-
This content has been downloaded from IOPscience. Please scroll
down to see the full text.
Download details:
IP Address: 128.122.253.228This content was downloaded on
21/04/2015 at 20:38
Please note that terms and conditions apply.
General nanomoulding with bulk metallic glasses
View the table of contents for this issue, or go to the journal
homepage for more
2015 Nanotechnology 26 145301
(http://iopscience.iop.org/0957-4484/26/14/145301)
Home Search Collections Journals About Contact us My
IOPscience
-
General nanomoulding with bulk metallicglasses
Ze Liu1,2 and Jan Schroers1,2
1Department of Mechanical Engineering and Materials Science,
Yale University, New Haven, CT06520, USA2Center for Research on
Interface Structures and Phenomena, Yale University, New Haven, CT
06511,USA
E-mail: [email protected] and [email protected]
Received 24 November 2014, revised 30 January 2015Accepted for
publication 19 February 2015Published 18 March 2015
AbstractBulk metallic glasses (BMGs) are ideal for nanomoulding
as they possess desirable strength formolds as well as for moldable
materials and furthermore lack intrinsic size limitations.
Despitetheir attractiveness, only recently Pt-based BMGs have been
successfully molded into poresranging 10100 nm (Kumar et al 2009
Nature 457 86872). Here, we introduce a quantitativetheory, which
reveals previous challenges in lling nanosized pores. This theory
considers, inaddition to a viscous and a capillary term, also
oxidation, which becomes increasingly moreimportant on smaller
length scales. Based on this theory we construct a
nanomouldingprocessing map for BMG, which reveals the limiting
factors for BMG nanomoulding. Based onthe quantitative prediction
of the processing map, we introduce a strategy to reduce the
capillaryeffect through a wetting layer, which allows us to mold
non-noble BMGs below 1 m in air. Anadditional benet of this
strategy is that it drastically facilitates demoulding, one of the
mainchallenges of nanomoulding in general.
Keywords: bulk metallic glasses, nanomoulding, nanoimprinting,
wetting/dewetting, surfaceoxide
(Some gures may appear in colour only in the online journal)
1. Introduction
A critical requirement for any moldable material is that
itsintrinsic length scale is small compared to the size of themolds
features. Polymers are ultimately limited by theirchain lengths and
crystalline materials are limited by theirgrain sizes. Bulk
metallic glasses (BMGs) lack intrinsic sizelimitation on the
nanoscale [15]. In addition to this crucialattribute, BMGs are hard
and tough compared to othermaterials used for nanomoulding, yet
they can be softenedinto a state in which they can be readily
molded [2]. Theseproperties are often combined with favorable
chemistry forbiocompatibility, electrochemical activity,
antibacterialbehavior, and biodegradability.
Despite their attractiveness, until recently, BMGs couldnot be
molded into features below 100 nm. Kumar et al [1]was rst to mold
Pt57.5Cu14.7Ni5.3P22.5 into pores of10100 nm in diameter and
lengths that correspond to an
aspect ratio of over 20. The ability to mold BMGs in
acontrollable manner has triggered broad research in
exploringapplications, such as electrochemical catalysts [69],
microfuel cell [10], MEMS/NEMS [5, 11, 12],
programmablebiomaterials [1315] and wastewater treatment
materials[16, 17]. Besides these applications, molded BMG
nanos-tructures have also enabled the scientic community to
studysize and connement effects on mechanical properties [1825],
deformation modes [2629], crystallization [30, 31] andow behaviors
[23]. However, Kumars ndings and all fol-lowing BMG nanomoulding
has been limited to very few,noble-metals-based alloys such as
Pt57.5Cu14.7Ni5.3P22.5 [1]and Pd43Ni10Cu27P20 [3234] with a high
resistance to oxideformation. Enabling the broad potential
applications of BMGnanomoulding requires the development of
nanomouldingstrategies for more economic BMGs. Most BMGs,
however,readily oxidize, particularly when heated into the
supercooledliquid region (SCLR) like during thermoplastic
forming
Nanotechnology
Nanotechnology 26 (2015) 145301 (9pp)
doi:10.1088/0957-4484/26/14/145301
0957-4484/15/145301+09$33.00 2015 IOP Publishing Ltd Printed in
the UK1
-
(TPF) [35]. Such oxidizing results in the formation of a
rigidoxide layer which dramatically affects and ultimately
prohi-bits the TPF lling of nanosized pores for the vast majority
ofBMGs. We show here a general description for
quantitativeprediction of molding of BMGs into nanoscale
features,which accounts for possible oxide formation. Our
modelincludes a threshold pressure and a capillary
contribution,which are required to break the oxide lm and overcome
thecapillary force to enter into a nanopore, respectively.
Inaddition, a viscous term, which when compared to thestrength of
the mold, denes the possible moldable aspectratios. Based on this
quantitative model which we use toconstruct a nanomoulding
processing map forZr35Ti30Cu8.25Be26.75, we introduce a strategy to
reduce thecapillary effect, which allows us to fabricate nanorods
in airfrom highly reactive, non-noble based BMGs such
asZr35Ti30Cu8.25Be26.75, Zr44Ti11Cu10Ni10Be25 andMg65Cu25Y10.
2. Materials and methods
The nanomoulding experiments are operated in air with
aheating-cell equipped Instron 5569 machine (50 kN maximumload
capacity). The processing temperature forZr35Ti30Cu8.25Be26.75,
Zr44Ti11Cu10Ni10Be25 andMg65Cu25Y10 with/without oils is 435 C and
175 C, respec-tively. The total processing time for all of the
samples is2min and the maximum loading force is 50 KN. Demouldingof
Zr35Ti30Cu8.25Be26.75 and Zr44Ti11Cu10Ni10Be25 nanorods isachieved
through etching Al2O3 templates with KOH solution(3 mol L1,
temperature of 85 C). For Mg65Cu25Y10, we useH3PO4 solution
(concentration of 85%, temperature of 85 C)due to the high
reactivity of the alloy with both alkaline andacidic solutions.
To demonstrate the use of BMGs as molds and moldablematerials,
we choose as an example Pt57.5Cu14.7Ni5.3P22.5 as amoldable
material and Pd43Ni10Cu27P20 as a mold material.Firstly, a
Pd43Ni10Cu27P20 BMG disc is thermoplasticallyformed into a
nanoporous Al2O3 template at 360 C. Themaximum loading force and
processing time are 5 KN and2 min, respectively. Pd43Ni10Cu27P20
nanorods arrays to beused subsequently as a mold are exposed by
etching theAl2O3 template. We then place a Pt57.5Cu14.7Ni5.3P22.5
discwhich is covered by a thin layer of high temperature oil on
thePd43Ni10Cu27P20 nanorods array mold and thermoplasticallyform
the mold-replicate combination at 270 C. Demouldingis realized by a
small mechanical force applied at roomtemperature.
3. Processing map for BMG nanomoulding
Thermoplastic compression molding, a process in whichmoldable
material is heated to ow into micro/nanoporeunder an applied
pressure, has been widely used in surfacepatterning for polymers
[3639] and BMGs [3, 12, 4045]due to its simplicity and scalability.
Typically for a cylindrical
pore with diameter, d, the required pressure to drive a
liquidinto a pore, p, can be calculated from HagenPoiseuille
law
= p Ld
4 , (1)max
where is the liquids viscosity and max is the maximumshear
strain rate. max is located at the liquid-mold interfacewhen
assuming a parabolic velocity distribution. This owresistance
linearly scales with the viscosity of the liquid(gure 1(a)).
Typically for creep ow of BMGs in theirSCLRs, ~ 1 s ,max 1 and
their minimum accessible viscositycan vary signicantly among BMG
formers ranging 105109 Pa s [46]. For > 109 Pa s, the molding
pressure of 10 GPais prohibitively high to ll an aspect ratio of
three. Figure 1(a)and equation (1) also suggest that the viscous
resistance ofBMG liquids is size independent, which has been
experi-mentally conrmed for molds larger than approximately
onemicron [47]. However, for nanosized molds,
-
oxide layer should be of the form of
=p
Ef
Ev
h
d, , , (5)0 c
where c is the strength of the oxide layer, and h the
thicknessof the oxide layer. From linear theory of small deection
andvon-Mises yielding criterion, equation (5) can be written as
= +
pv v
h
d
16
3 1, (6)0
c
2
2
which denes the pressure to break the oxide layer. Thispressure
barrier scales with (1/d)2 and is more dramatic thanthe
linear-scaling entering barrier from surface dewetting(second term
in equation (2)). This scaling behavior revealsthat the oxide layer
barrier is indeed dominating the llingrequirements of small pores.
Considering the nonlinear effect,the more accurate pressure to
break the oxide layer (p0/E) isrelated to the maximum deection
value (wm/h) when rstyielding at clamped edges [48]
+
= + +
+ +
v
v v E
a
h
w
hv v
w
hv v
w
h
1
14
1
90(1 )(113 13 )
1
1890(1 ) (40 26 ) , (7)
2
2
c2
m
m2
2 m3
=
+p
E v
h
a
w
h
w
h
16
3
1
1, (8)0
2
4m m
3
where = + v v(1 )(173 73 ).1360
By substituting themechanical properties of the oxide layer (E
200 GPa,c 1 GPa, and v 0.3) and the oxide thickness (1 and10 nm)
into equation (7), the maximum deection value (wm/h) at yielding is
calculated. Then substituting this value intoequation (8), the
critic pressure p0 to rupture the oxide layer isnally obtained
(gure 1(c)). It is obvious that when moldinginto small pores, the
surface oxide layer becomes increasinglyimportant. However, this is
only the case for reactive alloyssuch as Zr-BMGs. For alloys with
high corrosion resistancesuch as Pt-BMGs, the contribution of an
oxide layer can beneglected.
The maximum aspect ratio that can be achieved inmolding nanorods
is ultimately limited by the strength of themold (0), which denes
the upper bound for the appliedpressure. As a consequence, the
maximum aspect ratio thatcan be realized is given by
=
+L
d d
1
4
4 cos. (9)
max max0
Equations (2), (7), (8) and (9) describe the generalnanomoulding
of supercooled BMGs (gures 1(a)(c)). Thisdescription can be used to
construct a BMGs processingmap, which reveals the dominating effect
of nanomouldingat different pore diameters. Taking
Zr35Ti30Cu8.25Be26.75 forexample, parameters are ~ 1 s ,max 1 = 107
Pa s, = 150, = 1 J m2. Furthermore, we assume a maximum strength
of0 = 300 MPa for Al2O3 template (with 200 nm diameter
Figure 1. Thermoplastic based nanomoulding of bulk metallic
glasses. (a) Contribution of the viscosity term to the forming
pressure for BMGliquids owing into a nanopore (equation (1)) for an
aspect ratio of three, and a maximum shear strain rate of max 1 s1.
(b) For nanosizedmolds, the capillary effect should be included in
the pressure (second term in equation (2)) which inversely linearly
scales with the diameterof a pore. (c) When nanomoulding
supercooled BMGs in air, the BMG surface is typically covered with
an oxide layer, which deforms as arigid lm under normal pressure
exerted by the BMG liquid. The pressure needed to break the oxide
layer scales with the square of thediameter of a nanopore. (d)
Processing map for Zr35Ti30Cu8.25Be26.75 which is constructed based
on (a)(c), with max 1 s1, = 107 Pa s,= 150, = 1 J m2, and h= 10 nm.
The molding pressure is limited by the strength of the mold,
approximately 300 MPa [49] for Al2O3template assuming a 0.2%
elastic limit. Formation of nanorods is initially limited by the
pressure requirement to break the oxide layer (blackline) and/or
overcome capillary force (red line), which are the onset boundaries
of nanomoulding.
3
Nanotechnology 26 (2015) 145301 Z Liu and J Schroers
-
pores) from a 0.2% elastic limit [49]. Combiningequations (2),
(3), and (5), the processing map can be con-structed (gure 1(d)),
where the oxide thickness is set ash = 10 nm, the upper bound in
our experiments. We measuresuch a thickness of the oxide layer by
oxidizingZr35Ti30Cu8.25Be26.75 thin plate at 435 C for 4 min in
air.Subsequently, the plate is stretched to break the surfaceoxide
layer and the thickness is determined using SEM(gure 5).
4. Results and discussions
4.1. Nanomoulding non-noble BMGs by adding a wetting layer
The processing map (gure 1(d)) reveals the dominating
effects controlling molding into different size pores. Prior
to
this work, Zr-BMGs have not been molded into features
smaller than 1 m. For this size region, the processing map
reveals that the dominating barrier originates from either
the
Figure 2. (a) Molding reactive BMGs by reducing capillary force
through the use of an interlayer wetting layer. Al2O3 templates are
coatedwith high temperature oil (DuPont Kryto) prior to the molding
operation. (b) After thermoplastic forming Zr35Ti30Cu8.25Be26.75
into a200 nm Al2O3 template at 435 C for 2 min. Left: without the
use of a wetting layer, right: with the use of high temperature oil
as a wettinglayer. (c) Nanorods do not form without oil. (d)
Formation of nanorods under the same conditions as (c) but with
oil.
4
Nanotechnology 26 (2015) 145301 Z Liu and J Schroers
-
capillary effect or surface oxide. To reduce these barriers,
andhence enable nanomoulding with non-noble BMGs, we pro-pose a
simple strategy through introducing a wetting layer tothe Al2O3
template surface (gure 2(a)). Specically, weimmerse Al2O3 templates
and/or BMGs into high temperatureoil prior to the molding
operation. Oils typically wet ceramicsand metals. We choose DuPont
Kryto oil which wets Al2O3template and Zr35Ti30Cu8.25Be26.75 well
(contact angle< 45, gure 6). By reducing the surface energy
throughthe surface wetting oil, we successfully
moldZr35Ti30Cu8.25Be26.75 and Zr44Ti11Cu10Ni10Be25 into 100
nmnanopores. One of the results for 200 nm nanorods withaspect
ratio over ve is shown in gure 2(d). In contrast,
molding under the same processing conditions but without oildoes
not result in any lling of the 200 nm pores (gure 2(c)).
We attribute the successful nanomoulding of Zr-BMGwhen using an
oil layer to the reduced surface energy ofBMG liquids which
decreases the capillary (second term inequation (2)). When forming
supercooled Zr35Ti30Cu8.25Be26.75into a nanopore wetted with oil,
the capillary term inequation (2) changes to d4( )/ ,oil BMG oil
and the interfacialtension between oil and supercooled BMG can be
estimated by = + oil BMG oil BMG oil BMG [50]. The capillary
termthus becomes ( ) d4 1 / / ,BMG oil BMG which suggests thatthe
capillary effect can be reduced by using oil. In addition, the
Figure 3. Nanomoulding Zr44Ti11Cu10Ni10Be25 and Mg65Cu25Y10 by
using oil-wetted Al2O3 templates. Zr44Ti11Cu10Ni10Be25
andMg65Cu25Y10 are molded at 435 C and 175 C, respectively. The
total processing time is 2 min and the maximum loading force is 50
KN.(a) and (d) show the as-thermoplastic formed samples. The dark
color in both of the BMGs suggests that nanorods formed. (b)(c)
SEMimages of Zr44Ti11Cu10Ni10Be25 nanorods after removing the Al2O3
template through KOH etching. (e)(f) Mg65Cu25Y10 nanorods withAl2O3
template partially etched by using H3PO4. The reactivity of
Mg65Cu25Y10 with both alkaline and acidic solutions which are
required toetch the Al2O3 template causes degradation of the
nanorods.
5
Nanotechnology 26 (2015) 145301 Z Liu and J Schroers
-
Figure 4. BMGs used as mold or moldable materials. (a)
Requirements for a BMG mold/moldable material combination are that
one BMGhas sufciently softened, which is achieved by processing in
its supercooled liquid region close to its crystallization
temperature. The otherBMG to be used as a mold should have a 20%
higher Tg than that of the moldable BMG. Such a difference
guarantees several orders ofmagnitude difference in strength (ow
stress). (b) Schematics of nanomoulding Pt-BMG using Pd-BMG
nanorods as a mold. (c)(d)Pd43Ni10Cu27P20 nanorods after demoulding
from Pt57.5Cu14.7Ni5.3P22.5. (e)(f) Replicated nanoholes in
Pt57.5Cu14.7Ni5.3P22.5.
Figure 5. Surface oxide layer on Zr35Ti30Cu8.25Be26.75, formed
during processing in air at 435 C. Subsequently the sample is
deformed torupture the surface oxide, resulting in a total
processing time of 4 min. (a) Optical microscopy image reveals the
rupture of the surface oxidelayer. (b)(c) By using SEM (the sample
stage is tilted 30 degrees), the oxide thickness is measured to
about 156 nm, which corresponds toan oxidation rate of 0.65 nm s1
by assuming a linear oxidation process. Within our processing
protocol, it takes 15 s to heat the sample,which gives the upper
bound thickness of the oxide layer to 0.65*15 nm 10 nm.
6
Nanotechnology 26 (2015) 145301 Z Liu and J Schroers
-
changing of sticky boundary to lubrication boundary may helpto
reduce the owing resistance and thus enable the fabricationof
nanorods with larger aspect ratio, which is still
underinvestigation.
To demonstrate the general applicability of this strategy,we
also mold Zr44Ti11Cu10Ni10Be25 and Mg65Cu25Y10 intopores of 200 nm
in diameter (gure 3). When applying an oillm, both alloys can be
nanomolded. The appearance of theMg65Cu25Y10 nanorods is affected
by the subsequent H3PO4etching to remove the Al2O3 template.
4.2. BMGs used as molds and moldable materials
Besides using BMGs as ideal moldable materials, BMGsdrastic
temperature dependent strength can also be utilized asa means to
use them as molds. For example the softeningtemperatures of
thermoplastics are usually below 200 C,which is lower than the
glass transition temperatures (Tg) formost BMGs. Hence, BMGs are
high strength at these tem-peratures, hence can be used as molds
for thermoplasticpolymers. Due to the large range of glass
transition tem-peratures among BMGs, nanomoulding one BMG
withanother BMG mold is also possible. The selection of
theappropriate BMGs depends on their relative glass
transitiontemperatures. This attribute and requirement are
schemati-cally shown in gure 4(a), where the viscosity (or ow
stress)of BMGs can decrease by several orders of magnitude whenthe
temperature approaches Tg. Compared with other moldmaterials such
as silicon and Al2O3 molds, BMG nanomoldsare advantageous due to
their high elasticity, strength, andtoughness. These attributes
warrant long mold life, ability todemould, and precise and
inexpensive fabrication of the mold.
To demonstrate this concept, Pd43Ni10Cu27P20 nanorodarrays of
200 nm in diameter are fabricated by replicatingAl2O3 mold at 360
C. Subsequently, these nanorod arrays actas a mold and are
replicated at 270 C with oil-wettedPt57.5Cu14.7Ni5.3P22.5.
Demoulding at room temperature canbe readily achieved by small
mechanical forces. This method,in contrast to all previous BMG
nanomoulding methods,allows the reuse of the mold. As such it
represents a steptowards commercial usage of BMG nanostructures
since itdrastically reduces fabrication costs which in the past
inclu-ded the disposable Al2O3 template.
5. Conclusions
We identied the governing contributions for most generalBMG
nanomoulding. These are viscous, capillary, and sur-face oxide
layer fracture. The viscous contribution dominatesfor supercooled
BMGs with large viscosity, when lled intohigh aspect ratios. The
capillary term scales with the char-acteristic dimension of the
nanopore and is determined byinterface wetting properties (surface
energy/contact angle).The surface oxide layer controls the lling
barrier at smallerscales since it scales with the square of the
diameter of thenanopore. Our general description allows for
quantitativepredictions, which can be used to create a processing
map forBMG nanomoulding.
The processing map reveals the limiting factors, whichhave
prevented nanomoulding for the majority of BMG alloysin the past.
We used this knowledge to develop a moldingstrategy using a wetting
layer to reduce the entering barriers.Using this method, reactive
BMGs based on Zr and Mg canbe nanomolded under practical
conditions. Furthermore, thewetting layer also dramatically reduces
interfacial forces,which suggest the possibility for a
non-disposable moldmethod for BMG nanomoulding.
Acknowledgment
This work was primarily supported by the National
ScienceFoundation under MRSEC DMR-1119826. Facilities use
wassupported by YINQE. We thank Dr Sungwoo for help
castingMg-BMG.
Appendix
A.1. Oxidation rates
We oxidized Zr35Ti30Cu8.25Be26.75 in air at 435 C and
sub-sequently stretched the BMG to break its surface oxide
layer.The total processing time is 4 min. The ruptured surfaceoxide
layer is clearly revealed under an optical microscope(gure 5(a)).
Using SEM (gures 5(b)(c), the sample stage istilted 30), the oxide
thickness is measured to 156 nm,which corresponds to a 0.65 nm s1
oxidation rate if assuming
Figure 6. Wetting of high temperature oil (DuPont Kryto) on
Al2O3 template and polished Zr35Ti30Cu8.25Be26.75.
7
Nanotechnology 26 (2015) 145301 Z Liu and J Schroers
-
a linear oxidation. During the TPF-based molding, we
usuallyconsume 15 s to reach the processing temperature, whichdenes
the upper bound thickness of the oxide layer to10 nm (0.65*15
nm).
A.2. Wetting properties of high temperature oil with
templatesand BMGs
The contact angles of oil drops on surfaces of Al2O3 andpolished
Zr35Ti30Cu8.25Be26.75 are smaller than 45 (gure 6),which suggests
the high temperature oil (DuPont Kryto) usedin our experiments wet
both of them well.
References
[1] Kumar G, Tang H X and Schroers J 2009 Nanomoulding
withamorphous metals Nature 457 86872
[2] Schroers J 2010 Processing of bulk metallic glass Adv.
Mater.22 156697
[3] Saotome Y, Itoh K, Zhang T and Inoue A 2001
Superplasticnanoforming of Pd-based amorphous alloy Scr. Mater.
4415415
[4] Kumar G, Desai A and Schroers J 2011 Bulk metallic glass:
thesmaller the better Adv. Mater. 23 46176
[5] Sharma P, Kaushik N, Kimura H, Saotome Y and Inoue A2007
Nano-fabrication with metallic glassan exoticmaterial for
nano-electromechanical systemsNanotechnology 18 035302
[6] Brower W E, Matyjaszczyk M S, Pettit T L and Smith G V1983
Metallic glasses as novel catalysts Nature 301 4979
[7] Sekol R C et al 2013 PdNiCuP metallic glass nanowires
formethanol and ethanol oxidation in alkaline media Int. J.Hydrog.
Energy 38 1124855
[8] Carmo M et al 2011 Bulk metallic glass nanowire
architecturefor electrochemical applications ACS Nano 5 297983
[9] Zhao M, Abe K, Yamaura S, Yamamoto Y and Asao N
2014Fabrication of PdNiP metallic glass nanoparticles andtheir
application as highly durable catalysts in
methanolelectro-oxidation Chem. Mater. 26 105661
[10] Sekol R C et al 2013 Bulk metallic glass micro fuel cell
Small9 20815
[11] Nakayama K S et al 2010 Controlled formation andmechanical
characterization of metallic glassy nanowiresAdv. Mater. 22 872
[12] Schroers J, Pham Q and Desai A 2007 Thermoplastic formingof
bulk metallic glassa technology for MEMS andmicrostructure
fabrication J. Microelectromech. Syst. 162407
[13] Schroers J, Kumar G, Hodges T M, Chan S and Kyriakides T
R2009 Bulk metallic glasses for biomedical applications Jom61
219
[14] Padmanabhan J et al 2014 Engineering cellular response
usingnanopatterned bulk metallic glass ACS Nano 8 436675
[15] Zberg B, Uggowitzer P J and Lofer J F 2009 MgZnCa
glasseswithout clinically observable hydrogen evolution
forbiodegradable implants Nat. Mater. 8 88791
[16] Lin B, Bian X F, Wang P and Luo G P 2012 Application of
Fe-based metallic glasses in wastewater treatment Mater. Sci.Eng. B
177 925
[17] Wang J Q et al 2012 Rapid degradation of Azo dye by
Fe-based metallic glass powder Adv. Funct. Mater. 22256770
[18] Guo H et al 2007 Tensile ductility and necking of
metallicglass Nat. Mater. 6 7359
[19] Lee C J, Huang J C and Nieh T G 2007 Sample size effect
andmicrocompression of Mg65Cu25Gd10 metallic glass Appl.Phys. Lett.
91 081601
[20] Jang D C and Greer J R 2010 Transition from a
strong-yet-brittle to a stronger-and-ductile state by size
reduction ofmetallic glasses Nat. Mater. 9 2159
[21] Magagnosc D J et al 2013 Tunable tensile ductility in
metallicglasses Sci. Rep. 3 1096
[22] Tnnies D, Maa R and Volkert C A 2014 Room
temperaturehomogeneous ductility of micrometer-sized metallic
glassAdv. Mater. 26 571521
[23] Shao Z et al 2013 Size-dependent viscosity in the
super-cooledliquid state of a bulk metallic glass Appl. Phys. Lett.
102221901
[24] Chen D Z et al 2013 Nanometallic glasses: size
reductionbrings ductility, surface state drives its extent Nano
Lett. 1344628
[25] Gu X W et al 2014 Mechanisms of failure in
nanoscalemetallic glass Nano Lett. 14 585864
[26] Volkert C A, Donohue A and Spaepen F 2008 Effect of
samplesize on deformation in amorphous metals J. Appl. Phys.
103083539
[27] Chen C Q, Pei Y T and De H J T M 2010 Effects of size on
themechanical response of metallic glasses investigated throughin
situ TEM bending and compression experiments ActaMater. 58
189200
[28] Greer J R and De H J T M 2011 Plasticity in
small-sizedmetallic systems: intrinsic versus extrinsic size effect
Prog.Mater. Sci. 56 654724
[29] Wang Y B et al 2014 Ultrahigh-strength
submicron-sizedmetallic glass wires Scr. Mater. 84-85 2730
[30] Gopinadhan M et al 2013 Finite size effects in
thecrystallization of a bulk metallic glass Appl. Phys. Lett.
103111912
[31] Sun Y T et al 2014 Understanding glass-forming
abilitythrough sluggish crystallization of atomically thin
metallicglassy lms Appl. Phys. Lett. 105 14
[32] Mukherjee S et al 2012 Palladium nanostructures from
multi-component metallic glass Electrochim. Acta 74 14550
[33] Liu X, Shao Y, Li J F, Chen N and Yao K F 2014
Large-areaand uniform amorphous metallic nanowire arrays preparedby
die nanoimprinting J. Alloys Compd. 605 711
[34] Mukherjee S et al 2013 Tunable hierarchical
metallic-glassnanostructures Adv. Funct. Mater. 23 270813
[35] Schroers J 2008 On the formability of bulk metallic glass
in itssupercooled liquid state Acta Mater. 56 4718
[36] Chou S Y, Krauss P R and Renstrom P J 1996
Imprintlithography with 25-nanometer resolution Science 272 857
[37] Rohr T, Ogletree D F, Svec F and Frechet J M J 2003
Surfacefunctionalization of thermoplastic polymers for
thefabrication of microuidic devices by photoinitiated graftingAdv.
Funct. Mater. 13 26470
[38] Lendlein A, Jiang H Y, Junger O and Langer R 2005
Light-induced shape-memory polymers Nature 434 87982
[39] Nie Z H and Kumacheva E 2008 Patterning surfaces
withfunctional polymers Nat. Mater. 7 27790
[40] Chu J P et al 2007 Nanoimprint of gratings on a bulk
metallicglass Appl. Phys. Lett. 90 034101
[41] Li N, Xu X N, Zheng Z Z and Liu L 2014 Enhancedformability
of a Zr-based bulk metallic glass in asupercooled liquid state by
vibrational loading Acta Mater.65 40011
[42] Li N et al 2013 A thermoplastic forming map of a
Zr-basedbulk metallic glass Acta Mater. 61 192131
[43] Xia T, Li N, Wu Y and Liu L 2012 Patternedsuperhydrophobic
surface based on Pd-based metallic glassAppl. Phys. Lett. 101
081601
8
Nanotechnology 26 (2015) 145301 Z Liu and J Schroers
-
[44] Li N, Xia T, Heng L P and Liu L 2013 Superhydrophobic
Zr-based metallic glass surface with high adhesive force Appl.Phys.
Lett. 102 251603
[45] Chen W, Liu Z, Robinson H M and Schroers J 2014
Flawtolerance versus performance: a tradeoff in metallic
glasscellular structures Acta Mater. 73 25974
[46] Kato H et al 2006 Fragility and thermal stability of Pt-and
Pd-based bulk glass forming liquids and their correlation
withdeformability Scr. Mater. 54 20237
[47] Chiu H M, Kumar G, Blawzdziewicz J and Schroers J
2009Thermoplastic extrusion of bulk metallic glass Scr. Mater.61
2831
[48] Chien W Z 1947 Large deection of a circular clamped
plateunder uniform pressure Acta Phys. Sin. 7 6
[49] Ko S et al 2006 Mechanical properties and residual stress
inporous anodic alumina structures Thin Solid Films 515 19327
[50] Fowkes F M 1964 Attractive forces at interfaces Ind.
Eng.Chem. 56 40
9
Nanotechnology 26 (2015) 145301 Z Liu and J Schroers
1. Introduction2. Materials and methods3. Processing map for BMG
nanomoulding4. Results and discussions4.1. Nanomoulding non-noble
BMGs by adding a wetting layer4.2. BMGs used as molds and moldable
materials
5. ConclusionsAcknowledgmentAppendixA.1. Oxidation ratesA.2.
Wetting properties of high temperature oil with templates and
BMGs
References