Iselt 2012
Iselt 2012
Iselt 2012
The electrochemical preparation of Fe-Ga nanowires is reported. The direct adaptation of thin film deposition conditions to nanoporous
anodic aluminum oxide templates with a pore diameter of 70 nm results in an inhomogeneous growth with segmented nanowires due
to incomplete re-dissolution of hydroxides. The modification of the electrolyte by complexing the metal ions, therefore suppressing
the hydroxide formation, allowed the fabrication of smooth, homogeneous Fe-Ga nanowires. A negligible composition gradient over
wire length and low oxygen content are obtained for these wires.
© 2012 The Electrochemical Society. [DOI: 10.1149/2.002303eel] All rights reserved.
Manuscript submitted October 8, 2012; revised manuscript received November 20, 2012. Published December 15, 2012.
Fe100-x Gax (x = 15 to 35 at.%), a non rare-earth alloy with large SCE. A potential pulse technique is used with a deposition potential
magnetostriction and high mechanical strength, has recently attracted E1 and an “off ” potential E2 that are applied for t1 = t2 = 10 s. The
increasing attention for sensors and actuators. The further device potential values and number of cycles are discussed in the results
miniaturization required for these applications necessitates innovative, section. All experiments are performed at room temperature without
tunable preparation routes for low dimensional structures. Electro- stirring.
chemical deposition is a promising low cost method to fabricate com- Nanowire morphology was characterized at membrane cross sec-
plicated geometries. In particular, nanowire arrays with high aspect tions by high-resolution scanning electron microscopy (Leo 1530
ratios are preferably prepared by electrodeposition within nanoporous Gemini/Zeiss). The integral composition was obtained by energy
templates. dispersive X-ray spectroscopy (X-Flash detector, Esprit, Bruker sys-
Recently, much progress has been made in developing suitable tems). The oxygen content of the filled template was determined by
electrolytes and parameters for both Fe-Ga thin film and nanowire EDX and corrected for the Al/O ratio of the empty region of the tem-
electrodeposition.1–4 In particular the fabrication and magnetic prop- plate in order to estimate the oxygen content of the nanowires. After
erties of multilayered nanowires have been investigated.5–8 Our pre- removing the working electrode from the template back side the struc-
vious EQCM investigations of Fe-Ga alloy deposition from a simple ture was analyzed by X-ray diffraction in Bragg-Brentano geometry
aqueous electrolyte revealed a complex mechanism comprising both (XRD, Bruker D8 Advance, Co-Kα radiation).
electrochemical and chemical deposition processes.9 An induced co-
deposition was proposed for the Fe-Ga alloy formation at potentials
more positive than that of single Ga deposition. Simultaneously, large Results and Discussion
amounts of mainly gallium oxide/hydroxide are formed and incorpo-
rated into the deposit due to a pH increase, caused by strong hydrogen Transferring the previously reported film deposition conditions
evolution. Applying alternating potential pulses allowed to re-dissolve using electrolyte A9 to nanoporous templates with pore diameters
these (hydr)oxides and results in dense, homogeneous and oxygen free below 100 nm leads to non-stationary deposition behavior and irreg-
films.10 In this study we report the preparation of Fe-Ga nanowires ular nanowire growth. An increase in cathodic currents with cycle
in anodic aluminum oxide (AAO) templates starting from these number is observed for both potential steps E1 and E2 (Figure 1a).
promising thin film results. The electrolyte and the deposition con- The obtained nanowires achieve an average length of about 30 μm
ditions were appropriately modified in order to obtain homogeneous after 180 cycles, but with a large variation in length (Figure 1c). The
nanowires. desired composition around Fe80+-2 Ga20+-2 was achieved for pulse
potentials of E1 = −1.375 V and E2 = −0.70 V but with a slight
variation in Ga content along the wire axis. The oxygen content of
Materials and Methods the filled template is significantly increased compared to the empty
Hexagonally ordered AAO templates were prepared by the well- region of the template indicating large amounts of oxygen incorpo-
established two-step anodization process.11,12 Both the remaining alu- rated in the nanowires. More detailed SEM investigation (Figure 1d)
minum and the barrier layer were removed from the back side to reveals short nanowire segments of about 200 nm length with a
obtain freestanding membranes.13 The typical pore diameter is 70 nm rough surface that are separated by thin gaps. As can be seen in
and the pore length approximately 35 μm. On the back side of the Figure 1d) these segments typically have a thickening at the lower end
template a 30 nm Au layer was evaporized as the first electrical con- and are narrower at the top. Here, the pore diameter is not completely
tact. Additional Cu electrodeposition seals the pores and enhances the filled.
mechanical stability of the template. This Au/Cu layer serves as the According to the previously proposed deposition mechanism,9
working electrode. each pulse would consist of simultaneous Fe-Ga alloy deposition
The aqueous electrolytes contain 0.06 M Ga2 (SO4 )3 · 18H2 O and and (mainly Ga) hydroxide precipitation during the first potential step
0.3 M (electrolyte A) or 0.03 M (electrolyte B) FeSO4 · 7H2 O as the E1 and re–dissolution of hydroxide during the second potential step
Fe and Ga sources, respectively. Boric acid (0.5 M) is added as buffer E2 . Restricted mass transport conditions in the confined geometry of
and ascorbic acid (0.04 M) as antioxidant. The modified electrolyte the AAO template, however, hinder particularly the pH compensation
(B) additionally contains 0.15 M sodium citrate. The pH of electrolyte and thus the complete re-dissolution of hydroxides. This can explain
A is adjusted to 1.5 to avoid gallium hydroxide precipitation while both the incomplete filling of the pore diameter, the unusual non-
electrolyte B can be operated at a less acidic pH of 3.0 compact appearance of the segments and the high oxygen content of
Electrodeposition experiments are performed in a three electrode the nanowires. With further length growth of the nanowires the un-
arrangement using a potentiostat/galvanostat HEKA PG 310. Platinum filled pore length is reduced, thereby improving the mass transport
is used as the counter electrode and a Saturated Calomel Electrode conditions and increasing the deposition rates as it is clearly seen in
(SCE, 241mVSHE ) as the reference electrode. All potentials refer to Figure 1a). Along with the changing pH conditions within the pores
this may be the reason for the observed variation in Ga content along
the nanowire length. Finally, the large variation in nanowire length is
z
E-mail: h.schloerb@ifw-dresden.de attributed to the strong hydrogen evolution due to the low pH solution.
Downloaded on 2015-05-11 to IP 192.231.202.205 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).
D14 ECS Electrochemistry Letters, 2 (3) D13-D15 (2013)
Figure 1. Current-time behavior of Fe-Ga deposition within AAO templates (a) for the standard electrolyte A and (b) for the modified electrolyte B with the
corresponding HR-SEM cross sections of Fe-Ga nanowires in AAO templates deposited from electrolyte A (E1 = −1.35 V, E2 = −0.75 V, t1 = t2 = 10 s, 180
cycles, c) overview, d) detail) and electrolyte B (E1 = −1.5 V, E2 = −0.9 V, t1 = t2 = 10 s, 500 cycles, e) overview, f) detail).
The resulting gas bubbles can grow to different sizes before they es- the second step E2 is almost zero, which effectively enables the com-
cape out of the pores. This can lead to a partial blocking of single pensation of pH and concentration gradients of the metal ions. The
pores and thus to differences in growth rates and varying lengths. cross-sectional SEM overview (Figure 1e) shows uniform nanowires
As a conclusion from these experiments it is obvious that compact reaching a length of about 6 μm after 500 cycles. High resolution
alloy nanowires which completely fill the pores and with low oxygen SEM (Figure 1f) reveals compact nanowires almost completely filling
content can only be achieved with a completely different deposition the pore diameter. The surface morphology is smooth and no defects
mechanism where hydroxide formation and re-dissolution are elimi- are visible. The desired composition of Fe81±2 Ga19±2 was obtained at
nated. This is expected to be possible via an appropriate complexation potentials E1 = −1.5 V and E2 = −0.9 V. As it is seen in Figure 2a,
of the metal ions. As shown by McGary et al.1 citrate seems well within the accuracy of the EDX there is no gradient in Ga content
suited as a complexing agent. Stabilizing the metal ions, in partic- along the wire length. Only a marginal increase in the oxygen content
ular Ga3+ , through complexation allows operating the electrolyte at was obtained for the filled template compared to the empty template
less acidic pH. This significantly reduces the hydrogen evolution and indicating a significant decrease in nanowire oxygen content com-
also has a positive effect on the stability of the template. However, pared to the samples deposited from electrolyte A. X-ray diffraction
simply adding citrate to the standard electrolyte A while keeping the (Figure 2b) shows a single reflection that can be assigned to the (110)
metal ion ratios constant revealed pure Fe wires containing almost reflection of the thermodynamically stable disordered Fe-Ga phase,
no Ga. A co-deposition of Fe and Ga was only achieved for a metal with gallium atoms randomly distributed within the bcc lattice of iron.
ion ratio Ga3+ :Fe2+ ≥ 2:1. The citrate concentration was adapted As the (100) and (200) reflections are absent these results indicate a
to the overall metal ion content (Fe2+ + Ga3+ ) in order to obtain preferred (110) orientation along the wire axis. These results are con-
Me1 [Cit]1 complexes, that are reported to be most stable at slightly firmed by detailed TEM investigations (EDX and nanodiffraction).
acidic conditions.14–16 Magnetization measurements performed parallel and perpendicu-
Figure 1b) presents the electrochemical data of pulse deposited lar to the wire axis reveal an almost identical magnetization behavior
Fe-Ga nanowires from the modified electrolyte B. In contrast to the (Figure 2). As the magnetization behavior of nanowire arrays are
standard electrolyte the current-time transients reveal a very uniform commonly dominated by shape anisotropy,17 this uncommon obser-
deposition behavior with increasing number of cycles. Although the vation implies contributions of other magnetic anisotropies like mag-
potentials are more negative compared to the standard electrolyte A netocrystalline or magnetostrictive anisotropy. Both are negligible for
the obtained pulse currents are lower, mainly due to the lower con- Fe, but Fe-Ga exhibits a high magnetostriction. As we have shown ex-
centration of Fe2+ in the solution. Interestingly, the current during emplarily for Ni nanowires, exhibiting a much lower magnetostriction,
Downloaded on 2015-05-11 to IP 192.231.202.205 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).
ECS Electrochemistry Letters, 2 (3) D13-D15 (2013) D15
Summary
The preparation of Fe-Ga nanowires in AAO templates is reported.
A simple adaptation of the deposition conditions from thin film exper-
iments leads to segmented nanowires with an inhomogeneous length
distribution. In order to obtain continuous, smooth nanowires a mod-
ification of the electrolyte is required. Complexing the metal ions,
in particular the Ga3+ ions, allows an increase in the electrolyte pH
from 1.5 to 3.0 and avoids the precipitation of hydroxides. Defect free
Fe80 Ga20 nanowires are successfully obtained up to a length of several
micrometers with low oxygen content and a negligible composition
gradient over nanowire length. The almost isotropic magnetic behav-
ior indirectly confirms the formation of the highly magnetostricitve
disordered Fe-Ga alloy.
Acknowledgment
The authors thank Sabine Neitsch for template preparation and
Sebastian Fähler for assistance in the interpretation of the magnetic
measurements.
References
1. P. D. McGary, K. S. M. Reddy, G. D. Haugstad, and B. J. H. Stadler, J. Electrochem.
Soc., 157, D656 (2010).
2. S. M. Reddy, E. C. Estrine, D.-H. Lim, W. H. Smyrl, and B. J. H. Stadler, Elec-
trochem. Comm., 18 127 (2012).
3. P. D. McGary and B. J. H. Stadler, J. Appl. Phys., 97, 10R503 (2005).
4. P. D. McGary, L. W. Tan, J. Zou, B. J. H. Stadler, P. R. Downey, and A. B. Flatau, J.
Appl. Phys., 99, 308 (2006).
5. N. Lupu, H. Chiriac, and P. Pascariu, J. Appl. Phys., 103, 07B511 (2008).
Figure 2. Characteristics of Fe-Ga nanowires obtained from the complexed 6. N. Lupu, P. Pascariu, C. Gherasim, and H. Chiriac, IEEE Trans. Magn., 44, 3005
electrolyte B (E1 = −1.5 V, E2 = −0.9 V, t1 = t2 = 10 s, 500 cycles): a) (2008).
Gallium content vs. nanowire length determined by EDX, b) XRD diagram 7. S. M. Reddy, J. J. Park, S.-M. Na, M. M. Maqableh, A. B. Flatau, and B. J. H. Stadler,
Adv. Funct. Mater., 21, 4677 (2011).
(substrate reflections are marked with *), c) hysteresis loops measured parallel
8. S. M. Reddy, J. J. Park, M. M. Maqableh, A. B. Flatau, and B. J. H. Stadler, J. Appl.
and perpendicular to the wire axis. Phys., 111, 07A920 (2012).
9. D. Iselt, K. Tschulik, S. Oswald, D. Pohl, L. Schultz, and H. Schlörb, J. Electrochem.
Soc., 159, H633 (2012).
the stress between the template and Ni nanowires is sufficient to in- 10. D. Iselt, U. Gaitzsch, S. Oswald, S. Fähler, L. Schultz, and H. Schlörb, Electrochim.
duce a magnetostrictive anisotropy, which can compete with shape Acta, 56, 5178 (2011).
11. H. Masuda and K. Fukuda, Science, 268, 1466 (1995).
anisotropy.12 Hence the almost coinciding hysteresis curves are an 12. A. Kumar, S. Fähler, H. Schlörb, K. Leistner, and L. Schultz, Phys. Rev. B, 73, 064421
additional, indirect confirmation of a highly magnostrictive material (2006).
like disordered Fe-Ga. 13. V. Hähnel, S. Fähler, P. Schaaf, M. Miglierini, C. Mickel, L. Schultz, and H. Schlörb,
In the following, the obtained results for the modified electrolyte Acta Mater., 58, 2330 (2010).
14. T. B. Field, J. L. McCourt, and W. A. E. McBryde, Can. J. Chem., 52, 3119
are evaluated and discussed in comparison to the simple electrolyte. (1974).
The stabilization of the metal ions through complexing shifts the re- 15. C. H. F. Chang, T. P. Pittner, R. E. Lenkinski, and J. D. Glickson, J. Am. Chem. Soc.,
duction potentials to more negative values. This might explain the 99, 5858 (1977).
required higher negative deposition potential of E1 = −1.5 V. As 16. M. Clausen, L.-O. Öhman, and P. Persson, J. Inorg. Biochem., 99, 716 (2005).
17. H. Schlörb, V. Hähnel, M. S. Khatri, A. Srivastav, A. Kumar, Ludwig Schultz, and
the Ga[Cit] complex (log K1 (Ga3+ ) = 5.7418 ) is more stable than S. Fähler, Phys. Status Solidi B, 247, 2364 (2010).
the Fe[Cit] complex (log K1 (Fe2+ ) = 3.0818 ) Fe2+ is preferentially 18. L. G. Sillen and A. E. Martell, Stability Constants of Metal-Ion Complexes, Section
deposited and only a minor portion of Ga is co-deposited. The posi- II: Organic Ligands, Vol. 17, The Chemical Society, London (1964).
Downloaded on 2015-05-11 to IP 192.231.202.205 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).