Effect of shot peening on the residual stresses and microstructure of tungsten cemented carbide

    Effect of shot peening on the residual stresses and microstructure of tungsten cemented carbide Chengxi Wang, Chuanhai Jiang, Fei Cai, Yuantao Zhao, Kaiyuan Zhu, Ze Chai PII: DOI: Reference:

S0264-1275(16)30101-0 doi: 10.1016/j.matdes.2016.01.101 JMADE 1302

To appear in: Received date: Revised date: Accepted date:

12 July 2015 19 January 2016 21 January 2016

Please cite this article as: Chengxi Wang, Chuanhai Jiang, Fei Cai, Yuantao Zhao, Kaiyuan Zhu, Ze Chai, Effect of shot peening on the residual stresses and microstructure of tungsten cemented carbide, (2016), doi: 10.1016/j.matdes.2016.01.101

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT Effect of shot peening on the residual stresses and

T

microstructure of tungsten cemented carbide

IP

Chengxi Wang, Chuanhai Jiang*, Fei Cai, Yuantao Zhao, Kaiyuan Zhu, Ze Chai

SC R

School of Materials Science and Engineering, Shanghai Jiao Tong University, No. 800, Dongchuan Road, Shanghai 200240, PR China

NU

Abstract

MA

Shot peening is conducted on pretreated WC-10 wt.% Co composite. X-ray stress analyzer coupled with X-ray diffraction line profiles analysis is employed to determine residual stresses and microstructure of peened samples. Variations of

TE

D

morphology due to different treatments are detected by scanning electron microscope. The results show that the compressive residual stresses in WC and Co phase increase

CE P

by 48% and 70% respectively after SP. The surface topography and dislocation densities are improved substantially, while the domain size decreases dramatically.

AC

Compared with the unaffected region, it is observed that the microstrain becomes severe in the affected region, and the microhardness improves greatly and reaches its maximum in a nanocrystalline layer formed at the top surface of the specimen, representing it is not subjected to the inverse Hall-Petch effect. It is also noted that Hertzian effect induces a higher shear stress in the subsurface, which results in the inflection points revealed on the distribution of residual stresses and microstructure.

Keywords: Shot peening, Tungsten cemented carbide, Residual stress, Microstructure.

*Corresponding author. E-mail address: [email protected]

ACCEPTED MANUSCRIPT 1. Introduction WC-Co cemented carbides have been widely used to make tools for metal cutting and

T

rock drilling or wear parts for several decades. They exhibit excellent hardness,

IP

strength and wear resistance resulting from the combination of hard carbides and

SC R

ductile binder [1-4]. However, the inherent high hardness and low ductility always mean a limited toughness. And tensile stress also remains after cooling from the liquid

NU

phase or solid state sintering due to the different thermal expansion coefficients between the carbides and binder phase [5-8]. The limited fracture toughness and

MA

tensile residual stress are detrimental to the service life, especially in extreme environments [9]. It is well known that the mechanical properties of WC-Co

D

cemented carbides depend strongly on their microstructural parameters such as grain

TE

size and mean free path [10, 11]. For example, nanocrystalline WC-Co cemented

CE P

carbides often exhibit simultaneous improvements in hardness and toughness [12]. However, wear resistance, corrosion resistance and other properties are usually related to surface phenomena in general, and material failures occur on the surface in most

AC

cases. So the surface performance plays an important role in their extensive application.

Shot peening (SP) is one of the most popular surface treatments to enhance the service life of engineering components [13, 14]. It is a cold working process where highspeed shots impinge repeatedly on the work-pieces and induce the compressive residual stress (CRS) into the plastic deformed surface layers [15,16]. The CRS could effectively inhibit crack growth under both static and cyclic loading, improving the fatigue life and resistance to stress corrosion cracking (SCC) [17]. Moreover, SP can also significantly change the microstructure, such as refining the domain size. The

ACCEPTED MANUSCRIPT refined domain often improves the hardness and strength, as the ultra-fine microstructure increases the grain (sub-grain) boundaries and interfaces, resulting in a

T

more difficult travel through by the dislocations. Up to now, some obvious

IP

achievements have been reported about SP effect on the brittle ceramics. Pfeiffer and

SC R

Frey [18-20] have successfully enhanced Al2O3 and Si3N4 surface with high residual stresses using tailored SP parameters. Pratik P. Shukla and Jonathan Lawrence [21] published their new result about micro-shot peening of zirconia-advanced ceramics,

NU

which demonstrated that the surface roughness and fracture toughness increased in

MA

spite of the slight reduce of hardness. Tomaszewski et al. [22] also reported the increased hardness and refined grain size of alumina ceramics after SP. Nevertheless,

D

the investigation of SP on tungsten cemented carbide is insufficient. The present work

TE

attempts to shed light on the effect of SP treatment on the residual stresses and

CE P

microstructure of WC-10 wt.% Co composite and discuss the influencing mechanism.

2. Experimental

AC

The tungsten cemented carbide (TCC) composite used in this study was YG10 with nominal addition of 10 wt.% Co. Before SP, all specimens were pretreated with grinding and polishing and then cut into square shape with effective dimensions of 20 by 20 by 5 mm. SP was performed on a traditional air blasting machine, and ZrO2 ceramic beads with average diameter of 0.25 mm were employed as the medium. The nozzle diameter was 15 mm and the stand-off distance was 100 mm. The coverage of all specimens was 100% and other parameters were given as follows: jet pressure of 0.65 MPa, working time of 100 seconds, and Almen peening intensity of 0.23 mmA. X-ray Stress Analyzer (LXRD, Proto, Canada, Cu-Kα radiation, 30 KV, 25 mA, Ni filter) was employed to collect the residual stress information of surface layers after

ACCEPTED MANUSCRIPT SP by utilizing sin2ψ method. The shifts of WC (212) and Co (114) diffraction profiles were detected. XRD patterns of TCC were measured by Rigaku Ultima IV X-

T

ray diffractometer (Cu- Kα radiation, 40 KV, 30 mA), with a scan speed of 2°/min,

IP

and a step of 0.02°, to analyze the microstructures of shot peened surface layers.

SC R

Instrumental profile was obtained from the standard sample of WC powder annealed at 400°C for 4h. The surface topography and cross-section microstructure were observed on scanning electron microscope (SEM, JSM-7600F, JEOL, Japan) with the

NU

voltage of 15 kV. The hardness of TCC was measured on DHV-1000 Digital

MA

Microhardness Tester with a loading force of 9.8 N and a holding time of 15 s. Each layer was measured for five times and the average value was recorded as a final result.

D

In order to investigate the variations of residual stresses, microstructure and hardness

TE

along the depth, an electrochemical corrosion peeling method was performed instead of cross-section testing. Firstly, the whole surface was carefully wiped with ethanol to

CE P

diminish the effect of foreign impurities and adsorbate such as cracked beads from the SP process. Secondly, the thin top surface layers of TCC were etched uniformly step

AC

by step via iterative electrolytical method with saturated NaCl solution at room temperature. Finally, when a layer with a certain thickness was removed, the surface of sample was wiped again with ethanol prior to the measurement. Regarding X-ray diffraction profiles analysis, the Voigt method was employed to evaluate the microstructure. The measured profile h is always the convolution of physical profile (instrumental profile) g and structurally broadened profile f. In this study, the measured profile of annealed WC powder was applied as the physical profile g. The structurally broadened profile is composed of Cauchy and Gaussian components. The relationship of integral breadth β according to Voigt method is shown as follows [23]:

ACCEPTED MANUSCRIPT Gh 2  Gf 2  Gg 2 , Ch  Cf  Cg

(3)

T

where the subscript G and C denote the Gaussian and Cauchy components,

IP

respectively. The superscript h, f and g denote the measured line profile, the

SC R

structurally broadened profile and the physical profile, respectively [23]. It is assumed that the Cauchy component is only donated by domain size, while the Gaussian component is solely contributed by microstrain.  Gf and  Cf can be figured out from

NU

Eq. (3). Subsequently, the domain size D and microstrain ε can be obtained according

   cos( ) f C

,

Gf 4 tan( )

(4)

D

D

MA

to the following equations:

TE

where λ is the wavelength of incident X-ray and θ is the diffraction angle of each peak

CE P

at exact reflection position.

3. Results and discussion

AC

3.1. Surface morphology and residual stress distribution Surface morphologies of WC-10 wt.% Co both before and after SP are illustrated in Fig. 1 (a), and (b), respectively. The striking contrast demonstrates that surface topography has been dramatically changed after treatment. Before SP, obvious residual furrows and other defects resulted from the grinding can be observed in spite of its relatively smooth surface owing to the followed mechanical polishing. However, the furrows and other machined striation marks disappear after the fine-shots impinging on the coarse surface. In addition, no micro-crack is found under the

ACCEPTED MANUSCRIPT specific set-condition applied, which means the selected parameters are suitable and

MA

NU

SC R

IP

T

the surface topography is improved.

Fig. 1 SEM images of WC-10 wt.% Co surface before (a) and after (b) SP.

D

Residual stresses of WC and Co phase along the depth before and after SP are shown

TE

in Fig. 2 (a) and (b), respectively. The pretreatment of grinding and polishing induces

CE P

CRS in both WC and Co phases due to the predominant squeezing action and burnishing effect of mechanical impact. Generally, the pretreatment besides mechanical grinding and polishing has an influence on the surface properties such as

AC

hardness and indentation size effect [24, 25]. As can be seen from Fig. 2, before SP, the peak of CRS lays at the top surface where the maximum plastic deformation appears, with a value of -1063 MPa for WC and -451 MPa for Co. The lower CRS in Co phase is due to the different original stress states before grinding and polishing, that is, tensile stress in Co phase while a compressive one in WC phase. This can be attributed to their different thermal expansivities during the cooling process from liquid phase or solid state sintering. It can also been observed that CRS value decreases sharply from the top surface, and the depth of maximum CRS before SP was smaller. At around 40 µm from the top surface, CRS has already tended to be stable. After SP, the peak value emerged at the top surface as well, and they were -

ACCEPTED MANUSCRIPT 1576 and -768 MPa in WC and Co phase, respectively. SP increases both the CRS value and its maximum depth in two phases obviously. It is noted from these results

T

that inflection points exist in subsurface of both phases, with a depth of around 10 and

IP

30 µm before and after SP, respectively. Before SP, the plastic deformation at

SC R

inflection point is usually regarded as the contribution of Hertzian pressure generated by vertical force during the pretreatment process. The vertical force perpendicular to the surface, however, is much smaller than the tangential force parallel to the top

NU

surface, and the latter plays a dominant role in the whole process. Therefore, the

MA

inflection points are not conspicuous. During SP, shot media with a high speed and a small diameter give the work-piece a great vertical force, and Hertzian effect

AC

CE P

TE

D

generates maximum shear stress in subsurface other than the top surface [26].

Fig. 2. Distribution of residual stresses in WC (a) and Co (b) phase along the depth. The relationship between Hertzian pressure and the vertical pressure force is shown as follows [26]:

P0 

3 F  2 a 2

(1)

ACCEPTED MANUSCRIPT where P0 is the Hertzian pressure, F is the vertical pressure force and a is the half width of the connect zone. The distinct distance Zτ,max where shear stress has the

=0.47a

IP

τ,max

(2)

SC R

Z

T

maximum value below surface can be given as:

During the SP treatment, the vertical pressure F on the composite is far bigger than

NU

that during the pretreatment process. Therefore, enhanced Hertzian effect enlarges the residual stress in the composite after SP. As for materials with higher hardness such

MA

as tungsten cemented carbide, the half width of the contact zone would be much smaller under the same peening intensity. Then Hertzian effect would become more

D

predominant compared with that in softer materials. In fact, the distribution of

TE

residual stresses in the two phases shown in Fig. 1 is the resultant stresses of both the

CE P

pretreatment and SP. However, the maximum of CRS in WC and Co phases after SP are also improved by 48% and 70%, respectively. During the process of SP, the top surface layer intends to stretch as the shots impact on it, but is restrained by the

AC

subsurface layer because the upper layer suffers more elongation than the below one. Therefore, the top surface layer is in compression. A crack can hardly form or propagate in this case. Undoubtedly, the higher CRS induced in the surface layers by SP would also be beneficial to the resistance of fatigue and SCC. 3.2. Distribution of domain size and microstrain XRD patterns with indexing of the top surface of WC-10 wt.% Co composite before and after SP are shown in Fig. 3. It can be found that no new phase is generated after SP. In addition, the location, intensity and width of diffraction peaks are almost the same before and after SP. As the refined domain size always contribute to the

ACCEPTED MANUSCRIPT broadened diffraction peak, and it can be concluded that the domain size in the top surface does not change significantly. Only face centered cubic (FCC) phase of Co is

T

observed with weak diffraction intensity, while hexagonal close-packed (HCP)

IP

structure is not detected, which is likely due to its low content in this study. The

SC R

existence of Co phase (FCC) at room temperature indicates that high tungsten or carbon dissolution in the binder phase prohibits the crystal transformation from FCC

AC

CE P

TE

D

MA

NU

to HCP.

Fig. 3. X-ray patterns of the top surface of WC-10 wt.% Co composite before and after SP. In order to further investigate the influence of SP on the microstructure under the top surface, SEM images of cross-section before and after SP are taken and shown in Fig. 4.

SC R

IP

T

ACCEPTED MANUSCRIPT

NU

Fig. 4 Cross-section SEM images of WC-10 wt.% Co before (a) and after (b) SP

MA

treatment.

It can be seen that the grain (particle) sizes diminished sharply after SP compared

D

with those without SP treatment. Now that the specific shot impact can severely

TE

decrease the grain (particle) sizes at the deformation layer, it is reasonable to deduce that those microstructure (domain) with smaller sizes could also be diminished.

CE P

Unfortunately, SEM images cannot provide sharper resolution of the variation of much smaller microstructures. In fact, the grains along with those microstructures in

AC

smaller dimensions, which are usually called domain, involve various types of mistake-deformation, such as the sub-grain, dislocation cell, mean distance between dislocations or other defects at the diffraction direction. However, XRD patterns could provide valuable information about domain size at the deformation layer. Based on XRD line profiles, Voigt method [23] is employed to calculate the domain size and microstrain of WC phase. Before calculation, all diffraction profiles are treated at the same procedure. First, the Kα2 profiles are separated from Kα1 profiles and subtracted subsequently according to Pearson type VII function. Then the background is corrected. In order to improve the accuracy of calculation, WC (001), WC (100) and WC (101) reflections are all selected to calculate the domain size, microstrain and

ACCEPTED MANUSCRIPT dislocation density. The microstructure of Co phase is not calculated because its weak peak would increase the artificial errors greatly.

T

The domain size and microstrian distribution along the depth from the top surface are

IP

shown in Fig. 5. The variation trends of domain sizes calculated from all three

SC R

diffraction profiles are nearly the same and the values are located at the same level. This result shows a similarity with the trend of residual stresses along the depth. The

NU

top surface appears the minimum domain size which is less than 10 nm. Increasing with the depth, the variations of domain sizes also appear inflection points at about 30

MA

μm from the top surface. Around 90 μm, the domain size rises to be stable and the values calculated from WC (100), WC (001) and WC (101) are 196, 227 and 156 nm,

D

respectively. The difference in domain sizes between different crystal planes indicates

TE

the anisometric dimension at different directions. Due to the fact that the ultra-fine

CE P

microstructure has increased volume fraction of interfaces, nanocrystalline structure

AC

in the top surface can considerably improve the material strength and stiffness [27].

CE P

TE

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

Fig. 5. Domain size and microstrain along the depth: calculated from WC (100) (a),

AC

WC (001) (b), and WC (101) (c) diffraction profiles of WC-10 wt.% Co composite. With respect to microstrain, the maximum values corresponding to the most severe deformation in the top surface are 1.6×10-3, 4.6×10-3 and 1.6×10-3 for WC (100), WC (001) and WC (101), respectively. At the depth of around 30 μm, the microstrain presents the inflection points, and becomes stable at more than 80 μm. Because of the isotropy characteristic of polycrystalline WC-Co composite, there is not much difference among the domain size and microstrain calculated from WC (100), WC (001) and WC (101). The inflection points of domain size and microstrain correlate to that in residual stresses. This demonstrates that Hertzian pressure induced by SP can decrease domain size and increase microstrain in the subsurface as well.

ACCEPTED MANUSCRIPT 3.3. Distribution of dislocation densities In comparison to tensile testing, the elastic and plastic deformations at the surface

T

layers induced by SP is not serious. So the dislocation type is mostly the in-plane

IP

misfit dislocation herein. Based on the results of domain size and microstrain, the

SC R

distributions of dislocation densities in depth could be obtained by Williamson function [28], which is given as follows:

b



2

1

2

NU



2 3

D

(5)

MA

where ρ denotes the dislocation density,  2 denotes the weighted average of  2

D

after multiple measuring, and b is the mold of Burgers vector of WC. According to

TE

Reference [29], b as 0.29 nm is adopted to perform the calculation. In term of the

CE P

calculated domain size and microstrain, the corresponding dislocation densities along the depth are calculated and shown in Fig. 6. From the diffraction profiles of WC (100), WC (001) and WC (101), it could be found that the calculated dislocation

AC

density variations show a similar down trend. The dislocation densities plunge from the top surface in the first 20 µm, then swell the inflection points and finally decrease gradually until getting steady. The insets in Fig. 6 show the inflection points described herein.

AC

CE P

TE

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

Fig. 6. Depth distribution of dislocation densities calculated from WC (100) (a), WC (001) (b), and WC (101) (c) diffraction profiles of WC-10 wt.% Co composite. On the top surface, the dislocation densities calculated from WC (100), (001) and (101) planes reach 4.8×1015, 8.0×1015 and 5.7×1015 m-2, respectively. Before grinding, polishing and SP, the dislocation densities of original samples on the top surface maintain 2~4×1013 m-2, which have been increased almost two orders of magnitude after SP. The high dislocation densities are resulted mainly from two processes.

ACCEPTED MANUSCRIPT Firstly, the grinding and polishing treatments lead to the severe elastic-plastic deformation in the very thin upper layers. Secondly, the small shot beads with high

T

speed transmit high kinetic energy to the surface layers during SP and then induce

IP

further elastic-plastic deformation. During the grinding and polishing processes, Co

SC R

phase often offers little resistance to deformation because of the tensile stress inherited from the cooling procedure. At the beginning of deformation, composite compression mainly reduces the tensile stress in Co phase. Nevertheless, with the

NU

development of dislocation gliding, proliferating and tangling, the Co phase will be

MA

work-hardened, and its resistance to deformation will be enhanced. With the further progress of plastic deformation, the contact stress between Co grains and WC grains

D

will reach to a marginal value which exceeds the proof stress of WC. Then the WC

TE

grains begin to transfer from elastic deformation to plastic deformation. The same situation also occurs during the SP process, and the deformation becomes more severe

CE P

compared with pretreatment. Accordingly, the dislocation densities increase dramatically after these surface treatments.

AC

3.4. Distribution of microhardness Fig. 7 shows the microhardness distribution of peened composite along the depth. From the top surface to 20 μm, the microhardness decreases violently. Also striking, there is a rebounding of the microhardness variation at the depth of 25 μm. After that, it gradually descends to a stable value at 80 μm. The microhardnesss in the top surface, the kick point and the stable region are 1563, 1385 and 1327 HV, respectively. The inflection point is partly attributed to the CRS at that position. If the stable region is regarded as the un-peened matrix, a significant microhardness increase of 18% can be obtained in the top surface after SP.

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

Fig. 7. Depth distribution of microhardness of shot peened WC-10 wt.% Co

D

composite.

TE

The variation trend of microhardness is very similar to those of residual stresses,

CE P

domain sizes and dislocation densities. As SP is a work-hardening process which can refine the domain size and improve the dislocation density. Especially, the formation of nanocrystalline layer in the top surface with higher dislocation density will be

AC

helpful to the improvement of microhardness. As suggested by many reports, there is an inverse Hall-Petch response existing in ceramic materials while grain sizes are no more than 100 nm. However, the findings here do not comply with the inverse HallPetch, and they present a normal relationship between the grain size and hardness, which are in good agreement with other results [30]. The still applicable Hall-Petch relation to refined domain size at the deformation layer suggests that the deformation mechanism remains the same. The only detection of the FCC cobalt phase in this study, means there is no or little formation of HCP lamellae here. Combined with other reports [31-33] on deformation of cemented carbides, the

ACCEPTED MANUSCRIPT deformation mechanism can be accordingly predicted as follows: The FCC Co phase deforms initially among the carbide skeleton owing to its easily open slip system. The

T

binder phase then was strain hardened with the accumulation of dislocation motion,

IP

multiplication and tangling. This can be confirmed by the remarkable increase of

SC R

dislocation density calculated from X-ray diffraction profiles analysis. There is no or little phase transition from FCC to HCP occurring during this step. Along with the dislocation interactions, the work-hardened Co phase attains a certain critical value

NU

(which in general represents the proof stress of the neighboring WC phase at the

MA

specific grain size and crystal structure). This could enable the WC phase to deform at the micro-localized region, which subsequently leads to the deformation of the whole

D

carbide skeleton. It is noteworthy that the stress and strain states between two phases

CE P

4. Conclusions

TE

are dynamic (such as redistributing and cancelling) all through the SP process.

The residual stresses and microstructure of peened WC-10 wt.% Co were investigated.

AC

The SP treatment resulted in high CRS and deep deformation layers. The maximum CRS in WC phase and Co phase were improved by 48% and 70%, respectively. At the same time, the nanocrystalline layer formed at the top surface, and the minimum domain size decreased to less than 10 nm. Along the depth, the domain size appeared the minimum value at the top surface, and then gradually increased before it was stable. Microstrain, dislocation densities and microhardness showed the similar change trends. The most serious microstrain appeared at the top surface, and the maximum value was 4.6×10-3. The dislocation densities reached 4.8~8.0×1015 m-2 after SP, which increased two orders of magnitude compared with the pristine matrix. The variation of residual stresses and microstructure can be well explained in terms of

ACCEPTED MANUSCRIPT Hertzian pressure generated by the vertical force during the SP process. Furthermore, the micro-hardness was also improved by 18% on the top surface owing to the refined and

higher

dislocation

densities.

The

largest

microhardness

of

T

domains

IP

nanocrystalline showed the normal Hall-Petch effect.

SC R

References

[1] Anhai Li, Jun Zhao, Dong Wang, Xinliang Gao, Hongwei Tang, Three-point

NU

bending fatigue behavior of WC-Co cemented carbides, Mater. Des. 45 (2013) 271-

MA

278.

[2] L. Llanes, Y. Torres, M. Anglada, On the fatigue crack growth behavior of WC-

D

Co cemented carbides: kinetics description, microstructural effects and fatigue

TE

sensitivity, Acta Mater. 24 (2002) 2381-1393.

CE P

[3] Omayma A.M. El-Kady, Effect of nano-yttria addition on the properties of WC/Co composites, Mater. Des. 52 (2013) 481-486.

AC

[4] G. S Upadhyaya, Materials science of cemented carbides—an overview, Mater. Des. 22 (2001) 483-489. [5] Aaron Krawitz, Eric Drake, Residual stresses in cemented carbides-An overview, Int. J. Refract. Met. Hard Mater. 49 (2015) 27-35. [6] D.Mari, B. Clausen, M.A.M. Bourke, K. Buss, Measurement of residual thermal stress in WC-Co by neutron diffraction, Int. J. Refract. Met. Hard Mater. 27 (2009) 282-287.

ACCEPTED MANUSCRIPT [7] C. Larsson, M. Oden, X-ray diffraction determination of residual stresses in functionally graded WC-Co composites, Int. J. Refract. Met. Hard Mater. 22 (2004)

T

177-184.

SC R

Co composites, Mater. Sci. Eng., A. 359 (2003) 338-342.

IP

[8] D.L. Coats, A.D. Krawitz, Effect of particle size on thermal residual stress in WC-

[9] J.A.M. Ferreira, M.A. Pina Amaral, F.V. Antunes, J.D.M. Costa, A study on the

NU

mechanical behaviour of WC/Co hardmetals, Int. J. Refract. Met. Hard Mater. 27

MA

(2009) 1-8.

[10] Thomas Klunsner, Stefan Marsoner, Reinhold Ebner, Reinhard Pippan, Johannes

D

Glatzle, Arndt Puschel, Effect of microstructure on fatigue properties of WC-Co hard

TE

metals, Procedia Eng. 2010;2: 2001-2010.

CE P

[11] K. Jia, T.E. Fischer, B.Gallois, Microstructure, hardness and toughness of nanostructured and convertional WC-Co composites, Nanostruct. Mater. 10 (1998)

AC

875-891.

[12] A.S. Kurlov, A. Leenaers, S. van den Berghe, M. Scibetta, H. Schrottner, A.A. Rempel, Microstructure of nanocrystalline WC powders and WC-Co hard alloys, Rev. Adv. Mater. Sci 27 (2011) 165-172. [13] B.J. Foss, S. Gray, M.C. Hardy, S. Stekovic, D.S. McPhail, B.A. Shollock, Analysis of shot-peening and residual stress relaxation in the nickel-based superalloy RR1000, Acta Mater. 61 (2013) 2548-2559. [14] Pratik P. Shukla, Philip T. Swanson, Colin J. Page, Laser shock peening and mechanical SP processes applicable for the surface treatment of technical grade ceramics: A review, J. Eng. Manufac. 5 (2013) 639-652.

ACCEPTED MANUSCRIPT [15] D.J. Child, G.D. West, R.C. Thomson, Assessment of surface hardening effects from SP on a Ni-based alloy using electron backscatter diffraction techniques, Acta

T

Mater. 59 (2011) 4825-4834.

IP

[16] Lechun Xie, Chuanhai Jiang, Weijie Lu, KeZhan, Yanhua Chen, Investigation on

Mater. Sci. Eng., A. 528 (2011) 3423-3427.

SC R

the residual stress and microstructure of (TiB+TiC)/Ti-6Al-4V composite after SP,

NU

[17] Y.K. Gao, X.R. Wu, Experimental investigation and fatigue life prediction for 7475-T7351 aluminum alloy with and without SP-induced residual stresses, Acta

MA

Mater. 59 (2011) 3737-3747.

TE

France, 2005, P. 326-331.

D

[18] W. Pfeiffer, T. Frey, Advances in SP of silicon nitride ceramics, ICSP9, Paris,

CE P

[19] Wulf Pfeiffer, Johannes Wenzel, SP of brittle materials-status and outlook, Mater. Sci. Forum. 638 (2010) 799-804.

AC

[20] Wulf Pfeiffer, Tobias Frey, Strengthening of ceramics by SP, J. Eur. Ceram. Soc. 26 (2006) 2639-2645. [21] Pratik P. Shukla, Jonathan Lawrence, Micro-SP of ZrO2 advanced ceramics: An examination of surface integrity, J. Mater. Sci. 50 (2015) 1728-1739. [22] Tomaszewski H, Stalony-Dobrazanski F, Ostachowski P, Sztwi-ertnia K, Faryna M, Srutowanie ceramiki konstrukcyjnej, Ceram. Mater. 62 (2010) 266-271. [23] TH.H. De Keijser, J.I. Langford, E.J. Mittemeijer, A.B.P. Vogels, Use of the voigt function in a single-line method for the analysis of X-ray diffraction line broadening, J. Appl. Cryst. 15 (1982) 308-314.

ACCEPTED MANUSCRIPT [24] Joshua D. Gale, Ajit Achuthan, David J. Morrison, Indentation size effect (ISE) in copper subjected to severe plastic deformation (SPD), Metall. Mater. Trans. A. 45

T

(2014) 2487-2497.

IP

[25] George M. Pharr, Erik G. Herbert, and Yanfei Gao, The indentation size effect:

Annu. Rev. Mater. Res. 40 (2010) 271-292.

SC R

A critical examination of experimental observations and mechanistic interpretations,

NU

[26] H. Wohlfahrt, The influence of peening conditions on the resulting distribution of

MA

residual stress, ICSP2, Chicago, USA, 1984, P. 316-331. [27] Chang-Soo Kim, Ted R. Massa, Gregory S. Rohrer, Interface character

D

distributions in WC-Co composites, J. Am. Ceram. Soc. 3 (2008) 996-1001.

TE

[28] G.K. Williamson, R.E. Smallman, Dislocation densities in some annealed and

CE P

cold-worked metals from measurements on the X-ray debye-scherrer spectrum, Philos. Mag. 1 (1956) 34-36.

AC

[29] Yu.V. Milman, S. Luyckx, V.A. Goncharuck, J.T. Northrop, Mechanical properties in bending testing and mechanical behaviour of submicro and micro WCCo grades at elevated temperatures, 15th International Plansee Seminar, Plansee Holding AG, Reutte, 2 (2001) 75-90. [30] James A. Wollmershauser, Boris N. Feigelson, Edward P. Gorzkowski, Chase T. Ellis, Ramasis Goswami, Syed B. Qadri, et al., An extended hardness limit in bulk nanoceramics, Acta Mater. 69 (2014) 9-16. [31] V. K. Sarin and T. Johannesson, On the deformation of WC-Co cemented carbides, Met. Sci. 9 (1975) 472-476.

ACCEPTED MANUSCRIPT [32] J.J. Roa, E. Jimenez-Pique, J.M. Tarrago, M. Zivcec, C. Broeckmann, L. Llanes, Berkovich nanoindentation and deformation mechanisms in a hardmetal binder-like

T

cobalt alloy, Mater. Sci. Eng. A 621 (2015) 128-132.

IP

[33] J.M. Tarrago, J.J. Rosa, E. Jimenez-Pique, E. Keown, J. Fair, L. Llanes,

AC

CE P

TE

D

MA

NU

compression, Hard. Mater. 54 (2016) 70-74.

SC R

Mechanical deformation of WC-Co composite miscropillars under uniaxial

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

AC

CE P

TE

D

Graphical Abstract

ACCEPTED MANUSCRIPT Highlights:

AC

CE P

TE

D

MA

NU

SC R

IP

T

 Shot peening induces high compressive residual stress in WC phase and adjusts stress state in Co.  Domain size decreases and a nanocrystalline layer is formed near the top surface.  Increasing trends of dislocation densities, microstrain, microhardness are similar.