j m a t e r r e s t e c h n o l . 2 0 1 4;3(4):338–343

Available online at www.sciencedirect.com

www. jmrt .com.br

Original Article

Electron backscatter diffraction (EBSD)microstructure evolution in HPT copper annealedat a low temperature�

Alexander P. Zhilyaeva,b,∗, Semyon N. Sergeevb, Terence G. Langdona

a Materials Research Group, Faculty of Engineering and the Environment, University of Southampton, Southampton, United Kingdomb Institute for Metals Superplasticity Problems, Khalturina, Ufa, Russia

a r t i c l e i n f o

Article history:

Received 6 June 2014

Accepted 25 June 2014

Available online 28 July 2014

Keywords:

Hardness

High-pressure torsion

Homogeneity

a b s t r a c t

Detailed EBSD analysis was performed on copper specimens processed by high-pressure

torsion at P = 6 GPa for one whole turn and subsequently annealed at a temperature of

100 ◦C for 15, 30 and 60 min. The basic microstructural parameters (mean grain size, GB

statistics, microtexture) were evaluated in the mid-radius areas of the HPT disks. Micro-

hardness of all samples was measured across the two diameters and interlinked to the

microstructures observed. Small but noticeable changes of microhardness in HPT copper

after annealing were detected. The changes were interlinked to microstructural parameters

acquired by EBSD. The relationships obtained are discussed in terms of the microstructure

and microtexture evolution during low temperature annealing.

Severe plastic deformation © 2014 Brazilian Metallurgical, Materials and Mining Association. Published by Elsevier

Editora Ltda.

in pure copper and its alloys in order to get the optimum

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

1. Introduction

The processing of ultrafine-grained (UFG) and nanostructuredmetallic materials using severe plastic deformation (SPD) [1]is a new and promising method of enhancing the propertiesof metals and alloys for advanced structural and functionalapplications [2,3]. Traditionally, there have been two maintechniques for producing UFG materials using either equal-

channel angular pressing (ECAP) [4] or high-pressure torsion(HPT) [5,6] but other techniques are now available such asaccumulative roll bonding (ARB) [7], multiaxial forging [8],

� Paper presented in the form of an abstract as part of the proceedings21st to 25th 2014.

∗ Corresponding author.E-mail addresses: AlexZ@anrb.ru, A.Zhilyaev@soton.ac.uk (A.P. Zhil

http://dx.doi.org/10.1016/j.jmrt.2014.06.0082238-7854/© 2014 Brazilian Metallurgical, Materials and Mining Associa

Este é u

twist extrusion [9], plain strain machining (PSM) [10] and oth-ers [11]. SPD processing is attractive because the straining ispractically unlimited due to the unchanging sample geometryand shape. However, there is a general tendency for a satura-tion in grain refinement for high melting temperature metals[12] and recovery (and even recrystallization) for low meltingmaterials [13] processed by continuing SPD methods.

There is a considerable interest in creating UFG structure

of the Pan American Materials Conference, São Paulo, Brazil, July

yaev).

combination of low wear rate and high conductivity. How-ever, pure copper has some distinct drawbacks, such as lowstrength and low thermostability, which tends to restrict its

tion. Published by Elsevier Editora Ltda.

m artigo Open Access sob a licença de CC BY-NC-ND

o l . 2

atapofHdDnNmeHait2

pgartd

HiimamtsrH3

j m a t e r r e s t e c h n

pplications. There are numerous reports on the applica-ion of SPD for grain refinement of pure copper by ECAPnd HPT where the thermostability of ultrafine-grained cop-er was analyzed. Probably the first comprehensive reportsn copper were published in 1999 for ECAP [14] and 2000or HPT [15]. In the latter report, the microhardness of anPT disk of 98.5% purity Cu (P = 5 GPa, N = 5 turns) dropsown at an annealing temperature of about 180–200 ◦C. FromSC data with a heating rate of 40◦/min, the most pro-ounced peak was detected for the HPT sample strained for

= 1 turn at 220 ◦C, which corresponds to the microhardnesseasurements [15]. A similar temperature (∼190 ◦C) of recov-

ry was also reported for higher purity HPT copper (99.9%).igh purity cooper (99.99%) [16] shows similar behavior whennnealing at 134, 269 and 405 ◦C (0.3, 0.4 and 0.5 of melt-ng point) for 1 h. The microhardness drops down from 130o 80 Hv as the annealing temperature increases from 134 to69 ◦C.

There are some reports on low thermostability of UFG cop-er at room temperature in which recovery processes andrain growth have been detected in a period of one monthnd one year [17]. Other reports [18] show no significant grainefinement in copper subjected to rolling at liquid nitrogenemperature suggesting that recovery processes take placeuring rolling.

In practice, the simplest way to achieve UFG copper is byPT, which permits the processing of disks suitable for exper-

ments on highly strained materials. In an earlier report [19],t was demonstrated that there is a slight increase of Vickers

icrohardness in HPT copper subjected to low-temperaturennealing. Careful monitoring of the microhardness andicrostructure in HPT aluminum and copper stored at room

emperature for long periods of time [20] has not revealed any

ignificant changes of these parameters. Thus, the presenteport was undertaken to study microstructure evolution ofPT copper specimens subjected to annealing at 100 ◦C for 15,0 and 60 min.

25 µm

a b

Fig. 1 – Microstructure (a) and micorotext

0 1 4;3(4):338–343 339

2. Experimental materials and procedure

Disks of copper were used as the starting material. Thematerial was purchased from Goodfellow Cambridge Ltd.,Huntingdon, UK, and the typical chemical composition wasgiven as (in ppm): Ag 500, Bi < 10, Pb < 50, O 400, other metals<300. The HPT specimens were in the form of disks havingdiameters of 10 mm and thicknesses of about 1 mm. Thesedisks were processed at room temperature by HPT for atotal of N = 1 turn under an applied pressure of P = 6 GPa.The processing was conducted under quasi-constrained con-ditions where there is a small outflow of material aroundthe periphery of each disk during processing. Parts of theprocessed specimens were annealed at 100 ◦C for a period of15, 30 or 60 min. During annealing, the HPT disks were placedon the flat surface of a thermocouple and hence the temper-ature was stable to within ±2◦. Prior to all measurements,the specimens were mechanically polished on 1000 grit SiCpaper followed by a final polish using a diamond suspensioncontaining monocrystalline diamond with a size of 3 �m withsubsequent electro-chemical polishing at room temperatureusing an electrolyte of HNO3:CH3OH = 1:3 with a voltage of 10 V.These samples were employed for microhardness measure-ments and electron backscatter diffraction (EBSD) analysis.

The microhardness was measured with steps of 0.5 mmalong the diameters of the disks using an FM-300 testerequipped with a Vickers indenter using a load of 100 gf anda dwelling time of 15 s.

The EBSD analysis was performed using a TESCAN MIRA3LMH FEG scanning electron microscope equipped with anEBSD analyzer “CHANNEL 5”, and a rectangular grid with scanstep of 50 nm was used. The EBSD analysis was performedfor regions located near the center of the disk, near the mid-

radius (2.5 mm from the center) and near the edge (∼4.5 mmfrom the center.) The acquired data were subjected to standardclean-up procedures involving a grain tolerance angle of 5◦ and

001 001

TD TD

RDRD

111

RD

TD

Texture name: harmonic: L=16, HW=5.0Calculation method: harmonic series expansionSeries rank (1): 16Gaussian smoothing: 5.0єSample symmetry: triclinic

max=4.1933.3022.6002.0481.6131.2701.0000.787

ure (b) of Cu specimen prior to HPT.

340 j m a t e r r e s t e c h n o l . 2 0 1 4;3(4):338–343

Misorientation angle

Num

ber

Fra

ctio

n

Misorientation angle [degrees]

0.00

0.02

0.04

0.06

0.08

0.10

10 20 30 40 50 602 µm

a b

ution (b) at mid-radius of Cu disk after HPT (P = 6 GPa, N = 1).

140

150

130

120

110

Annealing time (min)

Inte

grat

ed H

v

0 10 20 30 40 50 60

Fig. 3 – Integrated microhardness at x = r/2 as a function of

Fig. 2 – Microstructure (a) and GB misorientation distrib

a minimum grain size of three pixels. The grain sizes weremeasured using the linear intercept as the distances betweenhigh-angle boundaries with misorientations above 15◦.

3. Experimental results

Fig. 1 presents (a) an initial microstructure and (b) the micro-texture of a copper specimen prior to high-pressure torsion.The average grain size by EBSD was larger than 25 �m withwell-defined grain boundaries and twins in the interior ofgrains. The pole figures in Fig. 1(b) reflect the inherent textureof extruded rods of the primary material.

Fig. 2 shows the inverse pole figure map and grain boundarymisorientation distribution of HPT copper processed under aload of P = 6 GPa for one whole revolution where the EBSD wastaken at the mid-radius of the disk. Significant grain refine-ment with grains elongated along the torsional direction isobserved. The mean grain size by EBSD was about 200–300 nmwhich is a typical grain size for HPT copper. The fractionof low-angle boundaries was below 10%, which is consistent

with earlier reports [3]. Fig. 3 represents the integrated Vickersmicrohardness taken at a distance of r/2, where r is the mid-radius of the HPT disk. A slight increase in Hv is observed forHPT copper annealed for 15 min. Although it is in the range of

60.0 40.0

2-Theta

Cu

200.0

400.0

Inte

nsity

1/2 [C

ount

1/2 ]

Fig. 4 – X-ray profile

annealing time.

the experimental error, the gain in microhardness is about 4%and it was systematically detected.

An X-ray analysis of HPT-processed copper in Fig. 4 gives asize of coherent domains of 287.5 ± 6.3 nm, which is in goodagreement with the EBSD data. Fig. 5 shows a Taylor factor map

80.0 100.0

[degrees]

of HPT copper.

j m a t e r r e s t e c h n o l . 2 0 1 4;3(4):338–343 341

2 µm

2 µm

2 µm5 µm

A B

C

D

Fig. 5 – Taylor factor map of HPT copper: (a) initial, and after annealing at T = 100 ◦C for (b) 15 min, (c) 30 min and (d) 60 min.

2 µm

2 µm

2 µm 2 µm

a b

c d

Fig. 6 – Kernel average misorientation of HPT copper: (a) initial and after annealing at T = 100 ◦C for (b) 15 min, (c) 30 min and(d) 60 min.

342 j m a t e r r e s t e c h n o l . 2 0 1 4;3(4):338–343

Table 1 – Experimental parameters of HPT copper subjected to annealing at 100 ◦C for 15, 30 and 60 min.

Specimen d (�m) KAM (◦) Taylor factor LAB (%) d (nm) Microstrain‹�2›1/2 (×10−3)

Dislocation density(m−2, ×1014)

EBSD EBSD EBSD EBSD X-ray X-ray EBSD X-ray

HPT Cu 0.33 0.537 3.318 20.4 193.8 2.125 7.322 1.484HPT Cu + ann. 15′ 0.36 0.553 3.295 23.1 185.0 2.150 7.540 1.573HPT Cu + ann. 30′ 0.39 0.563 3.359 21.1 210.0 2.300 7.677 1.482

r

HPT Cu + ann. 60′ 0.34 0.549 3.317 20.1

for HPT copper after processing in Fig. 5(a) and after annealingat 100 ◦C for 15 min in Fig. 5(b), 30 min in Fig. 5(c) and 60 min inFig. 5(d). The average value of the Taylor factor (TF) decreasesslightly from 3.318 for HPT copper to 3.295 for the HPT diskannealed for 15 min. For the sample annealed for 30 min, theTF increases to a value of 3.359 and then it decreases again to3.317 for the specimen annealed for 60 min. These variationsof the Taylor factors are not in agreement with changes inthe Vickers microhardness. The Kernel average misorientation(KAM) maps for copper specimens are shown in Fig. 6. Theaverage KAM also increases for the specimens annealed for15 and 30 min and then slightly decreases for the HPT copperannealed for 60 min.

Table 1 provides a comprehensive summary of all experi-mental data obtained in this investigation. The area weightedgrain size obtained from EBSD shows a gradual increase withannealing time. Also, there is a noticeable increase in the frac-tion of low-angle boundaries (LAB) for the specimen annealedfor 15 min. X-ray analysis revealed a small decrease in thecoherent domain size from 193.8 nm (HPT copper) to 185 nm(after annealing for 15 min) and a gradual increase in themicrostrain level.

4. Microstructure evolution during lowtemperature annealing

In pure metals, there are a limited number of factors influenc-ing hardening: the Hall–Petch relationship and the dislocationhardening and texture (corresponding to a change in the Tay-lor factor). The texture can change during significant changingof the microstructure as in recrystallization and grain growth.During low temperature, annealing at 100 ◦C it is difficult toexpect significant changes in texture and the average valuesof the Taylor factor for all specimens clearly support this. Onthe basis of the experimental parameters, it is possible to eval-uate the dislocation density either using the average KAM(� = �/(b·h), where � is the average angle in radians, b is theBurgers vector and h is the step size of 50 nm in EBSD), or usingthe microstrain level measured by X-rays (� = 2

√3‹�2›1/2/(b·d)),

where ‹ε2›1/2 is the microstrain and d is the coherent domainsize. The two final columns in Table 1 represent the dislocationdensities calculated on the basis of X-ray analysis and EBSDexperiments, respectively. The dislocation density calculatedfrom KAM corresponds to the geometrically necessary dislo-

cations (GND) and it is about 3 times higher than the densityof the statistically stored dislocations (SSD) estimated fromX-ray analysis. Both values show a slight increase in the Cuspecimen annealed for 15 min.

200.0 2.250 7.486 1.522

5. Summary

Integrated microhardness measurements at the mid-radiusposition show a systematic increase for the HPT copper sub-jected to low temperature annealing at 100 ◦C for 15 min. Themost correlated parameters that may be responsible for thischange are the fraction of low-angle grain boundaries and thesize of the coherent domains.

Conflict of interest

The authors declare no conflicts of interest.

Funding

This work was supported in part by the European ResearchCouncil under ERC Grant Agreement No. 267464-SPDMETALS(APZ & TGL).

e f e r e n c e s

[1] Valiev RZ, Islamgaliev RK, Alexandrov IV. Bulknanostructured materials from severe plastic deformation.Prog Mater Sci 2000;45:103–89.

[2] Valiev RZ, Sabirov I, Zhilyaev AP, Langdon TG. Bulknanostructured metals for innovative applications. JOM2012;64:1134–42.

[3] Valiev RZ, Zhilyaev AP, Langdon TG. Bulk nanostructuredmaterials: fundamentals and applications. Hoboken, NewJersey: Wiley; 2014. pp. 450.

[4] Valiev RZ, Langdon TG. Principles of equal-channel angularpressing as a processing tool for grain refinement. ProgMater Sci 2006;51:881–981.

[5] Zhilyaev AP, Langdon TG. Using high-pressure torsion formetal processing: fundamentals and applications. ProgMater Sci 2008;53:893–979.

[6] Pippan R, Scheriau S, Taylor A, Hafok M, Hohenwarter A,Bachmaier A. Saturation of fragmentation during severeplastic deformation. Annu Rev Mater Res 2010;40:319–43.

[7] Tsuji N, Saito Y, Lee SH, Minamino Y. ARM (AccumulativeRoll Bonding) and other new techniques to produce bulkultrafine grained materials. Adv Eng Mater 2003;5:33844.

[8] Salishchev GA, Valiakhmetov OR, Galeyev RM. Formation ofsubmicrocrystalline structure in the titanium VT8 and itsinfluence on mechanical properties. J Mater Sci

1993;28:28982902.

[9] Beygelzimer Y, Orlov D, Varyukhin V. A new severe plasticdeformation method: twist extrusion. In: Zhu YT, LangdonTG, Mishra RS, Semiatin SL, Saran MJ, Lowe TC, editors.

o l . 2

j m a t e r r e s t e c h n

Ultrafine grained materials II. Warrendale, PA: TMS; 2002. p.297304.

[10] Shankar MR, Chandrasekar S, King AH, Compton WD.Microstructure and stability of nanocrystalline aluminum6061 created by large strain machining. Acta Mater2005;53:4781–93.

[11] Zhu YT, Valiev RZ, Langdon TG, Tsuji N, Lu K. Processing ofnanostructured metals and alloys via severe plasticdeformation. MRS Bull 2010;35:977–81.

[12] Zhilyaev AP, Gimazov AA, Langdon TG. Recent developmentsin modelling of microhardness saturation during SPDprocessing of metals and alloys. J Mater Sci 2013;48:4461–6.

[13] Xu C, Horita Z, Langdon TG. The evolution of homogeneity inprocessing by high-pressure torsion. Acta Mater2007;55:203–12.

[14] Komura S, Horita Z, Nemoto M, Langdon TG. Influence of

stacking fault energy on microstructural development inequal-channel angular pressing. J Mater Res 1999;14:4044–50.

[15] Jiang HG, Zhu YT, Butt DP, Alexandrov IV, Lowe TC.Microstructural evolution, microhardness and thermal

0 1 4;3(4):338–343 343

stability of HPT-processed Cu. Mater Sci Eng2000;A290:128–38.

[16] Cizek J, Prochazka I, Cieslar M, Kuzel R, Matej Z, Cherkaska V,et al. Influence of ceramic nanoparticles on grain growth inultra fine grained copper processed by high pressure torsion.Phys Stat Sol C 2007;4:3587–90.

[17] Schafler E, Kerber MB. Microstructural investigation of theannealing behavior of high-pressure torsion (HPT) deformedcopper. Mater Sci Eng 2007;A462:139–43.

[18] Konkova T, Mironov S, Korznikov A, Semiatin SL.Microstructure instability of cryogenically deformed copper.Scripta Mater 2010;63:921–4.

[19] Zhilyaev AP, Shakhova I, Belyakov A, Kaibyshev R, LangdonTG. Wear resistance and electroconductivity in copperprocessed by severe plastic deformation. Wear2013;305:89–99.

[20] Zhilyaev AP, Shakhova I, Belyakov A, Kaibyshev R, LangdonTG. Effect of annealing on wear resistance andelectroconductivity of copper processed by high-pressuretorsion J. Mater Sci 2014;49:2270–8.