Appunti completi surface technology

INDEX

INDEX 1

1 Vacuum technology 6

1.1 Molecular flow 7

1.2 Throughput 7

1.3 Conductance 7

1.4 Evacuation speed 7

1.5 Evacuation time 8

1.6 Outgassing 8

1.7 Pumping systems 8

1.7.1 Rotary pumps 8

1.7.2 Diffusion pumps 9

1.7.3 Turbomolecular pumps 9

1.7.4 Cryopumps 10

1.7.5 Diaphragm pump 10

1.7.6 Scroll pump 10

1.8 Vacuum gauges 11

1.8.1 Capacitance vacuum gauge 11

1.8.2 Pirani vacuum gauge 11

1.8.3 Ionization vacuum gauges 11

Exercises about vacuum technology 12

Exercise 1 12

Exercise 2 12

Exercise 3 12

2 Microstructure and sintering 13

2.1 Structure of coatings and thin films 13

2.1.1 Early stages of deposition 13

2.1.2 Film growth 13

2.1.3 Structure zone models 13

2.1.4 Effect of impurities 14

2.1.5 Effect of energetic particles 14

2.2 Sintering 15

2.2.1 Sintering processes 15

2.2.2 Densification 15

2.2.3 Driving energy 15

2.2.4 Transport mechanisms 16

2.2.5 Neck growth and densification 17

2.2.6 Shrinkages of particle arrays 17

2.2.7 Factors affecting shrinkage 17

1 of 76

2.2.8 Hot pressing 18

2.2.8.1 Hot pressing applications 18

2.2.8.2 Mechanisms of hot pressing 18

Exercises about coating microstructure and sintering 20

Exercise 4 20

Exercise 5 20

Exercise 6 21

3 PVD processes 21

3.1 Vacuum Evaporation 21

3.1.1 Evaporation of compounds 22

3.1.2 Evaporation of alloys 22

3.1.3 Deposition geometry 22

3.1.4 Thickness uniformity 23

3.1.5 Conformal coverage 23

3.1.6 Film purity 23

3.1.7 Evaporation sources 24

3.1.7.1 Resistively-heated sources 24

3.1.7.2 Electron-beam sources 25

3.1.8 Vacuum evaporation configurations 25

3.1.9 Process control 25

3.1.10 Advantages of vacuum evaporation 26

3.1.11 Disadvantages of vacuum evaporation 26

3.1.12 Applications of vacuum evaporation 26

3.1.12.3 Future trends 28

Exercises about vacuum evaporation 28

Exercise 7 28

Exercise 8 28

Exercise 9 29

Exercise 10 29

3.2 Glow discharges and plasmas 30

3.2.1 Plasma discharges 30

3.2.2 Collision processes 31

3.2.3 Ion-surface interactions 32

3.2.4 Sputter yield 33

3.2.5 Sputter deposition 33

3.2.6 Sputtering of alloys 34

3.2.7 Thermal history of the substrate 34

3.2.8 Sputtering processes 34

3.2.8.1 DC Sputtering 34

2 of 76

3.2.8.2 RF Sputtering 35

3.2.8.3 Magnetron Sputtering 36

3.2.8.4 Reactive Sputtering 36

3.2.9 Transport of sputtered species 37

3.2.9.1 Process control 37

3.2.9.2 Contamination 38

3.2.10 Advantages of sputter deposition 38

3.2.11 Disadvantages of sputter deposition 38

3.2.12 Applications of sputter deposition 38

3.3 Arc Plasma Deposition 39

3.3.1 Cathodic arc deposition 39

3.3.2 Particle ejection 40

3.3.3 Filtered cathodic arc 40

3.3.4 Arc source configurations 40

3.3.5 Reactive arc deposition 41

3.3.6 Arc deposition system 41

3.3.7 Advantages of arc deposition 41

3.3.8 Applications of arc deposition 41

3.4 Ion Beam Assisted Deposition 41

3.5 High Power Pulsed Magnetron Sputtering 42

3.6 Plasma Immersion Ion Implantation 43

3.6.1 PIII and deposition 43

Exercises about sputtering 43

Exercise 11 43

Exercise 12 44

Exercise 13 45

Exercise 14 45

4 Advanced coatings for tribological applications 46

4.1 Coating failure 47

4.2 Coating composition 47

4.3 Tribological coatings 47

4.4 Contact stresses 48

4.5 Load-carrying capacity 49

4.6 Coating structures 49

4.6.1 Graded coatings 49

4.6.2 Duplex coatings 50

4.6.3 Multilayer coatings 50

4.6.4 Superlattice coatings 51

3 of 76

4.6.5 Nanocomposite coatings 51

4.6.6 Diamond-Like Carbon 52

4.6.6.1 Carbon hybridization 52

4.6.6.2 Structure of DLC (a-C:H) 52

4.6.6.3 Deposition processes 52

4.6.6.4 Thermal stability of DLC 53

4.7 Hardness 53

4.7.1 Indentation 54

5 Stress and fatigue 55

5.1 Residual stresses 55

5.1.1 Stress measurement 56

5.1.2 Thermal stress 56

5.1.3 Stress in PVD films 57

5.1.4 Stress in CVD coatings 57

5.1.5 Scratch test 57

5.2 Fatigue resistance of coated components 58

5.2.1 Fatigue resistance - NiP 58

5.2.2 Fatigue resistance - Cr 58

5.2.3 Fatigue resistance – Zn 58

Exercises about residual stresses 59

Recall of elastic behavior of material 59

Exercise 15 59

Exercise 16 60

6 Atmospheric plasma processes 61

6.1 High-pressure plasmas 61

6.2 Instabilities in atmospheric plasmas 61

6.3 Corona discharge 62

6.4 Dielectric Barrier Discharge 62

6.5 Atmospheric pressure plasma jets 62

6.6 Surface modification treatments 62

6.7 Deposition of coatings 63

6.8 Collisions with metastable species 63

6.9 Structure of coatings 63

6.10 Deposition rate 64

6.11 Future applications of AP plasma jets 64

6.12 Concluding remarks 64

7 CVD processes for surface treatment 64

7.1 Thermochemical treatments 65

4 of 76

7.2 Thermochemical processes 65

7.3 Carburizing 65

7.3.1 Gas carburizing 66

7.3.2 Carburizing reactions 67

7.3.3 Carbon potential 67

7.3.4 Control of carburizing 67

7.3.5 Two-stage carburizing 68

7.3.6 Carbonitriding 69

7.3.7 Residual stress 69

7.3.8 Plasma carburizing 69

Exercise about carburizing 69

Exercise 17 69

7.4 Nitriding 71

7.4.1 Fe-N phase diagram 71

7.4.2 Nitride formation 71

7.4.3 Nitrided iron 71

7.4.4 Nitrided steels 72

7.4.5 Nitrided stainless steels 72

7.4.6 Nitriding process 72

7.4.7 Nitriding potential 72

7.4.8 Process control 72

7.4.9 Oxidizing treatment 73

7.4.10 Properties of treated components 73

7.4.11 Plasma nitriding 74

7.4.12 Plasma nitriding techniques 74

7.4.12.1 Cold-wall technique 74

7.4.12.2 Active Screen Plasma Nitriding 74

7.4.12.3 Hot-wall technique 75

7.4.13 Advantages of Plasma Nitriding 75

7.4.14 Control of Plasma Nitriding 76

7.4.15 Plasma Nitriding of Al alloys 76

7.4.16 Al alloys – nitride layer 76

7.4.17 Al alloys – metal oxidation 76

7.4.18 Age-hardening alloys 76

5 of 76

11-03-20

Esame anche in italiano; scegliere un topic e parlare per circa 15 min, poi ci saranno altre

domande sul resto del corso.

1 Vacuum technology

PVD works in vacuum. We shall start understanding what is vacuum. Vacuum means that the

pressure of the gas is lower then the atmospheric pressure. This is needed in order to:

1. Avoid collision of atoms. The most general setup is made up of a source of atoms or

molecules and a substrate where these molecules are collected in order to allow the coating

to grow. So, if atoms can move freely from the source to the substrate we can maximize the

deposition rate and then we have a more efficient process.

2. Collect particles with high energy, because through an electric field we can accelerate charged

particles producing a subset of particles with quite high average energy. They can produce

specific effects that are not obtained through conventional process. In order to accelerate

PVD processes

these particles up to high energy we need:

Sufficiently long distance between subsequent collision;

• A low-density gas.

• PVD processes operate under vacuum, that is the total pressure

So, the distance between collision is expressed by the mean free path, that is the average of the

distances made by the particle without colliding with other particles. In particular, being λ the

the gas phase is lower than the atmospheric pressure.

mean free path, we have that:

Pressure reduction is required to increase the mean free path ( )

The expression for λ is derived from kinetic theory of gases. It means that the mean free path is

larger if we reduce the pressure of the gas, that is if we reduce the density of gas.

that is the average distance travelled by the gas molecules

In the expression for λ, σ is a cross-section. What does it mean? To every particle X in the system,

we associate a sphere whose surface is defined cross-section, such that we say that a particle A

between collisions. The kinetic theory of gases shows that

has an (elastic) interaction with particle X (with a momentum transfer causing a scattering effect)

whenever the center of the particle A touches the cross-section of X.

Moreover, we associate a specific cross-section on the basis of the interaction

we are considering. This means that, changing the set of interactions in which

we are interested on, the cross-section value, in general, changes as well. It is

worth to say that it is possible that the cross-section value can sometimes be

larger than the physical dimension of the particle (for example in electrostatic

interaction). At pressure P < 10 mbar the mean free path is so large that, typically, particles

k is the Boltzmann constant, T is the absolute temperature, is

-3

interact only with the walls of the vacuum chamber.

the cross section for the collision and P is the gas pressure.

We can distinguish two different regimes of flow:

The cross section represents an area associated with each

Molecular flow. In molecular flow, the mean distance between molecular collision is large

• compared to the dimensions of the system, and it is achievable with low pressure and low gas

particle; the interaction (e.g. a collision) occurs when the centre

density.

Viscous flow. In viscous flow the mean free path is much smaller than the chamber dimensions,

• of another particle hits this area.

and is obtained with high pressure. The viscous flow can be either laminar or turbulent,

depending on the gas velocity. -3

At pressure below 10 mbar, is so large that molecules

Flow regimes are distinguished according to the value of the Knudsen number:

typically collide only with the walls of the vacuum chamber.

We distinguish in the following way:

Molecular flow for Kn > 1

• Intermediate flow for 0,01 < Kn < 1

• Viscous flow for Kn < 0,01

•

At pressure below 10 mbar we have λ > 6 cm. Moreover, in the whole process, flow could be

-3

viscous somewhere (like in the pump) and molecular in other parts (like in the vacuum chamber).

The most important variable in the design of a vacuum machinery is the pressure which the

pumping system must be able to maintain in the working chamber. It is possible to distinguish

between several types of vacuum:

low vacuum, for P = 10 -1 mbar, applied in drying and food processing;

3

• medium vacuum, for P = 1-10 mbar, applied in steel degassing, vacuum distillation;

-3

• high vacuum, for P = 10 - 10 mbar, applied in PVD coatings;

-3 -6

• ultrahigh vacuum, for P < 10 mbar, applied in surface analysis and MBE (molecular beam

-6

• epitaxy). 6 of 76

randomly and collisions between molecules are rare

compared to collisions with the walls.

In order to design a vacuum system, a relationship between

the gas flow rate and the pressure difference should be

known for any portion of the system.

1.1 Molecular flow A typical example is given by

Under the hypothesis of molecular flow regime, the collisions between A

an opening of area A that

molecules are rare compared to the ones with walls of the chamber. In order to

separates two chambers

design a vacuum system, a relationship between the gas flow rate and the P P

1 2

maintained at low pressures,

pressure difference should be known for any portion of the system.

P and P (P > P ).

1 2 1 2

Let us consider a separator with opening of area A.

The gas flux φ is the number of moles that strike an element of surface per unit time and area and

is given by:

where n is the gas concentration (moles per unit volume) and ṽ is the average molecular speed .

The factor ¼ is due to the fact that:

Particles can move to the left or the right but only the particles that cross the area A are

• considered, so we consider ½;

Particles can move also in all directions but only the component perpendicular to the area A is

• considered so we have ; if we make the average value of all the cosines function we

obtain the other factor ½.

The net flux through the opening is:

From the equation of state of an ideal gas, n = N/V = P/RT and

The net rate at which gas molecules crosses the opening is

that is the amount of molecule that go through the area in unit time.

1.2 Throughput

The throughput Q is the quantity of gas that passes a plane in a unit time. It is expressed as the

volume of gas per unit time multiplied by its pressure:

Moreover, if the temperature is constant, throughput can be related to molar or mass flow:

1.3 Conductance

The conductance C of a duct is the ratio of the throughput to the pressure drop at constant

temperature:

Now, taking into account the Maxwell-Boltzmann distribution of velocities: being M the

molar mass of the gas molecules, then:

These formulas are used to detect leakage effect and He gas is used due to its low mass

The conductance of a pipe can be written taking into account the transmission probability α, that

is the probability that a particle that enters the pipe will exit it through the opposite end. Thus,

simply correcting the expression for C:

For example, the following expression is true for a circular tube: Evacuation speed (1)

being D and L, respectively, the diameter and the length of the tube. It is worth to notice that

conductance of a tube becomes larger as the diameter increases and the length decreases.

The speed S of any type of vacuum pump can be defined as

p S = Q/P

Transmission probabilities can be calculated by the Monte Carlo method to track molecular

p p

where P is the pressure at the inlet to the pump and Q is the

p

motions through various configurations. throughput at that point.

The speed S has the same dimensions of conductance, volume

p 3 -1

per unit time; the SI units are m s .

1.4 Evacuation speed C

With reference to a typical

Between the vacuum chamber and the pump, the throughput is constant,

vacuum system, we can write S, P

because of conservation of mass (there are no temperature gradients or

P = Q/S at the inlet to the duct S , P

p p

Vacuum

other things). The speed of any type of vacuum pump can be defined as:

P = Q/S at the inlet to the pump chamber

p p

It is possible to obtain that: P-P = Q/C over the duct length Valve

p

then 1/S – 1/S = 1/C Pump

p

or S = S [1/(1+S /C)]

p p

We notice that the effective pumping speed cannot exceed the smaller quantity between the

pump speed and the duct conductance. This is equivalent to say that, between the pump and the

vacuum chamber there exist a resistance given by the reciprocal of the duct conductance. A

general rule is to increase the diameter of the duct and the length of connections as much as

possible. 7 of 76

13-03-20

1.5 Evacuation time

The change in pressure in the vacuum chamber is related to the pumping speed of the system by

the equation

where P is the pressure measured at a specified point in the system that change in time, S is the

speed at that point, V is the system volume and Q is the throughput associated with gas leakage

t

and interior surface outgassing.

If duct conductance C, pump speed S , and throughput Q are assumed to be independent of

p t

pressure and time, the equation can be easily integrated:

P is the initial pressure and P = Q /S is the ultimate pressure (minimum pressure attainable in the

i u t

chamber).

1.6 Outgassing

The total throughput Q includes many gas sources, which mainly consist of:

t

1. Desorption from surfaces, that is temporary;

2. Diffusion from materials: polymers can contain in particular dissolved gas;

when we reduce the pressure in the chamber the gas go out the material, but it

is temporary.

3. Permeation through materials: we have a continuous gas flow from the exterior

to the inside of the chamber, so it is permanent.

4. Gas leakage: it is due to micro-cavities for example, where gas passes through it.

The increase in pressure caused by these effect is so much (in particular with high and ultra high

vacuum) that increases the theoretical time calculated before.

Initially, the volume gas is removed and the pressure decreases exponentially (e ) in a short

-St/V

time. Subsequently, the pressure reduction is controlled by surface desorption (mainly water

vapour) and later by diffusion (e.g. absorbed gases, water and solvents from polymers, hydrogen

from metals). Finally the ultimate pressure is attained, with constant contribution from permeation

and gas leakage.

1.7 Pumping systems

A system, initially at atmospheric pressure, is first evacuated by a medium-vacuum pump

(forepump or preliminary), that is able to discharge gas at atmospheric pressure, until a pressure is

reached where a high-vacuum pump becomes effective and can be used for final evacuation to

the desired pressure.

Many types of vacuum pumps exist, only the most common ones will be examined.

1.7.1 Rotary pumps

Most mechanical vacuum pumps are of rotary type and are

sealed with oil. A small quantity of gas from the system is

isolated, compressed and discharged to the atmosphere with Rotary pumps (2)

each rotation of the piston or vane. The pumping speed is roughly constant at high pressures and

The spring needs to push the blade of the rotor in contact with

decreases as the ultimate pressure (P ) is approached.

the stator in order to give a sealing effect (useful to avoid oil at 0

At P , the back leakage equates the forward flow and the

the bottom of our stator that can contaminate both the 0

pumping speed drops to zero.

chamber if it has a high vapor pressure and the exhaust gas

that forms a sort of emulsion of micro droplets inside gas).

In the two-stage pump, lower pressures are reached by connecting

the exhaust of the first stage to the intake of the second one. In

this case we have the product of the compression in the first stage

and that in the second one. S = S (1-P /P )

p 0 0 p

In order to characterize the performance of this pump we use this

graph. The pumping speed is roughly constant at high pressures

and decreases as the ultimate pressure (P ) is approached. At P ,