Electrodeposited magnetic nanowires: Arrays, field-induced assembly, and surface functionalization

Article abstract of DOI:10.1016/S0304-8853(02)00523-1

Electrodeposition into nanoporous templates provides a means of fabricating large quantities of magnetic nanowires with diameters in the range 5 nm-10 μm, and lengths up to 60 μm. Several recent developments in nanoporous templates and Ni nanowires are described. Templates formed by α-particle irradiation of single-crystal mica are shown to contain diamond-shaped nanopores with uniform size and orientation. Ni prisms deposited in these templates show anisotropic magnetic properties along all three axes of the prisms. The manipulation of isolated Ni nanowires in a variety of suspensions, and a quantitative analysis of the dynamics of the self-assembly of these nanowires under the control of external magnetic fields is described. Surface functionalization with porphyrin molecules yields fluorescent Ni nanowires that have potential for use in biotechnology and other applications.

Full text of DOI:10.1016/S0304-8853(02)00523-1

Journal of Magnetism and Magnetic Materials 249 (2002) 146–155
Electrodeposited magnetic nanowires: arrays, field-induced
assembly, and surface functionalization
C.L. Chiena,*, L. Sunb, M. Tanasea, L.A. Bauerc, A. Hultgrena, D.M. Silevitcha,
G.J. Meyerc, P.C. Searsonb, D.H. Reicha
a Department of Physics and Astronomy, Bloomberg Center, The Johns Hopkins University, 3400 N. Charles St.,
Baltimore, MD 21218, USA
b Department of Materials Science and Engineering, The Johns Hopkins University, Baltimore, MD 21218, USA
c Department of Chemistry, The Johns Hopkins University, Baltimore, MD 21218, USA
Abstract
Electrodeposition into nanoporous templates provides a means of fabricating large quantities of magnetic nanowires
with diameters in the range 5 nm–10 mm, and lengths up to 60 mm. Several recent developments in nanoporous templates
and Ni nanowires are described. Templates formed by a-particle irradiation of single-crystal mica are shown to contain
diamond-shaped nanopores with uniform size and orientation. Ni prisms deposited in these templates show anisotropic
magnetic properties along all three axes of the prisms. The manipulation of isolated Ni nanowires in a varietyof
suspensions, and a quantitative analysis of the dynamics of the self-assembly of these nanowires under the control of
external magnetic fields is described. Surface functionalization with porphyrin molecules yields fluorescent Ni
nanowires that have potential for use in biotechnologyand other applications.
r 2002 Elsevier Science B.V. All rights reserved.
PACS: 75.75.+a; 75.30.Gw; 75.60.Jk
Keywords: Nanowire arrays; Template fabrication; Anisotropy; Magnetic manipulation; Surface functionalization
1. Introduction
including the abilityto fabricate large quantities of
nanowires, and to introduce composition modula-
tion along the wires’ length [2–4]. The template
material maintains the nanowires within its thick-
ness and the nanowires are usuallyperpendicular
to the plane of the template. Byusing templates
with many, parallel-walled nanopores, arrays of
nanowires can be grown with wire diameters from
5 nm to 10 mm, wire lengths up to 60 mm, and areal
densities up to 1011 cmꢀ2 (limited bywire dia-
meter). However, since manyessential character-
istics of the nanowires, such as shape, size, and
Nanowires are quasi-one-dimensional entities
with large aspect ratios and lateral dimensions in
the 1–100 nm range. One approach to fabricating
such structures that is attracting increasing atten-
tion is electrodeposition into nanoporous tem-
plates [1]. This approach offers several advantages,
*Corresponding author. Tel.: +1-410-516-8092; fax: +1-
410-516-8260.
E-mail address: clc@eta.pha.jhu.edu (C.L. Chien).
0304-8853/02/$ - see front matter r 2002 Elsevier Science B.V. All rights reserved.
PII: S 0 3 0 4 - 8 8 5 3 ( 0 2 ) 0 0 5 2 3 - 1

C.L. Chien et al. / Journal of Magnetism and Magnetic Materials 249 (2002) 146–155
147
surface roughness, directlyreflect the properties of
2. Nanowire fabrication
the pores, templates with well-controlled pore
morphologyare crucial for nanowire fabrication
bythis technique.
Of particular interest are magnetic nanowires
made of ferromagnetic elements such as Co or Ni,
and alloys such as Permalloy. Due to the large
aspect ratio of the nanowires, their magnetic
properties are dominated byshape anisotrop.y
In fabricating nanowires byelectrodeposition,
one side of the porous template is usuallycoated
with a metal layer, which serves as a working
electrode in a three-electrode deposition cell. The
electrodeposition takes place inside the nanopores,
starting from the metal layer. While the principle
of the electrodeposition process is simple, the
qualityof the nanowires depends criticallyon the
detailed composition of the electrodeposition
solution and the deposition characteristics. The
Nickel nanowires used in the experiments de-
scribed here were grown from a solution of 20 g lꢀ1
NiCl2 ꢁ 6H2O, 515 g lꢀ1 Ni(H2NSO3)2) ꢁ 4H2O, and
20 g lꢀ1 H3BO3 at ꢀ1.0 V (Ag/AgCl).
After
a nanowire has been magnetized, its
magnetization remains along the wire’s long axis,
with onlytwo possible orientations. Such a
nanowire maybe verywell approximated as a
magnetic dipole, and depending on the geometrical
arrangement and the proximityof the nanowires,
inter-wire dipolar interactions can playan im-
portant role. In an arrayof magnetic nanowires in
a template, the dipolar interactions are negligible
for wires with low densities but substantial for
wires with high densities. For free magnetic
nanowires removed from the template and sus-
pended in a solution, dipolar forces between
nearbywires help determine the type of ordered,
multi-wire structures that can be assembled. Since
1993, when arrays of electrodeposited magnetic
nanowires were reported [1], manygroups have
activelypursued a varietyof magnetic nanowires,
see, e.g. Ref. [5]. Most of these efforts have been
directlyat magnetic nanowire arrays in templates,
but recently, manipulation of magnetic nanowires
in suspension has been reported [6]. Together with
the multifunctionalitymade possible bymulti-
section nanowires, freestanding magnetic nano-
wires provide manynew avenues for fruitful
exploration.
In this work, we describe several recent devel-
opments in nanoporous templates and Ni nano-
wires. Using single-crystal mica, we have created
arrays of diamond-shaped nanopores with uni-
form size and orientation. In Ni prisms deposited
in these templates, we have observed anisotropic
magnetic properties along all three axes of the
prisms. We also describe the manipulation of
isolated Ni nanowires in a varietyof suspensions,
and studies of interactions between these nano-
wires. Finally, we describe briefly surface-function-
alized, fluorescent Ni nanowires and various
prospects for future applications of nanowires.
2.1. Nanoporous templates
The qualityof the electrodeposited nanowires
also depends criticallyon the nanoporous tem-
plate, which defines the geometrical shape and
areal densityof the nanowires. A varietyof
templates with a range of pore sizes and pore
densities are available. These include polycarbon-
ate membranes [7] and mica crystals [8] containing
etched nuclear particle tracks, alumina templates
formed from anodic oxidation of aluminum [9],
and diblock co-polymers [10]. The pores in
alumina and in diblock co-polymers are partially
or highlyordered, and have veryhigh pore
densities (108–1011 cmꢀ2). On the other hand,
while pores made bynuclear track etching are
randomlylocated, the pore densitycan be varied
over a wide range bycontrolling the fluence of the
energetic nuclei. The template material also limits
the maximum operating temperature of the
nanowire array. In the case of organic templates,
this can be quite low.
Commerciallyavailable ultrafine filters (e.g.,
Nuclepore, Anodisc) can be and have been used
as templates, but theyare often less than
satisfactorysince their intended purpose is not
nanowire fabrication. The pore densities are
generallyquite high, and the resulting proximity
of neighboring nanowires can lead to strong inter-
wire interactions that can influence the magnetic
response of arrays. This is particularly true for

148
C.L. Chien et al. / Journal of Magnetism and Magnetic Materials 249 (2002) 146–155
commercial alumina templates. As for commercial
polycarbonate membranes, in addition to their low
operating temperature, their most severe problem
is that the pores are often not parallel. The pores
can be splayed more than 201 from the normal to
the template plane, severelycompromising the
interpretation of the arrays’ measured properties.
Manyaspects of the studyof magnetic nanowires
require templates with characteristics that are not
available commercially.
cross section, whereas those in mica have a
diamond shape. To fabricate pores in mica, a-
particles with energy6 MeV from a 100- mCi Cf252
source are used to form damage tracks in 5-mm
thick mica single crystals along the [0 0 1] direction.
The collimation of the a-particles, 51 in our case,
can be controlled bythe separation between the
source and the mica. The densityof the particle
tracks, and hence that of the nanowires in the
array, is proportional to the irradiation time, and
2
for our apparatus, the densityof pores per cm is
2.2. Diamond-shaped nanopores in mica
103 times the irradiation time in seconds [11]. The
damage tracks are then etched by20 wt% HF
(11.4 M). Because the etching rate along the tracks
is much faster than both the lateral etch rate and
The nuclear-track-etched nanopores in polycar-
bonate membranes are approximatelycircular in
Fig. 1. (a, b) Scanning electron microscope images of etched particle tracks in single-crystal mica. (c) Projection of the atomic
arrangement in the (0 0 1) plane of mica illustrating that the pore walls correspond to the oxygen-terminated planes in the mica
structure.

C.L. Chien et al. / Journal of Magnetism and Magnetic Materials 249 (2002) 146–155
149
that for undamaged mica, when the tracks are
etched through, nanopores with small cross section
and small taper angle (0.021) are created.
shape anisotropy. This anisotropy can be deter-
mined bymeasurements with the magnetic field
applied parallel (Hjj) and perpendicular (H>) to
the wires’ long axis. For Ni wires with circular
cross section, such as those grown in polycarbo-
nate membranes, there is no magnetic anisotropy
within the plane perpendicular to the wire axis [1].
An arrayof Ni nanowires grown in a single-
crystal mica template is a collection of aligned Ni
prisms, each of which can be defined bythe three
axes: c (along the wire), and a and b (along the two
diagonals of the diamond). The magnetic proper-
ties for fields applied in the plane perpendicular to
c are no longer isotropic, and the results for Ha
and Hb are significantlydifferent. We stress that
because the Ni prisms are all co-aligned and
because inter-wire interactions are negligible since
n is small, the magnetic properties measured from
an arrayof Ni prisms are those of a single Ni
prism.
Representative hysteresis loop of Ni prisms with
the magnetic field along the a-, b-, and c-axis are
shown in Fig. 2, demonstrating the anisotropy
between the three symmetry axes. The easy axis is
the c-axis, as the hysteresis loop along the c-axis
has the highest remnant magnetization Mr: As the
size of the Ni nanowires is decreased, both the
coercivityand remnant magnetization (expressed
as the squareness, defined as Mr=Ms; where Ms is
the saturation magnetization) increase as shown in
The most unusual aspect of the nanopores in
single-crystal mica is that all the pores are
diamond-shaped with inner angles veryclose to
601 and 1201 as shown in Figs. 1(a) and (b).
Furthermore, all of the pores in a given sample
have the same size and orientation [12]. By
correlating the results of scanning electron micro-
scopyand X-raydiffraction on the etched mica
crystals, we find that the orientation of the
diamonds is exactlythe same as that of the mica
unit cell. Furthermore, the four sides of the
diamonds are parallel to the four oxygen-termi-
nated planes within the unit cell as shown in
Fig. 1(c). This establishes that the diamond-shaped
pores have their origin in the mica crystal
structure. The uniform diamond shape, regardless
of pore size, arises because the oxygen-terminated
planes are those with the slowest etch rate, and the
pore axes are aligned because the template are
single crystals.
3. Magnetic properties of Ni nanowire arrays in
mica
The strong magnetic anisotropyfound in
ferromagnetic nanowires is due principallyto
Fig. 2. Hysteresis loops of an array of Ni nanowires d ¼ 100 nm and 5 mm long with the field applied along the a- and b-axis of the
diamond pores and along the wire (or c)-axis.

150
C.L. Chien et al. / Journal of Magnetism and Magnetic Materials 249 (2002) 146–155
Fig. 3. Coercivity(a) and squareness (b) of Ni nanowires in mica as a function of effective diameter with field applied along the wire
axis.
Fig. 3. Note that the squareness of Ni wires
deposited in commercial Nuclepore templates is
noticeablysmaller for the same wire diameter [1].
This is due to the splaying of the nanopores in the
commercial templates noted above.
A series of hysteresis loops of the Ni prisms have
been measured with the magnetic field applied at
an angle y with respect to the c-axis. The remnant
magnetization MrðyÞ; shown in Fig. 4, clearly
illustrates uniaxial anisotropywith a period of p:
The solid curve that describes excellentlythe
experimental data is not a fit but simply jcos yj:
This is because the wire axis is the magnetic easy
axis, and regardless of the direction of H; at H ¼
0; the magnetization will be along the wire
direction. The remnant magnetization MrðyÞ is
therefore just Mrð0Þcos y; the projection of the
remnant magnetization along the wire direction
onto the measured direction.
Within the plane of the diamond, the magnetic
Fig. 4. (Top) Remnance as a function of the angle y at which
the field is applied with respect to the wire direction. (Bottom)
properties are not isotropic. This is illustrated by
the saturation field as a function of the angle f
with respect to the a-axis. As shown in Fig. 4 the
Saturation field in the plane of the diamond cross section as a
function of y with respect to the a-axis.
saturation field shows
a uniaxial symmetry.
Because the magnetocrystalline anisotropy for Ni
is small, the measured anisotropyis predominantly
due to the diamond shape. The measured results
can be well described byassuming two uniaxial
anisotropyconstants Ka and Kb along the a- and b-
axis
4. Assembly and functionalization of suspended
magnetic nanowires
In addition to their unique properties in
templated arrays, magnetic nanowires have inter-
esting behavior when removed from the templates,
and placed in fluid suspensions. First, their large
magnetic shape anisotropyand high remnant
magnetization makes suspended nanowires highly
Hsat ¼ ð2Ka=MsÞcos2 f þ ð2Kb=MsÞsin2 f;
with Ka ¼ 1:13 ꢂ 106 and Kb ¼ 8:2 ꢂ 106 erg cmꢀ3
ð1Þ
.

C.L. Chien et al. / Journal of Magnetism and Magnetic Materials 249 (2002) 146–155
151
orientable and easilymanipulated in small external
4.2. Chain formation from suspended nanowires
magnetic fields. This can be used to control the
inter-wire dipolar forces, and hence to obtain
assemblyof ordered structures. In addition, the
nanowires can be surface-functionalized with a
varietyof organic molecules that can influence
how the nanowires interact with their environ-
ment. These properties give magnetic nanowires
potential for applications in biotechnologyand
separations chemistry. Some preliminary results
have been reported previously[6]. Here we
describe experiments that explore effects of differ-
ent fluids on the production of nanowire suspen-
sions, the assemblyof magnetic nanowires into
ordered chains, and the functionalization of
magnetic nanowires with fluorescent porphyrin
molecules.
The response of suspensions of nanowires to
magnetic fields was investigated under optical and
fluorescence microscopes. Even in the most viscous
solvents, the nanowires align parallel to the field,
and move in response to magnetic field gradients.
Controlled assemblyof the nanowires maybe
achieved byappliyng a small external magnetic
field Ho20 G to a nanowire suspension. Since the
initiallyisolated wires all align parallel to H; their
tendencytoward random aggregation is sup-
pressed, and instead the inter-wire interactions
lead to the formation of head-to-tail chains. This
occurs both in the suspension, and on the bottoms
of our glass experimental cells, where the chains
can extend over hundreds of microns [6]. We have
studied the dynamics of chain formation on the
cell bottoms via video microscopy. Fig. 5 shows
four video frames from the joining of two short
chains composed of two and three nanowires,
respectively, carried out in ethylene glycol at
T¼ 201C and H ¼ 4 G. From these data, the
chains’ locations can be determined for each
frame, and the end-to-end separation vs. time for
the event in Fig. 5 is shown in as Trace 3 in Fig. 6.
The large circles indicate the data points corre-
sponding to the frames in Fig. 5. Fig. 6 also shows
data for a second event in ethylene glycol (Trace
4), along with two events observed in water at
T¼ 351C (Traces 1 and 2). The separation vs. time
4.1. Preparation of nanowire suspensions
Cylindrical Ni nanowires were grown as de-
scribed in Section 2, using 50-mm-thick alumina
filter templates (Anodisc, Whatman, Inc.) with a
nominal minimum pore diameter of 100 nm. These
templates yield wires with average radius
a ¼ 180720 nm. The nanowires were removed
from the alumina templates byreaction in 0.5 M
KOH at 501C, and collected either byplacing a
small magnet on the side of the flask or by
centrifugation. The magnetic collection procedure
exposes the wires to fields in excess of 1 kG, and
permanentlymagnetizes them with their magnetic
moments parallel to the long axis. With either
procedure, after several rinsings the wires could be
resuspended in a varietyof solvents with sonica-
tion. Nanowires suspended in low viscosityliquids
such as water, ethanol, and isopropanol precipi-
tate in the course of several minutes. In addition,
aggregation occurs due to inter-wire magnetic
forces. To minimize aggregation and precipitation,
the nanowires were suspended in more viscous
media such as 1:1 hexadecane (C16H34):octadecane
(C18H38) mixtures or ethylene glycol. These
suspensions are much more stable. For example,
in hexadecane:octadecane the nanowires remained
suspended for periods of days at room tempera-
ture.
Fig. 5. Video microscopyimages showing attractive interaction
between two Ni nanowire chains in ethylene glycol, coaligned in
a 4 Oe external field.

152
C.L. Chien et al. / Journal of Magnetism and Magnetic Materials 249 (2002) 146–155
curves are qualitativelysimilar for all joining
events, but the significantlyhigher viscosityof
of an extended dipole with magnetic charges
7Qm ¼ 7Mpa2 separated by L; where M is the
wire’s magnetization. This approximation is accu-
rate to within 0.05% for distances greater than
5 mm from the wires’ ends. The magnetic force
between wires or chains is given bythe equivalent
Coulomb forces between these magnetic charges.
For all the events shown in Fig. 6 the wires or
chains are nearlyco-linear. In this case, the force
ethylene glycol (mEG ¼ 0:02 P at T¼ 201C; mH O
¼
2
7 ꢂ 10ꢀ3 P at T¼ 351C) results in much slower
motion. The chain lengths and physical parameters
for these events are given in Table 1.
All of the nanowire motion studied here occurs
at verysmall Renyolds number ( Ro10ꢀ5), and
therefore viscous drag is the dominant hydrody-
namic effect [13]. Thus, a nanowire or nanowire
between two wires or chains of lengths L1 and L2 is
ꢀ
chain’s velocityis
v ¼ F=D; where D is the
1
r2
1
1
f ðrÞ ¼ ꢀ Q2m
ꢀ
ꢀ
appropriate drag coefficient and F is the total
applied force [14]. For nanowires moving along a
flat substrate, the inter-wire magnetic forces are
the dominant contribution to F: At distances large
compared to the wire radius a; the magnetic field
of a wire or chain of wires of total length L is that
ðr þ L1Þ2 ðr þ L2Þ2
ꢁ
1
þ
;
ð2Þ
ðr þ L1 þ L2Þ2
where r is the end-to-end separation. The time-
dependence of rðtÞ is determined from the one-
*
dimensional equation of motion dr=dt ¼ Df ðrÞ;
*
where D ¼ D1D2=ðD1 þ D2Þ is the reduced drag
coefficient. We note that rðtÞ has simple limiting
cases of the form
rðtÞ ¼ r0ð1 ꢀ t=t0Þa
ð3Þ
when rbL1; L2; where a ¼ 15 (the limit of two point
dipoles), and when r5L1; L2; where a ¼ 13: This
latter case provides a good description of the data
for Event (2) in Fig. 6, but Eq. (3) does not
describe most data sets unless the exponent is
allowed to varyin the range 0 :24oao0:29:
The full equation of motion with Eq. (2) maybe
integrated analytically to give the inverse of rðtÞ in
the form
Fig. 6. Separation vs. time for four chain-formation events in a
4 Oe external field. Events (1) and (2) were in water, and events
(3) and (4) were in ethylene glycol. Large circles (a)–(d) on curve
(3) correspond to the four panels in Fig. 4. The solid lines are
fits described in the text and in Table 1.
tðrÞ ¼ kðF1ðL þ r0; L1; L2Þ ꢀ F1ðL þ r; L1; L2ÞÞ; ð4Þ
where L ¼ ðL1 þ L2=2Þ: The somewhat lengthy
expression for the function F1 is given in Ref. [6].
Table 1
Parameters describing the four joining events shown in Fig. 6
k (s mmꢀ5
)
*
D=m (mm)
*
Dcalc=m (mm)
Event
Solvent
L1 (mm)
L2 (mm)
r0 (mm)
1
2
3
4
H2O
H2O
EG
6.4
57.2
29.3
9.1
16.4
124.0
17.0
37.2
28.9
19.8
16.6
1.48 ꢂ 10ꢀ6
2.13 ꢂ 10ꢀ7
1.59 ꢂ 10ꢀ5
5.09 ꢂ 10ꢀ5
25.2
243.8
45.9
17.2
89.2
31.7
17.3
EG
10.2
27.5
*
*
L1;2 are the wire or chain lengths. r0 and k are fitting parameters, as described in the text. D=m and Dcalc=m are the ratio of the reduced
drag coefficient and the fluid viscosity, as derived from k; and calculated from standard hydrodynamics, respectively. The uncertainties
in L1; L2; and r0 are all 70.1 mm. The uncertaintyin k is 71%. See text for discussion of the drag coefficients.

C.L. Chien et al. / Journal of Magnetism and Magnetic Materials 249 (2002) 146–155
153
Fig. 7. Ni nanowires functionalized with Hematoporphyrin IX, and immobilized on a glass slide. Prior to immobilization, the
nanowires were suspended in ethanol (a, d), water (b, e), or ethylene glycol (c, f) for 5 days. The upper panels (a)–(c) show optical
images of the nanowires, while the lower panels (d)–(f) show fluorescence images of the same nanowires.
This yields a description of the motion in terms of
two parameters, the initial separation r0; and k ¼
D=QmL1L2; which measures the relative strength
results from non-uniformities in the pore sizes of
*
the alumina templates. Values of D=m are also
given in Table 1.
2
*
*
of the drag and magnetic forces. As shown bythe
solid curves through all the data sets in Fig. 6, this
model provides a good description for a wide
range of chain lengths, initial separations, and
fluid viscosities. The values of r0 and k determined
in these fits are given in Table 1.
As noted in Section 2, accurate determination of
the remnant magnetization of nanowires grown in
alumina is difficult due to inter-wire interactions
in the templates. However, extrapolating the data
in Fig. 3, we estimate that these nanowires
have MB0:7Msat; where Msat ¼ 485 Oe is the
room temperature, bulk saturation magnetization
of Ni. This allows us to obtain an estimate for
The verysimilar numbers obtained for D=m for
joining events such as (1) and (4), that involve
individual nanowires, suggest that the drag coeffi-
cient is proportional to the fluid viscosity. This
result is not unexpected, since in the limit where a
nanowire or chain is far from the cell bottom, its
drag coefficient is D ¼ 6pmr; where the effective
radius r maybe determined byapproximating the
wire or chain as a prolate ellipsoid [15]. Calculated
*
values Dcalc=m based on this approximation are
given in Table 1, and show good general agree-
ment with the measured values. Scaling with fluid
viscosityis again clearlyseen. Such experiments
could be refined in the future byusing nanowires,
such as those grown in mica templates, that have a
more uniform cross section, and with an improved
analysis of the drag forces that treats wire–
substrate interactions.
*
the reduced drag coefficient D from k; but we
caution that there is a nearly50% overall
uncertaintyin these numbers, arising principally
from variations in the nanowires’ radius that

154
C.L. Chien et al. / Journal of Magnetism and Magnetic Materials 249 (2002) 146–155
4.3. Surface functionalization of Ni nanowires with
fluorescent porphyrins
ited magnetic nanowires. As studies and applica-
tions of nanowires increase in sophistication, it will
become increasinglynecessaryto develop template
materials optimized for nanowire growth. The
highlyuniform diamond-shaped pores in single
crystal mica templates described here represent an
important first step toward achieving this goal, as
shown byour magnetic anisotropymeasurements
which demonstrate that the pore geometryaccu-
ratelydetermines the nanowires’ shape. In liquid
suspensions, magnetic nanowires offer new possi-
bilities for magnetic self-assembly, and by coating
them with organic molecules, additional function-
alities maybe introduced. This gives magnetic
nanowires potential for use in biotechnologyor
chemical separation, where their surface-deter-
mined properties such as binding, fluorescence,
or solubilitycan be combined with their magnetic
properties.
Byfunctionalizing the surfaces of suspended
nanowires with organic molecules, additional
control and enhancement of their properties and
behavior can be achieved. One example of this is
the modification of their optical properties by
coating with fluorescent chromophores. We note
that the production of luminescent non-magnetic
nanowires was recentlyreported [16a,b]. Here we
describe the surface functionalization of Ni nano-
wires with a fluorescent porphyrin.
Magnetic nanowires were functionalized with
fluorescent Hematoporphyrin IX, [8,13-bis(1-
hydroxyethyl)-3,7,12,17-tetra-methyl-21H,23H-por-
phine-2,18-dipropionic acid], which has two
intense, red fluorescence bands (lmax ¼ 626 and
696 nm in ethanol), and two carboxylic acid groups
that bind stronglyto metal oxides and to the native
oxide films on metals, such as nickel [17a,b].
Reaction of the nanowires with millimolar concen-
trations of Hematoporphyrin IX for 24 h in ethanol
at room temperature resulted in surface binding.
The carboxylic acid functional groups on
Hematoporphyrin IX most likely bind to the
nanowires as carboxylate linkages, or possibly as
ester links [17a,b]. In either case it is important to
characterize the stabilityof the surface linkages
under a varietyof conditions. The stabilitywas
tested in three common solvents, ethanol, ethylene
glycol, and water. The functionalized nanowires
were placed in the pure solvent for 5 days. They
were then collected magnetically, resuspended in
ethanol solution, and spun out onto glass slides.
Fig. 7 shows optical and fluorescence microscopy
images of single, porphyrin-functionalized Ni
nanowires. The observation of fluorescence along
all of the nanowires suggests a uniform surface
coverage and shows high stabilityin all three
solvents, indicating a wide range of applicabilityof
optical coatings of this type.
Acknowledgements
This work was supported byNSF Grant DMR-
0080031, AFOSR F49620-1-0384 and F49620-02-
1-0307.
References
[1] T.M. Whitney, J.S. Jiang, P.C. Searson, C.L. Chien,
Science 261 (1993) 1316.
[2] A. Blomdel, J.P. Meier, B. Boudin, J.-Ph. Ansermet, Appl.
Phys. Lett. 65 (1994) 3019.
[3] L. Piraux, et al., Appl. Phys. Lett. 65 (1994) 2484.
[4] K. Liu, N. Nagadawithana, P.C. Searson, C.L. Chien,
Phys. Rev. B 57 (1995) 7381.
[5] A. Fert, L. Piraux, J. Magn. Magn. Mater. 200 (1999)
338.
[6] M. Tanase, L.A. Bauer, A. Hultgren, D.M. Silevitch, L.
Sun, D.H. Reich, P.C. Searson, G.J. Meyer, Nano Lett. 1
(2001) 155.
[7] R.L. Fleischer, P.B. Price, R.M. Walker, Nuclear Tracks
in Solids: Principles and Applications, Universityof
California Press, Berkeley, CA, 1975.
5. Summary
[8] G.E. Possin, Rev. Sci. Instr. 41 (1970) 772.
[9] M. Shiraki, Y. Wakui, T. Tokushima, N. Tsuya, IEEE
Trans. Magn. 21 (1985) 1465.
These studies demonstrate some of the novel
and potentiallyuseful properties of electrodepos-
[10] T. Thurn-Albrecht, et al., Science 290 (2000) 2126.

C.L. Chien et al. / Journal of Magnetism and Magnetic Materials 249 (2002) 146–155
155
[11] L. Sun, P.C. Searson, C.L. Chien, J. Mater. Sci. 35 (2000)
1097.
[15] H. Lamb, Hydrodynamics, 6th Edition, Section 339,
Dover, New York, 1945.
[12] L. Sun, P.C. Searson, C.L. Chien, Appl. Phys. Lett. 74
(2000) 2803.
[16] (a) P.A. Smith, et al., Appl. Phys. Lett. 77 (2000) 1399.
(b) B.R. Martin, et al., Adv. Mater. 11 (1999) 1021.
[17] (a) P.V. Kamat, M.A. Fox, Chem. Phys. Lett. 102 (1983)
379.
[13] L. Landau, E. Lifshitz, Fluid Mechanics, Pergamon,
London, 1959.
[14] R.J. Hunter, Introduction to Modern Colloid Science,
Oxford, New York, 1993.
(b) K. Kalyanasundaram, et al., J. Phys. Chem. 91 (1987)
2342.