G. DERCZ, I. MATU£A: EFFECT OF BALL MILLING ON THE PROPERTIES OF THE POROUS Ti–26Nb ...
795–803
EFFECT OF BALL MILLING ON THE PROPERTIES OF THE
POROUS Ti–26Nb ALLOY FOR BIOMEDICAL APPLICATIONS
VPLIV KROGLI^NEGA MLETJA NA LASTNOSTI POROZNE
ZLITINE Ti–26Nb ZA BIOMEDICINSKE APLIKACIJE
Grzegorz Dercz, Izabela Matu³a
University of Silesia, Institute of Materials Science, 75 Pu³ku Piechoty Street 1 A, 41-500 Chorzów, Poland
grzegorz.dercz@us.edu.pl
Prejem rokopisa – received: 2016-09-09; sprejem za objavo – accepted for publication: 2017-02-21
doi:10.17222/mit.2016.269
The current study investigated the effect of the ball-milling time on the structural characteristics and pore morphology of a
biomedical porous Ti–26Nb (a/%) alloy. An open-cell porous material was synthesized by mechanical alloying and sintering.
Commercially available elemental metal powders of Ti and Nb were used as the starting materials. Elemental metal powders
with a nominal composition of Ti–26Nb (a/%) were milled in a planetary ball mill. During the testing, the powders were milled
for two milling times: 50 h and 70 h. During the shorter milling time, the speed of 200 min–1 was applied, and during the longer
milling time, the speed of 400 min–1 was applied. The powders were cold pressed under a pressure of 750 MPa and then sintered
at 1000 °C for 24 h. The effects of the powder milling time on the microstructure and mechanical properties of the porous
structure were investigated with optic microscopy (OM), scanning electron microscopy (SEM), X-ray diffraction (XRD),
microhardness, nanoindentation and nanomachining tests. The hardness and elastic modulus were calculated from the
load-displacement data obtained with nanoindentation using a three-sided, pyramidal, diamond (Berkovich) indenter tip. The
X-ray diffraction results confirmed the presence of the  and  phases. An analysis of diffraction patterns revealed a decrease in
the lattice parameters for the 70 h milling. In summary, it should be pointed out that the material has a hierarchical structure,
obtained during the successive stages of milling. The observed grains are composed of many smaller grains, the size of which
decreased with an increase in the milling time.
Keywords: mechanical alloying, porosity, nanoindentation, Ti–26Nb alloy
V {tudiji so preiskovali vpliv razli~no dolgega ~asa krogli~nega mletja na strukturne zna~ilnosti in morfologijo por biomedi-
cinske porozne Ti-26Nb zlitine (a/%). Porozni material z odprtimi celicami je bil sintetiziran z mehanskim legiranjem in
sintranjem. Osnovni metalur{ki prahovi z nominalno vsebnostjo Ti-26Nb (a/%) so bili mleti v planetarnem krogelnem mlinu.
Med testiranjem so bili prahovi mleti pri dveh razli~nih ~asih: 50 h in 70 h. Med kraj{im ~asom mletja je bila uporabljena hitrost
mletja 200 min–1 in med dalj{im ~asom mletja je bila uporabljena hitrost 400 min–1. Prahovi so bili hladno stiskani pod tlakom
750 MPa in nato sintrani pri 1000 °C za 24 h. Preiskovali so vpliv razli~nega ~asa mletja prahu na mikrostrukturo in na
mehanske lastnosti porozne strukture z opti~nim mikroskopom, s SEM- in XRD-analizo ter dolo~evanjem mikrotrdote in
mehanske nanoobdelave. Trdoto in modul elasti~nosti so izra~unali iz meritev odvisnosti med obremenitvijo in globino vtisa
tristranske diamantne konice (po Berkovichu). Rezultati rentgenske difrakcije so potrdili prisotnost  in  faz. Skratka, poudariti
je treba, da ima material hierarhi~no strukturo, ki je posledica faze mletja. Pregledana zrna so sestavljena iz mnogo manj{ih
podzrn, katerih velikost se je zmanj{evala s podalj{evanjem ~asa mletja.
Klju~ne besede: mehansko legiranje, poroznost, nanovtiskovanje, zlitina Ti-26Nb
1 INTRODUCTION
Titanium-based alloys seem to be the preferred ma-
terials for orthopaedic applications. They are characte-
rized by high biocompatibility, great corrosion resis-
tance, light weight and good mechanical properties and
have a relatively low modulus in comparison to the other
materials used as implants.1 However, titanium-based
alloys used for implants have important disadvantages.
Firstly, popular alloying additives (Al, V) are dangerous
for the health and life of a patient.2 Secondly, the metals
used for implants have too high an elastic modulus in
comparison to the mechanical properties of the bone.
This difference between the mechanical properties in-
duces stress between the implant and the bone.3 In recent
years, due to intensive investigation, we have seen the
production and development of various Ti-based alloys,
among which there are Ti-Mo-based, Ti-Nb-based,
Ti-Zr-based and Ti-Ta-based alloys.2
This work focuses on niobium as the additive to
titanium because it has been recognized as an extremely
promising element. Nb is highly biocompatible, has a
low ionic in-vitro cytotoxicity, and most importantly –
there is no evidence of mutagenicity or carcinogenicity.4
The Ti-Nb-based alloys are attracting more researchers
to study them due to their low modulus, good biocom-
patibility and shape-memory effect. Medical applications
of SME (shape-memory effect) alloys will be especially
increased in the near future, so the development and
basic or extensive research of the Ti-based alloys are
necessary.1,2,4,5 A series of studies confirmed the presence
of the shape-memory effect and superelasticity in the
alloys with the composition of Ti-26Nb (a/%) and also in
approximate compositions of the elements.4–7 Most of
Materiali in tehnologije / Materials and technology 51 (2017) 5, 795–803 795
MATERIALI IN TEHNOLOGIJE/MATERIALS AND TECHNOLOGY (1967–2017) – 50 LET/50 YEARS
UDK 621.926:669.293:669.295:620.192.47 ISSN 1580-2949
Original scientific article/Izvirni znanstveni ~lanek MTAEC9, 51(5)795(2017)
these alloys are obtained with a conventional method
such as arc melting. One of the alternatives to the widely
used arc melting is powder metallurgy (PM). In the
scientific literature, there are evidences of successful
preparations of titanium-based materials with this pro-
duction method.8–10 This production method has a
number of significant advantages. It allows creating
alloys of materials with large differences between their
melting temperatures or densities. The most important
advantages of PM for the production of biomaterials are
its low cost and the possibility to produce highly porous
materials.4,11
The aim of the present paper is to determine the
possibility to produce a Ti–26Nb (a/%) alloy using the
powder-metallurgy method. Simultaneously, this re-
search is to develop a technology of producing Ti-based
alloys and estimate the influence of changing parameters
during the mechanical alloying. Such research is the
starting point for further development of nickel-free
alloys with the shape-memory effect based on titanium
and niobium. In this article, the influence of the milling
time on the microstructure and mechanical properties of
the produced alloy is investigated.
2 EXPERIMENTAL PART
Commercial elemental powders of Ti (Atlantic
Equipment Engineers (AEE), 99.7 %, <20 μm) and Nb
(Atlantic Equipment Engineers (AEE), 99.8 %, < 5 μm)
were used as the initial materials for the synthesis of the
alloy. Elemental metal powders with the nominal com-
position of Ti-26Nb (a/%) were milled for two different
milling times, 50 h and 70 h, in a planetary ball mill
Fritch PULVERISETTE 7 premium line. To prevent the
powder from oxidation as much as possible, the process
was carried out in an argon-protective atmosphere.
During the shorter milling time, the speed of 200 min–1
was applied, and for the longer milling time (with
additional 20 h), the speed of 400 min–1 was applied. The
powders were cold isostatically pressed under a 750 MPa
pressure without any substances that improve the
porosity (e.g., space holders). Then the material was
sintered at 1000 °C for 24 h and cooled in the furnace to
room temperature.
The crystalline structure and phase content of the
sintered materials were tested with X-ray diffraction.
The refinement of the X-ray diffraction pattern was
carried out using Rietveld’s whole X-ray profile fitting
technique with the DBWS 9807a program.12 The profile
function used to adjust the calculated diffractograms to
the observed ones was the pseudo-Voigt one.13,14 The
weight fraction of each component was determined
based on the optimized scale factors with the use of the
relation proposed by Hill and Howard.15,16 Based on the
Williamson–Hall method, the crystallite sizes (D) and
lattice strain <a/a> for the phases appearing in the
material at various milling times were determined. The
apparatus factors were eliminated using LaB6 (SRM
660a).
The morphology of the initial powders and the
microstructure of the sintered material were tested using
an optic microscope and scanning microscope JEOL
JSM 6480 with an accelerating voltage of 20 kV.
Additionally, a chemical analysis was performed using
an EDS detector manufactured by IXRF using the
standard calibration method.
An analysis of the microstructure was performed on
microscopic images with the use of ImageJ, which is the
software used for image processing and analysis. The
size and shape of the grains were determined using the
planimetric method. The study also included determining
the inhomogeneity of the size of the grains. A quantita-
tive analysis of the morphology of the material was
performed using the stereological parameters presented
below:
a) the size of the area of a section of grain a (μm2),
b) dimensionless lengthening factor (Equation (1):
f
h
wg1
= (1)
where: h – height and w – width of the smallest
rectangle described on the object,
c) dimensionless shape factor – the circularity:
f
F
Lg 2 2
4
=
π
(2)
where: F – area of the analyzed object; L – perimeter of
the analyzed object,
d) grain-size change-ability factor:


= x
a
(3)
where: x – grain-size standard deviation; a – grain
mean value,
the number of analyzed elements per area unit of the
image: NA [ ]1 2/mm .
The thermal behavior of the martensitic transforma-
tion was studied using a differential scanning calorimeter
(DSC) Mettler Toledo DSC-1. Transformation tempera-
tures were determined from the thermograms measured
at a heating rate of 10 deg/min and a thermal range of
–120 °C to 600 °C.
The surface morphology, hardness and elastic modu-
lus were tested using a Hysitron Triboindenter Ti950
with AFM QScope 250. The measurements were
calculated from the load-displacement data obtained with
the nanoindentation, using a three-sided, pyramidal,
diamond (Berkovich) indenter tip. An area of 40 μm ×
40 μm was analyzed at a scanning frequency of 0.25 Hz.
The microhardness measurement was carried out parallel
to the nanoindentation tests. The Vickers microhardness
measurement was taken at a load of 500 N and a loading
time of 10 s on a 401MVD microhardness tester.
G. DERCZ, I. MATU£A: EFFECT OF BALL MILLING ON THE PROPERTIES OF THE POROUS Ti–26Nb ...
796 Materiali in tehnologije / Materials and technology 51 (2017) 5, 795–803
MATERIALI IN TEHNOLOGIJE/MATERIALS AND TECHNOLOGY (1967–2017) – 50 LET/50 YEARS
3 RESULTS AND DISCUSSION
SEM micrographs of the initial metal powders are
presented in Figure 1. The titanium-powder morphology
is irregular, having sharp corners and rough surfaces.
The dispersion of the Ti particles is wide and ranges
from a few μm to even 50 μm. The niobium powder has a
much wider dispersion of the particles in comparison to
the Ti powder.
Some of the particles have the size below 10 μm, and
the shape of these particles is irregular. Simultaneously,
there are also very large particles, the sizes of which are
greater even than 50 μm and these particles have an
irregular, polyhedral shape. As a result of the work-hard-
ening effect and high density of defects such as dislo-
cations and vacancies created during long ball-milling
times, the fracture and welding of powders reach an
equilibrium state. This state leads to the formation of
rather equiaxed and small particles.16,17 However, due to
the higher bonding strength of the finer particles, their
ability to undergo further plastic deformation is de-
creased and a higher force is required to fracture the
small particles during the ball-milling process.18 It is
worth mentioning that the reduction in the particle size at
a given milling time can also be influenced by the
milling technique.
The morphologies of the Ti-26Nb (a/%) powders at
different milling stages are shown in Figure 2. The
powder after the shortest milling time shows globular
particles. In the picture with the higher magnification
(Figure 2b), some cracks were observed, which is the
beginning of the cyclic process of material milling
during high-energy ball milling. In contrast, in the
samples after 70 h of milling, the particles became larger
and the shape of the particles became less globular.
Moreover, the particles of this sample are cold welded
and the powder particles, in particular, are composed of
the welded layers of the material. This behavior of the
material during the high-energy milling is consistent
with the conduct and following steps of mechanical
alloying.19 As shown by the other studies, the mechani-
cal-alloying method leads to a decreased particle size,
the -Ti phase being stabilized by Nb, and an increased
surface area between Ti and Nb. This promotes good
homogeneity of the solid solution.20
Figure 3 shows a summary of diffraction patterns of
the material after the mechanical synthesis, depending on
the milling time of the initial powders. The X-ray
qualitative analysis showed that only  and  phases are
present in the material after milling for 50 h and 70 h.
Change of the profile lines of individual diffraction
patterns are clearly visible.
The broadening of the particle-size distribution at
longer ball-milling times is a typical behavior of high-
energy ball-milling process.17,19,21,22 For the  and 
phases, it was observed that a decrease in the crystallite
size took place simultaneously with an increase in the
milling time (Table 1). In addition, in the case of the 
phase there was a tendency towards a decrease in the
lattice strain <a/a> (Table 1). The estimated size of the
crystallites for the powder after milling for 70 h is 9(1)
nm and 16(2) nm for the  and  phases, respectively.
The observed <a/a> lattice strain values indicate
growth in the  phase (3.43E-03 %) and in the  phase
G. DERCZ, I. MATU£A: EFFECT OF BALL MILLING ON THE PROPERTIES OF THE POROUS Ti–26Nb ...
Materiali in tehnologije / Materials and technology 51 (2017) 5, 795–803 797
MATERIALI IN TEHNOLOGIJE/MATERIALS AND TECHNOLOGY (1967–2017) – 50 LET/50 YEARS
Figure 3: X-ray diffraction patterns of the powder after 50 h and 70 h
of the milling process
Figure 1: SEM micrographs of the initial Ti and Nb powders at
different magnifications
Figure 2: SEM micrographs of particles after different milling times:
a), b) 50 h of milling and c), d) 70 h
these alloys are obtained with a conventional method
such as arc melting. One of the alternatives to the widely
used arc melting is powder metallurgy (PM). In the
scientific literature, there are evidences of successful
preparations of titanium-based materials with this pro-
duction method.8–10 This production method has a
number of significant advantages. It allows creating
alloys of materials with large differences between their
melting temperatures or densities. The most important
advantages of PM for the production of biomaterials are
its low cost and the possibility to produce highly porous
materials.4,11
The aim of the present paper is to determine the
possibility to produce a Ti–26Nb (a/%) alloy using the
powder-metallurgy method. Simultaneously, this re-
search is to develop a technology of producing Ti-based
alloys and estimate the influence of changing parameters
during the mechanical alloying. Such research is the
starting point for further development of nickel-free
alloys with the shape-memory effect based on titanium
and niobium. In this article, the influence of the milling
time on the microstructure and mechanical properties of
the produced alloy is investigated.
2 EXPERIMENTAL PART
Commercial elemental powders of Ti (Atlantic
Equipment Engineers (AEE), 99.7 %, <20 μm) and Nb
(Atlantic Equipment Engineers (AEE), 99.8 %, < 5 μm)
were used as the initial materials for the synthesis of the
alloy. Elemental metal powders with the nominal com-
position of Ti-26Nb (a/%) were milled for two different
milling times, 50 h and 70 h, in a planetary ball mill
Fritch PULVERISETTE 7 premium line. To prevent the
powder from oxidation as much as possible, the process
was carried out in an argon-protective atmosphere.
During the shorter milling time, the speed of 200 min–1
was applied, and for the longer milling time (with
additional 20 h), the speed of 400 min–1 was applied. The
powders were cold isostatically pressed under a 750 MPa
pressure without any substances that improve the
porosity (e.g., space holders). Then the material was
sintered at 1000 °C for 24 h and cooled in the furnace to
room temperature.
The crystalline structure and phase content of the
sintered materials were tested with X-ray diffraction.
The refinement of the X-ray diffraction pattern was
carried out using Rietveld’s whole X-ray profile fitting
technique with the DBWS 9807a program.12 The profile
function used to adjust the calculated diffractograms to
the observed ones was the pseudo-Voigt one.13,14 The
weight fraction of each component was determined
based on the optimized scale factors with the use of the
relation proposed by Hill and Howard.15,16 Based on the
Williamson–Hall method, the crystallite sizes (D) and
lattice strain <a/a> for the phases appearing in the
material at various milling times were determined. The
apparatus factors were eliminated using LaB6 (SRM
660a).
The morphology of the initial powders and the
microstructure of the sintered material were tested using
an optic microscope and scanning microscope JEOL
JSM 6480 with an accelerating voltage of 20 kV.
Additionally, a chemical analysis was performed using
an EDS detector manufactured by IXRF using the
standard calibration method.
An analysis of the microstructure was performed on
microscopic images with the use of ImageJ, which is the
software used for image processing and analysis. The
size and shape of the grains were determined using the
planimetric method. The study also included determining
the inhomogeneity of the size of the grains. A quantita-
tive analysis of the morphology of the material was
performed using the stereological parameters presented
below:
a) the size of the area of a section of grain a (μm2),
b) dimensionless lengthening factor (Equation (1):
f
h
wg1
= (1)
where: h – height and w – width of the smallest
rectangle described on the object,
c) dimensionless shape factor – the circularity:
f
F
Lg 2 2
4
=
π
(2)
where: F – area of the analyzed object; L – perimeter of
the analyzed object,
d) grain-size change-ability factor:


= x
a
(3)
where: x – grain-size standard deviation; a – grain
mean value,
the number of analyzed elements per area unit of the
image: NA [ ]1 2/mm .
The thermal behavior of the martensitic transforma-
tion was studied using a differential scanning calorimeter
(DSC) Mettler Toledo DSC-1. Transformation tempera-
tures were determined from the thermograms measured
at a heating rate of 10 deg/min and a thermal range of
–120 °C to 600 °C.
The surface morphology, hardness and elastic modu-
lus were tested using a Hysitron Triboindenter Ti950
with AFM QScope 250. The measurements were
calculated from the load-displacement data obtained with
the nanoindentation, using a three-sided, pyramidal,
diamond (Berkovich) indenter tip. An area of 40 μm ×
40 μm was analyzed at a scanning frequency of 0.25 Hz.
The microhardness measurement was carried out parallel
to the nanoindentation tests. The Vickers microhardness
measurement was taken at a load of 500 N and a loading
time of 10 s on a 401MVD microhardness tester.
G. DERCZ, I. MATU£A: EFFECT OF BALL MILLING ON THE PROPERTIES OF THE POROUS Ti–26Nb ...
796 Materiali in tehnologije / Materials and technology 51 (2017) 5, 795–803
MATERIALI IN TEHNOLOGIJE/MATERIALS AND TECHNOLOGY (1967–2017) – 50 LET/50 YEARS
3 RESULTS AND DISCUSSION
SEM micrographs of the initial metal powders are
presented in Figure 1. The titanium-powder morphology
is irregular, having sharp corners and rough surfaces.
The dispersion of the Ti particles is wide and ranges
from a few μm to even 50 μm. The niobium powder has a
much wider dispersion of the particles in comparison to
the Ti powder.
Some of the particles have the size below 10 μm, and
the shape of these particles is irregular. Simultaneously,
there are also very large particles, the sizes of which are
greater even than 50 μm and these particles have an
irregular, polyhedral shape. As a result of the work-hard-
ening effect and high density of defects such as dislo-
cations and vacancies created during long ball-milling
times, the fracture and welding of powders reach an
equilibrium state. This state leads to the formation of
rather equiaxed and small particles.16,17 However, due to
the higher bonding strength of the finer particles, their
ability to undergo further plastic deformation is de-
creased and a higher force is required to fracture the
small particles during the ball-milling process.18 It is
worth mentioning that the reduction in the particle size at
a given milling time can also be influenced by the
milling technique.
The morphologies of the Ti-26Nb (a/%) powders at
different milling stages are shown in Figure 2. The
powder after the shortest milling time shows globular
particles. In the picture with the higher magnification
(Figure 2b), some cracks were observed, which is the
beginning of the cyclic process of material milling
during high-energy ball milling. In contrast, in the
samples after 70 h of milling, the particles became larger
and the shape of the particles became less globular.
Moreover, the particles of this sample are cold welded
and the powder particles, in particular, are composed of
the welded layers of the material. This behavior of the
material during the high-energy milling is consistent
with the conduct and following steps of mechanical
alloying.19 As shown by the other studies, the mechani-
cal-alloying method leads to a decreased particle size,
the -Ti phase being stabilized by Nb, and an increased
surface area between Ti and Nb. This promotes good
homogeneity of the solid solution.20
Figure 3 shows a summary of diffraction patterns of
the material after the mechanical synthesis, depending on
the milling time of the initial powders. The X-ray
qualitative analysis showed that only  and  phases are
present in the material after milling for 50 h and 70 h.
Change of the profile lines of individual diffraction
patterns are clearly visible.
The broadening of the particle-size distribution at
longer ball-milling times is a typical behavior of high-
energy ball-milling process.17,19,21,22 For the  and 
phases, it was observed that a decrease in the crystallite
size took place simultaneously with an increase in the
milling time (Table 1). In addition, in the case of the 
phase there was a tendency towards a decrease in the
lattice strain <a/a> (Table 1). The estimated size of the
crystallites for the powder after milling for 70 h is 9(1)
nm and 16(2) nm for the  and  phases, respectively.
The observed <a/a> lattice strain values indicate
growth in the  phase (3.43E-03 %) and in the  phase
G. DERCZ, I. MATU£A: EFFECT OF BALL MILLING ON THE PROPERTIES OF THE POROUS Ti–26Nb ...
Materiali in tehnologije / Materials and technology 51 (2017) 5, 795–803 797
MATERIALI IN TEHNOLOGIJE/MATERIALS AND TECHNOLOGY (1967–2017) – 50 LET/50 YEARS
Figure 3: X-ray diffraction patterns of the powder after 50 h and 70 h
of the milling process
Figure 1: SEM micrographs of the initial Ti and Nb powders at
different magnifications
Figure 2: SEM micrographs of particles after different milling times:
a), b) 50 h of milling and c), d) 70 h
(2.31E-03 %). The analysis of the diffraction patters
obtained using the Rietveld method showed that that
high-energy milling has an influence on the lattice
parameters.
Table 1: Changes of the average crystallite sizes (D) and lattice
distortions (<a/a>) of the  and  phases of the milled powders
Phase Parameters
Milling time (h)
50 70

D (nm) 22(2) 16(1)
<a/a> (%) 2.76E-03 3.43E-03

D (nm) 15(2) 9(1)
<a/a> (%) 2.04E-03 2.31E-03
Table 2 presents lattice parameters determined for
individual phases and the corresponding ICDD data
sheets. In the case of the  phase, a slight deviation of
the a0 lattice parameters from the catalogue data was
observed. A similar type of cell contraction was
discovered for the a0 parameter of the  phase and for
the c0 parameter, but these changes are smaller, resulting
from the shift systems for the phases of the hexagonal
system. It should be stressed that for the sample after 70
h of milling, the deviation of both lattice parameters can
be noted. It is probably the effect of the alternate cold
welding and deposition of the particles during the
mechanical synthesis.
Table 2: Lattice parameters and contents of  and  phases of the
milled powders
Phase
Lattice parameters (nm)
ICDD*

a0 0.2970
c0 0.4720
 a0 0.3307
* International Centre for Diffraction Data® – a scientific organization
dedicated to collecting, editing, publishing and distributing powder-
diffraction data for the identification of materials
The quantitative phase analysis of the material after
each stage show a successive progressive synthesis of the
as-prepared powders in relation to the  phase (77.5(11))
% mass fraction and 82.6(12) % mass fraction, for 50 h
and 70 h milling times, respectively).
The X-ray qualitative analysis showed that only 
and  phases are present in the samples annealed at
1000 °C for 24 h from the powder previously milled for
50 h and 70 h, as shown in Figure 4. The XRD shows
that there were no obvious diffraction peaks of elemental
Nb remaining in the sintered Ti–26Nb (a/%). The
diffraction peaks attributable to the  and  phases from
the pattern confirm a duplex microstructure. M. Tahara et
al.23 confirmed only the  phase for Ti-26Nb (a/%)
produced by arc melting. The predominant phase in all
the samples was the body-centered-cubic (bcc)  phase.
This indicates that Nb was diffused into Ti, thereby lead-
ing to the formation of  +  phases after the sintering.
This is significant because two-phase Ti alloys, with the
major  phase and the minor  phase, possess good com-
prehensive properties including high yield strength, ex-
cellent corrosion resistance and good fracture toughness.
The presence of unalloyed Ti particles in materials
may arise from different facts. According to S. N.
Patankar and F. H. Froes24, it could be correlated with the
brittleness of the Nb particles, as compared to the Ti
particles, since these brittle particles adhere to the sur-
faces of ductile particles. Another reason could be the
limited solubility of Nb in -Ti. It is known that iso-
morphous -stabilizer elements, such as Nb, exhibit
slower diffusivities in Ti than the eutectoid elements of
Fe, Co and Ni.25
The structural analysis of the diffraction patters
obtained using the Rietveld method showed that the
lattice parameters for all the tested samples become
reduced. In comparison to the ICDD data, it was found
that high-energy milling and annealing have the smallest,
but significant effect on the reduction of the size of a unit
cell of the  and  phases. The quantitative phase anal-
ysis (Table 3) of the material after annealing showed a
significant synthesis of the as-prepared powders in rela-
tion to the  phase 96.6(12) w/% and 98.8(12) w/% for
samples previously milled for 50 h and 70 h, respecti-
vely).
Table 3: Lattice parameters and contents of  and  phases after
annealing (A) the samples at 1000 °C for 24 h from the powder
previously milled for 50 h and 70 h
Phase
Lattice parameters (nm)
Phase
Contents (w/%)
ICDD*
Rietveld
50 h +
A
70 h +
A
50 h +
A
70 h +
A

a0 0.2970 0.2898(2)
0.2891
(2) 
3.4
(6)
1.2
(4)
c0 0.4720 0.4715(4)
0.4710
(4)

96.6
(12)
98.8
(12)
 a0 0.3307 0.3287(2)
0.3286
(2)
* International Centre for Diffraction Data®– a scientific organization
dedicated to collecting, editing, publishing and distributing
powder-diffraction data for the identification of materials
Optical and SEM micrographs of the studied samples
are presented in Figures 5 and 6. They clearly show the
G. DERCZ, I. MATU£A: EFFECT OF BALL MILLING ON THE PROPERTIES OF THE POROUS Ti–26Nb ...
798 Materiali in tehnologije / Materials and technology 51 (2017) 5, 795–803
MATERIALI IN TEHNOLOGIJE/MATERIALS AND TECHNOLOGY (1967–2017) – 50 LET/50 YEARS
Figure 4: X-ray diffraction patterns for the samples annealed (+A) at
1000 °C for 24 h from the powder previously milled for 50 h and 70 h
changes in the morphology of the material after sintering
with respect to various milling times. The common
feature of the samples is a hierarchical structure. After
the milling, large-sized particles caused the formation of
a porous material. The pores were large and intercon-
nected (Figures 5a and 5c) in spite of the applied
pressure of 750 MPa during the isostatic pressing.
In the case of the sample milled for 50 h (Figure 5b),
an anomalous grain growth was observed. SEM micro-
graphs (Figure 6) revealed further differences between
the obtained microstructures. The powders ball-milled
for a long time cannot move and deform effectively
during the compaction. This may be primarily due to the
effects of the decrease in the particle size, solid-solution
strengthening and work hardening of the particles during
the ball-milling process.26,27 Amongst these factors, the
work-hardening effect appears to be the main deter-
minant in governing the plastic behavior of the powder
particles during the compaction. In the ball-milling
process, work hardening is caused by the interaction of
dislocations with each other through the continuous
mechanical impact of the balls on the powder particles.16
A quantitative analysis of the microstructure was
conducted including the determination of the cross-sec-
tional area of the grains and the subgrains, as well as the
average circularity of the subgrains of the microsections
based on microscopic images, the results of which are
presented in Tables 4 to 6. Significant changes between
the samples in the average a of the grain cross-sectional
area can also be observed. For the samples milled for
50 h and 70 h, the average section area of grains was
9.09 μm2 and 3.55 μm2, respectively (Table 4). A change
in the time and rotations per minute during milling
causes a difference in the parameter.
Table 4: Results for the cross-sectional area of the grains for the
samples milled for different times
Parameter Sample Minimumvalue
Maximum
value
Average
value Std. dev.
a (μm2)
50 h 0.51 267.57 9.09 12.68
70 h 0.51 32.35 3.55 3.45
Table 5: Results for the circularity of the grains for the samples milled
for different times
Parameter Sample Minimumvalue
Maximum
value
Average
value Std. dev.
fg2
50 h 0.10 1.00 0.60 0.15
70 h 0.11 0.96 0.59 0.15
Table 6: Results for the dimensionless lengthening factor of the grains
for the samples milled for different times
Parameter Sample Minimumvalue
Maximum
value
Average
value Std. dev.
fg1
50 h 0.32 3.31 1.06 0.37
70 h 0.28 3.30 1.06 0.37
Histograms of the size of the area of the section of
the grains in the samples milled for 50 h and 70 h are
presented in Figures 7 and 8, respectively. For the sam-
ple milled for 70 h, a larger fraction of the grains with
the section area in the range of 0.5–35 μm can be ob-
served. Milling for 70 h caused a significant fragmen-
tation of the grains with the section area below 30 μm2.
G. DERCZ, I. MATU£A: EFFECT OF BALL MILLING ON THE PROPERTIES OF THE POROUS Ti–26Nb ...
Materiali in tehnologije / Materials and technology 51 (2017) 5, 795–803 799
MATERIALI IN TEHNOLOGIJE/MATERIALS AND TECHNOLOGY (1967–2017) – 50 LET/50 YEARS
Figure 6: a), b) SEM micrographs of Ti-26Nb (a/%) after different
milling times – 50 h and c), d) 70 h and annealing
Figure 5: a), b) Optical micrographs of Ti-26Nb (a/%) at different
magnifications, produced in different milling times – 50 h and c), d)
70 h and annealing
Figure 7: Histogram of the grain average area for the sample milled
for 50 h and annealed
(2.31E-03 %). The analysis of the diffraction patters
obtained using the Rietveld method showed that that
high-energy milling has an influence on the lattice
parameters.
Table 1: Changes of the average crystallite sizes (D) and lattice
distortions (<a/a>) of the  and  phases of the milled powders
Phase Parameters
Milling time (h)
50 70

D (nm) 22(2) 16(1)
<a/a> (%) 2.76E-03 3.43E-03

D (nm) 15(2) 9(1)
<a/a> (%) 2.04E-03 2.31E-03
Table 2 presents lattice parameters determined for
individual phases and the corresponding ICDD data
sheets. In the case of the  phase, a slight deviation of
the a0 lattice parameters from the catalogue data was
observed. A similar type of cell contraction was
discovered for the a0 parameter of the  phase and for
the c0 parameter, but these changes are smaller, resulting
from the shift systems for the phases of the hexagonal
system. It should be stressed that for the sample after 70
h of milling, the deviation of both lattice parameters can
be noted. It is probably the effect of the alternate cold
welding and deposition of the particles during the
mechanical synthesis.
Table 2: Lattice parameters and contents of  and  phases of the
milled powders
Phase
Lattice parameters (nm)
ICDD*

a0 0.2970
c0 0.4720
 a0 0.3307
* International Centre for Diffraction Data® – a scientific organization
dedicated to collecting, editing, publishing and distributing powder-
diffraction data for the identification of materials
The quantitative phase analysis of the material after
each stage show a successive progressive synthesis of the
as-prepared powders in relation to the  phase (77.5(11))
% mass fraction and 82.6(12) % mass fraction, for 50 h
and 70 h milling times, respectively).
The X-ray qualitative analysis showed that only 
and  phases are present in the samples annealed at
1000 °C for 24 h from the powder previously milled for
50 h and 70 h, as shown in Figure 4. The XRD shows
that there were no obvious diffraction peaks of elemental
Nb remaining in the sintered Ti–26Nb (a/%). The
diffraction peaks attributable to the  and  phases from
the pattern confirm a duplex microstructure. M. Tahara et
al.23 confirmed only the  phase for Ti-26Nb (a/%)
produced by arc melting. The predominant phase in all
the samples was the body-centered-cubic (bcc)  phase.
This indicates that Nb was diffused into Ti, thereby lead-
ing to the formation of  +  phases after the sintering.
This is significant because two-phase Ti alloys, with the
major  phase and the minor  phase, possess good com-
prehensive properties including high yield strength, ex-
cellent corrosion resistance and good fracture toughness.
The presence of unalloyed Ti particles in materials
may arise from different facts. According to S. N.
Patankar and F. H. Froes24, it could be correlated with the
brittleness of the Nb particles, as compared to the Ti
particles, since these brittle particles adhere to the sur-
faces of ductile particles. Another reason could be the
limited solubility of Nb in -Ti. It is known that iso-
morphous -stabilizer elements, such as Nb, exhibit
slower diffusivities in Ti than the eutectoid elements of
Fe, Co and Ni.25
The structural analysis of the diffraction patters
obtained using the Rietveld method showed that the
lattice parameters for all the tested samples become
reduced. In comparison to the ICDD data, it was found
that high-energy milling and annealing have the smallest,
but significant effect on the reduction of the size of a unit
cell of the  and  phases. The quantitative phase anal-
ysis (Table 3) of the material after annealing showed a
significant synthesis of the as-prepared powders in rela-
tion to the  phase 96.6(12) w/% and 98.8(12) w/% for
samples previously milled for 50 h and 70 h, respecti-
vely).
Table 3: Lattice parameters and contents of  and  phases after
annealing (A) the samples at 1000 °C for 24 h from the powder
previously milled for 50 h and 70 h
Phase
Lattice parameters (nm)
Phase
Contents (w/%)
ICDD*
Rietveld
50 h +
A
70 h +
A
50 h +
A
70 h +
A

a0 0.2970 0.2898(2)
0.2891
(2) 
3.4
(6)
1.2
(4)
c0 0.4720 0.4715(4)
0.4710
(4)

96.6
(12)
98.8
(12)
 a0 0.3307 0.3287(2)
0.3286
(2)
* International Centre for Diffraction Data®– a scientific organization
dedicated to collecting, editing, publishing and distributing
powder-diffraction data for the identification of materials
Optical and SEM micrographs of the studied samples
are presented in Figures 5 and 6. They clearly show the
G. DERCZ, I. MATU£A: EFFECT OF BALL MILLING ON THE PROPERTIES OF THE POROUS Ti–26Nb ...
798 Materiali in tehnologije / Materials and technology 51 (2017) 5, 795–803
MATERIALI IN TEHNOLOGIJE/MATERIALS AND TECHNOLOGY (1967–2017) – 50 LET/50 YEARS
Figure 4: X-ray diffraction patterns for the samples annealed (+A) at
1000 °C for 24 h from the powder previously milled for 50 h and 70 h
changes in the morphology of the material after sintering
with respect to various milling times. The common
feature of the samples is a hierarchical structure. After
the milling, large-sized particles caused the formation of
a porous material. The pores were large and intercon-
nected (Figures 5a and 5c) in spite of the applied
pressure of 750 MPa during the isostatic pressing.
In the case of the sample milled for 50 h (Figure 5b),
an anomalous grain growth was observed. SEM micro-
graphs (Figure 6) revealed further differences between
the obtained microstructures. The powders ball-milled
for a long time cannot move and deform effectively
during the compaction. This may be primarily due to the
effects of the decrease in the particle size, solid-solution
strengthening and work hardening of the particles during
the ball-milling process.26,27 Amongst these factors, the
work-hardening effect appears to be the main deter-
minant in governing the plastic behavior of the powder
particles during the compaction. In the ball-milling
process, work hardening is caused by the interaction of
dislocations with each other through the continuous
mechanical impact of the balls on the powder particles.16
A quantitative analysis of the microstructure was
conducted including the determination of the cross-sec-
tional area of the grains and the subgrains, as well as the
average circularity of the subgrains of the microsections
based on microscopic images, the results of which are
presented in Tables 4 to 6. Significant changes between
the samples in the average a of the grain cross-sectional
area can also be observed. For the samples milled for
50 h and 70 h, the average section area of grains was
9.09 μm2 and 3.55 μm2, respectively (Table 4). A change
in the time and rotations per minute during milling
causes a difference in the parameter.
Table 4: Results for the cross-sectional area of the grains for the
samples milled for different times
Parameter Sample Minimumvalue
Maximum
value
Average
value Std. dev.
a (μm2)
50 h 0.51 267.57 9.09 12.68
70 h 0.51 32.35 3.55 3.45
Table 5: Results for the circularity of the grains for the samples milled
for different times
Parameter Sample Minimumvalue
Maximum
value
Average
value Std. dev.
fg2
50 h 0.10 1.00 0.60 0.15
70 h 0.11 0.96 0.59 0.15
Table 6: Results for the dimensionless lengthening factor of the grains
for the samples milled for different times
Parameter Sample Minimumvalue
Maximum
value
Average
value Std. dev.
fg1
50 h 0.32 3.31 1.06 0.37
70 h 0.28 3.30 1.06 0.37
Histograms of the size of the area of the section of
the grains in the samples milled for 50 h and 70 h are
presented in Figures 7 and 8, respectively. For the sam-
ple milled for 70 h, a larger fraction of the grains with
the section area in the range of 0.5–35 μm can be ob-
served. Milling for 70 h caused a significant fragmen-
tation of the grains with the section area below 30 μm2.
G. DERCZ, I. MATU£A: EFFECT OF BALL MILLING ON THE PROPERTIES OF THE POROUS Ti–26Nb ...
Materiali in tehnologije / Materials and technology 51 (2017) 5, 795–803 799
MATERIALI IN TEHNOLOGIJE/MATERIALS AND TECHNOLOGY (1967–2017) – 50 LET/50 YEARS
Figure 6: a), b) SEM micrographs of Ti-26Nb (a/%) after different
milling times – 50 h and c), d) 70 h and annealing
Figure 5: a), b) Optical micrographs of Ti-26Nb (a/%) at different
magnifications, produced in different milling times – 50 h and c), d)
70 h and annealing
Figure 7: Histogram of the grain average area for the sample milled
for 50 h and annealed
The circularity of grains was 0.60 for 50 h of milling
and 0.59 for 70 h of milling, which corresponds to the
dimensionless lengthening factor for both samples
(Tables 5 to 6). The coefficient of the grain variability
was also determined. For all the samples milled for 50 h,
it was 1.4, and for the sample milled for 70 h, it was 0.9.
It means that for the sample milled for a shorter time, the
size of the cross-sectional area is much more diversified
in comparison to the sample milled for 70 h. The longer
milling time (70 h) caused a greater fragmentation of the
material. The difference in the average section area of
the grains was significant, especially when compared to
the size of the powder after milling. No significant diffe-
rences in the size of the particles depending on the
milling time were noticed. The number of the elements
that were tested per area unit of the image was also
determined; for the samples annealed and previously
milled for 50 h and 70 h, the number of elements per
area was estimated to be 82379 [ ]1 2/mm and 199951
[ ]1 2/mm respectively.
Distribution maps of the elements for the sintered
samples revealed single regions rich in titanium in the
case of the material milled for 50 h (Figure 9) Also, a
slightly higher concentration of Ti was observed inside
the grains. In contrast, milling for 70 h allowed us to
obtain a material with an even distribution of the ele-
ments on its surface (Figure 10). In the case of this alloy,
the increase in the milling time allows a more
homogeneous structure.
Figures 11 and 12 show the effect of an analysis with
the Hysitron Tribointender Ti950 with AFM QScope
250, which allowed us to compare the surface topogra-
phies of the produced samples. For the material milled
for 50 h and sintered, a higher diversification of the
sample surface was observed. On the sample, initially
synthetized for 70 h of milling, we can observe smaller
and more regular grains in comparison to the sample ex-
posed to the shorter milling time, which was confirmed
with a stereological analysis.
G. DERCZ, I. MATU£A: EFFECT OF BALL MILLING ON THE PROPERTIES OF THE POROUS Ti–26Nb ...
800 Materiali in tehnologije / Materials and technology 51 (2017) 5, 795–803
MATERIALI IN TEHNOLOGIJE/MATERIALS AND TECHNOLOGY (1967–2017) – 50 LET/50 YEARS
Figure 8: Histograms of the grain average area for the sample milled
for 70 h and annealed. The upper histogram presents the distribution
of the average area of the grains smaller than 36 μm2 for this sample
Figure 10: Distribution maps of the elements for the sample milled for 70 h and annealed
Figure 9: Distribution maps of the elements for the sample milled for 50 h and annealed
Another important aspect of the produced material is
its properties. DSC results showed (Figure 13) the
presence of an endothermic peak on the heating curve for
both samples. In the case of the sample milled for 70 h,
the peak was slightly shifted toward higher temperatures.
The peaks occurred at about 490 °C and 520 °C for the
samples milled for 50 h and 70 h, respectively. The pre-
sence of the peaks can indicate the presence of a partial
phase transformation   . Moreover, the milling time
has an impact on the temperature of transformation.
However, no significant changes were noticed on the
cooling curves.
The nanoindentation method allowed the determina-
tion of the reduced Young’s modulus and hardness of the
material, and the results of this analysis of the material
after different milling times and sintering are presented
in Table 7.
Table 7: Results of the nanoindentation analysis – hardness and
modulus of both samples
Samples Contact depth(nm)
Hardness
(GPa) Modulus (GPa)
50 h 89(14) 0.99(0.31) 48(19)
70 h 42(12) 3.86(1.71) 95(32)
The measurement is based on the curves of the load
and the penetration depth of the indenter (Figures 14 and
15). After 50 h of milling, the material obtained much
lower values of hardness and modulus in comparison to
the material milled for 70 h. For the samples milled for
50 h and 70 h, the hardness was 0.99±0.31 GPa and
3.86±1.71GPa, respectively. A similar situation applied
to the values of the modulus. For the material milled for
50 h, it was 48±19 GPa, and for the material milled for
70 h, it was 95±32 GPa. Such low modulus values,
especially in the case of the sample with the shorter
milling time, are very promising as they are close to the
value of the modulus of a human bone.
G. DERCZ, I. MATU£A: EFFECT OF BALL MILLING ON THE PROPERTIES OF THE POROUS Ti–26Nb ...
Materiali in tehnologije / Materials and technology 51 (2017) 5, 795–803 801
MATERIALI IN TEHNOLOGIJE/MATERIALS AND TECHNOLOGY (1967–2017) – 50 LET/50 YEARS
Figure 13: DSC analysis of the samples milled for 50 h and 70 h and
also annealed
Figure 12: Surface morphology of the area of 40 μm × 40 μm for the
sample milled for 70 h and annealed
Figure 11: Surface morphology of the area of 40 μm × 40 μm for the
sample milled for 50 h and annealed
The circularity of grains was 0.60 for 50 h of milling
and 0.59 for 70 h of milling, which corresponds to the
dimensionless lengthening factor for both samples
(Tables 5 to 6). The coefficient of the grain variability
was also determined. For all the samples milled for 50 h,
it was 1.4, and for the sample milled for 70 h, it was 0.9.
It means that for the sample milled for a shorter time, the
size of the cross-sectional area is much more diversified
in comparison to the sample milled for 70 h. The longer
milling time (70 h) caused a greater fragmentation of the
material. The difference in the average section area of
the grains was significant, especially when compared to
the size of the powder after milling. No significant diffe-
rences in the size of the particles depending on the
milling time were noticed. The number of the elements
that were tested per area unit of the image was also
determined; for the samples annealed and previously
milled for 50 h and 70 h, the number of elements per
area was estimated to be 82379 [ ]1 2/mm and 199951
[ ]1 2/mm respectively.
Distribution maps of the elements for the sintered
samples revealed single regions rich in titanium in the
case of the material milled for 50 h (Figure 9) Also, a
slightly higher concentration of Ti was observed inside
the grains. In contrast, milling for 70 h allowed us to
obtain a material with an even distribution of the ele-
ments on its surface (Figure 10). In the case of this alloy,
the increase in the milling time allows a more
homogeneous structure.
Figures 11 and 12 show the effect of an analysis with
the Hysitron Tribointender Ti950 with AFM QScope
250, which allowed us to compare the surface topogra-
phies of the produced samples. For the material milled
for 50 h and sintered, a higher diversification of the
sample surface was observed. On the sample, initially
synthetized for 70 h of milling, we can observe smaller
and more regular grains in comparison to the sample ex-
posed to the shorter milling time, which was confirmed
with a stereological analysis.
G. DERCZ, I. MATU£A: EFFECT OF BALL MILLING ON THE PROPERTIES OF THE POROUS Ti–26Nb ...
800 Materiali in tehnologije / Materials and technology 51 (2017) 5, 795–803
MATERIALI IN TEHNOLOGIJE/MATERIALS AND TECHNOLOGY (1967–2017) – 50 LET/50 YEARS
Figure 8: Histograms of the grain average area for the sample milled
for 70 h and annealed. The upper histogram presents the distribution
of the average area of the grains smaller than 36 μm2 for this sample
Figure 10: Distribution maps of the elements for the sample milled for 70 h and annealed
Figure 9: Distribution maps of the elements for the sample milled for 50 h and annealed
Another important aspect of the produced material is
its properties. DSC results showed (Figure 13) the
presence of an endothermic peak on the heating curve for
both samples. In the case of the sample milled for 70 h,
the peak was slightly shifted toward higher temperatures.
The peaks occurred at about 490 °C and 520 °C for the
samples milled for 50 h and 70 h, respectively. The pre-
sence of the peaks can indicate the presence of a partial
phase transformation   . Moreover, the milling time
has an impact on the temperature of transformation.
However, no significant changes were noticed on the
cooling curves.
The nanoindentation method allowed the determina-
tion of the reduced Young’s modulus and hardness of the
material, and the results of this analysis of the material
after different milling times and sintering are presented
in Table 7.
Table 7: Results of the nanoindentation analysis – hardness and
modulus of both samples
Samples Contact depth(nm)
Hardness
(GPa) Modulus (GPa)
50 h 89(14) 0.99(0.31) 48(19)
70 h 42(12) 3.86(1.71) 95(32)
The measurement is based on the curves of the load
and the penetration depth of the indenter (Figures 14 and
15). After 50 h of milling, the material obtained much
lower values of hardness and modulus in comparison to
the material milled for 70 h. For the samples milled for
50 h and 70 h, the hardness was 0.99±0.31 GPa and
3.86±1.71GPa, respectively. A similar situation applied
to the values of the modulus. For the material milled for
50 h, it was 48±19 GPa, and for the material milled for
70 h, it was 95±32 GPa. Such low modulus values,
especially in the case of the sample with the shorter
milling time, are very promising as they are close to the
value of the modulus of a human bone.
G. DERCZ, I. MATU£A: EFFECT OF BALL MILLING ON THE PROPERTIES OF THE POROUS Ti–26Nb ...
Materiali in tehnologije / Materials and technology 51 (2017) 5, 795–803 801
MATERIALI IN TEHNOLOGIJE/MATERIALS AND TECHNOLOGY (1967–2017) – 50 LET/50 YEARS
Figure 13: DSC analysis of the samples milled for 50 h and 70 h and
also annealed
Figure 12: Surface morphology of the area of 40 μm × 40 μm for the
sample milled for 70 h and annealed
Figure 11: Surface morphology of the area of 40 μm × 40 μm for the
sample milled for 50 h and annealed
The nanosized indentation measurements of the
mechanical properties are from the selected specific
areas, which neutralize the effects of the pores or defects
on the results. A large measurement uncertainty is even
more surprising. Assuming that the material is mechani-
cally widely heterogeneous, it seems necessary to apply
heat treatment to the material. The microhardness
measurement of the sample milled for 50 h revealed a
higher value of the microhardness and it is 288.6(±31.0)
HV0.5, which is around 2.82 GPa. For the sample milled
for 70 h, the revealed value is 256.2(±14.6) HV0.5,
which is approximately 2.51 GPa. However, we observed
a relatively high uncertainty of the measurement, which
made it difficult to clearly evaluate the differences in the
microhardness measurement.
4 CONCLUSIONS
Based on the investigation into the microstructure
and mechanical properties of the Ti-26Nb (a/%) alloy,
the main conclusions can be drawn:
• It was proved that the microstructure, the degree of
porosity and mechanical properties can be adjusted
by means of mechanical alloying.
• A longer milling time (70 h) resulted in an increased
degree of structure fragmentation. Young’s modulus
increased twice as well.
• The porous samples sintered from the powders milled
for 50 hours have a very low elastic modulus of
48(19) GPa, which is similar to that of natural bones.
• Ball-milling for 50 h led to the formation of a rela-
tively equiaxed shape and powder particles of a
smaller size and with a more asymmetrical particle-
size distribution than that of the powders ball milled
for 70 h.
• Based on XRD, it was found that the process of
high-energy milling for 50 h and 70 h makes it pos-
sible to obtain a solid nanocrystalline solution  + .
Acknowledgments
This work was supported by the Polish National
Science Centre (Polish: Narodowe Centrum Nauki, abbr.
NCN) under the research project no. UMO-2011/03/
D/ST8/04884
5 REFERENCES
1 M. Long, H. J. Rack, Titanium alloys in total joint replacement – a
materials science perspective, Biomaterials, 19 (1998) 18,
1621–1639, doi:10.1016/S0142-9612(97)00146-4
2 Y. Li, C. Yang, H. Zhao, S. Qu, X. Li, Y. Li, New Developments of
Ti-Based Alloys for Biomedical Applications, Mater., 7 (2014) 3,
1709–1800, doi:10.3390/ma7031709
3 M. Niinomi, Mechanical biocompatibilities of titanium alloys for
biomedical application, J. Mech. Behav. Biomed. Mater., 1 (2008) 1,
30–42, doi:10.1016/j.jmbbm.2007.07.001
4 A. Biesiekierski, J. Wang, M. Gepreel, C. Wena, A new look at
biomedical Ti-based shape memory alloys, Acta Biomater. 8 (2012)
5, 1661–1669, doi:10.1016/j.actbio.2012.01.018
5 S. Miyazaki, H.Y. Kim, Basic characteristics of titanium–nickel
(Ti–Ni)-based and titanium–niobium (Ti–Nb)-based alloys, Shape
Memory and Superelastic Alloys, Appl. Technol., (2011), 15–42,
doi:10.1016/B978-1-84569-707-5.50002-X
6 M. Tahara, H. Y. Kim, H. Hosoda, S. Miyazaki, Cyclic deformation
behavior of Ti-26 at.% Nb alloy, Acta Mater., 25 (2009), 2461–2469,
doi:10.1016/j.actamat.2009.01.037
7 H. Y. Kim, Y. Ikehara, J. I. Kim, H. Hosoda, S. Miyazaki, Marten-
sitic transformation, shape memory effect and superelasticity of
Ti–Nb binary alloys, Acta Mater. 54 (2006), 2419–2429,
doi:10.1016/j.actamat.2006.01.019
8 G. Dercz, I. Matu³a, M. Zubko, A. Liberska, Structure characteri-
zation of biomedical Ti-Mo-Sn prepared by mechanical alloying
method, Acta Phys. Pol. A, 130 (2016), 1029–1032, doi:10.12693/
APhysPolA.130.1029
9 I. Matu³a, G. Dercz, M. Zubko, L. Paj¹k, Influence of high energy
milling time on the Ti-50Ta biomedical alloy structure, Acta Phys.
Pol. A, 130 (2016), 1033–1036, doi:10.12693/APhysPolA.130.1033
10 X. Rao, C. L. Chu, Y. Y. Zheng, Phase composition, microstructure,
and mechanical properties of porous Ti-Nb-Zr alloys prepared by a
two-step foaming powder metallurgy method, J. Mech. Behav.
Biomed., 34 (2014), 27–36, doi:10.1016/j.jmbbm.2014.02.001
11 G. Ryan, A. Pandit, D. P. Apatsidis, Fabrication methods of porous
metals for use in orthopaedic applications, Biomaterials, 27 (2006)
13, 2651–2670, doi:10.1016/j.biomaterials.2005.12.002
G. DERCZ, I. MATU£A: EFFECT OF BALL MILLING ON THE PROPERTIES OF THE POROUS Ti–26Nb ...
802 Materiali in tehnologije / Materials and technology 51 (2017) 5, 795–803
MATERIALI IN TEHNOLOGIJE/MATERIALS AND TECHNOLOGY (1967–2017) – 50 LET/50 YEARS
Figure 15: Diagram of reduced modulus to contact depth for both
samples, after 50 h and 70 h of milling
Figure 14: Diagram of hardness to contact depth for both samples,
after 50 h and 70 h of milling
12 H. M. Rietveld, A Profile Refinement Method for Nuclear and
Magnetic Structures, J. Appl. Cryst., 3 (1969), 65–69, doi:10.1107/
S0021889869006558
13 R. J. Hill, C. J. Howard, Quantitative phase analysis from neutron
powder diffraction data using the Rietveld method. J. Appl. Cryst.,
20 (1987), 467–474, doi:10.1107/S0021889887086199
14 G. Dercz, D. Oleszak, K. Prusik, L. Paj¹k, Rietveld-based quanti-
tative analysis of multiphase powders with nanocrystalline NiAl and
FeAl phases, Rev. Adv. Mater. Sci., 8 (2008), 764–768
15 Y. Li, Y. Cui, F. Zhanga, H. Xua, Shape memory behavior in Ti–Zr
alloys, Scripta Mater., 64 (2011) 6, 584–587, doi:/10.1016/
j.scriptamat.2010.11.048
16 A. Nouri, P. D. Hodgson, C. Wen, Effect of ball-milling time on the
structural characteristics of biomedical porous Ti–Sn–Nb alloy,
Mater. Sci. Eng. C, 31 (2011), 921–928, doi:10.1016/j.msec.2011.
02.011
17 G. Dercz, I. Matu³a, M. Zubko, J. Dercz, Phase composition and
microstructure of new Ti–Ta–Nb–Zr biomedical alloys prepared by
mechanical alloying method, Powder Diffr., (2017), 1–7,
doi:10.1017/S0885715617000045
18 L. Lü, M. O. Lai, Mechanical Alloying, Kluwer Academic Pub-
lishers, Boston, 1998
19 C. Suryanarayana, Mechanical alloying and milling, Prog. Mater.
Sci., 46 (2001) 1–2, 1–184, doi:10.1016/S0079-6425(99)00010-9
20 A. Omran, K. Woo, H. B. Lee, Mechanical properties of –Ti-
35Nb-2.5Sn alloy synthesized by mechanical alloying and pulsed
current activated sintering, Metall. Mater. Trans. A, 43 (2012) 12,
4866–4874, doi:10.1007/s11661-012-1298-y
21 G. Dercz, B. Formanek, K. Prusik, L. Paj¹k, Microstructure of
Ni(Cr)-TiC-Cr3C2-Cr7C3 composite powder, J. Mater. Process. Tech.,
162 (2005), 15–19, doi:10.1016/jmatprotec.2005.02.004
22 G. Dercz, L. Paj¹k, B. Formanek, Dispersion analysis of NiAl-
TiC-Al2O3 composite powder ground in a high-energy attritorial mill,
J. Mater. Process. Tech., 175 (2006), 334–337, doi:10.1016/
j.jmatprotec.2005.04.060
23 M. Tahara, H. Y. Kim, H. Hosoda, S. Miyazaki, Cyclic deformation
behavior of a Ti–26 at.% Nb alloy, Acta Mater., 57 (2009),
2461–2469, doi:10.1016/j.actamat.2009.01.037
24 E. K. Molchanova, Phase Diagrams of Titanium Alloys (Translation
of Atlas Diagram Sostoyaniya Titanovyk Splavov), Israel Program
for Scientific Translations, Jerusalem, 1965, 116
25 S. N. Patankar, F. H. Froes, Transformation of mechanically alloyed
Nb-Sn powder to Nb3Sn, Metall. Mater. Trans. A, 35 (2004),
3009–3012, doi:10.1007/s11661-004-0248-8
26 P. J. James, Particle Deformation During Cold Isostatic Pressing of
Metal Powders, Powder Metall., 20 (1977), 199–204, doi:10.1179/
pom.1977.20.4.199
27 L. Lu, M. O. Lai, G. Li, Influence of sintering process on the mecha-
nical property and microstructure of ball milled composite compacts,
Mater. Res. Bull., 31 (1996), 453–464, doi:10.1016/S0025-5408(96)
00024-4
G. DERCZ, I. MATU£A: EFFECT OF BALL MILLING ON THE PROPERTIES OF THE POROUS Ti–26Nb ...
Materiali in tehnologije / Materials and technology 51 (2017) 5, 795–803 803
MATERIALI IN TEHNOLOGIJE/MATERIALS AND TECHNOLOGY (1967–2017) – 50 LET/50 YEARS
The nanosized indentation measurements of the
mechanical properties are from the selected specific
areas, which neutralize the effects of the pores or defects
on the results. A large measurement uncertainty is even
more surprising. Assuming that the material is mechani-
cally widely heterogeneous, it seems necessary to apply
heat treatment to the material. The microhardness
measurement of the sample milled for 50 h revealed a
higher value of the microhardness and it is 288.6(±31.0)
HV0.5, which is around 2.82 GPa. For the sample milled
for 70 h, the revealed value is 256.2(±14.6) HV0.5,
which is approximately 2.51 GPa. However, we observed
a relatively high uncertainty of the measurement, which
made it difficult to clearly evaluate the differences in the
microhardness measurement.
4 CONCLUSIONS
Based on the investigation into the microstructure
and mechanical properties of the Ti-26Nb (a/%) alloy,
the main conclusions can be drawn:
• It was proved that the microstructure, the degree of
porosity and mechanical properties can be adjusted
by means of mechanical alloying.
• A longer milling time (70 h) resulted in an increased
degree of structure fragmentation. Young’s modulus
increased twice as well.
• The porous samples sintered from the powders milled
for 50 hours have a very low elastic modulus of
48(19) GPa, which is similar to that of natural bones.
• Ball-milling for 50 h led to the formation of a rela-
tively equiaxed shape and powder particles of a
smaller size and with a more asymmetrical particle-
size distribution than that of the powders ball milled
for 70 h.
• Based on XRD, it was found that the process of
high-energy milling for 50 h and 70 h makes it pos-
sible to obtain a solid nanocrystalline solution  + .
Acknowledgments
This work was supported by the Polish National
Science Centre (Polish: Narodowe Centrum Nauki, abbr.
NCN) under the research project no. UMO-2011/03/
D/ST8/04884
5 REFERENCES
1 M. Long, H. J. Rack, Titanium alloys in total joint replacement – a
materials science perspective, Biomaterials, 19 (1998) 18,
1621–1639, doi:10.1016/S0142-9612(97)00146-4
2 Y. Li, C. Yang, H. Zhao, S. Qu, X. Li, Y. Li, New Developments of
Ti-Based Alloys for Biomedical Applications, Mater., 7 (2014) 3,
1709–1800, doi:10.3390/ma7031709
3 M. Niinomi, Mechanical biocompatibilities of titanium alloys for
biomedical application, J. Mech. Behav. Biomed. Mater., 1 (2008) 1,
30–42, doi:10.1016/j.jmbbm.2007.07.001
4 A. Biesiekierski, J. Wang, M. Gepreel, C. Wena, A new look at
biomedical Ti-based shape memory alloys, Acta Biomater. 8 (2012)
5, 1661–1669, doi:10.1016/j.actbio.2012.01.018
5 S. Miyazaki, H.Y. Kim, Basic characteristics of titanium–nickel
(Ti–Ni)-based and titanium–niobium (Ti–Nb)-based alloys, Shape
Memory and Superelastic Alloys, Appl. Technol., (2011), 15–42,
doi:10.1016/B978-1-84569-707-5.50002-X
6 M. Tahara, H. Y. Kim, H. Hosoda, S. Miyazaki, Cyclic deformation
behavior of Ti-26 at.% Nb alloy, Acta Mater., 25 (2009), 2461–2469,
doi:10.1016/j.actamat.2009.01.037
7 H. Y. Kim, Y. Ikehara, J. I. Kim, H. Hosoda, S. Miyazaki, Marten-
sitic transformation, shape memory effect and superelasticity of
Ti–Nb binary alloys, Acta Mater. 54 (2006), 2419–2429,
doi:10.1016/j.actamat.2006.01.019
8 G. Dercz, I. Matu³a, M. Zubko, A. Liberska, Structure characteri-
zation of biomedical Ti-Mo-Sn prepared by mechanical alloying
method, Acta Phys. Pol. A, 130 (2016), 1029–1032, doi:10.12693/
APhysPolA.130.1029
9 I. Matu³a, G. Dercz, M. Zubko, L. Paj¹k, Influence of high energy
milling time on the Ti-50Ta biomedical alloy structure, Acta Phys.
Pol. A, 130 (2016), 1033–1036, doi:10.12693/APhysPolA.130.1033
10 X. Rao, C. L. Chu, Y. Y. Zheng, Phase composition, microstructure,
and mechanical properties of porous Ti-Nb-Zr alloys prepared by a
two-step foaming powder metallurgy method, J. Mech. Behav.
Biomed., 34 (2014), 27–36, doi:10.1016/j.jmbbm.2014.02.001
11 G. Ryan, A. Pandit, D. P. Apatsidis, Fabrication methods of porous
metals for use in orthopaedic applications, Biomaterials, 27 (2006)
13, 2651–2670, doi:10.1016/j.biomaterials.2005.12.002
G. DERCZ, I. MATU£A: EFFECT OF BALL MILLING ON THE PROPERTIES OF THE POROUS Ti–26Nb ...
802 Materiali in tehnologije / Materials and technology 51 (2017) 5, 795–803
MATERIALI IN TEHNOLOGIJE/MATERIALS AND TECHNOLOGY (1967–2017) – 50 LET/50 YEARS
Figure 15: Diagram of reduced modulus to contact depth for both
samples, after 50 h and 70 h of milling
Figure 14: Diagram of hardness to contact depth for both samples,
after 50 h and 70 h of milling
12 H. M. Rietveld, A Profile Refinement Method for Nuclear and
Magnetic Structures, J. Appl. Cryst., 3 (1969), 65–69, doi:10.1107/
S0021889869006558
13 R. J. Hill, C. J. Howard, Quantitative phase analysis from neutron
powder diffraction data using the Rietveld method. J. Appl. Cryst.,
20 (1987), 467–474, doi:10.1107/S0021889887086199
14 G. Dercz, D. Oleszak, K. Prusik, L. Paj¹k, Rietveld-based quanti-
tative analysis of multiphase powders with nanocrystalline NiAl and
FeAl phases, Rev. Adv. Mater. Sci., 8 (2008), 764–768
15 Y. Li, Y. Cui, F. Zhanga, H. Xua, Shape memory behavior in Ti–Zr
alloys, Scripta Mater., 64 (2011) 6, 584–587, doi:/10.1016/
j.scriptamat.2010.11.048
16 A. Nouri, P. D. Hodgson, C. Wen, Effect of ball-milling time on the
structural characteristics of biomedical porous Ti–Sn–Nb alloy,
Mater. Sci. Eng. C, 31 (2011), 921–928, doi:10.1016/j.msec.2011.
02.011
17 G. Dercz, I. Matu³a, M. Zubko, J. Dercz, Phase composition and
microstructure of new Ti–Ta–Nb–Zr biomedical alloys prepared by
mechanical alloying method, Powder Diffr., (2017), 1–7,
doi:10.1017/S0885715617000045
18 L. Lü, M. O. Lai, Mechanical Alloying, Kluwer Academic Pub-
lishers, Boston, 1998
19 C. Suryanarayana, Mechanical alloying and milling, Prog. Mater.
Sci., 46 (2001) 1–2, 1–184, doi:10.1016/S0079-6425(99)00010-9
20 A. Omran, K. Woo, H. B. Lee, Mechanical properties of –Ti-
35Nb-2.5Sn alloy synthesized by mechanical alloying and pulsed
current activated sintering, Metall. Mater. Trans. A, 43 (2012) 12,
4866–4874, doi:10.1007/s11661-012-1298-y
21 G. Dercz, B. Formanek, K. Prusik, L. Paj¹k, Microstructure of
Ni(Cr)-TiC-Cr3C2-Cr7C3 composite powder, J. Mater. Process. Tech.,
162 (2005), 15–19, doi:10.1016/jmatprotec.2005.02.004
22 G. Dercz, L. Paj¹k, B. Formanek, Dispersion analysis of NiAl-
TiC-Al2O3 composite powder ground in a high-energy attritorial mill,
J. Mater. Process. Tech., 175 (2006), 334–337, doi:10.1016/
j.jmatprotec.2005.04.060
23 M. Tahara, H. Y. Kim, H. Hosoda, S. Miyazaki, Cyclic deformation
behavior of a Ti–26 at.% Nb alloy, Acta Mater., 57 (2009),
2461–2469, doi:10.1016/j.actamat.2009.01.037
24 E. K. Molchanova, Phase Diagrams of Titanium Alloys (Translation
of Atlas Diagram Sostoyaniya Titanovyk Splavov), Israel Program
for Scientific Translations, Jerusalem, 1965, 116
25 S. N. Patankar, F. H. Froes, Transformation of mechanically alloyed
Nb-Sn powder to Nb3Sn, Metall. Mater. Trans. A, 35 (2004),
3009–3012, doi:10.1007/s11661-004-0248-8
26 P. J. James, Particle Deformation During Cold Isostatic Pressing of
Metal Powders, Powder Metall., 20 (1977), 199–204, doi:10.1179/
pom.1977.20.4.199
27 L. Lu, M. O. Lai, G. Li, Influence of sintering process on the mecha-
nical property and microstructure of ball milled composite compacts,
Mater. Res. Bull., 31 (1996), 453–464, doi:10.1016/S0025-5408(96)
00024-4
G. DERCZ, I. MATU£A: EFFECT OF BALL MILLING ON THE PROPERTIES OF THE POROUS Ti–26Nb ...
Materiali in tehnologije / Materials and technology 51 (2017) 5, 795–803 803
MATERIALI IN TEHNOLOGIJE/MATERIALS AND TECHNOLOGY (1967–2017) – 50 LET/50 YEARS