An Evaluation of a Borided Layer Formed on Ti-6Al-4V Alloy by Means of SMAT and Low-Temperature Boriding

materials

Article An Evaluation of a Borided Layer Formed on Ti-6Al-4V Alloy by Means of SMAT and Low-Temperature Boriding

Quantong Yao 1, Jian Sun 2,*, Yuzhu Fu 1, Weiping Tong 1,* and Hui Zhang 1 1 Key Laboratory of Electromagnetic Processing of Materials, Ministry of Education, Northeastern University, Shenyang 110819, Liaoning, China; [email protected] (Q.Y.); [email protected] (Y.F.); [email protected] (H.Z.) 2 Department of Materials Science and Engineering, Hefei University of Technology, Hefei 230009, Anhui, China * Correspondence: [email protected] (J.S.); [email protected] (W.T.); Tel.: +86-551-6290-4557 (J.S.); +86-24-8368-2376 (W.T.)

Academic Editor: Daolun Chen Received: 6 October 2016; Accepted: 25 November 2016; Published: 8 December 2016 Abstract: In this paper, a nanocrystalline surface layer without impurities was fabricated on Ti-6Al-4V alloy by means of surface mechanical attrition treatment (SMAT). The grain size in the nanocrystalline layer is about 10 nm and grain morphology displays a random crystallographic orientation distribution. Subsequently, the low-temperature boriding behaviors (at 600 ◦C) of the SMAT sample, including the phase composition, microstructure, micro-hardness, and brittleness, were investigated in comparison with those of coarse-grained sample borided at 1100 ◦C. The results showed that the boriding kinetics could be signiﬁcantly enhanced by SMAT, resulting in the formation of a nano-structured boride layers on Ti-6Al-4V alloy at lower temperature. Compared to the coarse-grained boriding sample, the SMAT boriding sample exhibits a similar hardness value, but improved surface toughness. The satisfactory surface toughness may be attributed to the boriding treatment that was carried out at lower temperature.

Keywords: Ti-6Al-4V alloy; SMAT; low-temperature boriding; hardness; toughness

1. Introduction Titanium and its alloys are widely used in artiﬁcial bone implants, aircraft manufacturing, and kitchenware due to their excellent chemical and physical properties, such as high strength-to-weight ratio, good biocompatibility, and low corrosion/oxidation rate [1–4]. However, the poor wear resistance extensively limits their further applications [5,6]. Among various surface treatments, boriding is a kind of thermo-chemical treatment which can generate a hard borided layer on the surface of titanium alloys. Although the borided layer presents low thickness, it effectively enables the improvement of the tribological performance of titanium and its alloys [6]. Nevertheless, because of the low boron atomic diffusivity, the boriding process is conventionally performed at 1000 ◦C ± 100 ◦C for a long duration time, which may induce the formation of a large amount of porosity in the borided layer, resulting in the deterioration of surface toughness. The borided layer, with high brittleness, may bring some negative effects for the further application of titanium materials. Therefore, reducing the boriding temperature of titanium alloys has become a hot topic in the thermo-chemical treatment ﬁeld. It is well known that the nano-grained materials are characterized by a large number of nanocrystalline boundaries and dislocation substructures, which can act as fast atomic diffusion pathways [7]. Therefore, inducing a nanocrystalline layer on the surface of metallic materials seems to be an effective method to enhance the atomic diffusion kinetics, leading to a reduction in the

Materials 2016, 9, 993; doi:10.3390/ma9120993 www.mdpi.com/journal/materials Materials 2016, 9, 993 2 of 10 thermo-chemical treatment temperature. In past decades, surface mechanical attrition treatment (SMAT) has been an effective method that enables coarse grains to reﬁne into nanoscale grains at the surface of various metallic materials [8–11]. Previously, studies have shown that the nitriding, chromizing, and boriding of various metallic materials could be performed at lower temperature with the assistance of SMAT [12–15]. For example, T. Balusamy showed that SMATed EN8 steel can be borided with a reasonable case depth at 650 ◦C for 7 h [16]. Xu et al. have shown that boriding of SMATed H13 steel could be achieved at 650 ◦C for 8 h [17]. In this paper, we employed one of the most widely used Ti-6Al-4V alloys to study the low-temperature pack boriding behavior (600 ◦C) by assisting with a nanocrystalline layer. The phase composition, microstructure, and micro-hardness of the SMAT borided sample were studied by using X-ray diffraction (XRD), scanning electron microscopy (SEM), and transmission electron microscopy (TEM). In particular, the brittleness of the surface layer on the borided SMAT sample was investigated in comparison with that of coarse-grained sample borided at 1100 ◦C.

2. Results and Discussion

2.1. Microstructure Characterizations of the SMAT Sample Figure1 shows the microstructure characterizations of SMAT Ti-6Al-4V alloy by using SEM, XRD and TEM. It can be seen that an obvious deformation layer of 15 µm thickness was distinguished from the matrix, as shown in Figure1a. The grain boundaries and microstructure are no longer clearly identified by SEM observation in the deformation layer. Figure1b shows XRD patterns of the Ti-6Al-4V alloy before and after SMAT. An evident broadening of the Bragg reflections was found on the SMAT sample by comparing the full width at half maximum (FWHM), which may be attributed to the grain refinement and micro-strain development. Additionally, the broadening rate of Bragg reflections of α-Ti is more evident than that of β-Ti. Generally, dual-phase structural Ti-6Al-4V alloy coexists in the hexagonal-closed-packed (HCP) α-Ti structure and face-centred-cubic (FCC) β-Ti structure. Since the stacking fault energy is different between the α-Ti (HCP) and the β-Ti (FCC), the plastic deformation more easily occurs in the lower stacking fault energy structure of α-Ti (HCP) [18]. The similar results can be also observed by surface nanocrystallization process of dual/multi-phase structural alloy in other literatures [19–21]. There is another variation that should be paid attention to: the solid-state transformation (α-Ti→β-Ti) induced by SMAT. This can be attributed to the increase in strain energy, which may account for the earlier-stated structural instability due to the Gibbs–Thompson effect [22]. Thus, this structural instability due to grain size reduction and strain may ultimately cause a polymorphic change from α-Ti (HCP)→β-Ti (FCC), which has a lower ∆G or higher structural stability [23]. Further TEM observation and corresponding selected area electron diffraction (SAED) illustrated that the microstructure of the uppermost-treated surface layer is characterized by ultrafine, equiaxed grains with random crystallographic orientations, as shown in Figure1c,d. The average grain size in the top surface layer is approximately 10 nm. The surface chemical composition of the SMAT sample was also detected by EDS in Figure2. No oxides and contaminants could be detected in the SMAT sample, owing to the fact that we used the Ti-6Al-4V alloy as the container and impacting balls during the SMAT process. Materials 2016, 9, 993 3 of 10 Materials 2016, 9, 993 3 of 10

Materials 2016, 9, 993 3 of 10

Figure 1. The microstructure characterizations of the SMAT Ti-6Al-4V alloy: (a) SEM micrographs; Figure 1. The microstructure characterizations of the SMAT Ti-6Al-4V alloy: (a) SEM micrographs; (b)Figure XRD patterns;1. The microstructure (c) TEM image characterizations (dark field); (d of) SAED the SMAT pattern. Ti-6Al-4V alloy: (a) SEM micrographs; (b) XRD patterns; (c) TEM image (dark ﬁeld); (d) SAED pattern. (b) XRD patterns; (c) TEM image (dark field); (d) SAED pattern.

Figure 2. The EDS analysis of the sample surface after 120 min SMAT. FigureFigure 2. 2.The The EDS EDSanalysis analysis ofof the sample surface surface after after 120 120 min min SMAT. SMAT.

2.2.2.2. Thermal Thermal Stability Stability of of SMAT SMAT Ti-6Al-4V Ti-6Al-4V 2.2. Thermal Stability of SMAT Ti-6Al-4V In Inorder order to to obtain obtain a asuitable suitable pack pack boridingboriding temperature,erature, the the thermal thermal stability stability of of the the nano-grain nano-grain was investigated by isothermal anneal for 5 h at various temperatures in a vacuum furnace. The wasIn investigated order to obtain by aisothermal suitable pack anneal boriding for 5 temperature,h at various temperatures the thermal stability in a vacuum of the nano-grainfurnace. The was effect of the annealing temperature on grain size was studied by TEM. As shown in Figure 3a,b, the investigatedeffect of the by annealing isothermal temperature anneal for on 5 hgrain at various size was temperatures studied by TEM. in a vacuumAs shown furnace. in FigureThe 3a,b, effect the of nano-grain is still at the nano-scale (<100 nm) when the annealing temperature is at 550 °C and 600 thenano-grain annealing is temperature still at the nano-scale on grain size (<100 was nm) studied when by the TEM. annealing As shown temperature in Figure is3 ata,b, 550 the °C nano-grain and 600 °C. With the annealing temperature increased to 650 °C, a large number of sub-micron◦ scale◦ grains is still°C. With at the the nano-scale annealing (<100 temperature nm) when increased the annealing to 650 °C, temperature a large number is at 550of sub-micronC and 600 scaleC. Withgrains the can be found on the surface layer, as shown in Figure 3c. Additionally, the grain size is about annealing temperature increased to 650 ◦C, a large number of sub-micron scale grains can be found on can100~300 be found nm. Meanwhile,on the surface corresponding layer, as shownSAED patterns in Figure present 3c. Additionally,a discontinuous the circular grain distribution, size is about the100~300 surfaceindicating nm. layer, Meanwhile,that as grain shown size corresponding in no Figure longer3c. Additionally,belongs SAED patternsto the thenanoscale. present grain sizea Whendiscontinuous is about the annealing 100~300 circular nm.temperature distribution, Meanwhile, correspondingindicating that SAED grain patterns size no presentlonger belongs a discontinuous to the nanoscale. circular When distribution, the annealing indicating temperature that grain

Materials 2016, 9, 993 4 of 10

◦ size noMaterials longer 2016 belongs, 9, 993 to the nanoscale. When the annealing temperature increases to 7004 of 10C, the grain size is about 300~500 nm, as shown in Figure3d. According to the above TEM observations, the temperatureincreases to 700 of obvious°C, the grain abnormal size is grainabout 300~5 growth00 nm, could as beshown conﬁrmed in Figure at 3d. a rangeAccording of 600~650 to the ◦C. above TEM observations, the temperature of obvious abnormal grain growth could be confirmed at Considering that the boron atoms’ diffusion in the nanocrystalline layer are very fast, which may a range of 600~650 °C. Considering that the boron atoms’ diffusion in the nanocrystalline layer are hinder the growth of nanoscale grains to some extent, a boriding temperature of 600 ◦C (5 h) should be very fast, which may hinder the growth of nanoscale grains to some extent, a boriding temperature chosenof 600 in the °C following(5 h) should boriding be chosen experiment. in the following boriding experiment.

Figure 3. TEM images and corresponding SAED patterns of 120 min SMAT sample annealed for 5 h Figure 3. TEM images and corresponding SAED patterns of 120 min SMAT sample annealed for 5 h at at different temperatures: (a) 550 °C; (b) 600 °C; (c) 650 °C; and (d) 700 °C. different temperatures: (a) 550 ◦C; (b) 600 ◦C; (c) 650 ◦C; and (d) 700 ◦C. 2.3. Microstructure Characterization of the Borided Layer 2.3. Microstructure Characterization of the Borided Layer Figure 4 shows the cross-sectional SEM micrographs of a SMAT sample borided at 600 °C for 5 Figureh and a4 coarse-grainedshows the cross-sectional sample borided SEM at 1100 micrographs °C for 5 h. Borided of a SMAT layers sample in the boridedSMAT sample at 600 and◦C for 5 h andcoarse-grained a coarse-grained sample samplepresent boridedcompletely at 1100different◦C formorphology. 5 h. Borided A gray layers borided in the layer SMAT with sample a and coarse-grainedthickness of 15 μ samplem was found present on completelythe surface of different the borided morphology. SMAT sample, A gray which borided is obviously layer with thicker than that formed on the borided coarse-grained sample. Additionally, the borided layer of a thickness of 15 µm was found on the surface of the borided SMAT sample, which is obviously the coarse-grained sample presents a tooth-shaped morphology which extends into the sample thicker than that formed on the borided coarse-grained sample. Additionally, the borided layer of the interior. Similar tooth-shaped crystal whiskers cannot be found in the borided SMAT sample. In coarse-grainedaddition, the sample borided presents layer of athe tooth-shaped SMAT sample morphology is discontinuous, which indicating extends intothat the diffusion sample interior.of Similarboron tooth-shaped atoms in the crystal surface whiskers of the SMAT cannot sample be foundis mostly in thealong borided the nanocrystalline SMAT sample. boundaries In addition, and the boridedother layer crystallographic of the SMAT defects. sample Figure is discontinuous, 5 shows XRD indicatingpatterns of thatthe borided the diffusion layer of of the boron two kinds atoms of in the surfacesamples. of the SMATThe XRD sample patterns is mostly provide along phase the nanocrystallineinformations of boundariesabout 20 μm and thickness other crystallographic from the defects.outermost Figure surface5 shows to XRDthe interior. patterns The ofBragg the borideddiffraction layer peaks of of the the two borided kinds SMAT of samples. sample present The XRD patternsobvious provide broadening phase after informations boriding treatment, of about indicati 20 µmng thickness that the borides from theare in outermost the nano-scale. surface This to the interior.result The can Bragg be attributed diffraction to peaksthe nanocrystalline of the borided remaining SMAT sample in the presentnano-scale obvious without broadening obviously after increasing at 600 °C for 5 h. Therefore, the borides of the SMAT sample could be obtained in the boriding treatment, indicating that the borides are in the nano-scale. This result can be attributed to the nano-scale in the following boriding process. TEM images and their corresponding SAED patterns nanocrystalline remaining in the nano-scale without obviously increasing at 600 ◦C for 5 h. Therefore, of the borided layer of the SMAT sample are presented in Figure 6. It can be seen that the borided the borideslayer consists of the SMATof nano-scaled sample bori coulddes be(20~30 obtained nm). Additionally, in the nano-scale the borides in the are following composed boriding of Ti, TiB, process. TEMTiB images2, and and Ti3B their4 by measuring corresponding the SAED SAED patterns. patterns These of the results borided are in layer agreement of the SMATwith the sample XRD are presentedanalysis. in FigureThe Ti63B.4 It is can detected be seen on thatthe surface the borided of the layerSMAT consists sample,of but nano-scaled it cannot be borides detected (20~30 on the nm). Additionally,coarse-grained the borides sample. areThis composed result can be of attributed Ti, TiB, TiB to2 the, and Ti3 TiB4 3thatB4 by was measuring decomposing the into SAED TiB patterns.and TheseTiB results2 at a arehigh in boriding agreement temperature with the of XRD 1100 analysis. °C. Additionally, The Ti3B oxides4 is detected are not ondetected the surface on the of surface the SMAT sample,of the but SMAT it cannot sample be and detected the coarse-grained on the coarse-grained sample from sample.XRD and ThisSAED result analysis. can be attributed to the ◦ Ti3B4 that was decomposing into TiB and TiB2 at a high boriding temperature of 1100 C. Additionally, oxides are not detected on the surface of the SMAT sample and the coarse-grained sample from XRD and SAED analysis. Materials 2016, 9, 993 5 of 10 MaterialsMaterials 2016 2016, 9,, 9939, 993 5 of5 10 of 10

Materials 2016, 9, 993 5 of 10

Figure 4. Cross-sectional SEM micrographs of the borided layer of (a) the SMAT sample and (b) the Figure 4. Cross-sectional SEM micrographs of the borided layer of (a) the SMAT sample and (b) the Figurecoarse-grained 4. Cross-sectional sample. SEM micrographs of the borided layer of (a) the SMAT sample and (b) the coarse-grainedcoarse-grained sample. sample. Figure 4. Cross-sectional SEM micrographs of the borided layer of (a) the SMAT sample and (b) the coarse-grained sample.

Figure 5. XRD patterns of the borided layer of the SMAT sample and coarse-grained sample.

FigureFigure 5. 5.XRD XRD patterns patterns of of the theborided borided layerlayer of the SMAT SMAT sample sample and and coarse-grained coarse-grained sample. sample. Figure 5. XRD patterns of the borided layer of the SMAT sample and coarse-grained sample.

Figure 6. TEM images and corresponding SAED patterns of the borided layer of the 120 min SMAT sample: (a) light field; (b) dark field; (c) electron diffraction patterns.

Figure 6. TEM images and corresponding SAED patterns of the borided layer of the 120 min SMAT FigureFigure 6. 6.TEM TEM images images and and corresponding corresponding SAEDSAED patterns of of the the borided borided layer layer of of the the 120 120 min min SMAT SMAT sample: (a) light field; (b) dark field; (c) electron diffraction patterns. sample:sample: (a )(a light) light ﬁeld; field; (b ()b dark) dark ﬁeld; field; ( c(c)) electronelectron diffractiondiffraction patterns. patterns.

MaterialsMaterials 20162016,, 99,, 993 993 66 of of 10 10

2.4. Hardness of the Nitrided Layer 2.4. Hardness of the Nitrided Layer In Figure 7, the micro-hardness of various samples along the cross-sectional direction from the outermostIn Figure surface7, the to micro-hardness the interior was of measured various samples by a micro-hardness along the cross-sectional tester. The directionmicro-hardness from the of theoutermost coarse-grained surface sample to the interioris 340 HV was by measured repeated measurement. by a micro-hardness The hardness tester. of The the micro-hardness SMAT sample isof 550 the HV, coarse-grained which is 1.6 sample times higher is 340 HVthan by that repeated of the matrix. measurement. The hardness The hardness of the SMAT of the sample SMAT furthersample increased is 550 HV, to which 1210 isHV 1.6 following times higher the thanboridi thatng treatment. of the matrix. Such The a high hardness hardness of the value SMAT is generallysample further achieved increased at a higher to 1210 temperature HV following for the a longer boriding duration treatment. time Such in conventional a high hardness boriding value treatmentsis generally [24]. achieved Since the at aimproved higher temperature hardness is forma ainly longer caused duration by boride time formation in conventional and boron boriding atom diffusion,treatments the [24 thickness]. Since the of improvedthe hardening hardness layer is can mainly be considered caused by borideas the depth formation of the and boron boron atoms’ atom diffusion.diffusion, As the shown thickness in the of theSMAT+Boriding hardening layer curve, can the be consideredhardening layer as the concentrates depth of the within boron 20 atoms’ μm whichdiffusion. is the As range shown of inthe the nanocrystalline SMAT+Boriding layer. curve, This the means hardening that boron layer atom concentrates diffusion within was mainly 20 µm alongwhich nanocrystalline is the range of the boundaries. nanocrystalline The layer.boron This atoms means cannot that borondiffuse atom into diffusion the crystal was lattice mainly of along the matrixnanocrystalline below the boundaries. nanocrystalline The boron layer atoms at cannota low diffuse temperature into the crystalof 600 lattice °C. As of theshown matrix in below the ◦ coarse-grained+Boridingthe nanocrystalline layer atcurve, a low the temperature borided layer of 600 of C.coarse-grained As shown in thesample coarse-grained+Boriding presents a striking contrastcurve, the with borided that of layerthe SMAT of coarse-grained sample. Although sample the presentscoarse-grained a striking sample contrast is borided with at that the of high the ◦ temperatureSMAT sample. of 1100 Although °C, the the effective coarse-grained thickness sample of the hardening is borided layer at the is high obviously temperature thinner of than1100 thatC, ofthe the effective coarse-grainedthickness sample. of the hardening The hardness layer of is obviouslythe outermost thinner surface than achi thateved of the 1520 coarse-grained HV, but it rapidlysample. decreasesThe hardness within of 20the μ outermostm. This result surface can be achieved attributed 1520 to HV, the butlow itdiffusivity rapidly decreases of boron withinatoms in20 theµm. Ti-6Al-4VThis result alloy. can Therefore, be attributed the toboriding the low kinetics diffusivity are effectively of boron atomsenhanced in theby Ti-6Al-4Vnanocrystalline alloy. layerTherefore, assistance. the boriding kinetics are effectively enhanced by nanocrystalline layer assistance.

FigureFigure 7. HardnessHardness variations variations along along the the depth depth of of both both the the SMAT SMAT boriding boriding sample sample and and the coarse-grainedcoarse-grained boriding boriding sample. sample.

2.5.2.5. Toughness Toughness of the Borided Layer InIn order order to to evaluate the the toughness of of the the bo boridedrided layer, the sharp indentation measurement methodmethod was utilized. TheThe sharpsharp indentersindenters (conic (conic or or pyramidal) pyramidal) are are normally normally used used for for the the analysis analysis of theof thetoughness toughness in opaque in opaque ceramic ceramic materials materials because becaus the contacte the pressurecontact ispressure independent is independent of the indentation of the indentationsize, and failure size, propagates and failure from propagates the corners from of the the residual corners impression of the residual [25]. The impression material toughness [25]. The is materialestablished toughness by three is experimental established by parameters: three experime the boridedntal parameters: layer thickness, the borided the indentation layer thickness, impression the indentationsize, and the impression applied load size, [26 and]. Before the applied the measurement, load [26]. Before the surfacesthe measurement, of the various the surfaces samples of were the variousultrasonically samples cleaned were ultrasonically in acetone for cleaned 15 min toin removeacetone the for contaminants15 min to remove attached the contaminants to the sample attached surfaces. toThe the cleaned sample surfaces surfaces. were The immediately cleaned surfaces subjected were to immediately the Vickers hardnesssubjected testto the using Vickers a load hardness of 30 kg. testGenerally, using therea load are of two 30 typeskg. Generally, of brittle crackthere modesare two from types Vickers of brittle indentations: crack modes Palmqvist from crackingVickers indentations:mode and radial-median Palmqvist cracking cracking mode mode. and After radial the-median measurement, cracking markedlymode. After different the measurement, indentation markedlymorphologies different of various indentation samples morphologies are presented of vari in Figureous samples8. It seems are presented that the in Palmqvist Figure 8. cracking It seems that the Palmqvist cracking mode is the main crack type in sample surfaces after boriding

Materials 2016, 9, 993 7 of 10 Materials 2016, 9, 993 7 of 10 modetreatment. is the The main main crack difference type in sampleis that no surfaces obvious after Palmqvist boriding cracking treatment. can The be mainidentified difference from Figure is that no8a. obviousThis result Palmqvist means crackingthat the surface can be identiﬁedtoughness from of the Figure SMAT8a. samples This result has means no obviously that the increase surface toughnessafter boriding of the treatment. SMAT samples However, has no in obviously sharp contrast increase with after the boriding SMAT treatment. sample, the However, coarse-grained in sharp contrastsample within Figure the SMAT 8b presents sample, themarked coarse-grained Palmqvist sample cracking in Figure after 8theb presents boriding marked treatment. Palmqvist The crackingnumerous after Palmqvist the boriding cracking treatment. originates The numerousfrom the Palmqvistangle position cracking of originatesthe pyramidal from the Vickers angle positionindentation of the and pyramidal extends Vickersto the indentationsurroundings. and Thus extends, it tocan the be surroundings. seen that the Thus, boriding it can process be seen thatdecreases the boriding the toughness process decreases of the Ti-6Al-4V the toughness samp ofle, the whereas Ti-6Al-4V the sample, SMAT whereas sample the maintains SMAT sample the maintainstoughness theof the toughness Ti-6Al-4V of thesample Ti-6Al-4V by nanocrystalline sample by nanocrystalline layer assistance. layer assistance.

Figure 8. SEM micrographs of the Vickers indentation of both (a) the SMAT boriding sample and (b) Figure 8. SEM micrographs of the Vickers indentation of both (a) the SMAT boriding sample and (b) the coarse-grained boriding sample. the coarse-grained boriding sample.

3. Materials and Methods 3. Materials and Methods Ti-6Al-4V cylinders with 49 mm in diameter, having a chemical composition of (in wt.%) Ti-6Al-4V cylinders with 49 mm in diameter, having a chemical composition of (in wt.%) 5.5~6.75, Al; 3.5~4.5, V; 0.3, Fe; 0.08, C; 0.05, N; 0.015, H; 0.2, O; and the remaining balance, Ti, were 5.5~6.75, Al; 3.5~4.5, V; 0.3, Fe; 0.08, C; 0.05, N; 0.015, H; 0.2, O; and the remaining balance, Ti, were used in the present study. Firstly, the Ti-6Al-4V cylinders were annealed at 600 °C for 10 h to relieve used in the present study. Firstly, the Ti-6Al-4V cylinders were annealed at 600 ◦C for 10 h to relieve the the residual stress. Secondly, the Ti-6Al-4V cylinders were cut into sheets with a thickness of 3 mm residual stress. Secondly, the Ti-6Al-4V cylinders were cut into sheets with a thickness of 3 mm by wire by wire cut electrical discharge machining (WEDM). Thirdly, Ti-6Al-4V sheets were polished by 60 cut electrical discharge machining (WEDM). Thirdly, Ti-6Al-4V sheets were polished by 60 #~1000 # #~1000 # abrasive papers to remove the WEDM marks. Finally, Ti-6Al-4V sheets were ultrasonically abrasive papers to remove the WEDM marks. Finally, Ti-6Al-4V sheets were ultrasonically cleaned in cleaned in acetone for 10 min. acetone for 10 min. The SMAT process was performed by using a SPEX/8000M mill (SPEX® SamplePrep, The SMAT process was performed by using a SPEX/8000M mill (SPEX® SamplePrep, Metuchen, Metuchen, NJ, USA) which is shown in Figure 9. The Spex mills are widely used tools for NJ, USA) which is shown in Figure9. The Spex mills are widely used tools for synthesizing synthesizing nanocomposites. Since they have the ability to provide vibration at high energy, they nanocomposites. Since they have the ability to provide vibration at high energy, they have an have an obvious potential to input sustained stress onto a sample surface, which induces the grains obvious potential to input sustained stress onto a sample surface, which induces the grains to reﬁne to refine to the nano-scale. There are a great number of studies that describe the Spex mill used in to the nano-scale. There are a great number of studies that describe the Spex mill used in surface surface nanocrystallization process [7,12,27]. In order to prevent the incorporation of iron pollution nanocrystallization process [7,12,27]. In order to prevent the incorporation of iron pollution products products into the sample surface, the cylindrical container of the Spex mill was made from into the sample surface, the cylindrical container of the Spex mill was made from Ti-6Al-4V alloy. Ti-6Al-4V alloy. The cleaned Ti-6Al-4V alloy sheet was held in place via mechanical locking at one The cleaned Ti-6Al-4V alloy sheet was held in place via mechanical locking at one end of the cylindrical end of the cylindrical container of the Spex mill, and 20 Ti-6Al-4V alloy balls 10 mm in diameter container of the Spex mill, and 20 Ti-6Al-4V alloy balls 10 mm in diameter were used to provide the were used to provide the desired impact on the surface of the Ti-6Al-4V alloy sheet. The impact desired impact on the surface of the Ti-6Al-4V alloy sheet. The impact veloctiy of Ti-6Al-4V balls veloctiy of Ti-6Al-4V balls induced by shaking the cylindrical container of the Spex mill was 15 m/s. induced by shaking the cylindrical container of the Spex mill was 15 m/s. In order to prevent the In order to prevent the sample from oxidizing, the SMAT process was conducted under an argon sample from oxidizing, the SMAT process was conducted under an argon atomsphere. atomsphere.

Materials 2016, 9, 993 8 of 10 Materials 2016, 9, 993 8 of 10

Figure 9. A schematic view of the Spex mill. Figure 9. A schematic view of the Spex mill.

The SMAT sample was ultrasonically cleaned in acetone for 5 min, then immediately subjected to seal byThe a SMATstainless sample steel was container ultrasonically for the cleanedpack boriding in acetone process for 5 min,with then B4C immediately powder media subjected (99.9 to wt.%,seal 320 by mesh). a stainless Additionally, steel container the boriding for the packprocess boriding was performed process with in a B4Cmuffle powder furnace media at 600(99.9 °C for wt.%, ◦ 5 h.320 For mesh). comparison,Additionally, the coarse-grained the boriding sample process was was pack performed borided in at a1100 muffle °C for furnace 5 h. After at 600 boridingC for 5 h. ◦ treatment,For comparison, the boridingthe container coarse-grained cooled sample down in was the packfurnace. borided at 1100 C for 5 h. After boriding treatment,The grain the size, boriding micro-strain, container and cooled phase down informat in theion furnace. of nanocrystalline layer and borided layer were obtainedThe grain with size, X’ Pert micro-strain, Pro PW3040/60 and phase X-ray information diffractometer of nanocrystalline (PANalytical, Lelyweg, layer and EA borided Almelo, layer Thewere Netherlands), obtained with (XRD) X’ Pertusing Pro Cu PW3040/60 Kα radiation X-ray (40 diffractometerkV, 40 mA). Small (PANalytical, angular steps Lelyweg, of 2θ EA = Almelo,0.03° α θ ◦ wereThe taken Netherlands), to measure (XRD) the intensity using Cu of K eachradiation Bragg (40 diffraction kV, 40 mA). peak. Small The angular grain size steps and of 2micro-strain= 0.03 were weretaken derived to measure from the the breadth intensity at ofhalf each maximum Bragg diffraction intensity peak.of measured The grain Bragg size diffraction and micro-strain peaks by were usingderived the Scherrer-Wilson from the breadth equation at half maximum [12]. Cross-sectional intensity of measured morphologies Bragg diffractionof various peaks samples by using were the observedScherrer-Wilson by using a equation Zeiss Ultra [12]. 55 Cross-sectional scanning electron morphologies microscope of (Zeiss, various Jena, samples Freistaat were Thüringen, observed by Germany)using a Zeiss(SEM). Ultra Phase 55 scanning information electron of microscope outermost (Zeiss, surface Jena, layer Freistaat of Thüringen,borided samples Germany) were (SEM). characterizedPhase information by a TECNAI of outermost G20 transmission surface layer electron of borided microscope samples were(FEI,characterized Hillsboro, OR, by USA) a TECNAI (TEM) G20 fromtransmission their corresponding electron microscope selected area (FEI, electron Hillsboro, diffraction OR, USA) (SAED) (TEM) patterns. from their The corresponding TEM samples selectedwere groundarea electronand mechanically diffraction polished, (SAED) patterns.followed Theby ion TEM thinning samples at werea lower ground temperature. and mechanically polished, followedThe micro-hardness by ion thinning values at a loweralong temperature.the surface to interior were measured with a Wolpert L101 MVD VickersThe micro-hardness micro-hardness values tester along (Buehler, the surface Lake to interiorBluff, IL, were USA) measured with witha load a Wolpertof 25 g L101 and MVD a durationVickers time micro-hardness of 10 s. Micro- testerhardness (Buehler, values Lake Bluff,were IL,calculated USA) with by a the load measured of 25 g and geometry a duration of timethe of Vickers10 s. Micro-hardnesspyramid import. values The weresurface calculated toughness by the was measured evaluated geometry by a ofWolpert the Vickers 450 SVA pyramid Vickers import. macro-hardnessThe surface toughness tester (Buehler, was evaluated Lake Bluff, by aIL, Wolpert USA) with 450 SVAa load Vickers of 30 kg macro-hardness and duration time tester of (Buehler, 10 s. TheLake pyramidal Bluff, IL, indentation USA) with was a load characteri of 30 kgzed and by durationusing the time Zeiss of Ultra 10 s. 55 The SEM. pyramidal indentation was characterized by using the Zeiss Ultra 55 SEM. 4. Conclusions 4. Conclusions A surface nanocrystalline layer with a thickness of about 15 μm was fabricated on the surface A surface nanocrystalline layer with a thickness of about 15 µm was fabricated on the surface of Ti-6Al-4V alloy after 120 min SMAT. The average grain size of the outermost surface was refined of Ti-6Al-4V alloy after 120 min SMAT. The average grain size of the outermost surface was refined to about 10 nm with a random crystallographic orientation. The thermal stability of nanocrystalline to about 10 nm with a random crystallographic orientation. The thermal stability of nanocrystalline could be maintained below 650 °C. could be maintained below 650 ◦C. The low-temperature boriding kinetics of the SMAT sample was significantly enhanced by The low-temperature boriding kinetics of the SMAT sample was significantly enhanced by nanocrystalline layer assistance. The borided layer of the SMAT sample was composed of TiB, TiB2, nanocrystalline layer assistance. The borided layer of the SMAT sample was composed of TiB, TiB2, Ti3B4, and Ti with supersaturated boron atoms. Ti3B4, and Ti with supersaturated boron atoms. Compared to the coarse-grained borided sample, the SMAT borided sample exhibits a similar hardness value, but improved surface toughness. Meanwhile, the obvious brittleness crack was not found on the borided layer of the SMAT sample. The excellent toughness of borided layer of the SMAT sample may be attributed to the low boriding temperature.

Materials 2016, 9, 993 9 of 10

Compared to the coarse-grained borided sample, the SMAT borided sample exhibits a similar hardness value, but improved surface toughness. Meanwhile, the obvious brittleness crack was not found on the borided layer of the SMAT sample. The excellent toughness of borided layer of the SMAT sample may be attributed to the low boriding temperature.

Acknowledgments: This research was supported by the National High Technology Research and Development Program (2012AA03A508) and the National Natural Science Foundation of China (U1360102 and 51275344). Author Contributions: The research work presented in this article was carried out in a national collaborative team. The team consists of two supervisors, Weiping Tong, Jian Sun, and three researchers, Quantong Yao, Yuzhu Fu, and Hui Zhang. The two supervisors provided the original idea and supervised the research work; Quantong Yao, Yuzhu Fu, and Hui Zhang obtained and analyzed the experimental data; Quantong Yao wrote the manuscript. Conﬂicts of Interest: The authors declare no conﬂict of interest.

References

1. Wen, M.; Wen, C.; Hodgson, P.; Li, Y. Improvement of the biomedical properties of titanium using SMAT and thermal oxidation. Colloids Surf. B Biointerfaces 2014, 116, 658–665. [CrossRef][PubMed] 2. Huang, R.; Lu, S.; Han, Y. Role of grain size in the regulation of osteoblast response to Ti-25Nb-3Mo-3Zr-2Sn alloy. Colloids Surf. B Biointerfaces 2013, 111, 232–241. [CrossRef][PubMed] 3. Wu, Y.; Zitelli, J.P.; TenHuisen, K.S.; Yu, X.; Libera, M.R. Differential response of Staphylococci and osteoblastes to varying titanium surface roughness. Biomaterials 2011, 32, 951–960. [CrossRef][PubMed] 4. Ando, Y.; Noda, Y.; Kobayashi, A. Photo-catalytic titanium oxide film deposition by atmospheric TPCVD using vortex Ar plasma jet. Vacuum 2014, 110, 190–195. [CrossRef] 5. Santecchia, E.; Hamouda, A.M.S.; Musharavati, F.; Zalnezhad, E.; Cabibbo, M.; Spigarelli, S. Wear resistance investigation of titanium nitride-based coatings. Ceram. Int. 2015, 41, 10349–103799. [CrossRef] 6. Attar, H.; Prashanth, K.G.; Chaubey, A.K.; Calin, M.; Zhang, L.C.; Scudino, S.; Eckert, J. Comparison of wear properties of commercially pure titanium prepared by selective laser melting and casting processes. Mater. Lett. 2015, 142, 38–41. [CrossRef] 7. Tong, W.P.; Sun, J.; Zuo, L.; He, J.C.; Lu, J. Study on wear and friction resistance of nanocrystalline Fe nitrided at low temperature. Wear 2011, 271, 653–657. [CrossRef] 8. Huang, R.; Han, Y. The effect of SMAT-induced grain refinement and dislocations on the corrosion behavior of Ti-25Nb-3Mo-3Zr-2Sn alloy. Mater. Sci. Eng. C 2013, 33, 2353–2359. [CrossRef][PubMed] 9. Arifvianto, B.; Suyitno, S.T.; Mahardika, M.; Dewo, P.; Iswanto, P.T.; Salim, U.A. Effect of surface mechanical attrion treatment (SMAT) on microhardness, surface roughness and wettability of AISI 316L. Mater. Chem. Phys. 2011, 125, 418–426. [CrossRef] 10. Maawad, E.; Brokmeier, H.G.; Wagner, L.; Sano, Y.; Genzel, C. Investigation on the surface and near-surface characteristics of Ti-2.5Cu after various mechanical surface treatments. Surf. Coat. Technol. 2011, 205, 3644–3650. [CrossRef] 11. Yasuoka, M.; Wang, P.; Zhang, K.; Qiu, Z.; Kusaka, K.; Pyoun, Y.-S.; Murakami, R. Improvement of the fatigue strength of SUS304 austenite stainless steel using ultrasonic nanocrystal surface modification. Surf. Coat. Technol. 2013, 218, 93–98. [CrossRef] 12. Wu, X.; Tao, N.R.; Hong, Y.; Xu, B.; Lu, J.; Lu, K. Microstructure and evolution of mechanically-induced ultrafine grain in surface layer of Al-alloy subjected to USSP. Acta Mater. 2002, 50, 2075–2084. [CrossRef] 13. Wang, Z.B.; Tao, N.R.; Tong, W.P.; Lu, J.; Lu, K. Diffusion of chromium in nanocrystalline iron produced by means of surface mechanical attrition treatment. Acta Mater. 2003, 51, 4319–4329. [CrossRef] 14. Gunes, I.; Kayali, Y. Investigation of mechanical properties of borided Nickel 201 alloy. Mater. Des. 2014, 53, 577–580. [CrossRef] 15. Balusamy, T.; Sankara Narayanan, T.S.N.; Ravichandran, K.; Park, S., II; Lee, M.H. Pack boronizing of AISI H11 tool steel: Role of surface mechanical attrition treatmen. Vacuum 2013, 97, 36–43. [CrossRef] 16. Balusamy, T.; Sankara Narayanan, T.S.N.; Ravichandran, K. Effect of surface mechanical attrition treatment (SMAT) on boronizing of EN8 steel. Surf. Coat. Technol. 2012, 213, 221–228. [CrossRef] 17. Xu, L.; Wu, X.; Wang, H. Influence of surface nano-structured treatment on pack boriding of H13 Steel. J. Mater. Sci. Technol. 2007, 23, 525–528. Materials 2016, 9, 993 10 of 10

18. Liu, Y.; Li, M. Nanocrystallization mechanism of beta phase in Ti-6Al-4V subjected to severe plastic deformation. Mater. Sci. Eng. A 2016, 669, 7–13. [CrossRef] 19. Liu, Y.G.; Li, H.M.; Li, M.Q. Characterization of surface layer in TC17 alloy treated by air blast shot peening. Mater. Des. 2015, 65, 120–126. [CrossRef] 20. Chui, P.; Sun, K.; Sun, C.; Yang, X.; Shan, T. Effect of surface nanocrystallization induced by fast multiple rotation rolling hardness and corrosion behavior of 316L stainless steel. Appl. Surf. Sci. 2011, 257, 6787–6791. [CrossRef] 21. Hassain, F.Z.; Ketabchi, M. Nano grained AZ31 alloy achieved by equal channel angular rolling process. Mater. Sci. Eng. A 2011, 528, 6426–6431. [CrossRef] 22. Schwarz, R.B.; Koch, C.C. Formation of amorphous alloys by the mechanical alloying of crystalline powders of pure metals and powders of intermetallics. Appl. Phys. Lett. 1986, 49, 146–148. [CrossRef] 23. Manna, I.; Chattopadhyay, P.P.; Banhart, F.; Fecht, H.J. Formation of face-centered-cubic zirconium by mechnical attrtion. Appl. Phys. Lett. 2002, 81, 4136–4138. [CrossRef] 24. Atar, E.; Kayali, E.S.; Cimenoglu, H. Characteristics and wear performance of borided Ti6Al4V alloy. Surf. Coat. Technol. 2008, 202, 4583–4590. [CrossRef] 25. Sun, J.; Tong, W.P.; Zhang, H.; Zuo, L.; Wang, Z.B. Evaluation of surface-modiﬁed 20CrMo by plasma nitriding coupled with ion sputtering and SMAT. Surf. Coat. Technol. 2012, 213, 247–252. [CrossRef] 26. Campos, I.; Rosas, R.; Figueroa, U.; velázquez, C.V.; Meneses, A.; Guevara, A. Fracture toughness evaluation using Palmqvist crack models on AISI 1045 borided steels. Mater. Sci. Eng. A 2008, 488, 562–568. [CrossRef] 27. Ebrahimi-Kahrizsangi, R.; Abdellahi, M.; Bahmanpour, M. Self-ignited synthesis of nanocomposite powders induced by Spex mills; modeling and optimizing. Ceram. Int. 2015, 41, 3137–3151. [CrossRef]

© 2016 by the authors; licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).