M A T E R I A L S C H A R A C T E R I Z A T I O N 5 9 ( 2 0 0 8 ) 2 3 – 3 1

Kinetics of borided 31CrMoV9 and 34CrAlNi7 steels

Gözde Çelebi Efe, Mediha İpek, İbrahim Özbek, Cuma Bindal⁎

Sakarya University, Engineering Faculty, Department of Metallurgical and Materials Engineering, 54187 Esentepe Campus, Sakarya, Turkey

A R T I C L E D A T A

⁎ Corresponding author. Tel.: +90 264 2955759E-mail address: bindal@sakarya.edu.tr (C. B

1044-5803/$ – see front matter © 2006 Elsevidoi:10.1016/j.matchar.2006.10.007

A B S T R A C T

Article history:Received 1 August 2006Received in revised form10 October 2006Accepted 17 October 2006

In this study, kinetics of borides formed on the surface of 31CrMoV9 and 34CrAlNi7 steelsborided in solid medium consisting of Ekabor II at 850–900–950 °C for 2, 4, 6 and 8 h wereinvestigated. Scanning electron microscopy and optical microscopy examinations showedthat borides formed on the surface of borided steels have columnar morphology. Theborides formed in the coating layer confirmed by X-ray diffraction analysis are FeB, Fe2B,CrB, and Cr2B. The hardnesses of boride layers are much higher than that of matrix. It wasfound that depending on process temperature and time the fracture toughness of boridelayers ranged from 3.93 to 4.48 MPa m1/2 for 31CrMoV9 and from 3.87 to 4.40 MPa m1/2 for34CrAlNi7 steel. Activation energy, growth rate and growth acceleration of boride layercalculated according to these kinetic studies revealed that lower activation energy results inthe fast growth rate and high growth acceleration.

© 2006 Elsevier Inc. All rights reserved.

Keywords:BoridesHardnessFracture toughnessKineticsGrowth rateGrowth acceleration

Table 1 – The chemical composition of test materials (wt.%)

Steels C Cr Ni Al Cu Mo Mn V Si

31CrMoV9 0.32 2.6 0.2 0.01 0.06 0.2 0.6 0.2 0.2534CrAlNi7 0.33 1.7 1.0 0.7 0.1 0.2 0.5 0.03 0.22

1. Introduction

As an effective thermochemical surface hardening technique,boriding can offer a superior alternative to conventionalsurface hardening processes, such as carburizing, nitriding orcarbonitriding [1]. In boriding process, boron atoms diffuse intothe surface of the workpiece to produce hard boride layerswithout requiring high technology [2–4]. Boriding surfacehardening process can be applied to a wide variety of materialse.g. ferrous, nonferrous, and cermet materials [5,6]. Processinvolves heating well-cleaned materials in the range of 700 to1000 °C, preferably for 1 to 12 h, in contact with a boronaceoussolid powder, paste, liquid or gaseous medium [7,8]. Sincesurface hardening process with boriding has superior charac-teristic than the other thermochemical techniques it has beenwidely used in industrial applications where the control offriction and wear is important [3,7,9]. In general, depending onprocess temperature, time and boron potential of mediumsingle-phase Fe2B or two phases (FeB, Fe2B) are obtained bydiffusing of boron atoms into the surface of metallic materials

; fax: +90 264 2955601.indal).

er Inc. All rights reserved

[6,7,10]. The single-phase Fe2B layer is preferred to the twophases layer because FeB, which is slightly harder than Fe2B, isalso much more brittle and has a coefficient of thermalexpansion that is almost three times less than that of Fe2B[6,12]. This significant difference between the coefficients ofthermal expansion of the two layers, coupled with thebrittleness of FeB, can cause cracks to form between the twolayers [11]. These cracks often lead to flaking and spallingwhena mechanical load is applied. By controlling the boridingprocess parameters, especially by limiting the boron potentialof the boriding media, the Fe2B phase can be consistentlyachieved during pack boriding [12,13] or formed FeB phase can

.

Fig. 1 –Microstructure of (A) 31CrMoV9 and (B) 34CrAlNi7 steels borided at 950 °C for 6 h.

24 M A T E R I A L S C H A R A C T E R I Z A T I O N 5 9 ( 2 0 0 8 ) 2 3 – 3 1

transform into Fe2B phase through heat treatment at hightemperatures (950–1100 °C) [12].

In this study, 31CrMoV9 and 34CrAlNi7 steels were boridedin solidmedium, kinetics parameters of borides formed on the

Fig. 2 –SEM micrographs of (A) 31CrMoV9 steel borided at 850 °C(C) 34CrAlNi7 steel borided at 850 °C for 8 h and (D) 34CrAlNi7 st

surface of test materials were investigated and how thegrowing rate and growth acceleration of borides changedepending on test parameters such as process time, temper-ature and chemical composition were determined.

for 8 h, (B) 31CrMoV9 steel borided at 900 °C for 8 h,eel borided at 950 °C for 4 h.

Fig. 3 –The variation of the depth of boride layer for (A)31CrMoV9 and for (B) 34CrAlNi7 steels as a function ofprocess time including contour diagrams.

25M A T E R I A L S C H A R A C T E R I Z A T I O N 5 9 ( 2 0 0 8 ) 2 3 – 3 1

2. Experimental Details

2.1. Substrate Materials

Substrate materials used in this study were 31CrMoV9 and34CrAlNi7 steels. The chemical compositions of the testmaterials are given in Table 1. Samples of 31CrMoV9 steelwere cube shaped in the size of 10×10×10mm and samples of34CrAlNi7 steel were semi-cylinder shaped with the 20 mm indiameter and 10 mm of height. All the samples were groundusing 600-grid emery paper to get surface finish beforeboriding heat treatment.

2.2. Boriding Heat Treatment

Boriding heat treatment was carried out in the solid mediumconsisting of Ekabor-II powder mixture in an electricalresistance furnace under atmospheric pressure at 850, 900and 950 °C for 2, 4, 6 and 8 h. followed by cooling in air.

2.3. Characterization and the Mechanical Tests

Microstructure analysis of borides formed on the test materi-als was performed with optical microscope and a JEOL JSM-5600model Scanning ElectronMicroscope (SEM). The presenceof the borides formed in the coating layer was confirmed bymeans of X-ray diffraction (XRD) analysis. The distributions ofalloying elements in boride layer (which elements accumu-lated in boride teeth and between them) for 34CrAlNi7 weredetermined by EDS parallel to surface. The hardness and thefracture toughness of borides were measured using Vickersmicrohardness tester. The equation used for calculating thefracture toughness was Kc=X ·P /c3/2 where X is residual-indentation coefficient, P the load and c the half of theindentation crack length The parameter X depends on thehardness-to-elasticity modulus ratio (H/E) of the boride layer.It equals to 0.028 (E/H)1/2. The value of E is approximately290 GPa for fracture toughness calculations [14].

2.4. Kinetics

Provided that boron diffused unidirectionaly and grownparabolically, the change of boride layer thickness with timecan be described by:

d2 ¼ Dt ð1Þ

where d is the boride layer depth (cm), t is the processduration (s) and D is the growth rate constant (cm2/s). As it iswell-known the main factor limiting layer growth is thediffusion of boron. It is possible to claim that the relationshipbetween growth rate constant, D, and activation energy, Q,and the temperature, T, can be expressed as an Arrheniusequation:

D ¼ Do:expð−Q=RTÞ ð2Þ

where D is the diffusion coefficient (cm2/s), Do is a pre-exponential constant, Q is the activation energy (cal/mol), T isthe absolute temperature in Kelvin and R is the gas constant(J/mol K) [15].

In addition the growth rate and growth acceleration ofborides formed on the steel substrates were calculatedaccording to the Eqs. (3) and (4) as follows;

Go ¼ d=t ð3Þ

26 M A T E R I A L S C H A R A C T E R I Z A T I O N 5 9 ( 2 0 0 8 ) 2 3 – 3 1

where Go is growth rate (μm/h), t is time (h) and d is the depthof boride layer (μm).

a ¼ Go=t ð4Þ

where a is the growth acceleration.

3. Results and Discussion

3.1. Microstructure and Characterization

Metallographic analysis showed that boride layers formed onthe 31CrAlNi7 and 34CrAlNi7 steel surface have a dense andcolumnar structure with zigzag teeth shape at the coating–matrix interface (Figs. 1 and 2). In coating layer both FeB andFe2B phases were confirmed by optical and SEM microscope.

It can be seen in Table 1 that the amount of alloyingelements in 34CrAlNi7 is higher than that of 31CrMoV9. Itwas observed that boride layer–matrix interface of 34CrAlNi7steel is slightly smoother than that of 31CrMoV9 steel dueto the alloying elements that it has. Increased Al and Niof matrix alloy makes the boride layer–matrix interfacesmoother [7].

Fig. 4 –X-ray diffraction patterns for borided 31CrMoV9 (A) 850 °C–850 °C–8 h, (E) 900 °C–8 h (F) 950 °C–8 h.

Depending on process time, temperature and chemicalcomposition of substrates the depth of boride layer rangedfrom 36 to 211 μm (Fig. 3). The boride layer thickness increaseswith increasing boriding time and temperature for each steel.It is seen that the thickness of boride layer formed on thesurface of 31CrMoV9 steel is longer than that of 34CrAlNi7steel for the same boriding time and temperature 34CrAlNi7steel that alloy has more Al and Ni, and less V.

Borides formed on steel substrates were identified bymeans of X-ray diffraction analysis (Fig. 4). XRD patternsshow that the boride layer consists of borides such as MB andM2B (M=Metal; Fe, Cr). Also some carbide which did notdissolve at the chosen boriding temperature was detected inthe boride layer. It was found that the dominant phases in theboriding layer are Fe2B and FeB. With increasing time andtemperature, Fe2B phase content decreases and FeB phasecontent increases for 31CrAlNi7 steel. The amount of FeBphase on 34CrAlNi7 steel decreases with increasing processtime and temperature. Probably two reasons can cause thisdecreasing:

a) Higher alloying element especially Al, Ni and Cu content ofthis steel makes diffusion of B atoms difficult and FeB

8 h, (B) 900 °C–8 h (C) 950 °C–8 h and for borided 34CrAlNiV9 (D)

27M A T E R I A L S C H A R A C T E R I Z A T I O N 5 9 ( 2 0 0 8 ) 2 3 – 3 1

phase containing more B atoms occurs. For long durationtime at process temperature results in transformation ofFeB→Fe2B thus the amount of FeB decreases. This cannotsuggest for all structure, because growth is anisotropic anddepends on the texture [7].

b) Boron potential of boriding media decreases so that thereis insufficient boron potential for FeB phase formation[12].

EDS line analysis was performed for determining whichelement accumulated among the Fe2B phase and matrix in

Fig. 5 –SEM micrograph and EDS line analysis

the 34CrAlNi7 steel (Fig. 5). In this region, Al, Si and a smallscale of Ni and Cu are determined. Since Al, Si and Cu donot dissolve in the boride layer and Ni diffuses from thesurface into the matrix [7], we can come to the conclusionthat these elements accumulated among the Fe2B phaseand matrix.

3.2. Hardness and Fracture Toughness

The hardness of the boride layer is much higher than that ofthe substrate because of the presence of hard FeB and Fe2B

of 34CrAlNi7 steel borided at 900 °C for 8 h.

Fig. 6 – (A–F) Micro-hardness profiles of specimens borided at 850 °C, 900 °C and 950 °C for 2, 4, 6 and 8 h.

28 M A T E R I A L S C H A R A C T E R I Z A T I O N 5 9 ( 2 0 0 8 ) 2 3 – 3 1

phases (see Fig. 6). Microhardness measurements were donefrom the surface to interior along a line to see variation ofhardness of boride layer, transition zone and matrix,respectively. Hardness of borides formed on the surface of31CrMoV9 steel ranged from 1273 to 1672 HVN and34CrAlNi7 steel ranged from 1389 to 1600 HVN whereasVickers hardness of the untreated 31CrMoV9 steel is 271HVN and 34CrAlNi7 steel is 318 HVN. The highest hardnessvalue was obtained after the boriding treatment of 8 h. Theboride layer hardness of 31CrMoV9 steel is higher than thatof 34CrAlNi7 steel because of easy diffusion of B atoms intothe 31CrMoV9 steel.

Indentation fracture toughness tests were performed onpolished cross-sections of the boride layers formed on thesurface of borided materials at 850–900–950 °C, for 4–6–8 h.Depending on the process time and temperature, fracturetoughness of borides formed on surfaces of 31CrMoV9 steelthat ranged from 3.93 to 4.48 MPa m1/2 and from 3.87 to4.40 MPa m1/2 for 34CrAlNi7 steel (Fig. 7). It was noticed thatthe longer boriding time results in the lower fracturetoughness. It is possible to claim that this is a consequenceof formation of hard FeB on the surface of steel substrate.

3.3. Kinetics

Some kinetic parameters such as time and temperature have tobe known for controlling the boriding treatment. In this study,the effects of the boriding time and temperature on the growthkinetics of the boriding layer were investigated. Primarily,diffusion coefficient values were calculated using the Eq. (1)(Fig. 8).

Calculated diffusion coefficient values were given in theTable 2. The plot of LnD versus 1/T reveals a linear relationship(Fig. 9) and activation energy of 230 kJ/mol for 31CrMoV9 and270 kJ/mol for 34CrAlNi7 steels were measured from the slopeof the straight lines. The activation energy value for 31CrMoV9which is lower than that of 34CrAlNi7 steel, is an expectedresult because the boride layer thickness of 31CrMoV9 ishigher than that of 34CrAlNi7 steel. This indicates that borideson the surface of the 31CrMoV9 steel forms easier and growsquicker compared to 34CrAlNi7 steel.

In addition, growth rate and the growth accelerations of theborides were determined (Figs. 10 and 11). It was found thatthere is a good agreement among activation energy, growthrate and acceleration.

29M A T E R I A L S C H A R A C T E R I Z A T I O N 5 9 ( 2 0 0 8 ) 2 3 – 3 1

4. Conclusion

The following results may be drawn from the present study.

1. Borides formed on the steel surface have a columnar anddense morphology.

Fig. 7 –The variation of the fracture toughness as a function ofprocess time including contour diagrams for (A) 31CrMoV9and for (B) 34CrAlNi7 steels.

Fig. 8 –Square of the boride layer thickness of borided31CrMoV9 (A) and 34CrAlNi7 (B) steels as a function ofprocess time.

2. Thickness of boride layer formed on the surface of31CrMoV9 steel is higher than that of 34CrAlNi7 steel.

3. Borides formed on steel surface confirmed by X-raydiffraction analysis are mainly FeB and Fe2B.

4. The hardness of borides formed on surfaces of 31CrMoV9steel ranged from 1283 to 1672 HVN for 34CrAlNi7 steel1389–1600 HVN. Fracture toughness of borides formed onsurfaces of 31CrMoV9 steel ranged from 3.93 to 4.48 MPam1/2 and from 3.87 to 4.40 MPa m1/2 for 34CrAlNi7 steel.

5. Activation energy of 230 kJ/mol for 31CrMoV9 and 270 kJ/mol for 34CrAlNi7 steels was determined. It is possible toclaim that diffusivity of B atoms in 34CrAlNi7 is lower thanthat of 31CrMoV9 because of the alloying elements it has.

Table 2 – Diffusion coefficient values of 31CrMoV9 and34CrAlNi7 steels at different temperature

Boriding temperature Diffusion coefficient (cm2/sn)

(°C) 31CrMoV9 34CrAlNi7

850 2.5×10−9 2.45×10−9

900 6.9×10−9 3.83×10−9

950 17.4×10−9 11.28×10−9

Fig. 9 – lnD versus 1/T for (A) 31CrMoV9 and for (B) 34CrAlNi7steels.

30 M A T E R I A L S C H A R A C T E R I Z A T I O N 5 9 ( 2 0 0 8 ) 2 3 – 3 1

6. The growth acceleration of borides in 31CrMoV9 is higherthan that of 34CrAlNi7 steel and increases with increasingtemperature.

Acknowledgements

The authors thank Gebze High Technology Institute forperforming XRD and ASSAN Aluminum Company forperforming SEM studies. The authors thank the technicianErsan Demir for his assistance in performing experiments.

Fig. 10 –The variation of growth rate as a function of processtime including contour diagrams for (A) 31CrMoV9 and for (B)34CrAlNi7 steels.

R E F E R E N C E S

[1] Yu LG, Khor KA, Sundararajan G. Boriding of mild steelusing the spark plasma sintering (SPS) technique. SurfCoat Technol 2002;157:226–30.

[2] Hutching WF. Tribology: friction and wear of engineeringmaterials. London: Edward Arnold; 1992. p. 214–26.

[3] Ozbek I, Bindal C. Mechanical properties of boronized AISI W4steel. Surf Coat Technol 2002;154:14–20.

[4] Maragoudakis NE, Stergioudis G, Omar H, Pavlidou E, TsipasDN. Elsevier Mater Lett 2002;57:949–52.

[5] Usta M, Ozbek I, Ipek M, Bindal C, Ucisik AH. Thecharacterization of boride pure tungsten. Surf Coat Technol2005;194:330–4.

[6] Uslu I, Comert H, Ipek M, Celebi FG, Ozdemir O, Bindal C. Acomparison of borides formed on AISI 1040 and AISI P20steels. Mater Des 2006;80:1321–5.

[7] Sinha AK. Boriding (Boronizing). Heat treating, vol. 4. ASMHandbook; 1997. p. 437–47.

[8] Ingole S, Liang H, Usta M, Bindal C, Ucisik AH. Multi-scalewear of a boride coating on tungsten. Wear 2005;259:849–60.

[9] Erdemir A, Bindal C. Formation and self-lubricatingmechanisms of boric acid on borided steel surfaces. Surf CoatTechnol 1995;76–77:443–9.

Fig. 11 –The variation of growth acceleration of borides withtemperature.

31M A T E R I A L S C H A R A C T E R I Z A T I O N 5 9 ( 2 0 0 8 ) 2 3 – 3 1

[10] Celebi G, Ipek M, Bindal C, Ucisik AH. Some mechanicalproperties of borides formed on AISI 8620 steel. Mater Forum2005;29.

[11] Spence TW, Makhlouf MM. Characterization of the operativemechanism in potassium fluoborate activated pack boridingof steels. Mater Process Technol 2005;168:127–36.

[12] Yu LG, Chen XJ, Khor KA, Sundararajan G. FeB/Fe2B phasetransformation during SPS pack-boriding: boride layer growthkinetics. Acta Mater 2005;53:2361–8.

[13] Martini C, Palombarini G. Mechanism of thermochemicalgrowth of iron borides on iron. J Mater Sci 2004;39:933–4.

[14] Ucisik AH, Bindal C. Fracture toughness of boride formed onlow-alloy steels. Surf Coat Technol 1997;94–95:561–5.

[15] Yu LG, Khor KA, Sundarajan G. Boride layer growth kineticsduring boriding of molybdenum by the Spark Plasma Sinter-ing (SPS) technology. Surf Coat Technol 2006;201:2849–53.