Multigram synthesis of copper nanowires using ac electrodeposition intoporous aluminium oxide templates

Genaro A. Gelves, Zakari T. M. Murakami, Matthew J. Krantz and Joel A. Haber*

Received 7th March 2006, Accepted 15th May 2006

First published as an Advance Article on the web 31st May 2006

DOI: 10.1039/b603442j

Multigram quantities of Cu nanowires ca. 25 nm in diameter and mm in length have been

produced by AC electrodeposition into porous aluminium oxide (PAO) templates. Multiple,

large-area Al electrodes (5 6 11 cm or 10 6 25 cm) are anodized in parallel at 25.0 V in 0.3 M

H2SO4(aq) using custom built baths. The pores are efficiently filled by applying 200 Hz sine waves

at 10 Vrms between the anodized Al and Cu plate counter electrodes immersed in a 0.50 M

CuSO4(aq) solution. Dissolution of the PAO template in 0.6 M H3PO4 to free the Cu nanowires

results in significant coarsening of the nanowires, whereas dissolution of the PAO template in

1.0 M NaOH(aq) results in retention of the Cu nanowire diameters corresponding to the pore

diameter of the PAO template. Liberated Cu nanowires were characterized by scanning electron

microscopy, X-ray diffraction, and X-ray photoelectron spectroscopy.

Introduction

The synthesis and applications of nanowires, nanobelts, nano-

rods, and related nanostructures has recently been reviewed.1

In the past few years, significant advances have been made in

the solution-based synthesis of metal nanorods and nanowires.

In particular, Murphy and co-workers have demonstrated

great control over the aspect ratio and surface chemistry of

gold nanorods and nanowires, using a surfactant mediated

technique.2–6 Xia’s and several other groups have developed

solution based techniques to synthesize high aspect ratio silver

nanowires.1,7–17 The relative rarity and expense of these

metals may preclude the widespread use of these nanowires

in engineering polymer composites, for instance. Recently, a

solution-based synthesis of high aspect-ratio copper nanowires

has been reported.18 This method appears promising, but

relatively large nanowire diameters (90–120 nm) are obtained.

Template-directed syntheses in the pores of porous alumi-

nium oxide (PAO) or track-etched polymer membranes have

been widely used to produce uniformly sized nanofibers and

nanowires of many materials and composite structures.19–22 As

described in the references given below, DC and AC

electrodeposition of metals into PAO templates has been

investigated for decades. Liberation of the metal nanoparticles

by dissolution of the alumina has been used to produce small

quantities (¡1 mg) of metal nanoparticles or metal nanowires

whose fundamental properties and applications have been

investigated (notably by Moskovits, Martin, Masuda,

Mallouk, and Gosele, as well as others). The well-developed

approach of DC electrodeposition of metals into PAO

template membranes affords control over the crystallinity of

nanowires23–28 and enables precise modulation of the compo-

sition along the length of the nanowire.29–36 These nanowires

display a plethora of interesting properties with diverse

applications.26,29–42 Unfortunately, the multiple processing

steps required to produce the rather fragile free-standing

PAO membranes and then to prepare electrodes suitable for

DC electrodeposition may prevent this method from being

used to produce metal nanowires in sufficient quantity for

many potential applications.

The less developed approach of AC electrodeposition into

PAO templates through the resistive, rectifying barrier layer

without separation from the Al substrate also affords good

quality filling of the pores with single metals, if optimized

electrodeposition conditions are used.43–54 High uniformity of

deposition has been demonstrated by the group of Gosele

using current-controlled deposition sequences.45,49,52 We have

elected to develop voltage-controlled deposition sequences,

due to this method’s insensitivity to electrode area. We have

previously reported a detailed study of the effect of electro-

deposition parameters upon the uniformity of pore-filling of

PAO templates with copper,55 and have communicated the

preparation of polymer nanocomposites containing Cu nano-

wires prepared using AC electrodeposition into PAO tem-

plates.56 We believe that it is technically feasible to produce kg

quantities of metal nanowires using the ac electrodeposition

process, as the anodization and electrodeposition processes are

fundamentally similar to the Anolok1 process (a trademark

of Alcan International Ltd) that is used to anodize and

then colorize by metal deposition large Al pieces (7 6 1.8 m)

for architectural applications (http://www.hmfltd.co.uk).

Although the Anolok1 process does not produce well-ordered

pores and the quantity of metal deposited is small (little metal

is required for colorization), straightforward adoption of the

processes reported herein could be implemented.

Polymer matrix nanocomposites containing nanodisperse

fillers such as exfoliated clays, carbon black, multi-walled and

single-walled fullerene nanotubes, and metal nanoparticles

have received great and increasing attention due to their

combination of processibility with enhanced or unique barrier,

electrical, mechanical, or bioactive properties.57–64 RelativelyDepartment of Chemistry, University of Alberta, Edmonton, Alberta,Canada T6G 2G2. E-mail: joel.haber@ualberta.ca

PAPER www.rsc.org/materials | Journal of Materials Chemistry

This journal is � The Royal Society of Chemistry 2006 J. Mater. Chem., 2006, 16, 3075–3083 | 3075

Dow

nloa

ded

by U

nive

rsity

of

Ten

ness

ee a

t Kno

xvill

e on

07

Mar

ch 2

013

Publ

ishe

d on

31

May

200

6 on

http

://pu

bs.r

sc.o

rg |

doi:1

0.10

39/B

6034

42J

View Article Online / Journal Homepage / Table of Contents for this issue

little investigated are polymer matrix nanocomposites contain-

ing nanowires composed of metals, likely due to the lack of

synthetic methods suitable for producing even gram quantities

of nanowires. Were gram quantities of metal nanowires

available inexpensively, the combination of their high thermal

and electrical conductivity and the potential ease of modifying

their surface properties by surface functionalization with small

molecule thiols, amines, and carboxylates would afford

numerous opportunities for tuning and enhancing properties

in applications such as polymer matrix nanocomposites. We

have developed a simple, scalable, versatile, bench-top electro-

synthesis of gram quantities of high-aspect ratio Cu nanowires.

With appropriate modification of the electrodeposition condi-

tions, the process described here could be applied to the

multigram synthesis of other metals and compound nanowires.

Experimental

Herein we describe somewhat different processing conditions

for two differently sized Al electrodes for growth of the PAO

films and for electrodeposition to deposit the Cu nanowires.

Electrode preparation

For the smaller scale process, aluminium electrodes were pre-

pared by cutting Al plate (99.99+%, Alfa Aesar, 1 mm thick)

into 5 6 11 cm pieces, which were annealed at 500 uC in air for

24 h. The electrodes were etched in 60 uC, 1.0 M NaOH(aq) for

2 min, and both faces immediately electropolished in a 1 uCdouble-walled, cooled cylindrical bath containing 500 mL of

solution consisting of a mixture of 115 mL 70% perchloric

acid, 90 mL 2-butoxyethanol, 600 mL ethanol and 150 mL

water. Graphite foil (99%, Alfa Aesar, 1 mm thick) was used

for the counter electrodes (one on each side of the Al, 2 cm

from and parallel to the Al face), with electrical contacts made

using copper alligator clips. Three pulses with a current density

of 280 mA cm22 and 15 s duration were applied with a 5 min

period between pulses to allow the bath to cool to 1 uC.

The electropolished electrodes were anodized following the

two-step process of Masuda.65 Five electrodes were anodized

in parallel by immersing them to a depth of 9.5 cm in 3.0 L of

0.3 M H2SO4(aq) at 2 uC in a double walled, chilled cylindrical

bath, as shown in Fig. 1a. Each Al electrode was separated by

a stainless steel 316 mesh (Alfa Aesar, 150 mesh, 0.066 mm

diameter) counter electrode (6 counter electrodes in total)

1.5 cm from the face of each Al plate. Electrical contact was

again made by attaching an alligator clip to the top of each

plate connected in parallel to the positive terminal, while the

6 counter electrodes were connected in parallel to the negative

terminal of a Hewlett Packard 6024A DC power supply

(0–70 V and 0–12 A). The electrodes were first anodized at

25.0 V (producing a current of 1.5 A) for 2 h, removed from

the anodizing bath and the alumina film removed by etching

for 30 min in a 60 uC mixture of 0.20 M H2CrO4 and 0.60 M

H3PO4, followed by reanodization in the H2SO4 electrolyte for

4, 6, or 8 h, producing pores ca. 20 nm in diameter and 18, 27,

or 36 mm deep, respectively. At the completion of the second

anodization the barrier layer thickness was reduced by

decreasing the voltage in 2 V increments every 2 min until

15 V was reached, then in 1 V increments every 1 min until 9 V

was reached, which potential was held for 5 min and then the

power supply was turned off. The electrodes were soaked in

deionized water for 5 min and then dried with compressed air.

For the larger scale process, electrodes were prepared by

cutting Al plate into 10 6 25 cm pieces, which were neither

annealed nor electropolished, but instead chemically polished

before their first use. To chemically polish the electrodes they

were individually immersed for 3 min in a bath at 80–85 uCconsisting of 4 L of a mixture in the ratio of 784 ml 85%

H3PO4, 98 mL glacial HNO3, 40 g NaNO3, and 118 mL H2O

(caution: this bath evolves NOx).66 They were anodized

following a two-step process similar to that described above,

in an anodization tank shown in Fig. 1b, composed of 0.75 inch

thick plexiglass walls, containing 30 L of 0.30 M H2SO4. Each

Al electrode was separated by a graphite foil (1 mm thick, Alfa

Aesar, 97%) counter electrode (10.6 6 25 cm, 11 counter

electrodes in total) 2.0 cm from the face of each Al plate.

Electrical contact was again made by attaching an alligator

clip to the top of each plate connected in parallel to the

positive terminal, while the 11 counter electrodes were

connected in parallel to the negative terminal, of a Lambda

LK-351-FM power supply (0–40 V and 0–30 A). The two-step

anodization process at 25.0 V and barrier layer thinning

procedure described above was used.

The temperature was controlled by a combination of

circulating the acid mixture through a heat-exchange coil

immersed in a cooling bath using a peristaltic pump and by

Fig. 1 Photographs of (a) double-walled, cooled 3 L and (b) 30 L

plexiglass anodization baths.

3076 | J. Mater. Chem., 2006, 16, 3075–3083 This journal is � The Royal Society of Chemistry 2006

Dow

nloa

ded

by U

nive

rsity

of

Ten

ness

ee a

t Kno

xvill

e on

07

Mar

ch 2

013

Publ

ishe

d on

31

May

200

6 on

http

://pu

bs.r

sc.o

rg |

doi:1

0.10

39/B

6034

42J

View Article Online

circulating cooling fluid through glass radiators immersed on

both sides of the anodizing tank. Keeping the bath at 0–4 uCis critical to maintain a constant anodization current. If the

bath heats up to above 4 uC, the chemical etch rate at the

pore bottom increases, reducing the barrier layer thickness,

enabling the anodization current to increase and producing a

runaway feedback loop, until the current limit of the power

supply is reached. At this point the voltage decreases and

moves away from the conditions that enable attainment of

good pore-ordering. Extra circulation of the anodizing

solution was provided by two mechanical stirrers in addition

to the circulation of the solution with the peristaltic pump.

Electrodeposition of Cu

The pores of the alumina films were filled with Cu using ac

electrodeposition. The edges and top boundary of the anodized

electrodes were coated with nail polish. This is necessary

because micrometer sized cracks formed at all 90u edges

provide a low resistance deposition pathway that prevents

deposition into the 25 nm diameter pores, if exposed to the

deposition solution. Previously, ultrasound was used to

improve wetting of the pores prior to electrodeposition;55

however, this step was eliminated because we found it

damaged the surface of the electrodes and reduced our ability

to reuse the Al plates. Electrodes were wet with distilled water

and then individually immersed for 5 min in the deposition

solution consisting of 600 mL (for 5 6 11 cm electrodes) or 4 L

(for 10 6 25 cm electrodes) of 0.50 M CuSO4–0.285 M

H3BO3(aq). Each electrode was filled by applying a continuous

200 Hz sine wave at 10 Vrms for 10–15 min between the

anodized Al and Cu plate (99.999%, Alfa Aesar) counter

electrodes 2 cm from both faces of the Al electrode. The

deposition signal was generated using a Tabor Electronics

8023 function generator, the output of which was amplified

with a 1600 W Pyramid PB1281X car stereo amplifier powered

with a LOKO 45 A, 16 V DC power supply or a car battery.

The nail polish was removed from the edges of the electrodes

with acetone. Bulk Cu deposition was removed by immersing

the electrode in 60 uC, 0.6 M H3PO4 for 1 min, followed by

cleaning the surface with a Kimwipe.

Liberation of Cu nanowires

Two methods for dissolving the PAO template to liberate the

nanowires have been used. The first method attempted was

dissolution of the alumina in 60 uC, 0.6 M H3PO4(aq) for

30 min; however, this method produced significant coarsening

of the nanowires. Therefore, the second method of dissolving

the alumina in 40 uC, 1.0 M NaOH(aq) for 3 min is preferred.

The nanowires were cleaned for 10 min in a 1 : 1 mixture

of 0.10 M NaOH(aq)–MeOH, filtered using filter paper

(Whatman, ,1 mm pore size), rinsed with methanol, and

transferred to a volume of 100 or 250 ml of methanol. The

nanowires were then sonicated for a period of 1 h in a

sonication bath of 135 W average power and 38.5–40.5 kHz

frequency by applying the ultrasound for 5 min with a 5-min

rest period between sonications. After sonication, dispersed

copper nanowires were collected by filtration on Osmonics Inc.

nylon membranes (0.45 mm pore size) and dried under reduced

pressure. Yield: 0.87 mg cm22 (ca. 0.39 g per batch of

5 electrodes) for 4 h anodized and 1.68 mg cm22 (ca. 0.75 g per

batch of 5 electrodes) for the 8 h anodized 5 6 11 cm

electrodes and ca. 0.8–1.4 mg cm22 (ca. 3.5–6.2 g per batch of

10 electrodes) for the 10 6 25 cm electrodes anodized for 8 h.

Length distribution of NaOH-liberated Cu nanowires

Copper nanowires were grown in pores anodized for 8 h using

5 6 11 cm electrodes, and liberated using 1.0 M NaOH(aq).

The Cu nanowires were cleaned and sonicated as described

above. The resultant suspension of Cu nanowires in MeOH

was spin cast at 3000 rpm on 2.5 6 2.5 cm glass slides. SEM

images were collected of the well-dispersed nanowires and the

lengths of >1000 individual nanowires was measured using the

Image J image analysis software program (Wayne Rasband,

National Institutes of Health, USA, http://rsb.info.nih.gov/ij).

Preparation of nanowire samples for XPS and XRD analysis

Suspensions of Cu nanowires in MeOH obtained from the

liberation process were dried under reduced pressure in glass

vials containing 1 6 1 cm silicon wafers on the bottom. The

nanowires sedimented and dried onto the Si slide. The X-ray

photoelectron spectra (XPS) and XRD patterns were imme-

diately collected, limiting air exposure to ,30 min. The

diffraction peak of the silicon wafer was removed by excluding

the diffraction spot arising from the single crystal silicon

from the integration of the 2D image of X-ray diffraction to

produce the 2h plot.

Depth profiling of pore-filling

Following electrodeposition into 5 6 11 cm electrodes the

surfaces were cleaned (0.5 mm removed from the surface) or

ion-milled (10 mm removed from the surface) using an Oxford

Ion Fab 300+ ion mill. The surfaces were characterized by

SEM, and the percentage of pores filled to the top of the new

surface was determined using the cell counter plugin developed

for ImageJ 1.35j image analysis software (Wayne Rasband,

National Institutes of Health, USA, http://rsb.info.nih.gov/ij).

Instrumentation

Characterization of the liberated nanowires was accomplished

by scanning electron microscopy (SEM) using a JEOL 6301F

FESEM equipped with an energy dispersive X-ray spectro-

meter (EDX), X-ray diffraction (XRD) using a Bruker AXS

D8 Discover with GADDS detector, X-ray photoelectron

spectroscopy (XPS) with a Kratos Axis Ultra or Kratos Axis

165 spectrometer, and transmission electron microscopy

(TEM) using a JEOL 2010 equipped with a CCD camera

and EDX spectrometer.

Results and discussion

Results

As briefly described previously,56 gram quantities of ca. 25 nm

diameter Cu nanowires are produced by AC electrodeposition

into porous aluminium oxide (PAO) films followed by

dissolution of the alumina in 1.0 M NaOH(aq) (preferred) or

This journal is � The Royal Society of Chemistry 2006 J. Mater. Chem., 2006, 16, 3075–3083 | 3077

Dow

nloa

ded

by U

nive

rsity

of

Ten

ness

ee a

t Kno

xvill

e on

07

Mar

ch 2

013

Publ

ishe

d on

31

May

200

6 on

http

://pu

bs.r

sc.o

rg |

doi:1

0.10

39/B

6034

42J

View Article Online

in 0.6 M H3PO4(aq). Slightly different processes have been

developed for two different size Al electrodes (see the

Experimental section for details). In the smaller scale process

both faces of five 5 6 11 cm, 1.0 mm thick Al electrodes

interdigitated between stainless steel mesh counter electrodes

are anodized in a chilled 3.0 L bath using the method of

Masuda,65 Fig. 1a. In the larger scale process, both faces of

ten 10 6 25 cm, 1.0 mm thick Al electrodes interdigitated

between graphite foil counter electrodes were anodized in a

chilled 30 L bath, Fig. 1b.

Anodization of large-area electrodes. In previous work

anodizing small-area Al electrodes (1 6 1.5 cm),55 we observed

that these electrodes would frequently ‘‘burn-out’’; that is,

anodize rapidly at a defect by shunting all current to that

point, resulting in formation of a pit, and preventing uniform

anodization of the remainder of the electrode. This was an

especially common occurrence at the solution/air interface

when anodizing bare Al plate electrodes. To prevent this,

we found it helpful to make sure that the electrodes were free

of deep scratches or other obvious defects and to paint a

protective line of epoxy or nail-polish around the Al plate

at the level of the solution/air interface. In contrast, the

large-area electrodes described here are far less susceptible to

‘‘burn-out’’, so it is not necessary to protect the electrodes at

the solution/air interface.

AC electrodeposition into PAO on large area electrodes.

Micrometer-sized edge cracks. Anodization of both faces of

bare Al plate results in the formation of ca. 10 micrometer

sized cracks on all 90u edges as a result of the volume

expansion accompanying conversion of Al to Al2O3 (Fig. 2a).

If these edge-cracks are not coated before electrodepositon

under continuous ac conditions, most Cu will be deposited in

them, because the diffusion limitation into the ca. 10 mm

diameter cracks is far less than the diffusion limitation into the

ca. 20 nm diameter, 18 or 36 mm deep pores (Fig. 2b). Pulsed

electrodeposition conditions may be used to minimize the

effect of this parasitic pathway by introducing a ‘‘hold-time’’

between deposition pulses to allow diffusion of Cu2+ into the

pores, but at the expense of greatly increasing the time

required for the electrodeposition to be completed. We have

found it to be more efficient to coat the edges of the electrodes

with nail polish to cover the cracks and prevent electro-

deposition and enable continuous AC conditions to be used.

The nail polish is easily removed with acetone after the

electrodeposition.

Effect of counter electrode on stability of deposition solution

and yield. In prior work using small-area electrodes we utilized

platinum or stainless steel counter-electrodes immersed in a

similar CuSO4(aq) solution (the H3BO3 was 0.57 vs. 0.285 M in

the present work).55 However, for these large-area electrodes

we observed that the deposition solution could only be used

once or twice before deposition quality was adversely affected

and the yield of Cu nanowires decreased substantially. We

observed that the pH decreased with each electrode filled with

Cu, we believe as a result of oxidation of water occurring at

the counter electrode during the reduction of Cu in the PAO.

By using Cu counter electrodes, we do not deplete the Cu2+

concentration in the solution and the pH is stable for

deposition into tens of electrodes, as dissolution of Cu is the

most favorable electrochemical process. In addition, the

deposition time is shorter using Cu counter electrodes than

when using stainless steel counter electrodes.

Effect of pore depth on yield. The fraction of pores filled as a

function of depth beneath the initial electrode surface is

comparable to that reported previously for small-area electro-

des (1 6 1.5 cm) filled using pulsed sine wave electrodeposition

conditions, except that the pores described here are deeper.

The fraction of pores filled 0.5 mm below the surface of 8 h

anodized (ca. 36 mm deep pores) is ca. 20% (Fig. 3a), while the

fraction filled to a level 10 mm below the initial surface is 60%

(Fig. 3b). As shown in Fig. 4, for the 5 6 11 cm electrodes,

the mass of nanowires produced per cm2 of electrode area

increases approximately linearly with pore depth, over the

range of anodization times investigated (4, 6 and 8 h corres-

ponding to 18, 27 and 36 mm deep pores, respectively).

Fig. 2 SEM of (a) micron size crack that forms on 90 degree edges

and (b) the bulk deposition of Cu that occurs at the crack if it is not

covered.

3078 | J. Mater. Chem., 2006, 16, 3075–3083 This journal is � The Royal Society of Chemistry 2006

Dow

nloa

ded

by U

nive

rsity

of

Ten

ness

ee a

t Kno

xvill

e on

07

Mar

ch 2

013

Publ

ishe

d on

31

May

200

6 on

http

://pu

bs.r

sc.o

rg |

doi:1

0.10

39/B

6034

42J

View Article Online

Effect of electrode size on yield. One of the limitations we

have encountered in the scale-up of the process is our ability to

supply sufficient current during each deposition pulse. In

particular, during the first seconds of electrodeposition a much

higher current is required, because the bulk concentration

of Cu2+ is present at the pore bottoms. As the deposition

continues, the Cu2+ concentration decreases to some steady-

state value and a lower current is required. However, it

appears that this initial, high-current period is critical to

nucleating and depositing Cu in all of the pore bottoms. The

amplifier we used is limited to Irms 30 A peak current (before it

goes into protection mode). The smaller 5 6 11 cm electrodes

initially require Idc 15 A current (measured dc current supplied

to the amplifier) which declines to a steady-state Idc 5 A

current after 60 s. The larger 10 6 25 cm electrodes required

Idc > 30 A current initially, and declined to a steady-state

current of Idc 15–20 A after 60 s. We observe a higher yield

with the 5 6 11 cm electrodes (1.68 mg cm22, ca. 0.15 g per

electrode, ca. 0.75 g per batch) than for the larger 10 6 25 cm

electrodes (0.8–1.4 mg cm22, ca. 0.35–0.62 g per electrode, ca.

3.5–6.2 g per batch)—the yield per cm2 for the 10 6 25 cm

electrodes was only 50–80% of the yield obtained for the

smaller electrodes.

Effect of liberation process on morphology and dispersion.

Scanning electron microscopy indicates that the nanowires

liberated by dissolving the PAO template in 0.6 M H3PO4

[Fig. 5 (a–b)] are significantly less well dispersed than those

liberated by dissolving the PAO template in 1.0 M NaOH(aq)

[Fig. 5 (c–d)]. In addition, the morphology of copper

nanowires liberated using H3PO4 does not replicate that of

the porous alumina templates used. An average diameter of

69 ¡ 19 nm for H3PO4-liberated nanowires indicates a

significant increase in diameter (coarsening) when compared

to the average diameter of 25 ¡ 4 nm for nanowires liberated

in NaOH. In addition, the H3PO4-liberated nanowires display

noticeable oscillations in diameter along the nanowire length.

Although the pores of the PAO template are quite well filled

to depths of 30 mm, the length of the liberated nanowires is

significantly less. As shown in Fig. 6, even when grown in

pores ca. 36 mm deep, after fragmentation during the liberation

process the average nanowire length is ca. 2 mm and 90% of the

nanowires are ,3 mm in length. For these 25 nm diameter

wires, however, this still translates into an average aspect ratio

of ca. 80.

XRD patterns collected of nanowires immediately after

liberation (,30 min of air exposure) by both methods show no

evidence of oxides, CuO or Cu2O (Fig. 7). XPS collected

immediately after liberation (,30 min air exposure) indicates

little surface oxidation of the nanowires liberated by either

method (Fig. 8). The presence of surface oxide (CuO) is more

evident, as detected by XPS, after exposure of the liberated

copper nanowires to atmosphere for 3 d, but the nanowires

remain largely unoxidized. The Cu L3M4.5M4.5 Auger spectra

in Fig. 8c for nanowires exposed to air for ,30 min or 3 d,

Fig. 3 SEM images of top surfaces of ion-milled Cu-filled templates

(5 6 11 cm electrodes) to show degree of pore-filling as a function

of depth: (a) 0.5 and (b) 10 mm removed from the top of the 36 mm

deep-pores.

Fig. 4 Yield of Cu nanowires from PAO templates prepared by

anodization for different time in 0.3 M H2SO4 solutions (4, 6 or 8 h).

This journal is � The Royal Society of Chemistry 2006 J. Mater. Chem., 2006, 16, 3075–3083 | 3079

Dow

nloa

ded

by U

nive

rsity

of

Ten

ness

ee a

t Kno

xvill

e on

07

Mar

ch 2

013

Publ

ishe

d on

31

May

200

6 on

http

://pu

bs.r

sc.o

rg |

doi:1

0.10

39/B

6034

42J

View Article Online

however, clearly show that the NaOH-liberated nanowires

have a larger fraction of oxidized Cu(I) on their surface than

do the H3PO4-liberated nanowires.67,68

Discussion

In our previous work reporting a detailed study of the effect of

electrodeposition parameters upon the uniformity of pore-

filling of PAO templates with Cu, we noted that continuous ac

deposition conditions produced significant damage in the PAO

templates.55 In that work we focused upon the use of pulsed

depositions with both sine and square-wave forms, and noted

that square wave forms generally yielded more uniform pore

filling. For practical reasons we have utilized continuous

200 Hz sine waves for the electrodeposition of Cu in this work:

(1) continuous ac depositions require a fraction of the time to

complete compared to pulsed depositions, and (2) the low cost

stereo amplifier used in this work amplifies sine waves with

good fidelity, but not square waves. It is possible to use

continuous sine waves, because, as we noted previously, the

damage to the PAO was pronounced in larger pore diameter

templates grown in oxalic acid (ca. 35 nm diameter pores), but

minor for the smaller pore diameter templates grown in

sulfuric acid (ca. 20 nm diameter pores). We also noted that

damage only occurred when deposition was continued until

bulk deposition of Cu on the surface of the electrode was

observed. Therefore, the large-scale synthesis of Cu nanowires

reported herein has been limited to the smaller diameter pores

grown in sulfuric acid with the electrodeposition stopped as

soon as bulk deposition was observed. In addition, the voltage

Fig. 5 SEM of (a) and (b) 0.6 M H3PO4-liberated Cu nanowires and (c) and (d) 1.0 M NaOH-liberated Cu nanowires.

Fig. 6 (a) SEM image of copper nanowires grown in pores anodized

for 8 h (36 mm deep) and liberated using 1.0 M NaOH and dispersed

onto a glass slide, and (b) length distribution of copper nanowires

measured using similarly prepared samples.

3080 | J. Mater. Chem., 2006, 16, 3075–3083 This journal is � The Royal Society of Chemistry 2006

Dow

nloa

ded

by U

nive

rsity

of

Ten

ness

ee a

t Kno

xvill

e on

07

Mar

ch 2

013

Publ

ishe

d on

31

May

200

6 on

http

://pu

bs.r

sc.o

rg |

doi:1

0.10

39/B

6034

42J

View Article Online

used for copper electrodeposition in this work was 10 Vrms

rather than the 14 Vrms used previously. We have not observed

any evidence of damage to the PAO templates using these

deposition conditions.

Based on the Pourbaix diagram of copper in an aqueous

environment, we expect that Cu nanowire surfaces will be less

susceptible to oxidation during the H3PO4-liberation process

than during the NaOH-liberation.69 From studies of anodic

oxidation of copper in aqueous solutions, it is widely accepted

that a duplex film forms, composed of a bottom layer of Cu2O

formed during the initial stage of oxidation, and a top layer of

Cu(OH)2 or CuO formed primarily at the expense of the Cu2O

layer. The composition of the top layer is dependent on the

conditions of the oxidation treatment and the composition of

the aqueous electrolyte.70,71 Copper surfaces exposed to acidic

and basic solutions have shown different mechanisms of

oxidation.72 Acidic conditions resulted in Cu2O-rich oxide

layers, whereas strongly basic conditions (pH 13) resulted in a

CuO-rich layer after 24 h exposure. The presence of CuO is

observed from the weak signals obtained for the satellite peaks

in the Cu 2p XP spectra of both NaOH- and H3PO4-liberated

nanowires (Fig. 8b).67 However, the Cu L3M4.5M4.5 Auger

spectrum of NaOH-liberated nanowires has an intense signal

corresponding to Cu2O (at BE 569.8 eV) along with a less

intense signal corresponding to Cu(0) at BE 567.9 eV.68 In

contrast, a signal corresponding primarily to Cu(0) was

obtained for H3PO4-liberated nanowires indicating less surface

oxidation (Fig. 8c). Hence, we concluded that surface

oxidation of copper nanowires occurred to some degree after

liberation, but the kinetics of oxide formation is much slower

when H3PO4 solutions were used. Unfortunately, surface

quantification of Cu(I), Cu(II) and Cu(0) species is not possible

from XPS and Auger peaks.73,74 From these results, we believe

that because the nanowires are held parallel to each other by

the residual PAO template and collapse together as a result of

van der Waals forces,75 the metallic surfaces exposed during

H3PO4 liberation come in contact and irreversibly cold-weld,76

such that each apparent nanowire is actually a fused bundle of

Fig. 8 X-Ray photoelectron spectra of sputter-cleaned copper foil and copper nanowires (a) survey (b) Cu 2p peak (c) Cu L3M4.5M4.5 Auger peak.

The spectra correspond to sputter-cleaned Cu foil (A), Cu nanowires liberated from PAO templates using 1.0 M NaOH(aq) and exposed to air for

,30 min (B) and 3 d (C), and Cu nanowires liberated using 0.6 M H3PO4(aq) and exposed to air for (D) ,30 min and (E) 3 d. Satellite peaks on Cu

2p spectra labelled with an * indicate the presence of Cu(II) species, such as CuO.67 Dashed lines indicate the binding energy of the most intense

component of the L3M4.5M4.5 Auger peak from Cu(0) to Cu(I) and Cu(II).68

Fig. 7 X-Ray diffractograms of Cu nanowires exposed to air less

than 30 min (a) liberated in 1 M NaOH solutions, and (b) liberated in

0.6 M H3PO4 solutions. Standard powder patterns for (c) Cu, (d)

Cu2O, and (e) CuO.

This journal is � The Royal Society of Chemistry 2006 J. Mater. Chem., 2006, 16, 3075–3083 | 3081

Dow

nloa

ded

by U

nive

rsity

of

Ten

ness

ee a

t Kno

xvill

e on

07

Mar

ch 2

013

Publ

ishe

d on

31

May

200

6 on

http

://pu

bs.r

sc.o

rg |

doi:1

0.10

39/B

6034

42J

View Article Online

several nanowires. In contrast, liberation in NaOH solution

produces a surface oxide or hydroxide layer77,78 that inhibits

the fusing processes, ultimately yielding a finer nanowire

dispersion (Fig. 9).

The coarsening phenomenon and the diameter fluctuations

observed for H3PO4-liberated nanowires could be related

to Rayleigh instability in the copper nanowires.79 Rayleigh

instability of other types of nanowires has generally been

observed after annealing of the nanoparticles at high tempera-

tures, as it requires surface and volume diffusion.80 The

morphology variation in the nanowires liberated in H3PO4

is apparent following the etching step at 60 uC and before

any application of ultrasound. The reason nanowires with

modulated diameters are obtained from H3PO4-liberation and

not from NaOH-liberation is the subject of current study, but

we believe the phenomenon could be caused by enhanced

diffusion of copper atoms promoted by factors such as surface

chemistry, surface tension, pH, and temperature during the

liberation process.

One of the advantages of the AC deposition technique over

the DC deposition technique is the lower consumption of high-

purity Al. As shown in Fig. 10, to perform DC electrodeposi-

tion into PAO templates grown on Al substrates several

processing steps are required before deposition: (1) the Al

substrate is dissolved using HgCl2 or CuCl(aq), (2) the barrier

layer at the base of the pores is removed by dissolution in

H3PO4(aq) (the pores are also widened), (3) a metallic contact

is deposited on one side of the membrane. Although DC

electrodeposition enables greater control over the crystallinity

and structure of the deposited nanowires, it is not easily

scalable to produce grams of nanomaterials in the laboratory.

In contrast, AC electrodeposition requires no additional

processing steps, but finding conditions enabling uniform

pore-filling and efficient utilization of the template is

challenging. Most importantly, with AC deposition (once

efficient voltage controlled deposition conditions are

developed) scale-up to the multi-gram scale is relatively

straightforward. With AC deposition the Al substrate is not

sacrificed. Once the nanowires are liberated by dissolution of

the PAO template, the Al substrate can be re-anodized.

Starting with 1 mm thick plate, we are able to recycle each

electrode 12 times before it is completely consumed.

Conclusions

We have successfully developed a simple, inexpensive, bench

top electrochemical method for producing gram quantities of

Cu nanowires uniformly 25 nm in diameter and with average

lengths of ca. 2 mm when liberated by dissolving the PAO

template in NaOH(aq). The method of liberation significantly

impacts the diameter and dispersion of the nanowires, with

H3PO4(aq)-liberated nanowires exhibiting larger diameters,

less surface oxidation, and more agglomeration. A yield of

1.68 mg Cu per cm2 of Al has been obtained. This methodo-

logy is amenable to adaptation for the synthesis of other

metals or compound nanowires in gram quantities.

Acknowledgements

The assistance of the Department of Chemistry Electronics

Shop as well as the collection of SEM images by Mr. George

Braybrook, are gratefully acknowledged. We thank the

National Institute of Nanotechnology (NINT-NRC), the

Alberta Centre for Surface Engineering and Science

(ACSES), and the University of Alberta Nanofab for use of

XRD, XPS, and Ion-milling facilities. Funding was provided

to JAH as an NSERC Discovery grant (#238526) and jointly

to JAH and U. Sundararaj as an NSERC Strategic project

grant (#306589).

References

1 Y. N. Xia, P. D. Yang, Y. G. Sun, Y. Y. Wu, B. Mayers, B. Gates,Y. D. Yin, F. Kim and Y. Q. Yan, Adv. Mater., 2003, 15, 353–389.

2 A. Gole and C. J. Murphy, Langmuir, 2005, 21, 10756–10762.3 A. Gole and C. J. Murphy, Chem. Mater., 2005, 17, 1325–1330.4 L. F. Gou and C. J. Murphy, Chem. Mater., 2005, 17, 3668–3672.5 C. J. Murphy, T. K. San, A. M. Gole, C. J. Orendorff, J. X. Gao,

L. Gou, S. E. Hunyadi and T. Li, J. Phys. Chem. B, 2005, 109,13857–13870.

6 C. J. Murphy, T. K. Sau, A. Gole and C. J. Orendorff, MRS Bull.,2005, 30, 349–355.

7 C. Li, X. G. Yang, B. J. Yang, Y. Yan and Y. T. Qian, Mater.Lett., 2005, 59, 1409–1412.

8 Y. G. Sun, B. Gates, B. Mayers and Y. N. Xia, Nano Lett., 2002, 2,165–168.

9 Y. G. Sun, B. Mayers, T. Herricks and Y. N. Xia, Nano Lett.,2003, 3, 955–960.

10 Y. G. Sun, B. Mayers and Y. N. Xia, Nano Lett., 2003, 3, 675–679.11 Y. G. Sun and Y. N. Xia, Adv. Mater., 2002, 14, 833–837.12 Y. G. Sun, Y. D. Yin, B. T. Mayers, T. Herricks and Y. N. Xia,

Chem. Mater., 2002, 14, 4736–4745.

Fig. 9 Scheme comparing the surface chemistry and coarsening

mechanism of the Cu nanowires during liberation in 0.6 M

H3PO4(aq) and 1.0 M NaOH(aq).

Fig. 10 Schematic diagram comparing DC and AC electrodeposition

of metals into PAO templates.

3082 | J. Mater. Chem., 2006, 16, 3075–3083 This journal is � The Royal Society of Chemistry 2006

Dow

nloa

ded

by U

nive

rsity

of

Ten

ness

ee a

t Kno

xvill

e on

07

Mar

ch 2

013

Publ

ishe

d on

31

May

200

6 on

http

://pu

bs.r

sc.o

rg |

doi:1

0.10

39/B

6034

42J

View Article Online

13 Z. H. Wang, X. Y. Chen, J. W. Liu, M. Zhang and Y. T. Qian,Chem. Lett., 2004, 33, 1160–1161.

14 B. Wiley, Y. G. Sun, B. Mayers and Y. N. Xia, Chem. Eur. J.,2005, 11, 454–463.

15 S. H. Zhang, Z. Y. Jiang, Z. X. Xie, X. Xu, R. B. Huang andL. S. Zheng, J. Phys. Chem. B, 2005, 109, 9416–9421.

16 B. Wiley, Y. G. Sun and Y. N. Xia, Langmuir, 2005, 21, 8077–8080.17 Z. H. Wang, J. W. Liu, X. Y. Chen, J. X. Wan and Y. T. Qian,

Chem. Eur. J., 2004, 11, 160–163.18 Y. Chang, M. L. Lye and H. C. Zeng, Langmuir, 2005, 21,

3746–3748.19 J. C. Hulteen and C. R. Martin, J. Mater. Chem., 1997, 7,

1075–1087.20 C. R. Martin, Science, 1994, 266, 1961–1966.21 H. Masuda and K. Fukuda, Science, 1995, 268, 1466–1468.22 S. Shingubara, J. Nanopart. Res., 2003, 5, 17–30.23 M. E. T. Molares, V. Buschmann, D. Dobrev, R. Neumann,

R. Scholz, I. U. Schuchert and J. Vetter, Adv. Mater., 2001, 13,62–65.

24 H. Pan, B. H. Liu, J. B. Yi, C. Poh, S. Lim, J. Ding, Y. P. Feng,C. H. A. Huan and J. Y. Lin, J. Phys. Chem. B, 2005, 109,3094–3098.

25 M. L. Tian, J. G. Wang, J. Snyder, J. Kurtz, Y. Liu, P. Schiffer,T. E. Mallouk and M. H. W. Chan, Appl. Phys. Lett., 2003, 83,1620–1622.

26 M. L. Tian, J. U. Wang, J. Kurtz, T. E. Mallouk andM. H. W. Chan, Nano Lett., 2003, 3, 919–923.

27 J. G. Wang, M. L. Tian, T. E. Mallouk and M. H. W. Chan,J. Phys. Chem. B, 2004, 108, 841–845.

28 G. Yi and W. Schwarzacher, Appl. Phys. Lett., 2000, 76,1630–1630.

29 F. D. de Horne, L. Piraux and S. Michotte, Appl. Phys. Lett., 2005,86, 152510.

30 L. Gravier, S. Serrano-Guisan and J. P. Ansermet, J. Appl. Phys.,2005, 97, 10C501.

31 C. D. Keating and M. J. Natan, Adv. Mater., 2003, 15, 451–454.32 S. R. Nicewarner-Pena, R. G. Freeman, B. D. Reiss, L. He,

D. J. Pena, I. D. Walton, R. Cromer, C. D. Keating andM. J. Natan, Science, 2001, 294, 137–141.

33 A. K. Salem, J. Chao, K. W. Leong and P. C. Searson, Adv.Mater., 2004, 16, 268–271.

34 A. K. Salem, M. Chen, J. Hayden, K. W. Leong and P. C. Searson,Nano Lett., 2004, 4, 1163–1165.

35 A. K. Salem, C. F. Hung, T. W. Kim, T. C. Wu, P. C. Searson andK. W. Leong, Nanotechnology, 2005, 16, 484–487.

36 J. G. Wang, M. L. Tian, T. E. Mallouk and M. H. W. Chan, NanoLett., 2004, 4, 1313–1318.

37 M. Chen and P. C. Searson, Adv. Mater., 2005, 17, 2765–2768.38 T. R. Kline, W. F. Paxton, T. E. Mallouk and A. Sen, Angew.

Chem., Int. Ed., 2005, 44, 744–746.39 W. E. Paxton, A. Sen and T. E. Mallouk, Chem. Eur. J., 2005, 11,

6462–6470.40 W. F. Paxton, K. C. Kistler, C. C. Olmeda, A. Sen, S. K. St Angelo,

Y. Y. Cao, T. E. Mallouk, P. E. Lammert and V. H. Crespi, J. Am.Chem. Soc., 2004, 126, 13424–13431.

41 A. K. Salem, P. C. Searson and K. W. Leong, Nat. Mater., 2003, 2,668–671.

42 J. G. Wang, M. L. Tian, N. Kumar and T. E. Mallouk, Nano Lett.,2005, 5, 1247–1253.

43 D. AlMawlawi, N. Coombs and M. Moskovits, J. Appl. Phys.,1991, 70, 4421–4425.

44 D. AlMawlawi, C. Z. Liu and M. Moskovits, J. Mater. Res., 1994,9, 1014–1018.

45 J. Choi, G. Sauer, K. Nielsch, R. B. Wherspohn and U. Gosele,Chem. Mater., 2003, 15, 776–779.

46 D. N. Davydov, J. Haruyama, D. Routkevitch, B. W. Statt,D. Ellis, M. Moskovits and J. M. Xu, Phys. Rev. B, 1998, 57,13550–13553.

47 D. N. Davydov, P. A. Sattari, D. AlMawlawi, A. Osika,T. L. Haslett and M. Moskovits, J. Appl. Phys., 1999, 86,3983–3987.

48 F. Li, L. Zhang and R. M. Metzger, Chem. Mater., 1998, 10,2470–2480.

49 K. Nielsch, F. Muller, A.-P. Li and U. Gosele, Adv. Mater., 2000,12, 582–586.

50 C. Preston and M. Moskovits, J. Phys. Chem., 1993, 97,8495–8503.

51 D. Routkevitch, A. A. Tager, J. Haruyama, D. Almawlawi,M. Moskovits and J. Xu, IEEE Trans. Electron Devices, 1996, 43,1646–1658.

52 G. Sauer, G. Brehm, S. Schneider, K. Nielsch, R. B. Wherspohn,J. Choi, H. Hofmeister and U. Gosele, J. Appl. Phys., 2002, 91,3243–3247.

53 M. Sun, G. Zangari and R. M. Metzger, IEEE Trans. Magn., 2000,36, 3005–3008.

54 M. Sun, G. Zangari, M. Shamsuzzoha and R. M. Metzger, Appl.Phys. Lett., 2001, 78, 2964–2966.

55 N. J. Gerein and J. A. Haber, J. Phys. Chem. B, 2005, 109,17372–17385.

56 G. A. Gelves, U. Sundararaj and J. A. Haber, Macromol. RapidCommun., 2005, 26, 1677–1681.

57 R. H. Baughman, A. A. Zakhidov and W. A. de Heer, Science,2002, 297, 787–792.

58 O. Breuer and U. Sundararaj, Polym. Compos., 2004, 25, 630–645.59 N. Cioffi, L. Torsi, N. Ditaranto, G. Tantillo, L. Ghibelli,

L. Sabbatini, T. Bleve-Zacheo, M. D’Alessio, P. G. Zamboninand E. Traversa, Chem. Mater., 2005, 17, 5255–5262.

60 P. C. LeBaron, Z. Wang and T. J. Pinnavaia, Appl. Clay Sci., 1999,15, 11–29.

61 S. S. Ray and M. Okamoto, Prog. Polym. Sci., 2003, 28,1539–1641.

62 J. Sandler, A. H. Windle, P. Werner, V. Altstadt, M. V. Es andM. S. P. Shaffer, J. Mater. Sci., 2003, 38, 2135–2141.

63 J. K. W. Sandler, J. E. Kirk, I. A. Kinloch, M. S. P. Shaffer andA. H. Windle, Polymer, 2003, 44, 5893–5899.

64 R. Schueler, J. Petermann, K. Schulte and H. P. Wentzel, J. Appl.Polym. Sci., 1997, 63, 1741–1746.

65 H. Masuda, F. Hasegwa and S. Ono, J. Electrochem. Soc., 1997,144, L127–L130.

66 Eur. Appl. Pat., 325097A1, 1989.67 P. E. Larson, J. Electron Spectrosc. Relat. Phenom., 1974, 4,

213–218.68 N. S. McIntyre and M. G. Cook, Anal. Chem., 1975, 47,

2208–2213.69 M. Pourbaix, in Atlas of electrochemical equlibria in aqueous

solutions., Pergamon, London, England, 1966, pp. 384–392.70 W. Kautek and J. G. Gordon, J. Electrochem. Soc., 1990, 137,

2672–2677.71 H. H. Strehblow and B. Titze, Electrochim. Acta, 1980, 25,

839–850.72 Y. Feng, K. S. Siow, W. K. Teo, K. L. Tan and A. K. Hsieh,

Corrosion, 1997, 53, 389–398.73 P. E. Laibinis and G. M. Whitesides, J. Am. Chem. Soc., 1992, 114,

9022–9028.74 H. D. Speckmann, S. Haupt and H. H. Strehblow, Surf. Interface

Anal., 1988, 11, 148–155.75 S. J. Lee, A. R. Morrill and M. Moskovits, J. Am. Chem. Soc.,

2006, 128, 2200–2201.76 G. S. Ferguson, M. K. Chaudhury, G. B. Sigal and

G. M. Whitesides, Science, 1991, 253, 776–778.77 X. M. Liu and Y. C. Zhou, J. Mater. Res., 2005, 20, 2371–2378.78 Z. W. Liu and Y. Bando, Chem. Phys. Lett., 2003, 378, 85–88.79 M. E. Toimil-Molares, A. G. Balogh, T. W. Cornelius,

R. Neumann and C. Trautmann, Appl. Phys. Lett., 2004, 85,5337–5339.

80 A. Fukunaga, S. Y. Chu and M. E. McHenry, J. Mater. Res., 1998,13, 2465–2471.

This journal is � The Royal Society of Chemistry 2006 J. Mater. Chem., 2006, 16, 3075–3083 | 3083

Dow

nloa

ded

by U

nive

rsity

of

Ten

ness

ee a

t Kno

xvill

e on

07

Mar

ch 2

013

Publ

ishe

d on

31

May

200

6 on

http

://pu

bs.r

sc.o

rg |

doi:1

0.10

39/B

6034

42J

View Article Online