Furnace Atmoshpere

8/10/2019 Furnace Atmoshpere

1/48

Furnace Atmospheres

No. 3Gas Nitriding and Nitrocarburising.


2/48

Preface

2

This booklet is part of a series on heat treatment, brazing and solderingprocess application technology and expertise available from Linde Gas.The booklet focuses on the use of furnace atmospheres; however, a briefintroduction to each process is also provided. In addition to this work onnitriding and nitrocarburising, the series includes:

3 Furnace Atmospheres No. 1 Gas Carburising and Carbonitriding

3 Furnace Atmospheres No. 2 Neutral Hadening and Annealing

3 Furnace Atmospheres No. 3 Nitriding and Nitrocarburising

3 Furnace Atmospheres No. 4 Brazing of Metals

3 Furnace Atmospheres No. 5 Sub-zero treatment of steels

3 Furnace Atmospheres No. 6 Low pressure carburising and high

pressure gas quenching

3 Furnace Atmospheres No. 7 Tube Annealing

The author gratefully acknowledges the support of the following com-panies and persons in providing information and images: Ipsen GmbH,

ABB AB, Nitrex Metal Inc, Sulzer Metaplas GmbH, GE Sensing & Inspec-tion Technologies AB, Bodycote Vrmebehandling AB, Professor BrigitteHaase, Bremerhaven University and Professor Marcel Somers, TechnicalUniversity of Denmark.


3/48

3

Table of contents

I. Introduction 4

II. Process Selection 7

III. The Process Steps in Nitriding and Nitrocarburising 9

A. Prior Heat Treatment Condition 9B. Cleaning 9

C. Preheating and Pre-Oxidation 10

D. Nitriding 10

E. Nitrocarburising 11

F. Vacuum, High Pressure and Plasma Nitriding and

Nitrocarburising 12

G. Post Oxidation 13

H. Austenitic Nitrocarburising 13

I. Combined Processes 14

J. Cooling/Quenching and Post Treatment 14

IV. Properties of Nitrided and Nitrocarburised Steels 16

A. Hardness and Wear Resistance 16

B. Static and Fatigue Strength 18

C. Selection of Nitriding/Nitrocarburising Case Depth 18

D. Corrosion Resistance and Surface Appearance 18

E. Dimensional Changes 19

F. Properties of Nitrided/Nitrocarburised Stainless Steel 20

V. Atmosphere Surface Interaction 22

A. The Atmosphere Nitriding Potential 23

B. The Atmosphere Carbon Potential 24

C. Atmosphere Analysis 24

1. Ammonia and Hydrogen Analysis 24

2. Oxygen Probe Analysis 25 3. FTIR Gas Analysis of Nitrocarburising Atmospheres 26

D. Layer Growth Determination 26

E. Guidelines for Regulating the Atmosphere 27

1. Nitriding 27

2. Nitrocarburising 27

3. Post Oxidation Control 29

VI. Compound Layer and Diffusion Zone Formation 30

A. Nitriding 30

B. Nitrocarburising 31

1. Furnace Interior Influence 33

2. Influence of Amount of Active Gas 34

3. Steel Alloy Content Effect 35

C. Pore Formation 35

VII. The NITROFLEXSolution 36

A. Gas Supply 36

1. Nitriding 39


B. NITROFLEXProcess Recipes 40

C. Case Studies 40


2. Austenitic Nitrocarburising 41

3. Nitriding 42

VIII. Safety 43

A. Toxicity and Asphyxiation 43

B. Flammability 43

IX. Concluding Remarks 45

X. References 46


4/48

4 Introduction

Nitriding and nitrocarburising of steel parts give unique improve-ments in wear and corrosion resistance that cannot be obtained by

carburising or carbonitriding. Increased fatigue strength is alsoobtained. These improvements can be understood when examiningthe surface microstructure and hardness after treatment illustratedin Figure 1. The outermost compound layer of a nitrided or nitrocar-burised steel is 230 m thick and consists of the iron/nitrogen/car-bon -phase with variable compositions of carbon and nitrogen andthe -phase with the virtually stoichiometric chemical formula Fe4N.This layer is called the compound layer, sometimes also referredto as the white layer or the ceramic layer. Corrosion resistance andtribological properties (friction and wear) are mainly determined bythe compound layers properties, which differ notably from those ofthe base material.

Under the compound layer there is a diffusion zone, which goesdeeper into the steel, typically 0.1 0.5 mm. Load bearing capacity,and static and fatigue strength are largely determined by the hard-ness and depth of the diffusion zone.

Because of the benefits of a shortened production cycle, limited dis-tortion, elimination of post grinding and attractive aesthetic surfaceappearance, a great number of parts subjected to wear and fatigueare nitrocarburised today instead of being carburised or carboni-trided, as was previously the case. Examples of applications for thenitriding and nitrocarburising of low alloy steels are gears, crank-

shafts, camshafts and parts in sliding contact such as cylinders andpistons where good tribological properties are needed. An elegant

demonstration of the benefits of nitrocarburising was given in apaper by Dawes and Tranter [1] in which the application of a screenwiper linkage was described. The strength increase associated withnitrocarburising resulted in a weight decrease of 62%. The verygood tribological properties made it possible to eliminate the bronzebearings previously used. The corrosion resistance was very goodand corresponded to a neutral salt corrosion resistance of 250 hours.Finally, the parts were made aesthetically attractive with a black sur-face produced by a post oxidation treatment. High alloy steels usedfor forging and extrusion tools are other examples that benefit frombeing nitrided or nitrocarburised.

Atmosphere composition and control are of crucial importance forthe nitriding/nitrocarburising result with respect to final propertiessuch as wear and fatigue resistance. This is exemplified in Figure 46,showing four very different compound layer morphologies, merely asthe result of the use of different gas compositions during the process(See section V.E.2, page 27, for explanation).

Nitriding and nitrocarburising are thermochemical processes, asare carburising and carbonitriding, see Table 1. In these processes,nitrogen and/or carbon are transferred from the process medium,normally gas, to the surface of the treated steel. An elevated temper-ature is required in order to ensure fast transfer of carbon/nitrogen

I. Introduction

Figure 1. Micrographs showing compound layer and diffusion zone on nitro-

carburised: a) Low carbon steel with ferritic-pearlitic core microstructure.b) 5 % Cr tool steel with tempered martensite core microstructure. The diffusion

zone is the dark etched region below the compound zone.

Hardness

Diffusion

Diffusion

Hardness

- and/or -phase compound layer


5/48

5Introduction

from the gas to the steel surface and also to allow carbon/nitrogento diffuse into the steel at an appreciable rate.

An important feature of nitriding and nitrocarburising is that they arelow temperature methods whereas carburising and carbonitridingare high temperature methods. Here low temperature refers to atemperature below that where phase transformation to austenitestarts (A1), and high temperature is above said temperature. A valu-able consequence is notably reduced distortion of treated parts.This can often save time and costs by eliminating the need for postgrinding to meet dimensional tolerance requirements. The productioncycle of a part therefore becomes faster and cheaper. A limitation

caused by the lower temperature is that the diffusion rate for nitro-gen and carbon is modest, which sets limits on the case depths thatcan be obtained.

Carburising and carbonitriding give a surface hardness in the rangeof 750-850 HV that is largely independent of the steel type, whereasnitriding and nitrocarburising give a wide possible range of surfacehardness determined by the steel selection.

Austenitic nitrocarburising is a process that has characteristics inbetween the high temperature methods of carburising and carboni-triding and the low temperature processes of nitrocarburising and

nitriding.

A consequence of the low process temperature, the short processtime and the elimination of productions steps is low energy con-sumption. Table 2 shows an example in which the energy saving wasabout 50 % when the process route was changed to nitrocarburising.

Table 1. Characteristics of thermochemical processes involving nitrogen and/or carbon.

Process Temperature Typical Element Case depth Sur face DistortionC (F) process time transferred mm hardness HV

Carburising 850-950 (1562-1742) 4-10 h C 0.2-1.5 750-850

Carbonitriding 750-900 (1382-1652) 2-5 h C+N 0.1-0.8 750-850

Austenitic

nitrocarburising 600-700 (1112-1292) 2-4 h N+C 0.1-0.5 750-850

Nitrocarburising 560-580 (1040-1076) 2-4 h N+C 0.05-0.2 450-1200

Nitriding 500-510 (932-950) 5-100 h N 0.05-0.8 450-1200Increased

distortion

Table 2. An example of energy requirements for two process

routes, one being nitrocarburising [2].

Process step Energy requirement, KWh

Harden, temper Nitrocarburising

and nickel plate

Heat to processing 265 163

temperature

Hold at processing 27 66

temperature

Quench 5

Heat to tempering 94

temperatureHold at tempering 44

temperature

Electroplating 20

Total 455 229

Classical gas nitriding was developed for the purpose of increasingfatigue strength and load bearing capacity without involving sig-nificant distortion of treated components. Specially alloyed nitridingsteels are used in order to achieve a high hardness. Very long nitrid-ing times, from ten to hundreds of hours, have been and are used toobtain sufficient case depths.

Nitrocarburising began to grow drastically with the developmentof the salt bath process Tenifer (Tufftride) and the gaseous process


6/48

6 Introduction

Nitrocarburising

Nitriding

Nitemper in the sixties. Compared to classical nitriding, nitrocarbu-rising is a short-time process, typically lasting 30 minutes to 4 hours,and is performed at a higher temperature, about 570 C (1058 F)compared to 500-510 C (932-950 F) for gas nitriding. It is typicallyapplied to low alloy steels but also to unalloyed steels and castirons as well as to tool steels.

There are numerous names on the market to describe nitriding andnitrocarburising. Some of them are given in Table 3. NITROFLEXisLindes trademark for nitriding and nitrocarburising processes.

The process medium can be salt, gas or plasma. The salt bath proc-

esses are losing market to atmospheric gas pressure processes dueto the environmental problems with salts, which contain cyanide.The use of plasma processes has steadily increased in recent dec-ades although the number of installations is still limited in compari-son with atmospheric pressure processes.

Table 3. Nitriding and nitrocarburising features andprocess names.

Temperature Time Name or Media

C (F) h trade name

500-510 5-100 Gas nitriding NH3/N2/H2

(932-950) Oxynitriding NH3/Air

NH3/ H2O

NH3/ N2O

Plasma nitriding N2/H2

High pressure N2, NH3

nitriding

560-580 0.5-5 NITROFLEX NH3/N2/CO2/

(1040-1076) (H2/CO/C3H8)

Nitemper NH3/Endogas

Nitrotec NH3/Endogas/air

Nitroc NH3/Exogas

Tenifer/Tufftride Salt

Plasma N2/H2/CH4

nitrocarburising

Post oxidation Air, H2O, N2O


7/48

Part properties

Static and fatigue strength Contact loadfatigue

Abrasive wearresistance

Adhesive wearresistance

Corrosionresistance

High Moderate

High alloy and

nitriding steels

Steel

Compound layer

Diffusion layer

Process

Low alloy steels, cast

irons, sintered steels

High and low alloy

steels

High and low alloy steels, cast irons, sintered steels

Minor influence High hardness, high -content

High hardnessand depth

Moderate hardnessand depth

High hardness anddepth

Minor influence

Dense,

postoxide

Gas nitridning Nitrocarburising Gas nitridning and nitrocarburising Nitrocarburising +postoxidation

Table 4. Relation between part properties, steel selection and nitriding/nitrocarburising parameters.

7

The functional properties of a part, such as fatigue and static

strength, or wear and corrosion resistance, are the basis forspecifying the proper process and steel as illustrated in Figure 2.

The steel alloy and carbon content and the type of prior heat treat-

ment determine the core strength of the part and will also influencethe development of the compound layer structure, case depth andhardness (see Figures 17 and 19). Surface fatigue strength and wearresistance are therefore to a great extent dependent on the steelspecified and will improve with increased alloy content.

The functional part properties that essentially depend on the com-pound layer are wear resistance, tribological properties, corrosionresistance and general surface appearance. Both abrasive andadhesive wear resistance increase with hardness and with minimisedporosity of the compound layer. Porosity can be positive in lubricatedmachinery parts as the pores act as lubricant reservoirs. The

compound layer depth has to be deep enough not to be worn away.The diffusion layer (depth, hardness and residual stress) determinessurface fatigue resistance and resistance to surface contact loads.

Cost is a factor that limits the number of options. Variable costsincrease proportionally to increased time for nitriding/nitrocarbu-rising. Increased treatment temperature will also increase runningcosts due to higher energy losses and wear of equipment (See Table2). Capital costs are highest for vacuum and plasma equipment, butthe variable cost for utilities (electricity and gas) is much lower thanfor conventional atmospheric furnaces.

II. Process Selection

Figure 2.

The steps to process specification starting from required part properties.

The interdependence between required functional properties,the selected steel and nitriding/nitrocarburising specifications isillustrated in Table 4.

Required functional propertiesFatigue, wear, corrosion resistance

Compound layerStructure, chemistry,

hardness, depth

Diffusion zoneHardness, depth

SteelAlloy, content, heat treat

condition, core hardness

Process specification

Temperature, time nitridningpotential, gas composition

Process Selection


8/48

8

Part geometry and dimensions may limit the number of possiblealternatives. Very long objects such as shafts or axles cannot betreated in standard box or sealed quench furnaces, for instance,but tend to be treated in cylindrical pit furnaces. Requirements fordimensional tolerances may also limit the number of alternativeswith respect to treatment temperature as increased treatmenttemperature leads to increased dimensional changes. Cooling optionsare also limited as increased cooling severity will increase distortion.

Nitriding and nitrocarburising have significant cost benefits thatinclude:

1. Cycle times:Short cycle times lead to reductions in energy and

atmosphere costs. In addition, the furnace capacity and thefurnace equipment life will increase. This will also increase theproduction capacity of the manufacturing facility.

2. Materials:Many engineering components are made fromexpensive highly alloyed steels, not because the properties theyoffer are required throughout the material, but because they arerequired at or near the surface. The use of lower cost materialscombined with nitriding/nitrocarburising can reduce costs.

3. Energy costs:There are a number of possibilities to decreaseenergy costs. These include decreasing cycle times, as described

above, and decreasing process temperature. Nitriding/nitrocarburising processes operate at lower temperatures thanconventional methods, resulting in energy savings.

4. Dimensional scraps:The low temperature nitriding/nitrocarbu-rising processes reduce distortion, compared to high temperatureprocesses.

5. Elimination of post grinding:It is possible in certain situationsto predict the dimensional change after treatment. The limiteddistortion makes it possible to eliminate tolerance adjustmentoperations such as post grinding.

6. Additional processes:With gaseous treatments there is no needto remove salt residues from the surface after treatment, thuseliminating post cleaning operations. Nitrocarburising treatmentsproduce a compound layer which is scuff and corrosion resistant.

This layer does not need to be ground after treatment, and ismore flexible than the white layer that is produced by traditionalnitriding treatments. Processes also produce an excellent surfacefinish that requires no post operation and that can be useddirectly on the assembly.

Process Selection


9/48

9

A nitriding/nitrocarburising cycle has three major steps: 1) heating

to temperature, 2) holding at temperature for a sufficient time toreach the required nitriding depth, and 3) cooling. This is il lustratedin Figure 3, which also shows the optional additional steps of pre-heating/pre-oxidation and post-oxidation used in nitrocarburising.

III. The Process Steps inNitriding and Nitrocarburising

There may also be pollution from manufacturing machinery in the

form of hydraulic fluids, tool wear debris, chips, turnings, blastingagents and abrasives, and, if machines are used for different metalssuch as aluminium, even residues from non-ferrous metals. Anti-cor-rosives used to protect parts from rust in storage and transport maybe a further source. Contaminations may be in the form of surfacefilms or layers, or particles.

Haase [3] studied the effect of different additives in cutting oils onnitriding results with results shown in Figure 4. An increased amount ofadditive reduced hardness and gave uneven and locally zero compoundlayer thickness. In another study Persson and Troell [4] found that thespecific chemicals sulphur and phosphorous added to the cutting oil as

well as sodium, boron, and calcium in cutting fluids all had a negativeimpact on compound layer formation. They also found a negativeinfluence if fluids were allowed to dry on the surface before nitrid-ing/nitrocarburising.Time

Temperature Preheat/

preoxidize at350-450C(662-842F)

Heating

Nitride/Nitrocarburiseat 500-580C(932-1076F) for 30minutes to 100 hours

Oxidize at450-550C(842-1022F)

Cool ingas or oil

Figure 3. Nitriding/nitrocarburising cycle.

A. Prior Heat Treatment Condition

For parts subjected to high stress, the normal state of the steel priorto nitriding or nitrocarburising is hardened and tempered at a tem-pering temperature at least 20-30 C higher than the nitriding/nitro-carburising temperature in order to prevent loss of hardness duringnitriding/nitrocarburising. If nitriding/nitrocarburising is conducted

primarily to increase resistance to wear and scuffing, steels inannealed or normalized conditions can be used. Cast irons may benitrided or nitrocarburised in the annealed state.

For parts that have been subject to turning, drilling or any other ma-chining or cold forming operation, it is necessary to release internalstresses by stress-relieving annealing. After stress relieving the partdimensions are adjusted by fine machining or grinding to meet thetolerance requirements before nitriding/nitrocarburising. The tem-perature for stress relieving should preferably be 20-30 C above thenitriding/nitrocarburising temperature in order to avoid stress reliev-ing and concurrent distortion during nitriding/nitrocarburising.

B. Cleaning

Cleaning is an important process step before nitriding/nitrocarburis-ing as surface contaminants disturb nitride layer formation. In manu-facturing steps before heat treatment, contamination sources arelubricants, coolants and cuttings oils used in machining and grinding.

commercial product

600 700 800 900

fatty ester

mineral oil (base)

S/P additive (1%)

S additive (1%)

S additive (2%)

P additive (1%)

P additive (2%)

Surface hardness HV0.5

Figure 4. Uneven nitriding results due to the effect of surface passivation. Micro-

graphs showing compound layer after a) grinding and b) milling. c) Surface hard-

ness of steel after machining with mineral oil containing different additives. Steel

42CrMo4 [3].

The Process Steps in Nitriding and Nitrocarburising

b) milleda) ground 10 m

10 m

c)


10/48

10 The Process Steps in Nitriding and Nitrocarburising

Cleaning agent residues can lead to passive surface layers if not re-moved by thorough rinsing. Their negative effect on surface hardnessafter nitriding depends on their volatility; inorganic salts such as sili-cates and phosphates have high melting points and do not vaporizeat nitriding temperature. Furthermore, they form glassy films, whichmay completely prevent diffusion or nitrogen uptake. However, if acleaning installation provides rinsing stages, they can be removedcompletely.

Water-based cleaning solutions have to a large extent replaced chlo-rinated hydrocarbons that are no longer permitted for environmentalreasons. In addition to water (> 80%), the washing liquid consists ofactive cleaning agents such as surfactants, inorganic builders, andcomplex agents and anti corrosives. The surfactants remove oil filmswhereas particles are removed by the inorganic builders.

The amount of the metal ions calcium and magnesium define the

water hardness. Both these ions are detrimental if left on the surfacebefore nitrocarburising as this may lead to spots with thin or no com-pound layer [5]. The washing agents dissolved in the water shouldreact with and bind the Ca and Mg ions into chemical complexes,thereby eliminating the deposits of free ions and the risk of weakspots. If the water hardness is high, then the amount of washingagent and complex building capacity may not be sufficient to bindall Ca and Mg ions into complexes to avoid spot defects. Low waterhardness is therefore preferred to guarantee good and even nitrocar-burising results.

Plastic deformation in the surface region prior to nitrocarburising will

also reduce compound layer growth. A 25 m plastic deformationzone has been found to reduce the layer thickness by 30% [4].

C. Preheating and Pre-Oxidation

Preheating in air at a temperature in the range of 350-450 C (662-842 F) for 30-60 minutes is a standard procedure before nitrocarbu-rising for a number of reasons:

The process time in the nitrocarburising furnace, which is morecostly than the preheat furnace, is reduced.

Heating in air leads to surface oxidation that is found to acceler-

ate compound layer nucleation and growth during nitrocarburis-ing, see Figure 5.

Residues on the part surfaces are oxidized and vaporized, result-ing in cleaner parts and improved nitriding ability.

Safety is ensured for salt bath nitriding/nitrocarburising by re-moving any water that has adhered to the parts.

One possible reason for the positive effect of pre-oxidation is thatoxide formation results in a notable increase in the surface area,

which facilitates nitrogen uptake and the nucleation and growth ofthe nitride compound layer [6]. Other possible explanations are thatnucleation and growth of -carbonitride is facilitated and/or that thereverse process of nitriding, i.e. desorption of nitrogen, is retarded byoxidation [7].

D. Nitriding

Gas nitriding is typically performed in convection furnaces, either ofpit type as in Figure 6 or a box furnace, at a temperature in the range

of 500-520 C (932-968 F) and in an ammonia atmosphere. The am-monia may be diluted with nitrogen or hydrogen. The parts to be ni-trided are placed on fixtures or in baskets, which are transferred toand loaded into the furnace. The furnace cover or door is then closed.To ensure precision with regard to compound layer and diffusion zonethickness, it is important with enough good furnace temperatureuniformity, typically 5 C (9 F).

0

Distance from surface, nm

0

5

100 200 300

10

15

20

50

Nitrogenconcentration,atom

%

Pre-oxidized,400C (752F), 15 min

No pre-oxidation

Figure 6.

Example of a nitriding furnace installation [courtesy of Nitrex Metal Inc].

Figure 5. Nitrogen concentration profiles after short-time nitriding of the cold

work steel X155CrVMo12-1 with and without pre-oxidation [3].

1Sours of process gas(tanks, process gas generator)

2Nitrex Gas Nitriding Process Control System

3Nitrex Furnace

4Turbo coling system

5Closed-loop water cooling system

6Racking7Exhaust gas neutralizing equipment


11/48

11

A nitriding cycle consists of the steps illustrated in Figure 7. Nitrogenpurging of the furnace is conducted to remove air before ammonia islet into the furnace. This purge is carried out to eliminate the risk ofexplosion, as ammonia and oxygen form an explosive mixture withina certain concentration range. For this reason it is also advantageousto perform heating to nitriding temperature in nitrogen. When thenitriding temperature is reached, ammonia is let into the furnace. Inthe beginning a high f low rate is used to increase the efficiency ofnitrogen transfer to the steel surface.

Figure 7. Nitriding steps and connected gas consumption.

As soon as a compound layer is formed, the nitriding rate is control-

led by diffusion from the layer into the steel. The ammonia flow ratemay then be lowered just to give a nitrogen transfer rate from gas tosurface which keeps up with the diffusion rate [8]. Nitriding is contin-ued until the desired nitriding depth is reached.

After finishing the nitriding step, the furnace interior is purged withnitrogen to remove the ammonia gas in order to ensure safety.

Cooling should continue in nitrogen to avoid discoloration by oxida-tion. In pit furnaces the retort is normally lifted out of the furnace

ConditioningPurging Processing Cool and Purge

Temperature

Time

Gas

flowr

ates

Temperature

N2 NH3 H2

and put into a cooling station. In furnaces without a retort, coolingtakes place in the furnace.

Nitriding furnaces have to be tight for safety reasons and also be-cause of the odour resulting from possible leakage of ammonia gas.

E. Nitrocarburising

In principle, the same type of furnaces can be used in nitrocarburis-ing as in gas nitriding; however, one special feature of nitrocarburis-ing is that the final cooling is usually fast. Brick-lined sealed quenchfurnaces with an oil quenching capability of the same type as forcarburising are therefore used (see Figure 8a). Other common solu-tions are box-type atmosphere furnaces, often with fibre lining, andbatch furnaces with a vacuum pumping option for quick atmosphereconditioning and with integrated gas cooling (Figure 8b), as well asmetallic retort furnaces of the type shown in Figure 6. Table 5 gives

specific advantages and disadvantages of each type of furnace.

Table 5. Specific features of each furnace type.

Furnace type Advantage Disadvantage

Brick-lined Slow ammonia Slow changeatmosphere furnace dissociation in atmosphere

Modular construction

Metallic retort Fast change in Fast ammoniafurnace atmosphere dissociation

Low nitriding potential Lifetime of retort without hydrogen addition

Evacuation possible

Figure 8. Furnaces for gaseous nitrocarburising: a) Sealed quench furnace with

integrated oil quench bath. b) Side view of one chamber vacuum/atmosphere

furnace with integrated gas cooling also showing the gas system and c) Cross

section of the same furnace as in b) [Courtesy of Ipsen International GmbH].

After the parts have been cleaned, they are loaded into baskets orfixtures and transferred to the furnace for heating (preceded by pre-heating) to process temperature, 570-580C (1058-1076F), and keptat that temperature for a time that yields the desired compound layerand diffusion depth. As in the case of nitriding, close temperatureuniformity, typically 5 C (9 F), is required. In addition to containing


a b c


12/48


13/48

13

advantage of almost no effluents (see Table 6), the possibility totailor the phase constituency in the compound layer to be pure or pure , and the possibility to nitride to only yield a diffusion zonewithout the compound layer. Plasma nitriding can be used at verylow process temperatures, down to 400-450 C (752-842 F), whichmeans that hardened and tempered steels can maintain their hard-ness. Lower distortion is a connected benefit. Because of the ionbombardment, surfaces are activated to make it also possible to ni-tride stainless steels and even the nonferrous metals aluminium andtitanium.

Table 6. Comparison of effluent emission data for gaseous and

plasma nitrocarburising [10].

Plasma Gaseous Reduction in %

Amount of gas used

m3/h (100ft3/h) 0.6 (0.21) 6.0 (2.12) 90 (32)

Total carbon emission via

CO/CO2, mg/m3 504 137 253 99.63

Total amount of NOXgas,

mg/m3 1.2 664 99.82

Output of residual carbon

bearing gas, mg/h 302 823 518 99.96

Output of residual NOXgas 0.72 3 984 99.98

The combination of plasma nitriding with atmosphere nitrocarburisinghas been shown to give excellent surface appearance, wear and cor-

rosion resistance. An example is the Corr-I Dur process, see Figure 11.

G. Post Oxidation

A remarkable improvement in corrosion resistance is obtained ifnitrocarburising is followed by a short oxidation in the temperaturerange 450-550C (842-1022F). The aim is to create a Fe3O4oxidelayer with a thickness of about 1 m formed on top of the compoundlayer, as shown in Figure 12. Fe2O3must not be formed because itdeteriorates both the aesthetic surface appearance and corrosionresistance. If done properly, the oxidation treatment gives the proc-essed parts an aesthetically attractive black colour with high surfacecorrosion resistance.

The first gaseous post oxidation process was developed by Lucas,England, and is called Nitrotec [1]. It is based on the Nitemper proc-ess with an added oxidation step in air. Other oxidation methodsusing water vapour [12] or nitrous oxide (N2O) [13] have later beensuccessfully developed and found to yield good properties.

Oxide layer 1-3 m

Compound layer 5-25 m

Diffusion 100-500 m

Oxide, ( 3 m)

Compound zone

Core microstructure

H. Austenitic Nitrocarburising

Austenitic nitrocarburising is developed in order to create thickercases that can sustain greater surface loads or bending stresses. Itis performed at a temperature above the temperature for the partialtransformation of the steel to austenite. At the process temperatureaustenite enriched with carbon and nitrogen is formed beneath the

compound layer. Upon cooling after finalised nitrocarburising someof this austenite will remain as retained austenite and some willtransform into bainite, pearlite or martensite. A subzero treatmentwill transform the retained austenite further into martensite with ahardness in the range of 750 to 900 HV. Alternatively, a temperingoperation can be carried out to transform the retained austenite intobainite/martensite. This will also raise the hardness both in the diffu-sion zone and in the compound layer as shown in Figure 13.

Figure 13. The influence of tempering at 350C (662 F) for 2 hours on the

hardness profile across an austenitic nitrocarburised case produced at 700 C

(1292 F) in a mixture of 14 % ammonia and 86% endogas [14].

Figure 11. Micrograph cross section of a steel specimen treated in

the CORR-I-DUR process [11].

0

Distance from surface, mm

0

Hardness,

HV0.1

100

200

300

400

500

600

700

800

0.1 0.2 0.3 0.4

/Fe3C

layer

/

layer

225HV0.5

200HV0.5

CORE

Austenitic nitrocarburisedat 700C (1292F)

Austenitic nitrocarburisedat 700C (1292F) and temperedat 350C (662F) for 2 hours

Figure 12. Compound layer with oxide layer at top.



14/48

14

I. Combined Processes

PVD (Physical Vapour Deposition) and CVD (Chemical Vapour Deposi-tion) are processes that produce coatings such as TiN or CrN with veryhigh hardness. These coatings are thin and their load bearing capac-ity is limited, which makes the coating susceptible to failure becauseof cracking or flaking. Performing nitriding or nitrocarburising as apreceding operation before the PVD or CVD coating improves bearingcapacity. Combined processes have therefore been developed. Figure14 shows a cross section of equipment capable of performing bothplasma nitriding and PVD coating. An example of the resulting hard-ness profile after nitriding + PVD/TiN coating is also shown in theFigure.

the hardness in the diffusion zone. This effect is pronounced only inunalloyed or low-alloy steels. For steels containing high concentra-tions of nitride forming alloying elements such as chromium and alu-minium, the precipitation hardening effect of nitrides dominates thehardness contribution and the solution hardening effect is negligible.

0

0

Hardness

HV

0 1 10 100 1000

500

1000

1500

2000

2500

PVDPlasmanitriding

Distance, m

Figure 14. a) Cross section of a unit for both plasma nitriding and PVD coating.

b) Result of plasma nitriding followed by PVD-TiN coating. Note logarithmic

distance scale. [Courtesy of Sulzer Metaplas GmbH].

Figure 15. Microstructure showing precipitates of needle-like nitrides in the

diffusion zone.

An increased cooling rate increases the compressive residual stress inthe case. Both hardness and compressive residual stresses contributeto fatigue hardness, which means that the best fatigue hardness isobtained with high cooling rates. Increasing part dimension leads toa decreased cooling rate for a given cooling setup. This leads to low-ered hardness as seen in Figure 16.

0

100100

200

300

400

500

600

700

800

900

1000

0 0.1 0.2 0.3 0.4 0.5

Hardness.H

V

Depth, mm

10 mm

40 mm

Figure 16. Hardness of bars with two diameters after gaseous nitrocarburising

followed by oil quenching. Steel approx. corresponding to 16MnCr5. [15].

J. Cooling/Quenching and Post Treatment

Cooling after nitriding is usually performed in nitrogen gas with no ormoderate convection, whereas the cooling step after nitrocarburis-

ing can be accelerated by quenching in oil or by forced gas cooling(water quenching is an alternative after salt bath processes). If thecooling rate is low, it will lead to the precipitation of needle-like ironnitrides in the diffusion zone when unalloyed steels are treated asshown in Figure 15. With increased cooling intensity the amount ofnitrogen, kept in solid solution in ferrite, is increased. This increases

If high fatigue strength or surface contact load bearing capacity isrequired, a fast cooling should be applied. If wear and/or corrosionresistance is the main objective, the properties of the compoundlayer are decisive. Compound layer properties are to a lesser extentdependent on the cooling rate. Therefore the cooling rate is unimpor-tant in such cases. Productivity is of course increased with increasedcooling rate.

The number of available cooling options depends on the type of fur-nace used as illustrated in Table 7. Water, the fastest type of quench

medium (not included in the table), is used only in connection withsalt bath nitriding/nitrocarburising. Oil quenching, the second fast-est, is common as it is the standard quench method in sealed quenchfurnaces for which gaseous nitrocarburising was primarily developed,such as in the Ipsen Nitemper process. Gas quenching is growing inimportance for several reasons: Gas quenching is the only method


Compoundzone

Nitrideprecipitates

Process gasVacuumpumpset

Window

Infrared-temperatur-measurement

Substrateholder

Pulsed DCgenerator forplasmanitriding

Coatingchamber

Powersupplies

CircularEvaporators

Vacuum measurementand control systems

+

+


15/48

15

that can be integrated into single chamber furnaces of both vacuumand atmosphere types. Gas quenching has environmental advantagescompared to oil quenching, both for the external and work environ-ment. Gas quenching also has the advantage that post cleaning canbe eliminated. It is possible to increase and vary the cooling rate byincreasing and varying the gas pressure and velocity.

Table 7. Cooling options for different types of nitriding/nitrocarburising furnaces.

Furnace type Possible cooling options

Sealed quench furnace with integrated oil quench Oil

Gas, slow convection

Box furnace Gas, slow convection

Box furnace with integrated high speed gas circulation Gas, high convection

Pit retort furnace Gas, slow convection

Furnace line with integrated high pressure gas cooling Gas, high pressure and high convection

Single chamber Gas, slow convection

Single chamber/high pressure High pressure, high convection

Furnace line with integrated high pressure gas cooling High pressure, high convection

Atmosphere furnaces

Vacuum furnaces(including plasma furnaces)



16/48

16

IV. Properties of Nitrided and Nitrocarburised Steels

Nitriding and nitrocarburising of parts is carried out to improve wearresistance and strength, in particular fatigue strength, and corrosionresistance. Sometimes improved aesthetic surface appearance is alsoa desired property.

A. Hardness and Wear Resistance

The case on nitrided/nitrocarburised steel is divided into the com-pound zone, typically 5-20 m thick, and the diffusion zone, with athickness typically of several tenths of a mill imetre. The hardness ofthe compound zone is about constant through its thickness with theexception of the outer porous zone, where hardness is reduced due

to porosity. It is higher than the diffusion zone hardness, which con-tinuously decreases from the surface into the steel interior.

Compound layer hardness after nitrocarburising is about 700 HV forlow alloy steels and hardness increases with increasing alloy con-tent in the steel as shown in Figure 17. The measured hardness fallsas the degree of porosity in the outermost compound layer surfaceincreases. Generally porosity is greater for low alloy steels than forhigh alloy steels.

ficient time for the precipitation of iron or alloying element nitrides(Figure 15), which reduces the hardening effect from solid solutionhardening, see Figure 18. This is why low alloy steels are normallyquenched in water or oil after nitrocarburising.

600

0.00

Alloy content, weight %

700

800

900

1000

1100

1200

5002.00 4.00 6.00 8,00 10.00 12.00 14.00

Hardness,

HV

unalloyed

1.5% Cr

1% Cr

4% Cr

5% hot worktool steel

12% Cr

tool steel

13% Cr stainless steel

Figure 17. Approximate relation between hardness of the compound layer and

alloying content (Sum of Cr+Mo+V+W) of the treated steel [16].

There are two mechanisms which determine diffusion zone hardness.First, solid solution hardening is a mechanism that is of high impor-tance for low alloy steels. The process temperature determines thedegree of solid solution of nitrogen, carbon and alloying elements.Quench rate from the process temperature determines how muchcan be maintained in solid solution. A slow cooling rate allows suf-

The second hardening effect is precipitation hardening. This harden-ing process is predominant with alloyed steels. A dramatic conse-

quence of this mechanism is that a hardness range as wide as 300-1300 Vickers is obtained depending on which steel has been nitridedor nitrocarburised, see Figure 19.

The hardness of the compound layer will determine wear resistance.Increased hardness in the compound layer gives increased resist-ance to abrasive wear, in which abrasive particles such as sand weara surface. An important point is that compound zone hardness afternitriding or nitrocarburising can give rise to low level abrasive wearsituations. This means a very low wear rate when the surface hard-ness is higher than the hardness of the abrasive particles. The limiteddepth of the compound layer is a drawback and thus nitriding or

nitrocarburising can only be used successfully in mild abrasive wearsituations.

Abrasive wear resistance is proportional to hardness and for nitrocar-burised or nitrided steels is therefore dependent on the type of steelused (See Figure 19). For alloyed steels hardness and thus abrasive

Depth below surface, mm

H

ardness,

HV

0.00

150

170

190

210

230

250

270

290

0.05 0.10 0.15 0.20 0.25 0.30 0.35

water

air

oil

Properties of Nitrided and Nitrocarburised Steels

Figure 18. Influence of quenching on diffusion zone hardness after salt bath

nitrocarburising of steel C15 [17].


17/48

17

wear resistance is superior to that of carburised steels. However,abrasive wear will increase drastically when the thin case of a nitro-carburised steel part is worn off. The thicker case of a carburised partwill result in longer resistance to wear.

Figure 20 shows that both the dry and lubricated adhesive wearresistance of nitrocarburised steels is superior to that of carburisedsteels, and is higher than could be expected from hardness alone.The compound layer, which is a ceramic, gives low friction and alow tendency to weld opposing steel surfaces. The oil retentionproperties of the porous outer zone of the compound layer offerself lubrication properties similar to those of sintered non-ferrous

bearing materials. These properties act together to give excellentadhesive wear resistance. Adhesive wear resistance improves with

an increased ratio / in the compound layer [18] and is highest for100% -phase layers [19]. The initial lubricated wear rate of nitrocar-burised parts is higher than the steady state wear rate, probably dueto porosity. In certain situations this can be an advantage as running-in wear will smoothen the contact surfaces, which leads to a lowsteady state wear rate.

Nitriding or nitrocarburising improves the service life of hot and coldwork tools. The adhesive wear resistance of the compound layercontributes to this effect, as does the fact that the hardness of thediffusion zone of hot work tool steels is maintained at temperaturelevels of the order of 500 C (932 F). Aluminium extrusion dies is an

application in which nitriding and nitrocarburising have proven toyield improved wear resistance and endurance [21].

0

Nitriding 510C (950F)/24 h

0

0.1 0.2 0.3 0.4

Depth, mm

200

400

600

800

1000

1200

Nitrocarburising 570C (11058

F)/2 h

0

0.01

Depth, mm

200

400

600

800

1000

1200

HardnessHV

0.02 0.1 Logscale0.04 0.2 0.4

HardnessHV

5 % Cr hot work tool steel

Al alloyed steel

casehardening steel

lowalloyed steel

5% Cr hot work tool steel

low alloyedsteel

1,5% Cr air hardening steel

mild steel

1% Cr oilhardening steel

1,5 % Crair hardening steel

Figure 19. Typical hardness after nitriding and nitrocarburising. Please note the logarithmic length scale in the right hand figure [13].

Figure 20. Adhesive wear resistance measured by a pin/disc test: a. With lubrication. b) Dry (without lubrication). [20]

Sliding distance, m

Wearvolume,mm3

0

0.000

0.001

0.002

0,003

0.004

0.005

0.006

0.007

0.008

0.009

426 1096 3059 5044 7096 9118 10110

Carburised

Nitrocarburised

Sliding distance, m

Wearvolume,mm3

0

0.000

200 400 500 600 700 1200 1400

0.010

0.020

0.030

0.040

0.050

0.060

0.070

0.080

0.090

Carburised

Nitrocarburised



18/48

18

Table 8.

Bending fatigue strength improvement after nitriding [22].

Type of steel Bending fatigue strength increase,

Nitrided/Not nitrided, %

Smooth test bars Notched test bars

Nitriding Cr-Mo-Al

steels Cr-V 28-41 64-130

Cr-Mo-V

Oil hardening Cr-Mo-Mn

steels C-steel 43-55 7-96

Cr-Mo

The fatigue strength is improved by a state of compressive residualstresses in the surface layer after nitriding/nitrocarburising. This isalso the case for the diffusion zone as illustrated in Figure 21. Thereis also a state of compressive stress in the compound zone if it con-sists of a -layer, whereas tensile surface stress has been predictedbut not clearly confirmed for -layers [23]. The highest compressivestresses are obtained when nitrocarburising is completed with a fastquench in water or oil. As distortion will increase with increasedquench severity, it is necessary to find the quench rate that gives thebest combination of strength and minimised distortion.

Fatigue crack initiation in the compound zone has been observed,

especially after gas nitriding. Therefore there is sometimes a needto perform gas nitriding without the formation of the compound layer

Figure 21. Residual stress profiles after nitriding and nitrocarburising [24].

B. Static and Fatigue Strength

Static strength increases with increased surface layer hardness andrelative case depth (= case depth divided by part thickness). Relatedto nitriding process parameters this means that static strength in-creases with increased nitriding/nitrocarburising time and also withincreased cooling rate after nitrocarburising. In order to achieve thebest effect from solution and precipitation hardening, it is essentialthat cooling starts from the nitrocarburising temperature before thetemperature drops. A drop in temperature from 550 (1022) to 450C(842F) can reduce static strength by more than 20 % [1].

The improvement in fatigue strength is greatest for specimens withnotches that act as stress raisers, see Table 8. This is because thehard case will withstand high stress much better than the core mate-rial.

(often called white layer after nitriding). This is possible by properadjustment of the nitriding potential of the nitriding atmosphere (seesection VI, page 30). In rare cases the white layer is removed afternitriding to obtain the best fatigue properties.

C. Selection of Nitriding/Nitrocarburising Case Depth

The case depth is determined by process time and temperature andby the type of steel nitrided. After nitriding/nitrocarburising thereare two depths that are of interest: 1) the compound zone thicknessand 2) the diffusion zone depth. Which depth is important dependson the application involved, as outlined in Table 9.

Depth below surface, mm

Residulstress,

MPa

0

600

500

400

300

200

100

0

0.2 0.4 0.6 0.8 1

Nitrided

Nitrocarburised

Table 9.

Selection of surface properties after nitriding/nitrocarburising.

Type of application Primary objective Depth

Abrasive wear Compound zone hardness Enough for wear life

Adhesive wear Compound zone hardness, Enough for wear life

porosity and phase

constituency

Corrosive protection Compound layer porosity Enough to ascertain a

and phase constituency dense layer

Contact & bending Diffusion zone hardness Enough to incorporate

fatigue and residual stress state maximum point of stress

D. Corrosion Resistance and Surface Appearance

Compared to other thermal or thermochemical surface hardeningmethods on steels, nitriding and nitrocarburising are unique in thatcorrosion resistance is improved because of the superior electro-chemical properties of the ceramic surface layer consisting of or / carbonitride. Nitrocarburised parts have excellent humiditycorrosion resistance that is better than that of mild steel. Althoughbetter than that of steels, salt water corrosion resistance is modest.Corrosion resistance is greatly improved by a post oxidation step thatcreates a thin oxide layer on top of the compound layer. As shownin Table 10, nitrocarburising can replace chromium plating or othersurface treatments for salt spray corrosion resistance. By additional

Table 10. Corrosion data [1, 2, 25].

Type of finish Salt spray corrosion resistance

(ASTM B117), hours to failure

Phosphate/Oiled 64

Passivated zinc plating 64

Electroless nickel plating 64

Passivated cadmium plating

(10/12 m) 348

Passivated cadmium plating

(25/35 m) 3800

Nitrocarburised 17

Nitrocarburising + post oxidation(10/20 m) 645

Nitrocarburising + post oxidation

(25/35 m) 3800

Hard chrome plating 120



19/48

19

oiling, waxing or painting it is possible to increase the corrosion re-sistance of nitrocarburised parts even more [1, 2, and 25].

Nitriding and nitrocarburising gives the surface a light grey surfaceappearance. By post-oxidizing, possibly combined with oiling or wax-ing, a dark black, shiny surface appearance can be obtained, seeFigure 22.

Figure 22. Example of surface appearances after nitrocarburising.

Nitriding or nitrocarburising of stainless steels will deteriorate corro-sion resistance because the passive chromium oxide layer that pro-vides the corrosion protection of stainless steels will be destroyed.(Certain plasma nitriding techniques explained in IV.F are exceptionsto this).

E. Dimensional Changes

Compound layer growth results in dimensional growth. A solid barwill therefore grow in diameter during nitriding/nitrocarburising. Fig-ure 23 shows that dimensional growth is proportional to compoundlayer thickness. A guideline is that diameter growth is 30-50 % ofthe compound layer thickness. Thick walled (wall thickness > 25mm)hollow cylindrical parts will decrease on the inner and increase onthe outer diameter by about the same factor as the growth of a solidbar.

Wall thickness, mm

Diameterchange,m

1

50.0

2 3 4 5 6 7 8 9 10

0.0

50.0

100.0

150.0

250.0

250.0

300.0

Inside diameter

Outside diameter

Wall thickness, mm

Diameterchange,m

0

10

5

0

5

10

15

20

25

30

35

40

2 4 6 8 10 12 14 16 18

Inside diameter

Outside diameter

Nitrocarburising time, min

C45

C45

16MnCr5

16MnCr5

Diametergrowth

Compoundzonethickness

Diametergrowthm

2

4

6

8

10

12

14

16

18

20

0

0 30 60 90 120 150 180 210

nitriding/nitrocarburising. The inner diameter of cylinders will shrinkfor large wall thicknesses but will grow for thin thicknesses as shownin Figures 24-25. For a real part with a complex shape the dimensionalchange is more difficult to predict, especially for a non-uniform part.

Figure 23. Growth of the compound layer and of the diameter of cylindrical

parts made from l6MnCr5 and C45 steel, depending on the nitrocarburising time[26].

Figure 24. Diameter change as a function of the wall thickness of hollow cylin-

ders (70 mm diameter) nitrided for 72 hours [27].

Figure 25. Diameter change after nitrocarburising (2h) followed by oil quench-

ing for cylinders with varying wall thickness [15].

Figure 26 shows that both the inner and outer diameter changesare positive in the range from 30 to 100 m for thin-walled longcylinders. The figure also shows that the least diameter change isobtained after salt bath treatment (Tenifer) with water quench butthat the scatter of measured diameter changes is highest for that

treatment.

The dimensional change is determined not only by compound layergrowth but also and primarily by the thermal stress history during

Figure 26. Measured diameter changes after nitrocarburising of 300 mm long

cylinders with outer diameter 108 mm and wall thickness 4 mm. Nitrocarburis-

ing was conducted by Tenifer with water quench and Nitemper with both oil and

atmosphere cooling [15].


Cooling method

Diameterchange,m

0

20

40

60

80

100

120 Water (Tenifer) Oil Atmosphere

Inner diameter

Outer diameter


20/48

20

F. Properties of Nitrided/Nitrocarburised Stainless Steel

It is not possible to carry out standard gas nitriding or nitrocarburis-ing of stainless steels in a reproducible way without deterioratingcorrosion resistance. The first reason is that the thin chromium oxidecovering the surface of stainless steels that gives them such goodcorrosion resistance acts as a barrier to nitriding. A second reason isthat the nitrogen that actually enters into the steel will form chro-mium nitrides and thereby deplete the chromium concentration in thematrix. Although high hardness can be achieved, the nitriding willresult in reduced corrosion resistance.

It is possible to partly eliminate these drawbacks with plasma nitrid-ing. The plasma activates the surface, enabling nitrogen to be trans-ferred and diffused into the steel. It has additionally been shownto be possible to avoid chromium nitride formation by performingnitriding at a low temperature. The nitriding will then result in asurface layer consisting of austenite with a highly supersaturated

concentration of nitrogen, called the S-phase. The hardness of thesupersaturated S-phase is very high, as shown in Figure 27. Corrosionresistance is maintained, as shown in Figure 28. The highly super-saturated S-phase is not thermochemically stable but will decomposeat temperatures above approximately 700C (1292F).

0

Hardness,

HK

400C (752F)/15h

40

800

1200

1600

2000

450C (842F)/15h

Distance from surface, m

10 20 30 40 500

0 20 40 60 80 100

10

20

30

40

Test time (min)

untreatedCorrosiondepth,m

550C (1022F)

500C (932F)

450C (842F)

120

0

0

200

400

600

800

1000

1200

1400

5 10 15 20 25 30

Hardness

,HV

T= 718K (445C, 851F); 22h KN = 0.293 KN = 2.49

T=780K (507C, 944F); 6h KC = inf

AISI 316

Depth, m

Figure 27. Hardness profiles of low temperature plasma nitrided 18/8 type

stainless steel [10].

Figure 28. Corrosion resistance of stainless steel AISI 316 plasma nitrided at 450

(842), 500 (932) and 550C (1022F) compared to untreated steel. Dipping test

in 50%HCL+25%HNO3+25%H2O [10].

Somers [28] has developed a method to perform gaseous nitridingof stainless steel without plasma. The problem of nitriding inhibitiondue to the chromium oxide surface is overcome by using a specialcoating applied to the steel surface before nitriding. Examples ofresults achieved with this process are given in Figure 29.

Figure 29. Surface microstructure and hardness gradients obtained after gas

nitriding at two nitriding potentials of stainless steel, type AISI 316 [28].

Examples of S-phase nitriding and carburising processes that havebeen developed are given in Table 11.


445C 22H445C 22H


21/48

21

Table 11. Different processes for S-phase nitriding and carburising of stainless steel [10].

Company Process Element Temperature, C (F) Medium

Birmingham LTPN N


22/48

22

During nitriding, ammonia (NH3) in the furnace atmosphere decom-poses into hydrogen and nitrogen at the surface, enabling nitrogenatoms to be adsorbed at the steel surface and to diffuse further intothe steel as illustrated in Figure 30a. In nitrocarburising it is addition-ally necessary to have a carbonaceous gas transferring carbon to thesteel surface. This is illustrated in Figure 30b by the carbon depositingsurface reaction with carbon monoxide, CO. A compound layer con-sisting of and nitride is formed when a sufficiently high surfacenitrogen (and carbon) concentration forms on the steel surface.

There are three major stages of nitrogen/carbon transfer:

1. From the gas to the steel surface

2. Diffusion through the compound layer

3. Diffusion into the diffusion zone

(Additional stages not considered here are transport within the gasboundary layer and carbon diffusion from the steel matrix into thecompound zone.) The concentration gradients driving the transportof nitrogen and carbon are shown in Figure 31.

V. Atmosphere Surface Interaction

NH3

H2

N

2NH3

2N + 3H2

H

CO

2

H O2

CO + H C + H O22

C

Figure 30. Schematic illustration of the nitriding and carburising processes.

CN

(gas)

CN

(surface)

CC(gas)

CC (surface)

dcN /dx

dcC /dx

dcN /dx

dcC /dx

Compound layer Diffusion zoneGas

Figure 31. Concentrations and concentration gradients of nitrogen and carbon.

a b

Atmosphere Surface Interaction


23/48

23

The flux of nitrogen and carbon from the gas to the steel surface isproportional to the concentration differences between the gas andthe surface:

dmN/dt(surface)= k1[cN (gas) cN (surf)]

dmC/dt(surface)= k2[cC (gas) cC (surf)]

Here m denotes mass, t time, c concentration per volume unit and k1and k2are reaction rate coefficients.

The transfer of nitrogen and carbon from the surface further into thesteel is controlled by diffusion. Diffusion rates follow Ficks first law,which for the compound layer and diffusion zone are respectively:

dm/dt(comp layer)= DCompdc/dx

dm/dt(diff zone)= DDiffdc/dx

Balance of mass requires that all three mass transfer rates are equal:

dm/dt(surface)= dm/dt(comp layer)= dm/dt(diff zone)

The slowest of the three stages controls the nitrogen and carbontransfer rates. For a compound layer consisting of alternating --layers, the rate will be determined by the phase with the slowestdiffusion properties.

A. The Atmosphere Nitriding Potential

The nitrogen activity in the gas is determined by the equilibrium,

NH3= N + 3/2 H2 4.1

with the chemical equilibrium constant

K = aN PH23/2/ PNH3

where PH2and PNH3are the partial pressures of hydrogen and am-monia respectively. Accordingly the nitrogen activity, aN, can beexpressed as

aN = K PNH3/ PH23/2

where aNis proportional to the concentration, cN. The ratio

rN = PNH3/ PH23/2

4.2

is commonly called the nitriding potential.

If 100% NH3(ammonia) is added to the furnace, which is fairly com-monplace in gas nitriding, some ammonia dissociates into nitrogen

and hydrogen gas when the ammonia hits the hot furnace interior.The part of the ammonia which does not dissociate is called residualammonia, and is what causes nitriding according to reaction 4.1. Inthe expression for the nitrogen potential,

rN= PNH3/PH23/2

is PNH3 the residual ammonia partial pressure and PH2is the partialpressure of hydrogen, formed by the ammonia dissociation (in addi-tion possibly with separately added hydrogen).

The ammonia dissociation may be described by the reaction,

1NH3 (1 )NH3 (residual) +3/2H2+1/2N2

where is the degree of ammonia dissociated. The total number ofmoles after dissociation is

(1 ) + 3/2+ 1/2= 1+

For a nitriding atmosphere produced from 100% ammonia, the partialpressures can therefore be expressed as

PNH3= (1 )/(1 + )PH2 = 3/2/(1 + )

PN2 = 1/2/(1 + ) 4.3

and the nitriding potential as

rN = (1 )(1 + )1/2/(3/2)3/2 4.4

(This method of calculating the nitriding potential is only fullyaccurate if the amount of ammonia consumed in the nitriding reac-

tion 2NH3 2N+3H2can be neglected in comparison to the availableamount of ammonia.)

The nitriding potential can alternatively be expressed as a function ofthe residual ammonia or the hydrogen partial pressure according tothe equations

rN= 8/3 PNH3/(1 PNH)3/2

rN=(1 4/3PH2)/PH23/2

As each dissociated ammonia molecule produces two gaseous mol-

ecules (1/2N2+3/2H2), the furnace pressure will increase as a resultof the volume expansion. A measure of the furnace pressure is there-fore an indication of the degree of ammonia dissociation and accord-ingly also of the nitriding potential [29, 30].



24/48

24

B. The Atmosphere Carbon Potential

Similar to nitriding, the carburising reaction

CO+H2 C + H2O

has an equilibrium constant expressed by

K= aC PH2O/ PCO PH2

where P stands for the partial pressure, and the carbon activity, aC , isproportional to the carbon concentration,

aC = const. cC

The atmosphere carbon potential can alternatively be expressed bythe ratios

rC1 = PCO PH2 /PH2O

rC2 = PCO2/PCO2

Endogas or exogas which both contain carbon monoxide have beenand are used as CO sources. Pure CO is an alternative carbon sourcebut the high cost is a barrier to its use. The alternative developedwith NITROFLEXtechnology is to use carbon dioxide, CO2 , which byreacting with the hydrogen inside the furnace produces CO accordingto the reaction,

H2+ CO2= CO + H2O

The hydrogen needed for this reaction originates from dissociatedammonia or from a separate addition of hydrogen. Figure 32 showsthat CO forms according to this reaction. Furnace walls, retort andload catalyse the reaction, which means that the reaction rateshown will be different for different furnaces.

Table 12. Calculated quasi equilibrium compositions of various

nitrocarburising atmospheres at 580C (1076F).

Added gas Quasi equilibrium composition (vol-%) Activities1)

mixture N2 H2 CO H2O CO2 NH2 aN aC aO50% Endogas 24.1 25.9 9.0 0.76 0.68 38.4 1620 22.4 0.07250% NH3

40% Endogas10% Air 29.2 24.3 6.7 2.46 1.77 35.4 1650 4.81 0.25

50% NH35% CO260% N2 58.8 15.5 2.9 2.98 1.45 18.4 1680 1.09 0.48

35% NH35% CO25% CO

55% N2 54.4 15.9 6.4 2.22 2.31 18.6 1640 3.32 0.34

35% NH35% CO2

20% H245% N2 45.4 26.6 3.4 3.50 1.16 19.9 806 1.88 0.33

1)The reference state for the activity values are nitrogen gas at 1 atm, graphite and wustite (FeO), fornitrogen, carbon and oxygen respectively.

0 2.5

0

Carbonmon

oxidecontent,%

1

5 7.5 10 12.5 15

2

3

4

5

6

Furnace, gas flow rate

IPSEN RTQ 1,4 m3/h

IPSEN RTQ 1,2 m3/h

UTAB OMG, 4 m3/h

Carbon dioxide addition, %

Figure 32.

Relation between the added amount of CO2and produced amount of CO.

The atmosphere carbon activities of atmospheres based on blendsbetween ammonia/endogas and ammonia/exogas are excessivelyhigh and practice has shown that soot deposits may be a problemin furnaces running with these atmospheres for a long time. An at-mosphere based on a blend between ammonia/CO2has much lowercarbon activity, see Table 12.

C. Atmosphere Analysis

The nitriding potential is determined by gas analysis. In nitrocarbu-rising the gas analysis aims to determine both the nitriding and the

carbon potentials. The analysis of nitrocarburising atmospheres isdifficult due to the complex gas composition (NH3, H2, CO, CO2andH2O) with fairly high water vapour content, which may result in theprecipitation of ammonium carbonate, which clogs the analyser andanalysing sample gas pipes. The use of a heated sample gas systemis one way to inhibit condensation.

1. Ammonia and Hydrogen Analysis

The simplest method of determining the nitriding potential is ammo-nia analysis with the ammonia burette technique shown in Figure 33.The analysing principle is based upon that ammonia gas dissolves inwater. The valves at the top and bottom of the graduated containerin the figure are first placed in purging position. Closing the upper

Figure 33. Ammonia analyser



25/48

25

valve leads to a gas sample being locked into the container. Dur-ing analysis the bottom valve is opened to allow water from the lefthand container into the measuring container. The un-dissociated am-monia will thereby dissolve in the water, causing the water to rise inthe graduated container. The more un-dissociated ammonia there is,the higher the water level will rise.

On-line analysis can be performed with infrared analysis of ammoniaor hydrogen analysis with a thermal conductivity analyser or a com-bination of both. The principle of IR analysis is based on the ability ofmulti-atomic gas molecules to absorb an IR wavelength that is spe-cific to each gas molecule. The principle is illustrated in Figure 34.A beam of infrared light is split into two separate beams, one ofwhich passes through a cell containing the sample gas, the otherthrough a reference cell, which in the figure is filled with nitrogen.A rapidly alternating splitter (on the left of the figure) separates theradiation reaching the detector on the right of the figure. A measure

of the amount of ammonia is obtained by comparing the detectorsignals for the two beams.

the resistor and thus a resistance change, which can be measured asa signal. Figure 35 shows the principle of analysis.

Hydrogen can also be analyzed with a sensor with a similar appear-ance to an oxygen probe as shown in Figure 36. It uses a measuring

tube material with the ability to sensor the difference in the hydro-gen partial pressure between a reference gas inside and the actualfurnace atmosphere outside the tube. The analysis signal is a directvalue of the hydrogen concentration in the furnace atmosphere.

NH3

N2

Light source Shutter Chopperwheel Sample cell Detector

Sample gas outlet Sample gas inlet

Reference cell Calibration cell

Figure 34. Principle of an infrared NH3analyser [Courtesy of ABB AB].

Hydrogen gas, being a one-atom gas, cannot be analyzed by IR anal-ysis. However, due to the very high thermal conductivity of hydrogengas, analysis based on thermal conductivity can be used. A classicthermal conductivity detector design utilizes a Wheatstone bridge inwhich a resistor is in contact with the gas to be analyzed. Changesin the hydrogen concentration will result in a temperature change in

Figure 35. Principle of a thermal conductivity H2analyser [Courtesy of ABB AB].

Furnace Wall

Protection TubeHead

Pressure sensorQuarz TubeMeasurment

Tube

Furnace atmosphere

Measurment TubeNH3/H2/N2/CO/CO2

pa(H2) pi(H2) H2

pa(H2) = pi(H2)

Pressure Balance

Bridge

current

60C

TemperaturecontrollerHeater

Sample gasReference gas

Figure 36. a) Principle and b) appearance of the hydrogen sensor [31] [Courtesy

of Ipsen International GmbH]

In cases where nitriding is performed with a variable unknown am-monia/nitrogen mixture, two hydrogen analysers can be used toaccurately establish the nitriding potential. The analysers are posi-tioned so that one analyses hydrogen in the furnace atmosphere,PH2 , and the second the hydrogen concentration after the completedissociation of ammonia, PH2. From these two measurements thenitriding potential is expressed by

rN = (PH2 PH2)/[(3/2 PH2) PH2]

An alternative atmosphere analysis is hydrogen analysis combinedwith oxygen probe readings.

2. Oxygen Probe Analysis

Oxygen probe analysis can be used to gain a measure of the nitridingpotential even though oxygen is not active in the nitriding process.The atmosphere oxygen partial pressure is proportional to the water/hydrogen ratio and thus the oxygen probe signal A can be expressedby the following equation

A= PH2O /PH2

The principle of the oxygen probe method is based on the fact thatthe inlet gas composition is known, as is its water content, whichmeans that PH2Ois also known. The hydrogen concentration, PH2,


a.

b.


26/48

Controlunit

FTIR and hydrogenanalyzers

Datahandling

Exhaust

Heatedsampleline

26

can be expressed as a function of the degree of ammonia dissocia-tion (see equation 4.3) and the nitriding potential from equation 4.4.A is accordingly an indirect measure of the nitriding potential.

With two separate oxygen probes the nitriding potential can be de-termined without any knowledge of the ingoing gas composition.The first probe measures the oxygen potential A in the actual furnaceatmosphere, whereas the second measures the oxygen potentialafter passing the atmosphere through an ammonia cracker, whichcompletely dissociates all ammonia ( = 1). The second oxygenprobe measures the signal:

B = PH2ODiss/ PH2

Diss

where the superscript Diss refers to the atmosphere compositionafter the ammonia cracker. If the nitriding atmosphere consists onlyof ammonia, it can be shown that the ratio between the two oxygen

probe signals is a measure of the degree of ammonia dissociationinside the furnace according to the equation [32]

D = A /B = 1/

One development of the oxygen probe to determine the nitridingpotential is the type of sensor shown in Figure 37. The outer elec-trode is in direct contact with the furnace atmosphere. The furnaceatmosphere is also led into the inner side of the ceramic tube butpasses through a catalyser that promotes complete dissociation ofall residual ammonia before it comes into contact with the innerelectrode. The resulting voltage DU is a measure of the difference

in oxygen potentials between the actual furnace atmosphere andthe atmosphere of completely dissociated ammonia. This voltage isa direct indication of the degree of ammonia dissociation, , in thefurnace atmosphere expressed by [32, 33].

DU = 0.0992 T log

absorption at one specific IR wavelength as in conventional IR spec-troscopy. The principle setup of such a system is shown in Figure 38and is connected to a furnace as in Figure 39, which also includes ahydrogen analyser.

An oxygen probe signal will also be a measure of the atmospherecarbon activity, aC, according to the relations

aC= K1PCO/(PO2)1/2= K2(PCO)

2/PCO2= K3PCOPH2/PH2O

where K1, K2, K3are constants.

3. FTIR Gas Analysis of Nitrocarburising Atmospheres

FTIR (Fourier Transform Infrared) analysis gives as a result a wave-length-dependent interference pattern called an interferogram thatmakes it possible to quantitatively determine the concentrations ofall multi-atomic gas species in a gas sample, in contrast to detecting

Figure 37. Oxygen probe as a nitrogen sensor [34].

ComputerMonitor

Plotter

Printer

Datastation

Spectrometer

This method also suffers from the clogging and corrosion problemscaused by condensation, but for short time analysis it is a method ofobtaining a complete gas analysis footprint not only from NH3andH2but also from CO, CO2, H2O, N2O etc., as indicated in Figure 40. It isused for calibrating individual furnaces.

Figure 38. Basic components of a computerized FTIR spectrometer system.

Figure 39. Analyser set up including FTIR and hydrogen analysis.

D. Layer Growth Determination

In addition to analysing the atmosphere, it is possible to register thecompound layer thickness and its growth with a sensor that utiliseselectromagnetic principles. It is inserted into the furnace as shown inFigure 41.

Figure 40. Example of FTIR analysis result.

Outer electrodeInner electrode

Gasentrance Electrolyte

Protection tube

Catalyst



27/48

27

the ammonia flow rate. A control system incorporating a hydrogensensor and using the combination of ammonia and hydrogen isshown in Figure 42. Hydrogen could alternatively be supplied fromcracked ammonia.

It is also possible to dilute ammonia with nitrogen, thereby affectingthe nitriding potential. The controllability range, however, is thenlimited, but from the viewpoints of safety and economy it may beadvantageous.

Frequencygenerator

Isolation Furnace wall

Transport rod

Slide

Measuringdevice420 mA

CoilKiNit-Sensor

Furnace atmosphere

Figure 41. Schematic principle of the installation of an electromagnetic sensor

for registration of compound layer growth. [35].

E. Guidelines for Regulating the Atmosphere

1. Nitriding

Controlled nitriding is realized by a control system that determinesthe actual nitriding potential and adjusts the atmosphere compositionto the chosen set point value. This can be achieved either by manualanalysis and flow adjustments or by automatic closed loop control.

At a high f low rate most ammonia supplied for nitriding to a furnaceremains un-dissociated, but at low flow rates it dissociates moreeasily into nitrogen and hydrogen as the residence time in the fur-nace is sufficiently long. One way to perform controlled nitriding is

therefore to start the process with a high ammonia flow rate (= highresidual ammonia = high nitriding potential) in order to build up thecompound layer as fast as possible. Later on the ammonia flow rateis decreased to typically give a residual ammonia concentration ofabout 60 vol-percent [8].

An alternative method of controlling the nitrogen activity is by hydro-gen addition. In such cases the atmosphere nitrogen activity can bevaried over a much wider range than is possible by simply changing

Furnaceatmosphere

HydroNitSensor

Temperature

H2

Controller

NH3

2. Nitrocarburising

The atmosphere for nitrocarburising consists of 20-50 % ammonia,2-20 % carbon dioxide and the balance nitrogen, the specific com-position depending on which furnace equipment is used and whichproperties are desired. Experiments have shown that an addition ofabout 5 vol-percent CO2is often sufficient (see Figure 32).

Figure 43 gives examples of analysis results for residual NH3in anitrocarburising process where 50% NH3+ 50 % endogas were intro-duced into a sealed quench ceramic lined furnace. The figure shows a

Figure 42. Nitriding potential control (Courtesy Ipsen International GmbH).

Figure 43. Examples of residual ammonia concentration variations during

nitrocarburising cycles at 570C (1058F) in a sealed quench furnace.

The b cycle was the same as the a cycle with the exception that the

furnace was preconditioned with an active atmosphere in the b cycle.

0

1

15

Residual ammonia concentration, vol%

0

2

3

4

5

30 45 600

1

15

Processtime,

h

Residual ammonia concentration, vol%

0

2

3

4

5

30 45 60

Processtime,

h

Loading intofurnace

Loading intoprechamber Loading into

prechamber

Loading intofurnace


a b


28/48

28

large variation in residual ammonia content both during one processcycle and between two different cycles. The same gas flow ratios andmixing ratios were used. Obtained depths on the compound layer anddiffusion zone correlated positively with residual ammonia analysisresults.

The shortest possible cycle for a certain compound layer thickness isachieved if the atmosphere composition is changed during the cycle,with high nitrogen activity and a carbon activity promoting -nuclea-tion and growth in the first part of the process.

For medium and high carbon steels, carbon is donated to the com-pound layer by the steel. The atmosphere carbon activity is thenless important, which explains why carbon-free atmospheres workin such cases even though such atmospheres are pure nitridingand not nitrocarburising atmospheres. With low carbon steels, onthe other hand, it is necessary that the atmosphere has a balanced

carbon activity to ensure a good compound structure with mainly -phase.

1600

10 20 30 40 50 60 70 80 90 % NH3

0.2

0.4

0.6

0.8

1.0

1.2

1.4

1.6

1.8

1000

1100

1200

1300

1400

1500

0

5

aC

aN

aN

aC

1800

% CO2

0.9

1.0

1.2

1500

1600

1700

2.5 5 10 15 20 30

1.1

aC

aN

aN

aC

Figures 44-45 show how adjusting the atmosphere concentrationsof CO2and NH3can change the atmosphere carbon and nitrogen ac-tivities. This possibility is of great value when optimising compoundlayer properties for different steels.

With the NITROFLEXsystem it is possible to use a boost techniquewith high nitrogen activity in the first stage of the process and a low-er activity in a second stage. There is an upper limit for the nitrogenactivity corresponding to excessive porosity formation. In this wayit is possible to control the degree of porosity and compound layerthickness.

Examples of the influence of atmosphere composition on compoundlayer microstructure are shown in Figure 46. For high nitrogen activi-ties entailing a high ratio for PNH3 /PH2

3/2, the compound layer is thickand has extensive porosity, as illustrated in the photographs on theleft, whereas a lower nitrogen activity gives a dense layer, as illus-

trated in the photographs on the right. By increasing the atmospherecarbon activity, the porosity decreases and the amount of -phase inthe compound layer increases, see the lower right photograph.

Figure 46. Four examples of compound layer depth and morphology as the

result of different gas recipes. Treatment temperature and time were the same

in all four cases. Unalloyed 0.1% C steel. Gas atmospheres:

a) 35% NH3, 5% CO2 , balance N2, aC = 0.93, aN = 2390,

b) 33% NH3, 5% CO2 , 5% CO, balance N2, aC = 2.95, aN = 2320,

c) 28% NH3, 4% CO2, balance N2, aC = 1.54, aN = 1300,

d) 60% NH3, balance endogas, aC=23.5, aN= 1690 [16, 36].

In order to lower the nitrogen activity during the later part of thetreatment, the addition of carbon dioxide could be lowered or com-pletely shut off. It is also possible to lower the ammonia addition inorder to lower the nitrogen activity; however, the carbon activity willdrastically decrease as well.

In order to attain high carbon concentrations, one alternative is toadd propane to the atmosphere. This has given positive results forthe obtained hardness and wear resistance of treated parts. This isbecause the carbon provided by the propane addition favours nu-cleation and growth of the -phase. This is il lustrated in Table 13,which also shows wear results. Care must be taken not to create soot

Figure 44. Theoretically calculated carbon and nitrogen activities at 580 C

(1076 F) for different ammonia additions to an atmosphere based on

N2+ 5%CO2.

Figure 45. Theoretically calculated carbon and nitrogen activities at 580C

(1076F) for different carbon dioxide additions to an atmosphere based on

N2+35%NH3.


a b

c d


29/48

29

with the propane addition. Other hydrocarbons such as ethylene andpropylene are also commonly used to provide carbon [37].

3. Post Oxidation Control

To obtain a result with an adhering iron oxide that provides corrosionprotection it is necessary that only the Fe3O4oxide is formed. If Fe2O3forms, the result will be poor surface appearance with varying colourand flaking of the oxide.

In an atmosphere with oxygen partial pressure higher than thatrequired for the formation of Fe2O3 , the equilibrium conditions areexpected to allow all three iron oxides to form. However, FeO willonly form at temperatures above 570C (1058F). The post oxidationtemperature should therefore be below that for formation of FeO,although FeO formation is very slow up to 700C (1292F) [38].

Figure 47 shows the equilibrium oxygen pressures over Fe2O3andFe3O4plotted versus inversed temperature (1/K) as straight lines. Itcan be seen that the equilibrium oxygen pressure stability range forFe3O4is approximately 10-31 to 10-30 atm at 450C (842F). Table

14 indicates that, in order to avoid Fe2O3but to form Fe3O4at 580C(1076F), the following condition must be fulfilled:2.4 1026< PO2< 3.7 10

25or 0.30 < PH2O /PH2


30/48


31/48

31Compound Layer and Diffusion Zone Formation

More nitrogen atoms thus have to diffuse into the steel to reach acertain depth compared to when no alloy elements trap nitrogen.

The nitriding rate increases if small concentrations of oxygen arepresent in the nitriding atmosphere as shown in Table 15. The tablealso shows that oxygen addition favours the formation of -phase.

The nitrogen take-up and thus also the layer growth rate increaseswith increased surface area (surface roughness) as shown in Figure53.

The addition of CO2and NO2has also been found to increase thenitriding rate. Small additions of water, however, seem to retardnitriding, see Figure 52.

0.0E+00

1.0E02

2.0E02

3.0E02

4.0E02

5.0E02

6.0E02

7.0E02

8.0E02

9.0E02

1.0E+01

1.1E01

1.2E01

1.3E01

1.4E01

m/m

0

Time (103s)

10 20 30 40 50

CrN+Fe2N1-z

CrN+Fe4N1-x

CrN

40/60 +300 ppm O

2

38/62 +100 ppm H

2O

38/62 +300 ppm H

2O

38/62

40/60

60

Table 15. Influence of oxygen addition on compound layer

thickness and phase composition in gas nitriding at 30 min-

utes at 550 C (1022 F). Steel DIN C10 [41].

Nitriding atmosphere Compound layer, Phase composition, m vol%

No oxygen 1.5-2.0 34 47

No oxygen but pre-oxidised 2.9-3.5 67 25

Oxygen added during nitriding 2.5-3.5 67 25

B. Nitrocarburising

In nitrocarburising the compound layer starts to form by nucleationof cementite even if the carbon activity of the gas is lower than thatof cementite [44, 45]. A possible explanation is that the gas/surfacereaction delivering carbon to the surface (the heterogeneous watergas reaction) is faster and kinetically favoured compared with thenitriding (ammonia decomposition) reaction during heating beforereaching the nitrocarburising temperature. There is additionallyevidence that cementite formation at moderate atmospherecarbon activity is favoured by the presence of ammonia in the

atmosphere. [46]. Within the order of minutes after reaching thenitrocarburising temperature, phase is nucleated on the primaryformed cementite. is favoured because its crystal structure is similarto that of cementite. The phase layer then grows at the expenseof cementite, which is consumed by transformation to phase,leading to an almost homogeneous phase layer. Later forms at

Compo

undlayerthickness,m

Surface roughness Rz,m

0.1

0

2

4

6

8

10

12

14

1.0 10.0 100.0

Figure 53. Dependence of the thickness of the compound layer on the mean

surface roughness Rz. 0.45%C carbon steel, nitriding temperature 570C

(1058F), nitriding time 3 hours [43]

Figure 52. Effect of oxygen and water vapour on weight gain during nitriding

in a thermo-balance at 550C (1022F) of Fe20Cr powder. The ratio NH 3/H2is

indicated at each graph [42].


32/48

32

the interface between the substrate and the -layer. Redistributionof nitrogen and carbon at the /substrate interface will eventuallycreate a second phase layer between the -layer and the substrate-phase. The compound layer will therefore ultimately consist ofthree alternating layers of //. This is indirectly shown by the Nand C concentration profiles shown in Figure 54.

The nitrogen surface concentration increases with increased processtime and increased nitriding potential (Figure 55), whereas thecarbon surface concentration decreases (Figure 54). The totalamount of nitrogen in the compound layer increases, whereas thetotal amount of carbon is constant or decreases with increasedtreatment time. Carbon is redistributed as shown in Figure 54 with adepletion of carbon in the intermediate -layer, an accumulation ofcarbon in the phase adjacent to the core ferrite/cementite matrixand a positive carbon concentration gradient in the outer phasecompound layer. For nitrocarburising of medium and high carbon

steels, carbon originating from the steel matrix is incorporated in thecompound layer, resulting in carbon enrichment of the phase, (seethe lower part of Figure 54).

0

5

10

15

20

25

N-content(at.-%

)

0 5 10 15

Depth, m

120 min

240 min

120 min

240 min

0

C-content(at.-%

)

2

4

6

8

0 5 10 15

Depth, m

4.0

0

Carbon,

wt.%

4.5 4.0 5.5 6.0 6.5 7.0 7.5 8.0

0.5

1.0

1.5

2.0

2.5

03.0

Nitrogen, wt.%

+

'

+ '

'+

cem +

Nitrogenconcentration,atomic%

10

24

Nitridning potential, Pa1/2 103

15 20 25 30 35 40

25

26

27

28

29

30

experiment

absorption isotherm

local equilibrium at /'

Surface

Interface

The contact between the - and the -phases observed at theinterface compound layer/matrix is consistent with the phasediagram evaluated by Du [39] and shown in Figure 56. (It does

not, however,

Documents

Furnace Atmoshpere