Tp

Xa

b

a

ARRAA

KTPIM

1

iabcrawabih

a[fltmcsmt

0h

Materials Science and Engineering A 550 (2012) 131– 137

Contents lists available at SciVerse ScienceDirect

Materials Science and Engineering A

jo ur n al hom epage: www.elsev ier .com/ locate /msea

he investigation on the unstable flow behavior of Ti17 alloy in �+�hase field using processing map

iong Maa, Weidong Zenga,∗, Kaixuan Wanga,b, Yunjin Laia,b, Yigang Zhoua

State Key Laboratory of Solidification Processing, Northwestern Polytechnical University, Xi’an 710072, ChinaWestern Superconducting Technologies Co. Ltd., Xi’an 710016, China

r t i c l e i n f o

rticle history:eceived 15 February 2012eceived in revised form 26 March 2012ccepted 13 April 2012vailable online 21 April 2012

a b s t r a c t

In this study, five existing instability criteria (Prasad’s, Murty’s, Sematin’s, Gegel’s and Malas’s criterion)were employed to delineate the unstable flow regions for Ti17 alloy during hot deformation. Experi-mental stress–strain data obtained from isothermal hot compression tests in the temperature range of780–860 ◦C and strain rate range of 0.001–10 s−1 were utilized to develop the instability maps. Thesedomains were validated through detailed microstructure observation. The material exhibited stable flow

eywords:i17 alloyrocessing mapnstability criteria

icrostructure

at lower strain rates while adiabatic shear band and flow localization occurred at higher strain rates(≥1 s−1). It was observed that the maps at strain of 0.9 developed by Prasad’s and Murty’s criteria havea good ability for predicting unstable flow at high strain rates and entire temperatures. However, theSemiatin’s criterion under-predicts the instability regions in the test temperatures and strain rates. Onthe contrary, the Gegel’s and Malas’s criteria over-predict the instability domains, especially for the low

strain rate region.

. Introduction

Titanium alloys have a wide range of applications, especiallyn the aerospace industry [1,2]. Ti17 (Ti–5Al–2Sn–2Zr–4Mo–4Cr)lloy has been extensively accepted in the aerospace applicationsecause of its attractive combination of properties such as high spe-ific strength, excellent fracture toughness and good heat/corrosionesistance [3]. However, the alloy is rather difficult to deform into

complex shape because of its poor formability. In addition, its hotorkability is limited by the narrow range of forging temperature

nd the generation of various deformation defects, such as shearands, cracking and flow instability [4,5]. Therefore, the character-

zation of deformation behavior is essential for the optimization ofot forging processes of titanium alloy.

In order to characterize the hot forming processes, a series ofpproaches and methods have been proposed. Front and Ashby6] introduced a deformation map showing the area of dominantow mechanisms in the plot of normalized stress vs. deformationemperature. The map can be used to identify the dominant defor-

ation mechanism for a given test condition. However, the mapan be only available to pure materials or sample alloys under

teady-state conditions. Raj [7] suggested the damage nucleationap, presenting the nucleation of deformation damage as a func-

ion of temperature and effective strain rate. Semiatin and Lahoti [8]

∗ Corresponding author. Tel.: +86 29 88494298.E-mail address: zengwd@nwpu.edu.cn (W. Zeng).

921-5093/$ – see front matter © 2012 Elsevier B.V. All rights reserved.ttp://dx.doi.org/10.1016/j.msea.2012.04.045

© 2012 Elsevier B.V. All rights reserved.

and Montheillet et al. [9] considered that plastic instability behaviorwas related to the work-hardening rate and the strain rate sensitiv-ity parameter, and then established a phenomenological criterionby using test methods. Prasad [10,11] has developed the so-calleddynamic material model under which the workpiece is consideredas a dissipater of power. However, Prasad et al. assumed the strainrate sensitivity parameter (m) a constant, which was questionedby some scholars. For this reason, Murty et al. [12,13] proposed amuch stricter plastic instability a criterion in which m is a variable.Gegel [14] derived a novel stability criteria based on the second lawof thermodynamics theory and Liapunov stability equation. Malas[15] replaced � in Gegel criterion with m and then built anothergroup of stability criteria. However, there is no unique instabilitycriterion that can be applied to delineate the stable and unstableregion for all kinds of materials, designer needs to establish a suit-able instability criterion or examine the applicability of the existingtheories for the intended materials based on detailed microstruc-tural observations [16,17].

The present work was mainly focused on the characterization ofthe flow instability during hot working of Ti17 titanium alloy basedon the various instability criteria. Further, the flow instability wasvalidated for Ti17 alloy forging by microstructural observation.

2. Experimental procedure

The material used for compression test was Ti17 alloy. Thechemical composition is given in Table 1, and the initial microstruc-ture is a typical lamellar microstructure with 15–25 �m length and

132 X. Ma et al. / Materials Science and Engineering A 550 (2012) 131– 137

Table 1Chemical composition (in wt.%) of Ti17 titanium alloy.

05

∼t

tisifsaadmpTttoJ

3c

utdrcop

3

[dt

Al Cr Mo Sn Zr Fe

5.02 3.93 3.88 2.37 1.95 0.

1 �m thick � lamellas as shown in Fig. 1. The � transus tempera-ure for this material was calculated to be approximate 905 ◦C.

To investigate the hot workability and microstructure evolu-ion in the �+� phase field, compression tests were performedn the temperature ranges of 780–860 ◦C with 20 ◦C interval andtrain rates of 0.001, 0.01, 0.1, 1 and 10 s−1. Compression test spec-mens of 12 mm in height and 8 mm in diameter were machinedrom the as-received materials and both ends were grooved forufficient retention of the glass lubricants during each test sos to reduce the friction. All the tests were performed in argontmosphere under constant strain rate and uniform heating con-itions (300 s) using a Gleeble-1500 thermo-mechanical simulatorachine. The high reduction of specimens was 60%. After the com-

ression, the specimens were cooled in water to room temperature.o observe the evolution of microstructure during compressionests, the compression-tested specimens were sectioned in parallelo the stress axis, and the metallographic observation was carriedut on an Olympus/PMG3 optical microscope (OM) and the JEOLSM-6390A scanning electron microscope (SEM).

. Determination of instability parameters of variousriteria

In the field of manufacture industry, to obtain a desired prod-ct with the optimum properties at lower cost, it is very importanto control the microstructure evolution in such a manner that itoes not contain any type of flow instabilities. It has been wellecognized that the flow localization is one of the most signifi-ant damaging mechanisms that can limit the intrinsic workabilityf materials. Several researchers have proposed various kinds oflastic instability criteria [8,10,12,14,15].

.1. Phenomenological criterion

Based on the principle of force equilibrium, Semiatin and Lahoti

8] have suggested that the workability parameter ̨ which canescribe the relation between the rate of flow softening andhe strain rate sensitivity parameter is used to predict the flow

Fig. 1. Initial microstructure of Ti17 alloy.

C N H O Ti

0.01 0.01 0.003 0.12 Balance

localization in hot forging of titanium and its alloys. This parameteris defined as:

˛ = 1ε̇

∂ε̇

∂ε

∣∣∣∣T

=(

− 1�

∂�

∂ε

∣∣∣∣ε̇

)1m

= − �

m(1)

where � is the normalized work hardening parameter:

� = − 1�

∂�

∂ε

∣∣∣∣ε̇

(2)

and m is the strain rate sensitivity parameter:

m = ∂ log �

∂ log ε̇(3)

Typically, the value of ̨ has been observed [18] to be more thanfive in titanium alloys for noticeable non-uniform flow. Then, theinstability criterion can be expressed mathematically as:

�1 = 5 − ̨ < 0 (4)

3.2. Prasad’s criterion

Prasad et al. [10] have developed an instability criterion on thebasis of extremum principles in irreversible thermodynamics asapplied to the continuum of large plastic flow. The flow instabilitywill occur during hot forging if

�2(ε̇) = ∂ ln (m/(m + 1))∂ ln ε̇

+ m < 0 (5)

Unstable flow during hot deformation is predicted whendimensionless instability parameter �2(ε̇) becomes negative. Thiscriterion has been applied to a large variety of materials and thepredictions were validated with the microstructural observations[19]. However, it is worth emphasizing that, this criterion is derivedbased on the assumption that m is a constant and flow stress followsthe constitutive equation of power law [16].

3.3. Murty’s criterion

Murty et al. [11,12] considered that strain rate sensitivityparameter m is not a constant, and suggested that the efficiencyof dissipation can be calculated directly by obtaining J from thenumerical integration procedure. This procedure has been carriedout as follows.

The efficiency of power dissipation for a non-linear dissipatermay be expressed as

� = J

Jmax= 2

(1 − 1

�ε̇

∫ ε̇

0

�dε̇

)(6)

Then a new condition for the instability has been established as

�3 (ε̇, T) = 2m

�− 1 < 0 (7)

This instability criterion is valid for any type of constitutiverelation due to a strict theory basis, but it is difficult to take thecalculation of integration.

nd Engineering A 550 (2012) 131– 137 133

3

t

0

s

w

s

[�ca

�

�

�

�

isscr

4

4

s�adTtesstibfiwt

Fig. 2. Typical true stress–true strain curves obtained on Ti17 alloy at 860 ◦C anddifferent strain rates.

higher than 830 ◦C and strain rates faster than 1 s−1 as shown inFig. 3.

Table 2Values of �1 parameter (Eq. (4)) calculated at different temperatures and strain ratesand a strain of 0.9.

Temperature (◦C) Strain rate (s−1)

0.001 0.01 0.1 1 10

780 1.88 2.88 1.13 0.92 0.84

X. Ma et al. / Materials Science a

.4. Gegel’s and Malas’s criteria

Considering a Lyaponov function L (�, s), Gegel [14] suggestedhe following conditions for the stable material flow

< m ≤ 1 (8)

∂�

∂ ln ε̇< 0 (9)

≥ 1 (10)

∂s

∂ ln ε̇< 0 (11)

here s is the temperature sensitivity and can be written as

= 1T

∂(ln �)∂(1/T)

(12)

Similarly, on the basis of the Lyaponov function L (m, s), Malas15] proposed that the parameter m can be used to substitute the

in Gegel’s criterion. Apart from Eqs. (8), (10) and (11), the newondition for material flow is given by the strain rate sensitivity ms

∂m

∂ ln ε̇< 0 (13)

Then the instability criteria can be written as

4 = m < 0 (14)

5 = − ∂�

∂ ln ε̇< 0 (15)

6 = s − 1 < 0 (16)

∂s

∂ ln ε̇< 0 (17)

8 = − ∂m

∂ ln ε̇< 0 (18)

It should be pointed out that in both Gegel’s and Malas’s flownstability criteria, the flow stress with respect to strain rate curvehould be convex in nature and the material should exhibit flowoftening with increase in temperature. Furthermore, as per Malas’sriteria, the strain hardening should increase with decreasing strainate and increasing temperature.

. Results and discussion

.1. Hot deformation behavior of Ti17 alloy

Typical true stress–true strain curves at 860 ◦C and differenttrain rates are shown in Fig. 2, which are representative of the+� deformation behavior. All the curves exhibit flow softening atll strain rates. In addition, the stress–strain curves of Ti17 alloyisplay significant oscillatory flow at higher strain rates (≥1 s−1).hose broad oscillations may be induced by instable deforma-ion. However, it should be noted that different mechanisms canxhibit similar stress–strain behavior and conclusions based on thehapes of the curves may be erroneous. For example, oscillatorytress–strain curves have been observed during dynamic recrys-allization as well as during unstable flow. Flow softening mayndicate lamellar globularization [20], DRX [21], adiabatic heating

and [22], voids and cracks formation [23], etc. In view of this dif-culty, further analysis of the flow stress in terms of its variationith temperature and strain rate should be conducted to evaluate

he mechanisms of hot deformation.

Fig. 3. Instability map developed as per Semiatin’s criterion for Ti17 alloy. Theshaded regime corresponds to flow instability (�1).

4.2. Variant instability maps

Different instability parameters have been formulated (Eqs. (4),(5), (7), (14)–(18)) and plotted in Figs. 3–7. The values of ̨ evaluatedusing the flow stress data of Ti17 alloy at different temperaturesand strain rates and a strain of 0.9 are given in Table 2. It can beseen from Table 2 that the stable flow predicted by the criterionwas at lower strain rate since �1 is greater than zero. At the highstrain rate of 10 s−1, the value of �1 is below zero and the magnitudeincreases with temperature. Therefore, plastic instability and flowlocalization are predicted to occur in this material at temperature

800 2.93 2.85 1.88 0.88 0.54820 2.32 2.06 2.79 2.08 0.23840 6.82 2.67 1.56 0.66 −0.44860 8.18 1.59 0.52 −0.2 −1.95

134 X. Ma et al. / Materials Science and Engineering A 550 (2012) 131– 137

F tour nc

oidppatae

Fc

ig. 4. Instability map developed as per Prasad’s criterion for Ti17 alloy. The conorresponds to flow instability (�2).

The instability map developed by Prasad’s criteria using the databtained from the compression tests at strain of 0.9 is presentedn Fig. 4. The contour number represents the efficiency of powerissipation (�), and the shaded regime corresponds to instabilityarameter (�2). It can be seen that the efficiency of power dissi-ation � increases with the increasing of deformation temperaturend decreasing of strain rate. The peak of � appears in the high

emperature and low strain rate regime, indicating a good work-bility. Also, the criterion predicts a flow instability region in thentire temperature range and strain rate range of 1–10 s−1. It may

ig. 5. Instability map developed as per Murty’s criterion for Ti17 alloy. The contour norresponds to flow instability (�3).

umber represents the efficiency of power dissipation (�), and the shaded regime

be noted that the yielding and oscillation in the flow curves (Fig. 2)and the phenomenological criterion instability predictions are alsoin support of this prediction.

Fig. 5 shows the predicted stable and unstable domains ofmaterial flow in the temperature and strain rate window usingthe Murty’s instability criterion. It could be observed that themap developed as per criterion shows similar domains as that of

Prasad’s flow instability criterion (Fig. 4). However, the domains ofthis instability map are a little wider than the unstable domainsrevealed in Fig. 4.

umber represents the efficiency of power dissipation (�), and the shaded regime

X. Ma et al. / Materials Science and Eng

Fig. 6. Instability map developed as per Gegel’s criteria for Ti17 alloy. The shadedregime corresponds to flow instability (�4, �5, �6, �7).

Gldwb0oc

bsabrFaPtr

Fr

Figs. 6 and 7 depict the instability maps developed based onegel’s and Malas’s criteria. The two maps exhibit almost simi-

ar kind of unstable domains for Ti17 titanium alloy. However, theomains of these instability maps are interconnected and muchider than the unstable domains of Prasad’s and Murty’s insta-

ility map. For example, the unstable domain (at 790–845 ◦C and.001–0.1 s−1) predicted by Gegel’s and Malas’s criteria was notccurred in the instability maps predicted by Prasad’s and Murty’sriteria.

It has been observed that the flow instability maps developedy Prasad’s and Murty’s criteria are similar (Figs. 4 and 5) andhow a broad unstable domain which lied in the entire temperaturend high strain rate (0.316–10 s−1). The unstable domain predictedy Semiatin’s criterion (Fig. 3) was small and matched with theegion (at 830–860 ◦C and 1–10 s−1) of the instability map shown inig. 5. On the other hand, the unstable domain predicted by Gegel’snd Malas’s criteria are much wider than the unstable domains ofrasad’s and Murty’s instability map at high strain rate, and alsohese criteria can be predicted the flow instability at low strainate.

ig. 7. Instability map developed as per Malas’s criteria for Ti17 alloy. The shadedegime corresponds to flow instability (�4, �6, �7, �8).

ineering A 550 (2012) 131– 137 135

4.3. Microstructural validation and comparison of instabilitycriteria

Microstructural study on the deformed specimens was carriedout to confirm the predictions made by the previous criteria, andto make a comparison of the instability criteria.

Due to the strict theoretical inferences and calculations ofMurty’s criterion, validation of all the unstable domains of theMurty’s criterion based instability map would collectively corrob-orate these five instability maps (Figs. 3–7). The micrographs of thespecimens deformed at unstable regions are shown in Fig. 8. Thespecimen deformed at 780 ◦C/10 s−1 (representative area at lowtemperature and high strain rate) shows adiabatic shear band inthe microstructure (Fig. 8a). A similar observation was made byWang et al. [24] which suggested the adiabatic shear band maybe occurred at low strain of 0.36. Another deformation condition(860 ◦C/10 s−1) in the unstable domain manifests flow localiza-tion (Fig. 8b). The microstructure of the specimen deformed at780 ◦C/1 s−1, which is also lied in the unstable domain predictedby the Prasad’s, Murty’s, Gegel’s and Malas’s criteria, reveals asmall number of break-up and kinked � lamellas (Fig. 8c). And thespecimen deformed at 860 ◦C/1 s−1 shows intense kinked and sev-eral globularized � phase (Fig. 8d). The local kinking of � lamellasmay be induced the significant oscillatory flow of the stress–straincurves at high strain rates (≥1 s−1) in Fig. 2. The microstructure ofthe specimen deformed at 800 ◦C/0.001 s−1 which represents thedomain predicted by Gegel’s and Malas’s criteria is shown in Fig. 8e.It can be observed that the microstructure shows homogeneousdistribution of globularized � phase and a few residual lamellar�. Such a microstructure is not bad for the component as men-tioned in Ref. [4]. In contrast, typical microstructure in the stabledomain (860 ◦C/0.001 s−1) exhibits much homogenous distributionof equiaxed � (Fig. 8f), which is desirable and expected to augmentthe material’s performances.

Also it has been observed that instability map developed asper Gegel’s criterion as well as Malas’s criterion shows similar andmuch greater unstable domains than the other criteria. However,these domains are interconnected. Microstructure evolution of thesample deformed at 840 ◦C/0.01 s−1 is shown in Fig. 8g. It may benoted that this hot processing condition falls within the unsta-ble domain of map developed as per Gegel’s and Malas’s criteriabut lie outside the unstable domain of instability map developedby Prasad’s and Murty’s criteria. It can be found from Fig. 8g thatthe microstructure does not indicate any kind of flow instabil-ity.

The comparison of all the instability maps and the microstruc-ture validation shows that Semiatin’s criterion under-predicts theinstability regions in the deformation temperatures and strainrates. The reason for the under prediction of Semiatin’s criterioncould be attributed to the fact that the criterion was just developedon the basis of force equilibrium approach and applied for planestrain compression test [18]. That is, the parameter of ̨ is definedand derived by the rate of flow softening and the strain rate sensi-tivity parameter except for temperature factor. Gegel’s and Murty’scriteria found to over-predict the instability domains. These criteriacan predict the flow instability at high temperature and low strainrate (Figs. 6 and 7), which is not predicted by Prasad’s and Murty’scriteria under the test conditions. It is maybe related to the wedgecracking. However, the instability map developed by Prasad’s andMalas’s criteria could predict the unstable domains for Ti17 alloyduring the hot processing more precisely than the other instabilitycriteria.

It can be seen from the processing map of Ti17 alloy that all kindsof instability criteria are not consistent completely. Therefore, thesecriteria should be chosen carefully and considered comprehen-sively in practical application. In the present research, the Murty’s

136 X. Ma et al. / Materials Science and Engineering A 550 (2012) 131– 137

F ◦C/108

ib

5

vPIef

ig. 8. Microstructure evolution of Ti17 alloy deformed at strain of 0.9: (a) 78060 ◦C/0.001 s−1; (g) 840 ◦C/0.01 s−1.

nstability criteria should be selected for more accurate predictionsecause of its strict theoretical inferences and calculations.

. Conclusions

The instability of Ti17 alloy has been studied based on thearious existing instability criteria, including Semiatin’s criterion,

rasad’s criterion, Murty’s criterion, Gegel’s and Malas’s criteria.nstability maps were developed based on these criteria to delin-ate the unstable domains employing stress–strain data obtainedrom isothermal hot compression tests in the range of 780–860 ◦C

s−1; (b) 860 ◦C/10 s−1; (c) 780 ◦C/1 s−1; (d) 860 ◦C/1 s−1; (e) 800 ◦C/0.001 s−1; (f)

and 0.001–10 s−1. From the analysis of the obtained results andmicrostructure observation, the following conclusions are drawn:

1. It has been observed that Semiatin’s criterion under-predictsthe instability regions whereas Gegel’s and Malas’s criteria over-predict the instability domains in the studied temperatures andstrain rates at a certain strain of 0.9.

2. The maps at strain of 0.9 developed by Prasad’s and Murty’s cri-

teria have been found to predict the unstable domains for Ti17alloy with high reliability.

3. The material exhibited stable flow at lower strain rates whileflow instabilities occurred at higher strain rates (≥1 s−1) and

nd Eng

A

CtsHow

R

[[[[[

[

[

[

[

[

[[

X. Ma et al. / Materials Science a

entire temperatures. The microstructural manifestations ofthese instabilities are in the form of adiabatic shear band andflow localization.

cknowledgements

The authors are grateful to the financial support of the Newentury Excellent Talents in University with No. NCET-07-0696,he Doctorate Foundation of Northwestern Polytechnical Univer-ity with No. cx201211, Research Fund for the Doctoral Program ofigher Education of China with No. 20116102110015, and the Fundf the State Key Laboratory of Solidification Processing in NWPUith No. 35-TP-2009.

eferences

[1] C. Leyns, M. Peters (Eds.), Titanium and Titanium Alloys, Wiley-VCH GmbH &Co. KGaA, Weinheim, 2003.

[2] Y.G. Zhou, W.D. Zeng, H.Q. Yu, Mater. Sci. Eng. A 393 (2005) 204–212.

[3] R. Boyer, G. Welsch, E.W. Collings (Eds.), Materials Properties Handbook: Tita-

nium Alloys, ASM International, Materials Park, OH, 1994, pp. 453–463.[4] K.X. Wang, W.D. Zeng, Y.Q. Zhao, Y.J. Lai, Y.G. Zhou, J. Mater. Sci. 45 (2010)

5883–5891.[5] S.L. Semiatin, V. Seetharaman, I. Weiss, Mater. Sci. Eng. A 243 (1998) 1–24.

[[

[

ineering A 550 (2012) 131– 137 137

[6] H.J. Front, M.E. Ashby, Deformation Maps, Pergamon Press, New York, 1982.[7] R. Raj, Metall. Trans. A 12 (1981) 1089–1097.[8] S.L. Semiatin, G.D. Lahoti, Metall. Trans. A 12 (1981) 1705–1718.[9] F. Montheillet, J.J. Jonas, K.W. Neale, Metall. Trans. A 27 (1996) 232–235.10] Y.V.P.K. Prasad, Indian J. Technol. 28 (1990) 435–451.11] Y.V.R.K. Prasad, H.L. Gegel, Metall. Trans. A 15 (1984) 1883–1892.12] S.V.S.N. Murty, B.N. Rao, Mater. Sci. Eng. A 254 (1998) 76–82.13] S.V.S.N. Murty, B.N. Rao, Mater. Sci. Technol. 18 (2002) 571–574.14] H.L. Gegel, Synthesis of atomistic and continuum modeling to describe

microstructure, in: R.J. Arsenault, J.R. Beeler, D.M. Easterling (Eds.), ComputerSimulation in Material Science, ASM, Metals Park (Ohio), 1987, pp. 291–344.

15] J.C. Malas, Methodology for design and control of thermo-mechanical process,Ph.D. dissertation, Ohin University, Athens, OH, 1991.

16] X. Ma, W.D. Zeng, Y. Sun, Y.Q. Zhao, S.L. Wang, Y.G. Zhou, Rare Metal. Mater.Eng. 39 (2010) 756–761.

17] S.V.S.N. Murty, B.N. Rao, B.P. Kashyap, J. Mater. Process. Technol. 166 (2005)268–278.

18] S.L. Semiatin, J.J. Jonas, Formability and Workability of Metals: Plastic Instabilityand Flow Localization, ASM, Metals Park, OH, 1984.

19] Y.V.R.K. Prasad, S. Sasidhara (Eds.), Hot Working Guide: A Compendium ofProcessing Maps, Materials Park, OH, ASM International, 1997.

20] S.L. Semiatin, V. Seetharaman, I. Weiss, Mater. Sci. Eng. A 263 (1999) 257–271.21] Y.Y. Zong, D.B. Shan, M. Xu, Y. Lv, J. Mater. Process. Technol. 209 (2009)

1988–1994.

22] X. Ma, W.D. Zeng, B. Xu, Y. Sun, C. Xue, Y.F. Han, Intermetallics 20 (2012) 1–7.23] Y. Sun, W.D. Zeng, Y.Q. Zhao, X.M. Zhang, Y. Shu, Y.G. Zhou, Mater. Sci. Eng. A

528 (2011) 1205–1211.24] K.X. Wang, W.D. Zeng, Y.Q. Zhao, Y.J. Lai, X.M. Zhang, Y.G. Zhou, Mater. Sci.

Technol. 27 (2011) 21–28.