0.33Al0.67N coated insert before and after post heat treatment at 1173 K for 2 hours. It is clear that Ti0.33Al0.67N has decomposed more in the post heat treated samples.

(32)

35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 c-AlN (200) TiAlN (200) TiN (200) c-AlN (111) TiAlN (111) TiN (111) S S machined Intensity (a.u. ) 2θ as-deposited

Figure 13 shows the diffractograms of the as-deposited and machined Ti0.33Al0.67N coated insert.

Figure 14 shows the diffractograms of a Ti0.33Al0.67N coated insert before and after post heat treatment at 1173 K for 2 hours.

(33)

6 Discussion and Conclusion

It is a well know fact that a thin coating of Ti1-xAlxN dramatically enhance the life time of a cutting tool insert compared to an uncoated tool. Ongoing activities at the metal cutting industry is now directed towards understanding the fundamentals, i.e., the key controlling factors, of tool wear during metal cutting. With such understanding at hand, it will be possible to realize next generation tool coatings and/or coating concepts.

As outlined earlier, Ti1-xAlxN coatings in different compositions are widely used in many cutting applications world wide. It is well known that [rocksalt]-TiAlN transforms into [rocksalt]-TiN and [rocksalt]-AlN at elevated temperatures and at the same time improve its hardness properties, i.e., the system exhibits an age hardening effect. The optimum hardness is obtained at about 1100 K which also is the temperature range at the cutting edge during metal cutting. With this in mind, post heat treatment of the coating with the aim to mimic the cutting situation, at least in terms of temperature, has been a successful and important test during the development phase of Ti1-xAlxN. However, it is now time to improve these test conditions to reach the next level of understanding of coating wear during metal cutting. Biased by the fact that during a cutting operation, a normal force of about 1-3 GPa is applied to the cutting edge of the insert it has been suggested to couple the effect of temperature and pressure for the phase stability of, e.g., Ti1-xAlxN coatings. This diploma work is the first attempt in this direction dealing with high temperature high pressure (HPHT) of coated tools.

Several methods, both in situ and ex situ, can be used to analyze thin coatings such as Ti 1-xAlxN in an HPHT environment. For example, X-ray diffraction and resistance measurements could be used to study the HPHT phase behavior of Ti1-xAlxN. Resistance measurement has previously been performed both at HPHT as well as high temperature conditions on Ti1-xAlxN. The phase transformation of Ti1-xAlxN at elevated temperature was easily detected with this technique. Hence, it is reasonable to assume that this technique also will work at HPHT conditions. Moreover, the coating hardness is an important property of coatings on cutting tools. However, due to the relatively thin coatings on cutting tools ordinary hardness tests such as Vickers indentation can not be

(34)

performed since the hardness of the substrate will come into play with this method. Instead nanoindentation is widely used for hardness evaluation of thin coatings. The indentation load used in nanoindentation is in the mN range and the indenter only penetrates the top surface of the coating. Hence this technique allows for an understanding of the “pure” coating properties.

No studies concerning HPHT experiments of thin coating systems were found during this diploma work. The first HPHT technique that was found was the device called diamond anvil cell (DAC). It turned out that DACs have been used in many different application including geological sciences, medicine, materials science, etc.. Studies in the GPa pressure regime at a temperatures up to around 900K is doable with DAC. If the right percussions are taken, temperatures in the range of interest for this project could be achieved. In fact, such DAC devices are marketed and sold. Several factors have to be taken into account when choosing a DAC (see appendix B). DAC devises are available at most large synchrotron radiation labs around the world, for example at HASYLAB in Hamburg, Germany and at Argonne National Laboratory near Chicago, USA. If in situ synchrotron studies of the HPHT behaviors of Ti1-xAlxN are of interest, borrowing a DAC from one of these places would probably be possible.

There are several different methods to apply HPHT to large samples (from cubic millimeters and up). The large volume method that most frequently came across during this diploma work was the cubic press. The cubic press typically works on samples in the cubic millimeter range, however, sample volumes in the cubic micrometer range have been investigated. The cubic press could be an alternative to the diamond anvil cell. An alternative idea is to put the whole cutting insert into a large volume press and apply pressure to see what happens with how the thin film and substrate interacts at HPHT conditions. No study concerning this topic was found.

As already mentioned, nanoindentation generate a large pressure on a small area, just below the indenter. A lot of experience using nanoindentation exists within both the nanostructured materials group at Linköpings University as well as in R&D department

(35)

within Seco Tools AB. On top of that there are reports of phase transformations occurring using nanoindentation. All this makes it tempting to think about using nanoindentation on a heated sample to achieve HPHT. High temperature nanoindentation have been performed but not at temperatures high enough for our interests. It could perhaps be done at temperatures 1300 K in an inert environment but that would probably be very complicated, expensive and since it’s would also likely probably be the first time in history it was done it would also give very uncertain results.

The nanoindentation test performed during this diploma work shows results similar to those done by Hörling and others. As can be expected considering earlier results the hardness of Ti0.33Al0.67N rises with temperature up to a point where it declines fast. The nanoindentation itself does not give any information about the phases involved. However, the rapid change in hardness gives a clue that something has happened with the material, probably involving phase transformations. Phase transformations have been shown to occur at these temperatures, as mentioned in this diploma work. If HPHT treated samples could be tested using nanoindentation any phase transformation would most likely affect the measured hardness.

An actual cutting test is probably the first natural step to study the HPHT performance of coated cutting tools. The result of a cutting test tells what have happened to the materials (protective film, substrate and film/substrate) in a true cutting environment. It is, however, worth while noting that the cutting environment is much more complex then a pure HPHT condition and hence the evaluation may be difficult. Also, actual temperature and pressure acting on the cutting insert is coupled (depends on cutting parameters) in cutting test.

The cutting tests using Ti0.33Al0.67N coated inserts including initial post X-ray diffraction measurements over the cutting edge looks promising. Small change in the diffracted intensities of c-AlN (111) and c-AlN (200) between as-deposited and machined inserts indicate the presence of a phase transformation. Further studies are needed to confirm

(36)

these results, preferably by varying the cutting over a wide range yielding different pressure and temperature at the interesting area (cutting edge).

In appendix A, three methods for achieving HPHT conditions on Ti1-xAlxN are summarized and compared. Please note that the cubic press method could be used in a similar way on samples in the 10 mg range, at least according to the work performed on the MAX-phase Ti2AlC material. Based on the present knowledge, the cubic press method should be comparable to DAC on TiAlN. However, as a result of this diploma work, we believe that a DAC approach would be the better choice. The methods described in appendix A could be used separately but they will not, by themselves, result in a complete description of the HPHT behavior of Ti1-xAlxN. The best choice would probably be to combine the results of DAC and cutting tests. Another way, if applicable, would be to use a large volume press to apply HPHT to a whole cutting tool insert. The next step of this project should probably involve some kind HPHT tests and compare to machining test. The most suitable method would be a DAC, or possibly the cubic press technique. It would be interesting if a large volume press on the whole cutting tool insert could be applied. In situ measurements would possibly give more specific information about phase transformations, but it is likely that ex situ measurement, e.g., X-ray diffraction would be enough. In situ measurements could be done afterwards if the ex situ examination of the samples does not give enough information. Appendix C shows some possible ways to continue this project. As an alternative to in situ XRD, resistance measurement could be used to detect at what pressure and temperature phase transformations occur.

This diploma work is intended to be a starting point in a project that aims to give greater knowledge about the HPHT behavior and other wear mechanisms of Ti1-xAlxN. With this information is possible to optimize the composition, manufacturing process and product information for maximum cutting speed and tool life.

(37)

7 References

1. Modern metal cutting - a practical handbook. 1st ed. Sweden: Sandvik Coromant; 1996. .

2. Sundgren J-. Structure and properties of TiN coatings. Thin Solid Films 1985;128(1-2):21-44.

3. Chim YC, Ding XZ, Zeng XT, Zhang S. Oxidation resistance of TiN, CrN, TiAlN and CrAlN coatings deposited by lateral rotating cathode arc. Thin Solid Films

2009;517(17):4845-9.

4. Buhl R, Pulker HK, Moll E. TiN coatings on steel. Thin Solid Films 1981;80(1-3):265-70.

5. Johansson BO, Sundgren JE, Greene JE, Rockett A, Barnett SA. Growth and properties of single-crystal tin films deposited by reactive magnetron sputtering. Journal of Vacuum Science & Technology A-Vacuum Surfaces and Films 1985;3(2):303-7.

6. PalDey S, Deevi SC. Single layer and multilayer wear resistant coatings of (Ti,Al)N: A review. Materials Science and Engineering A 2003;342(1-2):58-79.

7. Panjan P, Navinšek B, Cvelbar A, Zalar A, Milošev I. Oxidation of TiN, ZrN, TiZrN, CrN, TiCrN and TiN/CrN multilayer hard coatings reactively sputtered at low temperature. Thin Solid Films 1996;281-282(1-2):298-301.

8. Adibi F, Petrov I, Hultman L, Wahlström U, Shimizu T, McIntyre D, Greene JE, Sundgren J-. Defect structure and phase transitions in epitaxial metastable cubic Ti0.5Al0.5N alloys grown on MgO(001) by ultra-high-vacuum magnetron sputter deposition. J Appl Phys 1991;69(9):6437-50.

9. Hörling A, Hultman L, Odén M, Sjölén J, Karlsson L. Thermal stability of arc evaporated high aluminum-content Ti1-xAlxN thin films. Journal of Vacuum Science and Technology A: Vacuum, Surfaces and Films 2002;20(5):1815-23. 10. Holleck H. Metastable coatings - prediction of composition and structure. Surface and

Coatings Technology 1988;36(1-2):151-9.

11. Hörling A. Thermal stability and age hardening of TiN-based thin films. Sweden: Department of Physics and Measurement Technology at Linköping University; 2005.

12. Hörling, A et al. Phase transformations in Ti1-xAlxN thin films. [Unpublished]

(38)

13. Kimura A, Hasegawa H, Yamada K, Suzuki T. Effects of al content on hardness, lattice parameter and microstructure of Ti1-xAlxN films. Surface and Coatings Technology 1999;120-121:438-41.

14. Horling A, Hultman L, Oden M, Sjolen J, Karlsson L. Mechanical properties and machining performance of Ti1-xAlxN-coated cutting tools. Surf Coat Technol 2005 FEB 21;191(2-3):384-92.

15. Hörling A, Hultman L, Odén M, Sjölén J, Karlsson L. Thermal stability of arc evaporated high aluminum-content Ti1-xAlxN thin films. Journal of Vacuum Science and Technology A: Vacuum, Surfaces and Films 2002;20(5):1815-23. 16. Askeland DR, Phulé PP. The science and engineering of materials. 5th ed. Canada:

Thomson; 2006. .

17. Kittel C, Kroemer K. Thermal physics. 2nd ed. Freeman; 1980. .

18. Haasen P. Physical metallurgy. 3rd ed. UK: Cambridge University Press; 1996. . 19. Nordling C, Österman J. Physics handbook for science and engineering. 8th ed.

Studentlitteratur; 2006. .

20. Paszkowicz W. High-pressure powder X-ray diffraction at the turn of the century. Nuclear Instruments and Methods in Physics Research, Section B: Beam Interactions with Materials and Atoms 2002;198(3-4):142-82.

21. Mayrhofer PH, Mitterer C, Hultman L, Clemens H. Microstructural design of hard coatings. Progress in Materials Science 2006;51(8):1032-114.

22. Schuh CA. Nanoindentation studies of materials. Materials Today 2006;9(5):32-40. 23. Flink A. Growth and characterization of ti-si-N thin films. Sweden: Department of

Physics and Measurement Technology at Linköping University; 2008.

24. Ma LW, Cairney JM, Hoffman MJ, Munroe PR. Deformation and fracture of TiN and TiAlN coatings on a steel substrate during nanoindentation. Surface and Coatings Technology 2006;200(11):3518-26.

25. Beake BD, Smith JF, Gray A, Fox-Rabinovich GS, Veldhuis SC, Endrino JL. Investigating the correlation between nano-impact fracture resistance and hardness/modulus ratio from nanoindentation at 25-500 °C and the fracture resistance and lifetime of cutting tools with Ti1-xAlxN (x = 0.5 and 0.67) PVD coatings in milling operations. Surface and Coatings Technology

(39)

26. Knutsson A, Johansson MP, Persson POÅ, Hultman L, Od́n M. Thermal

decomposition products in arc evaporated TiAlN/TiN multilayers. Appl Phys Lett 2008;93(14).

27. Bridgeman PW. The physics of high pressure. Dover ed. Canada: General Publishing Company; 1970. .

28. Li M, Gao CX, Ma YZ, Hao AM, He CY, Huang XW, Li YC, Liu J, Liu HW, Zou GT. In situ HPHT resistance measurement of (Fe0.125,mg 0.875)2SiO4 in a designed laser heated diamond anvil cell. Journal of Physics Condensed Matter 2007;19(42).

29. Qin J, Wang Z, Fang L, Wang F, Lei L, Li Y, Wang J, Kou Z, He D. In situ electrical resistance study of Ti2AlC to 6 GPa. Solid State Commun 2008;148(9-10):431-4. 30. Piermarini GJ. High pressure X-ray crystallography with the diamond cell at

NIST/NBS. Journal of Research of the National Institute of Standards and Technology 2001;106(6):889-920.

31. Katrusiak A. High-pressure crystallography. Acta Crystallographica Section A: Foundations of Crystallography 2007;64(1):135-48.

32. Horvath-Bordon E, Riedel R, Zerr A, McMillan PF, Auffermann G, Prots Y, Bronger W, Kniep R, Kroll P. High-pressure chemistry of nitride-based materials. Chem Soc Rev 2006;35(10):987-1014.

33. McMillan PF, Shebanova O, Daisenberger D, Cabrera RQ, Bailey E, Hector A, Lees V, MacHon D, Sella A, Wilson M. Metastable phase transitions and structural transformations in solid-state materials at high pressure. Phase Transitions 2007;80(10-12):1003-32.

34. Jayaraman A. Ultrahigh pressures. Rev Sci Instrum 1986;57(6):1013-31. 35. Lin J-, Santoro M, Struzhkin VV, Mao H-, Hemley RJ. In situ high

pressure-temperature raman spectroscopy technique with laser-heated diamond anvil cells. Rev Sci Instrum 2004;75(10 I):3302-6.

36. Brister K. X-ray diffraction and absorption at extreme pressures. Rev Sci Instrum 1997;68(4):1629-47.

37. Zhou Q, Ma Y, Cui Q, Cui T, Zhang J, Xie Y, Yang K, Zou G. Raman scattering system for a laser heated diamond anvil cell. Rev Sci Instrum 2004;75(7):2432-4. 38. Lin J-, Degtyareva O, Prewitt CT, Dera P, Sata N, Gregoryanz E, Mao H-, Hemley RJ.

Crystal structure of a high-pressure/high-temperature phase of alumina by in situ X-ray diffraction. Nature Materials 2004;3(6):389-93.

(40)

39. Liu L-. New high pressure phases of ZrO2 and HfO2. Journal of Physics and Chemistry of Solids 1980;41(4):331-4.

40. Diacell® HeliosDAC [Internet]UK: easyLab Technologies Ltd [cited 2009 10/24]. Available from:

http://www.easylab.co.uk/productsdetails.asp?ID=9&iflgBlueSet=1.

41. Ueno M, Onodera A, Shimomura O, Takemura K. X-ray observation of the structural phase transition of aluminum nitride under high pressure. Physical Review B 1992;45(17):10123-6.

42. Phatak NA, Saxena SK, Fei Y, Hu J. Synthesis of a new MAX compound

(Cr0.5V0.5)2GeC and its compressive behavior up to 49 GPa. J Alloys Compounds 2009;475(1-2):629-34.

43. Palosz B, Gierlotka S, Stel'makh S, Pielaszek R, Zinn P, Winzenick M, Bismayer U, Boysen H. High-pressure high-temperature in situ diffraction studies of

nanocrystalline ceramic materials at HASYLAB. J Alloys Compounds 1999;286(1-2):184-94.

44. Brister K. X-ray diffraction and absorption at extreme pressures. Rev Sci Instrum 1997;68(4):1629-47.

45. Thessieu C. Telephone call with seller at easyLab. 2009.

46. Shekar NVC, Sahu PC, Rajan KG. Laser-heated diamond-anvil cell (LHDAC) in materials science research. Journal of Materials Science & Technology 2003 NOV;19(6):518-25.

47. Palosz B, Gierlotka S, Stel'makh S, Pielaszek R, Zinn P, Winzenick M, Bismayer U, Boysen H. High-pressure high-temperature in situ diffraction studies of

nanocrystalline ceramic materials at HASYLAB. J Alloys Compounds 1999;286(1-2):184-94.

48. Xu J-, Huang E. Graphite-diamond transition in gem anvil cells. Rev Sci Instrum 1994;65(1):204-7.

49. Beake BD, Smith JF. High-temperature nanoindentation testing of fused silica and other materials. Philosophical Magazine A: Physics of Condensed Matter, Structure, Defects and Mechanical Properties 2002;82(10 SPEC.):2179-86. 50. Richter A, Chen C-, Smith R, McGee E, Thomson RC, Kenny SD. Hot stage

nanoindentation in multi-component al-ni-si alloys: Experiment and simulation. Materials Science and Engineering A 2008;494(1-2):367-79.

(41)

51. Murugavel R, Ramesh T, Prema P. Variable thermal loading analysis of polycrystalline copper. Materials and Design 2009;30(2):221-7.

52. Qin JM, Yao B, Jia XP, Shan CX, Zhang JY, Ma HA, Shen DZ. Characterizations of single-phased cubic Mg0.5Zn0.5O prepared at high pressure and high temperature. J Phys D 2008;41(15).

53. Yin L-, Li M-, Cui J-, Zou Z-, Liu P, Hao Z-. Diamond formation using Fe3C as a carbon source at high temperature and high pressure. J Cryst Growth

2002;234(1):1-4.

54. Lv R, Liu J, Li Y, Li S, Kou Z, He D. High pressure sintering of cubic boron nitride compacts with al and AlN. Diamond and Related Materials 2008;17(12):2062-6. 55. Qin J, He D, Chen C, Wang J, Hu J, Yang B. Phase segregation of

titanium-aluminium carbide (Ti2AlC) at high pressure and high temperature. J Alloys Compounds 2008;462(1-2).

56. Gladkaya IS, Kremkova GN, Bendeliani NA. The phase transition of hBN-cBN in the B-N-H-O system. Diamond and Related Materials 1996;5(12):1440-3.

57. Sidorov VA. Differential thermal analysis of magnetic transitions at high pressure: Néel temperature of NiO up to 8 GPa. Appl Phys Lett 1998;72(17):2174-5. 58. Fabrication and physical properties of SiC-GaAs nano-composites. ; 2006. .

59. Bull CL, Guthrie M, Klotz S, Philippe J, Strässle T, Nelmes RJ, Loveday JS, Hamel G. Toroidal anvils for single-crystal neutron studies. High Pressure Research 2005;25(4):229-31.

60. Abbaschian R, Zhu H, Clarke C. High pressure-high temperature growth of diamond crystals using split sphere apparatus. Diamond and Related Materials 2005;14(11-12):1916-9.

61. Horvath-Bordon E, Riedel R, Zerr A, McMillan PF, Auffermann G, Prots Y, Bronger W, Kniep R, Kroll P. High-pressure chemistry of nitride-based materials. Chem Soc Rev 2006;35(10):987-1014.

62. McMillan PF, Shebanova O, Daisenberger D, Cabrera RQ, Bailey E, Hector A, Lees V, MacHon D, Sella A, Wilson M. Metastable phase transitions and structural transformations in solid-state materials at high pressure. Phase Transitions 2007;80(10-12):1003-32.

63. Song O. Characterization of synthesized and treated gem diamonds. Metals and Materials International 2007;13(5):427-31.

(42)

8 Picture sources

[Figure 1]:

Holleck H. Metastable coatings - prediction of composition and structure. Surface and Coatings Technology 1988;36(1-2):151-9.

[Figure 2]:

Horling A, Hultman L, Oden M, Sjolen J, Karlsson L. Mechanical properties and

machining performance of Ti1-xAlxN-coated cutting tools. Surf Coat Technol 2005 FEB 21;191(2-3):384-92.

[Figure 3]:

Schuh CA. Nanoindentation studies of materials. Materials Today 2006;9(5):32-40.

[Figure 4]:

Hörling, A et al. Phase transformations in Ti1-xAlxN thin films. [Unpublished]

[Figure 5]:

Jayaraman A. Ultrahigh pressures. Rev Sci Instrum 1986;57(6):1013-31.

[Figure 6]:

Jayaraman A. Ultrahigh pressures. Rev Sci Instrum 1986;57(6):1013-31.

[Figure 7]:

Press Technology [Internet]: Smith Technologies [downloaded 6th sept2009]. Available from: http://www.megadiamond.com/technologyPress.aspx

[Figure 8]:

Qin J, Wang Z, Fang L, Wang F, Lei L, Li Y, Wang J, Kou Z, He D. In situ electrical resistance study of Ti2AlC to 6 GPa. Solid State Commun 2008;148(9-10):431-

[Figure 9]:

Hörling, A et al. Phase transformations in Ti1-xAlxN thin films. [Unpublished]

[Figure 10-14]:

(43)

9 Appendix

9.1 Appendix A – HPHT techniques suitable for Ti

1-x

Al

x

N

Table 2 HPHT techniques

Technique Description Measured Advantages Disadvantages

DAC Place a small sample of

the film into the DAC

The HPHT behavior of the film alone

The films properties could be measured alone Widely used in HP research Needs to be purchased/borrowed Working with small samples can be problematic

Large volume press Place the whole cutting tool insert into a large volume press

The HPHT behavior of the film/substrate together in a pure environment

Easy, not sample preparation needed

It could be hard to figure out what the specific properties of the film and what is of the film/substrate system Pressure gradients must be taken into account

No/little published about this Cutting test Use the cutting insert

in cutting application with different speeds and loads

The HPHT behavior of the film/substrate together in the real cutting environment

This is the environment the cutting tool is meant to be used in Is done and can be done at SECO tools factory

Other factors effecting the result as the air and the work-piece metal The pressure and

temperature cannot be varied freely

The DAC would give most accurate answers about HPHT behavior of Ti1-xAlxN but the results would quite far from the real cutting environment. On the other hand the cutting test would show the HPHT behavior in the cutting environment but would properly be not representative to the HPHT behavior alone. To apply HPHT in a large volume pressing apparatus would give answers about the HPHT behavior of Ti1-xAlxN in interaction with the substrate. Choosing the right pressure transmitting media so that it doesn’t interact with the film would be suitable for obtaining the best results of the HPHT behavior of Ti1-xAlxN.

(44)

9.2 Appendix B – DAC for Using on Ti

1-x

Al

x

N Samples

Table 3 DAC properties

Property Suitable

Comment

Pressure generation Gas-membrane is first choice The gas-membrane could be

used for operating the pressure remotely

Pressure transmitting media Cryogenically loaded N2

or

Cryogenically loaded Ar2

N2 would stabilize the nitride

Ar2 would be inert to the

sample

Pressure calibration SrB4O7:Sm2+ florescence Works in the same way as

the widely used ruby florescence method but is rarely affected by

temperature

Temperature calibration Grey body radiation Common method for

temperature

Heating equipment Gasket heating (internal

heating and maybe external heating could work as well)

Works at higher temperatures then external heating but is more easy to handle the laser heating which can achieve unnecessary high

temperature

Sample preparation PVD coating to steel plate

and removal of steel with hydrochloric acid

This is already done for DSC sample preparation

(the sample however must be pressed to a pellet)

Gasket Inconel gasket plates Since steel is quite weak at

1200 K Inconel would be more suitable as gasket material

(45)

The other option is to borrow a DAC, preferably using it at a place where materials tests are commonly done, for example at a synchrotron source. The advantage about using a already existing DAC is that it would probably be less time consuming if only a few tests are to be preformed, qualified help is probably easy to get and that in situ measurement would be more easy to perform.

However since ex situ measurement would probably answer most of the questions purchasing a DAC could be used, when purchased, whenever wanted and could be used of different materials in the future. In situ XRD or other measurements could be done in a DAC at place however it would probably take some time to do the arrangements.

9.2.1 Purchasing DACs

Three on-demand DAC manufactures/suppliers have been found. None of them are marketing laser-heated DAC on their web pages, even tough laser heating could be installed to some of the DACs. The UK based company easyLab has a gas-membrane driven DAC, Diacell® HeliosDAC, together with a resistivity heating device that can operate up to 1273 K. Together with necessary equipment this would cost about 200000SEK/20000€. However for drilling holes with 5 µm precision a micro driller would be suitable and that would cost additional 200000SEK/20000€ if purchased from easyLab. Ex situ measurements are less precise and thereby give uncertainty which makes the precision of the micro driller redundant. Thus a cheaper and less precise drill will probably be sufficient.

Table 4 DAC suppliers

Company Most

suitable

product

Specifications

easyLab Technologies Ltd

UK based company

http://www.easylab.co.uk/

Diacell® HeliosDAC, a gas-membrane driven DAC

Max pressure: 100 GPa Max temp.: 1273 K

Heating: Single internal gasket heater

Scimed GmbH

German based company

http://www.scimed.de/DAC_eng.htm

Non found, very confusing homepage N/A

D’anvils Ltd

Israeli based company

http://danvils.com/dac.html

(1) Screw-driven Opposing-plate DAC or (2) Screw-driven Piston-Cylinder DAC

Max pressure: (1) 70 GPa and (2) 200 GPa

Max temp.: 1200 K Heating: Not included

(46)

9.3 Appendix C – The Continuing of This Project

The next step of this project should be to do more compressive cutting tests similar to the one done during this diploma work. After the cutting tests have been performed there are several ways to continue this work and all of them will probably give the wanted knowledge. In situ measurements at HPHT conditions will, if they are performed correctly, give the most accurate answer about what phase transformations is active. However, in situ measurements are usually more difficult to perform while metastable to stable transformations are easily detected ex situ. Thus it would be recommended to start with ex situ measurements and only perform in situ measurements if the results obtained are insufficient. The film/substrate system is of interest for cutting applications, however, if the HPHT are applied to the film/substrate system it is possible that the Ti1-xAlxN behaves in another way then it would be if the film was studied alone. HPHT studies of the whole cutting insert could be done if the separate studies of Ti1-xAlxN and the cutting tests does not give the all the knowledge wanted.

The recommended path to increased knowledge about Ti1-xAlxN would be as argued above and is surrounded by an ellipse in the flowchart, Table 5, below. However if for some practical reason it would be appropriate to take another path two additional paths are shown in the flowchart. This is for instance if it would be possible to use a large volume press much sooner then obtaining (or borrow) a DAC or cubic press. Also this could be if it would turn out to be very easy to either install a DAC inside a XRD apparatus or gaining beam-time. Something to consider is to use in situ resistance measurements depending if it is easily done in the device used. In situ resistance measurements would give more specific information about where phase transformations occur but not as accurate as for example XRD.

References

Related documents

To test a heat detector the response time of the detector is measured, which means the time interval between the start of a temperature increase from the application temperature

For constants that are read by all or many of the threads, like a filter kernel in image or volume convolution, the constant memory should be used since it is cached and thereby

This has been shown through characterizations of worn coatings used in continuous metal cutting where it is seen that the spinodally decomposed domains are

Estos ejemplos que hemos usado para mostrar cada inteligencia, son los ejemplos de Tornberg (2009, pág. Además, para facilitar para los alumnos, hemos escrito las preguntas

Hence, in the case of continuous time series models and uniformly sampled data, one is actually interpolating the covariance function instead of the output as in the case

Det vi kan se från tabellen är att det är många som uppger att de aldrig skriver på något annat språk än svenska hemma men det kan även vara så att vissa av dessa elever

In Figure 5.14, a distinct difference is visible for the operating ANC system and the convergence of the system is seen during the time interval 2 − 2.3 seconds. A plot of the

To test the range of activities that can be detected using the plate reader assay, a test was performed in which the rate of pNPA hydrolysis was measured for a range of HCAII pwt