The Machinability of Ultrafine-grained Grade 2 Ti Processed by Equal Channel Angular Pressing

1. Introduction

The term machinability may be taken to imply that there is a property or quality of any given material which can be clearly deÀ ned and quantiÀ ed, thus indicating how easy (or difÀ cult) that mechanical operation can be. In fact, that term is not unambiguous, but the machinability of a mate-rial can be assessed by employing criteria such as: (i) tool

life or limiting rate of metal removal; (ii) cutting forces or power consumption; and (iii) surface À nish and chip mor-phology[1]_.

Machinability ratings for any given material can be ex-perimentally determined using a given type of cutting tool under a constant set of cutting conditions[2]_{. On this respect,} Yamane et al.[3] _{proposed a criteria known as “difÀ} cult-to-cut rating” (DTCR), expressed by the area of a radar chart that contemplates hardness, mechanical strength, elonga-tion, and thermal features of the material under consider-ation. These attributes maintain a close relationship with w w w. j m r t . c o m . b r

© 2012 Brazilian Metallurgical, Materials and Mining Association. Published by Elsevier Editora Ltda.

*Corresponding author.

E-mail address: [email protected] (A.I. S. Antonialli)

All processes of severe plastic deformation (SPD) are known to improve the mechanical strength of metals and alloys through microstructural reÀ nement. Although the literature contains a large number of investigations on the effect of À ne-grained microstructures on mechanical behaviour, data on the machinability is almost non-existent. This lack of information motivated the present work, in which the machinability of severely-deformed Grade 2 Ti is assessed in terms of cutting forces and the resulting product surface roughness. The SPD process here employed is Equal Channel Angular Pressing (ECAP), and the results are compared with those obtained on Ti and Ti-6% aluminum-4% vanadium (Ti6-4) alloy, both in the annealed condition. It was observed that the machining of ultraÀ ne-grained Ti in the as-deformed state, requires larger cutting forces than the necessary for the annealed material, whilst for the alloy, the forces are of the same order. Due to continuous chip generation taking place in commercially pure Ti but not in Ti6-4, the passive component of the cutting force and the average surface roughness of the À ne grained material are higher. Finally, whilst both annealed and À ne grained Ti wear the tool by an attrition mechanism, titanium alloy machining promotes only adhesion over the tool edge.

KEY WORDS: Severe plastic deformation; Equal channel angular pressing; Titanium; Machinability. © 2012 Brazilian Metallurgical, Materials and Mining Association. Published by Elsevier Editora Ltda.

ORIGINAL ARTICLE

The Machinability of UltraÀ

ne-grained Grade 2 Ti

Processed by Equal Channel Angular Pressing

Armando Ítalo Sette Antonialli

*, Anibal de Andrade Mendes Filho

,

Vitor Luiz Sordi

, Maurizio Ferrante

1_{Graduate Program on Materials Science & Engineering (PPG-CEM), Universidade Federal de São Carlos (UFSCar), São Carlos, Brazil.}

2_{Department of Materials Engineering (DEMa), Universidade Federal de São Carlos (UFSCar), São Carlos, Brazil.}

Manuscript received April 4, 2012; in revised form September 6, 2012.

Este é um artigo Open Access sob a licença de CC BY-NC-ND

Fig. 2 Components measured in a cutting force test[4]

tool life, cutting forces and temperatures, surface À nish-ing, and chip morphology. The smaller the chart area, the better the material machinability. Experimental data are referred to the DTCR of AISI 1045 steel, which is arbitrarily set as equal to the unity.

The DTCR radar chart concept is illustrated in Fig. 1, where Vickers hardness (H_V) and percent elongation (%EL) are plotted on the north-south axis. Ultimate tensile strength – TS (MPa) – and (k.Ŭ.c)-1/2 _{are plotted on the} east-west axis. The last parameter, where k (W/m.K) is the mate-rial thermal conductivity, Ŭ (g/cm3_{) density and c}

p (J/kg.K) the speciÀ c heat at constant pressure, is proportional to the cutting temperature.

Machinability can be compared by different ways and is determined by several diferent types of tests. For ex-ample, besides the main cutting force, the cutting force test measures the passive (F_p) and the feed components (F_f) necessary to machine the material under standardized con-ditions, see Fig. 2 for the identiÀ cation of said forces[4]_{. The} lower the force, the more machinable is the material. An-other currently used machinability criterion is the surface quality of the machined part[2]_.

Ti and its alloys are among the most difÀ cult materials to machine, mainly because of the metal reactivity at medium to high temperatures, from which a tendency to weld to the tool while machining leads to chipping and premature tool failure. Additionally, its low heat conductivity increases the temperature at the tool-workpiece interface. Finally, the low elastic modulus of Ti allows relatively large deÁ ections of the workpiece[5]_{, which affect adversely the tool life.} Rapid Á ank wear, crater wear, nose wear and notching de-fects, here summarized in Fig. 3[6]_{, are commonly found on} the machining of Ti and its alloys.

It is well known that severe plastic deformation (SPD)-processed materials are characterized by ultraÀ ne grains, leading to signiÀ cantly enhanced mechanical properties, including high cycle fatigue strength[7]_{; thus, such materials} are viewed as “advanced structural and functional materi-als of the next generation of metmateri-als and alloys”[8]_.

Among the available SDP techniques, Equal Channel An-gular Pressing (ECAP) has attracted a great deal of attention by its effectivity in producing bulk ultraÀ ne-grained billets and plates[9] _{– having submicron or even nanosized grains –} which are large enough for commercial applications[10]_{. In} general investigation on SPD processing of Ti or any other material deals with grain size reduction mechanisms, and related changes of mechanical behavior. For instance, Stol-yarov et al.[11] _{studied the inÁ uence of deformation by} roll-ing applied to previously ECAP processed Grade 2 Ti on the microstructure, tensile properties, and microstructural sta-bility; Sabirov et al.[12] _{investigated the fracture behavior,} whilst Zhao et al.[13] _{evaluated the effect of multiple ECAP} passes on the mechanical strength. However, there is a lack of studies directed to what may be called engineering prop-erties: machinability, workability, and corrosion behavior.

The aim of this paper is to evaluate how ECAP process-ing affects the machinability of Grade 2 Ti, an important piece of knowledge in the context of the evolution of SPD technology from laboratory to commercial scale.

2. Experimental Procedures

Three sets of samples, each one with three workpieces, were submitted to cutting force and surface À nish tests in order to assess the inÁ uence of SPD over machinability: (i) Fig. 3 Typical tool wear features found on titanium machining[6]

Rake face

End

clearance

face

1 - Flank wear 2 - Nose wear 3 - Notch wear 4 - Crater wear

Nose

radius

Flank

face

Fig. 1 Radar chart depicting difÀ cult-to-cut rating (DTCR) conception[3]

Cutting temperature

Cutting forces

TS (MPa)

Chip breakability

%EL

(HV)

(k.

ȡ

.c)

power is very small when compared to cutting power, as both cutting force feed component and feed rate are small. On the other hand, the main cutting force, which is the largest of all, may be calculated by Eq. (1)[17]_{, where k}

s is the speciÀ c cutting pressure (usually available in tool manufacturers cat-alogues for several materials and À xed tool geometry), while feed (f = 0.042 mm/rev) and depth of cut (a_p = 0.5 mm) are the chosen cutting parameters[17]_{. Therefore, knowing the} main cutting force it is possible to determine the speciÀ c cutting pressure for a given material in any tooling setup.

F

α

k

ȋ

a

Ǥ

f

ȌαͲǤͲʹͳ

After the machining tests, each sample was subjected to surface roughness measurement using a Time TR200 portable rugosimeter. Arithmetical mean roughness (Ra) was analyzed using Gauss À lter and keeping the evaluation length equal to À ve times the cut-off, chosen according to ISO 4287[18]_.

Additionally, both Á ank and rake faces of the worn tool edges were examined in a Philips XL30 Field-Emission Envi-ronmental Scanning Electron Microscope (ESEM-FEG) À tted with Energy-Dispersive Spectroscopy (EDS) attachments, in order to detect differences regarding the main wear mech-anisms acting in each test.

3. Results and Discussion

Fig. 6 shows data on cutting force obtained on Grade 2 Ti, samples 0X, 4XH and on the Ti6-4 alloy, sample #5. It can be seen that ECAP processing of the ultraÀ ne-grained Ti al-Fig. 5 Setup for cutting forces measurements

(1)

Table 1 Tooling and cutting parameters employed in the machining tests

Tooling

Holder DCLNR 2525M 12

Insert CNMG 12 04 12-QM S05F

Parameters

Cutting speed 63 m/min

Feed 0.042 mm/rev

Depth of cut 0.5 mm

Grade 2 Ti in the annealed condition (0X – coarse grained); (ii) same material after four ECAP passes performed at 300°C (4XH – ultraÀ ne grained)[14]_{; and (iii) annealed Ti-6%} aluminum-4% vanadium (Ti6-4) (#5 – coarse grained). As for DTCR[3]_{, Fig. 4 shows that from literature data the} param-eter resulted values equal to 3.14 for the alloy[15] _{and 1.70} and 1.90 for samples 0X and 4XH[14]_{, respectively. Therefore,} it is expected that after SPD machinability will be reduced, although by a small extent.

Machining tests were held on a Romi S20 conventional turning machine, using a 95° entering angle and î6° rake angle tool holder (manufacturer's code DCLNR 2525 M 12) and ISO HC S05-S15 grade CVD-coated (TiN+Al₂O₃+TiCN) car-bide inserts having 1.2 mm nose radius (manufacturer's code CNMG 12 04 12-QM S05F). Following the manufacturer's in-structions[16]_{, cutting speed, feed, and depth of cut were} kept, respectively equal to, 63 m/min, 0.042 mm/rev, and 0.5 mm as shown in Table 1. No cutting Á uid was employed.

All measurements of cutting forces were made setting the tool on a Kistler 9257B multicomponent dynamometer connected to a Kistler 5019B charge ampliÀ er and a NI PCI-6025E multifunctional data acquisition board. The NI Lab-VIEW 8.5 software was used to process signals received by the computer (Fig. 5), where F_p, F_f, and F_care, respectively, the passive, the feed, and the main cutting force compo-nents measured over the tool. Acquisition rate was À xed as 2 kHz, and a 1 kHz low pass À lter was applied.

The passive component of cutting force, commonly the smallest among all components, provides no load over ma-chining power, as it is orthogonal to the work plane. Feed

Fig. 4 DifÀ cult-to-cut rating (DTCR) for 0X, 4XH and #5 titanium

charge amplifier

dynamometer

Ff

Fc

Fp

sample

tool

acquisition

board

PC

Fig. 6 Cutting forces in 0X, 4XH, and #5 turning

(k.

ȡ

.c)

%EL

TS

HV

most doubled the main cutting force (F_c) required for the tests, reaching values close to that measured on alloy. Also, the feed force (F_f) required for the 4XH sample was three times larger than the value measured on sample 0X and of the same order than that of sample #5. Finally, the passive component of the cutting force (F_p) was even more affected by ECAP: it was four times larger than the force measured on sample 0X and twice as much as the value measured on the #5 sample. The Student t test, which is appropriated to compare two treatment means, statistically conÀ rms the equality between F_f and F_c components for 4XH and #5 and the superiority of the former sample in terms of F_p (Fig. 7), where t₀ is the test statistic[19]_.

From the above results it is clear that SPD decreases Ti machinability, as all cutting force components increase. However, this is in contrast with the radar chart DCTR values, which anticipate a relative equality between the coarse and the ultraÀ ne-grained sample (Fig. 4). In fact, the machinability of the ECAP-deformed Ti is even slightly lower than that of the alloy, again in desagreement with the radar chart data.

The large value of the cutting force passive component of ultraÀ ne-grained Ti may be ascribed to the type of chip generated, a feature which depends on the material duc-tility. The more ductile Ti was observed to produce a con-tinuous chip – a condition leading to huge shear stress over the rake surface of the tool – while the discontinuous chip produced by #5 alloy sample meant that such effect was absent. The SPD effect over the main cutting force reÁ ects on the speciÀ c cutting pressure (k_s), which is calculated by

Eq. (1) and plotted in Fig. 8. Regardless of its dependence on cutting thickness, this parameter may be used to esti-mate the cutting forces when employing other machining parameters, since the same tool geometry (95° entering and î6° rake angle) is maintained. It must be pointed out that k_s, when calculated for the coarse-grained Ti and the annealed Ti6-4, exhibits a very different value from that given in the tools suplier data[16]_{, which is known to assume} a positive tooling setup.

Data on mean surface roughness (Ra) of all specimens after the turning tests are shown in Fig. 9. For both an-nealed and ECAPed titanium, roughness data is highly scat-tered, with a Ra of about 2.00 —m, whilst for the alloy Ra is 1.00 —m and was obtained from a collection of less scat-tered data. The difference of surface quality is quite pro-nounced, corresponding respectively to Grades N8 and N7, as deÀ ned by the ISO 1302 Standard[20]_.

This difference on surface roughness may be attributed to different chip types obtained from the different mate-rials during the turning tests: coarse-grained or ultraÀ ne-grained Ti produced reasonably ductile continuous chips, that may scratch newly machined surface, thus causing damage to the samples surface, whilst the alloy produced a discontinuous type.

Fig. 10 shows SEM micrographs of the worn tool Á ank face after machining tests of annealed Ti (0X). Strong ad-hesion of workpiece material (rough Ti layers) is aparent over the exposed tool substrate material – W and Co on

Fig. 7 Student t test for cutting force components on 4XH and #5 machining

Fig. 10 Tool Á ank face after 0X samples machining Fig. 8 SpeciÀ c cutting pressure for 0X, 4XH, and #5

Fig. 15 Tool rake face after #5 samples machining Fig. 13 Tool rake face after 4XH samples machining

lighter areas. This is a typical indication that attrition is the main acting wear mechanism. When turning at relatively low speeds, the workpiece material usually builds on the cutting edge and tool fragments of microscopic size that are periodically torn from it, leading to large tool wear[1]_. Sequentially (Fig. 11), chip traces of Ti are often found on the tool rake surface, as the workpiece material is known to be very ductile.

SEM photographs taken on Á ank and rake faces of the tool used on 4XH turning experiments can be seen in Figs. 12 and 13, respectively. They show some typical attrition mechanism features, that is, tool material (W, Co) exposure and workpiece material (Ti) stuck on it, besides chip adher-ence. From Figs. 10 and 11, the same conclusions can be extended to coarse-grained annealed Ti.

Finally, Fig. 14 shows the Ti alloy worn tool Á ank face after the turning test. Adhesion of the workpiece material (Ti, Al, V) is apparent but there was no exposure of tool sub-strate, as the tool coating elements (Ti, Al, and O) can still be detected by EDS on the tool surface. This indicates that the tool wear mechanism acting on the alloy machining differs from that acting on both annealed and ECAP-deformed Ti. Klocke[4] _{deÀ nes adhesion as the formation of atomic bonds} between tool and workpiece. It is considered as a conse-quence of diffusion and mechanical snagging, which is differ-ent from attrition mechanism. As the hardness is higher than that of the metal, the tool-workpiece interface temperature is also higher, a condition which stimulates atomic diffusion between tool coating and the workpiece, both constituted by the same material.

Differently from what was observed on 0X and 4XH sam-ples, the alloy tool rake surface shows almost no chip trail (Fig. 15), since discontinuous chips are formed and as a con-sequence, there were no scratches over the tool surface. Said micrograph reinforces earlier assumptions put forward when discussing the results concerning the passive component of the cutting force and the corresponding surface roughness. That is, continuous chips generated when machining the 4XH sample100 are responsible for the higher F_p and R_a than those measured on sample #5.

Fig. 11 Tool rake face after 0X samples machining

Fig. 12 Tool Á ank face after 4XH samples machining

4. Conclusions

ECAP processing of Grade 2 Ti sharply decreases machin-ability; indeed, the main cutting force required on turn-ing these specimens reach the same level measured on the Ti6-4 alloy, and is much larger than that required by the coarse-grained Ti. Aditionally, the speciÀ c cutting pressure of SPD titanium and of the alloy are of the same order of magnitude. Because of the continuous chip generated when machining the 4XH sample, the passive component of the cutting force is also higher than that measured on Ti6-4 al-loy. By the same reason, the quality of the turned surfaces of both annealed and ECAP-deformed Ti was lower than that exhibited by the Ti6-4 material turned surface.

Attrition was found to be the main wear mechanism act-ing on Grade 2 Ti, regardless of its condition, since a strong adhesion of workpiece material followed by tool particles pull out was observed. On the other hand, adhesion without substrate material exposure took place when machining the Ti alloy.

Acknowledgements

The authors would like to thank all staff from Unicamp Machining Materials Laboratory for the support during the tests, and IFSP - São Carlos for use of their equipment. Thanks also to Sandvik Coromant for providing the cutting tools, Capes for granting a PhD scholarship and Fapesp for À nancial support.

References

Trent EM, Wright PK. Metal Cutting. 4.ed. Woburn: Butterworth-1.

Heinemann; 2000.

Zimmerman C, Boppana SP, Katbi K. In: ASM (ed.).

2. Metals

Handbook. v.16. 10.ed. Metals Park: ASM International; 1990, p. 639–47.

Yamane Y, Sekyia K, Narutaki N, Ezugwu EO. DifÀ culty in ma-3.

chining calculated from mechanical and thermal properties of difÀ cult to cut material. In: JSPE (org.). Proceedings of the 8th International Conference on Progress of Machining Technology. Matsue, Japan; 2006, p.497–501.

Klocke F. Manufacturing Processes 1: Cutting. Berlin: Springer; 4.

2011.

Chandler HE. Machining of Reactive Metals. In: ASM (ed.). 5.

Metals Handbook. v.16. 10.ed. Metals Park: ASM International; 1990, p. 844–57.

Ezugwu EO. Improvements in the machining of aero-engine 6.

alloys using self-propelled rotary tooling technique. J Mater Process Technol 2007; 185(1-3):60–71.

Semenova IP, Valiev RZ, Yakushina EB, Salimgareeva GH, 7.

Lowe TC. Strength and fatigue properties enhancement in ul-traÀ ne-grained Ti produced by severe plastic deformation. J Mater Sci 2008; 43(23-24):7354–9.

Valiev RZ. Paradoxes of Severe Plastic Deformation. In: 8.

Zehetbauer M, Valiev RZ (eds.). Nanomaterials by Severe Plastic Deformation. Weinheim: Wiley; 2004, p. 109–17. Stolyarov VV, Zhu YT, Alexandrov IV, Lowe TC, Valiev RZ. InÁ u-9.

ence of ECAP routes on the microstructure and properties of pure Ti. Mater Sci Eng A 2001; 299(1-2):59–67.

Valiev RZ, Langdon TG. Developments in the use of ECAP pro-10.

cessing for grain reÀ nement. Rev Adv Mater Sci 2006; 13(1): 15–26.

Stolyarov VV, Zeipper L, Mingler B, Zehetbauer M. InÁ uence of 11.

post-deformation on CP-Ti processed by equal channel angular pressing. Mater Sci Eng A 2008; 476(1-2):98–105.

Sabirov I, Valiev RZ, Semenova IP, Pippan R. Effect of equal 12.

channel angular pressing on the fracture behavior of commer-cially pure titanium. Metall Mater Trans A 2010; 41(3);727–33. Zhao X, Yang X, Liu X, Wang X, Langdon TG. The processing of 13.

pure titanium through multiple passes of ECAP at room tem-perature. Mater Sci Eng A 2010; 527(23):6335–9.

Mendes Filho AA, Sordi VL, Ferrante M. Optimization of tensile 14.

strength and ductility of Grade 2 Ti, conditioned by Severe Plas-tic Deformation. In: Trans Tech Publications (org.). Materials Science Forum. Nanjing, China; 2011, p. 803–8.

Collings EW. Physical Metallurgy of Titanium Alloys. In: Boyer R, 15.

Welsch G, Collings EW (eds.). Materials properties handbook: tita-nium alloys. Materials Park: ASM International; 1994, p. 1–122. Sandvik Coromant. Manual Técnico de Usinagem. São Paulo: 16.

Sandvik; 2005.

Diniz AE, Marcondes FC, Coppini NL. Tecnologia da Usinagem 17.

dos Materiais. 6.ed. São Paulo: Artliber; 2008.

ISO. ISO 4287:1997: Geometrical Product SpeciÀ cation (GPS) – 18.

Surface Texture: ProÀ le Method – Terms, deÀ nitions and surface texture parameters, 1997.

Montgomery DC. Design and analysis of experiments. 5.ed. New 19.

York: Wiley; 2000.

ISO. ISO 1302:2002: Geometrical Product SpeciÀ cations (GPS) 20.

– Indication of surface texture in technical product documenta-tion, 2002.