Surface nanocrystallization by ultrasonic nano-crystal surface modification and its effect on gas nitriding of Ti6Al4V alloy

Author’s Accepted Manuscript Surface Nanocrystallization by Ultrasonic Nanocrystal Surface Modification and its Effect on Gas Nitriding of Ti6Al4V Alloy Jun Liu, Sergey Suslov, Azhar Vellore, Zhencheng Ren, Auezhan Amanov, Young-Sik Pyun, Ashlie Martini, Yalin Dong, Chang Ye www.elsevier.com/locate/msea

PII: DOI: Reference:

S0921-5093(18)31166-3 https://doi.org/10.1016/j.msea.2018.08.089 MSA36860

To appear in: Materials Science & Engineering A Received date: 23 February 2018 Revised date: 28 July 2018 Accepted date: 26 August 2018 Cite this article as: Jun Liu, Sergey Suslov, Azhar Vellore, Zhencheng Ren, Auezhan Amanov, Young-Sik Pyun, Ashlie Martini, Yalin Dong and Chang Ye, Surface Nanocrystallization by Ultrasonic Nano-crystal Surface Modification and its Effect on Gas Nitriding of Ti6Al4V Alloy, Materials Science & Engineering A, https://doi.org/10.1016/j.msea.2018.08.089 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 galley proof before it is published in its final citable 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.

Surface Nanocrystallization by Ultrasonic Nano-crystal Surface Modification and its Effect on Gas Nitriding of Ti6Al4V Alloy Jun Liu1, Sergey Suslov2, Azhar Vellore3, Zhencheng Ren1, Auezhan Amanov4, Young-Sik Pyun4, Ashlie Martini3, Yalin Dong1*, Chang Ye1* 1

Department of Mechanical Engineering, University of Akron, Akron, OH 44325, United States

2

Qatar Environment and Energy Research Institute (QEERI), Qatar Foundation, Doha, Qatar

3

Department of Mechanical of Engineering, University of California - Merced, Merced, CA

95343, USA 4

Department of Mechanical Engineering, Sun Moon University, Asan 31460, Korea

[email protected] [email protected] *Corresponding author.

Abstract The effects of Ultrasonic Nanocrystal Surface Modification (UNSM) on the gas nitriding behavior of Ti6Al4V alloy have been investigated. Gas nitriding was performed at 700 and 800 °C. The microstructure after UNSM and gas nitriding was characterized using X-ray diffraction, scanning electron microscopy and transmission electron microscopy. Microstructural investigations revealed the formation of an approximately 10 μm thick severe plastic deformation (SPD) layer as well as nano-grains after UNSM treatment. The UNSM-treated Ti6Al4V alloy formed 0.26 μm and 1.35 μm thick nitride layers after nitriding at 700 °C and 800 °C, respectively, and UNSM resulted in an increased layer thickness relative to untreated samples at both temperatures. The results suggest that nitrogen adsorption and reaction capability were enhanced in the UNSM-treated Ti6Al4V alloy. This enhancement can be attributed to highdensity dislocations and grain boundaries that were introduced by UNSM and served as efficient channels for nitrogen diffusion.

1

Keywords Ti64; Gas nitriding; Ultrasonic nanocrystal surface modification; Surface severe plastic deformation; Nanocrystalline; Scratch Test. 1. Introduction Ti6Al4V, first developed in the mid-1950s, contains a mixture of 90% titanium, 6% aluminum, and 4% vanadium by weight. This material not only has extraordinary mechanical properties but also an inherent ability to osseointegrate, which is very favorable for orthopedic and dental implants. Nonetheless, poor antifriction characteristics and low hardness have limited its broad use in applications [1]. In order to improve both the surface hardness and wear resistance of Ti6Al4V, surface treatment techniques, such as boronizing, oxidation and nitriding [2–4] have been applied to form compound layers consisting of borides, oxides and nitrides. These compound layers are well known to be ceramic materials (TiB, TiO2, and Ti3N4) which exhibit considerable hardness, high corrosion resistance and high wear resistance. One of the most widely applied techniques among the above-mentioned surface treatments is gas nitriding [5]. However, gas nitriding is usually performed at a very high temperatures (~1050 °C) and for a very long duration (~100 hours), which may result in surface deterioration, precipitation, grain growth and phase transformation [6–9]. Therefore, the process conditions of gas nitriding should to be optimized to overcome these drawbacks. The hardness and wear resistance improvement induced by nitriding is mainly determined by the nitride layer thickness. The presence of grain boundaries, triple junctions, dislocations and atomic scale microstructural defects, can accelerate the diffusion of interstitial atoms in metallic materials, and thereby increasing the depth of nitride layer [10]. Surface severe plastic deformation (SSPD) techniques have received considerable attention due to their ability to achieve superior mechanical properties by generating grain refinement in the topmost material surface. Several SSPD techniques have been widely used, like shot peening [11], laser shock peening [12], surface mechanical attrition treatment (SMAT) [13], ultrasonic impact treatment (UIT) [14] and deep rolling [15]. The plastic deformation induced by these SSPD techniques leads to high-density defects, such as dislocations, twinning and subgrain boundaries. Additionally, with the accumulation of dislocations and twins, the initial coarse grains can be 2

subdivided into finer ones via dynamic recrystallization (DRX), which involves progressive increase in grain boundary misorientation. Also, a conversion of low angle boundaries into high angle boundaries is responsible for the grain refinement. The refined grains not only give rise to remarkable mechanical properties but also contribute to the increase of grain boundaries, triple junctions and many other defects which speed up the diffusion of interstitial atoms in metallic materials [16]. Ultrasonic Nanocrystal Surface Modification (UNSM) is an SSPD technique developed over the last of couple of years that can be used to improve the mechanical performance of metallic materials [17][18]. Like other SSPD processes, the plastic strain generated during UNSM process is induced by mechanical strikes that can be considered as cold forging. During a UNSM treatment, a tungsten carbide tip is driven by an ultrasonic horn, which can vibrate at a high frequency of 20 kHz. The amplitude of the vibration is 8~50 μm at the end of the tip. These strikes can generate SSPD with a certain depth from the top surface and thereby induce nanocrystallization in the surface layer. UNSM is implemented by a CNC machine, so the strikes (velocity and density) can be precisely controlled. In addition, the ultrasonic strikes is superimposed by a static load, making the strike intensity uniform [19]. This technique has been broadly applied to process titanium and its alloys [20–23], magnesium alloys [24,25], steel [26,27] and metallic glass [28], to improve surface hardness, fretting wear, fatigue resistance and fracture strength. The goal of this work is to study the effect of UNSM, as a pretreatment process, on the alteration of the near-surface microstructure of Ti6Al4V alloy and the effects on nitriding. To this end, comprehensive characterization was performed to study the nitriding behavior of Ti6Al4V at 700 °C and 800 °C with and without UNSM treatment. The nanostructured layer after UNSM was characterized using scanning electron microscopy (SEM), X-ray diffraction (XRD) and transmission electron microscopy (TEM). In addition, the microstructures of the nitrided Ti6Al4V alloy, with and without UNSM treatment, were studied using SEM and TEM. Finally, the mechanical properties of the nitride layers were evaluated using Vickers microhardness, Rockwell adhesion and scratch testing. We have demonstrated that UNSM induces surface nanocrystallization in Ti6Al4V and the resultant high density of dislocations and naoscale grain boundaries substantially increase nitriding efficiency.

3

2. Materials and methods 2.1. Sample preparation The experiments were carried out on α/β titanium alloy Ti6Al4V with the nominal chemical composition of 0.08 wt% carbon, 5.50-6.75 wt% aluminum, 3.5-4.5 wt% vanadium, 0.4 wt% max. iron, 88.10-90.92 wt% titanium and 0-0.3% other components. The material was received in the form of an annealed 6×6 inches square sheet with a thickness of 0.125 inches. Samples measuring 0.6×0.6 inches square were cut from the as-received sheet and then smoothed by mechanically abrading the sample surfaces with SiC papers from 320 to 1200 grits. In a UNSM process (Fig. 1), a tungsten carbide ball (tip) scans over the surface while striking it at a high frequency. The repeated, high-frequency strikes induce plastic deformation and crystalline refinement on material surfaces. UNSM was performed under the following conditions: distance between adjacent scans of 10 μm, scanning speed of 1000 mm/min, static load of 5 kg, amplitude of 24 μm and frequency of 20 kHz. A tungsten carbide tip with a 2.4 mm diameter was applied for all the UNSM treatments. Both non-treated and UNSM-treated samples were placed inside an OTF-1200X-S-UL furnace for gas nitriding. The treatment was carried out in Ammonia with a purity of 99.999% using a flow rate of 1000 ml/min. 700 °C and 800 °C were chosen as the working temperatures in order to avoid phase transformation and the duration was 6 hours for all samples. After isothermal exposure, the samples were cooled to room temperature in ammonia within the furnace. 2.2. Characterization To study the surface and subsurface microstructure evolution, the samples were cut in the middle and then the cross-sectional surfaces were grounded, polished to a mirror-like finish and then etched using Kroll’s Reagent (5 vol.% HF + 10 vol.%HNO3 + 85 vol.% H2O) supplied for 15 s. Then they were studied using a Tescan Lyra 3 SEM system. Chemical composition was examined using energy dispersive X-ray (EDX) integrated with the SEM. TEM characterization was carried out using an FEI Talos F200X TEM working at 200 kV, whereas the compositional anaylsis of samples was performed using TEM elemental mapping. The TEM samples were prepared using the lift-out technique in an FEI Versa 3D LoVac FIB-SEM Dual Beam system equipped with an FEI EasyLift system. To identify phases before and after UNSM treatment, 4

XRD characterization was performed using a Rigaku Ultima IV diffractometer with Cu Kα1radiation. The diffractometer, running at 40 kV/mA, was used to scan a 2θ range from 20° to 80° with a scan speed of 1.00 °/min. 2.3. Mechanical properties The microhardness for the control and UNSM-treated samples after 6 hours nitriding at 700 °C was measured using a Wilson Tukon 2100 system with a Vickers indenter under a 50 g load and a 10 s dwell time. The hardness of the rest of the samples was measured under a constant load of 400 gf and a 15 s holding time. Five measurements were performed for each reported data point. The adhesion of titanium nitride was measured using Rockwell HRC indentations under a load of 150 kg. The adhesion behavior of nitrided samples with and without the UNSM pretreatment was also evaluated and compared using scratch tests. The scratch hardness tests were performed following ASTM Standard G171–03 [29] using a Rtec Multifunction Tribometer. A constant normal load of 1 N was applied on a diamond sphero-conical tip indenter of radius 200 µm while moving the sample 5 mm laterally at a constant speed of 0.2 mm/s. The width of the scratch was then measured using white light interferometer images and the scratch hardness was calculated as 𝐻𝑆𝑝 = 𝑘𝑃 ⁄𝑤 2 , where P is the applied normal force (gf), w is the scratch width (µm), HSp is the scratch hardness number (GPa), and k is a geometric constant that takes the value 24.98 for the unit system used here. The scratch hardness number of each sample was determined as an arithmetic mean of a total of nine determinations (that is, 3 scratches and 3 width measurement locations per scratch). The scratch paths were then analyzed using SEM to study the failure mechanisms.

Figure 1 Schematic illustration of a UNSM process.

5

3. Results and discussion 3.1. Nanocrystallization by UNSM Fig. 2a shows the cross-sectional secondary electron images (SEI) of the UNSM-treated Ti6Al4V (without nitriding) microstructure exhibiting α/β phases. The α-Ti has a hexagonal close-packed (HCP) structure and can be identified as the dark gray areas on the image, and the β-Ti has a body-centered cubic (BCC) structure, identified as light gray regions. An approximately 15 μm thick severe plastic deformation (SPD) zone is distinguished by elongated β-Ti that appears nearly parallel to the top surface. This elongation resulted from the plastic strain introduced by ultrasonic strikes. Fig. 2b reveals clear evidence of the difference in the microstructure caused by the UNSM treatment. The SPD was also evidenced by the β fragments that were refined by UNSM. A similar observation of β phase refinement in Ti6Al4V alloy after shot peening was reported by Farokhzadeh et al. [16]. Fig. 2c shows the microstructure below the SPD layer. With less plastic deformation, the β particles maintain their random orientation and larger grain size as compared to the β particles in the SPD zone.

6

Figure 2. (a) Cross-section SEM image of the UNSM-treated Ti6Al4V alloy, (b) Magnification of the SPD layer identified as b in Fig. 2a, (c) Magnification of an interior region. Fig. 3 shows the XRD patterns of the Ti6Al4V samples before and after UNSM. It can be clearly seen that both samples consist of a α-Ti phase, which could be indexed to the {100}, {002}, {101}, {102}, {110}, {103} and {112} crystal planes. However, compared with the control sample, the full width at half maximum (FWHM) of the α-Ti peaks was broadened and the diffraction intensity peaks decreased significantly for the UNSM-treated sample. The distinct difference in the XRD patterns exhibited by the control and the UNSM-treated samples can be attributed to the reduction of grain size [30] and lattice distortion which result from the generation of residual stress and/or dislocations [31]. Recently, it was experimentally proven that grain boundaries, dislocations and other atomic level microstructural defects can effectively accelerate diffusion of gas interstitial atoms in metals [32–34]. In addition, the crystallinity analysis indicates that only the α-Ti phase can be clearly identified in the UNSM-treated 7

samples. In contrast, the pattern for the control sample contains a β-Ti peak, which could be indexed as the {110} crystal plane at around 39.63° (2θ). It is possible that there was some phase transformation from β-Ti to α-Ti during the UNSM process [35]. However, from direct microstructure observation (Fig. 2b), β-Ti can be clearly identified by the elongated white particles. It is possible that the severe broadening of the peaks for the α phase makes it unable to resolve the β-Ti peak at around 39.63° (2θ) in the XRD patterns.

Figure 3 (a) XRD patterns of control and UNSM-treated Ti6Al4V samples, and (b) comparison of FWHM of the {100}, {002} and {101} peaks of the control and UNSM-treated Ti6Al4V alloy. In order to confirm the generation of nano-grains, the microstructure of the Ti6Al4V alloy after UNSM was characterized using TEM. Fig. 4a and 4b show bright field and dark field images of the top surface of the material, respectively. It is observed that the grain size after UNSM treatment is at nanoscale. Grain boundaries are poorly defined with complicated non-uniform contrast, indicating non-equilibrium state as well as high internal stress [36][37][38]. Nanocrystallization mechanisms in Ti6Al4V alloy should be discussed separately due to the material’s duplex phase composition. It is known that twining dominates deformation in α-Ti at low strain due to its hcp structure [39][40]. As plastic strain accumulates, the generation of dislocations results in a lamellar structure with low angle misorientation [41]. When the plastic strain reaches a certain level, rearrangement of dislocations occurs in order to minimize the total system energy, which leads to the generation of low angle misoriented blocks and the formation of nanograins due to DRX [40]. On the other hand, in bcc β-Ti, dislocation slip dominates plastic deformation. It was suggested that the density increase of dislocations at the phase boundary 8

between α-Ti and β-Ti gives rise to a higher density of dislocation tangles (DTs) and dislocation walls (DWs), and then dislocations are triggered in the β grains [42]. With the increase of plastic strain caused by UNSM, subgrain boundaries are formed and gradually transform into high angle grain boundaries. Finally, high-strain rate deformation induced dynamic recrystallization results in formation of nanograins with random orientations [43].

Figure 4 Cross-sectional bright field (a), and dark field (b) TEM images of the top layer of the Ti6Al4V alloy after UNSM treatment. 3.2. Effects of gas nitriding on Ti6Al4V After gas nitriding, all of the nitrided Ti6Al4V samples had a homogeneous surface morphology with a uniform golden color, indicating the formation of TiN. These observations were in agreement with the findings reported by Zhecheva et al. [44] and Lee [45]. The formation of a nitride layer results from the inward diffusion of nitrogen and its subsequent reaction with the titanium and other alloying elements. Fig. 5a and b present the cross-sectional microstructure of the Ti6Al4V alloy nitrided at 700 °C, without and with UNSM pretreatment, respectively. Due to weak nitriding kinetics at this temperature, the compound layer in the sample without UNSM pretreatment is not uniform (Fig. 5a). In contrast, a uniform compound layer can be observed by the bright white line in the UNSM sample with a thickness of approximately 0.26 μm (Fig. 5b). Thus it is reasonable to claim that the nitrogen diffusion along dislocations and grain boundaries can be accelerated at this temperature. As shown in Fig. 5c and d, the cross-sectional SEM images of both the control and UNSM Ti6Al4V alloy after nitriding at 800 °C for 6 hours clearly reveal the presence of much thicker nitride layers than those at 700 °C. It is also observed that 9

the nitride layer of the UNSM-assisted sample (1.35 μm) is thicker than that in the control one (1.08 μm). The thickness of the nitride layer in both UNSM and control samples after nitriding at 700 and 800 °C are summarized in Fig. 6. It was found that the thickness of the nitride layer increases with temperature, and it was further increased significantly by UNSM pretreatment. The content of the nitride layer was analyzed using EDS. EDS tests of the UNSM-treated Ti6Al4V alloy at 800 °C nitriding show that nitrogen exists at spot 1 with 9.81% in wt (Table 1), confirming the presence of a titanium nitride layer. It is known that vanadium serves as a β-Ti stabilizer in Ti6Al4V. Therefore, it is reasonable to detect it at spot 2. EDS tests conducted at spot 3 showed aluminum and titanium in α-Ti, as expected. The microstructure of the UNSM Ti6Al4V alloy at 800 °C nitriding was characterized by TEM. Strong annealing at this temperature resulted in remarkable grain growth, which is evidenced in Fig. 5e. Overall, the grain size is at the microscale. Grain coarsening indicates a decrease in density of grain boundaries and dislocations which serve as fast diffusion paths for nitrogen, suggesting a decrease of nitride kinetics. Even so, the compound layer in the UNSM sample is still thicker than that of control. In addition to a compound layer, it is anticipated that the process generates a diffusion zone formed by an interstitial solid solution of nitrogen atoms in the titanium matrix. However, in our aforementioned EDS results, it was difficult to detect any nitrogen underneath the compound layer (spot 2 and spot 3). For a more detailed analysis of the microstructure, we performed elemental mapping of nitrogen using EDS in TEM. As shown in Fig. 5f, a nitrogen layer with high concentration can be clearly distinguished, indicating the presence of the compound layer. Nitrogen with a relatively lower concentration (in comparison with that of the compound layer) is distributed uniformly underneath the compound layer, suggesting the existence of a diffusion zone.

10

Figure 5 Cross-sectional SEM images of (a) control and (b) UNSM-treated Ti6Al4V alloy after nitriding at 700°C for 6 h. Cross-sectional SEM images of (c) control and (d) UNSM-treated Ti6Al4V after nitriding at 800°C for 6 h. (e) TEM image and (f) elemental mapping of nitrogen of the UNSM-treated Ti6Al4V alloy after nitriding at 800°C for 6 h.

Figure 6 Compound layer thicknesses measured in Ti6Al4V alloy after different processes.

11

Table 1 Point EDS (wt.%) results after nitriding the UNSM-treated Ti6Al4V alloy at 800°C for 6 h Element

Spot 1

Spot 2

Spot 3

N

9.81

-

-

Al

0.89

4.31

4.77

V

-

21.48

-

Ti

89.31

74.21

95.23

By comparing the 700 °C and the 800 °C cases, we may safely draw the conclusion that, due to high nitride kinetics at elevated temperature, the diffusion of nitrogen is significantly enhanced by raising the temperature by 100 °C. With the UNSM-treated Ti6Al4V, nitrided at either 700 °C or 800 °C, the surface abundance of dislocations and grain boundaries enables the formation of a thicker nitride layer, which covers the entire surface without spallation. Similar results were reported by Balusamy et al. [46] and Lin et al. [47] for SMAT-treated stainless steel (SS). Balusamy et al. observed a three-fold increase in thickness of the nitride layer formed on 304 SS using SMAT pretreatment with 5 mm diameter balls. Lin et al. claimed the depth of the nitride layer on 321 SS doubled (from 5 to 10 μm) when the nitriding treatment was carried out at 673 K for 4 hours. UNSM, similar to SMAT, results in an increase in dislocation density, and thus promotes the refinement of grain size with high free energy at the top surface, accelerating nitrogen diffusion, as schematically shown in Fig. 7. However, unlike SMAT, both the strike intensity and density in UNSM can be precisely controlled for optimal performance. This suggests that UNSM is an ideal pretreatment for assisting nitriding.

12

Figure 7 Illustration of nitrogen diffusion during gas nitriding on (a) control and (b) UNSMtreated Ti6Al4V. 3.3. Mechanical properties The mechanical properties of Ti6Al4V alloy before and after UNSM and nitriding were examined using Vickers hardness testing (Fig. 8). The average Vickers hardness of the UNSMtreated sample was 411.6 HV, showing an approximately 31% improvement as compared with that of the control one. This improvement originates from cold working and grain refinement strengthening [48]. Figure 8b shows the hardness of the control and UNSM-treated samples after 6 hours nitriding at 700 °C. The testing load was lowered to 50 g since the compound layers are very thin. It can be seen that, as compared with the control sample, the UNSM-treated one shows a 22.4% increase in hardness after nitriding. After 6 hours nitriding at 800 °C, the compound layer formed in the UNSM-treated sample increased the Vickers hardness to 602.2 HV (measured under the load of 400 g), indicating a nearly 17% improvement in surface hardness as compared with that of the control. For both cases, the improvements are due to the thicker compound layer in the UNSM-treated sample.

Figure 8 Vickers hardness of Ti6Al4V alloy after various processes. (a) Averaged Vickers hardness of the control and UNSM-treated samples, (b) after 6 hours after nitriding at 700 °C, and (c) after 6 hours nitriding at 800 °C.

13

HRC tests were carried out to investigate the adhesion strength between the compound layer and the Ti6Al4V substrate. The main difference between the control and the UNSM Ti6Al4V after 700 °C nitriding is that no cracks were present in the control Ti6Al4V sample. The many small cracks in the UNSM-treated Ti6Al4V sample after 700 °C nitriding imply the presence of the compound layer, which was not formed in the control at 700 °C due to inefficient lattice diffusion. After nitriding at 800 °C though, UNSM gave rise to a thicker compound layer, the two samples show no apparent difference in adhesion, with the presence of radial cracks in both the control and UNSM Ti6Al4V alloys. Nitriding at high temperature not only results in the formation of a thick compound layer but also leads to grain growth. This suggests that the former facilitates premature crack initiation and the latter decreases the substrate’s resistance to crack propagation [16].

Figure 9 HRC adhesion test: (a) control and (b) UNSM-treated Ti6Al4V after nitriding at 700°C for 6 h; (c) control and (d) UNSM-treated Ti6Al4V after nitriding at 800°C for 6 h.

The adhesion behavior during sliding of nitride samples with and without the UNSM pretreatment was evaluated and compared using scratch tests. The generated scratch grooves 14

were studied using SEM to examine possible failure mechanisms. Without UNSM pretreatment, tensile cracks arise and are limited to the scratch grooves after nitriding at 700 °C, as shown in Fig. 10a. This reveals the relatively poor strength of the Ti6Al4V substrate since there was no intact compound layer formed (Fig. 5a) and the stylus has penetrated into the Ti6Al4V substrate (Fig. 11b). The scratch hardness of the control Ti6Al4V is almost equal to that of the UNSM processed one (Fig. 11a), which results from the large penetration depth induced by the stylus that is far beyond the thickness of nitride layers (0.26 μm for the UNSM Ti6Al4V). Angular cracks can be observed at the edge of scratch grooves for both the control and UNSM samples (Fig. 10b and d). The cracks in the control samples are comparatively wide (Fig. 10b), which can be clearly visualized in the low magnification image shown in Fig. 10a. In Fig. 10d, it can be seen that the generated cracks are small but with high density for the UNSM processed workpiece. These cracks are formed due to the superimposition of two kinds of stresses: tensile stress parallel to the scratch groove caused by stylus moving action and tensile stress perpendicular to scratch groove induced by bending [49][50]. Thus, it is concluded that, due to preprocessing using UNSM, the generated nitride layer has a higher strength and better adhesion in comparison with that in the non-UNSM processed sample.

15

Figure 10 (a) SEM image showing the scratch on the surface of the control Ti6Al4V after nitriding at 700°C, (b) magnified SEM image showing angular cracks at the rim of the scratch at the location of b indicated in (a), (c) SEM image showing the scratch on the surface of the UNSM Ti6Al4V after nitriding at 700°C and (d) magnified SEM image showing angular cracks at the rim of the scratch at the location of d indicated in (c).

16

Figure 11 Measured scratch hardness (a) and scratch depth (b) after scratch tests at 1 N normal load.

The scratch hardness of Ti6Al4V after nitriding at 800 °C with UNSM pretreatment is higher than that of the sample without pretreatment (Fig. 11a), which originates from the thicker nitride layer assisted by the nano-grains. However, tensile cracks and angular cracks can be observed in both the control and UNSM processed samples, as shown in Figs. 12a and b, indicating the same degree of adhesion between the compound layer and the Ti6Al4V substrate. Moreover, coarsening grains (Fig. 5e) produced by the elevated nitriding temperature decrease the strength of the substrate for both the control and the UNSM Ti6A4V alloy, explaining the larger and

17

higher density of the tensile crack in comparison with that of the samples after nitriding at 700 °C (Fig. 10). After nitriding at 700 °C without UNSM pretreatment, the formed compound layers are thin, as shown in Fig. 5a and 5b, leading to a similar hardness and scratch test results. The hardness and scratch tests of the samples after 800 °C nitriding show more a significant difference than that of the samples after 700 °C nitriding. This is because the thicknesses of the compound layers are increased so these layers start to dominate the hardness and scratch tests. Therefore, the difference in thickness of the compound layers, which can be clearly observed in Fig. 5c and 5d, results in the difference of hardness and scratch tests performance.

Figure 12 (a) SEM image showing the scratch on the surface of the control Ti6Al4V after nitriding at 800°C and (b) a SEM image showing the scratch on the surface of the UNSM Ti6Al4V after nitriding at 800°C.

CONCLUSIONS In this work, UNSM was used as a pretreatment step to enhance the nitriding kinetics during gas nitriding of Ti6Al4V alloy. The available evidence suggests that: 1.

UNSM pretreatment introduced an approximately 15 μm thick severe plastic deformation

(SPD) zone with abundant dislocations and grain boundaries; 2.

UNSM can refine the grains in Ti6Al4V alloy to the scale of 30~130 nm;

18

3.

UNSM pretreatment increased nitrogen diffusion and enhanced nitriding kinetics, leading

to the formation of a thicker nitride layer at both 700 °C and 800 °C; 4.

Nitriding at 800 °C remarkably increased the grain size, resulting in a significant decrease

in strength of the Ti6Al4V substrate as well as in the adhesion between the nitride layer and the substrate.

ACKNOWLEDGMENTS The authors (J. Liu, Z. Ren, Y. Dong and C. Ye) are grateful for the financial support of this research by the start-up fund provided by the College of Engineering at The University of Akron. Scratch hardness experiments were performed with equipment purchased using funding from the US Army Research Office under contract/grant number W911NF1610549.

DATA AVAILABILITY The datasets supporting the conclusions of this article are included within the article.

REFERENCES [1]

R.H. Hu, J.K. Lim, Hardness and wear resistance improvement of surface composite layer on Ti-6Al-4V substrate fabricated by powder sintering, Mater. Des. 31 (2010) 2670–2675. doi:10.1016/j.matdes.2009.11.057.

[2]

E. Atar, E.S. Kayali, H. Cimenoglu, Characteristics and wear performance of borided Ti6Al4V

alloy,

Surf.

Coatings

Technol.

202

(2008)

4583–4590.

doi:10.1016/j.surfcoat.2008.03.011. [3]

M. Jamesh, T.S.N. Sankara Narayanan, P.K. Chu, Thermal oxidation of titanium: Evaluation of corrosion resistance as a function of cooling rate, Mater. Chem. Phys. 138 (2013) 565–572. doi:10.1016/j.matchemphys.2012.12.020.

[4]

K. Farokhzadeh, A. Edrisy, Fatigue improvement in low temperature plasma nitrided Ti– 6Al–4V alloy, Mater. Sci. Eng. A. 620 (2015) 435–444. doi:10.1016/j.msea.2014.10.008.

[5]

L. Ge, N. Tian, Z. Lu, C. You, Influence of the surface nanocrystallization on the gas nitriding

of

Ti-6Al-4V

alloy,

Appl.

Surf.

Sci.

286

(2013)

412–416. 19

doi:10.1016/j.apsusc.2013.09.105. [6]

K. Tokaji, T. Ogawa, H. Shibata, The Effects of Gas Nitriding on Fatigue Behavior in Titanium and Titanium Alloys, J. Mater. Eng. Perform. 8 (1999) 159–167. doi:10.1361/105994999770346990.

[7]

H.W. Lee, J.H. Kong, D.J. Lee, H.Y. On, J.H. Sung, A study on high temperature gas nitriding and tempering heat treatment in 17Cr-1Ni-0.5C, Mater. Des. 30 (2009) 1691– 1696. doi:10.1016/j.matdes.2008.07.023.

[8]

S. Malinov, A. Zhecheva, W. Sha, Relation between the microstructure and properties of commercial titanium alloys and the parameters of gas nitriding, Met. Sci. Heat Treat. 46 (2004) 286–293. doi:10.1023/B:MSAT.0000048836.23370.69.

[9]

B. Sarma, K.S. Ravi Chandran, Recent advances in surface hardening of titanium, Jom. 63 (2011) 85–92. doi:10.1007/s11837-011-0035-0.

[10] W.P. Tong, N.R. Tao, Z.B. Wang, J. Lu, K. Lu, Nitriding iron at lower temperatures., Science. 299 (2003) 686–688. doi:10.1126/science.1080216.c [11]

H. Lee, S. Mall, S. Sathish, Investigation into effects of re-shot-peening on fretting fatigue behavior

of

Ti-6Al-4V,

Mater.

Sci.

Eng.

A.

390

(2005)

227–232.

doi:10.1016/j.msea.2004.08.040. [12] C. Ye, S. Suslov, B.J. Kim, E.A. Stach, G.J. Cheng, Fatigue performance improvement in AISI 4140 steel by dynamic strain aging and dynamic precipitation during warm laser shock peening, Acta Mater. 59 (2011) 1014–1025. doi:10.1016/j.actamat.2010.10.032. [13] S. Anand Kumar, S. Ganesh Sundara Raman, T.S.N. Sankara Narayanan, R. Gnanamoorthy, Influence of counterbody material on fretting wear behaviour of surface mechanical

attrition

treated

Ti-6Al-4V,

Tribol.

Int.

57

(2013)

107–114.

doi:10.1016/j.triboint.2012.07.021. [14] M.A. Vasylyev, S.P. Chenakin, L.F. Yatsenko, Ultrasonic impact treatment induced oxidation

of

Ti6Al4V

alloy,

Acta

Mater.

103

(2016)

761–774.

doi:10.1016/j.actamat.2015.10.041. [15] R.K. Nalla, I. Altenberger, U. Noster, G.Y. Liu, B. Scholtes, R.O. Ritchie, On the influence of mechanical surface treatments-deep rolling and laser shock peening-on the fatigue behavior of Ti-6Al-4V at ambient and elevated temperatures, Mater. Sci. Eng. A. 355 (2003) 216–230. doi:10.1016/S0921-5093(03)00069-8. 20

[16] K. Farokhzadeh, J. Qian, A. Edrisy, Effect of SPD surface layer on plasma nitriding of Ti6Al-4V alloy, Mater. Sci. Eng. A. 589 (2014) 199–208. doi:10.1016/j.msea.2013.09.077. [17] I.H. Cho, G.H. Song, C.S. Kim, A. Nobuhide, A. Combs, J. Park, et al., Nano structured surface modification of tool steel and its beneficial effects in mechanical properties, J. Mech. Sci. Technol. 19 (2005) 2151–2156. doi:10.1007/BF02916512. [18] S.H. Han, J.W. Han, Y.Y. Nam, I.H. Cho, Fatigue life improvement for cruciform welded joint by mechanical surface treatment using hammer peening and ultrasonic nanocrystal surface modification, Fatigue Fract. Eng. Mater. Struct. 32 (2009) 573–579. doi:10.1111/j.1460-2695.2009.01361.x. [19] C. Ye, A. Telang, A.S. Gill, S. Suslov, Y. Idell, K. Zweiacker, et al., Gradient nanostructure and residual stresses induced by Ultrasonic Nano-crystal Surface Modification in 304 austenitic stainless steel for high strength and high ductility, Mater. Sci. Eng. A. 613 (2014) 274–288. doi:10.1016/j.msea.2014.06.114. [20] J. Liu, S. Suslov, S. Li, H. Qin, Z. Ren, G.L. Doll, et al., Electrically Assisted Ultrasonic Nanocrystal Surface Modification of Ti6Al4V Alloy, Adv. Eng. Mater. (2017). doi:10.1002/adem.201700470. [21] A. Amanov, I.S. Cho, D.E. Kim, Y.S. Pyun, Fretting wear and friction reduction of CP titanium and Ti-6Al-4V alloy by ultrasonic nanocrystalline surface modification, Surf. Coatings Technol. 207 (2012) 135–142. doi:10.1016/j.surfcoat.2012.06.046. [22] A.K. Gujba, Z. Ren, Y. Dong, C. Ye, M. Medraj, Effect of ultrasonic nanocrystalline surface modification on the water droplet erosion performance of Ti6Al4V, Surf. Coatings Technol. 307 (2016) 157–170. doi:10.1016/j.surfcoat.2016.08.054. [23] H. Zhang, R. Chiang, H. Qin, Z. Ren, X. Hou, D. Lin, et al., The Effects of Ultrasonic Nanocrystal Surface Modification on the Fatigue Performance of 3D-Printed Ti64, Int. J. Fatigue. 103 (2017) 136–146. doi:10.1016/j.ijfatigue.2017.05.019. [24] X. Hou, H. Qin, H. Gao, S. Mankoci, R. Zhang, X. Zhou, et al., A systematic study of mechanical properties, corrosion behavior and biocompatibility of AZ31B Mg alloy after ultrasonic nanocrystal surface modification, Mater. Sci. Eng. C. 78 (2017) 1061–1071. doi:10.1016/j.msec.2017.04.128. [25] A. Amanov, O. V. Penkov, Y.S. Pyun, D.E. Kim, Effects of ultrasonic nanocrystalline surface modification on the tribological properties of AZ91D magnesium alloy, Tribol. Int. 21

54 (2012) 106–113. doi:10.1016/j.triboint.2012.04.024. [26] B. Wu, P. Wang, Y.-S. Pyoun, J. Zhang, R. Murakami, Effect of ultrasonic nanocrystal surface modification on the fatigue behaviors of plasma-nitrided S45C steel, Surf. Coatings Technol. 213 (2012) 271–277. doi:10.1016/j.surfcoat.2012.10.063. [27] X.J. Cao, Y.S. Pyoun, R. Murakami, Fatigue properties of a S45C steel subjected to ultrasonic nanocrystal surface modification, Appl. Surf. Sci. 256 (2010) 6297–6303. doi:10.1016/j.apsusc.2010.04.007. [28] C. Ma, H. Qin, Z. Ren, S.C. O’Keeffe, J. Stevick, G.L.G.L. Doll, et al., Increasing fracture strength in bulk metallic glasses using ultrasonic nanocrystal surface modification, J. Alloys Compd. 718 (2017) 246–253. doi:10.1016/j.jallcom.2017.05.056. [29] ASTM, Standard G171-03, 2009, Standard Test Method for Scratch Hardness of Materials Using a Diamond Stylus, 3 (2009) 1–7. http://www.astm.org. [30] A.L. Patterson, The scherrer formula for X-ray particle size determination, Phys. Rev. 56 (1939) 978–982. doi:10.1103/PhysRev.56.978. [31] P.H. Lindenmeyer, R. Hosemann, Application of the theory of paracrystals to the crystal structure

analysis

of

polyacrylonitrile,

J.

Appl.

Phys.

34

(1963)

42–45.

doi:10.1063/1.1729086. [32] L. Shen, L. Wang, Y. Wang, C. Wang, Plasma nitriding of AISI 304 austenitic stainless steel with pre-shot peening, Surf. Coatings Technol. 204 (2010) 3222–3227. doi:10.1016/j.surfcoat.2010.03.018. [33] W.P. Tong, Z. Han, L.M. Wang, J. Lu, K. Lu, Low-temperature nitriding of 38CrMoAl steel with a nanostructured surface layer induced by surface mechanical attrition treatment,

Surf.

Coatings

Technol.

202

(2008)

4957–4963.

doi:10.1016/j.surfcoat.2008.04.085. [34] J. Liu, S. Suslov, S. Li, H. Qin, Z. Ren, C. Ma, et al., Effects of ultrasonic nanocrystal surface modification on the thermal oxidation behavior of Ti6Al4V, Surf. Coatings Technol. 325 (2017) 289–298. doi:https://doi.org/10.1016/j.surfcoat.2017.04.051. [35] M. Kheradmandfard, S.F. Kashani-Bozorg, C.L. Kim, A.Z. Hanzaki, Y.S. Pyoun, J.H. Kim, et al., Nanostructured β-type titanium alloy fabricated by ultrasonic nanocrystal surface

modification,

Ultrason.

Sonochem.

39

(2017)

698–706.

doi:10.1016/j.ultsonch.2017.03.061. 22

[36] R.Z. Valiev, A.V. Korznikov, R.R. Mulyukov, Structure and properties of ultrafine-grained materials produced by severe plastic deformation, Mater. Sci. Eng. A. 168 (1993) 141– 148. doi:10.1016/0921-5093(93)90717-S. [37] R.S. Mishra, V.V. Stolyarov, C. Echer, R.Z. Valiev, A.K. Mukherjee, Mechanical behavior and superplasticity of a severe plastic deformation processed nanocrystalline Ti–6Al–4V alloy, Mater. Sci. Eng. A. 298 (2001) 44–50. doi:10.1016/S0921-5093(00)01338-1. [38] Y.G. Ko, W.S. Jung, D.H. Shin, C.S. Lee, Effects of temperature and initial microstructure on the equal channel angular pressing of Ti-6Al-4V alloy, Scr. Mater. 48 (2003) 197–202. doi:10.1016/S1359-6462(02)00356-1. [39] X.L. Wu, K.M. Youssef, C.C. Koch, S.N. Mathaudhu, L.J. Kecskés, Y.T. Zhu, Deformation twinning in a nanocrystalline hcp Mg alloy, Scr. Mater. 64 (2011) 213–216. doi:10.1016/j.scriptamat.2010.10.024. [40] K.Y. Zhu, A. Vassel, F. Brisset, K. Lu, J. Lu, Nanostructure formation mechanism of αtitanium

using

SMAT,

Acta

Mater.

52

(2004)

4101–4110.

doi:10.1016/j.actamat.2004.05.023. [41] J.Z. Lu, L.J. Wu, G.F. Sun, K.Y. Luo, Y.K. Zhang, J. Cai, et al., Microstructural response and grain refinement mechanism of commercially pure titanium subjected to multiple laser

shock

peening

impacts,

Acta

Mater.

127

(2017)

252–266.

doi:10.1016/j.actamat.2017.01.050. [42] Y. Liu, M. Li, Nanocrystallization mechanism of beta phase in Ti-6Al-4V subjected to severe

plastic

deformation,

Mater.

Sci.

Eng.

A.

669

(2016)

7–13.

doi:10.1016/j.msea.2016.05.088. [43] S.J. Lainé, K.M. Knowles, P.J. Doorbar, R.D. Cutts, D. Rugg, Microstructural characterisation of metallic shot peened and laser shock peened Ti–6Al–4V, Acta Mater. 123 (2017) 350–361. doi:10.1016/j.actamat.2016.10.044. [44] A. Zhecheva, S. Malinov, W. Sha, Titanium alloys after surface gas nitriding, Surf. Coatings Technol. 201 (2006) 2467–2474. doi:10.1016/j.surfcoat.2006.04.019. [45] D.B. Lee, I. Pohrelyuk, O. Yaskiv, J.C. Lee, Gas nitriding and subsequent oxidation of Ti6Al-4V alloys, Nanoscale Res. Lett. 7 (2012) 21. doi:10.1186/1556-276X-7-21. [46] T. Balusamy, T.S.N.S. Narayanan, K. Ravichandran, I.S. Park, M.H. Lee, Plasma nitriding of AISI 304 stainless steel: Role of surface mechanical attrition treatment, Mater. Charact. 23

85 (2013) 38–47. doi:10.1016/j.matchar.2013.08.009. [47] Y. Lin, J. Lu, L. Wang, T. Xu, Q. Xue, Surface nanocrystallization by surface mechanical attrition treatment and its effect on structure and properties of plasma nitrided AISI 321 stainless steel, Acta Mater. 54 (2006) 5599–5605. doi:10.1016/j.actamat.2006.08.014. [48] E.O. Hall, The Deformation and Ageing of Mild Steel: III Discussion of Results, Proc. Phys. Soc. Sect. B. 64 (1951) 747–753. doi:10.1088/0370-1301/64/9/303. [49] K. Holmberg, A. Laukkanen, H. Ronkainen, K. Wallin, S. Varjus, A model for stresses, crack generation and fracture toughness calculation in scratched TiN-coated steel surfaces, Wear. 254 (2003) 278–291. doi:10.1016/S0043-1648(02)00297-1. [50] A. Ghabchi, S. Sampath, K. Holmberg, T. Varis, Damage mechanisms and cracking behavior of thermal sprayed WC-CoCr coating under scratch testing, Wear. 313 (2014) 97–105. doi:10.1016/j.wear.2014.02.017.

24