Morphology and preferred orientation of pulse electrodeposited magnesium

Article abstract of DOI:10.1149/1.3298883

A nanocrystalline magnesium material with a high specific surface area is expected to react rapidly and reversibly with hydrogen gas to yield magnesium hydride, a hydrogen storage medium. In this paper, the feasibility of the synthesis of magnesium materials for hydrogen storage applications by pulse electrodeposition of magnesium from ethereal electrolytes containing Grignard reagents was investigated. Deposition onto flat stainless steel electrodes established that, as in dc deposition, the morphology of the deposits varied widely with electrolyte composition and charge density. Irregular, nanocrystalline magnesium films were formed at low current density (0.4 mA cm^-2) and low charge density (1 C cm^-2) using butylmagnesium chloride electrolytes in dibutyl diglyme, while at a higher current density (15 mA cm^-2) in tetrahydrofuran, dense films were favored.

Full text of DOI:10.1149/1.3298883

Journal of The Electrochemical Society, 157 ͑4͒ E45-E49 ͑2010͒
E45
0
013-4651/2010/157͑4͒/E45/5/$28.00 © The Electrochemical Society
Morphology and Preferred Orientation of Pulse
Electrodeposited Magnesium
,
z
Rosalind J. Gummow* and Yinghe He
School of Engineering and Physical Sciences, James Cook University, Townsville 4811, Australia
A nanocrystalline magnesium material with a high specific surface area is expected to react rapidly and reversibly with hydrogen
gas to yield magnesium hydride, a hydrogen storage medium. In this paper, the feasibility of the synthesis of magnesium materials
for hydrogen storage applications by pulse electrodeposition of magnesium from ethereal electrolytes containing Grignard reagents
was investigated. Deposition onto flat stainless steel electrodes established that, as in dc deposition, the morphology of the deposits
varied widely with electrolyte composition and charge density. Irregular, nanocrystalline magnesium films were formed at low
current density ͑0.4 mA cm−2͒ and low charge density ͑1 C cm ͒ using butylmagnesium chloride electrolytes in dibutyl
−2
−
2
diglyme, while at a higher current density ͑15 mA cm ͒ in tetrahydrofuran, dense films were favored.
2010 The Electrochemical Society. ͓DOI: 10.1149/1.3298883͔ All rights reserved.
©
Manuscript submitted October 12, 2009; revised manuscript received December 10, 2009. Published February 9, 2010.
Magnesium hydride is an attractive material for solid-state hy-
drogen storage due to its high gravimetric energy density, simplicity,
high reversibility, and low cost. However, the speed of hydrogen
insertion and extraction from magnesium hydride is slow and the
thermodynamics is unfavorable. The kinetic restriction has been
largely overcome by the use of fine-grained, high specific surface
area powders, typically generated by milling large-grained materials
to a fine-grained product, and by the inclusion of suitable
magnesium specifically for the purpose of optimizing hydrogen stor-
1
8
age properties. Tatiparti and Ebrahimi studied the dc electrodepo-
sition of dendritic aluminum–magnesium deposits from ethereal so-
lutions for hydrogen storage applications using current densities
−
2
from 60 to 150 mA cm . Details of the hydrogen storage proper-
1
9
ties of their deposits have been published recently.
In this study, the direct synthesis of electrodeposited magnesium
with particle and grain sizes in the desired range by an appropriate
selection of deposition conditions was investigated. The effect of the
variation in the current density, electrolyte composition, and electro-
lyte concentration was studied.
1
-4
catalysts. Although milling produces particles with multiple de-
fects and grain boundaries near the surface, which favor nucleation
3
of the hydride, the milling process is time-consuming, requires a
high energy input, leads to the incorporation of impurities, and gives
an irregular product with a wide distribution of particle sizes.
Much attention has recently been directed to the deposition of
magnesium thin films using a variety of techniques including dc
magnetron sputtering, electron-beam physical vapor deposition,
plasma sputter, pulsed laser deposition, and oblique angle
Experimental
The Grignard reagents,
3 M methylmagnesium chloride
͑CH3MgCl͒ in THF, 2 M butylmagnesium chloride ͑BuMgCl͒ in
THF, and 1.45 M BuMgCl in dibutyl diglyme ͑DBG͒, and anhy-
5
-9
deposition. An alternative route for the direct deposition of fine
particles and thin films of magnesium is pulse electrodeposition.
This technique has been widely applied for the generation of metal-
drous aluminum chloride ͑AlCl Ͼ 99%͒ were obtained from
3
Sigma-Aldrich. All experiments were carried out in a glove box
under a protective ultrahigh purity argon atmosphere ͑BOC gas͒.
A two-electrode glass cell was used for the electrodeposition
experiments. All glassware were dried in a drying oven before use.
The cathode consisted of a 316 stainless steel ͑5 ϫ 15 ϫ 1 mm͒
sheet, polished with SiC paper, and cleaned and dried before use.
The anode consisted of a magnesium ribbon ͑Ͼ99% purity, Sigma-
Aldrich͒. The anode surface was scraped with a nylon scourer in the
glove box to remove any surface oxidation layer before loading into
the electrochemical cell. The electrolyte volume was approximately
5 mL. The deposition experiments were performed under constant
current conditions using a purpose built constant current power sup-
ply coupled to a signal generator to generate the pulse train. Typical
1
0,11
lic nanoparticles in both aqueous and nonaqueous electrolytes.
Precise control of grain size within a narrow range has been dem-
onstrated by a variation in the deposition parameters. This technique
has the potential for industrial application and the generation of
large amounts of material at low cost.
The electrodeposition of magnesium is not possible in aqueous
solutions due to the high electronegativity of magnesium, but it is
possible in organic solutions and ionic liquids. Several investigators
have successfully used dc electrodeposition to deposit magnesium
from ethereal solutions, and the size and morphology of electrode-
posited magnesium particles varies dramatically, depending on the
1
2-15
deposition conditions.
on the pulse electrodeposition of magnesium in ethereal electrolytes.
There have been only two papers to date
deposition parameters used were t = 2 ms and toff = 8 ms with a
on
1
6
−
2
Hassen et al. compared dc and pulse electrodeposited layers of
magnesium on mild steel substrates for corrosion protection using
electrolytes of methyl magnesium chloride in tetrahydrofuran ͑THF͒
and pulse electrodeposition resulted in a finer grain structure than dc
current density of 20–150 mA cm . The deposition time was typi-
cally 15–60 min. Additional measurements were performed using a
Princeton AG&G Versastat 3 potenstiostat/galvanostat in the fast
galvanostatic pulse mode. Deposited samples were washed in anhy-
drous THF ͑Aldrich͒ and vacuum dried before analysis.
1
7
deposition with grains of about 3 ␮m. Haas and Gedanken re-
cently reported the synthesis of magnesium nanopowders with an
average grain size of 4–6 nm by a sonoelectrochemical route, which
couples the electrochemical deposition on a Ti sonoelectrode surface
with a sonic pulse to release the electrodeposit from the surface. The
very small grain sizes reported for their products raise the possibility
of tailoring fine-grained magnesium deposits to optimize their use-
fulness as hydrogen storage materials by fine-tuning the deposition
conditions. Little work has been reported on the electrodeposition of
Powder X-ray diffraction ͑XRD͒ patterns were recorded using a
Siemens D5000 diffractometer with copper K␣ radiation. Samples
were sandwiched between two layers of polyimide Kapton tape
͑ChemTools͒ in the glove box to prevent exposure of the samples to
the atmosphere during XRD analysis. Grain sizes were determined
2
0
from the XRD peak broadening using the Scherrer equation after
2
1
peak fitting with the XFIT program ͑freeware͒. Pearson VII ͑PVII͒
functions were used for the fitting of the XRD line profiles. Correc-
tion for instrumental broadening was made using data from a lan-
thanum hexaboride ͑LaB ͒ standard ͑NIST 660a͒.
6
The morphology of the sample surfaces was examined with a
JEOL JSM-5410LV scanning electron microscope.
*
Electrochemical Society Active Member.
E-mail: rosalind.gummow@jcu.edu.au
z
Downloaded on 2016-06-06 to IP 144.82.108.120 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

E46
Journal of The Electrochemical Society, 157 ͑4͒ E45-E49 ͑2010͒
Table I. A summary of the electrodeposition parameters used in this study. iav = average current density and ipeak = peak current density. All
samples were deposited with ton = 2 ms, toff = 8 ms (duty cycle = 20%), and temperature = 25°C.
͓
AlCl3͔
͑M͒
iav
ipeak
Charge density
͑C cm ͒
͑mA cm−2͒
͑mA cm−2͒
−2
Sample
Electrolyte
Mg1
Mg2
Mg3
Mg4
Mg5
Mg6
Mg7
Mg8
3 M CH MgCl in THF
2 M BuMgCl in THF
1.45 M BuMgCl in DBG
0
0
0
15
15
0.4
15
15
30
3
75
75
2
75
75
150
15
15
54
54
1
54
54
54
0.18
1.8
3
1.5 M CH MgCl in THF
0.125
0.167
0.167
0.167
0.167
3
1 M CH MgCl in THF
3
1 M CH MgCl in THF
3
1 M CH MgCl in THF
3
1 M CH MgCl in THF
3
3
Results and Discussion
orientation of the Mg crystallites parallel to the 002 crystallographic
planes, indicating that crystal growth occurs along the c-axis. This
Pulse electrodeposition of magnesium.— Table I gives a sum-
mary of the deposition conditions of the samples prepared for this
16
agrees well with the results of Hassen et al.
The experiment was repeated using two other electrolytes, 2 M
study. An initial sample, Mg , was electrodeposited using a 3 M
1
BuMgCl in THF ͑Mg ͒ and 1.45 M BuMgCl in DBG ͑Mg ͒ for
2
3
solution of CH MgCl as electrolyte and deposition conditions as
3
comparison. DBG was selected for evaluation as it has a far lower
vapor pressure than THF and is therefore preferred from a safety
point of view. Due to the high viscosity and low ionic conductivity
of the DBG electrolyte, the maximum possible current density ob-
indicated in Table I. The deposit surface was matte and white. A
small section of the electrode was covered with very fine gray den-
drites. The XRD pattern of the white deposit in the range 30–80° 2␪
͑
Fig. 1a͒ shows sharp peaks, which can be indexed to the crystallo-
−2
tained was 0.4 mA cm . An enlarged plot of the powder XRD
graphic reflections of hexagonal magnesium ͑space group
P63/mmc͒, indicating a coarse-grained polycrystalline deposit. Ad-
ditional peaks from the stainless steel substrate are indicated in Fig.
. The ratio of the ͑002͒/͑101͒ peak intensities ͑Table II͒ for this
sample is 2.1 compared to 0.26 for a standard Mg sample. The
enhancement of the intensity of the 002 peak suggests a preferential
patterns for the 30–40° 2␪ region is shown for each electrolyte in
Fig. 1b-d. In each case, characteristic magnesium peaks were ob-
tained, but in the latter two electrolytes, there was no evidence of
significant preferential orientation, as seen from the ratios of the
1
͑
002͒:͑101͒ peak intensities in Table II. The values of the grain sizes
of the deposits calculated from the Scherrer formula based on the
line peak broadening of the ͑101͒ peak at 36.61° 2␪, after the cor-
rection for instrumental broadening, are given in Table II for each of
the samples. However, the Scherrer formula does not take into ac-
count the broadening of the XRD peaks due to microstrain and may
therefore only be considered to give approximate values for grain
size.
Scanning electron micrographs of the samples are shown in Fig.
2
a-c. Mg shows a dense coating consisting of large particles of the
1
order of 10 ␮m across, which appear to be largely oriented parallel
to the electrode surface. This confirms the preferential orientation
indicated by the XRD pattern ͑Fig. 1a͒. The thickness of the film
was approximately 70 ␮m, and the efficiency of the deposition was
7
5%. Mg deposited using BuMgCl in THF electrolyte is typical of
2
the crystalline deposits reported by other investigators for dc depo-
2
2
sition of magnesium in ethereal electrolytes. Randomly oriented,
pyramidal particles approximately 3–5 ␮m across are visible and
the surface is not completely covered by the film. In this case, the
film was approximately 60 ␮m thick and the deposition efficiency
was 57%. Mg shows a completely different morphology with an
3
irregular surface, with particles of the order of 2 ␮m across. This
Table II. Average crystallite size of electrodeposited magnesium
calculated using the Scherrer formula and the value of the (002)/
(
101) peak intensities.
Average
crystallite
size
Corrected peak
͑002͒/͑101͒
peak
width
Sample
͑°͒
͑nm͒
intensities
Mg1
Mg2
Mg3
Mg4
Mg5
Mg6
0.03
0.05
0.13
0.02
0.08
0.05
Ͼ200
160
60
Ͼ200
140
180
2.1
0.2
0.1
0.3
0.2
1.4
Figure 1. Powder XRD patterns of electrodeposited magnesium ͑a͒ Mg1
indexed to space group P63/mmc. ࡗ indicates peaks from the stainless steel
substrate and magnified images of 30–40° 2␪ region for ͑b͒ Mg , ͑c͒ Mg ,
1
2
and ͑d͒ Mg3.
Downloaded on 2016-06-06 to IP 144.82.108.120 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

Journal of The Electrochemical Society, 157 ͑4͒ E45-E49 ͑2010͒
E47
(
a)
b)
1
0 µm
Figure 3. Powder XRD patterns of electrodeposited magnesium samples ͑a͒
Mg , ͑b͒ Mg , and ͑c͒ Mg showing the effect of AlCl addition and current
4
5
6
3
density.
charge density of 0.3 C cm−2 per pulse. Their deposits would there-
fore have been made in the porous regime found at a low charge
density.
These three samples confirm that the choice of electrolyte and
deposition conditions has a profound effect on the morphology and
crystallite size of pulse electrodeposited magnesium. It is tempting
to speculate as to which morphology may be expected to be optimal
for hydrogen storage applications and whether the preferred orien-
tation is advantageous in any way. Dense, large-grained films like
(
10 µm
Mg would not be expected to perform well as hydrogen storage
1
media as the penetration of hydrogen would be limited. The prefer-
ential orientation of the deposit actually appears to facilitate dense
packing and minimizes the formation of voids, further limiting the
2
3
hydrogen flow. However, the results reported by Leon et al. for
hydrogen storage in 30 ␮m thick, preferentially orientated magne-
sium films formed by thermal evaporative deposition appear to in-
dicate that the ͓002͔ preferential orientation facilitates hydrogen ab-
sorption and desorption, in this case. The evaporatively grown films,
however, have a very fine grain structure and columnar growth mor-
phology, which is very different from the samples under consider-
ation here.
Mg has numerous voids between particles, probably due to the
2
less efficient packing of randomly oriented particles and a finer grain
(
c)
5 µm
structure compared with Mg . It is expected that both these factors
1
would be advantageous for hydrogen absorption. An irregular, fine-
grained sample with a high specific surface area like Mg3 may,
however, be expected to offer the best morphology, as it would
allow the gas to penetrate readily into the bulk of the film.
Figure 2. SEM micrographs of ͑a͒ Mg , ͑b͒ Mg , and ͑c͒ Mg showing the
1
2
3
dramatic change in morphology with different deposition conditions.
Effect of AlCl addition.— When AlCl is added to alkylmagne-
3
3
sium halide electrolytes, the ionic conductivity increases and com-
plex ionic species form in solution, resulting in improved voltage
2
2
stability. These enhanced electrolytes offer the possibility of in-
creased deposition efficiency at a high current density while avoid-
ing problems of electrolyte decomposition at high overvoltages.
Two samples, Mg4 and Mg5, were prepared with increasing ra-
tios of AlCl3 to CH3MgCl to assess the effect of AlCl3 addition on
the morphology and grain size of the pulse electrodeposited magne-
sium. Both samples gave white, matte deposits, with no dendrites.
Figure 3a and b shows the XRD patterns in the range 30–40° 2␪ for
samples with a ͓CH3MgCl͔:͓AlCl3͔ ratio of 12:1 and 6:1, respec-
tively. The XRD patterns show characteristic magnesium XRD
peaks indicating crystalline magnesium and no evidence of prefer-
ential orientation, as determined from the ratio of the ͑002͒:͑101͒
peak intensities ͑Table II͒. No aluminum peaks were present in the
XRD pattern, but this alone does not exclude the possibility of some
may be due to the much lower charge density in this case
−
2
15,22
͑
1 C cm ͒. Aurbach et al.
reported that the dc deposition of
magnesium from ethereal solutions results in a porous deposit, at
low charge density, with a different morphology from the pyramidal
crystals formed at higher charge density, although the morphology
of their sample produced using dc deposition is somewhat different
from that found in the present study. However, using dc deposition,
−
2
the porous morphology only persists below 0.5 C cm , after
which the angular crystallites form. It therefore appears that the
pulse deposition technique may extend the charge density range in
which the porous deposits are formed in DBG solutions. The sono-
1
7
electrochemical products generated by Haas and Gedanken, with
reported grain sizes from XRD line broadening of 25 nm, were
−
2
24
produced using 600 s pulses at 0.5 mA cm , corresponding to a
Al codeposition. However, Viestfrid et al. reported that, based on
Downloaded on 2016-06-06 to IP 144.82.108.120 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

E48
Journal of The Electrochemical Society, 157 ͑4͒ E45-E49 ͑2010͒
1
0 µm
(
(
a)
b)
(
(
(
a)
b)
c)
5 µm
5 µm
5 µm
1
0 µm
Figure 4. SEM micrographs of ͑a͒ Mg and ͑b͒ Mg .
4
5
the atomic absorption spectroscopy results, no codeposition of Al
occurs for Grignard reagent/Al salt ratios greater than 1/3. The cal-
culated values of the crystallite size are given in Table II. For Mg4
with a 12:1 ratio of ͓CH MgCl͔:͓AlCl ͔, there is no reduction in the
3
3
crystallite size, but as the proportion of AlCl in the solution in-
3
creases to 6:1, there is an increase in the XRD peak width and a
decrease in the crystallite size of the magnesium deposits to 140 nm.
Scanning electron microscope ͑SEM͒ images of Mg4 and Mg5
shown in Fig. 4a and b, respectively, revealed a very similar mor-
phology to Mg ͑Fig. 2b͒.
2
Effect of current density.— In pulse electrodeposition experi-
ments using aqueous electrolytes, an increase in current density
leads to an increase in overvoltage, which results in increased nucle-
ation rate and a reduction in grain size. A sample ͑Mg ͒ was elec-
trodeposited under the same deposition conditions as sample Mg ,
but with double the peak current density ͑150 mA cm ͒. Figure 5a
shows a scanning electron micrograph of sample Mg . It shows a
crystalline deposit with large, well-defined, pyramidal crystallites
approximately 5 ␮m across. In particular, the facets of the crystal
are clearly visible. The XRD pattern of this sample is shown in Fig.
c and indicates a large-grained material with some 002 preferential
orientation, although the preferred orientation is less marked than in
Mg ͑see Table II͒. This is confirmed by SEM, which shows some of
the crystallites orientated with the apex of the pyramids perpendicu-
lar to the surface. The crystallite size calculated from the XRD line
broadening is given in Table II. The increase in the current density
in this pulse regime results in grain sizes that are comparable, within
the accuracy of the technique, to those found at a lower current
density.
Figure 5. SEM micrographs of ͑a͒ Mg , ͑b͒ Mg , and ͑c͒ Mg .
6
7
8
11
6
5
−
2
−
2
sample Mg but with a peak current density of 15 mA cm and a
deposition time of 1 min to give a total charge density of
.18 C cm . Sparsely distributed submicrometer crystallites with
the same geometry as those deposited at a high charge density are
visible, and it can be seen how the small crystallites agglomerate to
form larger particles. It therefore appears that, under pulse deposi-
tion conditions in the THF-based solutions with a current density of
5 mA cm , the deposit maintains the same morphology even at a
low deposition charge density. This is in contrast to the findings of
5
6
−
2
0
3
1
−
2
1
1
5
Aurbach et al., where they consistently reported a different mor-
−
2
phology for dc deposits at a low charge density ͑Ͻ0.5 mA cm ͒.
It is speculated that an evenly dispersed coating of these small par-
ticles may react rapidly with hydrogen gas as they would be easily
accessible to the gas, and the small particle size would reduce the
diffusion distance of hydrogen into the bulk of the particles. It was
not possible to determine the grain size of these particles as the
Effect of charge density.— Figure 5b shows an SEM image of a
magnesium sample ͑Mg ͒ deposited under the same conditions as
7
Downloaded on 2016-06-06 to IP 144.82.108.120 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

Journal of The Electrochemical Society, 157 ͑4͒ E45-E49 ͑2010͒
E49
sample size was too small to allow for powder XRD analysis. Figure
c shows Mg prepared under the same conditions as Mg , but de-
posited for 10 min to give a charge density of 1.8 C cm . The
same angular particles are visible, but they are much larger than in
References
5
8
7
1. M. Dornheim, N. Eigen, G. Barkhordarian, T. Klaasen, and R. Bormann, Adv. Eng.
Mater., 8, 377 ͑2006͒.
. J. Huot, G. Liang, S. Boily, A. Van Neste, and R. Schulz, J. Alloys Compd.,
−
2
2
2
93–295, 495 ͑1999͒.
Mg and are agglomerated to form a layer which still includes fairly
7
3. A. Zaluska, L. Zaluski, and J. O. Ström-Olsen, J. Alloys Compd., 288, 217 ͑1999͒.
4. X. Yao, C. Wu, A. Du, G. Q. Lu, H. Cheng, S. C. Smith, J. Zou, and Y. He, J. Phys.
Chem. B, 110, 11697 ͑2006͒.
large voids.
5
6
. K. Yoshimura, Y. Yamada, and M. Okada, Surf. Sci., 566–568, 751 ͑2004͒.
. S. Singh, S. W. Eijt, M. W. Zandbergen, W. J. Legerstee, and V. L. Svetchnikov, J.
Alloys Compd., 441, 344 ͑2007͒.
Conclusions
In this work, it was demonstrated that, by varying the pulse
deposition conditions, the morphology, preferential orientation, and
grain size of the electrodeposited magnesium particles can be varied
from dense films with strong preferential 002 orientation to irregu-
lar, nanocrystalline films with a random crystallite orientation. The
irregular films deposited at a low current density and a low charge
density in DBG electrolytes are promising materials for hydrogen
storage applications as they have particle sizes in the 2–3 ␮m range
and grain sizes less than 100 nm. The process is also rapid, with
deposition times of 1 h compared to milling times of 20–100 h to
achieve a similar morphology.
7
. H. Akyıldız, M. Özenbaş, and T. Öztürk, Int. J. Hydrogen Energy, 31, 1379
͑2006͒.
8. A. Léon, E. J. Knystautas, J. Huot, and R. Schulz, J. Alloys Compd., 345, 158
͑2002͒.
9
. J. Qu, Y. Wang, L. Xie, J. Zheng, Y. Liu, and X. Li, J. Power Sources, 186, 515
2009͒.
0. Y. He and Y. Zhao, Nanotechnology, 20, 204008 ͑2009͒.
͑
1
11. H. Natter and R. Hempelmann, Z. Phys. Chem., 222, 319 ͑2008͒.
12. A. Brenner and J. L. Sligh, Trans. Inst. Met. Finish., 49, 71 ͑1971͒.
1
3. T. D. Gregory, R. J. Hoffman, and R. C. Winterton, J. Electrochem. Soc., 137, 775
1990͒.
4. C. Liebenow, J. Appl. Electrochem., 27, 221 ͑1997͒.
͑
1
15. D. Aurbach, A. Schechter, M. Moshkovich, and Y. Cohen, J. Electrochem. Soc.,
Further studies are underway to assess the hydrogen storage
properties of the pulse electrodeposited magnesium.
148, A1004 ͑2001͒.
1
1
6. S. B. Hassen, L. Bousselmi, E. M. Rezrazi, P. Berçot, and E. Triki, Surf. Coat.
Technol., 202, 3579 ͑2008͒.
7. I. Haas and A. Gedanken, Chem. Commun. (Cambridge), 2008, 1795 ͑2008͒.
Acknowledgments
18. S. S. V. Tatiparti and F. Ebrahimi, J. Electrochem. Soc., 155, D363 ͑2008͒.
9. M. Tanniru and F. Ebrahimi, Int. J. Hydrogen Energy, 34, 7714 ͑2009͒.
20. A. L. Patterson, Phys. Rev., 56, 978 ͑1939͒.
1
Dr. Shane Askew and Jen Whan of the Advanced Analytical Cen-
tre ͑AAC͒ at James Cook University ͑JCU͒ are thanked for their
assistance with the SEM analysis of the samples. Eddie Ness, also of
AAC at JCU, is thanked for the XRD data generation for the
samples examined in this study.
2
1. A. A. Coehlo and R. W. Cheary, X-ray Line Fitting Program XFIT, Deposited in
CCP14 Powder Diffraction Library, Engineering and Physical Science Research
Council, Daresbury Laboratory, Warrington, England ͑1996͒.
2
2. D. Aurbach, I. Weissman, Y. Gofer, and E. Levi, Chem. Rec., 3, 61 ͑2003͒.
23. A. Léon, E. J. Knystautas, J. Huot, and R. Schulz, Thin Solid Films, 496, 683
2006͒.
͑
James Cook University assisted in meeting the publication costs of this
article.
24. Y. Viestfrid, M. D. Levi, Y. Gofer, and D. Aurbach, J. Electroanal. Chem., 576,
183 ͑2005͒.
Downloaded on 2016-06-06 to IP 144.82.108.120 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).