NEW METHODS OF DAMAGE AND FAILURE ANALYSIS OF …konference.fmt.vsb.cz/work2014/files/New...Analysis... · data [4]. This is because the stress intensity factor (SIF) at point 1 in

NEW METHODS OF DAMAGE AND FAILURE ANALYSIS OF

STRUCTURAL PARTS

September 8 – 11, 2014

Ostrava

Czech Republic

i

NEW METHODS OF DAMAGE AND FAILURE ANALYSIS

OF STRUCTURAL PARTS

Book of Abstracts

6th International Conference

September 8 – 11, 2014 Ostrava

edited by

Prof. Ing. Bohumír Strnadel, DrSc.

VŠB – Technical University of Ostrava

17. listopadu 15/2172

708 33 Ostrava

Czech Republic

Vydavatelství Vysoká škola báňská – Technická univerzita Ostrava

Published in cooperation with the project Regional Materials Science and Technology Centre,

CZ.1.05/2.1.00/01.0040.

ii

Book of Abstracts

6th International Conference on NEW METHODS OF DAMAGE AND FAILURE ANALYSIS

OF STRUCTURAL PARTS.

VŠB - Technical University of Ostrava, Czech Republic

8 - 11 September 2014

Published by Vydavatelství VŠB – TU Ostrava

17. listopadu 15, 708 33 Ostrava

Czech Republic

Cover image: Corrosion products on fracture surface of steel AISI 304 after stress

corrosion cracking test, by Doc. Ing. Jan Siegl, CSc., image magnification of 1000x.

Proceedings were designed by

Ing. Pavel Židlík

Ing. Daniela Vedrová

ISBN 978-80-248-3488-7

iii

CONTENTS

Proposal of Proximity Rule on Transformation from Subsurface to Surface Flaw .................. 1 K. HASEGAWA, Y. LI, R. SERIZAWA, M. KIKUCHI

Improvement of the Herbert Pendulum Hardness Tester .......................................................... 3 T. KABURAGI, R. SUZUKI, M. MATSUBARA, T. KOYAMA, T. TASHIRO

Decreasing Thermal Stresses in Steam Generator Collector Weld’s Area Using

External Cooling ....................................................................................................................... 5 R. KRAUTSCHNEIDER, L. JOCH

Collapse Mechanism of Rectangular Tubes Subjected to Pure Bending .................................. 7 K. MASUDA

The Evaluation of Actual Material Properties of Low Alloy CrMoV Steel from the

Results of Small Punch Tests .................................................................................................... 9 K. MATOCHA, L. KANDER, O. DORAZIL, K. GUAN, Y. XU

Extra High Strength Steel Plates. Production, Possibilities and Limits .................................. 11 I. MIKA

Precision Evaluation for Realiability of Power Module Using Coupled Electrical-

Thermal-Mechanical-Analysis ................................................................................................ 13 H. MORITA, Q. YU

Comparative Strain Analysis of 34CrMo4 Steel and Inconel 738LC ..................................... 15 M. ŠTAMBORSKÁ, M. LOSERTOVÁ, K. KONEČNÁ, V. MAREŠ, L. HORSÁK, R. GALACZ

Internal Crack Growth Simulation Using S-version FEM ...................................................... 17 M. KIKUCHI, R. SERIZAWA, S. YAMADA

Collapse Evaluation of Double Notched Stainless Pipes Subjected to Combined

Tension and Bending ............................................................................................................... 19 R. SUZUKI, M. MATSUBARA, S. YANAGIHARA, M. MORIJIRI, A. OMORI, T. WAKAI

Development of a Crack Opening Displacement Assessment Procedure Considering

Change of Compliance at a Crack Part in Thin Wall Pipes Made of Modified

9Cr-1Mo Steel ......................................................................................................................... 21 T. WAKAI, H. MACHIDA, M. ARAKAWA, S. YOSHIDA, S. YANAGIHARA, R. SUZUKI,

M. MATSUBARA, Y. ENUMA

Stress Relaxation Small Punch Testing of P92 Steel .............................................................. 23 P. DYMÁČEK, F. DOBEŠ, M. JEČMÍNKA

ABI Testing of Reactor Pressure Vessel Steel ........................................................................ 25 P. HAUŠILD, J. SIEGL, A. MATERNA

Characterization of Heterogeneous Weldments ...................................................................... 27 V. ŠEFL, R. NOVÁKOVÁ, J. BYSTRIANSKÝ

iv

Resistance to Corrosion Cracking of Steel 100Cr6 in Humid Air under Higher Tensile

Stresses .................................................................................................................................... 29 S. LASEK, V. ČÍHAL, M. BLAHETOVÁ, E. KALABISOVÁ

Relationships between KIC and CVN at Temperatures Lower than NDT ............................... 31 M. HABASHI, M. TVRDY

Biomechanics - Probabilistic Reliability Assessment of Femoral Screws .............................. 33 K. FRYDRÝŠEK

Finite Element Human Model for Crash safety Assessment of Automobile in Frontal

Collision .................................................................................................................................. 35 Y. ZAMA

Biomechanics – Problematic Loosening of Locking Screws from Plates .............................. 37 R. ČADA, K. FRYDRÝŠEK

Biomechanics – Safety Factor Evaluation of Anterolateral Plates for Distal Tibia

Fractures .................................................................................................................................. 39 G. THEISZ, K. FRYDRÝŠEK

Cyclic Bending Deformation and Fracture of Al and Al-1.0mass%Mg Alloy ....................... 41 H. IKEYA, H. FUKUTOMI

Cyclic Instability of Steel-Titanium Bimetallic Composite Obtained by Explosive

Welding ................................................................................................................................... 43 A. KAROLCZUK, T. ŁAGODA

Including of Ratio of Fatigue Limits from Bending and Torsion for Estimation

Fatigue Life under Cyclic Loading ......................................................................................... 45 M. KUREK, T. ŁAGODA

Evaluation of Fatigue Crack Growth in Alpha Titanium Alloys ............................................ 47 O. UMEZAWA, M. HAMADA, T. TATSUMI

The Current Status of New Czech Corrosion Fatigue Evaluation Proposal for WWER

Nuclear Power Plants .............................................................................................................. 49 L. VLČEK

Effect of Repeated Heating on One-Point Rolling Contact Fatigue of High-Carbon

High-Chromium Steel Bar ...................................................................................................... 51 K. MIZOBE, R. SEGAWA, T. SHIBUKAWA, K. KIDA

Statistical Analysis of Accidents Due to Fatigue and Corrosion at Facilities Producing

High Pressure Gas ................................................................................................................... 53 T. SHIBUTANI, N. KASAI, H. KOBAYASHI, H. AKATSUKA, T. TAKAHASHI, T. YAMADA

Influence of Inductive Hardening on Wear Resistance in Case of Rolling Contact ............... 55 M. ŠOFER, R. FAJKOŠ, R. HALAMA

Effect of Surface Quality of Machined Railway Wheels on Fatigue Strength ....................... 57 R. FAJKOŠ, T. TKÁČ

v

Application of Ultrasonic Impact Treatment (UIT) for Improvement of Fatigue Life ........... 59 T. ISHIKAWA, K. HAYASHI

Cyclic Plastic Properties of Class C Steel Including Ratcheting: Testing and

Modelling ................................................................................................................................ 61 R. HALAMA, A. MARKOPOULOS, M. ŠOFER, P. MATUŠEK

Characterization of Vermiculite Particles after Mechanical Treatment .................................. 63 K. ČECH BARABASZOVÁ, G. SIMHA MARTYNKOVÁ

Advanced Numerical Modelling Methods for 1D Periodic Plasmonic Structure

Simmulations ........................................................................................................................... 65 L. HALAGAČKA, K. POSTAVA, M. VANWOLLEGHEM, B. DAGENS, J. BEN YOUSSEF,

J. PIŠTORA

Molecular Modeling of Antimicrobial Nanocomposites ........................................................ 67 D. HLAVÁČ, J. TOKARSKÝ

Antimicrobial Kaolinite Based Nanocomposites .................................................................... 69 S. HOLEŠOVÁ, M. HUNDÁKOVÁ, E. PAZDZIORA

Volatile Organic Molecules Sorption onto Carbon Nanotubes ............................................... 71 G. SIMHA MARTYNKOVÁ, D. PLACHÁ, L. ROZUMOVÁ, E. PLEVOVÁ

Optical Modelling of Microcrystalline Silicon Deposited by Plasma-Enhanced

Chemical Vapour Deposition on Low-Cost Iron-Nickel Substrate for Photo-Voltaic

Applications ............................................................................................................................ 73 Z. MRÁZKOVÁ, K. POSTAVA, A. TORRES-RIOS, M. FOLDYNA,

P. ROCA I CABARROCAS, V. VODÁREK, J. HOLEŠÍNSKÝ, J. PIŠTORA

Submicron Calcium Phosphate Particles Study Anchored on Clay Supports ......................... 75 L. PAZOURKOVÁ, G.SIMHA MARTYNKOVÁ, M. HUNDÁKOVÁ, M. VALÁŠKOVÁ

Preparation of Submicron Particles of Biologically Active Substances Using

Supercritical Fluids ................................................................................................................. 77 D. PLACHÁ, T. SOSNA, E. VACULÍKOVÁ, M. MIKESKA, R. DVORSKÝ

Preparation of Carbon Nano Fillers for Metalic Composites .................................................. 79 L. ROZUMOVÁ, G. SIMHA MARTYNKOVÁ

TiO2 – Based Sorbent for Lead Ions Removal ........................................................................ 81 J. SEIDLEROVÁ, M. ŠAFAŘÍKOVÁ, L. ROZUMOVÁ, I. ŠAFAŘÍK, O. MOTYKA

Properties of Kaolinite Treated by Different Temperatures .................................................... 83 M. TOKARČÍKOVÁ, K. MAMULOVÁ KUTLÁKOVÁ, J. SEIDLEROVÁ

Influence of Void on the Mechanical Property of Nanomaterial ............................................ 85 K. YODEN, Y. SAITO, Q. YU

Determination of Anisotropic Crystal Optical Properties Using Mueller Matrix

Spectroscopic Ellipsometry ..................................................................................................... 87 K. POSTAVA, R. SÝKORA, D. LEGUT, J. PIŠTORA

vi

Observation of Magnetic Fields around Plastic Deformation Area in Low Carbon

Alloy Steel ............................................................................................................................... 89 K. KIDA, M. ISHIDA, K. MIZOBE

Magneto-Plasmonic Properties of Au/Fe/Au Planar Nanostructures: Theory and

Experiments ............................................................................................................................. 91 J. VLČEK, M. LESŇÁK, P. OTIPKA

Image Analysys via Reaction Diffusion System for Edge Detection ..................................... 93 K. NAKANE, H. MAHARA, K. KIDA

Effects of Inclusion on the In-Plane Mechanical Performance of Micro-Lattice Plate .......... 95 K. USHIJIMA, W. J. CANTWELL, D. H. CHEN

Assessment of Structures Loaded at Creep ............................................................................. 97 S. VEJVODA, P. POPPELKA, P. RYŠAVÝ

Diffusion of Hydrogen in the TRIP 800 Steel ......................................................................... 99 J. SOJKA, P. VÁŇOVÁ, V. VODÁREK, M. SOZANSKA

Precipitation Reactions in a Copper - Bearing GOES ........................................................... 101 V. VODÁREK, A. VOLODARSKAJA, Š. MIKLUŠOVÁ, J. HOLEŠINSKÝ, O. ŽÁČEK

Concept of Damage Monitoring after Grinding for Components of Variable Hardness ...... 103 A. MIČIETOVÁ, J. PIŠTORA, Z. DURSTOVÁ, M. NESLUŠAN

Study on Reliability Evaluation Method ofAdhesion Strength of Resin .............................. 105 O. HONDA, Q. YU

Direct Bonding of Ti/Al by Metal Salt Generation Bonding Technique with Formic

Acid ....................................................................................................................................... 107 T. AKIYAMA, S. KOYAMA

Direct Bonding of SUS304 Stainless Steel by Metal Salt Generation Bonding

Technique with Formic Acid ................................................................................................. 109 T. TSUNETO, S. KOYAMA

Direct Bonding of A6061 Aluminum Alloy by Metal Salt Generation Bonding

Technique with Formic Acid ................................................................................................. 111 Y. TOMIKAWA, S. KOYAMA

Effect of Surface Modification by Aqueous NaOH Solution on Bond Strength of

A5052 Aluminum Alloy/Al and Cu/Al ................................................................................. 113 X. MA, S. KOYAMA

Direct Bonding of Cu/Cu by Metal Salt Generation Bonding Technique with Formic

Acid and Acetic Acid ............................................................................................................ 115 S. KOYAMA, N. HAGIWARA, I. SHOHJI

Delamination Property of Modelled Air Plasma Sprayed Therma Barriear Coatings:

Effect of Difference in Chemical Composition of Bond Coat .............................................. 117 M. HASEGAWA, S. YAMAOKA

vii

Microstructure Modification of CGDS and HVOF Sprayed CoNiCrAlY Bond Coat

Remelted by Electron Beam .................................................................................................. 119 P. GAVENDOVÁ, J. ČÍŽEK, J. ČUPERA, M. HASEGAWA, I. DLOUHÝ

Response of Alumina Foam to Tensile Mechanical Loading Including Stress

Concentrator Effect ............................................................................................................... 121 I. DLOUHÝ, Z. CHLUP, H. HADRABA, L. ŘEHOŘEK

Nondestructive Magnetic Monitoring of Grinding Damage ................................................. 123 M. ČILLIKOVÁ, B. MIČIETA, M. NESLUŠAN, D. BLAŽEK

viii

NEW METHODS OF DAMAGE AND FAILURE ANALYSIS OF STRUCTURAL PARTS

8 – 11, SEPTEMBER, 2014, OSTRAVA, CZECH REPUBLIC

1

PROPOSAL OF PROXIMITY RULE ON TRANSFORMATION FROM

SUBSURFACE TO SURFACE FLAW

K. HASEGAWA1*, Y. LI1, R. SERIZAWA2, M. KIKUCHI2

1Japan Atomic Energy Agency, Tokai-mura, Ibaraki-ken, 319-1195, Japan; email: [email protected]

2Tokyo University of Science, Yamazaki, Noda-shi, Chiba-ken, 278-8510, Japan

KEY WORDS: proximity rule, subsurface flaw, fatigue crack growth, stress intensity factor

If subsurface flaws are detected that are close to the component free surfaces, flaw-to-

surface proximity rule is used to determine whether the flaws should be treated as subsurface

flaws as is, or transformed to surface flaws. However, the criteria for the rules on transforming

subsurface to surface flaws differ among fitness-for-service codes.

A subsurface flaw located near a component surface is illustrated in Fig. 1, where a is the

half flaw depth, the length, and S the distance from subsurface flaw to component surface.

When S is short, the subsurface flaw is transformed to be a surface flaw with the depth of 2a+S.

ASME [1], JSME [2], and Swedish SSM[3] provide the proximity rules as follow;

4.0/ aSY . (1)

When S and a satisfy Eq. (1), the subsurface flaw is treated as a surface flaw, where Y is

the flaw-to-surface proximity factor.

It is reported that, from fatigue crack growth

experiments, the proximity factor is not constant, as shown

in Eq. (1). It is suggested that the Y should be the function

of the flaw aspect ratio of a/ based on the experimental

data [4]. This is because the stress intensity factor (SIF) at

point 1 in Fig. 1 increases with decreasing the aspect ratio

a/ under constant S. Large interaction between the crack

tip at point 1 and component free surface occurs for small

aspect ratio of a/.

The SIFs at points 1 and 2 for

subsurface flaws with various shapes in

plates were calculated by FEM analysis.

The applied load was membrane stress. The

relationship between the ratio of SIF at

points 1 and 2, K1/K2, and the distance S is

shown in Fig. 2, as a parameter of a/. The

K1/K2 increases with decreasing the

distance S and a/. When looking at the

same value of K1/K2, interaction of flaw

with small aspect ratio occurs at long

distance. For example, in case of

K1/K2 =1.1, flaw with a/ = 0.125 occurs at

S =2.3 mm, and flaw with a/ = 0.65 occurs

at S =1.0 mm. That is, smaller the aspect

ratio, longer the distance.

Fig. 1 Subsurface flaw near component

surface.

Fig. 2 Stress intencity factor ratio for the distance of

subsurface flaw.

Point , 2 K

Point , 1 K

2

1

S

2a

1.4

1.3

1.2

1.1

10 1 2 33 4 5

S, mm

K/K

12

a/ = 0.125

a/ = 0.250

a/ = 0.375

a/ = 0.500

a/ = 0.625

mailto:[email protected]



2

Figure 3 shows the constant interaction for distance and flaw aspect ratio. The exponent of

4.83 for K1/K2 came from fatigue crack growth rates [4]. The fatigue crack growth rate

da/dN =2.01×10-14K4.83 was obtained by 0.5CT (compact tension) specimens, where K is

the SIF range. To compare the behaviour of subsurface to surface flaw by fatigue growth,

equivalent interaction of (K1/K2 )4.83 is shown between the distance S/a and the aspect ratio a/.

Again, interaction for small a/ occurs at long distance. The fatigue test data on abrupt changes

from subsurface to surface flaw depths are shown as open circles in Fig. 3 [4]. It can be seen

that the test data are close to the curve of (K1/K2 )4.83 =1.3. Subsurface crack begins to penetrate

free surface when the crack growth rate at point 1 is 30% higher than that at point 2.

Fig. 3. Stress intencity factor ratios for subsurface flaw. Fig. 4. Proposal of proximity rule.

From the view point of codification, it is desired to be a simple expression. Based on the

curve of (K1/K2 )4.83 =1.3 and experimental data, a new proposal of proximity rule is given by;

Y = 0.8 – (a/) for 0 < a/ 0.6, and

Y = 0.2 for 0.6 < a/. (2)

It is concluded that the new proximity factor Y based on SIF interaction can be developed

as a function of aspect ratio.

Acknowledgement: The authors gratefully acknowledge the support by K. Saito, Hitachi

GE, and K. Miyazaki, Hitachi Ltd.

REFERENCES

[1] American Society of Mechanical Engineers Boiler & Pressure Vessel Code Section XI: Rules

for In-service Inspection of Nuclear Power Plant, 2013 Edition.

[2] The Japan Society of Mechanical Engineers S NA1: Rules on Fitness-for-Service for Nuclear

Power Plants (in Japanese), 2008.

[3] Swedish Radiation Safety Authority (SSM): A Combined Deterministic and Probabilistic

Procedure for Safety Assessment of Components with Cracks-Handbook, 2008.

[4] HASEGAWA, K., LI, Y., MIYAZAKI, K. SAITO, K.: Fatigue Crack Growth for Subsurface

Flaws near Component Surface and Proximity Rules, ASME PVP2013-97559, Paris, 2013.

1

0.8

0.6

0.4

0.2

00 0.60.40.2 0.8

Experiment

( / ) =1.2K K1 2

4.83

1.3

2.0

1.5

Aspect ratio, a/

S/

(=

Y)

a

1

0.8

0.6

0.4

0.2

00 0.60.40.2 0.8

( / ) =1.3K K1 2

4.83

1.3

Aspect ratio, a/

S/

(=

Y)

a

Y = 0.8-( / )a



3

IMPROVEMENT OF THE HERBERT PENDULUM HARDNESS TESTER

T. KABURAGI1, R. SUZUKI2*, M. MATSUBARA2, T. KOYAMA2, T. TASHIRO3

1Gunma Industrial Technology Center, 884-1 Kamesato-machi, Maebashi, Gunma 379-2147, Japan 2Faculty of Science and Technology, Gunma University, 1-5-1 Tenjin-cho, Kiryu, Gunma 376-8515, Japan; email: [email protected]

3Department of mechanical system engineering, Faculy of engineering, Gunma University, 1-5-1 Tenjin-cho, Kiryu, Gunma 376-8515, Japan

KEY WORDS: hardness, damping property, pendulum, measurement system

The Herbert hardness tester [1-3] is a typical pendulum-type hardness tester. Hardness of

materials is measured based on the swing angle of the pendulum in relation to the specimen. In

the present study, hardness is measured using a Habara-type Herbert pendulum hardness tester

[4] with a modified measurement system. We investigate the effect of each condition such as

the indenter tip radius of the Herbert pendulum, the swing cycle and the surface roughness of

the specimens on Herbert hardness. Moreover, we discuss the relationship between damping

hardness and conventional Brinell hardness.

Fig. 1 shows the measurement system for

Herbert hardness. The Herbert pendulum

swings on a specimen and its swing angle is

measured continuously with the two laser

displacement meters which are installed

independent from the Herbert pendulum.

Fig. 2 shows the swing angle of the

Herbert pendulum as a function of time. The

angular amplitude of the Herbert pendulum

decreases exponentially with time. The

original definitions of Herbert hardness is

evaluated by initial swing angle, S1, and the

time it takes for 10 swings, T, because the

swing time is measured using a stopwatch

and the swing angle is measured using a spirit

level by the visual evaluation. Therefore

Herbert hardness evaluated by original

definitions has a large variation. So we

improved the measurement system in order to

measure the swing angle accurately more

than original measurement system. In

addition, in place of the original definitions,

damping hardness is proposed as an

indication of hardness. Since damping

hardness is determined by the damping factor

obtained from the free damped vibration

waveform of the Herbert pendulum, its method of hardness evaluation more reasonable than

the original definitions. The following equation shows the envelope line connecting successive

local maximum amplitudes S(t).

teStS 0)( (1)

Fig. 1. Measurement system.

Fig. 2. Characteristics of the swing angle of the tester as

a function of time obtained by Herbert hardness tester.

WeightWeight

Specimen

Indenter

Weight

Laser displacement meter

-40

0

40

0 120

Sw

ing a

ngle

[d

egre

es]

Time [s]

t

T

S1

S2

S0

X teStS 0



4

The exponent α of Equation (1) is referred to as the damping factor and indicates the amount

of damping in the testing system.

Fig. 3 shows the relationship between the damping hardness and conventional Brinell

hardness. The dashed line is predicted value of the Brinall hardness from the damping hardness

and the indenter radius. For all indenter radii, the damping hardness increases with the decrease

in the Brinell hardness number. For a sample of the same Brinell hardness number, the damping

hardness increased with increasing indenter radius. The difference in the damping hardness

associated with the indenter radius increases as the Brinell hardness number decreases. The

increase in the contact area between the indenter and the specimen associated with the increase

in the indenter radius and the decrease in the hardness of the specimen, the resistance to

swinging of the pendulum tester increased, resulting in large damping.

Fig. 3. Brinell hardness predicted based on the damping hardness.

Here, in order to predict the Brinell hardness number the multivariate analysis is conducted

using the damping hardness. As a result, we obtained the following expression:

946.0

131.1

448.0

eR

HBW

, (2)

where HBW is Brinell hardness number, α is the damping hardness and R is the indenter radius.

The correlation coefficient exceeds 0.98. Therefore, it may be possible to accurately predict the

Brinell hardness based on the damping hardness, and this measurement system may be used

practically.

REFERENCES

[1] HERBERT, E.G.: "Some Recent Developments in Hardness Testing," The Engineer 135:686-68

(1923).

[2] BENEDICKS, C., CHRISTIANSEN: "Investigations on the Herbert Pendulum Hardness

Tester," Journal Iron and Steel Institute 110; 219-248 (1924).

[3] WILLIAMS, S.R.: Hardness and Hardness Measurements, American Society for Metals,

Cleveland (1942).

[4] HABARA, H., KAWAMITSU, T., HARIMOTO, K., AND INOUE, H.: "Restration of the

Herbert Pendulum Hardness Tester and its application (in Japanese)," Journal of Material

Testing Research Association of Japan, 43(4):248-254 (1998).

0

0.001

0.002

0.003

0.004

0 200 400 600 800

Dam

pin

g h

ard

nes

s

Brinell hardness

R1

R2

R4

累乗 (R2)Predicted value



5

DECREASING THERMAL STRESSES IN STEAM GENERATOR

COLLECTOR WELD’S AREA USING EXTERNAL COOLING

R. KRAUTSCHNEIDER1*, L. JOCH1

1Institute of Applied Mechanics Brno, Ltd., Resslova 972/3, 602 00 Brno; email: [email protected]

KEY WORDS: steam generator, thermal stresses, external cooling, stress corrosion cracking, dissimilar

metal weld

Presented paper deals with possibility of external cooling of steam generator weld’s area to

decrease internal tensile thermal stresses which are one of the causes of cracking in this area.

On various WWER-440 nuclear power plants (NPP) steam generators (SG) occurred quite

serious problem of cracking in weld joints connecting primary collectors to the SG vessel’s

nozzle (Fig. 1). The cause of this cracking is stress corrosion cracking (SCC) mechanism. The

crack rises and grows on the interface between different kinds of material, austenitic steel

08CH18N10T on one side and carbon steel 22K on the other (Fig. 1).

On the SG secondary side there is a space between the SG nozzle and the primary collector,

which is also called “pocket”. In this pocket, due to poor possibilities of effective blowdown,

exist the secondary media of higher corrosive potential. The existence of corrosive media

together with existing stresses can cause intergranular corrosion and cracking in this area.

Existing stresses are combination of thermal stresses from different thermal expansion

properties of austenitic and carbon steel, and external stresses on the SG nozzle from the

primary circuit. Thermal stresses are higher and thus more important.

So there are two approaches how to decrease the possibility of crack occurrence and growth.

The first one is to try keeping the pocket as clean as possible, it means to try to improve the

effectiveness of the pocket’s blowdown. And the second one is to try to decrease existing

(thermal) stresses. Of course both approaches can be combined.

Fig. 1. PGV-440 steam generator collector’s weld area.

Recently there were presented some studies of external cooling of this area from Russia.

Those studies were done on WWER-1000 steam generators, but from similar reason. And

SG

vessel

(22K)

SG collector

(08CH18N10

T)

SG nozzle

(22K)

22K

Dissimilar

metal weld



6

because the SG primary collector’s joint is similar on WWER-440, we tried to make similar

analyses here.

The external cooling of the SG primary collector weld joint is done by placing external

cooing sleeve around it, with inlet and outlet nozzle and with flowing air cooling media of

higher velocities.

Computational fluid dynamics (CFD) analyses and subsequent finite element analyses

(FEM) were done, with the objective to compare stresses in the weld area with and without

external cooling.

Results of these analyses showed that by proper external cooling of the weld area it is

possible not just to decrease significantly existing tensile stresses, but to change them into

compressive ones. This of course would have great impact on crack occurrence and growth.

REFERENCES

[1] JOCH, L.: Influence of thermal fields on dissimilar weld DN1105 of NPP Dukovany steam

generator’s collector (in Czech language), IAM Brno Report, 5071/13, Brno, 2013.

[2] LICKA, A.: Determination of residual life of SG primary collectors with graphite gasket. The

evaluation of the measurement results after running the second block and in steady state

operation at nominal power, including recommendations for further operation (in Czech

language), IAM Brno Report, 2643/98, Brno, 1998.

[3] KUTDUSOV, YU.F. et al.: Innovation of devices for stress reduction in welding joint 111 of

welding unit of “hot” heat-transfer manifold and socket DN1200 of PGV-1000 by air blowing

method on Rostov and Balakovo nuclear power plants (in Russian language), 8-th International

scientific and technical conference "safety assurance of NPP with WWER", OKB Gidropress,

ISBN: 978-5-94883-130-5, Podolsk, 2013.

[4] LYAKISHEV, S.L. et al.: Development and justification of measures on assurance of reliable

and safe operation of welded joints no. 111 of steam generator PGV-1000M (in Russian

language), 6-th International scientific and technical conference "safety assurance of NPP with

WWER", OKB Gidropress, Podolsk, 2009.

[5] TRUNOV N.B. et al.: Results of studies of metal fracture causes in the area of primary

collector-to-steam generator vessel welding and development of corrective measures (in

Russian language), 8th International Seminar on Horizontal Steam Generators, OKB

Gidropress, Podolsk, 2010.

http://www.gidropress.podolsk.ru/files/proceedings/seminar8/seminar8/authors-en.htm#trunov-nb



7

COLLAPSE MECHANISM OF RECTANGULAR TUBES

SUBJECTED TO PURE BENDING

K. MASUDA1*

1University of Toyama, Japan; email: [email protected]

KEY WORDS: FEM, pure bending, rectangular tube, buckling, effective width

Rectangular and square section tubes are widely used in mechanical equipment. Therefore,

a study of the collapse behaviour is important for both the design and analysis of weight-

efficient safety structures. In the present paper, the collapse behaviours of rectangular tubes

subjected to pure bending are investigated using the finite element method. Such bending

collapse has been investigated extensively [1], [2]. These studies have revealed the existence of

two types of collapse. The first type is a collapse due to buckling at the compression flange,

and the second type is a collapse due to plastic yielding at the flanges. Moreover, another type

of collapse exists. For a rectangular tube in which the web is wider than the flange, collapse

due to buckling occurs at the compression web [3]. In the present paper, previous evaluation

method of the maximum moment is refined, and systemized evaluation method is proposed.

The validity of this method is verified through comparison with the numerical results obtained

by FEM under various conditions.

REFERENCES

[1] KECMAN, D.: Bending collapse of rectangular and square section tubes, International Journal

of Mechanical Sciences, Vol. 25, 1983, pp. 623-36.

[2] LU, G., YU, T. X.: Energy absorption of structures and materials, CRC Press, Section 5, 2003.

[3] MASUDA, K., CHEN, D.H.: Prediction of Maximum Moment of Rectangular Tubes Subjected

to Pure Bending, Journal of Environment and Engineering, Vol. 6(3), 2011, pp. 554-566.



8



9

THE EVALUATION OF ACTUAL MATERIAL PROPERTIES OF LOW

ALLOY CrMoV STEEL FROM THE RESULTS OF SMALL PUNCH TESTS

K. MATOCHA1*, L. KANDER1, O. DORAZIL1, K. GUAN2, Y. XU2

1 MATERIAL & METALLURGICAL RESEARCH, Ltd., Ostrava, Czech Republic; email: [email protected]

2 East China University of Science and Technology, Shanghai, China

KEY WORDS: small punch test technique, actual mechanical properties, CEN workshop agreement,

residual lifetime

The Small Punch Test Technique is used, at the present time, to obtain actual tensile,

fracture and creep characteristics necessary for estimation and monitoring of residual lifetime

of critical components of industrial plants [1-3]. CWA 15627 “Small Punch Test Method for

Metallic Materials” was developed in Europe in 2007 [4]. Part B: A Code of Practice for Small

Punch Testing for Tensile and Fracture Behaviour is used for determination of yield and tensile

strength, Ductile Brittle Transition Temperature (DBTT) and fracture toughness of the metallic

materials.

The present paper summarizes the results of the bilateral project in the frame of Czech-

Chinese Scientific and Technological Cooperation focused on the determination and

comparison of the empirical correlations for estimation of FATT and yield strength, tensile

strength and JIC at laboratory temperature from the results of Small Punch Tests for low alloy

14MoV6-3 steel.

The participants of the project were MATERIAL & METALLURGICAL RESEARCH,

Ltd. Ostrava, Czech Rep. and East China University of Science and Technology, Shanghai,

China. Both tensile tests, Charpy impact tests and fracture toughness tests using standardized

test specimens and Small punch tests were carried out in both laboratories according to national

standards.

Results of standardized tensile tests and impact tests obtained in both laboratories are in

very good agreement. However the empirical correlations for determination of yield stress,

tensile strength and FATT from the results of Small punch tests are significantly different.

Factors affecting this experimentally proved fact are discussed.

Acknowledgement: The paper has originated during the solution of the project LH 12199

„The Comparison of Codes of Practice for Determination of Mechanical Properties by SP Tests

between EU and China“ in the frame of the programme of the Ministry of Education, Youth and

Sports KONTAKT II.

REFERENCES

[1] MATOCHA, K.: Determination of Actual Tensile and Fracture Characteristics of Critical

Components of Industrial Plants under Long term Operation by SPT. Proceedings of the ASME

2012 Pressure & Piping Division Conference PVP 2012, July 15-19, 2012, Toronto, Ontario,

Canada (CD-ROMM).

[2] FOULDS, J.R., JEWETT, C.W., BISBEE, L.H., WHICKER, G.A., VISWANATHAN, R.:

Miniature Sample Removal and Small Punch Testing for In-Service Component FATT.

Proceedings of the Robert I. Jaffee memorial Symposium on Clean Materials Technology.

ASM/TMS Materials Week, 2-5 November 1992, Chicago, Illinois, USA.




10

[3] ABENDROTH, M.: Identification of Creep properties for P91 Steels at High Temperatures Using

the Small Punch Test. In: Proc. of 1st International Conference “Determination of Mechanical

Properties of Materials by Small Punch and other Miniature Testing Techniques”. Ostrava,

Czech Rep., August 31 to September 2, 2010, pp. 39-43, ISBN 978-80-254-7994-0.

[4] CEN WORKSHOP AGREEMENT “Small Punch Test method for Metallic Materials” CWA

15627:2007 D/E/F, December 2007.



11

EXTRA HIGH STRENGTH STEEL PLATES. PRODUCTION,

POSSIBILITIES AND LIMITS

I. MIKA1*

1SSAB Swedish Steel s.r.o., Spartakovců 3, 70800 Ostrava, Czech Republic; email: [email protected]

KEY WORDS: high strength steel, welding, cutting, fatigue, buckling

The present limit in mass produced, high strength, low alloyed steels is approximately

1300 MPa of yield strength, or 650 HBW of hardness respectively. In principle there is not so

big problem to reach needed high strength of the steel. The biggest challenge now, is to assure

the toughness and technological properties in combination with high strength. The

technological properties especially mean welding and thermal cutting. To assure good balance

between strength/hardness, toughness and weldability of steel means to heat treat the steel to

optimal microstructure. This involves reaching martensitic structure with low possible content

of carbon and alloying elements, the same way as to guarantee extremely clean steel. Therefore

it is important, as to quenching process proceeds very fast. Right metallurgical treatment and

proper selection of raw materials has to assure extremely low contents of the elements like

S, P, Sn, Zn. Deep vacuum treatment is subsequently responsible for decreasing especially

hydrogen in liquid steel, which improves thermal cuttability of ready plates.

Nevertheless, practically always, when strength of the steel is increasing, namely more than

700 MPa of yield point, the steel has to be more alloyed and technological properties are

decreasing. Further, there are mechanical properties, like fatigue strength and stability of steel

construction which are practically not increasing with increasing of steel strength. There is

necessary to apply new technological procedures and constructional principles to overcome

these limitations. Very hard steels are also substantially more sensitive to steel corrosion

cracking which make a challenge in some kinds of structures made from these steels.



12



13

PRECISION EVALUATION FOR REALIABILITY OF POWER MODULE

USING COUPLED ELECTRICAL-THERMAL-MECHANICAL-ANALYSIS

H. MORITA1*, Q. YU1

1Department of Mechanical Engineering, Graduate School of Engineering, Yokohama National University; Tokiwadai 79-5, Hodogaya-ku, Yokohama, Japan; email:[email protected]

KEY WORDS: power-modules, electrical-thermal-mechanical-analysis, reliability

In recent years, with the development of power electronics technology, the power modules

have been used extensively. The problem of the power module is thermal fatigue of the junction

area between Si chip and substrate caused by cyclic temperature change. It is necessary to

evaluate the thermal fatigue life of the power module under the power cycle using coupled

electrical-thermal-mechanical analysis.

Power cycle test fixed current and sets up fixed on-off time. This conditions done regularly.

However, irregular the electric load imposed at the time of an actual vehicle running. A gap

exists between the thermal fatigues added by a power cycle test.

It is thought that the reliance valuation basis of a power module is a very high standard

compared with real usage environment. Then, if the evaluation technique in the conditions near

real usage environment is establishable. It will be thought that it becomes possible to cancel the

over-spec of a product and to make the product whose cost was cut down. In this research,

analysis in the conditions near real usage environment is conducted. It aims at establishment of

the reliability assessment method of the power module near a real operating condition.

Analysis model is shown in Fig. 1. Electrical-Thermal-Mechanical-Analysis tried two

pattern as follow.

1st pattern, 200 A current was impressed on

DC IN shown in Fig. 1. The changing time of the

electrical current is 2.0 s. The cooling time is 18 s.

This 1st pattern named is REGULAR LOAD

TEST.

2st pattern, direct current was impressed on DC

IN shown in Fig. 1. The current conditions

impressed on DC IN shown in Table 1. This 2st

pattern named is IRREGULARITY LOAD TEST.

The result of coupled electrical-thermal

analysis is shown in Fig. 2. Temperature

distribution changes by changing a setup of a

current value or On-Off time. This understands that

it sees Fig. 2.

This temperature distribution is applied to

Thermal-Mechanical analysis. The result obtained

by Thermal-Mechanical analysis is creep and

plastic strain. Based on Manson-Coffin’s Law, the

thermal figue life of solder joints is often evaluated

by the inelastic strain range, which is the sum of the

creep and plastic strain [1].

Fig. 1. Analysis model of power module.

Table 1 Analysis condition for coupled

Thermal-Electrical Analysis.

Step On [s] Off [s] Current [A]

1 4 3 50

2 6 20 100

3 3 18 200

4 2 15 150

5 5 10 60



14

One creep and plasitc strain comes out in a regular load test, and Manson-Coffin’s Law,

fatigue lives solder joint is one solution. But two or more solutions come out in an irregular

load test, and Manson-Coffin’s Law, fatigue lives solder joint are tow or more

solutions.Applyin this strain range to Manson-Coffin’s Law shown as following

43.1

01.01328

N , (1)

means strain range [1].

Miner’s Law was applied to irregular load test result. Miner’s Law shown as following,

i

i

N

nD , (2)

ni means actual number of cycles, Ni means fatigue lives of cycles [2].

Strain, arranged randamly, rearrranged in each pattern.Arranging for each pattern, each

pattern of fatigue life was known. The sum of adding fatigue life of each pattern is fatigue life

of irregular load.

The result of fatigue life of regular load test and irredular load are shown in Fig. 3.

Examining Fig. 3, fatigue life of irregular load test was 4 times of the fatigue life of regular

load test. This result of analysis was shown that, Load on a power cycle test is over-spac so

than the load applied at the time of actual running.

Fig. 2. Temperature date. Fig. 3. Fatigue life.

In this research, analysis in the conditions near real usage environment of thermal fatigue

life of power-modules. The rearranged by classifying each pattern, an irregular condition, to

apply the Manson-Coffin’s Law to each and apply Miner’s Law. By using method of this

research, it is possible by using the simulation also in operation in various patterns, to evaluate

the fatigue life. And in this research, precisione valuation for reality of power module using

coupled electrical-thermal-mechanical-analysis.

REFERENCES

[1] YU, Q., SHIRATORI, M.: Fatigue-Stremgh Prediction of Microelectronics Solder Joints Under

Thermal Cyclic Loading, IEEE Trans.Compon.,Packag.Manuf.Techol.,Part A 20(3),

pp. 266-273, 1997.

[2] MINER, M. A.: Cumulative Damage in Fatigue, Journal of Applied Mechanics, v01.12

pp. 159-164, 1945.

0

20

40

60

80

100

120

140

0 20 40 60

tem

per

ature

[°C

]

time[s]

regular load

irregularity load

0

10000

20000

30000

40000

50000

regular load irregular load

Fat

igue

life

[cycl

e]



15

COMPARATIVE STRAIN ANALYSIS

OF 34CrMo4 STEEL AND INCONEL 738LC

M. ŠTAMBORSKÁ1*, M. LOSERTOVÁ1, K. KONEČNÁ2, V. MAREŠ3, L. HORSÁK3,

R. GALACZ2

1Department of Non-ferrous Metals, Refining and Recycling, Faculty of Metallurgy and Materials Engineering, VŠB-Technical University of Ostrava, Czech Republic; email: [email protected]

2Department of Material Engineering, Faculty of Metallurgy and Materials Engineering, VŠB-Technical University of Ostrava, Czech Republic

3Center of Advanced Innovation Technologies (CAIT), VŠB-Technical University of Ostrava, Czech Republic

KEY WORDS: 34CrMo4, IN 738 LC, stress analysis

The article was focused on strain analysis of 34CrMo4 steel and IN 738 LC superalloy. The

34CrMo4 steel grade is a low-carbon steel with medium through-hardening for medium-duty

machine parts [1]. The INCONEL 738 LC alloy is Ni-based low carbon superalloy hardened

by fine precipitates and carbides and used for high temperature applications, as gas turbine

engines [2, 3].

Experimental analysis of plastic deformation on the surface of specimens using contactless

displacement sensing methods is advantageous to obtain deformation fields in pre-selected

areas. The digital image correlation (DIC) is one of the most advanced optical methods of

displacement sensing and subsequent determination of strains on the surface of examined

objects [4, 5]. The strain fields were calculated from the displacement fields by the Vic 2D

program for both above mentioned materials.

Tensile test was carried out on the Zwick / Roel Z150 device with deformation rate of

2.5 x10-3 s-1 on cylindrical specimens having the gauge length of 28 mm and the diameter of

5 mm. Evaluation and compilation of the true stress –strain diagrams for all six specimens were

carried out using image correlation software Vic 2D and scanning was performed using Canon

5D MARK II.

Material characteristics and stress – strain diagrams obtained for 34CrMo4 steel and IN 738

LC superalloy from standard uniaxial tensile test are listed in Table 1 and shown in the Fig. 1,

respectively.

Table 1 The values of the measured mechanical properties for 34CrMo4 steel and IN 738 LC superalloy.

34CrMo4 IN 738 LC

Specimens 1 2 3 Average value 1 2 3 Average value

Y.S. [MPA] 937 958 941 94528 725 721 722 723 4

U.T.S. [MPA] 1040 1041 1041 10411 945 864 914 908 79

εx [-] 0.446 0.436 0.390 0.4410.020 0.184 0.139 0.177 0.167 0.024

Fig. 2 shows the results of strain field εx [-] obtained by the VIC 2D software for 34CrMo4

steel. The values of strains for 34CrMo4 steel obtained by Vic 2D, shown in red in the necking

of the specimens, reach maximum values from 39 to 45%. The average values of tensile

characteristics of 34CrMo4 steel reached of 945 MPa and 1041 MPa for the yield strength and

ultimate tensile stress, respectively. The values of strains for IN738LC superalloy obtained by

Vic 2D, reach maximum values from 14 to 18%. In the case of IN 738LC, the average tensile

characteristics reached of 723 MPa and 908 MPa for the yield strength and ultimate tensile

stress, respectively.




16

Fig. 1. The stress-strain diagram for a) 34CrMo4 steel and b) IN 738 LC superalloy.

Fig. 2. The strain fields obtained by Vic 2D for a) 34CrMo4 steel and b) IN 738 LC superalloy.

Based on the results of comparative strain analysis the DIC method is suitable for applying

to both alloys with good plasticity and with higher fragility. The method allows very well

determining the magnitude of the strain fields and localization of the deformation area as well

as to detect the casting defects in the material affecting mechanical behaviour.

Acknowledgement: This article has been elaborated in the framework of the projects:

"Opportunity for young researchers", reg. Nr. CZ.1.07/2.3.00/30.0016, supported by

Operational Programme Education for Competitiveness and co-financed by the European

Social Fund and the state budget of the Czech Republic and "Regional Materials Science and

Technology Centre - Feasibility Program", reg. Nr. LO1203 funded by Ministry of Education,

Youth and Sports of the Czech Republic.

REFERENCES

[1] HENDRYCH, A., KVÍČALA, M., MATOLIN, V., ŽIVOTSKÝ, O., JANDAČKA, P.:

International Journal of Fracture, Vol. 168, 2011, No. 2, pp. 259-266,

DOI: 10.1007/s10704-010-9573-7.

[2] BALIKCI, E., MIRSHAMS, R.A., RAMAN, A.: Tensile Strengthening in the Nickel-Base

Superalloy IN738LC. Journal of Materials Engineering and Performance, Vol. 9(3), June 2000,

pp. 324-329.

[3] LOSERTOVÁ, M., KONEČNÁ, K., JUŘICA, J., JONŠTA, P.: Hydrogen Effect on Mechanical

Properties of IN738LC Superalloy, In: 20th Anniversary International Conference on

Metallurgy and Materials: METAL 2011 Metal 2011, pp. 1039-1043. ISBN 978-80-87294-24-6.

[4] ROSSI, M., PIERRON, F., ŠTAMBORSKÁ, M., ŠIMČÁK, F.: Experimental and Applied

Mechanics, Vol. 4, 2013, pp. 229-235, DOI: 10.1007/978-1-4614-4226-4_27.

[5] ŠIMČÁK, F., ŠTAMBORSKÁ, M., HUŇADY, R.: Deformation of materials by using digital

image correlation, CHEMICKE LISTY, Vol. 105, 2011, No. 4, pp. 564-567, ISSN 0009-2770.

a) b)

b) a)



17

INTERNAL CRACK GROWTH SIMULATION USING S-VERSION FEM

M. KIKUCHI1*, R. SERIZAWA2, S. YAMADA2

1Tokyo University of Science, 2641 Yamazaki, Noda, Chiba, 278-8510, Japan; email: [email protected] 2Graduation School of Tokyo University of Science, 2641 Yamazaki, Noda, Chiba, 278-8510, Japan

KEY WORDS: fatigue, inner crack, slant crack, fully automatic crack growth simulation system,

S-version FEM

In nuclear power plant, there is a proximity rule to evaluate multiple inner cracks [1]. Inner

cracks are generated inside of the structure in different manners. There are many parameters

which affects the growing processes of inner cracks. They are; locations, slant angles, aspect

ratio of each inner crack and distances between adjacent inner cracks. When multiple inner

cracks are detected, proximity rules are proposed. But due to the complexity of the problem, it

is necessary to verify proposed proximity rules. But experimental study is very difficult due to

existence of many parameters, and crack growth occurs inside of the structure. Numerical

simulation is needed for this purpose. This problem is simulated using S-version FEM, which

is fully automatic crack growth simulation system developed by authors [2]. Using S-FEM,

inner crack is modeled independently from global structure, and crack growth is easily

simulated. In maintenance code of nuclear power plant, initial defects are modeled as elliptical

cracks in a normal plane to tension loading direction, and growth rate is estimated on this plane.

But by using S-FEM, realistic defect shape is modeled, and crack growth by fatigue is

simulated. Usually, such small defects are subjected to multi-axial loading, and crack growth

behaviors are very complicated. Parametric studies are conducted for this problem, and

proximity rules are verified with numerical results. Three problems are simulated. One is effect

of initial defect shape, second is crack growth of slant inner crack and the last one is interaction

effect of multiple inner crakcs.

(1) Effect of initial defect shape.

Four initial defects shapes are assumed, as shown in Fig. 1 (a)-(d). Aspect ratios and initial

areas are same for all models. By the proximity rule by JSME [1], all initial defects should be

modeled as an elliptical shape, which is case (d) in these figures. Figure 2 (a)-(b) shows changes

of crack shape of star shape model. Number of cyclic loading is shown. At first, crack growth

occurs mainly at the smaller part along the crack front, and finally, crack shape becomes

circular. Figure 8 shows relations of crack area and number of cycles for all cases. It shows that

all results are very similar to each other. This results show that JSME code gives good

estimation on crack growth prediction.

(a) Star shape (b) Diamond shape (c) Arbitrary shape (d) Ellipse

Fig. 1. Four different initial defects shapes.



18

(a) 2.2x107 (b) 9.3x107

Fig. 2. Changes of crack shape for star shape model. Fig. 3. Comparison of crack growth rates.

(2) Effect of slant angle of inner crack.

Inner crack is initially generated from inclusions, inhomogeneities or some other initial

defects. At first, it may not grow in a perpendicular plane to principal stresses, and has some

slant angle to principal stress axis. Then initial slant inner crack is assumed and crack growth

process is simulated. By the JSME code, this slant crack is evaluated after it is mapped on a

plane perpendicular to the principal stress axis. In this case, pure mode I crack growth is

assumed. Simulation results verified this modeling is reasonable. Crack growth rate of slant

inner crack is compared with that of JSME code, and it is again verified that modeling by JSME

code is reasonable.

(3) Evaluation of interaction effect of multiple inner cracks.

Growing processes of multiple inner cracks are simulated. Figure 4 (a) and (b) shows

overlapping processes of two parallel inner cracks. Results are compared with JSME code.

Again it is verified that JSME code gives conservative evaluation for the effect of interaction

between multiple cracks.

(a) 1.3 x 107 (b) 2.3 x 107

Fig. 4. Crack growth processes.

REFERENCES

[1] JSME S NAl-2004: Codes for Nuclear Power Generation Facilities – Rules on Fitness-for-

Service for Nuclear Power Plants -, (2004), (In Japanese)

[2] KIKUCHI, M., WADA, Y., SHIMIZU, Y., LI, Y.: Crack Growth Analysis in a Weld-heat-

affected Zone Using S-version FEM, International Journal of Pressure Vessels and Piping,

90-91, (2011), pp. 2-8.



19

COLLAPSE EVALUATION OF DOUBLE NOTCHED STAINLESS PIPES

SUBJECTED TO COMBINED TENSION AND BENDING

R. SUZUKI1*, M. MATSUBARA1, S. YANAGIHARA2, M. MORIJIRI1, A. OMORI2, T. WAKAI3

1*Faculty of Science and Technology, Gunma University, 1-5-1 Tenjin-cho, Kiryu, Gunma 376-8515, Japan; email: [email protected]

2Department of Mechanical System Engineering, Graduate School of Engineering, Gunma University, 1-5-1 Tenjin-cho, Kiryu, Gunma 376-8515, Japan

34002 Narita-cho O-arai Ibaraki 3111393, JAEA, Japan

KEY WORDS: combined load, integrity assessment, plastic collapse, stainless steel, piping

The stainless steel piping is widely used in light water nuclear reactor plants and chemical

plants. By the end of 2011, approximately 30 percent of nuclear power plants in Japan had been

in operation for more than 40 years [1]. Aging of piping in old nuclear power plants is

significant concern to the safe operation of old nuclear power plants. In order to safely operate

old nuclear power plants, integrity assessment of aging piping is important and maintenance

has to be performed as necessary. Structural integrity assessment procedures for reactor

equipment are specified by the Japanese Society of Mechanical Engineers (JSME) and the

American Society of Mechanical Engineers (ASME) [2, 3]. The structural integrity evaluation

of stainless steel pipes cracked due to aging is generally performed using plastic collapse as a

failure criterion because large plastic deformation occurs in the ligament before the plastic

collapse. The plastic collapse point is obtained by the double elastic slope (DES) method and

the double elastic deformation (DED) method. A cracked pipe is typically subjected to

combined axial tension and bending in structural integrity evaluations. A circumferential crack

located in pipe cross section is more detrimental for guillotine break than an axial crack. In

many cases, circumferential cracked pipe can be treated as a single-edge cracked pipe. In this

study, the plastic collapse strength of asymmetry multiple circumferential notched stainless

steel pipes subjected to combined axial tension and bending is

investigated experimentally and is compared with the theoretical

plastic collapse strength. In addition, the potential is discussed for

the simplification of structural integrity evaluation of multiple

cracked piping.

Schematic illustration of a pipe with asymmetry multiple

circumferential notches is shown in Fig. 1 (a). The notch angles of

the two circumferential notches are 38 and 25 degrees, respectively.

The notch separation angle between the two notches is 27 degree.

The total notch angle that included the notch separation angle is

90 degrees. Schematic illustration of a single notched pipe with

notch angle of 90 degree is shown in Fig. 1 (b). Theoretical plastic

collapse limit curves were calculated based on elastic-perfectly

plastic body to compare with the experimental collapse strength of

the pipe with asymmetry multiple circumferential notches subjected

to combined axial tension and bending.

A specimen with 200 mm length was machined from SUS304

steel pipe with 32 mm diameter and 1.5 mm thickness. Two through

wall circumferential notches with notch angle 38 and 25 degrees

were cut in the specimen by a wire saw with 320 m diameter. The

notch tip radius was 160 m. A schematic illustration of the

(a) Multiple notches

(b) Single noth

Fig. 1. Cross sections of the

(a) multiple and (b) single

notched pipes.

90°

38° 27°

t

25°

t

90°



20

statically indeterminate fracture mechanics

testing equipment is shown in Fig. 2. This

equipment can apply arbitrary combined

axial tensile and bending loads to the

structural member [4]. The bending moment

is applied to the specimen using four point

bending method. A spring and two steel

round bars (SS400) were placed inside the

pipe in order to prevent local buckling at

transverse load points. The experiments

were carried out for the various load

patterns. The experimental plastic collapse

points were obtained by DES and DED

methods.

The corrected bending stress, b/y, at

plastic collapse point is plotted as a function

of the corrected membrane stress, m/y, in

Fig. 3. Here, b, m and y are bending

stress, membrane stress and yield stress of

pipe material, respectively. Theoretical

plastic collapse limit curves calculated for

the pipes with multiple notches (Fig. 1. (a))

and the single notch (Fig. 1. (b)) are also

plotted in Fig. 3. All experimental plastic

collapse points are over the theoretical

collapse limit curve for the pipe with

multiple notches. The theoretical plastic

collapse limit curve for the pipe with multiple notches is similar to the single notch. The

integrity of the asymmetry multiple circumferential notched stainless steel pipes subjected to

combined axial tension and bending can be evaluated conservatively using the theoretical

plastic collapse strength for the pipe with multiple notches calculated based on the elastic-

perfectly plastic model. Moreover, the integrity assessment of the pipe with multiple notches

with total notch angle 90 degree subjected to combined load can be performed more easily and

conservatively using the theoretical plastic collapse strength for the pipe with single notch with

notch angle 90 degree.

REFERENCES

[1] Nuclear Safety Review for the Year 2012, IAEA, 2012, pp. 32.

[2] SME Boiler and Pressure Vessel Code, Section XI: Rules for Inservice Inspection of Nuclear

Power Plant Components, 2011 edition, July 1.

[3] MATSUBARA, M., IZAWA, S., HIRAO, N., BUSUJIMA, K., KOYAMA, T., MACHIDA. K.,

KAWADA, D., SAKAMOTO, K., NEZU, K.: Development of Testing Equipment for Studying

Statically Indeterminate Fracture Mechanics, Proceedings ICPVT-10, 2003, pp. 481-486.

Fig. 2. Statically indeterminate fracture mechanics

experimental equipment.

Fig. 3. Statically indeterminate fracture mechanics

experimental equipment.

-0.5

0

0.5

1

1.5

2

2.5

3

-0.5 0 0.5 1 1.5

Co

rrec

ted

ben

din

g s

tres

s,

b/

y

Corrected membrane stress, m/v

Ccollapse limit for multiple

Collapse limit for single

DES

DED

Horizontal potention meter

Hydraulic cylinder for tensile load

Specimen

Displacement gauge

Hydraulic cylinder for bending load

Load cell

Vartical potention meter

Gonimeter



21

DEVELOPMENT OF A CRACK OPENING DISPLACEMENT ASSESSMENT

PROCEDURE CONSIDERING CHANGE OF COMPLIANCE AT A CRACK

PART IN THIN WALL PIPES MADE OF MODIFIED 9Cr-1Mo STEEL

T. WAKAI1*, H. MACHIDA2, M. ARAKAWA2, S. YOSHIDA2, S. YANAGIHARA3,

R. SUZUKI3, M. MATSUBARA3, Y. ENUMA4

14002 Narita-cho O-arai Ibaraki 3111393, JAEA, Japan; email: [email protected] 22-37-28 Eitai Koto-ku Tokyo 1350034, TEPCO Systems Corporation, Japan 31-5-1 Tenjin-cho Kiryu 3768515, Gunma University, Japan 42-34-17 Jingumae shibuya-ku Tokyo 1500001, Mitsubishi FBR Systems Inc., Japan

KEY WORDS: leak-before-break, crack opening displacement, Mod.9Cr-1Mo steel

This paper describes a crack opening displacement (COD) assessment procedure used in

Leak-Before-Break (LBB) assessment of sodium pipes of the Japan Sodium cooled Fast

Reactor (JSFR). For sodium pipes of JSFR, the continuous leak monitoring will be adopted as

an alternative to a volumetric test of the weld joints under conditions that satisfy LBB. The

sodium pipes are made of ASME Gr.91 (modified 9Cr-1Mo steel). Thickness of the pipes is

small, because the internal pressure is very low. Modified 9Cr-1Mo steel has a relatively large

yield stress and small work hardening coefficient comparing to the austenitic stainless steels

which are currently used in the conventional plants. In order to assess the LBB behavior of the

sodium pipes made of modified 9Cr-1Mo steel, the coolant leak rate from a through wall crack

must be estimated properly. Since the leak rate is strongly related to the crack opening

displacement (COD), an appropriate COD assessment method must be established to perform

LBB assessment. However, COD assessment method applicable for JSFR sodium pipes - thin

wall and small work hardening material - has not been proposed yet.

Taking non-linearity of the material and the geometry of JSFR pipes, a series of finite

element analyses (FEA) for the pipe containing a circumferential through-wall crack was

conducted. Based on the parametric FEA results, engineering formulae for COD evaluation

were established [1].

In this method, total elasto-plastic COD, EP, was calculated by classifying the components

of COD; elastic, EE, local plastic, LP, and fully plastic, FP, as follows,

FPLPEEEP . (1)

In order to estimate these COD components, elastic, elasto-plastic and plastic FEA were

performed.

The elastic COD, EE, was formulated based on the formulae of the GE/EPRI method [3]

with minor corrections. In accordance with the GE/EPRI methods [2], the COD corresponding

to small scale yielding condition is evaluated using an effective crack angle, eff, in the elastic

COD. The effective crack angle expresses the effect of increasing the COD due to local plastic

deformation around the crack tip as the crack length increases. For large work hardening

materials, such as austenitic stainless steels, the relationship between stress and COD can be

described from elastic to plastic regions smoothly. However, for small work hardening

materials, such as modified 9Cr-1Mo steel, the COD increased sharply. Therefore, in the

proposed method, local plastic COD, LP, is calculated separating from elastic COD, EE, by

using the following equations.



22

D

qX.p

LP VRX 0

20

4

, (2)

34

2321 XmXmXmmexpVD , (3)

where is the crack half angle, X is the ratio of primary load to plastic collapse load, R is the

radius of the pipe and 0 is strain at proportional limit. p, q are material constants. The

coefficients m1, m2, m3 and m4 are given in tabular form based on the parametric FEA results.

Based on the plastic FEA results, the COD produced after reaching fully plastic conditions,

FP, was formulated as follows;

nFP XhR 20 , (4)

where and n are coefficient and exponent of Ramberg-Osgood stress-strain relation of the

material, respectively. h2 are given in tabular form based on the plastic FEA results.

Fig. 1. Statically indeterminate fracture mechanics test

machine.

Fig. 2. Comparison between calcurations and

observations.

For the verification of the COD assessment method, a series of tests was conducted under

displacement controlled condition. The experimental apparatus is shown in Fig. 1. The

specimen was a tube with 31.8 mm in outer diameter and 1.5 mm in thickness. A

circumferential through wall crack was machined by electric discharging. The specimen was

subjected to tensile load and bending moment sequentially or concurrently. Figure 2 show an

example of the comparison between calculations and experimental results. In this case, as far

as the load was small, the calculated COD was in a good agreement with observations.

Acknowledgement: This paper includes results of “Technical development program on a

commercialized FBR plant” entrusted to JAEA by the Ministry of Economy, Trade and Industry

of Japan (METI).

REFERENCES

[1] WAKAI, T. et al.: Development of LBB Assessment Method for Japanese Sodium Cooled Fast

Reactors (JSFR) Pipes (2) -Crack Opening Displacement Assessment of Thin Wall Pipes Made

of Modified 9Cr-1Mo Steel-, ASME PVP 2010-25249.

[2] Electric Power Research Institute: Crack-Opening Area Calculations for Circumferential

Through-Wall Pipe Cracks, NP-5959-SR.



23

STRESS RELAXATION SMALL PUNCH TESTING OF P92 STEEL

P. DYMÁČEK1*, F. DOBEŠ2, M. JEČMÍNKA3

1CEITEC-IPM, Žižkova 22, 61662 Brno, Czech Republic; email: [email protected] 2Institute of Physics of Materials AS CR, Žižkova 22, 61662 Brno, Czech Republic 3CAIT-VŠB-TU Ostrava, 17. listopadu 15, 708 33 Ostrava-Poruba, Czech Republic

KEY WORDS: small punch test, stress relaxation, P92 steel

Small punch tests (SPT) are using specimens of a thin disc shape prepared from a small

amount of material that can be extracted directly from the surface of exposed parts without their

damage. In these tests, a puncher penetrates through the disc specimen into a hole [1]. Two

variations of this test type seem to have a good potential for use at elevated temperatures. First,

the test in which the puncher penetrates through the disc at a given constant rate of deflection

(i.e., central deflection measured in a direction perpendicular to the disc) and the necessary

force is measured; this test is marked as CDR (constant deflection rate). It has certain analogy

with the conventional tensile test. Second, the CF test (constant force) is a test in which the

puncher penetrates under constant load and the time dependence of the deflection is measured.

This test is similar to a conventional creep test. Both tests are run up to the rupture of the disc.

As a rule, the puncher is a ceramic ball or a bar with a hemispherical tip. In application within

the field of power- or thermal-generation industry, these tests should be performed at elevated

temperatures and they should be conducted in a protective atmosphere (usually Argon).

Recently, two other types of the small punch test have emerged: (i) the tested discs are

furnished with a precisely machined groove and their testing can then be compared with Charpy

impact tests [2, 3] and (ii) the loading mechanism is adjusted and the acting force is oscillating

[4]. In this way, the fatigue mechanisms and fatigue crack propagation can be studied.

New application of SPT could be employed as the stress relaxation testing at elevated

temperatures. Basically, the specimen has to be loaded at a given deflection rate to a specific

central deflection that conforms conditions of the membrane-stretching regime. The deflection

of the disc is then held constant and the force relaxes as the elastic strain is replaced with

inelastic creep strain. The force vs. time response during relaxation can be recalculated to stress

vs. time response, differentiated and divided by elastic modulus to give the creep rate and finally

its dependence on the stress.

As an experimental material for the study was chosen the ferritic-martensitic 9% chromium

steel P92, that was taken from a pipe with outer radius 800 mm and wall thickness 78 mm.

The relaxation small punch test SPT-R was done at temperature 600°C. The specimen was

deformed at constant rate of 0.25 mm/min to a deflection uR = 1.53 mm that produced the initial

force FR = 921 N and the recording provided force-time relation. Analogous dependence was

obtained from uniaxial tensile relaxation test at initial rate of 0.8 mm/min at 600°C. In both

tests was the maximum load reached at time of about 120 s.

It is possible to determine the initial relaxation force FR, residual force FRZ and the force

drop ΔF during the time that is needed for stabilization of the force. The conversion of force to

stress in uniaxial test is done by dividing the force by specimen cross section.

In small punch relaxation test we can use the parameter Ψ that was found from relation of

creep tests on standard specimens vs. small punch specimens:

F , (1)



24

where F is applied force, is applied

stress and Ψ is parameter that relates the

force with stress for the same time to

rupture in both types of creep tests. For

steel P92 and stress 300 MPa the factor

Ψ 3.1 [5].

The relaxation curves converted to stress shown in Fig. 1 show very good agreement. The

values of forces converted to stress values of initial stress R, residual stress RZ and stress drop

are summarized in Table 1. The research will continue on SPT relaxation testing of other

materials to verify the initial promising results.

Fig. 1. Stress relaxation diagram of P92 steel at 600 °C for uniaxial tensile

test and SPT-R.

Acknowledgement: This work was partly realised in CEITEC - Central European Institute

of Technology with research infrastructure supported by the project CZ.1.05/1.1.00/02.0068

financed from European Regional Development Fund.

REFERENCES

[1] CEN Workshop Agreement CWA 15627:2007, Small Punch Test Method for Metallic

Materials, Dec. 2007.

[2] CUESTA, I. I., RODRIQUEZ, C., BELZUNCE, F. J., ALEGRE, J. M.: Analysis of different

techniques for obtaining pre-cracked/notched small punch test specimens. Engineering Failure

Analysis, Volume 18, Issue 8, (2011), pp. 2282-2287.

[3] TURBA, K., GULCIMEN, B., LI, Y. Z.; et al.: Introduction of a new notched specimen

geometry to determine fracture properties by small punch testing. Engineering Fracture

Mechanics Vol.: 78 Issue: 16 (2011), pp. 2826-2833.

[4] VILLARRAGA, M. L., EDIDIN, A. A., HERR, M., KURTZ, S. M.: Multiaxial Fatigue

Behavior of Oxidized and Unoxidized UHMWPE During Cyclic Small Punch Testing at Body

Temperature. Crosslinked and Thermally Treated Ultra-High Molecular Weight Polyethylene

for Joint Replacements, ASTM STP 1445, S.M. Kurtz, R.Gsell, and J. Martell, Eds., ASTM

International, West Conshohocken, PA, 2003.

[5] DYMÁČEK, P., MILIČKA, K., DOBEŠ, F.: Analysis of potential factors influencing the

relation between small punch and conventional creep tests. Hunické listy (Metallurgical

Journal) 2010, vol. LXIII, pp. 50-53.

Table 1 Stress characteristics of SPT-R and uniaxial stress

relaxation test of P92 steel at 600°C.

Relaxation

test R

[MPa]

RZ

[MPa]

[MPa] [%]

tR

[h]

SPT-R 297 136 161 54 5.1

Uniaxial 298 137 162 54 4.9



25

ABI TESTING OF REACTOR PRESSURE VESSEL STEEL

P. HAUŠILD1*, J. SIEGL1, A. MATERNA1

1Czech Technical University in Prague, Faculty of Nuclear Sciences and Physical Engineering, Department

of Materials, Trojanova 13, 120 00 Praha 2, Czech Republic; email: [email protected]

KEY WORDS: instrumented indentation, automated ball indentation, reactor pressure vessel steel

The reactor pressure vessel is one of the key safety components in the complex safety

assessments of nuclear power plants. The reactor pressure vessel cannot be replaced (from both

technical and economical reasons) so it often becomes the component determining the

operational safety of the nuclear power plants. WWER 440 nuclear reactor pressure vessel is

fabricated by welding of thick walled ring-type components [1]. The reactor wall contains the

base metal (chromium-molybdenum-vanadium low alloy 15Ch2MFA steel), the multilayer

welding seam (10ChMFT steel) and the two-layer cladding (25 chromium/13 nickel non-

stabilized austenitic stainless steel Sv 07Ch25N13 and at least 2 passes of 18 chromium/10

nickel niobium stabilized Sv 08Ch18N10G2B austenitic stainless steel).

Mechanical properties can present a gradient through the wall thickness, which can hardly

be assessed by conventional testing such as tensile or Charpy tests.

The elastic-plastic material properties of WWER 440 nuclear reactor pressure vessel steels

were therefore assessed by instrumented indentation tests carried out across the wall thickness

(Fig. 1). The Automated Ball Indentation (ABI) test [2] is based on multiple instrumented

indentation cycles (at the same penetration location) on a polished metallic surface by a

spherical indenter. Each cycle consists of indentation, unload and reload sequences.

Fig. 1. Microstructure through wall-thickness of WWER 440 pressure vessel with position

of indents in base metal (15CH2MFA), weld (10ChMFT) and cladding.



26

Fig. 2. Estimated yield stress determined by the ABI test in the different positions

of the weld, base metal and cladding of WWER 440 pressure vessel wall.

The representative (true) stress and (plastic) strain curves were determined by the ABI test

and the yield stress was estimated in the different positions through the wall thickness in the

weld, base metal and cladding as shown in Fig. 2. True stress – plastic strain curves presented

well defined power law hardening in the base metal, weld and cladding, which proves the

robustness of the method. Estimated yield stress (as well as the whole true stress – plastic strain

curves) is lower in the cladding (around 400 MPa) than in the base metal (around 500 MPa).

The results obtained by ABI were in a good agreement with results obtained by tensile test.

Although the ABI test is a macroscopic technique, it estimates the properties on a small volume

of material (few grains), which is particularly useful in testing e.g. welds and irregularly shaped

heat affected zones. Especially a multilayer welding seam presented a large scatter which can

hardly be assessed by conventional tensile testing.

Acknowledgement: This work was carried out with the financial support of Technology

Agency of the Czech Republic in the frame of the research project TA03011266.

REFERENCES

[1] TIMOFEEV, B., BRUMOVSKÝ, M., VON ESTORFF, U.: The certification of

15Kh2MFA/15Cr2MoVA steel and its welds for WWER reactor pressure vessels, European

Commission, EUR 24581 EN, 2010.

[2] HAGGAG, F.M., NANSTAD, R.K., HUTTON, J.T., THOMAS, D. L., SWAIN R.L.: Use of

automated ball indentation to measure flow properties and estimate fracture toughness in

metallic materials, Applications of automation technology to fatigue and fracture testing,

ASTM 1092, A.A. Braun, N.E. Ashbaugh, and F.M. Smith, Eds., American Society for Testing

and Materials, Philadelphia, 1990, pp. 188-208.



27

CHARACTERIZATION OF HETEROGENEOUS WELDMENTS

V. ŠEFL1*, R. NOVÁKOVÁ1, J. BYSTRIANSKÝ2

1Institute of Chemical Technology, Technická 5 Praha 6, Czech Republic; email: [email protected] 2Department of Metals and Corrosion Engineering; Institute of Chemical Technology; Prague, Czech Republic

KEY WORDS: heterogeneous weld, austenitic steel, carbon steel, steam, structure

Heterogeneous welds are basically two or more dissimilar metals connected together by

welding. Dissimilar welds are often required in power plant engineering as different parts of

the water circuit are made of different materials. Other application is production of surface with

specific composition for specific environments or repair of damaged parts of components.

Major concern comes from the welding process - the material is annealed during the process

thus affecting the former structure of material. These changes correspond to the

recrystallization, precipitation and also diffusion of elements from two adjacent metals, mainly

chromium and carbon. All of these phenomena have strong effect on mechanical and corrosion

properties; carbides segregated on grain boundaries and grain coarsening strongly affect the

ductility of the material, diffusing carbon and chromium change the corrosion behaviour. Even

when the diffusion is low, dissimilar metals in the weld can form a galvanic cell due to the

different corrosion resistance. This can, combined with the effect of some precipitates, result in

attack of the fussion layer between the two metals, eventually leading to failure of the

component. Incorrect material selection, welding process and subsequent thermal treatment can

also lead to formation of microcracs providing another failure mechanism.

Fig. 1. Structure of heterogeneous weld.

In this work, we used model samples of carbon 22K steel welded with various austenitic

materials and samples of heterogeneous welds cut from steam generator of VVER 440 nuclear

power plant after 20 years of exposure. Model samples were thermally treated in order to

simulate different welding conditions and aging during exposure.

Structure of all heterogeneous weldments was studied using standard metallographical

methods, their chemical composition was verified using scanning electron microscope Tescan

VEGA 3 equipped with EDS probe, the carbon distribution was studied with microanalyzer

Heat affected zone

Pe

ak te

mp

era

ture

Tp

Solidified weld

Solid-liquid transition zone

Coarse prior austenite grains +fine prior delta ferrite grains

Grain growth zone

Grain refined zone

Intercritical zone

Over-tempered region

Unaffected BM

1400

1200

1000

800

600

FGHAZ

CGHAZ

Te

mpe

ratu

re

()

°C

0 0.2 0.4 0.6 0.8 1Carbon (wt %)

L

LL

L




28

COMEBAX equipped with WDS probe. The corrosion behaviour was mostly studied by

DL-EPR method (dual-loop electrochemical potentiodynamic reactivation); portion of the

model samples were placed in an autoclave and exposed to model environment at elevated

temperature. Kinetics of oxide layer growth and their composition across the weld was studied

by the techniques described above. Vicker´s hardness test, as the only method viable for

measurements in small-areas of the weld, was used to study mechanical properties. Possibility

of galvanic coupling between the two metals with different corrosion resistance was verified.

Acknowledgement: The authors gratefully acknowledge the support by (the Institution).



29

RESISTANCE TO CORROSION CRACKING OF STEEL 100Cr6 IN HUMID

AIR UNDER HIGHER TENSILE STRESSES

S. LASEK1*, V. ČÍHAL1, M. BLAHETOVÁ1, E. KALABISOVÁ2

1VŠB-Technical University of Ostrava, 17. listopadu 15/2172, Ostrava-Poruba, 70833, Czech Republic; email: [email protected]

2SRIMT, Prague

KEY WORDS: steel 100Cr6, stress corrosion cracking, humid air, deformation

In the automotive industry are introduced and applied high-strength steels for reducing of

weight and overall costs. Under specific conditions and environments the components made of

high-strength steels can be sensitive to stress corrosion cracking (SCC) or corrosion fatigue.

Cracks and failures of unknown origin have occurred on the parts made of high-carbon chrome

100Cr6 steel. The aim of the contribution is comparison and evaluation the resistance to SCC

of selected steel under atmospheric conditions with relative humidity 40-80% at room

temperature.

For SCC testing were used the tensile type of samples (working part Ø5.0 x 33 mm) made

of 100Cr6 steel (EN 1.3505) in two steelworks (A, B). Standard chemical composition of steel

(wt. %): 0.95 - 1.05% C, 1.35-1.65% Cr, 0.25-0.45% Mn, 0.15 to 0.35% Si, Ni ≤ 0.3%,

Cu ≤ 0.3%, P ≤ 0.03%, S ≤ 0.025%. This steel can be treated by soft annealing and/or

quenching.

The values of yield stress of A steel were Rp = 360-380 MPa, and B one Rp = 490-520 MPa.

The stress corrosion tests were conducted on the samples under constant tensile stress in the

range 600-610 MPa (true stress 612–642 MPa) at room temperature in laboratory atmosphere

(35-50% rel. humidity, first test series) and in humid air at 80% r.h. (second tests).

Stereomicroscopy and scanning electron microscopy were used for the surface and

corrosion study. The microscopic valleys (grooves and lines after tooling, like a stress

concentrators) were observed on samples surface after preparation. Under test conditions,

locally rusty spots or stains were observed, while some of them appeared as cracks at

macroscopic observation, Fig. 1. During test time (3700 h) the fracture has not occurred.

(a) (b)

Fig. 1. a) Surface of sample (A3) after exposition in humid atmosphere (3700 h, 25°C, 80% r.h.). Non-uniform

corrosion, rust stains and areas. b) Detail of surface attack, microscopic pits and microcracks.

Microscopic grooves and pits were formed probably during non-uniform local corrosion

under rust spots, see Fig. 1, where possible initiation of microcracks (SCC) is also shown.

Differences between A and B steel with respect to SCC resistance were not found out. The

initiation of microcracks is caused probably by carbide particles and microscopic pits.

1mm + σ




30

Small plastic deformations (3.5-5.0% for A steel and 1.6-2.5% for B one) and low

temperature logarithmic creep was detected and measured on the samples, Fig. 2. Registered

creep deformation was smaller in average for steel with higher yield stress (B).

Fig. 2. Low temperature creep registered on the samples (second series).

The mean slow strain rate ἑ = ∆ε/∆t was calculated in the range (5-9).10-10s-1. The

recommended initial strain rate that promoted cracking of ferritic or tempered steels in water is

10-6 s-1. The resistance of tested 100Cr6 steel to SCC is relatively high under tested conditions.

Acknowledgement: This paper was created in the project No. L01203 “Regional Materials

Science and Technology Centre” – Feasibility Program. Founded by Ministry of Education,

Young and Sports of Czech Republic.

REFERENCES

[1] ASTM G 49: Standard Practice for Preparation and Use of Direct Tension Stress-Corrosion

Test Specimens. 2000, 5 p.

[2] ISO 7539-4: Corrosion of metals and alloys. Stress corrosion testing. Part 4: Preparation and

use of uniaxially loaded tension specimens. 1989.

[3] LASEK, S., BLAHETOVÁ, M., ČÍHAL, V.: Stress Corrosion Cracking Study of Steam

Generator Bolt Steel. In Workshop: Fracture Damage of Structural Parts, VŠB-TU Ostrava,

2004, pp. 103-111.

0

0,1

0,2

0,3

0,4

0,5

0,6

0,7

0 200 400 600 800 1000

def

orm

atio

n

[%]

time t [h]

A4

B3

A3 B4



31

RELATIONSHIPS BETWEEN KIC AND CVN AT TEMPERATURES LOWER

THAN NDT

M. HABASHI1*, M. TVRDY2

1Research advisor at C.N.R.S. France; email: [email protected] 2Professor at VSB – Technical University of Ostrava, Czech Republic

KEY WORDS: fracture toughness KA or KIC, impact energy CVN, spink's model, internal hydrogen,

mild steel, metastable austenitic 2404 alloyed steel, fracture features

In the upper shelf and lower shelf of the transition curves CVN-T, several results have

shown the existence of different relationships between mechanical toughness KIC measured

either by linear elastic fracture mechanisms (LFEM) or by applying J-integral JIC and impact

energy CVN (Charpy-V notch). In the upper shelf the most commonly used relationship is that

of Barsom-Rolfe:

(KIC /σy )2 = 5(CVN/σy – 0.05). (1)

However, two analytical equations have been derived respectively by: a) Rithie, Francise

and Server and b) Rithie and Horn:

KA = 2.9 σy [exp ( σf /σy - 1)]1/2 ρ1/2, (2)

KA = (3/2 σy E σf )1/2 ρ1/2, (3)

where KA the apparent mechanical toughness and ρ the notch root radius.

Applying Spink's model, and thus plotting the variation of (KA/KIC)[1+(ρ/C)1/2] against

(ρ/C)1/2 with C the notch length, the results issued from literature and obtained at the upper

shelf, have shown that all the relations are linear with slopes which increase as the fracture

stress σf or the yield strength σy is higher. Furthermore; the characteristic distance ρ0 or the

effective notch root radius can be deduced from these slopes. The objective of this work is to

verify the validity of these relations at temperatures lower than the nil ductile temperature

(NDT). Knowing that in this field of temperatures, previous results have shown that CVN is

sensitive to the addition of elements in steel, heat treatments and eventually the existence of

defects such as those caused by internal hydrogen. Mild steel, with and without internal

hydrogen and a metastable austenitic 2404 alloyed steel are studied at -196°C (liquid nitrogen).

Standard Charpy specimens with different notch root radii varying from 0 to 0.7 mm are used

to measure KIC by applying J-integral and also to measure the impact energy CVN. For all;

bending tests, with high strain rate to measure the impact energy or with very slow strain rate,

were performed and the tests temperature was -196°C. For mild steel without internal hydrogen,

the changes in (KA/KIC)[1+(ρ/C)1/2] = Θ(ρ/C)1/2 and [CVN/(CVN)0 [1+(ρ/C)1/2 ] = Φ (ρ/C)1/2

are in good agreement with those obtained, in the upper sheld, by Ritchie et al in AISI 4340

steel in two different heat treatments, figure 1. However; in the case of mild steel severely

charged with internal hydrogen and containing more than 10 ppm H2, which promotes high

density of defects in the grain boundaries (hydrogen attack), the two linear relations are not

similar. The bi-phases 2404 alloyed steel (80% acicular martensite + 20% austenite) shows that

the slopes and the critical notch root radii of the linear relations are fairly the same. The

isothermal transformation of the residual austenite γ to the martensite α' during the

measurements of KIC with low strain rate (1.6710-5 m / s) is assumed to be responsible for this

difference. However; for all three cases studied here, in the lower shelf or from the results, in

the upper shelf, obtained by Rithie et al, the effective notch root radii whether measured by

fracture toughness or by impact energy tests are about the same, figure 2. The fracture type in



32

mild steel free from internal hydrogen is by micro-cleavage while in the presence of internal

hydrogen, micro-cleavage and inter-granular features, with large cracks are observed. After

fracture toughness tests, the fracture surface of the metastable 2404 alloyed steel shows a

mixture of cleavage-inter-granular features. The main conclusion is that by applying, from now

on, the Spink's model described and used above, large dimension specimens satisfying the

standard LFEM criterion are now not necessary. In homogeneous micro-structure steels, KA or

KIC could be deduced by measuring (CVN)o.

Fig. 1. (KA/KIC) and CVN/(CVN)0 against (ρ/C)1/2. Mild Steel.

Fig. 2. ρ0 measured from (KA/KIC) and from CVN/(CVN)0 for different steels

in the upper and lower shelds.

rC

rC

rC

0 0. 0 2. 0,4 0 6. 0 8.0

1

2

3

4

50 0. 0 2. 0 4. 0 6. 0 8.

0

1

2

3

4

5

Mild Steel

T= -196°C

[(C

VN

)/(C

VN

) 0)]

1/2

[1

+(

/C)1

/2]

(KA/K

ICr C)[1+( /C)1/2]

[(CVN)/(CVN)0] r C1/2 [1+( /C)1/2]

(KA/K

IC)[

1+

(/C

)1/2

]

( /C)1/2

0 20 40 60 80 100 120 1400

20

40

60

80

100

120

140

160

180

2000 20 40 60 80 100 120 140

0

20

40

60

80

100

120

140

160

180

200

Mil

d S

teel

(H

2)

AIS

I 4

34

0 S

tee

l(12

00

-80

0°C

)

Mild

Ste

el

' 24

04

All

oy

AIS

I 4

34

0 S

tee

l(87

0°C

)

0 fro

m (

KA/K

IC), µ

mm

0 from [(CVN)/(CVN)

0)]1/2, µmr

r



33

BIOMECHANICS - PROBABILISTIC RELIABILITY ASSESSMENT OF

FEMORAL SCREWS

K. FRYDRÝŠEK1*

1VŠB-Technical University of Ostrava, Faculty of Mechanical Engineering, Department of Mechanics of Materials, Czech Republic; email: [email protected]

KEY WORDS: biomechanics, traumatology, orthopaedics, femoral neck fracture, femoral screw,

strength analyses, beam, elastic foundation, probability, Monte Carlo Method, reliability assessment

Proximal femoral neck fractures, see

Fig. 1, remain a vexing clinical problem in

traumatology and are one of the common types

of trauma, see [1], [2]. One possible treatment

method for femoral neck fractures, is the

application of femoral screws made of

Ti6Al4V or stainless steel material. This paper

therefore aims to present a numerical model

(i.e. strength analysis) of femoral screws

together with a probabilistic reliability

assessment (i.e. an application of the

Simulation-Based Reliability Assessment

(SBRA) Method, Anthill SW, Monte Carlo

Method, see [3], [4] and [5]), which is a

modern and innovative stochastic trend. The

analytical model is based on the theory of

beams on an elastic foundation, see [4], where

the bone is approximated by the elastic

foundation.

Hence, the femoral screw is resting on an

elastic foundation prescribed by stiffness

k /Pa/. Three screws of length L = 90 mm are

applied in parallel positions on the elastic

foundation, see Fig. 1a. The force

F188.3;2129.2 N acting in one beam

(screw) can be defined via total loading force

Fm /N/, see Fig. 1b, by the equation

F = Fm/n=mkmkdyng/n. The variables are as

follows: m70;145 kg is the entire mass of a

patient; km0.72;0.82 is the coefficient of

mass reduction (i.e. the mass of one lower limb

is not acting, see Fig. 1b); kdyn1;4 is the

dynamic force coefficient; g = 9.81 ms-2 is the

gravitational acceleration; and n2;3 is the

coefficient of inequality in the division of force

Fm into three screws. These variables are defined via truncated histograms (stochastic

approach); see Fig. 2 for examples. The force F can be decomposed into forces F1=Fcos

and F2=Fsin ; see Fig. 3. The femoral screw angle 5;80 deg, which is defined by the

limiting angles of adduction and abduction, and the yield limit of material Re /MPa/, are

Fig. 1. Femoral screws in femur as beams on elastic

foundation and their loading.

Fig. 2. Examples of histogram for input parameter

Re /MPa/.

Fig. 3. Example of bending moment distribution for

cannulated femoral screw (result of 1 Monte Carlo

simulation).



34

likewise defined via histograms; see Fig. 2. According to the 2nd order theory and the theory of

beams on an elastic foundation, three linear differential equations for the intervals x1,2,3, can be

written as 0F iii

2

i

2

24

i

4

ZT kvdx

vd

dx

vdEJ together with 12 boundary conditions. EJZT /Nm2/ is

flexural stiffness, i /m/ is displacement and xi /m/ are coordinates. For more information, see

5 and 6. Hence, bending moments Moi /Nmm/, see Fig. 3, and maximum stresses

MAX–1129.2; –35.5 MPa, see Fig. 4, can be calculated.

Probabilistic reliability assessment can be

carried out via the SBRA Method by means of

the reliability function RF=Re–MAX /MPa/

(depending on load capacity, compared to the

extreme stress values with yield limit, see

Fig. 4). It is then obvious that if RF < 0, plastic

deformation occurs in the beam (i.e. the yield

limit of the material is overcome; see the 2D

histogram in Fig. 4. The probability that plastic

deformation will occur in the beam is

PRF<0=2.5110–5=0.00251%.

According the theory of beams on elastic

foundation in connection with SBRA Method,

own stochastic model for strength analyses of femoral screws intended for treatment of femoral

neck fractures was derived. The probability that yield limit is exceeded is 0.00251% and it is

acceptable. Hence, the femoral screws are safe and suitable for patient treatment. The presented

results were compared with 3D FE model with adequate results (deterministic approach).

Acknowledgements: Supported by the Czech projects TA03010804 and SP2014/17.

REFERENCES

[1] FRYDRÝŠEK, K., JOŘENEK J., UČEŇ, O., KUBÍN, T., ŽILKA, L., PLEVA, L.: Design of

External Fixators Used in Traumatology and Orthopaedics – Treatment of Fractures of Pelvis

and its Acetabulum, Procedia Engineering, vol. 48, 2012, pp. 164-173.

[2] FARHAD, N., BRADLEY, E.J., HODGSON, S.: Comparison of Two Tools for the

Measurement of Interfragmentary Movement in Femoral Neck Fractures Stabilised by

Cannulated Screws, Robotics and Computer-Integrated Manufacturing, vol. 26, issue 6, 2010,

pp. 610-615.

[3] FRYDRYŠEK, K.: Probabilistic Calculations in Mechanics 1, Department of Mechanics of

Materials, Faculty of Mechanical Engineering, VŠB - Technical University of Ostrava, Ostrava,

Czech Republic, 2010, ISBN 978-80-248-2314-0, p. 1-149, written in Czech language.

[4] FRYDRÝŠEK. K., TVRDÁ, K., JANČO, R. et al: Handbook of Structures on Elastic

Foundation. VŠB - Technical University of Ostrava, Ostrava, Czech Republic, 2013,

ISBN 978-80-248-3238-8, p. 1-1691.

[5] FRYDRÝŠEK. K.: Strength Analyses of Full and Cannulated Femoral Screws Made up from

Stainless Steel and Ti6Al4V, Calculation report, FME VŠB-Technical University of Ostrava,

Ostrava, Czech Republic, 2014, pp. 1-43.

Fig. 4. 2D histogram of output parameter RF /MPa/.



35

FINITE ELEMENT HUMAN MODEL FOR CRASH SAFETY ASSESSMENT

OF AUTOMOBILE IN FRONTAL COLLISION

Y. ZAMA1*

11-5-1, Tenjin-cho, Kiryu, Gunma, Japan; email: [email protected]

KEY WORDS: crash safety, frontal collision, injury mechanism, FEM

In the assessment for crash safety of automobile, crash test dummies such as Hybrid III and

THOR are utilized commonly. However, injury assessment by the dummy is only for specific

parts, which are head, neck chest and so on. Recently, finite element human model (FE human

model) for crash safety assessment has been developed in the world. The FE human model can

mimic human structure and kinetic characteristics of human body precisely as compared with

the crash test dummy. Moreover, cost of the crash safety assessment by using the model can be

reduced more than the crash test with the dummy. Therefore, crash test simulation by the model

have been carried out in order to clarify injury mechanism and evaluate crashworthiness of

automobiles.

Since 2004, the FE human model have been

developed in the project of Japan automobile

manufacture association (JAMA) in order to use

the model as common tool for crash safety

assessment. In this study, FE human model of

occupant in the event of frontal collision was

developed as shown in Fig. 1. The model

consisted of 90,000 elements with finite

element. A physique of the model was 50%tile

of American male, and UMTRI posture was

applied as occupant posture. This development

was performed in Japan automobile research

institute (JARI) as the previous work of author. The contents of this report already have been

presented and published in a journal [1].

As for the frontal collision of automobile, chest injury frequently occurred. Bio-fidelity of

chest part in the model is important in order to investigate injury mechanism in the real world.

Bending characteristics of ribs and clavicles in the model was validated with the experiment

results of bending test of the ribs and clavicles extracted from post mortem human subject

(PMHS)[2][3]. Force-displacement curve of the rib and clavicle in the current model was far

different from that of the PMHS ribs and clavicles, and material properties of the ribs and

clavicles in the model were estimated with force-displacement curve of the experiments. Rib

and clavicle models with the estimated material properties showed the similar force-

displacement curve of the experiment.

As for compression characteristics of chest part in the model, the characteristics of the

model was validated with experiment results of table top test by using only chest of PMHS. In

the table top test [4], the chest of PMHS was compressed with seat belt, and compression ratio

derived from chest displacement was evaluated. Figure 2(a) shows relationship between chest

compression ratio and force loading to the chest with seat belt for the model and PMHS. The

compression characteristics of the model was also far different from that of PMHS. The material

properties of organs and fresh in the model were modified. As the result, the compression

characteristics of the model showed good agreement with that of PMHS as shown in Fig. 2(b).

Fig. 1. FE human model of occupant for frontal

collision.



36

(a) Current model (b) Improved model

Fig. 2. Compression charavteristics of chest for the model and PMHS.

Finally, kinematics of the model in the

event of frontal impact was validated with the

result of PMHS frontal sled experiment [5].

In this experiment, trajectories of head,

thoracic spine, pelvis, greater trochanter,

patella and malleolus during frontal impact

were measured by the three dimensional

motion capture system. From the comparison

of the trajectories with the model and PMHS,

forward and upward motions of the model did

not coincide with those of PMHS. In order to

solve the issue, geometry and tensile property

of the fresh around the pelvis in the model

were optimized by considering of PMHS

experiment results. Fig. 3 shows trajectories

of the improved model and PMHSs. As the

results, forward and upward motions of the

improved model were similar to those of

PMHS. However, rotational motions after

100 ms from the collision in the model was

different from those in the PMHS.

Acknowledgement: The author would

like to express special thanks for Dr. Ejima

and Mr. Mikami in JARI.

REFERENCES

[1] ZAMA, Y., ANTONA, J., MIKAMI, K., EJIMA, S., KAMIJI, K., YASUKI, T.: Development

of Finite Element Human Model for Events of Frontal Impact, Transactions of JSAE 41, 2010,

pp. 1243-1248.

[2] KINDIG, M.: Tolerance to failure and geometric influence on the stiffness of human ribs under

anterior-posterior loading, Master thesis, University of Virginia, 2009.

[3] DUPREY, S.: Biomechanical response of the clavicle under bending, Proc. 34th Cong. French

Society of biomechanics, 2009.

[4] KENT, R.: Frontal thoracic response to dynamic loading: the role of superficial tissues,

viscera, and the rib cage, Proc. IRCOBI Conf., 2005.

[5] SHAW, C. G, PARENT, D., PURTSEZOV, S., KERRIGAN, J. R., SHIN, J., CRANDALL,

J. R.: Frontal impact PMHS sled tests for FE torso model development, Proc. IRCOBI Conf.,

2009.

belt

table

0

500

1000

1500

2000

0 5 10 15 20Fo

rce

[N]

Compression [%]

-σ

Avg

＋σ

Model

(PMHS)

0

500

1000

1500

2000

0 5 10 15 20

Forc

e [

N]

Compression [%]

-σ

Avg

+σ

Model

(PMHS)

(a) Side view

(b) Top view

Fig. 3. Trajectories of the model and PMHS.



37

BIOMECHANICS – PROBLEMATIC LOOSENING

OF LOCKING SCREWS FROM PLATES

R. ČADA1*, K. FRYDRÝŠEK1

1VŠB Technical University of Ostrava, Faculty of Mechanical Engineering, Department of Mechanical Technology, Czech Republic; email: [email protected]

KEY WORDS: biomechanics, traumatology, orthopaedics, locking bone screw, angularly stable plate,

osteosynthesis, titanium alloy

Angularly stable plates (Fig. 1) have been used in medical and veterinary practice for the

treatment of unstable fractures for several years. The principle is based on the fixation of a

screw in the plate by means of the screw thread. The thread on the screw head must have the

same gradient as the thread on the shank; often the thread on the screw head is finer and has

multiple starts. Angularly stable plates are produced either as straight plates or as anatomically

shaped plates. Plates and bone screws

(Fig. 2) are produced from austenitic steel

W. Nr. 1.4441 according to ISO 5832-1 and

titanium alloy Ti6Al4V according

to ISO 5832-3. For both these materials the

tensile strength is almost identical, at

860 - 1050 MPa. The modulus of tensile

elasticity is 1.135 GPa for titanium alloy and

2.1 GPa for stainless steel. Titanium alloy is

thus 145 % more elastic and can be surface-

treated by anodizing, creating titanium oxide

on the implant surface.

In the operating procedure, locking bone

screws in the angularly stable plate (Fig. 3)

are manually tightened using a torque clutch

with torque 1.5 Nm. Even if this procedure is

complied with, when extracting the titanium

alloy implant it is difficult (and sometimes

impossible) to loosen some locking bone

screws from the locking holes in the

angularly stable plate (Fig. 3). Often the

extraction causes wear of the internal

hexagon in the screw head (Fig. 4), leading

to stripping. This then necessitates a

complex process of drilling off the screw

heads and removing the screw shanks from

the bone using an extraction set, which

prolongs the length of time spent by the

patient in the operating theatre.

In order to solve the above-described problem, the self-locking properties of the thread

connection were verified by calculation, and the effect of the gradient angle of the screw head

on its self-locking properties was evaluated. The experiments were carried out using a

KRAFTWERK torque screwdriver, model 2039, 1-4 Nm (±6.0 %) according to ISO 6789

(Fig. 5), which was supplied with a calibration certificate.

Fig. 1. Angularly stable plate with locking bone screws

and their application (photos a, b Čada).

Fig. 2. Locking self-tapping bone screw and detail

of screw head (photos Čada).

Fig. 3. Locking bone screw and locking hole in angularly

stable plate (photo b Čada).

Fig. 4. Internal hexagon in the head of a locking self-

tapping bone screw (a – before, b – after using a

screwdriver), (photos Čada).



38

The distances between the opposite walls of the

2.5 mm hexagonal bit used in the experiments were

measured 10× during rotation of the bit using a

Mitutoyo digital micrometer: 0-25 mm, 0.001 mm.

Subsequently the mean value and the absolute

measurement error were calculated. The torque

required to strip the internal hexagon of a 3.5 mm

locking bone screw made of titanium alloy Ti6Al4V

was determined experimentally. It was also

determined whether the rotation-slippage of the end

of a hexagonal-bit screwdriver inside the bone

screw head when stripping the internal hexagon

(Fig. 6a) causes deformation in the circumference

of the screw head and thus increases friction in the thread. A comparison was made of the

measured values of the mean bone screw head diameter both before and after the stripping of

the internal screw head hexagon; absolute measurement errors were also coampared. It was

found that twisting of the shank did not lead to the loosening of the screw head from the

angularly stable plate, but instead caused the shank to fracture, creating a level fracture surface

(Fig. 6b). Experiments were performed to determine the effect of the torque and the influence

of the duration of tightening on the ability to unscrew the bone screw from the hole in the

angularly stable plate. The suitability of using hexagonal forms in screw heads was assessed

from the perspective of the active stresses.

The results of the experiments led to

the following recommendations: the torque

should be reduced from 1.5 Nm to a lower

value; conical (self-holding) screwdrivers

should be used when inserting screws and

cylindrical (non-self-holding)

screwdrivers should be used for extracting

screws; instead of the internal hexagon, the

screw head should use the TORX system,

which gives better torque transfer even

when the screwdriver is not fully inserted

into the aperture.

Acknowledgements: The authors gratefully acknowledge the funding from the projects

TA03010804 “Osteosynthesis of leg and arm fractures”, SP2014/193 “Research and

Optimization of Technologies for Higher Utility Properties of New Materials and Mechanical

Engineering Products” and SP2014/17 “Application of numerical and experimental methods

in the field of mechanics and biomechanics”.

REFERENCES

[1] ANTOSZEWSKI, B., EVIN, E., AUDY, J.: A study of the effect of type (Cu+Ti) and (Mo+Ti)

electro-spark coatings on fricion in pin-on-disc testing, Journal of Tribology, Vol. 130, No. 1

(2008), pp. 26-31, ISSN 0742-4787.

[2] FRYDRÝŠEK, K., JOŘENEK J., UČEŇ, O., KUBÍN, T., ŽILKA, L. and PLEVA, L.: Design

of External Fixators Used in Traumatology and Orthopaedics – Treatment of Fractures of

Pelvis and its Acetabulum, Procedia Engineering, Vol. 48, 2012, pp. 164-173, ISSN 1877-7058.

Fig. 5. Torque screwdriver with 2.5 mm bit and

angularly stable plate with locking bone screw

(photo Čada).

Fig. 6. Locking self-tapping bone screw in plate

(a – stripped internal hexagon in screw head after using

screwdriver, b – fracture surface after fracture of the screw

shank), (photos Čada).



39

BIOMECHANICS – SAFETY FACTOR EVALUATION

OF ANTEROLATERAL PLATES FOR DISTAL TIBIA FRACTURES

G. THEISZ1*, K. FRYDRÝŠEK1

1Department of Mechanics of Materials, Faculty of Mechanical Engineering, VŠB – Technical University of Ostrava, 17. listopadu 15/2172, 708 33 Ostrava, Czech Republic; email: [email protected]

KEY WORDS: biomechanics, traumatology, safety factor, internal fixation, anterolateral plate,

strength analysis

Two types of fracture osteosynthesis are used in medical

practice – external and internal fixation. This paper analyzes

anterolateral plates; see Fig. 1 (producer: Medin, a.s.). The plate is

used for the internal fixation of distal tibia fractures.

The material properties of bone can be described

with sufficient accuracy for individual bone parts using

a homogeneous isotropic material model. This model can be used

to describe tissue using the modulus of tensile elasticity E /MPa/

and the Poisson number μ /1/ for individual bone parts. In order to

provide an adequate description of reality, the bone model was

divided into cortical and spongy (cancellous) tissues. These

individual parts were each divided into 4 areas (giving a total of 8

areas displaying different material properties); see Fig. 2. In the

condyle areas the bone tissue is considerably harder and stronger

than in other parts of the bone, and so it shows a higher modulus

of tensile elasticity in the model of cancellous bone.

The analysis was performed on a selected simple intra-articular

fracture of the metaphysis and joint surface defined according to

the AO classification 43-C1.1; see Fig. 3. Coulomb friction

between the non-fused bone fragments is defined with friction

coefficient 0.4.

It is very difficult to determine the force acting upon the tibia

during walking. For this reason the force F was selected,

corresponding with the entire weight of the patient mp = 100 kg,

taking into account the dynamic loading of the bone by applying

the dynamic coefficient kDyn = 1.6.

The contact between the tibia and the talus is replaced by

the boundary condition of elastic support; this boundary condition

is a suitable replacement for the cartilage and mechanical contact

with the talus.

Fig. 2. Parts of the tibia

including modulus of elasticity.

Fig. 3. Stress distribution

according to HMH theory.

Fig. 1. Anterolateral plate.

N1.15691.. gmkF pDyn (1)



40

Fig. 3 shows the distribution

of reduced stress red according to

von Mises theory for a plate

consisting of material Ti6Al4V.

This calculation is performed for

non-fused bone. The maximum

stress reaches the value

of 641.24 MPa at the hole for

the K-wire. Fig. 4 shows the detail

of the maximum stress in the hole.

Variant calculations were

carried out for fused bone

(successful treatment) and non-

fused bone (unsuccessful treatment); see Tab. 1.

With regard to the minimum yield strength of the material, the safety factor is calculated

using the following equation:

y

redMAXyk

, (2)

where redMAX is the maximum reduced stress according to von Mises theory and y

is the minimum yield strength of the given material.

The situation of maximum loading on non-fused bone is an extreme state in which

the fracture fails to heal and the patient places excessive stress on the limb (with loading even

exceeding yield strength – see Tab. 1 – causing plastic deformation). In cases of successful

treatment the safety factor kσy > 10. From this perspective the analyzed anterolateral plate can

be considered safe. In cases of unsuccessful treatment, the safety factor may be kσy < 1. In such

cases the plate is unsafe and the fracture must be re-operated.

Table 1 Safety factor.

Material

Titanium

(Ti6Al4V)

Stainless

steel

(1.4441)

Minimum material yield strength σy /MPa/ 758 690

Non-fused bone

(unsuccessful treatment)

Maximum calculated stress σred MAX /MPa/ 641.24 728.61

Safety factor kσy /1/ 1.18 0.95

Fused bone

(successful treatment)

Maximum calculated stress σred MAX /MPa/ 45.76 63.74

Safety factor kσy /1/ 16.5 10.82

Acknowledgements: This work was supported by the Czech projects TA03010804 and

SP2014/17.

REFERENCES

[1] FRYDRÝŠEK, K., JOŘENEK J., UČEŇ, O., KUBÍN, T., ŽILKA, L., PLEVA, L.: Design of

External Fixators Used in Traumatology and Orthopaedics – Treatment of Fractures of Pelvis

and its Acetabulum, Procedia Engineering, vol. 48, 2012, pp. 164-173.

[2] COWIN, C. S.: Bone Mechanics Handbook. CRC Press, Florida, USA, 2001.

Fig. 4. Detail of distribution of reduced stress in K-wire hole.



41

CYCLIC BENDING DEFORMATION AND FRACTURE OF

Al AND Al-1.0MASS%Mg ALLOY

H. IKEYA1*, H. FUKUTOMI1

1Yokohama National University, Japan; email: [email protected]

KEY WORDS: aluminum, cyclic bending, grain size, work hardening

Reduction in weight is an urgent issue for automobiles in order to improve the fuel

efficiency for the decrease of CO2 gas emission. Replacement with light materials has been

challenged on every assembly of vehicles. Aluminum alloys are light-weight high strength

materials with relatively high electric conductivity and hence much attention has been paid by

automobile and its related industries. As a structural material, the application of aluminum

alloys to car body panels has been examined. In order to use the high electric conductivity

together with the weight advantage, application of aluminum alloys to the electric wires has

been expected. The replacement of electric wires made of copper with aluminum alloys gives

also one of the solutions for the exhaustion of copper resources at the same time.

Application of aluminum wires in vehicles, however, is quite limited at present, because

that strength, toughness, and electrical conductivity of aluminum are not enough in comparison

with copper wires. In order to extend the application area of aluminum wires, it is necessary to

develop aluminum alloys with mechanical properties, especially fatigue characteristics, better

than the present aluminum wires without losing high electrical conductivity.

Although many studies have been conducted on the fatigue of aluminum and aluminum

alloys (see e.g., [1]), studies on cyclic bending fatigue in the circumstances close to the actual

vehicle-fitted condition are limited and hence the fatigue process at the cyclic bending is not

clarified experimentally enough.

In this study, deformation and fracture behavior at the cyclic bending are investigated on

aluminum and Al-1.0mass%Mg specimens with well defined microstructures, as it is known

that grain size and crystal orientation distribution give strong effects on the deformation

behavior of polycrystalline materials. In addition, pure copper is examined for comparison.

Al-1.0mass%Mg alloy was produced by melting 99.99mass% pure Al and 99.9mass% pure Mg

in a high-frequency induction furnace. The purity of aluminum and copper used for cyclic

bending tests is 99.99mass% and 99.9mass%, respectively. Wire-shaped flat plates with the

dimensions of 0.6 mm in thickness and 0.8 mm in width were produced by drawing and used

for the cyclic bending test. The cyclic bending tests were carried out using a cyclic bending

machine at room temperature. Bending angle, bending rate and the maximum bending strain

are 90°, 50 rpm, and 0.02, respectively. Grain size was controlled by heat treatments. Table 1

shows the characteristic of three materials in this study.

Table 1 Characteristic of three specimens.

Yield strength

(MPa)

Tensile strength

(MPa)

Elongation

(%)

Grain size

(µm)

Al-1.0mass%Mg 74 – 102 30 – 42 8 – 15 60 - 210

Pure-Al 31 – 51 17 – 24 7 – 19 140 – 300

Pure-Cu 35 – 60 179 – 229 24 – 40 25 – 85

Figure 1a shows the relationship between grain size and cycles to failure. The maximum

number of cycles to failure is less than 5000 cycles in the present deformation conditions. It is

seen that the number of cycles to failure increases with a decrease in grain size. Thompson et

al. [2] reported that the concentration of dislocation sources increased with an increase in grain



42

size. If the slip activity is enhanced by an increase in grain size, it is expected that the specimens

with large grain size may form a long slip line, and dislocation density becomes high. It seems

that the failure at the cyclic bending is affected by the grain size through these microstructure

changes. Effect of microstructure on the failure can be also seen in the difference of the cycles

to failure for the same grain size. In Fig. 1a, distinct difference of the cycles to failure is seen

for pure Al and Al-1.0mass%Mg with a grain size of about 150 μm and Al-1.0mass%Mg and

Cu with a grain size of about 60 μm. This suggests that the slip line formation as well as the

dislocation microstructure plays an important role for the bending fatigue behavior. According

to the result in Fig. 1a it seems that lower stacking fault energy results in the higher cycles to

failure.

Figure 1b shows the relationship between work hardening exponent and cycles to failure.

Work hardening exponent is evaluated by the stress-strain curve up to a fracture strain. It is

seen that an increase in work hardening exponent results in an increase in cycles to failure.

However, the experimental results given in Fig. 1b suggest that cycles to failure vary

independently of the work hardening exponent in the same material. This indicates that work

hardening exponent might dominate the basic number of cycles to failure and microstructure

can contribute to the improvement in the failure resistance.

Fig. 1. Relationship between the cycles to failure and grainsize (a),

and the cycles to failure and work hardening exponent (b).

REFERENCES

[1] KAMP, N., GAO, N., STARINK, M. J., SINCLAIR, I.: Influence of Grain Structure and Slip

Planarity on Fatigue Crack Growth in Low Alloying Artificially Aged 2xxx Aluminum Alloys,

International Journal of Fatigue 29 (2007) 869-878.

[2] THOMPSON, A. W., BACKOFEN, W. A.: The Effect of Grain Size on Fatigue, Acta Met.

Vol. 19 (1971) 597-605.

:Al-1.0mass%Mg

:Pure-Al

:Pure-Cu

Cycles to failure, Cycle

10

Wo

rk H

ard

en

ing

Ex

po

nen

t,n

2 103 104

0

0.2

0.4

0.8

0.6

b

:Al-1.0mass%Mg

:Pure-Al

:Pure-Cu

Cycles to failure, Cycle

10

Gra

in S

ize,

m

2 103 104

0

100

200

400

300

a



43

CYCLIC INSTABILITY OF STEEL-TITANIUM BIMETALLIC COMPOSITE

OBTAINED BY EXPLOSIVE WELDING

A. KAROLCZUK1*, T. ŁAGODA1

1Opole University of Technology, ul. Mikołajczyka 5, 45-271 Opole, Poland; email: [email protected]

KEY WORDS: explosive welding, cyclic instability, fatigue

In an explosive welding process the energy of detonation is used to join metals. Using this

unique technology metals with highly dissimilar crystal structure can be welded. This advantage

is used for example in the following industry areas: chemical manufacturing, power generation,

shipbuilding, cryogenic gas production. The bonded metals during manufacture process absorb

large portion of kinetic energy and undergo large deformation that results in high hardened thin

joint zone. Fatigue resistance and fatigue phenomena of bimetal manufactured by explosive

welding are purely recognized.

According to the experimental fatigue analysis of bimetallic composite (S355J2+N –

Titanium Grade 1) performed under push-pull loading with force controlled amplitude the

bimetallic specimens exhibit ratcheting phenomena and cyclic instability (softening) [1]. The

example changes in the total strain amplitude a in the function of damage degree n=N/Nexp

(where: Nexp is the final fatigue life, N is current number of cycles) are shown in Fig. 1.

The observed cyclic instability can

be the result of instability of titanium

grade 1 or existence of residual stresses

[2]. The aim of the present article is to

propose fatigue characteristic in the

form of relation between strain

amplitude and number of cycles

associated with the given damage

degree n.

The basic mechanical properties of

parent materials are given in Table 1.

Since the mechanical properties of

parent materials are different the stress

amplitudes generated under push-pull

loading are not equal.

Table 1 Basic mechanical materials properties.

Material ReH, MPa Rm, MPa E, GPa , - A5, %

S355J2+N 382-395 598-605 204-220 0.27-0.30 24-34

Tytanium Grade 1 189-215 (Rp02) 308-324 100-104 0.37 43-56

As a result the stress based fatigue characteristic for the investigated bimetal cannot be

created. Only the strain based fatigue characteristic is reasonable because the elongations of

both materials under push-pull loading are equal. However, the standard Manson-Coffin

characteristic cannot be applied since it is based on separation of the total strain into elastic and

plastic parts and these parts are different in both materials. Based on the mentioned reasons the

new strain based fatigue characteristic is proposed in the following form

42 22 31

p

f

p

fa NpNp , (1)

Fig. 1. The cyclic instability of investigated bimetallic

composite.

N = 22980

N = 26570

N =104820

N =895970

1

2

3

4

5

6x 10

-3

exp

exp

exp

exp

0 0.2 0.4 0.6 0.8 1



44

where p1, p2, p3, p4 are parameters to be determined experimentally. The form of equation 1 is

very similar to the Manson-Coffin equation but without division into elastic and plastic parts.

The parameters of equation 1 are calculated using experimental data, i.e. total strain amplitudes

a and number of cycles N to the given damage degree n. The identification of parameters has

been done based on the following aim functions:

42 ,2,2,, 31

pp

a inNpinNpininer , (2)

k

iinernEr

1

2, , (3)

where: i it is the subsequent specimen, k is the total number of specimen (k=12). Minimization

of function (3) was performed using the Quasi-Newton line search method. The obtained fatigue

curves for three damage degrees n=[0.1; 0.5; 0.9] are presented in Fig. 2. For comparison

purpose the Manson-Coffin curve for S355J2+N steel is also presented in Fig. 2.

Fig. 2. Identified fatigue curves for three damage degrees

n=[0.1; 0.2; 0.3] with the Manson-Coffin curve for S355J2+N steel.

Based on the perforemed analysis the following conclusion are drawn: (i) The strain based

fatigue characteristic of S355J2+N-Titanium Gr. 1clad for n = 0.5 differes in large degree when

compared to steel characteristic; (ii) The proposed characteristic allows to estimate the number

of cycles to the given damage degree of S355J2+N-Titanium Gr. 1 clad under push-pull loading

Acknowledgement: The Project was financed from a Grant by National Science Centre

(Decision No. DEC-2011/03/B/ST8/05855).

REFERENCES

[1] KAROLCZUK A., KOWALSKI M., BAŃSKI R., ŻOK F.: Fatigue phenomena in explosively

welded steel–titanium clad components subjected to push–pull loading, Int. J. Fatigue 48, 2013,

pp. 101–108.

[2] KAROLCZUK A., KLUGER K., KOWALSKI M., ŻOK F., ROBAK G.: Residual stresses in

steel-titanium composite manufactured by explosive welding, Materials Science Forum 726,

2012, pp. 125-132.

[3] KAROLCZUK A., KOWALSKI M., ROBAK G., Modelling of titanium-steel bimetallic

composite behaviour under mechanical cyclic loading, Solid State Phenomena 199, 2013,

pp. 460-465.



45

INCLUDING OF RATIO OF FATIGUE LIMITS FROM BENDING AND

TORSION FOR ESTIMATION FATIGUE LIFE UNDER CYCLIC LOADING

M. KUREK1*, T. ŁAGODA1

1Faculty of Mechanical Engineering, Department of Mechanics and Machines Design, Opole University of Technology, ul. Mikołajczyka 5, 45-271 Opole, Poland; e-mail: [email protected]

KEY WORDS: fatigue life, multiaxial criteria

In the literature, there are many criteria of multiaxial fatigue. They are based on various

assumptions and parameters describing the process of fatigue. Among them, there is a special

group of criteria based on the concept of critical plane. Some of them in their equations take

into account the ratio of fatigue limits for bending and torsion. The paper presents the estimation

of the fatigue life under multiaxial cyclic loading of selected construction materials: two

aluminum alloys PA4 (6068) and PA6 (2017A), alloy steel S355JOWP (in past called

10HNAP) and cast iron GGG 40.

For the analysis authors used three different criteria, which are based on the concept of a

critical plane. Coefficients used in the expressions for the equivalent stresses are calculated on

the basis of classical fatigue limits. These are the criteria: the maximum normal and shear

stresses proposed by Macha [1] and the criterion of Carpintieri and Spagnoli [2], where the

critical angle of the plane is increased by the angle

451

2

32

af

af

, (1)

relative to the angle defined by the maximum normal stress.

General form of the equivalent stress according to the proposed criteria can be written as

),()()( tKtBtseq

(2)

where K and B are constants used to select a particular form of criteria.

Acknowledgement: The project financed from the funds of the National Centre of Science

– decision number 2011/01/B/ST8/06850.

REFERENCES

[1] MACHA E.: Generalization of fatigue fracture criteria for multiaxial sinusoidal loadings in the

range of random loadings, in: Biaxial and Multiaxial Fatigue, EGF 3 (Edited by M.W. Brown

and K.J. Miller), Mechanical Engineering Publications, London 1989, pp. 425–436.

[2] CARPINTERI A., SPAGNOLI A.: Multiaxial high–cycle fatigue criterion for hard metals, Int J

Fatigue 23, 2001, pp. 135–145.




46



47

EVALUATION OF FATIGUE CRACK GROWTH IN ALPHA TITANIUM

ALLOYS

O. UMEZAWA1*, M. HAMADA2, T. TATSUMI2

1Faculty of Engineering, Yokohama National University, Hodogaya, Yokohama, 240-8501, Japan; email: [email protected]

2Graduate School of Engineering, Yokohama National University, Hodogaya, Yokohama, 240-8501, Japan

KEY WORDS: subsurface crack, fatigue crack propagation, titanium alloy

Subsurface crack generation in the high-cycle fatigue of -titanium alloys is dominant at

lower stress regime and lower temperature [1]. The crack initiation sites appear at

crystallographic facets such as the (0001) transgranular cracking [2]. The fatigue crack growth

modelling that based on linear fracture mechanics under the Mode I condition provided a good

estimate of the stress intensity range of subsurface or surface and fatigue crack growth, enabling

the estimation of the crack propagation life for Ti-6Al-4V alloys [3]. The fatigue crack growth

rate calculated using the Paris rule, da/dN=C(KI)m, almost corresponded to the one obtained

from the analysis of the striation on the fracture surface. The calculated crack propagation life

was less than a tenth of the number of cycles to failure over 106. As a result, the subsurface

crack initiation (Stage I crack generation) process consumed a large number of cycles to failure.

In the present study, this evaluation was applied to or near -type titanium alloys with various

morphologies.

Three kinds of commercially pure titanium, i.e. CPTi JIS type 1 (O: 0.10 mass%),

2 (O: 0.13 mass%) and 3 (0.20 mass%), and four kinds of near α-type Ti-Fe-O (Fe: 0.994,

O: 0.386 mass%) materials, i.e. T specimen (parallel to transverse direction (TD)), L specimen

(parallel to rolling direction (RD)), CR specimen (cross-rolling) and CS specimen (grooved-

rolling) were used. Each material was hot-rolled and annealed. The L, T and CR materials

showed the pancaked grains elongated in both RD and TD, which were aligned in prior

grain. The CR showed equiaxed grains with {hkl}<110> fibre texture. Force-controlling tests

were done at 77 K and 293 K using a servo-hydraulic fatigue machine. The sinusoidal

waveform forcing was uniaxial with a minimum-to-maximum stress ratio, R (σmin/σmax), of

0.01. The fractured specimens showing subsurface fatigue crack initiation were chosen for.

Fatigue crack initiation sites and fracture surfaces were analysed by scanning electron

microscopy. In Ti-Fe-O materials, the planes of aligned facets in an initiation site showed

almost the same inclination against the principal stress axis and microcracking and/or its growth

[4]. Then the initiation site was approximated for an ellipse as initial crack [3]. The crack length,

2a, crack width, 2c, and distance from specimen surface to centre of ellipse, d, were determined,

where the direction of crack length was parallel to the initial crack propagating direction.

Maximum fatigue crack size in Stage II was taken from the ripple mark on fracture surface.

Striation marks on propagating plane were characterized experimentally to examine crack

growth rate, da/dN. A software system “SCAN” based on linear fracture mechanics was

adopted in the system for investigating subsurface crack [5], and then the crack size resulted

from its position gave the stress intensity factor range, KImax = KImax-KImin, using the SCAN.

Figure 1 shows da/dN-KImax relationship for Ti-Fe-O alloy failed at 293 K. Although the

crack growth rate of the L, T and CR were scattered because of their aligned grain

microstructure, their crack growth rates were approximately the same. The CS showed higher

the crack growth rate than the others. The crack growth rates among three CPTi alloys were

also almost the same where no influence of strength on the rate was detected.



48

Based on the da/dN-KI relationship, fatigue propagation cycle, Np, was calculated. The

Np was a smaller from a tenth to a hundredth than the number of cycle to failure, Nf. As a result,

the subsurface crack initiation process also consumed a large number of cycles to failure in

titanium alloys as well as Ti-6Al-4V alloys.

Fig. 1. Relationship between da/dN and ΔKImax for Ti-Fe-O alloys failed at 293 K.

REFERENCES

[1] UMEZAWA, O., NAGAI, K.: Subsurface Crack Generation in High-cycle Fatigue for High

Strength Alloys, ISIJ International 37, 1997, pp. 1170-1179.

[2] YOKOYAMA, H., UMEZAWA, O., NAGAI, K., SUZUKI, T., KOKUBO, K.: Finite Element

Analysis of Composite Materials, Boca Raton: CRC Press 2008. Cyclic Deformation,

Dislocation Structure and Internal Fatigue Crack Generation in Ti-Fe-O Alloy at Liquid

Nitrogen Temperature, Metallurgical Materials Transactions A 31A, 2000, pp. 2793-2805.

[3] HAMADA, M., UMEZAWA, O.: Evaluation of Subsurface Fatigue Crack Life in Forged Ti-

6Al-4V Alloys at Cryogenic Temperatures, ISIJ International 49, 2009, pp. 124-131.

[4] MORITA, M., UMEZAWA, O.: Slip Deformation Analysis Based on Full Constraints Model

for -Titanium Alloy at Low Temperature, Materials Transactions 52, 2011, pp. 1595-1602.

[5] NAKANISHI, S., IWAMATSU, F., SHIRATORI, M., MATSUSHITA, H.: Estimation of

Fatigue Crack Propagation of Subsurface Cracks by SCAN, Proceedings of the 2006 ASME

Pressure Vessels and Piping Conference PVP2006-ICPVT-11-93248, 2006.

L

T

CR

CS

LT

CRCS

10 10010

10

10

-4

-3

-2

da/

dN

(m

m/c

ycl

e)

K (MPa m)I max



49

THE CURRENT STATUS OF NEW CZECH CORROSION FATIGUE

EVALUATION PROPOSAL FOR WWER NUCLEAR POWER PLANTS

L. VLČEK1*

1Institute of Applied Mechanics Brno, Ltd., Resslova 972/3, Veveří, 602 00 Brno; email: [email protected]

KEY WORDS: low-cycle fatigue (LCF), Czech NTD A.M.E. standard, air and primary water

environment, nuclear power plant, WWER

Presented paper introduces an innovative principle of fatigue life assessment suggested for

WWER nuclear power plants. The subject of this work is to take into account the corrosion

environment influence in actual methodology of low-cycle fatigue assessment and prediction.

Due to Czech nuclear power plants operator requirement the project focused on base steel

materials, which are used in primary circuit of WWER-440, started in 2010. The basic idea of

Czech environmental fatigue correction factor has been introduced on international PVP

conference in 2013 [1]. The new project linked to the previous one is focused on the additional

area of welding joints. The aim of this paper is to summarize the current status of the Czech

proposal of corrosion fatigue assessment and prediction. Assessment procedures used for

fatigue life evaluation are stated in NTD A.M.E. standard [2]. The purpose is to take into

account the influence of primary water corrosion environment on fatigue life of components

and piping. The decrease of fatigue life due to primary water environment is generally realized

by so called fatigue life environmental correction factor. Such correction factor was originally

introduced in NUREG documents [e.g. 3] as a ratio of fatigue life in air at reference temperature

conditions to fatigue life in water at operating temperature conditions (FEN = Nair, RT / Nwater).

Such way defined environmental correction factor can’t be directly used for fatigue life

assessment and prediction under operating conditions of WWER nuclear power plants. Reasons

are lying on the side of different way of fatigue life assessment and prediction, which is used

on the WWER power plants. Therefore the redefinition of environmental correction factor FPR

was introduced as a ratio of total strain amplitude in air at operating temperature condition to

total strain amplitude in water at operating temperature condition [4]:

wateratairatPRF , (1)

where at air is total strain amplitude in air at operating temperature, at water is total strain

amplitude in water at the same operating temperature as at air.

Based on the new definition there were constructed dependencies of total strain amplitude

vs. environmental correction factor FPR. Environmental correction factor is related to the total

strain coming not from fatigue design curve, but from fatigue curve without the application of

safety factors on stress n = 2 and number of cycles nN =10. Dependencies at air vs. FPR were

constructed for the case of minimal (theoretical) influence and maximal (theoretical) influence

of primary water environment on fatigue life. Theoretical minimal and maximal influence of

corrosion environment on fatigue life is covered by design fatigue curves (so called S-N curves)

proposed by Russian authors [5]. With the aim of direct application in the frame of actual

mathematical description, which describe relations of S-N design curves, the coefficient for

water corrosion environment PR can be defined as the reciprocal value of FPR (PR = 1/FPR).

The project is completed by experimental verifications of proposed environmental

correction factor. Experimental work is based on LCF strain-controlled tests in primary water

environment of WWER-440. In the frame of finished project the dependency at air vs. FPR was

verified for base material, which is austenitic stainless steel 08CH18N10T (AISI 321). LCF



50

tests done in corrosion environment of primary water demonstrated that the use of theoretical

maximal correction proposed for austenitic stainless steel seems to be unfounded. Much closer

to the reality of fatigue life reduction due to effect of corrosion environment for base metal

08CH18N10T is the minimal theoretical proposal of correction.

General proposal of fatigue life assessment and prediction in water environment covers not

only the base steel materials, but also their welding joints. The subject of actually running

theoretical-experimental program covers similar metal welds of austenitic stainless steel

08CH18N10T. Moreover LCF tests in corrosion environment of dissimilar metal welds are

under preparation.

REFERENCES

[1] VLČEK, L.: Corrosion Fatigue Evaluation of Austenitic Stainless Steels: the New Proposal to

the Czech Standard in the Area of Nuclear Power Plants Type WWER, PVP 2013, July 14-18,

Paris, France.

[2] NTD A.M.E. standard: Normatively Technical Documentation of Association of Mechanical

Engineers, Section III, Strength assessment of equipment and piping of nuclear power plant type

WWER, NTD_ASI_Sekce_III_2013, č. 1 (in Czech).

[3] Effect of LWR Coolant Environments on the Fatigue Life of Reactor Materials, Final Report,

Argonne National Laboratory, U.S. Nuclear Regulatory Commission Office of Nuclear

Regulatory Research Washington, DC 20555-0001, NUREG/CR-6909, ANL-06/08, February

2007.

[4] VLČEK, L.: Fatigue life assessment of equipment under corrosion environment conditions,

stage I: Analysis of Russian Environmental fatigue Correction of Fatigue Design Curves and

Comparison with American Approach, Proposal of Environmental Correction Factor, IAM Brno

report No. 4736/10, Brno, 2010. (in Czech).

[5] FILATOV, V. M., EVROPIN, S. V., Strength calculation of NPP equipment and pipelines

during operation. Low- and high-cycle corrosion fatigue, International Journal of Pressure

Vessels and Piping, Vol. 81, 2004.



51

EFFECT OF REPEATED HEATING ON ONE-POINT ROLLING CONTACT

FATIGUE OF HIGH-CARBON HIGH-CHROMIUM STEEL BAR

K. MIZOBE1*, R. SEGAWA1, T. SHIBUKAWA2, K. KIDA1

1University of Toyama, Gofuku 3190, Toyama, 930-8555, Japan; email: [email protected] 2YSK Co., Ltd. 3103-6 Kitanokawachi-Otsu Nishimatsuura-gun Arita-Chou Saga 849-4166 Japan

KEY WORDS: single-ball rolling contact fatigue, repeated heating, SUJ2, grain refinement

Bearings are used under severe loadings. Flacking failure which is one of the bearing

fractures occur under contact stress. It is difficult to calculate the driving stress of crack tip

because the subsurface crack propagated under compressed stress field.

Lumdberg and Palmgren [1] found that the lifetime of bearings depended on the subsurface

crack initiation process and crack initiation occurred in late stages of bearing lifetime (106,

107 cycles). The present design criteria of the bearings is based on their concept. However, in

1999, Nélias’ research group [2] discovered that some cracks appeared during the early stages

of bearing lifetime (104, 105 cycles). According to their research, we need to focus on the two

bearing lifetime groups, one is the crack initiation at the low cycle stage (104, 105 cycles) and

the other was the crack propagation at the high cycle stage (106, 107 cycles).

Our research groups [3-4] developed the single-ball rolling contact fatigue (RCF) testing

machine in order to directly observe the crack initiation and propagation under contact stress.

This method has many advantages, such as, the single-ball contact stress, the simple shape of

the specimen and the large observation area. Generally, a bearing specimen of the RCF consists

of a retainer and two races. The retainer which includes three balls is sandwiched with the two

races. Fig. 1 shows the subsurface observation process of the RCF test. The balls rotate along

the upper race groove. The small contact area can be observed by straight cutting line. In the

single-ball RCF, we performed rolling contact fatigue between one ball and one shaft bar.

Single-ball RCF enables to observe the large contact area by sectioning the specimen.

In our previous work, we focused on the effect of repeated quenching on the rolling contact

fatigue. Repeated heating was widely used as the refinement method since Grange [6-7], who

investigated the relation between repeated heating and material strength of low carbon steel. In

the case of high-carbon high-chromium steel (JIS-SUJ2), refining the prior austenite grain size

improves material strength. We have investigated the prior austenite grain refinement of SUJ2

material by repeated heating [8-9]. In this study, we performed the single-ball RCF test of

repeatedly-heated SUJ2 bar and observed cracks originating from the non-metallic inclusions.

Fig. 2 is a schematic illustration showing the single-ball RCF mechanism. Fig. 3 is a photograph

of the device. All tests were performed using 17 mm diameter and 300 mm length SUJ2 shafts,

and 3/8 inch diameter SUJ2 balls. RCF tests were performed at 3000 rpm and maximum

Hertzian contact stress of 5.3 GPa. We prepared once-quenched samples and three-times-

quenched samples. After the single-ball RCF tests, cracks originating from the non-metallic

inclusions on the specimen’s cross section were observed by using a KeyenceVK9700 laser

confocal microscope. Because the ball always run on the same line, the inner stress was

analysed by Sackfield and Hills’s method [5].



52

Acknowledgement: This work was supported by Grant-in-Aid for JSPS Fellows 25-4761.

REFERENCES

[1] LUNDBERG, G., PALMGREN, A.: Dynamic Capacity of Roller Bearings (Generalstabens

litografiska anstalts förlag, Sweden 1947).

[2] NÉLIAS, D., DUMONT, M. L., CHAMPIOT, F., VINCENT, A., GIRODIN, D., FOUGERES,

R. FLAMAND, L.: Journal of Tribology, Vol. 121, No. 2, (1999), pp. 240-251.

[3] ROZWADOWSKA, J., KIDA, K., SANTOS, E. C., HONDA, T., KANEMASU, K.,

HASHIMOTO, K.: Advanced Materials Research, Vols. 418-420, (2011), pp. 1613-1617.

[4] HAZEYAMA, S., ROZWADOWSKA, J., KIDA, K., SANTOS, E. C., HONDA, T.,

KANEMASU, K., SHIBUKAWA, T.: Advanced Materials Research, Vol. 566, (2012),

pp. 182-186.

[5] SACKFIELD, A., HILLS, D. A.: The Journal of Strain Analysis for Engineering Design,

Vol. 18, No. 2, (1983), pp. 101-105.

[6] GRANGE, R. A., SHACKELFORD, E. R.: Method of Producing Fine Grained Steel, 1966, US

patent 3, 278, 345.

[7] GRANGE, R. A.: Effect of microstructural banding in steel, Metallurgical and Materials Trans.

A 2, 1971, pp. 417-426.

[8] MIZOBE, K., SANTOS, E. C., HONDA, T., KOIKE, H., KIDA, K.: Observation of non-

metallic inclusions on repeatedly quenched SAE 52100 bearing steel fracture surfaces,

International journal of materials and product technology 44(3/4), 2012, pp. 227-239.

[9] MIZOBE, K., HONDA, T., KOIKE, H., SANTOS, E. C., SHIBUKAWA, T., KIDA, K.:

Relationship between repeatedly quenching and fisheye cracks around TiN and Al2O3 inclusions

in high carbon bearing steel, Material Research Innovations 18, S1, 2014, pp. S60-S65.

Fig. 1. Schematic illustration of subsurface observation under RCF.

Fig. 2. Schematic illustration of single-ball RCF mechanism and

specimen sectioning for subsurface observation.

Fig. 3. Single-ball RCF testing apparatus.

Sectioning Observation

Rotation



53

STATISTICAL ANALYSIS OF ACCIDENTS DUE TO FATIGUE AND

CORROSION AT FACILITIES PRODUCING HIGH PRESSURE GAS

T. SHIBUTANI1*, N. KASAI1, H. KOBAYASHI2, H. AKATSUKA2, T. TAKAHASHI2,

T. YAMADA2

1Yokohama National University, 79-5 Tokiwadai, Hodogaya-ku, Yokohama, 240-8501, Japan; email: [email protected]

2Institute of High Pressure Gas Safety, Hulic Kamiyacho Building, 3-13, Toranomon 4-Chome, Minato-Ku, Tokyo 105-8447, Japan

KEY WORDS: failure mode analysis, high pressure gas safety act, fatigue, corrosion

Industrial plants, such as chemical plants, oil refinery use large amount of high pressure

gas to produce the product. However, facilities to use high pressure gas generally require high

safety level because they become the deterioration by high pressure or low temperature of the

gas. Moreover, the chemical composition of high pressure gas might cause explosion, fire

accident and corrosion.

In Japan, handling high pressure gas is restricted by High Pressure Gas Safety Act. When

an accident has taken place with respect to the high pressure gas, all who handle high pressure

gas shall submit a notification report of the accident to the government. The database of

accidents has been constructed by Institute of High Pressure Gas Safety.

This study provides a statistical analysis of notification reports of accidents from 2008 to

2011. The focus of the analysis is put on accidents at the facilities producing high pressure gas.

Authors have developed a method for classifying accidents of high pressure gas by using a tree

diagram.

Accidental events are classified into explosion, fire, leakage, rupture without the leakage,

and others. Most accidental events are the leakage. The leakage of high pressure gas is classified

into three types: the leakage from the pressurized component, the small leakage from looseness

of bolts, flanges, and valves, and the leakage from other factors such as human factors. As a

result of failure mode analysis, major failure modes of the leakage from the pressurized

component are fatigue and corrosion.

Accidents due to fatigue occur at cold evaporator (CE), compressed natural gas (CNG)

stand, and refrigeration equipment. Detail analysis revealed that fatigue cracking takes place at

brazing joints of pure copper (C1220) tubes at CE and refrigeration equipment. Copper tubes

are often used since copper has an excellent thermal conductivity. However, no fatigue design

is considered for brazing joints on those heat exchangers. Temperature changing or vibration

from a compressor may cause fatigue cracking at brazing joints of copper tubes.

At CNG stands, fatigue accidents occur around the compressor. In particular, some

accidents are related to flexible tubes. The flexible tube is used for an imperfect alignment of

pipes or absorbing displacements. If the flexible tube is used to absorb the vibration of

compressor, it shall be designed to prevent the fatigue cracking. However, many flexible tubes

at CNG stands are used without the fatigue design.

As for corrosion, large number of corrosion of carbon steel under insulator were caused by

the high temperature and wet environment, and the existence of chloride ion. The increased

number of corrosion on nozzles and the pipes in small diameter has been brought because the

parts are generally not paid attention at all.



54

Since many facilities are old, it has been misunderstood that many accidents are caused by

aging. However, the detail analysis in this study pointed out that fatigue accidents are caused

by design error (the lack of fatigue design). Also, corrosion accidents shall be prevented by

appropriate inspection plan (the lack of corrosion management).

Acknowledgement: The authors would like to thank T. SANO, and Y. Ueda for their support

for accident data arrangement.



55

INFLUENCE OF INDUCTIVE HARDENING ON WEAR RESISTANCE IN

CASE OF ROLLING CONTACT

M. ŠOFER1*, R. FAJKOŠ1, R. HALAMA1

1VŠB-TU Ostrava, 17. listopadu 15/2127, 708 33 Ostrava-Poruba, Czech Republic; email: [email protected]

KEY WORDS: rolling contact fatigue, wear, ratcheting, R8T

The main aim of presented paper is to show how heat treatment, in our case inductive

hardening, will affect the wear rates as well as the ratcheting evolution process below contact

surface in the field of line rolling contact. Used wear model is based on shear band cracking

mechanism [1] and non-linear kinematic and isotropic hardening rule of Chaboche and

Lemaitre. The entire numerical simulations have been realized in C# program. Results from

numerical simulations are subsequently compared with experimental data and metallographic

analysis.

All rolling contact wear tests were

performed in the Rolling Contact Fatigue

Laboratory at the Department of Mechanics

of Materials of VŠB-Technical University

of Ostrava on TUORS testing device [2]. All

eight samples of wheel specimen were made

of R8T steel, whereas the rail specimens

were made of class C steel.

The wheel specimens were organized

into four sets according to the location of

sample´s collection from railway wheel rim

and application of mentioned heat treatment.

The Hertzian contact pressure and the

creepage were 1200 MPa and 0.75%

respectively. All the wheel specimens realized 105 cycles in total. After each wear test, the

weight and diameter loss of the wheel specimen have been measured. The findings from

metallographic analysis of wheel specimen after realized 105 cycles were subsequently used in

the evaluation process of performed numerical study.

Exploited wear model uses shear band cracking mechanism, which is capable of predicting

the wear process according to the accumulation of plastic shear strain below contact surface.

For ratcheting prediction in particular depths below contact surface and in case of rolling/sliding

two-dimensional contact, the authors have used a non-linear kinematic hardening rule,

introduced by Lemaitre and Chaboche [3]. The authors in the numerical simulations also took

into account the variability of friction coefficient, which significantly influences the evolution

of ratcheting in early stage of the experiment.

The main aim was to compare experimentally and numerically obtained results with respect

to plastic shear deformation profile in the active material layer and the values of wear rates after

specified number of cycles. Relatively good conformity was found between these two

approaches.

Acknowledgement: This work has been elaborated in the framework of the project

Opportunity for young researchers, reg. no. CZ.1.07/2.3.00/30.0016, supported by Operational

Fig. 1. TUORS testing device.



56

Programme Education for Competitiveness and co-financed by the European Social Fund and

the state budget of the Czech Republic and in the framework of the IT4Innovations Centre of

Excellence project, reg. no. CZ.1.05/1.1.00/02.0070, supported by Operational Programme

Research and Development for Innovations and funded from the Structural Funds of the

European Union.

REFERENCES

[1] MAZZU, A.: A simplified non-linear kinematic hardening model for ratcheting and wear

assessment in rolling contact, Journal of Strain Analysis 43, 2008, pp. 349–360.

[2] HALAMA, R., FAJKOŠ, R., MATUŠEK, P., BÁBKOVÁ, P., FOJTÍK, F., VÁCLAVEK, L.:

Contact defects initiation in railroad wheels – Experience, experiments and modelling, Wear

271, 2011, pp. 174-185.

[3] LEMAITRE, J., CHABOCHE, J., L.: Mechanics of solid materials, Cambridge University

Press, Cambridge, 1994.



57

EFFECT OF SURFACE QUALITY OF MACHINED RAILWAY WHEELS ON

FATIGUE STRENGTH

R. FAJKOŠ1*, T. TKÁČ2

1 VŠB-TU Ostrava, 17. listopadu 15/2127, 708 33 Ostrava-Poruba, Czech Republic; email: [email protected]

2 BONATRANS GROUP a.s., Revoluční 1234, 735 94 Bohumín, Czech Republic

KEY WORDS: fatigue strength, surface layers, machining technologies, shot peening

Railway transport capacities all over the world have been growing, a phenomenon which

is accompanied by the requirement to increase axle loads of freight rolling stock. Apart from

new wheel designs for higher axle loads, growing have been also the requirements on their

safety and reliability, since these wheels are often used in extreme climactic conditions.

Cruising speeds of passenger trains have also been increasing, which likewise brings more

stringent requirements concerning the quality and safety of the supplied railway wheels.

This paper describes methods of

evaluating fatigue strength of railway

wheel webs and methods of evaluating

the quality of machined railway wheel

webs. Results of fatigue tests performed

on wheels machined in a standard way

are compared with wheels which have

been treated by shot peening, a

treatment frequently used to increase

the fatigue strength of wheel webs of

railway wheelset.

The principle of a fatigue test of

railway wheels is checking whether the

supplied wheels meet the parameters

defined in standard EN 13 262, i.e.

whether they can withstand 10 million

cycles with the test level of radial stress

amplitude set to 240 MPa at the critical

point. Schematically, this type of test is

carried out at BONATRANS GROUP

a.s., preferably on the electro-hydraulic

test equipment illustrated on 3D model

in Fig. 1 below.

In order for us to be able to qualify

the effect of shot peening and

machining quality on the resultant

fatigue strength of railway wheels on

real scale, the following experiment was

devised. In total four wheel variants

were tested, namely a wheel with an

unmachined web, a wheel with a

Fig. 1. A 3D model of the electro-hydraulic test equipment

used for fatigue strength tests of railway wheels.

Fig. 2. Comparison of stress levels of railway wheels with

different final finishing of the wheel web.



58

machined web, a wheel with an unmachined but shot peened web, and a wheel with a machined

and shot peened web.

The results of the tests for each of the above wheels with different machining technologies

and shot peening are for comparison purposes presented in Fig. 2. All the tests were carried out

on an Inova electro-hydraulic fatigue strength test machine at BONATRANS GROUP a.s.

The fatigue strength of wheels manufactured by BONATRANS GROUP a.s. is around

300 MPa, which provides an adequate spare strength capacity when conducting fatigue strength

tests at the stress amplitude level of 240 MPa required by the standard.

Wheels made from steel grades with a higher content of C (grades ER8, Class B and other),

are basically even better off because of the higher strength of their normalised structure which

develop in wheel web with higher content of C. However, this at least a 25% spare strength

capacity is not enough if the quality of the surface machining is substandard. If because of tool

post vibrations, or because of using a blunt cutting tool, or because of similar technological

shortcomings, fissures develop in the cut surface, the fatigue strength of such product decreases

rapidly.

To test the real fatigue strength of wheels machined using different technologies, designed

were flat bars. The designed shape of the test bodies allowed us to better capture the character

of stresses in the given part of the wheel, and at the same time enabled us to collect such bars

from the surface of a wheel with straight or only gently sloping fixed web. The width of the test

bar in the area of the fatigue failure was 24 mm, and the thickness of the sample was 12 mm.

Three variants of final surface treatment of the test samples collected from a wheel web were

selected for the experiment which will be described on full article.

Acknowledgement: This article has been elaborated in the framework of the project

Opportunity for young researchers, reg. no. CZ.1.07/2.3.00/30.0016, supported by Operational

Programme Education for competitiveness and co-financed by the European Social Fund and

the state budget of the Czech Republic.

REFERENCES

[1] ČSN EN 13262+A1: Railway applications–Wheelsets and bogies–Wheels–Product

requirements, May 2009.

[2] AAR M-107/M208.: AAR Manual of Standards and Recommended Practices, Wheels and

Axles, 2011.

[3] OKAGATA, Y., KIRIYAMA, K., KOAT, T.: Fatigue strength evaluation of the Japanese

railway Wheel, Fatigue Fact. Engng. Mater Struct 30, 356-371.

[4] STRNADEL, B.: Material Science II, Material degradation processes and design, Mining

Academy–Technical University of Ostrava, Ostrava 2008.

[5] MORAVEC, V.: Hardness and durability of dynamically loaded machine parts, Mining

Academy–Technical University of Ostrava, Ostrava 2007.

[6] KLESNIL, M., LUKÁŠ, P.: Fatigue at metal materials, Academia, Prague 1975, 222.

[7] BERETTA, S., CARBONI, M., LO CONTE, A., REGAZZI, D., TRASATTI, S., RIZZI, M.:

Crack Growth Studies in Railway Axles under Corrosion.

[8] BERETTA, S., CARBONI, M., FIORE, G., LO CONTE, A.: Corrosion–fatigue of A1N railway

axle steel exposed to rainwater, International Journal of Fatigue 32 (2010) 952–961.

[9] LUKÁŠ, P., KUNZ, L., WEISS, B., STICKLER, R.: Impact of short cracks and minor notches

on fatigue strength, Metal materials 5, 26, Bratislava 1988.



59

APPLICATION OF ULTRASONIC IMPACT TREATMENT (UIT) FOR

IMPROVEMENT OF FATIGUE LIFE

T. ISHIKAWA1*, K. HAYASHI2

1Nippon Steel & Sumikin Technology Co., LTD., 1-7-1 Yurakucho, Chiyoda-ku, Tokyo, 100-0006, Japan; email: [email protected]

2Nippon Steel & Sumikin Technology Co., LTD., KSP A101, 3-2-1 Sakado, Takatsu-ku, Kawasaki, Kanagawa, 213-0021, Japan

KEY WORDS: fatigue, residual stress, S-N curve, fatigue crack, repair, stress, welded joint

It is an important subject to prevent fatigue crack initiation for the structural integrity of

welded structural parts especially under cyclic loading. Post-weld treatment methods, such as

grinding, tungsten inert-gas (TIG) dressing, hammer peening, ultrasonic impact treatment (UIT)

etc., are applied to welded-toes as improving procedures against the fatigue of weld joints. The

UIT is one of the most powerful solutions for this subject, because of high improved fatigue

performance with high productivity and high durability in construction stage.

The UIT equipment consists of vibration exciter and transducer & horn, as shown in Fig. 1.

The procedure of UIT application to the welded-toe is shown in Fig. 2. The pins impact the

welded-toe with 25-30 m amplitude of 25-27 kHz vibration. The welded-toe is locally

deformed and reformed to the continuously smooth toe which can be confirmed to the eye.

Fig. 1. Tool of Ultrasonic Impact Treatment. Fig. 2. Process of UIT application to welded toe.

Figure 3 shows the example of S-N (stress versus number of loading cycles to failure)

curves of cruciform welded joint of as-welded, toe grinding, and UIT applied conditions [1].

As shown in Fig. 3, UIT can extend fatigue life ten times or more than the as-welded condition.

The mechanism of fatigue initiation property improvement by UIT is summarized in Fig. 4.

Root radius of welded toe increases from naturally formed shape (0.5-1 mm) to 3 mm as same

as pin head radius. It gives smaller stress concentration at the welded-toe. UIT also induces

compressive residual stress up to the yield strength level of steel, as shown in Fig. 5.

Furthermore, the microstructure near the welded-toe becomes ultra-fine grains by local heavy

plastic deformation.

UIT has been widely utilized in ships, bridges, earth moving equipment, crane garters, and

so on. Welded-toe grinding is well-known as improvement methods of fatigue life. For the ship-

building use, UIT has been certified by ship classifications societies as the alternative method

without any dust and less noise.

As the superior improved fatigue performance with UIT application [1, 2, 3], new S-N

curve with UIT in design codes or standards has been strongly demanded. Current guidance on

Vibration exciter

(25–27 kHz)

Pin

amp: 25–30

Transducer & Horn

weld

metalbase plate

welded toe

pin

impact by ultrasonic Continuously smooth

Welded toe formed

Before AfterApplying UIT



60

improved fatigue life prediction with hammer-peened welds, together with a modification to

the S-N curves is recently published in RP-C203

Fig. 3. Example of S-N curves of Welded joint [1]. Fig. 4. Residual stress distribution ahead of welded toe.

Fatigue Design of Offshore Steel Structures (2011) published by DNV ship classification

society. The S-N curve with UIT application can be referred as equivalent as ones with hammer-

peened welds, and may be expected more benefit in near future. In Japan, UIT has been widely

used for the bridge construction, and listed as recommended new technology of NEITES

authorized by Land, Infrastructure and Transportation Ministry.

(LTT is Low Temperature Transformation welding consumables)

Fig. 5. Mechanisms of improvement of fatigue crack initiation properties by UIT.

REFERENCES

[1] NOSE, T., OKAWA, T.: Approaches for Fundamental Principles 2: Total Solution for Fatigue

of Steel, Nippon Steel Technical Report, No.391, 2011, pp. 156-161.

[2] SHIMANUKI, H., NOSE, T.: Effect of Ultrasonic Impact Treatment on Fatigue Properties of

Structural Model, Proceeding of Japan Welding Society Vol. 81, 2007, No. 342.

[3] KAYAMORI,Y.,et.al.: Applicability of fatigue solutions to floating wind turbine structures,

International Symp. on Marine and Offshore Renewable Energy, 2013.

Distance from welded toe [mm]

Re

sid

ual S

tress a

hea

d o

f w

eld

ed

toe

, M

Pa

UIT treated

UIT treated

as welded

-600.0

-400.0

-200.0

0.0

200.0

400.0

600.0

0 2 4 8 10 12 14 166

・Root radius：sharp （0.5～1.0ｍｍ）

⇒ Stress concentration

・Residual Stress：Large tensile Res. Stress

⇒ increase appl,stress

・Microstructures：Coarse grains（ 20-50 μｍ）

⇒ Locally low strength

Increase to 3.0ｍｍ（as same as pin-tip)

⇒ Decrease Stress Concentration

Compressive Res.Stress by Plastic flow

⇒ Decrease appl.stress

Ultra-fine grains (1μｍ）⇒ increase strength

UIT Grinder LTT

TIG

○

◎

△

◎

◎－

－

－－

Radius～0.5mm WM

BMHAZ

Tensile R.S.

Radius

3mm 溶接金属

BM HAZ

Comp.Res.StressUltra-fine region

１mm

Comparison of

fatigue crack

Initiation

properties

improving effect

Before After



61

CYCLIC PLASTIC PROPERTIES OF CLASS C STEEL INCLUDING

RATCHETING: TESTING AND MODELLING

R. HALAMA1*, A. MARKOPOULOS1, M. ŠOFER1, P. MATUŠEK2

1VŠB-Technical University of Ostrava, Department of Mechanics of Materials, Centre of Excellence IT4Innovations, 17.listopadu 15, Ostrava, Czech Republic; email: [email protected]

2Bonatrans Group, Bohumín, Revoluční 1234, 735 94, Czech Republic

KEY WORDS: cyclic plasticity, ratcheting, FEM, low-cycle fatigue

Cyclic plasticity modelling of metals needs individual approach. There are specific theories

for various metallic materials including mainly phenomenological models useful for practical

applications [1].

This paper is focused on the stress-strain behavior of the Class C steel in cyclic plastic

domain and its FE simulation. An experimental study on the wheel steel specimens including

uniaxial as well as multiaxial tests has been realized in the laboratory at Department of

mechanics of materials of VŠB-TU Ostrava.

The main attention in this study was paid to study

ratcheting under nonproportional loading. The

specimens were subjected to tension-torsion tests on

the reconstructed test machine INOVA

100 kN/1000 Nm (Fig. 1) as in the previous study

performed on ST52 steel [2].

The extensometer EPSILON 3550 with 25.4 mm

gauge length was used to measure axial strain and

shear strain simultaneously. The testing specimen has

tubular testing part with outer diameter of 12.5 mm

and with inner diameter of 10 mm. The specimen was

used also for the case of uniaxial loading.

The uniaxial multistep test was performed under

strain rate of 0.01 s-1. A cyclically stable behavior of

the steel under higher amplitude loading was

observed in the uniaxial multistep test, see Fig. 2.

The load path in the tension/compression-torsion

tests was applied in accordance with McDowell’s

experiments [3] to obtain similar stress-strain history

as in a point on the surface under rolling-sliding line

contact case. All multiaxial force controlled tests

were realized under sinusoidal wave loading with

frequency of 0.1 Hz.

As a sample, results from the multiaxial

ratcheting test, which was obtained by the symmetric

tension/compression and by repeated torsion, are

presented at the Fig. 3. The case with axial stress

magnitude of 700 MPa and shear stress magnitude of

400 MPa was realized for the wheel steel Class C. As

the consequence of the repeated torsion applied to the

Fig. 1. Biaxial fatigue testing machine.

Fig. 2. Results of push-pull multistep test.



62

specimen the increase of the shear strain occurs cycle by cycle in the same direction as the

torque is applied.

The MAKOC model [4], which is based on AbdelKarim-Ohno kinematic hardening rule

and Calloch isotropic hardening rule, has been applied in subsequent finite element simulations.

The numerical results show very good prediction of stress-strain behaviour of the wheel steel.

Fig. 3. Results of multiaxial ratcheting test.

Acknowledgement: This work was supported by the European Regional Development

Fund in the IT4Innovations Centre of Excellence project (CZ.1.05/1.1.00/02.0070) and by the

OPVK project Opportunity for young researchers (CZ.1.07/2.3.00/30.0016) co-financed by the

ESF.

REFERENCES

[1] HALAMA, R., SEDLÁK, J., ŠOFER, M.: Phenomenological Modelling of Cyclic Plasticity,

Chapter in: P. Miidla (Ed.), Numerical Modelling, InTech, Rijeka, 2012, pp. 329-354.

[2] HALAMA, R., FOJTÍK, F., MARKOPOULOS, A.: Memorization and Other Transient Effects

of ST52 Steel and Its FE Description, Applied Mechanics and Materials 486, 2013, pp. 48-53.

[3] MCDOWELL, D.L.: Stress state dependence of cyclic ratchetting behaviour of two rail steels.

International Journal of Plasticity 11, 1995, pp. 397-421.

[4] HALAMA, R., ŠOFER, M., FOJTÍK, F.: Choice and Calibration of Cyclic Plasticity Model

with Regard to Subsequent Fatigue Analysis. Engineering Mechanics 19, 2012, pp. 87-97.

a x

a3

xy3



63

CHARACTERIZATION OF VERMICULITE PARTICLES AFTER

MECHANICAL TREATMENT

K. ČECH BARABASZOVÁ1,2*, G. SIMHA MARTYNKOVÁ1,2

1Nanotechnology Centre, VŠB-TU of Ostrava, 17. listopadu 15/2172, Ostrava-Poruba, Czech Republic; email: [email protected]

2IT4 Innovations Centre of Excellence, VŠB-TU of Ostrava, 17. listopadu 15/2172, Ostrava-Poruba, Czech Republic

KEY WORDS: vermiculite, atomic force microscopy, particles morphology, surface and size

The vermiculite particles are used increasingly for new functional materials. They are

strong contenders for use in polymer nanocomposites. There are many applications of

vermiculite particles as fillers (such as biopolymers nanocomposites [1, 2], lightweight additive

[3], catalyst [4], isolation, ceramics [5] etc.) since the material is natural, inexpensive and

relatively non-harmful for surrounding.

For many of applications is very important the input processing of vermiculite particles.

The vermiculite particles are normally carried out in energy intensive grinding mills such as

planetary mill, oscillating mill and jet mill. Short-time grinding of vermiculite particles requires

to the particle size reduction. But extended grinding lead to an intense structural degradation of

the lamellar shape, lateral size and particle thickness reduction and progressive amorphization

accompanied with formation of hard agglomerates. The changes of the structure and vermiculite

particle size has an influence on the properties of new nanocomposite materials.

Fig. 1. SEM images of the vermiculite particles after ball (VBb) and jet (VBj) milling.

The natural vermiculite particles from Brazil were grinding in jet (VBj) and ball (VBb)

mills. The shape of vermiculite particles has been studied using scanning electron microscopy

(SEM) and atomic force microscopy (AFM). The particles size (PS) changes were characterized

by the median particle size (d50), volume-weighted mean diameter (d43), mode diameter (dm)

and span value.

In Fig. 1 we can see that the VBb particles have the form of platelets with smooth surfaces

with the fact that individual particles showed sharp edges. After jet milling particles (VBj)

showed rounded and corrugated edges. The scanned data from AFM pictures were used for

description of size and thickness of the individual vermiculite particles. With the help of

topographic profiles were measured on the particles parameters of major length and width as

perpendicular profiles (Fig. 2).



64

Fig. 2. 3D AFM images of the vermiculite particles after ball (VBb) and jet (VBj) milling.

From PS parameters was found that grinding of original samples caused reduction of

particles (from 40 µm) after ball milling (VBb), d50 and d43 to values approx. 12 µm together

with the span value to the 1.5. The PS distribution curves had modal character.

Acknowledgement: This work was supported by the project No. SP2014/39 - Functional

nanostructured materials and CZ.1.05/1.1.00/02.0070 - IT for Innovations Centre of Excellence

project.

REFERENCES

[1] ZHANG, K., XU, J., WANG, K.Y., CHENG, L., WANG, J., LIU, B.: Preparation and

characterization of chitosan nanocomposites with vermiculite of different modification. Polymer

Degradation and Stability 94, 2009, pp. 2121-2127.

[2] GRYČOVÁ, E., ČECH BARABASZOVÁ, K.: Antibacterial properties of nanostructured

materials. Journal of Nanocomposites and Nanoceramics 3(1), 2012, pp. 7-13.

[3] LING, J., DAI, B.: TiO2 activation using acid-treated vermiculite as a support: characteristics

and photoreactivity. Applied Surface Science 258, 2012, pp. 3386-3392.

[4] QIUQIANG, CH., WU, P. DANG, Z., ZHU, N., LI, P., WU, J., WANG, X.: Iron pillared

vermiculite as a heterogeneous photo-Fenton catalyst for photocatalytic degradation of azo dye

reactive brilliant orange X-GN. Separation and Purification Technology 71, 2010, pp. 315-323.

[5] VALÁŠKOVÁ, M., SIMHA MARTYNKOVÁ, G., SMETANA, B., ŠTUDENTOVÁ, S.:

Influence of vermiculite on the formation of porous cordierites. Applied Clay Science 46, 2009,

pp. 196-201.



65

ADVANCED NUMERICAL MODELLING METHODS FOR 1D PERIODIC

PLASMONIC STRUCTURE SIMMULATIONS

L. HALAGAČKA1,2*, K. POSTAVA1, M. VANWOLLEGHEM3, B. DAGENS2, J. BEN YOUSSEF4,

J. PIŠTORA1

1Department of Physics and Nanotechnology Centre, Technical University of Ostrava, 17. listopadu 15, 708 33 Ostrava-Poruba, Czech Republic; email: [email protected]

2Institut d'Electronique Fondamentale, UMR CNRS 8622, Universite Paris-Sud XI, Orsay, France 3Institut d'Electronique, Microelectronique et Nanotechnologie, CNRS UMR 8520, Villeneuve-d'Ascq, France

4Laboratoire de Magnétisme de Bretagne, Universit de Bretagne Occidentale, EA 4522/CNRS, Brest, France

KEY WORDS: plasmonics, RCWA, magnetooptics, modeling

The RCWA method is well known approach for simulation of optical response of periodic

structures [1]. Depending on a type of simulated structure and required precision, the spectral

simulations could be time consuming. The reduction of computing time is than essential issue

in advanced simulations like a parameter sweep, simulations of thickness inhomogenity,

depolarization effects, etc. Moreover, the reduction is crucial point in data fitting with a model

where a model is recalculated over optimization algorithm.

In this paper we present our implementation of parallel RCWA method in MATLAB for

1D gratings. Our parallel RCWA split an initial spectral problem into series of individual and

independent problems. Those are solved in parallel by slaves and results are collected back by

master. The parallelization is shown on left subplot of Fig. 1 schematically. The Right subplot

shows comparison between ideal linear scaling and measured scaling of our code. A linear

scaling was achieved up to 256 CPUs.

Fig. 1. Parallelization of single spectral problem is shown schematically (left). Measured scalability of

implementation is compared with theoretical linear scaling (right).

The performance of the code is demonstrated by fitting of the optical data measured on real

sample with a developed model. The fabricated sample is the 1D periodic gold grating. The

benefit of the gold grating is, that it can support effect of excitation of surface plasmon

polaritons (SPPs). In study of p-reflectivity the excitation of the SPPs appears as a deep sharp

minima [2]. The excitation of SPPs is strongly dependent on the of incidence a beam, therefore

the structure is ideal for study of effect when the incident beam is focused; the angle of

incidence varies around central angle of incidence φ0 over certain interval <φ0-φs ,φ0+φs> .

Since the fabricated structure is not perfect the fitting of the data with model is needed in order

to determine geometry of the structure. In the numerical simulations the focused beam can be

.......

Problem definition:geometry, materials,spectral domain,...

distribution to workers

results collection

subproblem#1

subproblem#2

subproblem# ii

Parallel execution

O(n)

parRCWA

50 100 150 250200number of CPUs

1/

150

50

100

250

200

0

Scalability of spectral problem




66

approximated with series of plane waves weighted over Gaussian distribution function.

Assumption of 10 partial plane waves is enough for stable numerical simulations. On the other

hand it increases amount of eigenvalue decomposition by factor 10. Moreover, if the bandwidth

of the monochromator is too broad, depolarization occurs due to the wavelength dependence of

the optical properties of a sample. The normalized Gaussian distribution of the wavelengths

around chosen spectral point λ0 with standard deviation σw is assumed [1]. Modeling of the final

spectral resolution requires discretization of the spectral range around λ0 and calculation of the

optical response at all specific wavelengths weighted by corresponding distribution function.

For tabulated optical functions of used material we have used linear spline to obtain proper

values at any wavelength. By numerical test we found, that the use of of only three spectral

points is sufficient to describe depolarization effect from the finite bandwidth. The use of only

three spectral points, namely λ0 - σw, λ0, and λ0 + σw significantly reduces calculation time.

Acknowledgement: Partial support from the projects CZ.1.05/1.1.00/02.0070,

CZ.1.05/2.1.00/01.0040 (RMTVC), CZ.1.07/2.3.00/20.0074 (Nanobase), Czech Science

Foundation 205/11/2137 and SP2013/129 is acknowledged.

REFERENCES

[1] LI, L.: Use of Fourier series in the analysis of discontinuous periodic structures, Journal of

Optical Society of America: A, 13, 1996, pp. 1870-1876.

[2] HALAGAČKA, L., et. al.: Coupled mode enhanced giant magnetoplasmonics transverse Kerr

effect, Optics Express, 21 2013, pp. 21741-21755.

[3] GARCIA CAUREL, E., et. al.: Advanced Mueller ellipsometry instrumentation data analysis.

In Ellipsometry at the nanoscale, Springer, 2013, Engineering, p. 31.



67

MOLECULAR MODELING OF ANTIMICROBIAL NANOCOMPOSITES

D. HLAVÁČ1,2*, J. TOKARSKÝ1,2

1IT4Innovations, Centre of Excellence, VŠB-TU Ostrava, 17. listopadu 15/2172, Czech Republic;

email: [email protected] 2Nanotechnology Centre, VŠB-TU Ostrava, 17. listopadu 15/2172, Czech Republic

KEY WORDS: natural minerals, antimicrobial agents, molecular modeling, adhesion

Nowadays, growing demand of more effective antimicrobial agents is observed in many

areas involving food processing and packaging, water cleaning, hygiene or medicine.

Therefore, except finding new ones, various ways of improvements of existing agents are

searched. One of the most promising ways is the modification of administration (i.e. pure

solution, incorporation into composite material, etc.) especially by their docking on suitable

matrix because controlled release and, therefore, prolonged and more environmentally friendly

activity may be achieved. Wide range of possible matrices varying in effectivity, cost as well

as stability may be used. Natural minerals represent a reasonable choice since they are low-cost

environmentally stable materials. However, because docking capabilities of every mineral

differ from each other (in dependence on the surface area, the host-guest interaction and the

method of preparation), in the first step the selection of best ones may be done according to

knowledge of host-guest interaction. Therefore, docking capabilities of various natural minerals

were investigated using molecular mechanics and dynamics in Materials Studio modeling

environment. Various host matrices were compared according to calculated values of

interaction energies between antimicrobial agents and mineral surfaces. Obtained results were

compared to available experimental data in order to evaluate the possibility of prediction based

on knowledge of host-guest interaction.

Fig. 1. Model structures of antimicrobial agents and natural minerals: a) chlorhexidine, b) nystatin, c) kaolinite,

d) montmorillonite, e) vermiculite.

Acknowledgement: The authors gratefully acknowledge the support by the European

Regional Development Fund in the IT4Innovations Centre of Excellence project

(CZ.1.05/1.1.00/02.0070).



68



69

ANTIMICROBIAL KAOLINITE BASED NANOCOMPOSITES

S. HOLEŠOVÁ1,2*, M. HUNDÁKOVÁ1,2, E. PAZDZIORA3

1Nanotechnology centre, VŠB-Technical University of Ostrava, 17. listopadu 15/2172, 708 33 Ostrava-

Poruba, Czech Republic; email: [email protected] 2IT4Innovations Centre of Excellence, VŠB-Technical University of Ostrava, 17. listopadu 15/2172, 708 33 Ostrava-Poruba, Czech Republic

3Institute of Public Health Ostrava, Centre of Clinical Laboratories, Partyzánské náměstí 7, 702 00 Ostrava, Czech Republic

KEY WORDS: kaolinite, chlorhexidine, antimicrobial activity

The development of suitable materials with the ability to inhibit the growth of microbes is

one of the current topics of material and medical research. As far as treatment of oral infections

is concerned, the current market lacks any curative form for a local long-acting application that

would enable therapy without need to use systemic treatment. A solution might be offered by

anchoring the drug to a suitable carrier that can provide transport to the designated place in the

body, gradual release and hence suppression of side effects. Recently, increased attention is

paid to so-called inorganic carriers, often based on clay minerals. The use of clay minerals as

excipients in pharmaceutical formulations has been described by many authors [1, 2]. The

antimicrobial nanocomposites based on clay mineral montmorillonite are the most studied

systems. Our team mainly deals with investigation of antimicrobial nanocomposites, when clay

mineral vermiculite is used as a drug carrier [3, 4].

In this study we focused on antimicrobial nanocomposites based on kaolinite, which aren’t

much explored in past. Two series of nanocomposites were prepared. In the first case, kaolinite

(KAO) was used as the carrier for antibacterial drug and in the second case, kaolinite modified

with dimethyl sulfoxide (DMSO) was used. In both series, chlorhexidine dihydrochloride (CH)

acts as an active antimicrobial component. The resultant samples were characterized by X – ray

diffraction (XRD), infrared spectroscopy (IR) and scanning electron microscopy (SEM)

(Fig. 1).

Fig. 1. SEM pictures of KAO (left side) and KAO/DMSO (right side).

The antimicrobial activity of prepared composites against bacteria strains Staphylococcus

aureus, Escherichia coli and against yeast Candida albicans were evaluated by finding

minimum inhibitory concentration (MIC). The dilution and cultivation were performed on the

microtitration plate. Starting dispersion contained 10% (w/v) of nanocomposites and than this

dispersion was diluted by a threefold diluting method to concentrations 3.33%, 1.11%, 0.37%,

0.12%, 0.04% and 0.01%. A volume of 1 µl of glucose suspensions of bacterial strain was



70

added. Antimicrobial activity was monitored after the elapse of 30, 60, 90, 120, 180, 240 and

300 min, and then during 5 days, always in 24 h intervals. The MIC values in selected time

intervals for two bacteria strains and yeast are shown in Table 1.

Table 1 MIC (%) (w/v) values of prepared antimicrobial nanocomposites in selected time intervals

(90 min, 300 min, 1 day, 5 days).

Sample Staphylococcus aureus Escherichia coli Candida albicans

90 300 1 5 90 300 1 5 90 300 1 5

KAO_CH (1:1) 1.11 1.11 0.01 0.01 1.11 1.11 0.01 0.01 0.12 0.12 0.12 0.12

KAO_CH (2:1) 1.11 1.11 0.01 0.01 1.11 1.11 0.01 0.01 0.37 0.12 0.12 0.12

KAO_CH (4:1) 1.11 1.11 0.01 0.01 1.11 1.11 0.01 0.01 0.12 0.04 0.12 0.12

KAO/DMSO_CH (1:1) 0.37 0.37 0.01 0.01 1.11 0.37 0.01 0.01 0.12 0.04 0.12 0.12

KAO/DMSO_CH (2:1) 0.37 0.37 0.01 0.01 1.11 1.11 0.04 0.04 0.12 0.12 0.12 0.12

KAO/DMSO_CH (1:1) 3.33 1.11 0.01 0.01 3.33 1.11 0.01 0.01 0.12 0.12 0.12 0.12

It was found that prepared nanocomposites were very effective and they had different effect

against bacteria strains and yeast. In the case of gram-positive S. aureus we observed very good

efficiency in exposition after 24 h and longer. The MIC values decreased to the lowest

concentration 0.01% w/v. We obtained almost the same results against E. coli. All prepared

samples showed very good efficiency against yeast Candida albicans. We could observe not

only good activity in longer time intervals but the prepared samples, especially with higher CH

concentration, were already very effective at earlier time intervals. Important information was

that treatment with DMSO had not significant effect on antimicrobial activity.

These nanocomposites can be in future used for preparation of drugs for local treatment of

oral cavity with long-acting antimicrobial activity.

Acknowledgement: The authors gratefully acknowledge the support by the project

IT4Innovations Centre of Excellence, reg. no. CZ.1.05/1.1.00/02.0070.

REFERENCES

[1] CARRETERO, M.I., POZO, M., Clay and non-clay minerals in the pharmaceutical industry

Part I. Excipients and medical applications, Appl. Clay Sci. 46, 2009, pp. 73-80.

[2] AGGUZI, C., CEREZO, P., VISERAS, C., CARAMELLA, C., Use of clays as drug delivery

systems: possibilities and limitations, Appl. Clay Sci. 36, 2007, pp. 22-36.

[3] HOLEŠOVÁ, S., VALÁŠKOVÁ, M., PLEVOVÁ, E., PAZDZIORA, E., MATĚJOVÁ, K.,

Preparation of novel organovermiculites with antibacterial activity using chlorhexidine

diacetate, J. Colloid Interface Sci. 342, 2010, pp. 593-597.

[4] HOLEŠOVÁ, S., SAMLÍKOVÁ, M., VALÁŠKOVÁ, M., PAZDZIORA, E., Antibacterial

activity of organomontmorillonites and organovermiculites prepared using chlorhexidine

diacetate, Appl. Clay Sci. 83-84, 2013, pp. 17-23.



71

VOLATILE ORGANIC MOLECULES SORPTION ONTO CARBON

NANOTUBES

G. SIMHA MARTYNKOVÁ1,2*, D. PLACHÁ2, L. ROZUMOVÁ1,2, E. PLEVOVÁ3

1Nanotechnology Centre, VŠB-Technical University of Ostrava, 17. listopadu 15, 708 33 Ostrava-Poruba, Czech Republic; e-mail: [email protected]

2 IT4Innovations Centre of Excellence, VŠB-Technical University of Ostrava, 17. listopadu 15, 708 33 Ostrava-Poruba, Czech Republic

3Institute of Geonics AS CR,v.v.i. Ostrava, Studentska 1768, 708 00 Ostrava-Poruba, Czech Republic

KEY WORDS: carbon nanotubes, organics, adsorption, X-ray diffraction, molecular modelling

Studying of volatile organic compounds (VOCs) in ambient air and water sample is an

important analytical task because of VOCs’ great contribution to environmental pollution and

their potential threat to human health. The VOCs adsorbed on the adsorbent can be desorbed

by the methods of thermal desorption or solvent extraction, and then analysed using gas

chromatography (GC), or analysed for weight change at temperature using mass spectroscopy

(MS) and thermal gravimetric analysis [1].

Two types of fibrous carbons were studied:

carbon nanofibers and carbon nanotubes. Both

types were purified using acid treatment to remove

non-carbonaceous substances. The adsorption of

formaldehyde, dichlormethane and naphthalene on

the carbons was studied by experimental and

theoretical (molecular simulation) approaches.

Adsorption of organic molecules is the most

intensive at edges, places of defects or doping

atoms sites. Therefore 2 cases of doping were

studied theoretically and so phosphor and boron

atoms using molecular modelling environment of

software Accelerys.

Molecular models and experiment data are in

good agreement proving that higher adsorption has

happened in case of doped nanotubes. Average amount of organic matter was 8wt.% . Structural

changes for full system were observed using analytical methods: X-ray diffraction and infrared

spectroscopy.

Acknowledgement: We are grateful to project CZ.1.05/1.1.00/02.0070 – IT4Innovations

Centre of Excellence for financial support of this work.

REFERENCES

[1] LI, Q.L., YUAN, D.X., LIN, Q.M.: Evaluation of multi-walled carbon nanotubes as an

adsorbent for trapping volatile organic compounds from environmental samples, Journal of

Chromatography A, 1026 (2004) 283–288.

Fig. 1. Molecular model of H2 sorption onto

single wall carbon nanotubes doped with P.




72



73

OPTICAL MODELLING OF MICROCRYSTALLINE SILICON DEPOSITED

BY PLASMA-ENHANCED CHEMICAL VAPOUR DEPOSITION ON

LOW-COST IRON-NICKEL SUBSTRATE FOR PHOTO-VOLTAIC

APPLICATIONS

Z. MRÁZKOVÁ1,3*, K. POSTAVA2, A. TORRES-RIOS3, M. FOLDYNA3,

P. ROCA I CABARROCAS3, V. VODÁREK4, J. HOLEŠÍNSKÝ4, J. PIŠTORA1

1Nanotechnology Centre, Technical University of Ostrava, 708 33 Ostrava-Poruba, Czech Republic; email: [email protected]

2Department of Physics, Technical University of Ostrava, 708 33 Ostrava-Poruba, Czech Republic 3LPICM-CNRS, Ecole Polytechnique, 91128 Palaiseau, France 4Faculty of Metallurgy and Materials Engineering, Technical University of Ostrava, 708 33 Ostrava-Poruba, Czech Republic

KEY WORDS: in-situ ellipsometry, plasma-enhanced chemical vapour deposition, thin films,

crystalline silicon, solar cells

The ultimate goal of photovoltaic industry is to reduce the price per watt of generated solar

energy to achieve the grid parity. Research in photovoltaics is thus aimed to achieve low-cost

high-efficiency solar cells. The fabrication cost can be reduced by using less expensive

substrates, by deposition of thin silicon layers with good crystallinity, and by using

economically convenient methods of the deposition.

In this work we study thin microcrystalline silicon (μc-Si) films grown on a flexible

low-cost Fe-Ni alloy substrate by a low-temperature (175°) plasma-enhanced chemical vapour

deposition (PECVD) [1, 2]. Since the crystallinity and material quality of the microcrystalline

silicon change during its growth, the deposition results in an inhomogeneous material with a

rather complicated structure. In order to analyse the changing composition of this complex

material the real time spectroscopic ellipsometry has been used. In-situ ellipsometric data taken

at the photon energy from 2.8 to 4.5 eV every 50 seconds enabled us to study the evolution of

crystallinity of the microcrystalline silicon as it grows (shown in Fig. 1).

Fig. 1. Time evolution (each step correspond to 50 s) of the material composition acquired from the optical

modeling of measured in-situ data. The void, the microcrystalline silicon matrix, the crystalline and the

amorphous silicon fractions are marked in the figure, respectively.

void

c-Si

c-Si

a-Si

10 20 30 40 50 60 70 800

40

10

20

30

0

50

60

70

80

Measurement number

Volu

me f

raction (

%)



74

The transmission electron microscopy has been used to verify the conclusions from optical

modelling, confirming that (i) there is a thin nucleation layer formed at the beginning of the

material deposition on the substrate, from which silicon crystals start to grow, (ii) the volume

fraction of the crystalline silicon gradually increases as the cone crystals become larger in size,

forming a rough surface after their collisions and subsequent high crystalline fraction material

growth.

Acknowledgement: The authors acknowledge the financial support from projects

SP2014/86 and IT4Inovations CZ.1.05/1.1.00/02.0070.

REFERENCES

[1] TORRES RIOS, A., DJERIDANE, Y., NATH, M., REYDET, P. L., REYAL, J. P., ROCA I

CABARROCAS, P.: Epitaxial Growth of Crystalline Silicon on N42 Alloys by PECVD at

175°C for Low Cost and High Efficiency Solar Cells, EU PVSEC Proceedings, 2011, p. 2435.

[2] MRÁZKOVÁ, Z., TORRES-RIOS, A., RUGGERI, R., FOLDYNA, M., POSTAVA, K.,

PIŠTORA, J., ROCA I CABARROCAS, P.: In-situ spectroscopic ellipsometry of

microcrystalline silicon deposited by PECVD on flexible Fe-Ni alloy substrate for photovoltaic

applications, under review in Thin Solid Films.



75

SUBMICRON CALCIUM PHOSPHATE PARTICLES STUDY ANCHORED

ON CLAY SUPPORTS

L. PAZOURKOVÁ1*, G.SIMHA MARTYNKOVÁ1,2, M. HUNDÁKOVÁ1,2, M. VALÁŠKOVÁ1,2

1Nanotechnology Centre, VŠB – Technical university of Ostrava, 17. listopadu 15, 70833, Ostrava-Poruba, Czech Republic; email: [email protected]

2IT4 Innovations Centre of Excellence, VŠB – Technical university of Ostrava, 17. listopadu 15, 70833, Ostrava-Poruba, Czech Republic

KEY WORDS: calcium phosphate, clay mineral, wet precipitation

In recent years the synthetic calcium phosphates (mainly hydroxyapatite) are widely

studied due to their similarities to minerals occurring in human body [1]. The preparation

techniques include lot of methods [2], but utilization of clay minerals as support for calcium

phosphate is very spare [3, 4].

The aim of this study is to compare in-situ preparation of calcium phosphate with the main

component of hydroxyapatite (CPH) on pure clay minerals and sodium form of clay minerals,

to further usages and applications as biomaterial. The final samples were characterized using

X-ray powder diffraction (XRD) and scanning electron microscopy (SEM). The CPH particles

of different size and morphology were formed depending on the type of clay mineral and

chemistry of the montmorillonite (Mt) and vermiculite (Ver).

Fig. 1. Schematic for A) mixing and B) sonication preparation of calcium phosphate supported clay mineral.

The samples of CPH supported on clay minerals were prepared by wet precipitation, more

precisely by mixing and sonication. We used vermiculite and montmorillonite as supporting

clay minerals. Fig. 1 shows schema of preparation procedure.

X-ray diffraction patterns of CPH-NaMt composite (Fig. 2c, 3c) and sodium form of

montmorillonite (NaMt) were compared with CPH-Mt prepared in our previous study

(Fig. 2b, 3b) [5]. It was observed that main intensive reflections (d = 0.282 nm) of calcium

phosphate (with majority of hydroxyapatite) are present in both samples of pure and sodium

form of montmorillonite. In the case of vermiculite samples prepared by mixing method, the



76

samples with CPH show absence of d(001) reflection of pure vermiculite [5] and monoionic

sodium vermiculite (NaVer), respectively. The d(002) reflection of monoionic NaVer is shifted

from 1.225 nm to 1.471 nm of CPH-NaVer. This indicates that CPH influences clay structure

in the way of altering interlayer space of clays.

Fig. 2. XRD patterns of samples prepared using

mixing: a) CPH, b) CPH-Mt, c) CPH-NaMt.

Fig. 3. XRD patterns of samples prepared by

sonication: a) CPH2, b) CPH-Mt2, c) CPH-NaMt2.

From the SEM micrographs is evident that CPH is anchored on the surface of both clay

minerals in state of submicron particles. The CPH particles show different size and morphology

in dependence on type of clay mineral and preparation method.

Acknowledgement: The authors gratefully acknowledge the support by projects: Ministry

of Education, Youth and Sport of Czech Republic SP2014/82 and IT4 Innovations Centre of

Excellence project reg.no.cz.1.05/1.1.00/02.0070. Authors thank M. Heliová for SEM

micrographs.

REFERENCES

[1] ZHANG, J., LIU, W., SCHNITZLER, V., TANCRET, F., BOULER, J-M.: Calcium phosphate

cements for bone substitution: Chemistry, handling and mechanical properties, Acta

Biomaterialia 10, 2014, pp. 1035-1049.

[2] SADAT-SHOJAI, M., KHORASANI, M-T., DINPANAH-KHOSHDARGI, E., JAMSHIDI, A.:

Synthesis methods for nanosized hydroxyapatite with diverse structures, Acta Biomaterialia 9,

2013, pp. 7591-7621.

[3] AMBRE, A., KATTI, K.S., KATTI, D. R.: In situ mineralized hydroxyapatite on amino acid

modified nanoclays as novel bone biomaterials, Materials Science and Engineering C 31, 2011,

pp. 1017-1029.

[4] ROUL, J., MOHAPATRA, R., SAHOO, S.K., TRIBHUVAL, N.: Design and characterization

of novel biodegradable polymer-clay-hydroxyapatite nanocomposites for drug delivery

applications, Asian Journal of Biomedical and Pharmaceutical Sciences 2, 2012, pp. 19-23.

[5] PAZOURKOVÁ, L., ČECH BARABASZOVÁ, K., HUNDÁKOVÁ, M.: Preparation of

hydroxyapatite/clay mineral nanocomposite, In NANOCON 2013 5th international conference

October 16th – 18th 2013, 2013 Hotel Voroněž I Brno, Czech Republic, Tanger Ltd. Ostrava

2014, pp. 83-88. ISBN: 978-80-87294-47-5. In Press.



77

PREPARATION OF SUBMICRON PARTICLES OF BIOLOGICALLY

ACTIVE SUBSTANCES USING SUPERCRITICAL FLUIDS

D. PLACHÁ1*, T. SOSNA1, E. VACULÍKOVÁ1, M. MIKESKA1, R. DVORSKÝ2

1VŠB-Technical university of Ostrava, Nanotechnology Centre, 17. listopadu 15, 708 33 Ostrava Poruba, Czech Republic; email: [email protected]

2VŠB-Technical university of Ostrava, Faculty of Mining and Geology, 17. listopadu 15, 708 33 Ostrava Poruba, Czech Republic

KEY WORDS: nanoparticles, caffeine, aspirin, supercritical fluids, Spe-ed SFE-4

Numerous conventional methods are used to reduce the particle size, for example milling,

spray drying, re-crystallization using solvent evaporation, sieving and grinding. However, these

methods are characterized by the many disadvantages such as poor control on size, unsuitable

morphology, wide particle size distribution, exposure of particles to the locally high

temperature and loss of particles in spray drying and milling techniques [1]. Supercritical fluids

have wide range of industrial applications. Using of supercritical fluids for particle size decrease

seems to be a very effective method for many reasons, especially for working in mild

temperatures and for an ability of particle size reductions to nanometric levels.

The technique of particles micronization using supercritical fluids is convenient especially

for preparations of pharmaceuticals and other biologically active substances. Most of

pharmaceuticals are poorly soluble or insoluble in aqueous body fluid systems; this fact limits

their bioavaibility. The dissolution rate can be positively influenced by increase of particle

surface area through reduction of their size. Next prerequisites are suitable and uniform

morphology and narrow particle size distribution [1, 2].

Several supercritical fluid techniques for particles size reduction are known, such as RESS

– Rapid Expansion of Supercritical Solution, PGSS – Particles from Gas-Saturated

Solution/Suspension, GAS – Gas Anti Solvent, SAS – Supercritical Anti Solvent, SEDS –

Solution Enhanced Dispersion by Supercritical Fluid [1, 2].

The RESS technique using pure supercritical CO2 is an alternative how to quickly and

naturally reduce the particle size of various materials. Principle of the technique is: A treated

compound is dissolved in a supercritical fluid; consequently the solution is suddenly

depressurized through a nozzle and expands inside a chamber with much lower pressure. The

rapid depressurization of the supercritical phase causes decreased solubility of the solute which

precipitates as a powder in a gas phase [3].

The laboratory extraction system Spe-ed SFE-4 (Applied Separations) was used for size

reducing of biologically active substances such as caffeine (Fig. 1), aspirin and cimetidine with

application of RESS technique and supercritical CO2. This device is primarily determined to be

used as an extractor of non-polar organic compounds from solids; however the producer admits

that a nanoparticle production is possible.

The working pressure was set to 150 bar and several temperatures were applied (45, 50, 80

and 100°C). The CO2 flow was maintained at 5 l.h-1. The resulted particles of treated substances

(formed caffeine particles are presented on Fig. 2) were evaluated by using SEM, FTIR, XRD

and particle size distribution methods. Volumetric particle size distribution confirmed an

influence of the working temperatures on the particle size. The SEM evaluation is necessary to

observe morphology of formed particles (Fig. 1 and 2). No significant changes were observed

in chemical structure as confirmed by FTIR and XRD.



78

Fig. 1. Caffeine particles before treatment with

supercritical CO2.

Fig. 2. Caffeine particles after treatment with

supercritical CO2.

Acknowledgement: The authors gratefully acknowledge the support by the European

Regional Development Fund in the IT4Innovations Centre of Excellence

(CZ.1.05/1.1.00/02.0070), in the ENET Centre (CZ. 1.05/2.1.00/03.0069).

REFERENCES

[1] KESHAVARZ, A., KARIMI-SABET, J., FATTAHI, A., GOLZARY, A., RAFIEE-TEHRANI,

M., DORKOOSH, F. A.: Preparation and characterization of raloxifene nanoparticles using

Rapid Expansion of Supercritical Solution (RESS), The Journal of Supercritical Fluids, 63,

2012, pp. 169-179.

[2] SAMEI, M., VATANARA, A., FATEMI, S., NAJAFABADI A. R.: Process variables in the

formation of nanoparticles of megestrol acetate through rapid expansion of supercritical CO2.

The Journal of Supercritical Fluids, 70, 2012, pp. 1-7.

[3] SFC 526, The Micronization of Drug Particles by the Rapid Expansion of a Supercritical

Solution. Application Note. Applied Separations.



79

PREPARATION OF CARBON NANO FILLERS FOR METALIC

COMPOSITES

L. ROZUMOVÁ1*, G. SIMHA MARTYNKOVÁ1

1Nanotechnology Centre, IT4Innovations Centre of Excellence, VŠB-Technical University of Ostrava, 17. listopadu 15, 708 33 Ostrava-Poruba, Czech Republic; email: [email protected]

KEY WORDS: carbon nanotubes, silver

This paper summarises the research work carried out in the field of carbon nanotube (CNT)

metal matrix composites (MMCs). Much research has been undertaken in utilising CNTs as

reinforcement for composite material. However, CNT-reinforced MMCs have received the

least attention. These composites are being projected for use in structural applications for their

high specific strength as well as functional materials for their exciting thermal and electrical

characteristics [1].

The present paper focuses on preparation of metal matrix nanocarbon composites

(MMNCs) that include processing technique. Metal matrix nanocarbon composite (MMNCs)

was prepared high energy ball milling. Time of milling was set on 120 hours.

Fig. 1. XRD pattern of GAg100/120- enriched graphite with Ag (GAg-original sample).

Acknowledgement: We are grateful to project CZ.1.05/1.1.00/02.0070 – IT4Innovations

Centre of Excellence for financial support of this work. This paper has been elaborated in the

framework of the Nanotechnology – the basis for international cooperation project, reg. no.

CZ.1.07/2.3.00/20.0074 supported by Operational Programme 'Education for competitiveness'

funded by Structural Funds of the European Union and state budget of the Czech Republic.




80

REFERENCES

[1] BAKSHI, S. R., LAHIRI, D., AGARWAL, A.: Carbon nanotube reinforced metal matrix

composites – a review. International Materials Reviews 55, 2010, pp. 1-24.



81

TiO2 – BASED SORBENT FOR LEAD IONS REMOVAL

J. SEIDLEROVÁ1, M. ŠAFAŘÍKOVÁ2, L. ROZUMOVÁ1*, I. ŠAFAŘÍK2, O. MOTYKA1

1Nanotechnology Centre, VŠB – Technical University of Ostrava, 17. listopadu 15, 708 33 Ostrava;

email: [email protected] 2Institute of Nanobiology and Structural Biology of GCR, Na Sadkach 7, 370 05, Ceske Budejovice

KEY WORDS: TiO2, sorption, lead ions

The search of new technologies for removal of toxic metals from wastewaters has focused

the attention to sorption technologies which are based on metal adsorption on biological

materials (for example nut shells [1], maize leaves [2], tree ferns [3], grape stalk wastes [4]),

various composites or clay minerals. TiO2 is widely used in industry as the white pigment in

paints, as the filler and additive in cosmetics and pharmaceuticals, as photocatalyst, and also as

a sorbent [5]. The present study is focused on the sorption of Pb ions. Batch adsorption

experiments were carried out, aiming to remove lead ions from aqueous solutions using

nanopowder of TiO2.

Non-magnetic nanopowder of TiO2 and magnetically modified nanopowder of TiO2 were

employed for the sorption experiments. FeSO4 .7 H2O was used for synthesis of magnetically

responsive TiO2. The suspension of non-magnetic TiO2 and FeSO4 .7 H2O (at pH 12) was

treated by microwave irradiation for 10 min. The formed magnetically responsive composite

was captured using an appropriate magnetic separator or NdFeB magnet. Then magnetic TiO2

formed was air dried at ca 60°C [6]. A detailed study of the process was performed using various

concentrations of lead ions.

Non-magnetic and prepared magnetic material were characterized by using scanning

electron microscopy, X-ray diffraction methods and AFM. The particle size and specific surface

area were determined. The changes of Fe content in magnetically modified material after

sorption experiments were observed as well. A flame atomic absorption spectrometer was used

for determination the Pb and Fe concentration. Adsorption process has been modelled by

various sorption isotherms.

Acknowledgement: Authors thank to the financial support of Projects: GAČR No. 13

13709S/P503. This paper has been elaborated in the framework of the project New creative

teams in priorities of scientific research, reg. no. CZ.1.07/2.3.00/30.0055, supported by

Operational Programme Education for Competitiveness and co-financed by the European

Social Fund and the state budget of the Czech Republic and in the framework of the

Nanotechnology – the basis for international cooperation project, reg. no.

CZ.1.07/2.3.00/20.0074 supported by Operational Programme 'Education for competitiveness'

funded by Structural Funds of the European Union and state budget of the Czech Republic.

REFERENCES

[1] ORHAN, Y., BUYUKGUNGOR, H.: The removal of heavy metals by using agricultural

wastes. Water Sci. Technology, 1993, vol. 28, pp. 247-255.

[2] BABARINDE, N. A. A., BABALOLA, J. O., SANNI, R. A.: Biosorption of lead ions from

aqueous solution by maize leaf. Int. J. Phys. Science, 2006, vol. 1, pp. 23-26.

[3] HO, Y.S., CHIUB, W. T., HSUB, C. S., HUANGA, C. T.: Sorption of lead ions from aqueous

solution using tree fern as a sorbent. Hydrometallurgy, 2004, vol. 73, pp. 55-61.



82

[4] VILLAESCUSA, I., FIOL, N., MARTÍNEZ, M., MIRALLES, N., POCJ, J., SERAROLS, J.:

Removal of copper and nickel ions from aqueous solutions by grape stalks wastes. Water

Researcher, 2004, vol. 38, pp. 992-1002.

[5] PEHLIVAN, E., ALTUN, T., CETIN, S., BHANGER, M. I.: Lead sorption by waste biomass of

hazelnut and almond shell. J. Hazard. Mater. 167, 2009, pp. 1203-1208.

[6] SAFARIK, I., HORSKA, K., POSPISKOVA, K., MADEROVA, Z., SAFARIKOVA, M.:

Microwave assisted synthesis of magnetically responsive composite materials. IEEE Trans.

Magn. 49, 2013, pp. 213-218.

http://www.nh.cas.cz/people/safarik/0-IEEE-2013.pdf





83

PROPERTIES OF KAOLINITE TREATED BY DIFFERENT

TEMPERATURES

M. TOKARČÍKOVÁ1*, K. MAMULOVÁ KUTLÁKOVÁ1, J. SEIDLEROVÁ1

1Nanotechnology Centre, VŠB-Technical University of Ostrava, 17. listopadu 15/2172, 708 33 Ostrava-Poruba, Czech Republic; email: [email protected]

KEY WORDS: kaolinite, calcination, leaching test

This article deals with the influence of calcination on stability of kaolinite. Clay mineral

kaolinite Al2Si2O5(OH)4 is a suitable material for prepare photoactive composite with TiO2

nanoparticles (NPs). The composite was prepared by hydrolysis of kaolinite (SAK47) and

TiOSO4 (Precheza a.s.) as a TiO2 precursor. Prepared composite was dried at 105°C or

calcination at 600°C. Not only photoactive properties of prepared composite are important but

the stability and possible impact on the environment are important as well. Therefore,

composite stability was studied by leaching test in demineralization water and extraction agents

with different pH. However, determined concentration of leached aluminum was high,

particularly in extract obtained by leaching of calcined composite. Therefore, kaolinite

calcination was studied for deep understanding of the influence of method preparation on

photoactive composite properties and its stability. Kaolinite was calcined at different

temperatures (100°C, 200°C, 300°C, 400°C, 500°C, 600°C, 700°C and 800°C). After the

calcination of kaolinite at 400 – 650°C, the process of kaolinite dehydroxylation results in

formation of metakaolinite:

Al2Si2O5(OH)4 → Al2Si2O5(OH)xO2-x + (2-x/2) H2O (1)

with a low value of x (about 10% of residual hydroxyl groups in metakaolinite). Disordered

structure of metakaolinite possesses a huge reactive potential. X-ray powder diffraction patterns

of kaolinite calcined at 100°C (KA1) and 600°C (KA6) shows Fig. 1.

Fig. 1. XRPD patterns of kaolinite calcined at 100 °C (KA1) and 600 °C (KA6).

Legend: K - kaolinite, M - muscovite, Q - quartz.

Kaolinite treatment by different temperatures was leached in demineralization water and

extraction agent simulated acid rains. Leaching test was prepared in accordance with European

M

M

M

M

M

K

K

K

KK

QQ

Q

KA1

KA6

5 10 20 30 40 50

2-Theta - Scale




84

technical standard EN 12457-2. The influence of calcination was evaluated by determination of

aluminum, silicon and other elements in final extracts. Atomic emission spectrometry with

inductively coupled plasma (SPECTRO CIROS VISION) was used for determined of elements

concentration. Structure changes of calcined kaolinite were determined by rtg. diffraction

(Bruker D8 Advance diffractometer).



85

INFLUENCE OF VOID ON THE MECHANICAL PROPERTY OF

NANOMATERIAL

K. YODEN1*, Y. SAITO1, Q. YU1

1Department of Mechanical Engineering, Graduate School of Engneering, Yokohama National University Tokiwadai 79-5, Hodogaya-ku, Yokohama, Japan; email: [email protected]

KEY WORDS: Ag-nano, SiC-chip, void, nanomaterial, high-temperature power device

In recent years, many engineers develop electric vehicles. The power device are required

high performance, miniaturization and weight saving. Previous power device use Si-chip, so

the upper limit of the operating temperature is 150C. It enlarge cooler and prevent power

device to be smaller. Now SiC-chip are developed as an alternative to Si chip. SiC-chip can

work 250C or more. It is expected that SiC chip enable power device to become smaller.

However previous soldering melts high operating temperature. It is expected that bonding

technology using nanoparticles solve this problem. Sintering mechanism of nanoparticle is

complex and unexplained. It is known that sintered material of nanoparticle possesses many

micro voids. We have two tasks to use this nano-particle effectively. One is to clarify relation

void and sintering conditions. And two is to clarify the effect void give nanoparticle material.

In this study, I study influence that voids give the mechanical property of nanomaterial.

Fig. 1. Section having the void of a uniform shape. Fig. 2. Section having the void of a heterogeneous shape.

Fig. 1 and Fig. 2 are SEM (Scanning Electron Microscope) images of a cross-section of

Ag-nano sintered material which are sintered under same condition. The images show that void

shape, size, place and rate are various. To clear relation these elements and mechanical property,

I simulated some model using FEM analyse.

Fig. 3 is model.1 and model.2 which

are given by Fig. 1 and Fig. 2. Model.1

have voids which are uniform shape, and

model.2 have voids which are

heterogeneous shape. These models are

unified by 9.5% void rate. Size of these

models are one side of the square 80 m. I

cut these models in mesh one side is 1 m,

and vertical and horizontal strained force

of 1% in models.1 and 2. We investigated

the effects of void.

Fig. 3. model.1 (left) and model.2 (right).



86

Fig. 4 shows that when the strain is 1%,

vertical stress is higher than horizontal stress by

3.9% in model.1. Fig. 5 shows that when the

strain is 1%, vertical stress is higher than

horizontal stress by 18.5% in model.2. From

these results, I understood the heterogeneous

shape of void bring about anisotropy. Fig. 6 is

Young’s modulus which are calculated by

degree of leaning of the s-s diagram. Fig. 6

show that Young’s modulus in model.1 are

hardly a difference in vertical direction and a

horizontal direction. And the difference in the

case of bulk is approximately 46%. In model.2,

horizontal Young’s modulus is higher than vertical Young’s modulus by 14.3%. It means

model.2 have anisotropy. And the difference in the case of bulk is approximately 56%.

Thus, the approximately 10% of void lower 45-55% of Young's moduluses. And the

influence of void is depend on shape of void. And heterogeneous shape of voids make

anisotropy. So when metal nanoparticles materials are used, it is necessary to control a not only

rate of void but also shape of void by sintering condition.

REFERENCES

[1] YAMAGIWA, M., YU, Q., FUJITA, M., SHINOHARA, M., MURAKAMI, Y.:

”ReliabilityStudy of Mouting Structure for HighTemperature Power Semiconductor Device

Chip Using High Purity Aluminium”, Proceedings of Intersociety Conference on Thermal and

Thermomechanical Phenomena in Electronic Systems (ITherm2008), Orlando Florida, 2008.

Fig. 4. S-S diagram (model.1). Fig. 5. S-S diagram (model.2).

Fig. 6. Young’s modulus.

0

5

10

15

20

25

0 0,2 0,4 0,6 0,8 1

Stre

ss[M

Pa]

Strain[％]

Vartical

Horizontal

0

5

10

15

20

25

0 0,2 0,4 0,6 0,8 1

Stre

ss[M

Pa]

Strain[％]

Vertical

Horizintal

36,0 36,4 34,9 30,0

68,0 68,0

0,0

10,0

20,0

30,0

40,0

50,0

60,0

70,0

Yo

ng'

s m

od

ulu

s [G

Pa]

Horizontal Vertical

model.1

model.2

Al(bulk)



87

DETERMINATION OF ANISOTROPIC CRYSTAL OPTICAL PROPERTIES

USING MUELLER MATRIX SPECTROSCOPIC ELLIPSOMETRY

K. POSTAVA1,2, R. SÝKORA2*, D. LEGUT2, J. PIŠTORA2

1Department of Physics, Technical University of Ostrava, 17. listopadu 15, 708 33 Ostrava – Poruba, Czech Republic

2Nanotechnology centre, Technical University of Ostrava, 17. listopadu 15, 708 33 Ostrava – Poruba, Czech Republic; email: [email protected]

KEY WORDS: spectroscopic ellipsometry, anisotropic crystal, Mueller matrix, permittivity tensor

Recent development in spectroscopic ellipsometry and ellipsometric instrumentation

triggers wide applications of the technique to characterize anisotropic nanostructures, periodic

systems, and also crystals with reduced symmetry. Moreover, the Mueller matrix ellipsometry,

i.e. measurement of all 15 reduced Mueller matrix elements, enables a complete

characterization of reflection properties of the samples, including phenomena as mode

conversion and depolarization [1]. The main task is usually to determine spectra of all

components of the permittivity tensor. The number of independent components depends on

crystal symmetry.

In this paper the Mueller matrix ellipsometry in the spectral range from 0.73 to 6.4 eV

measured using dual rotating compensator ellipsometer RC2 (Woollam company) is applied to

study anisotropic crystals with tetragonal and monoclinic symmetry.

As a typical uniaxial sample we have characterized a rutile (TiO2) tetragonal crystal. The

optical axis of the sample is parallel to its surface. The sample is characterized at variable angle

of incidence and variable azimuthal rotation angle. Figure 1 shows typical Mueller matrix

spectra compared with data fit obtained using the matrix model based on light propagation in

anisotropic stratified media. The angle of incidence of 45° and the azimuthal angle of

approximately 45° were chosen.

The non-zero off-diagonal blocks of the matrix show mode conversion due to the optical

anisotropy. We discuss the sensitivity to determine ordinary and extraordinary optical functions

for various directions of the optical axis.

We also discuss application of Mueller matrix ellipsometry to determine optical functions

of crystals with monoclinic symmetry. The crystal is characterized by four independent

permittivity tensor elements (three diagonal and one off-diagonal) [2]. The crystals of

Cu(H2O)(C2H8N2)SO4 attract strong interest due to their magnetic properties. The crystal

exhibits monoclinic symmetry with the axis inclination of 105.5 degree. Obtained spectra from

Mueller matrix ellipsometry are compared with ab-initio models based on density function

theory (DFT).



88

Fig. 1. Spetra of Mueller matrix elements of the rutile crystal. The measured data (blue dots) are compared with

the model (red lines).

Acknowledgement: Partial support from the projects CZ.1.05/1.1.00/02.0070

(IT4Innovations), CZ.1.05/2.1.00/01.0040 (RMTVC), IRP 167/2014, and Czech Science

Foundation 13-30397S is acknowledged.

REFERENCES

[1] GARCIA-CAUREL, E., OSSIKOVSKI, R., FOLDYNA, M., PIERANGELO, A.,

DRÉVILLON, B., DE MARTINO IN A.: LOSURDO, M., HINGERL, K. (EDS.): Ellipsometry

at the Nanoscale, Springer-Verlag Berlin Heidelberg 2013.

[2] JELLISON JR., G. E., MCGUIRE, M. A., BOATNER, L. A., BUDAI, J. D., SPECHT, E. D.,

SINGH, D. J., Phys. Rev. B 84 (2011) 195439.

[3] SYKORA, R., LEGUT, D.: J. Appl. Phys. 115, (2014) 17B305.

Mueller matrix components

–0.3

–0.4

–0.5

1

–0.5

–0.4

–0.3

0

0.1

–0.1

0.98 –0.02

0

0.02

–0.04 –0.2

–0.1

0

0.1

0

0.04

0.08

0.3

0.2

0.1

0

0.8–

0.9––0.02

0.02

0

–0.1

0.1

0

–0.04

0

0.04

0.08

–0.2

–0.1

0

0.1

–0.3

–0.1

0

–0.9

–0.85

0.8–

2 4 6 2 4 6 2 4 6 2 4 6

Photon energy(eV)

–0.2



89

OBSERVATION OF MAGNETIC FIELDS AROUND PLASTIC

DEFORMATION AREA IN LOW CARBON ALLOY STEEL

K. KIDA1*, M. ISHIDA1, K. MIZOBE1

1University of Toyama, 3190, Gofuku, Toyama City, Toyama Prefecture, 930-8555, Japan; email: [email protected]

KEY WORDS: magnetic flux density, plastic deformation, scanning hall probe microscopy

Failure of machine elements is mainly caused by fatigue crack growth occurring as results

of cyclic stress concentration. Following the pioneering studies by Orowan and Irwin many

investigations have been carried out in order to understand the fatigue crack growth in steels.

One of the important topics in this field is plastic deformation occurring around the crack tip.

Non-destructive methods that can be related to the plastic deformation around small crack tip

area are necessary to study the crack growth. For our previous works, we developed a scanning

Hall probe microscope (SHPM) equipped with GaAs films and observed fatigue cracks growing

from artificial slits in steels (JIS, SUJ2 [1,2], S45C [3]). Furthermore, we applied this SHPM

technique to contact problem of tool steel (SKS93) [4], fatigue of welding part (SS400) [5],

tensile loading of thin plate (SKS93) [6] and plastic deformation around a Vickers indentation

(SKS93) [7]. A one-dimensional sensor was used in the first research. From the second research,

three-dimensional observations were carried out using three 10 µm-sized Hall films in order to

study the features of magnetic fields under various loadings.

Fig. 1. Vertical component of three-dimensional magnetic flux density (Bz) in a specimen. Observation covers

the area of (x, y) = (6.0 mm, 12.0 mm). The observations of magnetic flux density were done after

magnetization, before the tests (a), and after the second indentation test (b).

Plastic deformation area expands as the crack grows. In the present research we induced

two Vickers indentations along the center line on a low carbon alloy steel plate (JIS, S45C) and

observed the relation between magnetic field and plastic deformation. Demagnetization was

done by using a coil in order to normalize the initial magnetic field of the specimen. After the

demagnetization, a permanent magnet block was slid along the center line area of the specimen.

The size and residual magnetic flux density of the block were 1 – 10 – 10 mm3 and 99 mT,

respectively. Three components of the magnetic field, Bx, By and Bz were compared before and

after the indentation tests.



90

Fig. 1 is an example of the magnetic fields

before and after the tests. After the first and second

Vickers indentations were induced on the

specimen surface, they were observed. The load

and diagonal length of the indentations were

20 kgf and 300 μm. Fig. 1(a) shows a magnetic

field component (Bz) in a direction vertical to the

specimen surface before the test, and (b) is after the

second indentation test. The distance between

these indentations was 900 µm. The positions of

the first and second indentations were y = 5.7 mm

and 6.6 mm, respectively. We can see clearly the

changes in magnetic fields in a circle of Fig. 1(b).

Observation area was divided into 10 µm-width segments whose longitude axes are parallel to

the x-axis, and all of the magnetic components in the segments were compared to their initial

values observed before testing. The segment coordinates are numbered along the y-axis. The

peak to bottom value in each segment was calculated with the maximum and minimum values

in it. After the calculation, distributions of the peak to bottom values of Bx, By and Bz in all

segments were arranged along the y-axis. Fig. 2 shows the changes in the peak to bottom values

of Bz before and after the indentation tests. When comparing the magnetic components after

the first and second indentations, it is found that the bottom peak area expands toward the

second indentation area along the center line (y-axis). The left part of the curve measured after

the first indentation test corresponds to that after the second test. This means the magnetic fields

record the change in plastic deformation.

Acknowledgement: A part of this research was supported by Grants-in-Aid for Scientific

Research, JSPS (KAKENHI, Scientific Research (c), No. 23560089).

REFERENCES

[1] KIDA, K., TANABE, H., OKANO, K.: Changes in magnetic flux density around fatigue crack

tips, Fatigue & Fracture of Engineering Materials & Structures, 32, 3, 2009, pp. 180-188.

[2] KIDA, K., SANTOS, E. C., HONDA, T., KOIKE, H., ROZWADOWSKA, J.: Observation of

magnetic flux density around fatigue crack tips in bearing steel using a SHPM with a three-

dimensional small-gap probe, Int. J. Fatigue, 39, 2012, pp. 38-43.

[3] KIDA, K., SANTOS, E. C., URYU, M., HONDA, T., ROZWADOWSKA, J.,

SARUWATARI, K.: Changes in magnetic field intensities around fatigue crack tips of medium

carbon low alloy steel (S45C, JIS), Int. J. Fatigue, 56, 2013, pp. 33-41.

[4] KIDA, K., URYU, M., HONDA, T., SANTOS, E. C., SARUWATARI, K.: Three-Dimensional

Observation of Magnetic Fields in Alloy Tool Steel under Spherical Hertzian Contact, Materials

Research Innovations, 18, Supplement1, 2014, pp. 71-75.

[5] KIDA, K., HONDA, T., SANTOS, E. C., SARUWATARI, K., URYU, M., HOURI, K.,

TANABE, T., KANEMASU, K.: Three-dimensional Magnetic Microscopy of Early Stage

Fatigue in WMZ of Low Carbon Steel Plates (JIS-SS400), Materials Research Innovations, 18,

Supplement1, 2014, pp. 66-70.

[6] KIDA, K., URYU, M., HONDA, T., SANTOS, E. C.: Changes in Three-dimensional Magnetic

Fields of Star shaped JIS-SKS93 Plates Embedded in Clear Acrylic Cold Mounting Resin under

Tensile Loads, Materials Research Innovations, 18, Supplement1, 2014, pp. 89-93.

[7] HONDA, T., SANTOS, E. C., KIDA, K.: Scanning Hall probe microscopy of residual magnetic

fields around plastic deformation of Vickers indentations in carbon tool steel (JIS, SKS93),

Mechanics of Material, 69, 2014, pp. 262–269.

Fig. 2. Change in magnetic field component, Bz,

due to Vickers indentations.

0 4 8 12

0.7

0.8

0.9

1.0

y [mm]

Magn

etic

co

mpon

ent,

Bz

[mT]

Initial distribution

1st indentation1st indentation

2nd indentation

5.7 mm 6.6 mm



91

MAGNETO-PLASMONIC PROPERTIES OF Au/Fe/Au

PLANAR NANOSTRUCTURES: THEORY AND EXPERIMENTS

J. VLČEK1,3*, M. LESŇÁK2, P. OTIPKA3

1National Supercomputing Centre IT4Innovations, VŠB-TU Ostrava, Czech Republic; email: [email protected]

2Regional Materials Science and Technology Centre, VŠB-TU Ostrava, Czech Republic 3Dept. of Mathematics and Descriptive Geometry, VŠB-TU Ostrava, Czech Republic

KEY WORDS: magneto-plasmonics; response factors; sensitivity criteria

The non-reciprocity of magnetooptical

reflection response by surface plasmon

excitation in the planar Au/Fe/Au/glass

nano-systems with prism coupling is

studied. These structures are intended as

magnetic field sensor units combining

magneto-optical and surface-plasmon-

resonance effects. In order to simulate the

diffraction response of discussed structures

to external magnetic field theoretical model

based on matrix algorithm is applied. The

ability of MO-SPR systems to sensing of

magnetic field is analysed using the

response factor

( )pp pp

pp pp

R R

R R

r

, (1)

where Rpp denotes the reflectance of p-polarized beam; and, the sign in upper index relates to

the orientation of external magnetic field. Unlike our previous work [1] the newly proposed

sensitivity criteria F and K are applied (see Fig. 1).

Obtained theoretical results are compared with experiments realized using the measuring

device Multiskop (Optrel GbR, Germany). In particular, the intensity of reflected light is

detected in dependence to the angle of incidence at the wavelength 632.8 nm. We completed

this equipment by digitally controlled electro-magnet, which enables production of a predefined

magnetic field in transversal configuration.

REFERENCES

[1] VLČEK, J., LESŇÁK, M., PIŠTORA, J., ŽIVOTSKÝ, O.: Magneto-optical sensing of

magnetic field, Optics Communications 286, 2013, pp. 372-377.

Fig. 1. Sensitivity criteria F and K.

45 45.5 46 46.5 47-0.2

-0.15

-0.1

-0.05

0

0.05

0.1

0.15

0.2

0.25

0.3

incidence angle [deg]

r

Au 7 nm, Fe 8 nm, Au 14 nm

F

K = tan



92



93

IMAGE ANALYSYS VIA REACTION DIFFUSION SYSTEM FOR EDGE

DETECTION

K. NAKANE1*, H. MAHARA2, K. KIDA3

1Osaka University, Faculty of medicine, 1-7, Yamadaoka, Suita, Osaka, Japan; email: [email protected]

2Chiba University Hospital, 1-8-1, Inohana, Chuo-ku, Chiba-shi, 260-8677, Japan 3University of Toyama, Gofuku 3190, Toyama1-8-1, Inohana, Chuo-ku, Chiba-shi, 260-8677, Japan

KEY WORDS: reaction diffusion, edge detection, homology

To measure the particle size of the material and to analyze the images of the structure, we

need to detect the edge of the particle. Since the grain boundaries are not so clear, it takes a lot

of time to detect them, manually.

Here, we will present a support method to detect the edge from material images. By solving

non-linear partial differential equations numerically, we make blurry images clear. Combining

ordinary image analysis method, we will take the edge of grain. In this talk, we introduce the

results of our method, and discuss the possibility of this method.

(a) (b)

(c)

Fig. 1. (a) The image of SUJ2 (Q3T1). (b) The numerical result of reaction diffusion equation.

(c) The superimposed images of (a) and (b).

This method is developed for edge detection and figure-ground separation [1]. This method

can realize a high quality processing on noisy image compared with the Median filter [2].

Nomura et al. extended this algorithm in order to detect edge from a gray-scale image [3]. This

algorithm was applied, recently, for detection of blood vessels in fingertips [4].

The reaction-diffusion equations consist of three variables with diffusion terms. Two of

them are the FitzHugh-Nagumo equations that are a model of nerve membrane [5, 6]. Another



94

one is a variable for averaging threshold that is introduced by Nomura et al.. Precise equations

are described in the reference therein. Fig. 1 shows the result of the edge detection with this

method.

(d) (e)

Fig. 2. (d) Noise reduction result of (c). (e) The image that (d) is superimposed on (b).

By adding a manual correction (noise reduction and so on), we can detect the edge easily.

For this purpose, we need useful GUI system. To make a GUI that suits the purpose of the

engineer, it would be necessary to devise in each scene. We are very glad, if we could discuss

the direction of development.

REFERENCES

[1] NOMURA, A., ICHIKAWA M., MIIKE H., EBIHARA M., MAHARA H., SAKURAI T.:

Ralizing Visual Functions with Reaction-Diffusion Mechanism: Journal of the Physical Society

of Japan 72, 2003, pp. 2385-2395.

[2] EBIHARA M., MAHARA H., SAKURAI T., NOMURA A., MIIKE H., Image processing by a

discrete Reaction-Diffusion System: Proceeding of Visualization, Imaging and Image Processing

396, 2003, pp. 145-150.

[3] NOMURA, A., ICHIKAWA M., SIANIPAR R. H., MIIKE H.: Edge Detection with Reaction-

Diffusion Equations Having a Local Average Threshold: Pattern Recognision and Image

Analysis 10, 2008, pp. 289-299.

[4] NAKANE, K. AND MAHARA, H: A numrical method to detect the edge of vessels, in

preparation.

[5] FITZHUGH R.: Impulses and Physiological States in Theoretical Models of Nerve Menbrene:

Biophysical Journal 1, 1961, pp. 445-466.

[6] NAGUMO J., ARIMOTO S., YOSHIZAWA S.: An Active Pulse Transmission Line

Simulationg Nerve Axon: Proceeding of the IRE 50, 1962, pp. 2061-2070.



95

EFFECTS OF INCLUSION ON THE IN-PLANE MECHANICAL

PERFORMANCE OF MICRO-LATTICE PLATE

K. USHIJIMA1*, W. J. CANTWELL2, D. H. CHEN3

1Tokyo University of Science, 6-3-1 Niijyuku, Katsushika-ku, Tokyo, Japan; email: [email protected] 2Khalifa University of Science, PO Box 12778, Abu Dhabi, UAE 3Jiangsu University, 301 Xuefu Road, Zhenjiang, Jiangsu, P.R. China

KEY WORDS: micro-lattice structure, initial stiffness, yield strength, finite element method

Cellular structures, such as honeycombs, lattices and foams have been taken attention and

used for many structural applications owing to their superior mechanical performance per unit

mass. For decades, many researchers have been investigated the mechanical properties of

cellular structures with regular and irregular cells by using numerical, theoretical and

experimental approaches.

One of our authors have developed

the selective laser melting (SLM)

technique for manufacturing micro-

lattice structures at length scales of

microns. The micro-lattice structure can

be produced by using CAD data, so the

micro-architecture of the structure can be

changed easily to enhance the overall

mechanical properties such as initial

stiffness, plastic collapse or buckling

strength and impact energy absorption capacity. Figure 1 shows the example of micro-lattice

block with two types of inner cells.

The effects of hole and inclusions on the mechanical performance of honeycomb structures

have been investigated by other researchers. However, the micro-lattice structure investigated

has much potential for improving the mechanical properties by changing the strand’s length,

angles between adjacent strands.

In this study, the in-plane mechanical properties of the micto-lattice plate with inclusion is

analysed by using numerical analysis, Finite Element Method. The inclusion is centred in the

plate, and subjected to in-plane tension load. In our discussion, the effects of inclusions on the

initial stiffness and plastic collapse strength are discussed. The inclusions investigated here are

modelled by holes, softer (coarser) cells and harder (finer) cells.

Figure 2 shows analytical models investigated in this study. Also, Figures 3 shows

variations of initial stiffness E* and plastic collapse strength σ*pl with the defect width w. It can

be understood that the initial stiffness depends strongly on the shape of holes, and decreases

nonlinearly as the width w increases. On the contrary, the plastic collapse strength for each

plate is almost coincident under the same width w, and decreases linearly as w increases. That

is because the plastic collapse strength is mainly governed by the weakest point of the plate,

and the weakest point can be observed at the edge of the hole.

(a) with uniform cells (b) with non-uniform cells

Fig. 1. Photos of micro-lattice structure by SLM technique.



96

(a) Defect-w (b) Defect-s (c) Defect-d (d) Nu-1_2 (e) Nu-2_1

Fig. 2. Examples of lattice plates with holes or inclusion.

(a) for initial stiffness (b) for plastic collapse strength

Fig. 3. Variations of defect width w on material properties.

REFERENCES

[1] GIBSON, L. J., ASHBY, M. F.: Cellular Solids: Structure and Properties, 2nd ed. Cambridge:

Cambridge Press 1997.

[2] SILVA, M. J., GIBSON, L. J.: The Effects of Non-periodic Microstructure and Defects on the

Compressive Strength of Two-dimensional Cellular Solids, International Journal of Mechanical

Science 39, 1997, pp. 549-563.

[3] GUO, X. E., GIBSON, L. J.: Behavior of Intact and Damaged Honeycombs: a Finite Element

Study, International Journal of Mechanical Sciences 41, 1999, pp. 85-105.

[4] CHEN, C., LU, T. J., FLECK, N. A.: Effect of Imperfections on the Yielding of Two-

Dimensional Foams, Journal of the Mechanics and Physics of Solids 47, 1999, pp. 2235-2272.

[5] CHEN, C., LU, T. J., FLECK, N. A.: Effects of Inclusions and Holes on the Stiffness and

Strength of Honeycombs, International Journal of the Mechanical Sciences 43, 2001,

pp. 487-504.



97

ASSESSMENT OF STRUCTURES LOADED AT CREEP

S. VEJVODA1*, P. POPPELKA2, P. RYŠAVÝ1

1VÍTKOVICE ÚAM a.s., Mezírka 775/1, 602 00 Brno, Czech Republic; email: [email protected] 2SES Tlmače, a.s., Továrenská 210, 935 28 Tlmače, Slovak Republic

KEY WORDS: oxide layer (scale), tube, strain, stress, crack, exfoliation, delamination, fracture

toughness, stress intensity factor, creep strain, membrane stress, stress concentration, metal

The operational parameters of classical power plants are continually increased. Many times

there was not respected that used material has the limit of usability for reliable service at lower

operational temperatures than an increased operation temperature only upon 30°C. Parameters

effecting the damage from the oxide layer Fe3O4 (magnetite) on the tube head transfer surface

were analyzed for head resisting steel Cr-Mo-V, specification 15 128.5.

A study was made in accordance with [1] and [2]. The oxide layer continually arose on the

inside tube surface 38x6.1 mm made from steel 15 128.5. For the oxide layer Fe3O4 the

following were used: i = 4.5 [J.m-2]; EOX = 208000 [MPa]; KIc = 1.4 [MPa.m0.5] and Poisson

number = 0.262 [3].

The study was carried out for following parameters:

- thickness of the oxide layer d = 0.01 mm ÷ 1.0 mm;

- half-axis of the defect in the oxide layer c = 0.01 mm ÷ 1.0 mm;

- asperity of the tube inside surface r = 1.6 m; 3.2 m and 6.3 m;

- radius of the delaminated area of the oxide layer created on the tube metal

R = 0.025 mm ÷ 20.0 mm;

- wave length of the rough interface on the boundary oxide/metal = 0.025 mm ÷ 20.0 mm;

- coefficient f related with geometry, for real defect f =1.

Some results of the analysis:

- oxide layer (scale) with the defect of c =10 μm would crack at the limit strain ct = 0.12 %;

- interfacial defect of c = 10 m on the boundary oxide layer/metal would be begin growth

under compressive stresses when the critical strain reaches the value of ci = - 0.1%;

- buckling of the oxide layer under compressive stresses would arrive at the critical strain

cb -1.0 [%], when its thickness is d 1 mm and the radius of the delaminated area is

R 10 mm;

- circumferential shearing under

compressive stresses on the

bulging oxide layer d 1 mm of

the delaminated area

R = 20 mm located on the

boundary oxide layer/metal

would not occur at strain

cbf -0.01 [-];

- critical value of strain cs

necessary for exfoliation of the

oxide layer does not fall under -

1.90 [%] when the thickness of the oxide layer is d 1.0 mm, the length wave is 20 mm,

and roughness of r = 1.6 m.

Fig. 1. Exfoliation of the delaminated oxide layer Fe3O4 on the

inside surface of the tube




98

Exfoliation of the delaminated oxide layer Fe3O4 on the inside surface of the tube was

analyzed by program ANSYS for nominal operation conditions of the boiler (p = 22.1 MPa,

T = 625°C, t = 30 years). The calculated value of the stress intensity factor KI = 4.07 MPa.m0.5

is greater than the fracture toughness of the oxide layer KIc= 1.4 MPa.m0.5. The calculated strains

of the oxide layer were greater than their critical values for damage of the oxide layer as well.

When creep strain limits for given steels are not know, it is possible to use limits given in

[4]. It means 1% for membrane stresses, 2% for bending stresses and 5% for the stress

concentration area. Creep strains are usually calculated in areas of membrane stresses. The FEM

programs enable calculation of the whole structure, in which geometrical notches are present.

Compressive stresses and strains were calculated at these geometrical notches. The big

“compressive creep deformation” was calculated by the FEM in the small area of these notches

and until after stress relaxation the character of creep deformation was changed to tension. This

problem was discussed with creep specialists in the Czech Republic.

Fig. 2. Detail of analyzed structure;

points A, B, H – stress concentration.

Fig. 3. Point A, 2nd principle strain;

compressive stress, relaxation and growth of c.

Fig. 4. Creep strain intensity c at area

of membrane stresses.

Fig. 5. Creep strain intensity c at the stress concentration

area; boundary c =5 % is in red.

Acknowledgment: The authors gratefully acknowledge the support by Technological

agency of the Czech Republic for project No. TA02011179.

REFERENCES

[1] SCHÜTZE, M.: Modelling oxide scale fracture. Materials at High Temperatures, Volume 22,

Numbers 1-2, February 2005, pp. 147-154.

[2] SCHÜTZE, M, TORTORELLI, P.F., WRIGHT, I.G.: Development of a Comprehensive Oxide

Scale Failure Diagram. Oxid Met (2010) 73: pp. 389-418

[3] Gercek: Int. J. Rock Mech. Min. Sci. 44 (2007)

[4] Cases of ASME Boiler and Pressure Vessel Code, Case N-47-29, Class 1. Components in

Elevated Temperature Service. Section III, Div. 1, 1990.

http://www.ingentaconnect.com/content/stl/maht;jsessionid=109nflk13w0k3.alice



99

DIFFUSION OF HYDROGEN IN THE TRIP 800 STEEL

J. SOJKA1*, P. VÁŇOVÁ1, V. VODÁREK1, M. SOZANSKA2

1Faculty of Metallurgy and Materials Engineering, VŠB – Technical University of Ostrava, Czech Republic;

email: [email protected] 2Faculty of Materials Engineering and Metallurgy, Silesian University of Technology, Katowice, Poland

KEY WORDS: hydrogen electrochemical permeation, TRIP 800 steel, hydrogen embrittlement

The presented paper is devoted to the study of hydrogen diffusion in the C-Mn-Si TRIP 800

steel containing 0.2 wt. % of C, 1.5% of Mn and 1.5% of Si. The steel was tested in three

different states: in as-received state after hot and cold rolling and subsequent heat treatment;

and furthermore after 5% and 10% tensile deformation. The tensile deformation resulted in an

increase of mechanical properties (Re, Rm) and in a decrease of retained austenite content from

11.0% in as-received state to 2.2% after 10% tensile deformation. Hydrogen diffusion

characteristics were studied by means of electrochemical permeation method.

Electrochemical hydrogen permeation

tests were carried out using a Devanathan-

Stachurski two-component cell separated

by a steel membrane – working electrode.

The exit side of the working electrode was

palladium coated to prevent from

hydrogen atom recombination during

permeation experiments. Hydrogen

charging cell was filled with 0.05M

H2SO4, while the exit cell was filled with

0.1 M NaOH solution. The exit cell was

de-aerated by argon bubbling before and

during experiments. Firstly, the entry side

of the specimen was polarized anodically

at a current density of + 35 mA.cm-2. At the end of this period (5 minutes), H2SO4 charging

solution was renewed continuously to eliminate metallic ions from the solution. After that, two

build-up transients (BUT) were recorded, the first one at the charging current density of

-20 mA.cm-2, the second one at the charging current density of -35 mA.cm-2. Before ending the

experiment hydrogen charging was stopped and a decay transient (DT) was also recorded.

Hydrogen diffusion coefficients were calculated using the time-lag method according to

Eq. 1:

Lt

LD

6

2

, (1)

where L represents the membrane thickness and tL corresponds to the time where the permeation

current reaches 63% of its steady-state value. Sub-surface hydrogen concentration was

calculated using Eq. 2:

DF

LiCH

0 , (2)

where i is a steady-state current density and F is Faraday’s constant.

Fig. 1. Hydrogen diffusion coefficients Deff

for all studied states.

0E+00

1E-07

2E-07

3E-07

4E-07

5E-07

6E-07

7E-07

8E-07

9E-07

without deformation

5% tensile deformation

10% tensile deformation

first build up transient

second build up transient

decay transient




100

The results of hydrogen diffusion coefficients are presented in Fig. 1. It is obvious that the

lowest values of hydrogen diffusion coefficient were always obtained for the first BUT. It can

be related to hydrogen trapping in both reversible and irreversible traps. The hydrogen diffusion

coefficient corresponding to the first BUT was higher for the state after 10% tensile

deformation. This behaviour confirms the hypothesis that formation of martensite facilitates

hydrogen diffusion in the TRIP steel. Hydrogen diffusion coefficients corresponding to the

second BUT were markedly higher and confirmed that most of traps were filled by hydrogen

during the first BUT. Nevertheless, values of hydrogen diffusion coefficients still remained

lower in comparison with conventional steels having bcc lattice. For decay transients hydrogen

diffusion coefficients were usually situated between values obtained for the 1st and 2nd BUT. It

confirms that during the first BUT hydrogen trapping can be expected and during DT hydrogen

detrapping can be expected, influencing thus values of hydrogen diffusion coefficient.

Measured data were also fitted with the theoretical curves of normalized hydrogen flux Jt/J.

Measured data fitted very well with the theoretical curves for the second BUT in all studied

states. However, for the first BUT and for the DT, the measured data were shifted to longer

time in comparison with the theoretical curves confirming thus the important role of hydrogen

trapping and detrapping.

Hydrogen sub-surface concentration was calculated for the first BUT using Eq. 2. The

obtained results showed that the sub-surface concentration of hydrogen was rather high mainly

if the amount of retained austenite in the structure was not too low. In the as-received state,

hydrogen sub-surface concentration reached 12.6 ppm, while after 10% tensile deformation,

hydrogen concentration was 5.4 ppm. In this way the role of retained hydrogen as an important

and very probably irreversible hydrogen trap was confirmed. The high sub-surface

concentration of hydrogen in the studied TRIP steel can, at least partially, explain its rather high

susceptibility to hydrogen embrittlement [1].

The obtained results can be summarised as follows:

- The values of hydrogen diffusion coefficients in the TRIP 800 steel were rather low and

lay between 1.10-7 cm2.s-1 and 7.8.10-7 cm2.s-1;

- The highest values of hydrogen diffusion coefficient were observed during the 2nd build

up transient where the role of hydrogen trapping was limited;

- A decrease of retained austenite content resulted in an increase of hydrogen diffusion

coefficient;

- The comparison of experimental data with the theoretical model showed a good fitting for

the 2nd build up transient, while during the 1st build up transient and during the decay

transient a shift of experimental data to longer time was observed.

- Rather high sub-surface concentration of hydrogen was determined in the studied steel

especially for states with higher retained austenite content.

Acknowledgement: The authors are grateful to the Ministry of Education of the Czech

Republic for the financial support of the project No. LE13011 “Creation of a PROGRES 3

Consortium Office to Support Cross-Border Co-operation” and the project No. LO1203

"Regional Materials Science and Technology Centre - Feasibility Program".

REFERENCES

[1] SOJKA, J. at al.: Effect of hydrogen on the properties and fracture characteristics of TRIP 800

steels, Corrosion Science 53, 2011, pp. 2575-2581.



101

PRECIPITATION REACTIONS IN A COPPER - BEARING GOES

V. VODÁREK1*, A. VOLODARSKAJA1, Š. MIKLUŠOVÁ2, J. HOLEŠINSKÝ1, O. ŽÁČEK2

1 VŠB – TU Ostrava, Faculty of Metallurgy and Materials Engineering, Ostrava, Czech Republic;

email: [email protected] 2ArcelorMittal Frýdek-Místek, Czech Republic

KEY WORDS: GOES, precipitation processes, sulphides, nitrides, TEM

Magnetic properties of grain oriented electrical steels (GOES) depend strongly on the

sharpness of the Goss texture. It is believed that the perfection of the final texture is significantly

affected by structural inheritance during a complex processing route of GOES 1. Factors

which are considered to be very important for the formation of the Goss texture during high

temperature annealing include the size of the initial grains with the Goss orientation, their

orientation with respect to the other grains and the role of minor phases in grain boundary

pinning 2, 3. The processing technology of Cu – bearing GOES comprises of following

production steps: slab reheating – hot rolling and coiling – 1st cold rolling – decarburization

annealing – 2nd cold rolling – high temperature annealing – thermal flattening. The effect of

copper in GOES is not well understood. Copper is believed to play several roles 2:

1. Increase the volume fraction and stability of austenite during hot rolling in the two phase

( + ) region.

2. Small copper rich sulphides could inhibit grain growth during recrystallization processes

(decarburization annealing and high temperature annealing).

3. Precipitation of - Cu could positively affect distribution of AlN particles, which are

expected to be the most important inhibition phase 2. Ideal conditions for the

precipitation of - Cu represents a slow heating rate during the initial stages of the high

temperature annealing (less than 30°C/h) combined with the presence of many lattice

defects after the 2nd cold rolling. At temperatures of normal grain growth - Cu

precipitates are expected to dissolve.

4. Segregation of copper atoms at grain boundaries can modify their mobility.

This paper deals with minor phase evolution in a Cu-bearing GOES during the following

production steps of the AlN + Cu processing technology: hot rolling of slabs, 1st cold rolling +

decarburization annealing and a slow heating to the temperature of primary recrystalization

(620°C). Minor phases were investigated in TEM using carbon extraction replicas.

Chemical composition of the strip after hot rolling is shown in Table 1. Decarburization

annealing after the 1st cold rolling reduced the carbon content in the steel to 0.0029 wt.%.

Table 1 Chemical composition of the steel investigated (after hot rolling), wt.%.

C Mn Si S Cr Cu Altot. Ti N

0.03 0.25 3.16 0.004 0.024 0.50 0.014 0.004 0.009

In the state after hot rolling, ferrite grain boundaries were decorated by thin films of iron

carbides. This precipitation took place after coiling. Intragranular particles were mostly formed

by sulphides of manganese or complex sulphides of manganese and copper (up to 12 wt.%Cu).

The size of these particles reached up to several hundreds of nanometres. Copper rich sulphides

dissolved during hot rolling (Cu2S: Tsol = ca 950°C). A small number of precipitates consisted

of TiN or AlN particles. No Cu rich metallic particles were detected.



102

Parameters of hot rolling and 1st cold rolling + decarburization annealing are stated in

Table 2. Precipitation after decarburization annealing was much more pronounced. Grain

boundaries were decorated by both AlN and Si3N4 particles. In the GOES Si3N4 nitrides

represent a metastable phase, which is in the temperature interval of 700 – 900°C gradually

replaced by AlN phase. Intragranular precipitation of AlN was very variable. Local differences

in precipitation could be related to chemical heterogeneity of the strip inherited from hot rolling

in the two phase field. Re-precipitation of copper sulphides (Cu2S) and copper rich complex

sulphides of manganese and copper took place during decarburization annealing. The typical

size of these particles was several tens of nanometres. In many cases nucleation of AlN particles

on sulphides was observed. No copper rich metallic particles were detected. TiN particles were

not affected by decarburization annealing.

Table 2 Parameters of hot rolling and cold rolling + decarburization annealing (DCA).

Hot Rolling Cold Rolling + DCA

Thickness mm Tmax °C Tmin °C Tcoil. °C Thickness mm Tdecarb.°C

2.5 1250 970 600 0.65 840

Laboratory simulation of a slow heating rate to the temperature of the primary

recrystallization start (620°C) was carried out on the specimen after the 2nd cold rolling to the

final thickness of 0.3 mm. Heating rate in a protective nitrogen atmosphere was v = 25°C/h.

TEM analysis revealed identical minor phases as in the state after decarburization annealing,

no - Cu particles were observed. However, a number density of minor phase particles

decreased. It indicates that coarsening of precipitates occurred during the slow heating to the

temperature of primary recrystallization.

Results of investigations suggest that in the 0.5 wt.%Cu – bearing GOES precipitation of

- Cu does not play a crucial role. Copper atoms dissolve in sulphides exhibiting significantly

lower thermal stability than sulphides of manganese. Copper rich sulphides dissolve during hot

rolling and re-precipitation of copper rich sulphides takes place during decarburization

annealing. The most important inhibition phase in Cu – bearing GOES is AlN. Some nitrogen

is also bound in a metastable Si3N4 phase. Intensive precipitation of these nitrides occurs during

decarburization annealing. Slow heating to the temperature of primary recrystallization was

accompanied by coarsening of nitrides, no - Cu particles were detected.

Acknowledgement: This paper was created in the projects FR-TI3-053 and the project

No. LO1203 “Regional Materials Science and Technology Centre – Feasibility Program”

funded by Ministry of Education, Youth and Sports of the Czech Republic.

REFERENCES

[1] BERNIER, N., LEUNIS, E., FURTADO, C., VAN DE PUTTE, T. V., BAN, G.: EBSD Study of

Angular Deviations from the Goss Component in Grain-oriented Electrical Steel. Micron,

Vol. 54, 2013, p. 43.

[2] LOBANOV, M.L.: Upravlenije strukturoj i teksturoj eletrotechničeskoj anisotropnoj stali,

Abstract of Doctoral Thesis, Jekaterinburg 2010, pp. 48 (in Russian).

[3] HUMPHREYS, F. J., HATHERLY, M.: Recrystallization and Related Phenomena, Elsevier,

Amsterdam 2004.



103

CONCEPT OF DAMAGE MONITORING AFTER GRINDING

FOR COMPONENTS OF VARIABLE HARDNESS

A. MIČIETOVÁ1*, J. PIŠTORA2, Z. DURSTOVÁ1, M. NESLUŠAN1

1University of Žilina, Faculty of Mechanical Engineering, Univerzitná 1, 010 26 Žilina, Slovak Republic; email: [email protected]

2Nanotechnology Centre, VŠB TU Ostrava, 17. listopadu 15, 70833 Ostrava, Czech Republic

KEY WORDS: heat treatment, grinding, Barkhausen noise

Roll bearings are routinely heat treated in a variety of manners. Except induction or case-

hardening, conventional heat treatment is carried out to impart the high hardness and the

corresponding high resistance against friction and contact wear. Annealing process is always

performed after hardening to reduce the high internal stresses induced during rapid cooling.

Further heat treatment is sometimes required to enhance the toughness of hardened parts. These

parts are exposed to the elevated temperatures for a certain period within hardness of parts

decreases as a result of carbides coarsening, decrease of dislocation density and stress

relaxation. Hardness of such parts can vary in the range 38 to 62 HRC. Grinding operations are

usually involved in production of bearings to achieve the required surface roughness, shape and

dimension accuracy. Nowadays, additional requirements such as surface structure, hardness or

stress state are needed to be fully filled due to its substantial influence on functionality of parts

in operation.

Non-destructive monitoring of critical surfaces has to be carried out to reveal the parts

containing the unacceptable surface integrity. Magnetic method based on Barkhausen noise

(BN) is very often employed for such purpose, especially ground surfaces due to the high

sensitivity of BN emission to the thermally induced surface overtempering. BN originates from

irreversible Bloch Walls (BW) motion during cyclic magnetization due to existence of pinning

sites such as grain boundaries, dislocations, precipitates, other phases, etc. Ground parts can

suffer from thermally induced burn as a result of excessive heat generation in the wheel –

workpiece contact. Being so, BN emission increases in magnitude due to decreased pinning

strength of thermally softened layer produced by improper grinding as a result of carbides

coarsening and decrease of dislocation density (stress state is also altered) [1]. Thermally

softened layer contributing to the more enhanced BN signal received on the frees surface layer

can be easily contrasted and recognized when compared with untouched deeper regions in an

optical image due to reduced resistance against etching.

Concept for monitoring surface damage after grinding is based on the contrast between

poor BN emission of untouched structure and enhanced BN response (its rms value) of the

surface undergoing the elevated temperatures. This surface appears dark under optical

observation [2]. On the other hand, such concept can fail when the hardness of a component

decrease. Then the overtempering effect induced by grinding is shadowed by the previous heat

treatment regime since the both processes can represent nearly the same thermal load of the

surface. BN emission and its evolution (for instance, along with the progressive grinding wheel

wear) depends on the annealing temperature and the temperature in wheel – workpiece contact.

The higher annealing temperature is, the less pronounced contribution of the surface

overtempering induced by grinding itself would be expected which in turn correspond to the

less remarkable contrast between the deeper untouched and the near surface thermally softened

layer.

This study demonstrates that the evolution of BN and the BN features versus progressive

grinding wheel wear for components of low hardness differ from those of high hardness.



104

Fig. 1 illustrates that the evolution of BN in grinding bearing steel 100Cr6 of hardness

40 HRC opposes to the progressive increase of BN emission usually obtained when parts of

hardness 62 HRC are ground. Fig. 1 shows that BN values decrease when grinding wheel wear

is more developed. It means that the surface altered by grinding process is composed of

structure of high pinning strength considering BW motion. Increasing Peak Position indicates

that effect of surface hardening dominates as a result surface heating (induced by grinding

process) followed by self-cooling. This mechanism causes increasing hardness of near surface

regions when compared to the deeper regions. Being so, the concept for monitoring surface

integrity of such surfaces via BN technique should be reconsidered.

This paper discusses the specific aspects of the surfaces heat treated to variable hardness.

Surface characterization is determined by BN technique as well as the conventional ones such

as metallographic observation, residual stresses measurement and microhardness readings.

Acknowledgement: The authors gratefully acknowledge the support by Vega project n.

1/0097/12 and Regional Centre of Excellence reg. no. CZ.1.05/2.1.00/01.0040.

REFERENCES

[1] MOORTHY, V. et. all.: Evaluation of heat treatment and deformation induced changes in

material properties in gear steels using magnetic Barkhausen noise analysis, Conference ICBN

03, Tampere, Finland 2001.

[2] ČILLIKOVÁ, M., MIČÚCH, M., NESLUŠAN, M., MIČIETOVÁ, A.: Nondestructrive

micromagnetic evaluation of surface damage after grinding, Manufacturing technology 2013,

Vol. 13, No.2, pp. 152-157.

Fig. 1. BN (rms) and Peak Position versus number of ground rings, 40 HRC.

0

50

100

150

200

250

0 5 10 15 20 25 30 35 40

0

2

4

6

8

10

12

14

16

18

rms

Peak Position

ring n.

BN

, m

VP

ea

k P

ositio

n, m

a



105

STUDY ON RELIABILITY EVALUATION METHOD OF

ADHESION STRENGTH OF RESIN

O. HONDA1*, Q. YU1

1 Department of Mechanical Engineering, Graduate School of Engineering, Yokohama National University Tokiwadai 79-5, Hodogaya-ku, Yokohama, Japan; email: [email protected]

KEY WORDS: resin, young’s modulus, reliability evaluation method, adhesion

The recent spread of hybrid cars and electric cars has been supported by downsizing and

technological advance of the power modules. The power modules are composed by different

materials. Thus, the mismatch by the expansion difference between materials occurs because

of temperature increase in use. In this way, delamination occurs in the interface of encapsulant

(resin) and other parts, and it becomes the reliability issue of the product. Therefore the

establishment of the high heat-resistant resin packagfing technology is in demand. This study

is aimed for proposed of the adhesion reliability evaluation method of the resin packaging.

Specifically, it was intended to establish more quantitative method by improving previous

adhesion evaluation method

The delamination load in the

conventional resin adhesion

evaluation method is found by

pushing with a tool, and destroying

the resin part of specimen which

bonded resin to the substrate. The

delamination stress is provided by

inputting the load into analysis. The

test like that is called "pudding-cup

test" because of form of specimen.

The specimen and state of the

examination are shown in Fig. 1.

However, the conventional

pudding cup test can evaluate only

limited stress ratio of vertical

direction and shear direction.

Therefore, a new method that can

evaluate a ratio of various stress was

examined by improving a

conventional test method.

Specifically, Torsion occur in the

adhesion interface of the specimen

by attaching a jig on the specimen

like a cantilever and pushing it. The flat knob district is prepared into the specimen to attach

this jig. This test method is called "a new pudding cup test" as follows, and the state is shown

in Fig. 2.

The strength of the torsion can be controlled by adjusting the load point on a cantilever. If

the test is reproduced on analysis, an interfacial stress state at the time of the fracture becomes

clear. The result of preliminary analysis is shown in Fig. 3. From a Fig. 3, the shear stress in

Fig. 1. Pudding-cup test.

Fig. 2. New pudding-cup test.



106

the aspect (xz, yz plane) takes the maximum on the edge lap of the adhesion side, and the normal

stress also takes the maximum in one point in the edge lap.

Thus, the delamination

occurs in an edge lap at a

maximum point of the

normal stress. A fracture

condition is found by

plotting the normal and shear

stress at the starting point of

delamination.

Room temperature

(25 degrees), a ceramic

substrate, epoxy resin were

used as an examination

condition in this time. Three

points 0[mm], 40[mm] and

100[mm] from the center of

the pudding cup were chosen as a load point. The

universal Pulling-bending testing equipment was

used.

The plot of the delamination stress that is

found from the experiment and analysis is shown

in Fig. 4. Only the case that normal stress was

dominant was found out in the conventional

examination, but, by the new pudding cup test, the

delamination condition that the shear stress is

dominant became available to be found out. As

overall tendency, the graph is like a flat oval.

Therefore, the normal stress gives the bigger

contribution to the delamination compared to the

shear stress.

Thus, the packaging resin adhesion reliability evaluation method that could cover a wider

case was developed. In this time, it was carried out with room temperature, a ceramic substrate,

but the evaluation in wider condition will be possible in the future, because even if these

conditions are changed, the method of the examination does not change. If graphs such as Fig. 4

from various cases are found, it can be easily checked where delamination occurs in comparison

with a true product. Therefore, it will be able to contribute to the reliability evaluation of the

future product.

REFERENCES

[1] MIYOSHI, T., SHIRATORI, M., ODA, J.: Daigakukiso Zairyourikigaku, Zikkyou Shuppan:

2006.

[2] Technical Information Institute Co.,Ltd.: Jushi to Kinzoku no settyaku setsugou gijutsu,

Nihoninsatsu, 2012.

Fig. 3. Result of preliminary analysis.

Fig. 4. Delamination condition of resin.



107

DIRECT BONDING OF Ti/Al BY METAL SALT GENERATION BONDING

TECHNIQUE WITH FORMIC ACID

T. AKIYAMA1*, S. KOYAMA2

1Graduate School of Science and Technology, Gunma University, Japan; email: [email protected] 2Faculty of Science and Technology, Gunma University, Japan

KEY WORDS: surface modification, fracture, bonding strength, Ti, Al, formic acid

In the past, methods such as brazing, friction stir welding and laser welding have been used

for bonding titanium alloy and aluminum alloy. However, these techniques have some

shortcomings: (1) microcracks are developed owing to the softening of the weld zone; (2) the

gap in the weld zone results in corrosion; and (3) a high heat input is required to compensate

for high heat radiation from titanium and aluminum. Moreover, the adverse effect of halogens

present in the flux on the environment is also a cause of concern. In addition, tools used for

friction stir welding have a short lifespan and this translates to higher running costs.

Furthermore, aluminum is an excellent heat radiating and electricity conducting element;

therefore, it is difficult to bond titanium and aluminum using other welding methods. Because

of these limitations, solid-state bonding is considered to the most suitable method for bonding

materials at low temperatures. In recent years, as an alternative to fusion bonding, solid-state

bonding to assemble electronic parts have been proposed as packaging technologies for

miniaturizing medical equipment, and some progress has been made in their practical

application. Such components (low heat resistance and little mechanical strength) require a

bonding method of low bonding temperature and pressure. The problem is that at an actual

bonding surface, there is an oxide film and a machined layer. Therefore, the adhesion between

surfaces at the bond interface is the removal of the oxide film is needed to obtain high strength.

Recently, ultrasonic vibration or plasma processing has been studied as a method for breaking

and cleaning a superficial oxide film. Indeed, in an earlier study, we showed that modification

of an oxide film with formic acid greatly improves the strength of bonding between tins and tin

and copper [1, 2].

In this paper, the effect of metal salt generation processing on the bond strength of the solid-

state bonded interface of titanium and aluminum has been investigated by SEM observation of

the interfacial microstructures and fractured surfaces. A cylindrical Ti specimen (Table 1) with

dimensions of φ10 mm × 20 mm and cylindrical Al specimen (99.9% purity) with dimensions

of φ20 mm × 20 mm was used in this experiment. Titanium surfaces were modified by boiling

in formic acid (FA) for predetermined time. The faying surface of the aluminum was finished

by electrolytic polishing. Solid-state

bonding was performed in N2 gas at

bonding temperature of 733-773 K

under a pressure of 12 MPa (bonding

time of 900 s).

Fig. 1 represents the relationship between metal salt generation processing time and tensile

strength. As shown in Fig. 1, the optimum value of metal salt generation processing time was

observed. It is inferred that the processing time is long; excessive metal salt were generated and

the processing time is short; an oxide film were remained on the bonding surface.

Fig. 2 shows the relationship between the bonding temperature and tensile strength of the

joint. In order to illustrate the effect of metal salt generation processing, the corresponding

relationship for a non-modified joint are also shown. As shown in Fig. 1, the tensile strength

Table 1 Chemical composition of Titanium used in this study.

Elements H O N Fe C Ti

wt% 0.0012 0.109 0.004 0.034 0.004 Bal.



108

increased with bonding temperature irrespective of

metal salt generation processing and approached

about 50 MPa at a bonding temperature

approximately 25 K or more lower than that for the

non-modified joint.

To examine the factors determining fracture at

the bond interface, the area of the fractured surface

was observed. As shown in Fig. 3, when the surface

modification was not applied and the bonding

temperature of 733 K, substances are not found to

adhere to either surface. With a rise in bonding

temperature, substances came to be observed in a

pair in the fractured surfaces. When the metal salt

generation processing was applied (bonding

temperature of 733 K), the fractured surface started

to show ductile fracture characteristics, although it

was not observed when the metal salt generation

processing was not applied. Thus, it was

hypothesized that high-tensile strength joints were

obtained at a lower bonding temperature with metal

salt generation processing because the contact are

between atomic plane was increased in the bonding

process.

Acknowledgement: This study was supported

by Grant-in-Aid for Scientific Research (c)

(26820124) from Japan Society for the Promotion of

Science (JSPS).

REFERENCES

[1] KOYAMA, S., AOKI, Y., SHOHJI, I.: Effect of Formic Acid Surface Modification on Bond

Strength of Solid-State Bonded Interface of Tin and Copper, Materials Transactions 51: 2010,

pp. 1759-1763.

[2] KOYAMA, S., OYA, I.: Effect of Formic Acid Surface Modification on Bond Strength of Solid-

State Bonded Interface of Tin, J. Japan Inst. Metals 73: 2009, pp. 809-815.

Fig. 1. Modification time vs. tensile strength. Fig. 2. Effect of surface modification on the relation

between tensile strength of joint and bonding temperature.

Fig. 3. SEM images and EDX analysis results of

the fractured surfaces (Ti side).

T =

73

3 K

T =

75

3 K

non-modified

Ti

Ti

T =

75

3 K

T =

73

3 K

modified (FA)

σ = 11.8 MPa

σ = 23.7 MPa

σ = 21.9 MPa

σ = 58.3 MPa

10 μm



109

DIRECT BONDING OF SUS304 STAINLESS STEEL BY METAL SALT

GENERATION BONDING TECHNIQUE WITH FORMIC ACID

T. TSUNETO1*, S. KOYAMA2


KEY WORDS: metal salt generation bonding, fracture, bonding strength, SUS304, formic acid

In recent years, the demand for energy-efficient devices are increasing as societies around

the world are becoming more environmentally conscious. Efforts to address this demand are

being made in various fields including the medical equipment (automated bio chemical analyser

and artificial heart-lung machine). Therefore, SUS304 stainless steel which is excellent in

corrosion resistance, toughness and workability is widely used for manufacturing medical

devices. We propose assembling medical devices using solid-state bonding rather than the

fusion bonding method; this method is suitable for miniaturized medical equipment. Earlier

research showed that surface modification decreased the bonding temperature required to obtain

a high-bond-strength joint in the solid-state bonding of Cu/Sn [1].

In this investigation, we aimed to obtain a deeper understanding of the effect of metal salt

generation bonding process on the performance of a solid-state bonded joint of SUS304

stainless steel by SEM observation of interfacial microstructure and fractured surfaces. The

specimen to be bonded was a block 20 mm × 15 mm × 5 mm and a 5 mm × 100 mm × 0.178 mm

sheet cut from SUS304 stainless steel plate. SUS304 stainless steel surfaces were modified by

boiling in formic acid (50%) for 660 s. Solid-state bonding was performed in N2 gas at bonding

temperature of 1023-1123 K under a pressure of 147 N (bonding time of 1800 s). In addition,

the specimen was readied for solid-state bonding within 180 s after metal salt generation

processing to avoid oxidation or changes in the bonding surface due to moisture absorption.

After solid-state bonding, the peel strength was evaluated using universal testing machine at

room temperature and a displacement speed of 0.017 mm/s.

Fig. 1 represents the relationship

between the bonding temperature and peel

strength of the joint. In order to illustrate the

effect of metal salt generation processing,

the corresponding relationship for an

unmodified joint and metal salt generation

processing with formic acid is also shown.

As shown in Fig. 1, the peel strength

increase with bonding temperature

irrespective of metal salt generation

processing and approached 600 N at a

bonding temperature approximately 100 K

lower than that for the unmodified joint. At

a bonding temperature of 1073 K or more, the peel strength was higher about 300 N than that

of the non-modified joints. Therefore, it is inferred that the effect of metal salt generation

processing were exerted at about 1073 K. Moreover, in the case of the surface were modified

and a bonding temperature of 1123 K, fractures of the base metal a part in the joint occurred.

To examine the factors determine fracture at the bond interface, the area of the fractured

surface was observed with SEM. The results are shown in Fig. 2. As shown in Fig. 2, when the

Fig. 1. Bonding temperature vs Peel strength.



110

surface was not modified and the bonding

temperature of 1023 K, substances are not

found to adhere to either the surface. With a

rise in bonding temperature, substances came

to be observed in a part in the fractured

surfaces. When the surface was modified and

the bonding temperature of 1023 K, the

fractured surface started to show ductile

fracture characteristics, although it was not

observed when the surface modification was

not applied. Thus, it is hypothesized that high-

peel-strength joints were obtained at a lower

bonding temperature with metal salt

generation processing because the contact area

between SUS304 stainless steel was increased. It is

generally thought that SUS304 stainless steel

exposed to the atmosphere are immediately

covered with an oxide film. It is also well

known that these oxides are factors that

obstruct the increase in bond strength. It is

known that nickel (II) formate are formed by

boiling or exposing these oxide film and base

metal with formic acid. The GIRAS-IR

spectrum are shown in Fig. 3. Whereas the IR

spectrum of the non-modified sample shows

only spectra characteristic to siloxane

compounds as contaminations during the

polishing process, the IR spectrum of

modified sample shows IR absorption bands

characteristic to carboxylate at 1350 and

1650 cm-1, which indicates the existence of

nickel (II) formate at the surface. It is known

that at about 403 K, nickel (II) formate

undergoes an endothermic decomposition

reaction, as shown by following formula, to generate metallic nickel:

Ni(HCOO)2 → Ni + H2↑ + 2CO2↑. (1)

It is therefore thought that a high-peel-strength joint was obtained at a lower bonding

temperature with metal salt generation process because metal salt such as nickel (II) formate at

the bond interface underwent a decomposition reaction during bonding.

Acknowledgement: This study was supported by Grant-in-Aid for Scientific Research (c)

(26820124) from Japan Society for the Promotion of Science (JSPS).

REFERENCES



pp. 1759-1763.

Fig. 2. SEM micrographs of fractured surfaces of joints

after peel test: as polished and modified.

Fig. 3. GIRAS-IR analysis of Ni surface : as polished

and modified.

as polished modified

T=

10

23

KT

= 1

073

KT

= 1

12

3 K

10 μmA

bso

rban

ce(a

.u.)

modified

as polished

Wavenumbers, cm-1

700 1100 1500 1900

1350

1650



111

DIRECT BONDING OF A6061 ALUMINUM ALLOY BY METAL SALT

GENERATION BONDING TECHNIQUE WITH FORMIC ACID

Y. TOMIKAWA1*, S. KOYAMA2


KEY WORDS: metal salt generation bonding, fracture, bonding strength, A6061, formic acid

In recent years, the ease of recycling material and the demand for energy-efficient devices

are increasing as societies around the world are becoming more environmentally conscious.

Efforts to address this demand are being made in various fields including the automotive

industry where attempts have been made to reduce the weight of cars and vessel. In particular,

A6061 aluminum alloy have high strength, extrudability and are easily recyclable owing their

low melting point. Therefore, A6061 aluminum alloy is widely used for manufacturing various

structures. In solid-state bonding, materials are bonded together by applying heat and pressure

to promote interdiffusion without any significant deformation of the materials. Moreover, since

solid-state bonding involves only low bonding temperatures, the damage to a component is

lesser than that in other methods, thus making this method suitable for bonding precision

assemblies. However, in reality, a bonding surface has various defects such as surface

irregularities, an oxide film and processing layer, which act as inhibitors to successful bonding

[1]. During the early stages of solid-state bonding, surface irregularities on the bonding surfaces

from closing properly. Later, as these gaps close, the oxide films on the bonding surfaces

prevent the surface atoms from coming into contact with each other, thus lowering the bonding

strength. Therefore, it is necessary to pre-treat a bonding surface before solid-sate bonding to

remove any oxide film present on the surface.

In this paper, we examine the effect of acetic acid surface modification on the bond strength

of the solid-state bonded interface of A6061 aluminum alloy was investigated by SEM

observations of interfacial microstructures and fractured surfaces. A cylindrical Al alloy

specimen (Table 1) with dimensions of φ10 mm × 10 mm was used in this study. Before

undergoing metal salt generation, the bonding surface was polished using electrolytic polishing

method. The metal salt generation processing is carried out by boiling the aluminum alloy

surface in a 5% NaOH solution for predetermined time and in a fixed at around 373 K.

Solid-state bonding was performed in the atmosphere under the following conditions: bonding

pressure, 18 MPa; bonding time, 900 s; and bonding temperature, 693-733 K. After solid-state

bonding, the interfacial strength was evaluated using the tensile test. The tensile test was

performed using a universal testing machine at room temperature and a displacement speed

0.167 mm/s. To specify the kind of compound formed on the surface after metal salt generation

processing, the aluminum surface was identified by FT-IR.

Fig. 1 represents the

relationship between the

bonding temperature and

tensile strength of the joint.

In order to illustrate the effect of metal salt generation processing, the corresponding

relationship for unmodified joints, modified by aqueous NaOH solution joint and metal salt

generation processing with acetic acid is also shown. As shown in Fig. 1, the interfacial strength

increased with bonding temperature irrespective of metal salt generation processing and

approached a target interfacial strength (40 MPa) at a bonding temperature approximately 20 K

lower than that for the unmodified joint and modified by aqueous NaOH solution joint.

Table 1 Chemical composition of A6061 aluminum alloy.

Elements Si Fe Cu Mn Mg Cr Zn Ti Al

mass% 0.68 0.30 0.31 0.11 1.00 0.16 0.05 0.02 Bal.



112

Moreover, at a bonding temperature of 693 K, the modified joint had interfacial strength

approximately 3 times as large as the unmodified

joint and modified by aqueous NaOH solution joint.

Therefore, it is inferred that the effect of metal salt

generation processing were exerted at about 693 K.

To examine the factors determine fracture at the

bond interface, the area of the fractured surface was

observed with SEM. The results were shown in

Fig. 2. As shown in Fig. 2, when the surface was not

modified and the bonding temperature of 693 K,

substances are not found to adhere to either the

surface. With a rise in bonding temperature,

substances came to be observed in a pair in the

fractured surfaces. When the surface was modified

by aqueous NaOH solution and metal salt generation

processing were applied (bonding temperature of

713 K), the fractured surface started to show ductile

fracture characteristics, although it was not observed

when the surface modification was not applied.

Thus, it is hypothesized that high-interfacial strength

joints were obtained at a lower bonding temperature

with metal salt generation processing because the

contact area between bonding surface was increased.

Fig. 3 shows the FT-IR results of aluminum

surface that is modified by acetic acid for metal salt

generation after modifying the surface with aqueous

NaOH solution. As shown in Fig. 3, the difference in

the amount of aluminum oxide between non-metal

salt generated and metal salt generated aluminum

can also be observed. It is known that aluminum and

aluminum oxide are changed into aluminum acetate

by boiling the surface in acetic acid after boiling the

surface in aqueous NaOH solution. As for the

thermal decomposition of aluminum acetate,

literature review shows that the compound thermally

decomposed into Al2O3 granules, H2O, CO and CO2

at temperature ranging from 473-723 K. It is

therefore thought that a high-interfacial-strength

joint was obtained at a lower bonding temperature

with metal salt generation process because aluminum acetate at the bond interface underwent a

decomposition reaction during bonding.



REFERENCES

[1] KOYAMA, S., KEAT, T. S., AMARI, S., MATSUBARA, K., SHOHJI, I.: Effect of Surface

Modification by Aqueous NaOH Solution on Bond Strength of Solid-State Bonded Interface of

Al, Materials Transactions 54: 2013, pp. 1975-1980.

Fig. 1. Effect of surface modification of the

relation between interfacial strength and

bonding temperature.

Fig. 2. SEM micrographs of the fractured

surfaces of joint after tensile test.

Fig. 3. FT-IR analysis results.



113

EFFECT OF SURFACE MODIFICATION BY AQUEOUS NaOH SOLUTION

ON BOND STRENGTH OF A5052 ALUMINUM ALLOY/Al AND Cu/Al

X. MA1*, S. KOYAMA2


KEY WORDS: surface modification, fracture, bonding strength, A5052, Al, Cu, NaOH solution

Aluminum is widely used for manufacturing cars and electronic devices. Currently, the

most common method for bonding aluminum surfaces is brazing. However, brazing requires

positional accuracy and results in the formation of voids by the flux residue; therefore, to avoid

these problems, solid-state bonding methods are considered as a possible alternative. However,

solid-state bonding also suffers from some problems that need to be overcome. One of these

problems is the presence of an oxide film on aluminum surfaces, necessitating the need to

remove or destroy the oxide film without applying high temperature and high load. Moreover,

these techniques have some shortcomings: (1) microcracks are developed owing to the

softening of the weld zone; (2) the gap in the weld zone results in corrosion; and (3) a high heat

input is required to compensate for high heat radiation from aluminum. Furthermore, aluminum

is an excellent heat radiating and electricity conducting element; therefore, it is difficult to bond

aluminum using other welding methods. Because of these limitations, solid-state bonding is

considered to be the most suitable method for bonding materials at low temperatures [1].

In this study, a bonding

surface was treated with

NaOH (aq) for removing the

oxide film; moreover, the

effectiveness of this treatment was determined

by observing the bonding interfaces and

fractured surfaces of specimens. A cylindrical Al

specimen (99.9% purity) with dimensions of

φ20 mm × 15 mm, a cylindrical A5052

aluminum alloy (Table 1) with dimensions of

φ10 mm × 15 mm and cylindrical Cu specimen

(99.9% purity) with dimensions of φ10 mm ×

15 mm was used in this experiments. Before

surface modification, the bonding surface was

polished with a #800 emery paper. Surface

modification was carried out by boiling the

aluminum specimen in a 5% NaOH solution for

20 s with the solution temperature fixed at

around 373 K. The bonding surface was then

washed with methyl alcohol. Solid-state bonding was performed in N2 gas at bonding

temperature of 753 K under a pressure of 12 MPa (bonding time of 900 s). Furthermore,

grazing-angle incidence reflection-absorption infrared (GIRAS-IR) spectroscopy was

employed to obtain IR spectra for the modified aluminum surface at nanometer-scale depth.

The spectra were obtained using a Fourier transform infrared (FTIR) spectrometer (Thermo

Fisher Inc., Magna-750) equipped with a mercury-cadmium-telluride (MCT) detector using a

single reflection accessory (Harrick Inc., Seagull) at an incidence angle of 80°. An Au-coated

mirror surface was used as the reflecting surface and measurements were carried out at a

Table 1 Chemical composition of A5052 aluminum.

Elements Si Fe Cu Mn Mg Cr Zn Ti Al

mass% 0.11 0.21 0.04 0.05 2.60 0.18 0.02 0.00 Bal.

Fig. 1. The effect of surface modification on tensile

strength of joints.



114

wavenumber resolution of 8 cm-1. Moreover, the modifying actions of NaOH (aq) on the

aluminum surface, after boiling a 10-µm-thick aluminum foil in NaOH (aq) for 30 s, we

immediately subjected the specimen to differential scanning calorimetry (DSC, SII Seiko

Instruments, DSC6200).

Fig. 1 shows the effect of surface modification

on the bond strength of joint. As shown in Fig. 1,

the tensile strength increased with the surface

modification irrespective of bonding material.

However, when surface modification was applied

for the joint between A5052 aluminum alloy and

aluminum, the A5052/Al joint had a bigger ratio of

rise in strength than the Cu/Al joint. On the basis of

these results, it was inferred that the Cu/Al joint had

low tensile strength because of the formation of an

oxide film between the bonding surfaces, inhibiting

intimate contact between the atomic planes of Cu

and Al. It can be explained from difference in

volume of deformation of the joint after solid-state

bonding.

The GIRAS-IR spectra are shown in Fig. 2.

The IR absorption bands at about 3400 cm-1

correspond to a hydroxyl group, those at 1600 cm-1

correspond to hydrogen carbonate and those at less

than 950 cm-1 correspond to aluminum oxide. The

results shown in Fig. 2 revealed that the oxide

formation was decreased and Al(OH)3 was

generated upon surface modification. From this, it

can be inferred that the bonding surface of the

specimen used in this study is also covered with

Al(OH)3 when it was washed with methanol after

NaOH (aq) surface modification. Moreover, it is

also known from other studies that Al(OH)3

decomposed into particles of Al2O3 and H2O by an

endothermic and dehydration reaction at about

573 K. Actually, as shown in Fig. 3, the generation

and thermolysis of Al(OH)3 were supported by a bigger endothermic peak as recognized by a

gentle endothermic peak. Because the temperature used in solid-state bonding in this study was

at least 573 K or more, it can be inferred that the thermal decomposition of Al(OH)3 to H2O

(gas) and particles of Al2O3 occurred during the bonding process, thus resulting in the exposure

of atomic planes of aluminum. This caused the tensile strength of the joint that undergone

surface modification to increase.



REFERENCES




Fig. 2. FT-IR analysis results of the aluminum

surfaces with/without surface modification.

Fig. 3. DSC analysis results of the aluminum

sheet with surface modification.



115

DIRECT BONDING OF Cu/Cu BY METAL SALT GENERATION BONDING

TECHNIQUE WITH FORMIC ACID AND ACETIC ACID

S. KOYAMA1*, N. HAGIWARA2, I. SHOHJI1

1Faculty of Science and Technology, Gunma University, Japan; email: [email protected] 2Gracuate School of Engineering, Gunma University, Japan

KEY WORDS: metal salt generation bonding, fracture, bonding strength, copper, organic acid

As an alternative to fusion bonding, solid-state bonding to mount electronic parts have been

proposed as packaging technologies for miniaturizing electronic devices. Meanwhile, a number

of mounting parts made of resin have low heat resistance and little mechanical strength. Such

components require a bonding method of low bonding temperature and pressure. The problem

is that at an actual bonding surface, there is an oxide film and a machined layer [1]. Therefore,

the removal of the oxide film is needed to obtain high bond strength. Recently, ultrasonic

vibration or plasma processing has been studied as a method for breaking and cleaning a

superficial oxide film. Indeed, in an earlier study, we showed that modification of an oxide film

with formic acid greatly improves the strength of bonding of tin and copper [2].

In this paper, we examine the effect of organic acid surface modification on the bonding

strength of the solid-state bonded interface of copper was investigated by SEM observations of

interfacial microstructures and fractured surfaces. The specimen to be bonded was a block

15 mm × 15 mm × 5 mm cut from 99.9% ingot and a wire (99.9% purity) dimension of

φ1.2 mm. Copper surfaces were modified by boiling in formic acid and acetic acid for

predetermined time. Solid-state bonding was performed in a vacuum chamber at bonding

temperature of 423-673 K under a pressure of 588 N (bonding time of 0.9 ks).

Fig. 1 represents the relationship between the bonding temperature and peel strength of the

joint. In order to illustrate the effect of metal salt generation processing, the corresponding

relationship for a non-modified joint are also shown. As shown in Fig. 1, the peel strength

increased with bonding temperature irrespective of metal salt generation processing and

approached about 30 N at a bonding temperature approximately 150 K (with formic acid) and

100 K (with acetic acid) or more lower than that for the non-modified joint.

To examine the factors determining

fracture at the bond interface, the area of the

fractured surface was observed. As shown in

Fig. 2, when the surface was not modified,

approximately 0.5 µm-diameter copper

oxides were observed at the fractured

surfaces. In addition, ductile fracture

characteristics such as tear ridges were not

observed. When the surface was modified

with formic acid and acetic acid, the fractured

surface started to shown ductile fracture

mode, although it was not observed when the

surface modification was not applied. At the

same time, copper oxide was not observed at

the fractured surfaces. From these results, it was inferred that the joints bonded at low bonding

temperatures had high peel strength because of the removal of oxide film, thus allowing

intimate contact between the atomic planes of copper.

Fig. 1. Effect of surface treatment: relation between

bonding temperature and peel strength.



116

It is generally thought that copper exposed

to the atmosphere are immediately covered

with an oxide film. It is known that copper (II)

formate and copper (II) acetate are formed by

boiling the oxide film and the base metal with

formic acid and acetic acid for a long time.

Therefore, it is thought that at least

Cu(HCOO)2 and Cu2(CH3COO)4 were formed

on the surface layer when metal salt generation

process was applied. Moreover, it is known

that copper (II) formate and copper (II) acetate

undergoes an exothermic decomposition

reaction and metallic copper is generated at a

temperature using this study. It is therefore

thought that a high tensile-strength joint was

obtained at lower bonding temperature with

surface modification because copper (II)

formate and copper (II) acetate at the bond

interface underwent a decomposition reaction

after close contact between Cu/Cu was

achieved.

Fig. 3 represents the relationship between

the shelf time of the modified surface and peel

strength of the joint. In case the surface is

modified by formic acid, the peel strength was

decreased in several hours, but the case the

surface is modified by citric acid, the peel

strength was not decreased even for 168 hours.

Generally, it was understood that copper

formate is easy to dissolve in water but copper

acetate is hard to dissolve in water. Thus, it can

be inferred that a bonding surface modified

with acetic acid is also protected from

reoxidation.



REFERENCES






pp. 1759-1763.

Fig. 2. SEM micrographs of the fractured surfaces of

joints after tensile test with or without surface

modification.

Fig. 3. Effect of surface treatment: relation between

shelf time and peel strength.



117

DELAMINATION PROPERTY OF MODELLED AIR PLASMA SPRAYED

THERMA BARRIEAR COATINGS: EFFECT OF DIFFERENCE IN

CHEMICAL COMPOSITION OF BOND COAT

M. HASEGAWA1*, S. YAMAOKA1

179-5 Tokiwadai, Hodogaya-ku, Yokohama, 240-8501, Kanagawa, Division of Materials Science and Chemical Engineering, Faculty of Engineering, Yokohama National University, Japan; email: [email protected]

KEY WORDS: microstructure, yield stress, delamination toughness

Thermal barrier coatings (TBCs) have been widely used in order to increase the operating

temperature of hot section components in gas turbine blades and vanes [1]. TBCs are usually

composed of outer ceramics thermal barrier coating (TBC) layer and inner intermetallic bond

coat (BC) layer to protect the nickel base superalloy from high temperature and oxidation.

During the service, formation and growth of thermally grown oxide (TGO) occurs between

TBC and BC layer. This growth increases the thermal stress of TBCs, and finally, the increased

thermal stress results in the delamination of the coating. In order to understand the effect of the

difference in chemical composition and microstructure of BC layer on delamination properties,

mechanical properties are examined on modelled TBCs.

Nickel-platinum-aluminides (Ni-Pt-Al) and NiCoCrAlY alloy were selected as BC alloy.

Chemical compositions of BC alloys were Ni-43Al-9Pt, Ni-42Al-9Pt-0.3Hf and

Ni-25Al-19Co-16Cr-0.4Y (mol%). BC alloys were heat treated in a vacuum at 1413 K for

1 hour. After the treatment, modelled TBCs have been formed by air plasma-spraying process.

TBC layer of an 8 mass% Y2O3 partially stabilized ZrO2 was coated on the BC alloy in 250 m

thick. After the process, the TBCs were heat exposed in an air at 1323 K from 10 to 100 hours.

Changes in microstructure during heat exposure were characterized on the polished transverse

section of the TBCs by SEM and EBSD. Yield stresses of BC alloys were decided from the

result of Vickers hardness measurement. To evaluate the delamination toughness of the TBCs

under shear loading condition, pushout tests were performed [2, 3].

Ni-Pt-Al and NiCoCrAlY BC alloys are consist of single phase and (’) two phases,

respectively. phase of NiCoCrAlY BC alloy near the TGO/BC interface disappears during

heat exposure, due to the formation and growth of the TGO. In case of the TBCs with

NiCoCrAlY BC alloy, yield stress of the BC alloy decreases with the increase in heat exposure

time. However, as for the TBCs with Ni-Pt-Al BC alloy, yield stress was almost constant

independent of the heat exposure time. TGO thickness increases with the increase in heat

exposure time. TBCs with Ni-Pt-Al alloy shows thinner TGO thickness than that of the TBC

with NiCoCrAlY alloy in a same heat exposure time. Fracture of the TBCs occurs mainly at

TBC/TGO interface. TBCs having Ni-Pt-Al BC alloy shows higher shear strength and

delamination toughness than that of the TBCs with NiCoCrAlY BC alloy.

REFERENCES

[1] PADTURE, N. J., GELL, M., JORDAN, E. H.: Finite Thermal Barreir Coatings for Gas-

Turbine Engine Applications, Science 296, 2002, pp. 280-284.

[2] KIM, S. S., LIU, Y. F., KAGAWA, Y.: Evaluation of Interfacial Mechanical Properties under

Shear Loading in EB-PVD TBCs by the Pushout Method, Acta Mater. 55, 2007, pp. 3771-3781.

[3] HASEGAWA, M., ENDO, T., FUKUTOMI, H.: The Effect of Microstructure Change of Bond

Coat Layer in Air Plasma-Sprayed Thermal Barrier Coating System on Interfacial Mechanical

Property under Shear Loading, J. Japan Inst. Metals 73, 2009, pp. 802-808.



118



119

MICROSTRUCTURE MODIFICATION OF CGDS AND HVOF SPRAYED

CoNiCrAlY BOND COAT REMELTED BY ELECTRON BEAM

P. GAVENDOVÁ1*, J. ČÍŽEK1, J. ČUPERA1, M. HASEGAWA2, I. DLOUHÝ1

1Brno University of Technology, Institute of Materials Science and Engineering, NETME centre, Technická 2, Brno, Czech Republic; email: [email protected]

2Yokohama National University Faculty of Engineering, Division of Materials Science and Chemical Engineering, Yokohama, 240-8501, Japan

KEY WORDS: CoNiCrAlY, bond coat, thermal spray, cold kinetic deposition, electron beam

remelting, thermal barrier coating

The efficiency of gas turbine engines can be significantly improved by increasing their

operating temperatures. The engine components, especially the hot-end components, should

maintain their mechanical integrity at high-temperatures. The complex environment of the gas

turbine engine makes it extremely difficult for one material to meet all the different

requirements imposed on the various engine components. Coatings are thus required to provide

increased corrosion (and oxidation) protection at high temperatures in order to assure durability

and field performance of the base alloys. Ni-base turbine blades are generally protected by a

CoNiCrAlY overlay and a ceramic thermal barrier top layer, both thermally sprayed. The

method used for these coatings is usually high-velocity oxygen-fuel (HVOF) spraying, other

methods can be also applied however.

Promising results can be obtained with cold kinetic deposition method and/or cold gas

dynamic spraying (CGDS) [1]. It is a method that enables to get coatings with lower porosity,

higher relative density and better adhesion of the (bond) coating to substrate. From that point

of view it seems that this technique possesses a better potential of for bond coat fabrication

comparing to HVOF etc.

A subsequent remelting of the HVOF by electron beam (EB) irradiation opens another

opportunity to minimize the oxide content (because of vacuum condition needed for EB

technology) and the porosity (thanks to remelting and the melt rapid cooling) of the thermally

sprayed CoNiCrAlY bond coatings. The electron beam remelting process appears to be one of

the most convenient processes to reduce or fully remove the disadvantages of thermal spray

coatings. The bond coats remelted by electron beam produce modifications in the morphology

and phase composition [2] that could be further exploited for TBC performance optimisation.

In the present work two techniques are combined to optimise bond coat properties before

TBC application, the cold gas dynamic spraying (CGDS) and electron beam remelting (EB).

Results of the work focused on comparison of HVOF and CGDS CoNiCrAlY bond coats are

firstly presented. Than the effect of the electron beam remelting of the CoNiCrAlY coating

manufactured by HVOF and CGDS deposition techniques is deeply investigated. Scanning

electron microscopy, light microscopy, and, in addition, X-Ray diffraction techniques were

performed to characterize the phase modification and microstructure composition changes

before and after the treatment. The microstructural and phase analyses have been supported by

microhardness and nanohardness investigation and other necessary supporting techniques.

The bond coat having thickness of about 70 m prepared by both HVOF and CGDS

technique displayed the lower porosity for the CGDS microstructure. The CoNiCrAlY bond

coat to Inconel substrate interface displayed locations with very poor bonding, in larger extent

for the states prepared by HVOF comparing to CGDS. The bond coats prepared by both ways

being EB remelted up to depth of about 90 m are typical by removal of the defects on the




120

substrate to bond coat interface. The microstructure of the bond coat after this treatment has

been is formed by Inconel fine grain layer in depth of 50 m being followed by the surface

layer consisting of elongated dendritic microstructure. The longitudinal axis of dendrites has

been oriented predominantly perpendicularly to the Inconel surface. An increased porosity has

been observed in interdendritical space in larger extent for CGDS samples.

The results obtained thus showed that using the pulsed electron beam surface modification

technique produced positive changes in the bond coat layer as a necessary step for the thermal

barrier coating fabrication. In addition the EB treatment provided a smooth surface, low

porosity level comparing to bond coat surface without this modification. Although the

technology parameters window for successful application of the CGDS BC application and

subsequent EB remelting has been shown to be relatively narrow this procedure supplied good

basis for TBC application when compared to standard HVOF technique and subsequent vacuum

ageing. From the results, it appeared also that there is a potential for improvements of the bond

coat deposition process when applying low-temperature processing methods such as CGDS.

Thermal barrier coatings (TBC) that have been applied on the Inconel substrate with

modified bond coat surface suppose an effective engineering solution for the improvement of

in service performance of gas turbines components. The quality and further performance of

TBC, likewise all thermally sprayed coatings is strongly dependent on the adhesion between

the coating and the substrate as well as the adhesion between the metallic bond coat and the

ceramic top coat layer. The debonding of the ceramic layer or the bond coat layer will lead to

the collapse of the overall thermal barrier system. Though several possible problems can occur

in coating applications like residual stresses, local or net defects (like pores and cracks), one

could say that a satisfactory adhesion is the first and intrinsic need for a good coating. The

coating adhesion is also dependent on the pair substrate-coating materials, substrate cleaning

and blasting, coating application process, coating application parameters and environmental

conditions.

Acknowledgement: The works have been supported by the financial support from the

Operational Programme Education for Competitiveness No. CZ.1.07./2.3.00/30.0005 and

within the project NETME plus centre (Lo1202), project of Ministry of Education, Youth and

Sports under the “national sustainability programme”. Support of Czech Science Foundation

project GACR 13/35890S is further acknowledged.

REFERENCES

[1] UTU, D., MARGINEAN, G., BRANDL, W., CARTIS, I.: Improvement of the oxidation

behaviour of electron beam remelted MCrAlY coatings, Solid State Sciences 7, 2005,

pp. 459-464.

[2] UTU, D., BRANDL, W., MARGINEAN, G CARTIS, I., SERBAN, V.A.: Morphology and

phase modification of HVOF-sprayed MCrAlY-coatings remelted by electron beam irradiation,

Vacuum 77, 2005, pp. 451-455.

[3] LIMA, C.R.C., GUILEMANY, J. M.: Adhesion improvements of Thermal Barrier Coatings

with HVOF thermally sprayed bond coats, Surface & Coatings Technology 201, 2007,

pp. 4694-4701.

[4] RICHER, P., YANDOUZI, M., BEAUVAIS, L., JODOIN, B.: Oxidation behaviour, of

CoNiCrAlY bond coats produced by plasma, HVOF and cold gas dynamic sprayings, Surface &

Coatings Technology 204, 2010, pp. 3962-3974.



121

RESPONSE OF ALUMINA FOAM TO TENSILE MECHANICAL LOADING

INCLUDING STRESS CONCENTRATOR EFFECT

I. DLOUHÝ1*, Z. CHLUP1, H. HADRABA1, L. ŘEHOŘEK2

1Institute of Physics of Materials, Academy of Sciences of the Czech Republic, CEITEC IPM, Zizkova 22, 61662 Brno, Czech Republic; email: [email protected]

2Military Research Institute, Division of Materials Engineering, Brno

KEY WORDS: tensile test, ceramics foam, open porosity, tensile strength, compression strength

Ceramic foams with open cell porosity are of technological interest because of their

potential use in a number of industrial fields. Applications like catalytic substrates [1], high

temperature filters for melted alloys [2], in tissue engineering as a bone replacement material

[3, 4], insulation materials [5] etc. require high permeability, high surface area and good

insulation characteristics, but also a good response to different types of mechanical loading

typical for given applications. For ceramic foams, mechanical behaviour including the fracture

resistance of thus plays an important role in their potential applications.

At present, it is common to estimate the mechanical performance of ceramic foams by

compressive tests only [5]. The most frequently used mechanical parameter is a

compressive/crushing strength that is evaluated from the compression test curve either as

maximum force at the first relevant peak or as average force of plateau observed on the curve.

Very limited number of works has attempted also a modified bending test [6]. The most

complication supposes the measures to avoid to the crushing between the rollers and the tested

ceramics foam. Suitable thin sheet must be applied into interface between the roller and

specimen surface. It must be rigid and tough but not too much to assure load transfer [7].

Data of tensile tests of the ceramic foams are completely missing in the literature. For

interpretation of the mechanical behaviour and modelling the ceramic foam response [8] in the

given applications some material data are needed however. The aim of the paper thus can be

seen in summarisation of the knowledge obtained with the tensile test of ceramics foams of

different samples geometry, interpretation of data obtained and possible consequences relating

to foam material mechanical behaviour. Tensile loading of ceramic specimens brings

difficulties. It is necessary to solve efficient load transfer and ensure alignment of the specimen

with loading axis of the system which is not simple task. Brittleness of this material brings

difficulties with fixation of material into claws. It is impossible to use any fixing methods using

compression, friction, threaded joint and their combinations. Only one possibility is to employ

adhesion evoked by some kind of adhesive or resin.

Material used in the investigation was alumina based foam (85 vol. % Al2O3, 14 vol. %

SiO2, 1 vol. % MgO) commercially produced (Vukopor®A) by company Lanik typically used

e.g. for aluminium alloys melt filtration. The foam structure was produced by replication

technique consisting in slurry coating of polyurethane foam. Two types of cell sizes were

applied for investigations, 2.2 (±1.2) mm and 0.8 (±0.3 mm), respectively. The dimensions of

test specimens with both porosity types (10 and 60 PPI) were 10x10x30 mm3 and

15x15x40 mm3 respectively. Samples containing central internal sharp notch located in

10x30 mm cross-section have been also included in investigation.

Special fixture and testing rig was developed to assure transfer of the load to the ceramic

foam specimen. Tensile strength values were determined as maximum force from the loading

diagram related to cross-sectional area of the specimen at fracture.

mailto:@ipm.cz



122

As expected, the scatter of apparent tensile strength values is affected by specimen size.

Specimens with cross section of 15x15 mm2 showed noticeable lower data scatter because of

larger sampling volume averaging the structural heterogeneity. As shown also elsewhere [7, 9]

noticeable scatter of apparent tensile strength is caused (i) by the heterogeneity in distribution

characteristics of the cell sizes in the given sample and, in addition, (ii) by fatal macroscopic

material defects typical for this kind of material and fabrication technology applied. The

observed relations between the ceramic foam microstructure and measured apparent tensile

strength confirm very good susceptibility of the measurements to microstructural differences

however.

The specimen size (in the range of dimensions applied in this investigation) does not affect

substantially the average values of tensile strength as it has been shown based on the Weibull

statistical analysis. From the data sets for the specimen sizes applied and for both cell sizes,

having 10 and 60 PPI, it is obvious that there are no substantial differences in the quality of data

obtained from smaller and larger specimens. Both specimen geometries, having larger and

smaller cross-section exhibit the same slope in Weibull distribution. Specimens with higher PPI

have lower apparent tensile strength comparing to those with lower PPI. The higher cell size

the higher apparent tensile strength is. This is given mainly by strut thickness of the cell which

is significantly lower in samples with 60 PPI comparing to the 10 PPI ones.

The samples with central crack (sharp notch) have included into investigation

experimentally to determine the possible stress concentration effects in open cell porosity

structures. Data from pre-cracked samples were compared with results of samples having the

same cross-section as remaining part of section in the pre-cracked samples. For the pre-cracked

samples, the strength values have been found lower comparing to samples without crack.

Independently of quite large open porosity there is still certain stress concentration effect in

these structures. This must be taken into account in the most critical structural applications.

The quantitative data from the tests supported well the corresponding fractography

observations. Based on the analyses carried out and supporting finite element modelling critical

condition for the struts fracture have been established.

Acknowledgement: The works have been realised in CEITEC centre - infrastructure

supported by the project CZ.1.05/1.1.002/02.0068 financed from Europepan Regional

Development Fund. Support of Czech Science Foundation project GACR 14-11234S is further

acknowledged.

REFERENCES

[1] GARRIDO G. I., PATCAS F.C., UPPER G. ET AL.: Applied Catalysis A: General 338 (2008)

58.

[2] PATCAS F.C., GARRIDO G.I., KRAUSHAAR-CZARNETZKI B., Chemical Engineering

Science 62 (2007) 3984.

[3] MIAO X., TAN L.P., TAN L.S., HUANG X.: Materials Science and Engineering C 27 (2007)

274.

[4] REHOREK L. CHLUP Z., MENG D., et al.: Ceramics International 39 (2013) 8015–8020.

[5] GIBSON L. J., ASHBY M.F.: Cellular Solids, Cambridge University Press, Cambridge, 1999.

[6] BREZNY R., GREEN D.J., Acta Metall. Mater. 38 (1990) 2517.

[7] REHOREK L., DLOUHY I., CHLUP Z. et al.: Ceramics – Silikáty 53 (4) (2009) 237-241.

[8] MARCIAN P. MAJER Z., DLOUHY I., FLORIAN Z.: Chem. Listy Vol. 106 (2012),

pp. 476-477.

[9] DLOUHY, I., REHOREK L., CHLUP Z.: Key Eng. Mat. Vol. 409 (2009), pp. 168-175.



123

NONDESTRUCTIVE MAGNETIC MONITORING OF GRINDING DAMAGE

M. ČILLIKOVÁ1*, B. MIČIETA1, M. NESLUŠAN1, D. BLAŽEK2

1University of Žilina, Faculty of Mechanical Engineering, Univerzitná 1, 010 26 Žilina, Slovak Republic;

email: [email protected] 2Nanotechnology Centre, VŠB TU Ostrava, 17. listopadu 15, 70833 Ostrava, Czech Republic

KEY WORDS: grinding, Barkhausen noise, surface damage, wear, white layer

Ground surface can suffer from over tempering or overheating due to elevated temperatures

in the wheel – workpiece contact. For this reason, thermally softener or/and re-hardened layers

can occur in the near or sub surface regions. Surface containing the untempered brittle

martensite or thermally softened structure are assumed detrimental to components life due to

early crack initiation and premature failure. Surface integrity expressed in such term as residual

stresses, hardness and the corresponding structure alteration can varies in spite of the cutting

conditions are kept constant. Thus implementation of reliable monitoring concept should be

proposed for the real industrial applications.

Production of large bearing for wind power station involves grinding cycles as the final

operation substantially contributing to the functionality of bearings. Bearings undergo very

rigorous monitoring procedures to reveal unfavourable stress state, structure modifications or

crack initiation originating from grinding cycles or heat treatment. Manufacturer guarantees at

least 20 years failure – free operation of bearing. Monitoring of bearings (mainly raceways) of

diameter in the range 600 to 4000 mm after grinding does not allow implementation of chemical

activation to reveal the unfavourable surface alterations induced by grinding.

Magnetic Barkhausen noise (BN) has found the high industrial relevance for

characterization of ferromagnetic materials. BN originates from irreversible domain and mainly

Bloch Wall (BW) motion during cyclic magnetization. Pulsating magnetization on hysteresis

loop occurs as a result of BW interaction with stress fields and microstructure features such as

dislocations, secondary phases or grain boundaries hindering the smooth BW motion [1]. BW

jumps occur as soon as the strength magnetic field exceed the critical value equal the pinning

strength of the abovementioned pinning sites. BN is the stress and microstructure sensitive

technique. However, while the stress state affects mainly domain and the corresponding BW

alignment, microstructure features affect the free path of BW motion [2]. Although variety of

BN applications were reported, monitoring of grinding burn still dominates as a reliable method

to reveal surface over tempering. While untouched surface emits low BN value, enhanced BN

emission in thermally softened layers is due to thermally induced decrease of dislocation

density, precipitates coarsening and tensile stresses [2].

This paper reports about adoption of BN technique for surfaces of large diameter (for wind

power stations) after grinding. The paper researches the factors taking a key role in grinding

cycles based on BN technique and the corresponding conventional techniques for surface

inspection. To explain the significance of such approach the two main factors of grinding

operations are discussed as follows: grinding wheel wear and the lack of coolant. First aspect

is associated with higher BN values obtained after grinding of bearings of higher diameter,

while the second one represent the unexpected lack of coolant supply randomly occurring in

production.



124

Fig. 1. Influence of grinding wheel wear and lack of coolant on BN and the corresponding micrographs.

Fig. 1 demonstrates the influence grinding wheel wear and lack of coolant on BN.

Progressive grinding wheel wear enhances thermal load of ground surfaces. Higher

temperatures penetrate deeper beneath the surface; thickness of heat affected zone (HAZ)

contributing to the BN emission more than untouched structures values is increasing. Moreover,

compressive stresses induced in the earlier grinding cycles are shifted to the tensile ones. Effect

of grinding wheel wear takes the main role considering higher BN values indicated for bearings

of higher diameters due to higher volume material removed and higher allowances for the

finishing grinding operations, both contributing to the more developed grinding wheel wear in

the final phases of grinding cycles.

On the other hand, effect of coolant supply is more remarkable than that associated with

grinding wheel wear. Lack of coolant is more risky when the grinding wheel is more developed

(during grinding rings of higher diameter - see the red and blue lines in Fig. 1) than the absence

of coolant in the early phases of grinding cycle or grinding smaller rings (see the black line in

Fig. 1). The abrupt increase of BN values in dry grinding is due to accumulation of heat in the

ground surface which in turn corresponds with quite thick HAZ. The following steep decrease

of BN values is attributed to the surface over tempering when austenitizing temperature is

exceeded. The following self-cooling effect results in formation white layer. Surface cracking

which occurs together with white layers is due to high internal stresses induced by rapid cooling.

Implementation of BN technique in the abovementioned production has found the high

relevance serving the technologist and grinding staff to modify the grinding cycles and

eliminates the risky factors of grinding operations.

Acknowledgement: The authors gratefully acknowledge the support by Vega project n.

1/0701/12 and Regional Centre of Excellence reg. no. CZ.1.05/2.1.00/01.0040.

REFERENCES

[1] GATELIER-ROTHEA, C., CHICOIS, J., FOUGERES, R., FLEISCHMANN, P.:

Characterization of pure iron and carbon-iron binary alloy by Barkhausen noise

measurements: study of the influence of stress and microstructure, Acta Mater. 46/14, 1998,

pp. 4873-4882.

[2] MOORTHY, V. et all.: Evaluation of heat treatment and deformation induced changes in

material properties in gear steels using magnetic Barkhausen noise analysis, ICBN 03,

Tampere, Finland 2001.





Autor Prof. Ing. Bohumír Strnadel, DrSc.

Katedra, institut: Katedra materiálového inženýrství 636

Název: NEW METHODS OF DAMAGE AND FAILURE ANALYSIS

OF STRUCTURAL PARTS

Místo, rok, vydání: Ostrava, 2014, VI.

Počet stran: 124

Vydala: VŠB-TECHNICKÁ UNIVERZITA OSTRAVA

Tisk: Printo spol. s r.o.

Náklad: 150

Neprodejné

ISBN 978-80-248-3488-7

Documents

NEW METHODS OF DAMAGE AND FAILURE ANALYSIS OF …konference.fmt.vsb.cz/work2014/files/New...Analysis... · data [4]. This is because the stress intensity factor (SIF) at point 1 in