Enhanced oxidation resistance of magnesium nanorods · PDF fileEnhanced oxidation resistance of magnesium nanorods grown by glancing angle deposition Salih U. Baycaa,*, Mehmet F. Cansizoglub,

i n t e rn a t i o n a l j o u r n a l o f h y d r o g e n en e r g y 3 6 ( 2 0 1 1 ) 5 9 9 8e6 0 0 4

Avai lab le at www.sc iencedi rect .com

journa l homepage : www.e lsev ie r . com/ loca te /he

Enhanced oxidation resistance of magnesium nanorodsgrown by glancing angle deposition

Salih U. Bayca a,*, Mehmet F. Cansizoglu b, Alexandru S. Biris b,c, Fumiya Watanabe c,Tansel Karabacak b

aUniversity of Celal Bayar, Soma Vocational School, Soma, Manisa 45500, TurkeybUniversity of Arkansas at Little Rock, Department of Applied Science, Little Rock, AR 72204, USAcNanotechnology Center, University of Arkansas at Little Rock, Little Rock, AR 72204, USA

a r t i c l e i n f o

Article history:

Received 4 July 2010

Received in revised form

26 January 2011

Accepted 27 January 2011

Available online 31 March 2011

Keywords:

Magnesium

GLAD

Hydrogen storage

Oxidation

Thin film

TGA

* Corresponding author. Tel.: þ90 236 612 00E-mail address: [email protected] (S

0360-3199/$ e see front matter Copyright ªdoi:10.1016/j.ijhydene.2011.01.152

a b s t r a c t

Oxidation behavior of magnesium thin films and nanorods were investigated in the

temperature range of 25e550 �C by using thermal gravimetric analysis. Arrays of vertical

magnesium nanorods were deposited by the DC magnetron sputtering glancing angle

deposition technique, while the magnesium thin films were deposited using the same

system but at normal incidence. The morphologies and corresponding crystal structure of

the samples were analyzed by scanning electron microscopy, transmission electron

microscopy and X-ray diffraction methods, respectively. We report that the Mg thin films

showed oxidation induced weight gain starting from room temperature. On the other

hand, Mg nanorods did not show any indication of significant oxidation at temperatures

below 350 �C. Enhanced oxidation resistance of Mg nanorods was also confirmed by quartz

crystal microbalance measurements. At temperatures higher than 350 �C, Mg nanorods

started to get oxidized and their weight increased at a similar rate to that of Mg thin films.

We argue that reduced oxidation of Mg nanorods is mainly attributed to their single crystal

nature. Magnesium nanorods’ reduced oxidation can potentially play a key role in

hydrogen storage and gas sensing applications.

Copyright ª 2011, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights

reserved.

1. Introduction low and can lead to poor material properties especially at

Magnesium is among the lightest of all metals and as a result

it is used in an increased number of industrial applications

and processes as well as for products for which reduced

overall weight is essential, such as automobile parts, sporting

goods, and aerospace equipment [1]. Magnesium and its

alloys have also generated a high interest as advanced

materials for hydrogen storage and detection [2e11].

However, the oxidation resistance of magnesium is rather

63; fax: þ90 236 612 2002..U. Bayca).2011, Hydrogen Energy P

elevated temperatures. Several researchers have studied the

oxidation properties of magnesium and magnesium alloys

[12e15]. The oxide layer that can form at low temperatures

on the surface of magnesium and magnesium alloys is

generally in the form of MgO. The morphology of MgO layer is

porous and therefore it cannot act as an efficient barrier to

prevent the further oxidation of the metallic structure

[16e19]. Notable weight gain due to the oxidation process

was reported for Mg alloys at temperatures higher than

ublications, LLC. Published by Elsevier Ltd. All rights reserved.

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n en e r g y 3 6 ( 2 0 1 1 ) 5 9 9 8e6 0 0 4 5999

200 �C [20e22] for Mg ultrafine particles and Mg thin films [23]

at temperatures higher than 400 �C, respectively [24]. Studies

on the oxidation rate of magnesium and its alloys also

reported that they get further enhanced at temperatures

higher than about 400 �C [19,20,22e25]. However, there have

been reports indicating that thin films of Mg have been

observed to significantly get oxidized at temperatures as low

as 160 �C [26].

One of the most promising applications of the magnesium

based material systems is their use for solid-state hydrogen

storage environments because of their high storage capacity

(7.6 wt% for Mg), low material density, low cost, and vast

availability [27]. However, typical hydrogen absorption/

desorption temperatures of magnesium are elevated (�300 �C)due to the poor kinetics [28]. Recently, nanostructured

magnesium fabricated by the glancing angle deposition

(GLAD) technique [29e31] has been an attractive structure.

GLAD Mg nanostructures and nanostructures also incorpo-

rating catalyst additives [36e39] have been on the focus of

research for hydrogen storage and have significantly lower

absorption temperatures. Recently, Cansizoglu and Karabacak

reported significantly lower absorption temperatures of

hydrogen in Mg nanotrees produced by GLAD, where

a hydrogen storage value of w4.8 wt% was measured at

temperatures as low as 100 �C [40]. On the other hand, these

hydrogen storage experiments on nanostructured Mg either

with or without catalyst additives have been performed under

oxygen-free environments. There is a concern that high

surface area Mg nanostructures with small feature sizes can

seriously suffer from surface oxidation when exposed to

ambient or oxygen rich environments leading to low interac-

tion with gaseous hydrogen and poor absorption/desorption

kinetics. Therefore, there is a need to investigate the oxidation

properties of magnesium nanostructures in the presence of

oxygen, which will be quite critical especially for hydrogen

storage applications. It has been previously reported that

atomic hydrogen cannot diffuse trough the crystalline struc-

ture of the magnesium oxide films [26,41e43] which makes

MgO a diffusion barrier for H. In addition, Pranevicius et al. [44]

found that even small impurity levels of oxygen during

hydrogenation could result in the formation of a surface oxide

barrier.

The aim of the present study is to provide a detailed

investigation of oxidation properties of magnesium nanorods

fabricated by GLAD technique and compare this to that of

conventional magnesium thin films. Oxidation behavior of

magnesium samples has been studied using thermal gravi-

metric analysis (TGA) in dehumidified dry air supplied by

Airgas at temperatures ranging from room temperature

(w25 �C) to 550 �C. Mg nanorods and thin films have also been

characterized by X-ray diffraction (XRD) and scanning elec-

tron microscopy (SEM). The results of this study are believed

to be significant in elucidating the thermal behavior and

stability of GLAD-generated Mg nanostructures under various

temperature conditions. Such an understanding of the

behavior of various Mg structures in gaseous environments

could provide solutions to the current limitations faced by this

material for large scale usage in energy related applications

with a focus on their usage as the active component in high

efficiency hydrogen storage systems.

2. Experimental details

Glancing angle deposition (GLAD, also known as oblique angle

deposition) has been an effective technique to produce arrays

of various 3D nanostructures, in the form of nanosprings,

nanobeads, and nanorods [32e35]. During GLAD, the incident

beam of atoms is sent at a high angle (typically larger than 70�)as measured from the surface normal of a rotating or

a stationary substrate. Due to geometrical “shadowing effect”,

the incident vapor is preferentially deposited onto higher

surface features, which are better exposed to incident beam of

atoms. This preferential growth dynamic gives rise to the

formation of well-separated nanostructures, with dimensions

typically ranging from tens to hundreds of nanometers,

depending on the detailed growth parameters and material

being deposited.

In our experiment, samples of Mg thin films and GLAD

nanorods were deposited using a DC magnetron sputter

deposition unit. The substrates were polished alumina pieces

that were annealed at about 200 �C for 5 h under ambient

conditions before deposition in order to fully oxidize any

impurities present in the substrate. During the sputter depo-

sition of Mg, the base pressure was about 7.0 � 10�7 mbar, and

argon pressure was 2.7 � 10�3 mbar. The DC power used was

80 W and the distance between the center point of the

substrate holder and the Mg source (diameter of about 5 cm)

was about 20 cm. Conventional thin films samples were

deposited at an angle of q ¼ 0� and nanorods were deposited

using GLAD method at a deposition angle of q ¼ 84� (as

measured from the substrate surface normal). In addition,

both types of depositions were accompanied with an

azimuthal substrate rotation around the central substrate

surface normal axis with a speed of 1 rpm. Deposition rates

were measured to be 10.8 Ao/sec and 2.8 Ao/sec for normal

incidence and GLAD experiments, respectively. Silicon wafer

pieces have been used as substrates for the samples to be

analyzed for SEM and XRD. For TEM analysis, nanorods

deposited onto alumina substrates were scrapped off into

isopropyl alcohol and sonicated. Later this suspension was

placed on to holey carbon grids.

Thin films and nanostructured samples of Mg were

analyzed using SEM (FESEM-7000F, JEOL) for the morpholog-

ical investigation and XRD system (Bruker D8 discover) for the

crystal orientation (texture) investigation. TEM analysis was

performed by a JEOL JEM 2100F. Film thicknesses were

measured using the cross sectional SEM images.

Oxidation experiments of the Mg nanorods and Mg thin

films were performed by using TGA (METLER TOLEDO 815e) at

temperatures ranging 25e550 �C in dry air. A heating rate of

3 �C/min was used for the experiments. The weight change

was recorded as a function of time (for isothermal TGA

measurements) or temperature (for variable temperature

measurements). Isothermal TGA experiments were carried

out at constant temperatures of 150, 250, 300 and 350 �C. Thefinal mass values presented in this work were obtained by

subtracting the bare alumina weight data from that of coated

alumina samples of identical dimensions.

As it will be discussed in the following section, during some

of the TGA experiments, we observed a weigh loss for our

i n t e rn a t i o n a l j o u r n a l o f h y d r o g e n en e r g y 3 6 ( 2 0 1 1 ) 5 9 9 8e6 0 0 46000

nanorod samples. Mg has a high vapor pressure and may

show evaporation around 400 �C [23,52]. The origin of weight

loss although not entirely clear is believed to be due to either

the sublimation of Mg (due to magnesium’s high vapor pres-

sure at low temperatures) or the evaporation of water vapor

adsorbed onto the high surface area of the nanorod samples.

In order to investigate this observation, we utilized a new

quartz crystal microbalance (QCM) system that enabled us to

measure possible weight changes of the Mg nanorod coatings

under vacuum (i.e. free of water vapor on the surface of

nanorods) conditions. QCM crystals are typically used in

conventional thin film thickness monitoring units and are

sensitive to very small changes in the mass of a coating as

small as in the scale of nanograms. To perform the mass

change detection vs. temperature experiments using QCM

crystals as a substrate for our nanorod and thin film coatings,

we have developed a custom made chamber that can operate

at temperatures ranging from room temperature up to 500 �Cas well as at pressures from vacuum (10�3 bar) up to 50 bar.

The chamber has a dual QCM sample holder configuration.

One holder is reserved for a bare QCM crystal used for the

control sample while a 2nd holder is for the actual sample

measurements.We depositedMg nanorods onto QCM crystals

using GLAD as described above. Before Mg deposition, the

oscillation frequency of the bare QCM crystal wasmeasured at

atmospheric pressure and at a pressure of 10�3 bar while at

room temperature in order to calculate the pressure effect on

the crystal. The total Mg deposition amount on the crystal was

calculated comparing the frequency values before and after

Fig. 1 e Top view (top row) and cross section view (bottom row) s

of thickness about a) 250 nm and b) 1700 nm that were deposit

deposition. The sample to be tested was placed inside the

chamber and evacuated for 2 days at 10�3 bar of pressure. This

enabled us to remove the water present on the sample surface

and inside the chamber. Once the frequency stabilization was

achieved in both QCM crystal oscillations, the temperature

was increased from room temperature to 250 �C in 3 h at

a steady rate. Then the system was cooled down to room

temperature. The initial and finalmass values before and after

QCM tests were compared under the same conditions in order

to calculate the weight percentage changes in the mass of Mg

nanorod coatings.

3. Results and discussion

Both thin films and nanorod-structured samples of Mg were

grown with various thicknesses in order to understand their

morphological and crystal orientation evolution. Fig. 1 shows

the top view and cross sectional SEM images of some of theMg

thin film samples deposited at normal incidence. Mg thin

films quickly form a rough morphology during the early

deposition stages (Fig. 1a). As the film gets thicker (Fig. 1b),

a highly textured microstructure incorporating a cheese like

fine porosity was found to develop. This porosity might be due

to the limited mobility of the Mg atoms during the sputter

deposition process, while the substrate temperatures are

typically low (i.e. lower than about 50e100 �C), which is

correlated to lower adatommobility values [45] Sputtered thin

films also show a columnar microstructure as can be seen

canning electron microscopy (SEM) images of Mg thin films

ed at normal incidence.

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n en e r g y 3 6 ( 2 0 1 1 ) 5 9 9 8e6 0 0 4 6001

from the cross sectional SEM images in Fig. 1 (bottom row). In

addition, the XRD analysis presented in Fig. 2a for the sputter-

deposited Mg thin films show a pattern that promotes

a texture formation in the (002) direction which is indepen-

dent of the film thickness value.

On the other hand, GLAD Mg nanorods show a different

morphology than that of the Mg thin films. Unlike continuous

Mg films, as can be seen in Figs. 4 and 5, the nanorods form in

the shape of isolated columnar arrays with an average diam-

eter of about 120 nm, and separated by gaps in the range of

a few tens of nanometers. In addition, the XRD patterns of the

GLAD nanorods differed significantly from that of the thin

films deposited at normal incidence (Fig. 2b). Contrary to

strongly (002) oriented Mg thin films, the crystallographic

analysis of the nanorods presents peaks corresponding to the

crystalline directions of (101), (102) and (103) in addition to the

(002) peaks. These peaks other than (002) become more

intense when the nanorods get longer, while the change in

(002) orientation is minimal for thicknesses larger than

900 nm. The formation of crystal orientations other than the

preferential (002) texture in sputtered GLAD Mg nanorods is

Fig. 2 e X-ray diffraction (XRD) profiles of (a) Mg thin films

deposited by magnetron sputtering at normal incidence

and (b) Mg nanorods deposited by glancing angle

deposition (GLAD) show the evolution crystal orientation

as a film gets thicker. Line profiles have been off-set in the

y-axis for clarity.

believed to be due the shadowing effect and limited adatom

mobility of these crystal orientations [46,47]. During the initial

stages of GLAD, Mg atoms can form randomly oriented

islands. Some of these islands have higher vertical growth

rates due to the lower adatommobility on their crystal planes

(e.g. due to a more atomically rough crystal plane), where

others have faster lateral growth rate due to higher adatom

mobilities (e.g. on a atomically smooth closed packed crystal

plane like for example on Mg (002)). Due to shadowing effect,

obliquely incident atoms preferentially deposit on higher

islands of lower adatom mobility and higher vertical growth

rates, leading to nanorods with energetically un-preferential

crystal orientations. However, in our case it seems that

roughness of the GLAD Mg samples, and therefore the shad-

owing effect couldn’t develop strong enough at thicknesses

smaller than 900 nm, leading to still a dominant (002) texture.

As the nanorods get longer and surface gets rougher, shad-

owing effect can become more effective to promote the

growth of nanorods with other crystalline orientations.

TEM analysis on the GLAD Mg nanorods revealed a single

crystal structure (Fig. 3). In Fig. 3, single set of spots in the

selective area diffraction (SAD) pattern and high resolution

TEM image of a single nanorod both show that individual

nanorods of Mg have a single crystal formation. SAD pattern

corresponds to the hexagonal crystal structure on the sidewall

where the diffraction pattern was taken. Single crystal nature

of the individual Mg nanorods is similar to the previously

reported results on metallic nanorods deposited by GLAD

[46,48e50]. Single crystal structure is an important property in

explaining the oxidation behavior of the Mg nanorods.

Variable temperature TGA result of a conventional Mg thin

film is shown in Fig. 6. TGA datawas recorded forMg thin films

with a thickness of about 1100 nm. Actual mass values were

normalized to the initial value of the Mg thin film. These films

were found to continue gain weight due to the oxidation for

temperatures of up to 550 �C, with the exception of a small

plateau between 250 and 350 �C. Total change in the weight

was as high as about 85 wt% (weight percent) indicating an

oxygen rich stoichiometry compared to the crystalline MgO.

The rate of weight gain slightly slowed down as the temper-

ature increased from room temperature. This parabolic

behavior takes place due to the slower oxidation rate of Mg,

which originates from the diffusion barrier properties of the

oxide layer [23,52]. As the oxide layer gets thicker, it impedes

the oxygen from passing through and reaching the metallic

magnesium underneath the oxide layer [22,51]. The plateau

observed between temperatures of about 250e350 �Cmay also

be ascribed to the competing reactions of already slowed

down oxidation rate (i.e. smaller weight gain) and evaporation

of water vapor trapped in the small pores of Mg film (i.e.

weight loss).

The weight change profile of Mg nanorods (1050 nm in

height and morphology similar to the one shown in Fig. 5b),

for variable temperature TGA has a quite different behavior

than that of theMg films. In Fig. 6 it is observed that theweight

of the Mg nanorods decreased by about 2 wt% as the

temperature increased from 25 to 150 �C. This slight weight

lost is believed to be the result of the evaporation of water

vapor adsorbed on the surface of the magnesium nanorods.

The possibility of sublimation of magnesium has been

Fig. 3 e Transmission electron microscopy (TEM) results of an individual GLAD Mg nanorod are shown: (a) TEM selective

area diffraction (SAD) pattern showing an hcp crystal structure, (b) High resolution lattice imaging, and (c) TEM image.

Fig. 4 e Top view (a) and cross section view (b) SEM images

of Mg nanorods by GLAD of thickness about 200 nm.

i n t e rn a t i o n a l j o u r n a l o f h y d r o g e n en e r g y 3 6 ( 2 0 1 1 ) 5 9 9 8e6 0 0 46002

discarded after our QCM tests performed at temperatures

from 25 to 250 �C (described above) where no measurable

mass loss has been observed under water vapor free condi-

tions. Fig. 7, showed that the mass of the nanorods remained

relatively constant between temperatures from 150 to 350 �C.Above 350 �C, magnesium nanorods begin to gain weight with

increasing temperature up to 550 �C. This weight gain of about

6 wt%, corresponds to an oxidation process which is much

smaller than that of conventional Mg film (w85 wt%). We also

have recorded a weight gain due to the structural oxidation of

the Mg nanorod samples of about 12 wt% during the QCM

measurements. Although the QCM experiments were per-

formed under vacuum (10�3 bar) we believe that this level of

vacuum was not high enough to avoid oxidation.

Reduced oxidation values of Mg nanorods compared to

conventional thin films are believed to mainly originate from

their single crystal property. In a single crystal structure, there

are no grain boundaries present and therefore the diffusion

rate of oxygen is significantly reduced. In addition, possible

crystal orientation on the sidewalls of the nanorods might

present a less reactive plane for oxygen penetration and oxide

formation. For example, Mg basal (001) plane is about 25%

more reactive than (100) prismatic plane [23] in oxidation. On

the other hand, some of theMg nanorods have un-preferential

crystal orientations (Fig. 2) with non-closed pack crystal

planes, which could normally enhance diffusion of oxygen,

yet it seems that single crystal property dominated the

process leading to enhanced resistance to oxidation.

In addition, isothermal TGA analyses were carried out to

examine weight changes of the Mg nanorods at constant

temperatures of 150, 250, 300 and 350 �C. The TGA results for

these temperatures are shown in Fig. 7 for up to 180 min. All

the isothermal lines show a similar behavior. At the initial

times there is a mass loss mainly due to the evaporation of

water vapor followed by a slightly fluctuating behavior espe-

cially at temperatures higher than 150 �C. The fluctuation at

later times can originate from the competition between the

Fig. 5 e Top view (top row) and cross section view (bottom row) SEM images of Mg nanorods by GLAD of thickness about a)

900 nm and b) 1050 nm.

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n en e r g y 3 6 ( 2 0 1 1 ) 5 9 9 8e6 0 0 4 6003

weight gain due to the low oxidation of nanorods and evap-

oration of trapped water vapor in the smaller gaps of the

structures. However, the final weight gain after the initial

release of the water vapor is still insignificant (max value

0 100 200 300 400 500 600

1.0

1.2

1.4

1.6

1.8

Mg nanorods

).u.a( ssam dezila

mroN

Temperature (oC)

Mg thin film

Fig. 6 e Thermogravimetric analysis (TGA) results of

magnesium thin film andmagnesium nanorods are shown

as a function of temperature. Actual mass values of the

coatings were normalized to their initial values for

comparison.

being about 3 wt% at 300 �C). These results also show the

reduced oxidation property of Mg nanorods produced by

GLAD, which is consistent with the variable temperature TGA

and QCM results discussed above.

Fig. 7 e Isothermal TGA plots of magnesium nanorods at

150, 250, 300, and 350 �C show the change in dynamic

weight of nanorods as a function of time. Actual mass

values of the coatings were normalized to their initial

values for comparison.

i n t e rn a t i o n a l j o u r n a l o f h y d r o g e n en e r g y 3 6 ( 2 0 1 1 ) 5 9 9 8e6 0 0 46004

4. Conclusions

In this study, we report that Mg thin films showed significant

oxidation with weight gains of as high as about 85 wt%, in

temperature ranges from 25 to 550 �C. On the other hand, the

oxidation of Mg nanorods grown by GLAD was minimal with

a weight change of as little as 6 wt% in similar conditions.

Our QCM studies also further confirmed the reduced oxida-

tion of Mg nanorods. Because of their nanostructured large

surface area morphology, one might normally expect an

enhanced oxidation in nanorods of Mg. The unusual resis-

tance of Mg nanorods to oxidation is believed to originate

mainly from their single crystal nature where no grain

boundaries for fast-diffusion of oxygen are present. In addi-

tion, possible crystal orientation on the sidewalls of the

nanorods might present less reactive planes for oxygen

penetration and oxide formation. These results suggest that

GLAD Mg nanorod arrays with or without catalyst additives,

or their alloys can be attractive candidates especially for

hydrogen storage applications where reduced oxidation

property is highly desirable.

r e f e r e n c e s

[1] Kurth M, Graat PCJ, Carstanjen HD, Mittemeijer EJ. Surfaceand Interface Analysis 2006;38:931e40.

[2] Zaluska A, Zaluski L, Strom-Olsen JO. Journal of Alloys andCompounds 1999;288:217e25.

[3] Dehouche Z, Goyette J, Bose TK, Hout J, Schulz R. NanoLetters 2001;1:175e8.

[4] Huot J, Pelletier JF, Lurio LB, Sutton M, Schulz R. Journal ofAlloys and Compounds 2003;348:319e24.

[5] Oelerich W, Klassen T, Bormann R. Materials Transactions2001;42:1588e92.

[6] Barkhordarian G, Klassen T, Bormann R. Journal of Alloysand Compounds 2004;364:242e6.

[7] Au M. Materials Science and Engineering B 2005;117:37e44.[8] Chen J, Sakai T, Kitamura N, Takeshita HT, Kuriyama N.

Journal of the American Chemical Society 2001;123:6193e4.[9] Hanada N, Ichikawa T, Fujii H. The Journal of Physical

Chemistry B 2005;109:7188e94.[10] Yao X, Wu C, Du A, Lu GQ, Cheng H, Smith SC, et al. The

Journal of Physical Chemistry B 2006;110:11697e703.[11] Gutfleisch O, Schlorke-de Boer N, Ismail N, Herrich M,

Walton A, Speight J, et al. Journal of Alloys and Compounds2003;356:598e602.

[12] Cohen MS. Acta Metallurgica 1960;8:356e61.[13] Addiss Jr RR. Oxidation of magnesium single crystals and

evaporated films. Acta Metallurgica 1963;11:129e35.[14] Schwoebel PR, Brodie I. Journal of Vacuum Science and

Technology B 1995;13:1391e410.[15] Shih TS, Liu JB, Wei PS. Materials Chemistry and Physics

2007;104:497e504.[16] Zeng X, Wang Q, Lu Y, Ding W, Zhu Y, Zhai C, et al. Materials

Science and Engineering A 2001;301:154e61.[17] Wang XM, Zeng XQ, Wu GS, Yao SS, Li LB. Applied Surface

Science 2007;253:9017e23.[18] Ding W, Wang X, Zeng X, Wu G, Yao S, Lai Y. Materials

Letters 2007;61:1429e32.

[19] YouBS, ParkWW,Chung IS. ScriptaMaterialia 2000;42:1089e94.[20] Czerwinski F. Acta Materialia 2002;50:2639e54.[21] Czerwinski F. Journal of the Minerals, Metals and Materials

Society 2004;56:29e31.[22] Smeltzer WW. Journal of the Electrochemical Society 1958;

105:67e71.[23] Schwoebel RL. Journal of Applied Physics 1963;34:2776e83.[24] Liu T, Zhang Y, Li X. Scripta Materialia 2003;48:397e402.[25] Christopher J, Murthy IAPS, Swamy CS. Thermochimica Acta

1990;164:191e8.[26] Ostenfeld CW, Johansson M, Chorkendorff I. Surface Science

2007;601:1862e9.[27] Stioui M, Grayevsky A, Resnik A, Shaltiel D, Kaplan N. Journal

of Less Common Metals 1986;123:9e24.[28] Vijay R, Sundaresan R, Maiya MP, Murthy SS. Journal of

Alloys and Compounds 2006;424:289e93.[29] Tang F, Parker T, Li HF, Wang GC, Lu TM. Journal of

Nanoscience and Nanotechnology 2007;7:3239e44.[30] Tang F, Parker T, Li H-F, Wang G-C, Lu T-M. Nanotechnology

2008;19:465706.[31] He Y-P, Zhao Y-P. Crystal Growth & Design 2010;10:440e8.[32] Karabacak T, Wang GC, Lu TM. Journal of Vacuum Science

and Technology A 2004;22:1778e84.[33] Young NO, Kowal J. Nature 1959;183:104e5.[34] Robbie K, Brett MJ, Lakhtakia A. Nature 1996;384:616.[35] Karabacak T, Lu TM. Handbook of Theoretical and

computational nanotechnology. In: Rieth M, Schommers W,editors. New York: Americans Scientific Publishers; 2005.

[36] He Y, Zhao Y. Journal of Alloys and Compounds 2009;482:173e86.

[37] He Y, Zhao Y. Nanotechnology 2009;20:204008.[38] He Y, Zhao Y. Physical Chemistry Chemical Physics 2009;11:

255e8.[39] He Y, Zhao Y, Huang L, Wang H, Composto RJ. Applied

Physics Letters 2008;93:163114.[40] Cansizoglu M, Karabacak T. Hydrogen storage materials. In:

Akiba E, Tumas W, Chen P, Fichtner M, Zhang S, editors;2010. Materials Research Society Symposia Proceedings. vol.1216E, Warrendale, PA, USA, 1216-W05-03.

[41] Ostenfeld CW, Davies JC, Vegge T, Chorkendorff I. SurfaceScience 2005;584:17e26.

[42] Ostenfeld CW, Chorkendorff I. Surface Science 2006;600:136e1368.

[43] Hjort P, Krozer A, Kasemo B. Journal of Alloys andCompounds 1996;237:74e80.

[44] Pranevicius L, Wirth E, Milcius D, Lelis M, Pranevicius LL,Kanapickas A. Surface and Coatings Technology 2009;203:998e1003.

[45] Cansizoglu MF, Watanabe F, Wang PI, Karabacak T. MaterialsResearch Society Symposia Proceedings 2007;1042:S04e6.Warrendale, PA, USA.

[46] Tang F, Karabacak T, Morrow P, Gaire C, Wang GC, Lu TM.Physical Review B 2005;72:165402.

[47] Karabacak T, Mallikarjunan A, Sing JP, Ye D, Wang GC,Lu TM. Applied Physics Letters 2003;83:3096e8.

[48] Karabacak T,Wang PI, Wang GC, Lu TM. Materials ResearchSocietySymposiaProceedings2004;788:75.Warrendale,PA,USA.

[49] Karabacak T, Wang PI, Wang GC, Lu TM. Thin Solid Films2005;493:293e6.

[50] Djurfors B, Brett MJ, Ivey DG. Materials Research SocietySymposia Proceedings 2002;749:83. Warrendale, PA, USA.

[51] Fournier V, Marcus P, Olefjord I. Surface and InterfaceAnalysis 2002;34:494e7.

[52] R.R. Addiss Jr., Tech Report1. Project NR 036e038, ContractNonr-401(31), 1958; Acta, Met11, 129, 1963.