Manifestations in the magnetization of the hcp-Co nanowires due to interdependence of aspect ratio and c-axis orientation

Manifestations in the magnetization of the hcp-Co nanowires due to interdependenceof aspect ratio and c-axis orientationDaljit Kaur, Sujeet Chaudhary, and D. K. Pandya Citation: Journal of Applied Physics 114, 043909 (2013); doi: 10.1063/1.4816558 View online: http://dx.doi.org/10.1063/1.4816558 View Table of Contents: http://scitation.aip.org/content/aip/journal/jap/114/4?ver=pdfcov Published by the AIP Publishing

[This article is copyrighted as indicated in the abstract. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to ] IP:

180.149.52.49 On: Thu, 14 Nov 2013 04:37:32

Manifestations in the magnetization of the hcp-Co nanowires due tointerdependence of aspect ratio and c-axis orientation

Daljit Kaur, Sujeet Chaudhary, and D. K. Pandyaa)

Thin Film Laboratory, Physics Department, Indian Institute of Technology Delhi, Hauz Khas,New Delhi 110 016, India

(Received 22 June 2013; accepted 9 July 2013; published online 25 July 2013)

Effect of interdependence of aspect ratio (AR) and c-axis orientation of the hcp-cobalt nanowires

(NWs) on their magnetization behavior is reported in 40 and 100 nm diameter NWs. Experimental

evidence of periodically modulated magnetic state viz. large transverse-susceptibility arising due to

orientation of c-axis normal to NW-axis in 40 nm NWs and magnetic domain imaging is

demonstrated, which disappears at low AR owing to randomly oriented c-axes. The 100 nm

NWs exhibit a crossover in the easy-axis direction from longitudinal at high AR to transverse at

low AR and are explained on the basis of competition between different anisotropic contributions.VC 2013 AIP Publishing LLC. [http://dx.doi.org/10.1063/1.4816558]

I. INTRODUCTION

The magnetic response of ferromagnetic nanowire

arrays is being explored recently due to their diverse nano-

scale spintronics applications such as in patterned perpendic-

ular magnetic storage media, racetrack memory devices, spin

valves based devices, spin torque oscillators, and high reso-

lution magnetic sensors.1–6 There is a significant interest in

cobalt nanowires (NWs)4–6 because of the strong uniaxial

magnetocrystalline anisotropy along the c-axis of its hexago-

nal unit cell. Different methods are used to fabricate NWs

but electrodeposition stands out as a simple and efficient

method to deposit NWs in commercially available polycar-

bonate membranes.7 Interestingly, the c-axis orientation of

the Co NWs vis-a-vis NW-axis can be manipulated by

proper choice of electrodeposition parameters.8–11 Henry

et al. and Ferre et al. have shown that the c-axis lies parallel

to the NW-axis for wire diameters 2 R< 50 nm and per-

pendicular to the NW-axis for wire diameters of

2 R> 50 nm.12,13 The possibility of having an angle h> 0�

between c-axis (which also happens to be the easy-axis for

magnetization in cobalt) and the NW-axis and control of the

aspect ratio (length/diameter) of NWs opens an interesting

way to control their magnetization switching as rotation of

c-axis can allow a varying degree of competition between

magnetocrystalline anisotropy and shape anisotropy. In this

context, theoretical studies by Bergmann et al.14 and

Erickson et al.15 have predicted that in thinner Co NWs

(radius 40–50 nm) with c-axis oriented perpendicular to

NW-axis, a complex ground state develops in which magnet-

ization (M) gets frustrated. This is manifested in the form of

a “snake-like” ground-state with regards to M which exhibits

sinusoidal modulation along NW-axis so as to minimise the

free energy by decreasing the demagnetization energy.

These authors showed that the state is characterised by

higher transverse dc-susceptibility at low applied field

strengths.14,15 Remarkably, it is pointed out that if such a

state is realized in Co NWs then it should be possible to

tune/alter the modulation period by application of a modest

magnetic field thereby opening up avenues for various tech-

nological applications of NWs involving such a spin wave

character.16 There exists two experimental reports,17,18

where such a modulated domain state is evidenced via the

magnetic force microscopy (MFM) studies on Co NWs.

Whereas one report17 shows spatial modulation of M along

NW length, the other18 shows magnetic vortices with alter-

nating chirality along the wire-axis. However, the signature

of high transverse dc-susceptibility in the NWs is yet to be

experimentally demonstrated. This is primarily due to the

fact that there has been no detailed study of the texture con-

trol in the thinner Co NWs with diameter <50 nm. Recently,

we have used bath temperature as an effective parameter to

structurally tailor the c-axis orientation in the electrodepos-

ited hcp-Co NWs of diameter 100 nm.19,20 In the present

work, we have been able to fabricate Co NWs (2 R< 50 nm)

with c-axis perpendicular to NW-axis and have demonstrated

the occurrence of large transverse magnetization of 0.54 MS

(where MS is the saturation magnetization) at modest field

strength. In addition, the longitudinal and transverse magnet-

ization loops for 40 nm and 100 nm NWs are compared for

three different values of the aspect ratio.

II. EXPERIMENTAL DETAILS

Commercially available polycarbonate templates

(PCTs) with nominal pore diameter of 30 nm and 50 nm and

pore length of 6 lm were coated with 150 nm gold on one

side that serves as cathode for electrodeposition in the pores.

The cobalt NWs were then grown from an electrolytic bath

with the composition: 25 mM CoSO4�7H2O, 40 mM H3BO3

in 100 ml deionized water. Depositions were performed in a

three electrode cell using computer controlled Potentiostat

(Model CHI-1100 A). The exposed geometrical membrane

area of deposition was 0.07 6 0.01 cm2. Deposition is

stopped when the desired wire length is reached, which was

estimated from the total integrated charge passing between

the working electrode and the counter electrode. The bath

a)Author to whom correspondence should be addressed. E-mail: dkpandya@

physics.iitd.ac.in

0021-8979/2013/114(4)/043909/5/$30.00 VC 2013 AIP Publishing LLC114, 043909-1

JOURNAL OF APPLIED PHYSICS 114, 043909 (2013)

[This article is copyrighted as indicated in the abstract. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to ] IP:

180.149.52.49 On: Thu, 14 Nov 2013 04:37:32

pH was maintained at 4.5 and all depositions were taken at a

constant potential of �1.0 V. Six samples of the Co NWs of

lengths varying from 2500 nm to 100 nm in the two diameter

series were prepared under conditions as represented in

Table I. On selected samples, high-resolution transmission

electron microscope (HRTEM) images were acquired using

JEOL JEM 2100 F microscope equipped with electron gun

operated at 200 kV. Magnetization measurements were done

with vibrating sample magnetometer (VSM) and supercon-

ducting quantum interference device (SQUID) magnetome-

ter. The magnetization measurements were done on the

NW-array keeping the NWs embedded in the membrane

itself. The magnetic state of Co NWs was studied using the

MFM technique (Veeco Digital Instruments) employing

SmCo tip at a lift height ranging in 15–25 nm.

III. RESULTS AND DISCUSSION

Figures 1–3 show the TEM images of the samples S1,

S3, and S4, respectively. The lengths of the NWs are found

to be nearly 2500 nm in S1 and S4 while the length in sample

S3 varies from 60 nm to 125 nm with an average length of

100 nm. The diameter of NWs deposited in PC-30 membrane

(S1-S3 samples) is found to be �40 nm as observed in

Figs. 1(a) and 2(a) for samples S1 and S3, respectively. On

the other hand, the average diameter of NWs deposited in

PC-50 membrane (S4-S6 samples) is found to be 100 nm as

shown in Fig. 3(a) for sample S4. The sample S1 exhibits a

highly crystalline nature of the NW, such that the lateral size

of the crystallite is the NW diameter itself. As revealed from

the orientation of ð10�10Þ planes parallel to NW diameter, the

c-axis of hcp-cobalt in the S1 sample is oriented along the di-

ameter of the NWs.

However, as the length of NW is reduced from 2500 nm

(S1) to �100 nm (S3), the crystallite size is reduced and sev-

eral such crystallites can be clearly seen spanning the entire

NW diameter (cf. the HRTEM image of S3 NW, Fig. 2(b)).

For clarity and conciseness, the orientation of hcp-Co ð10�10Þplanes in two nearby crystallites is highlighted in Fig. 2(b),

which suggests that the c-axis in the S3 NW-sample takes

various possible orientations ranging from along the NW-

axis to along the NW-diameter. Thus, absence of preffered

orientation in the NWs of smaller aspect ratio is established

from the HRTEM studies.

Figure 3 shows one of the TEM micrographs recorded

on the S4 NWs (length 2500 nm, diameter 100 nm). Since

the diameter of the NW is large, the lattice planes could not

be clearly seen at higher magnification. However, from the

selected area diffraction (SAD) pattern shown as inset to Fig.

3(b), the weakly textured nature of the S4 NWs is confirmed

from the presence of polycrystalline rings corresponding to

different hcp-Co planes superposed with diffraction spots.

Figures 4(a)–4(c) show the hysteresis loops of 40 nm

NWs samples with field applied in parallel (longitudinal) and

perpendicular (transverse) direction to the NW-axis. By

comparing the two hysteresis loops for sample S1 shown in

TABLE I. Deposition time and calculated size for thinner and thicker cobalt

nanowire samples.

Sample

NW

diameter

(nm)

Deposition

time

(s)

NW

length

(nm)

Aspect

ratio

S1 40 1300 2500 63

S2 40 500 1000 25

S2 40 50 100 2.5

S4 100 360 2500 25

S5 100 140 1000 10

S6 100 15 500 5

FIG. 1. (a) TEM image of sample S1

40 nm diameter nanowires extracted

after dissolving PC membrane. (b) and

(c) HRTEM images of a single NW

showing the growth of ð10�10Þ planes

perpendicular to the nanowire-axis.

White arrows in (b) and (c) indicate

direction of NW-axis.

043909-2 Kaur, Chaudhary, and Pandya J. Appl. Phys. 114, 043909 (2013)

[This article is copyrighted as indicated in the abstract. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to ] IP:

180.149.52.49 On: Thu, 14 Nov 2013 04:37:32

Fig. 4(a), we observe that the NWs are more easily saturated

in transverse direction, while a relatively gradual and non-

saturating nature of M(H) is observed in the longitudinal

direction. There is also a large difference in the coercivity

(HC) values observed along the two applied field directions

(HllC¼ 700 Oe and H?C ¼ 160 Oe). This is due to the domi-

nance of shape anisotropy (SA) energy over the magneto-

crystalline anisotropy (MCA) energy of the NWs which tries

to maintain magnetisation M along the NW-axis, hence a

larger coercivity is observed along the longitudinal direction.

A closer look at the low field magnetization reveals that S1

possesses relatively large M at lower H (in transverse mag-

netization, H ? NW-axis).

As mentioned previously, the larger transverse dc-

susceptibility of S1 could be the indication of presence of the

so-called snake-state. Our S1 NWs sample fulfils both the

necessary microstructural requirements,15 i.e., NW diameter

<50 nm and the c-axis is normal to NW-axis. In their work,

the authors predicted that for such Co NW (radius¼ 40 nm),

the snake state will have lowest free energy till a field of

5500 Oe, above which the magnetization will cease to show

any modulation. As a signature of the modulated M in the

snake state, the relative magnetization values (M/MS) of our

NWs at two field strengths of 1% and 5% of 4pMS agree

excellently well to the predicted values. Our M/MS (field)

values are 0.28 (144 Oe) and 0.54 (740 Oe) as determined

from Fig. 4(a) (H ? NW), compared to 0.15 (173 Oe) and

0.46 (865 Oe) as reported in Ref. 15. In view of the lower

radii (20 nm) of NWs in the present work, it is concluded

that the magnetization behavior of S1 NWs (Fig. 4(a))

clearly provides experimental support in the form of high

transverse magnetic susceptibility as a signature of periodi-

cally modulated snake state in the NWs. The equivalent sam-

ple S4 of larger NW diameter of 100 nm obviously shows no

such signature.

To examine the ground domain state or remnant state,

MFM measurements were carried out on various 40 nm NWs

of sample S1 after extracting them from the membrane using

dichloromethane as solvent. Figure 5 shows one of the topo-

graphical AFM images of a NW and its corresponding MFM

image recorded at a lift height of 22 nm. Although the pres-

ence of un-dissolved template-residues affected the image

contrast, the presence of alternate bright and dark regions

along the NW’s length is clearly evident. Similar periodic

variation in the magnetization vector M along the NW length

is seen in the images of Liu et al.17 MFM images confirm the

existence of the expected “snake state” in the NWs of the

sample with c-axis perpendicular to the NW-axis.

As the aspect ratio decreases, the magnetization reversal

in 1000 nm long S2 NWs (Fig. 4(b)) is almost similar in both

the transverse and the longitudinal directions. It may be

noted that compared to S1 NWs, the HC in S2 NWs is greatly

reduced only in longitudinal direction (HllC¼ 350 Oe and

H?C ¼ 150 Oe). This indicates effective decrease in SA

energy on account of decrease in aspect ratio. The changes

in the “snake state” due to shorter NWs (e.g., S2, S3) can

also be understood by comparing Figs. 4(a) and 4(b). It was

argued in Ref. 15 that the period of modulation in M gets

reduced as the length of the NW is reduced and this in turn

tends to destabilise or suppress the “snake state” eventually.

FIG. 2. (a) TEM image of sample

S3 40 nm diameter nanowires after

dissolving the PC membrane. (b)

HRTEM image of a single NW show-

ing a number of small grains along the

nanowire diameter. White arrow in (b)

indicates direction of NW-axis.

FIG. 3. (a) TEM image and (b)

indexed SAD pattern of sample S4

100 nm diameter nanowires showing

the polycrystalline growth.

043909-3 Kaur, Chaudhary, and Pandya J. Appl. Phys. 114, 043909 (2013)

[This article is copyrighted as indicated in the abstract. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to ] IP:

180.149.52.49 On: Thu, 14 Nov 2013 04:37:32

In the present case, as we decrease the length to 1000 nm in

S2 and 100 nm in S3, the signature of higher transverse mag-

netic susceptibility vanishes, so much so that the transverse

and longitudinal magnetization behaviors become identical.

In S3 NWs, the observed coercivity and remanence values

are almost equal along both the directions (Fig. 4(c)). The

HC values are very small and there is non-saturating nature

of both the M-H loops which is due to the fact that the Co

grains are small sized possessing superparmagnetic (SPM)

nature, which is indeed confirmed from the HRTEM image

of S3 (Fig. 2(b)).

Figures 4(d)–4(f) show the transverse and longitudinal

hysteresis loops recorded on the thicker NWs samples (diam-

eter 100 nm) of different lengths (S4-S6). In sharp contrast to

thinner NWs sample S1 of similar aspect ratio, thicker S4

NWs with highest aspect ratio get easily saturated in the

longitudinal direction than transverse direction and the

FIG. 4. M-H loops with magnetic field applied along longitudinal and transverse directions to the NW-axis. (a)–(c) 40 nm diameter NW samples with lengths:

(a) 2500 nm, (b) 1000 nm, (c) 100 nm; and (d)–(f) 100 nm diameter NW samples with lengths: (d) 2500 nm, (e) 1000 nm, (f) 500 nm.

FIG. 5. (a) Topographical AFM image recorded on 40 nm diameter NW

extracted from sample S1 and (b) its MFM image showing alternate dark

and bright regions indicating the periodic modulation of magnetization.

Some small broken NW portions are also seen in the image.

043909-4 Kaur, Chaudhary, and Pandya J. Appl. Phys. 114, 043909 (2013)

[This article is copyrighted as indicated in the abstract. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to ] IP:

180.149.52.49 On: Thu, 14 Nov 2013 04:37:32

coercivity is similar in both the directions (HllC¼ 620 Oe and

H?C ¼ 630 Oe). As the length of the NWs is decreased to

1000 nm in sample S5, the coercivity as well as the rema-

nence in the longitudinal direction are increased while these

are decreased in the transverse direction (HllC¼ 800 Oe and

H?C ¼ 540 Oe). Given the polycrystalline nature of these

100 nm diameter NWs (as revealed by SAD pattern of this

sample in Fig. 2), the increased area under the M-H loop in

S5 sample indicates relatively higher domain wall pinning

compared to S4 sample. This is consistent with slightly

higher coercivity and remanence in longitudinal case of S5.

It should be noted that in samples S4 and S5 having aspect

ratio >10, the easy-axis is longitudinal due to dominance of

shape anisotropy.

However, on further decreasing the length to 500 nm in

sample S6 having aspect ratio of �5, we observe a large

reduction in the coercivity and remanence values in both the

longitudinal and transverse directions. Also, the sample S6

shows the tendency of early saturation in the transverse

direction than in the longitudinal direction exhibiting the

rotation of the easy-axis towards the transverse direction at

this smaller aspect ratio. As the samples S4-S6 are polycrys-

talline in nature, the MCA has no contribution. Thus, the ani-

sotropy could be due to shape anisotropy and/or dipolar

interaction among the NWs. Assuming a MS value of bulk

Co, the shape anisotropy field is �9 kOe. Thus for 2500 nm

long NWs (at high aspect ratio), the SA energy alone con-

trols the easy-axis of magnetisation which is seen to lie in

the longitudinal direction. On decreasing the length of the

NWs, the SA energy decreases, and the magnetostatic energy

due to interaction from the neighbouring wires increases.21

Since the magnetostatic interaction favours the easy-axis to

be transverse, the decreasing aspect ratio would switch from

longitudinal to transverse direction. With the porosity of

�10% in our PCT membranes (pore density �109 cm�2),

the dipolar field due to the interaction among the neighbour-

ing NWs is �2.7 kOe.22 The dominance of the in-plane

interaction field among the NWs thus accounts for the satura-

tion of the magnetization in the transverse direction in S6

sample, as is observed in Fig. 4(f).

IV. CONCLUSIONS

In conclusion, the effect of aspect ratio on both structural

and magnetic properties of thinner and thicker hcp-Co NWs

is studied. When c-axis is oriented perpendicular to the NW-

axis, we have demonstrated the existence of the “snake state”

via the presence of large transverse-susceptibility and the spa-

tial periodic modulation of magnetization vector along the

NW length. This could be useful for spin wave propagation

device applications. With reduction in aspect ratio, this mag-

netically modulated sate is shown to be suppressed consistent

with the randomness in c-axis orientation with regard to NW-

axis. The effective anisotropic contributions like shape anisot-

ropy and magnetocrystalline anisotropy are responsible for

the anisotropic behavior in NWs at different lengths. The

easy saturation along the longitudinal direction in the higher

aspect ratio thicker NWs of 100 nm diameter is an expected

outcome of the combined effect of their polycrystalline

microstructure and predominant shape anisotropy aided by

the magnetic interaction due to neighbouring NWs. The mag-

netic easy-axis changes to transverse direction at lower

lengths (500 nm) due to dominance of magnetostatic dipolar

interaction among the NWs in the array over the shape anisot-

ropy energy.

ACKNOWLEDGMENTS

One of the authors D.K. wants to acknowledge Council

of Scientific and Industrial Research, New Delhi for Senior

Research Fellowship. We would also like to acknowledge

Department of Science and Technology for magnetization

measurements through their national (MPMS XL-7) SQUID

facility at I.I.T. Delhi and Deepak Varandani for MFM

measurements.

1J. E. Wegrowe, D. Kelly, A. Franck, S. E. Gilbert, and J.-Ph. Ansermet,

Phys. Rev. Lett. 82, 3681 (1999).2J. Topp, D. Heitmann, M. P. Kostylev, and D. Grundler, Phys. Rev. Lett.

104, 207205 (2010).3S. S. P. Parkin, M. Hayashi, and L. Thomas, Science 320, 190 (2008).4L. G. Vivas, J. Escrig, D. G. Trabada, G. A. Badini-Confalonieri, and M.

V�azquez, Appl. Phys. Lett. 100, 252405 (2012).5A. V. Davydenko, A. V. Ognev, E. V. Pustovalov, and L. A.

Chebotkevich, Appl. Phys. Lett. 101, 252412 (2012).6A. Fern�andez-Pacheco, L. Serrano-Ram�on, J. M. Michalik, M. R. Ibarra,

J. M. De Teresa, L. O’Brien, D. Petit, J. Lee, and R. P. Cowburn, Sci. Rep.

3, 1492 (2013).7T. M. Whitney, J. S. Jiang, P. C. Searson, and C. L. Chien, Science 261,

1316 (1993).8N. B. Chaure, P. Stamenov, F. M. F. Rhen, and J. M. D. Coey, J. Magn.

Magn. Mater. 290, 1210 (2005).9M. Darques, A. Encinas, L. Vila, and L. Piraux, J. Phys. D: Appl. Phys.

37, 1411 (2004).10Y. Ren, J. Wang, Q. Liu, Y. Dai, B. Zhang, and L. Yan, J. Mater. Sci. 46,

7545 (2011).11J. Zhang, G. A. Jones, T. H. Shen, S. E. Donnelly, and G. Li, J. Appl.

Phys. 101, 054310 (2007).12Y. Henry, K. Ounadjela, L. Piraux, S. Dubois, J. M. George, and J. L.

Duvail, Eur. Phys. J. B 20, 35 (2001).13R. Ferr�e, K. Ounadjela, J. M. George, L. Piraux, and S. Dubois, Phys. Rev. B

56, 14066 (1997).14G. Bergmann, J. G. Lu, Y. Tao, and R. S. Thompson, Phys. Rev. B 77,

054415 (2008).15R. P. Erickson and D. L. Mills, Phys. Rev. B 80, 214410 (2009).16Z. H. Wang, V. L. Zhang, H. S. Lim, S. C. Ng, M. H. Kuok, S. Jain, and

A. O. Adeyeye, ACS Nano 4, 643 (2010).17Z. Liu, P. Chang, C. Chang, E. Galaktionov, G. Bergmann, and J. G. Lu,

Adv. Funct. Mater. 18, 1573 (2008).18Y. P. Ivanov, L. G. Vivas, A. Asenjo, A. Chuvilia, O. Chubykalo-Fesenko,

and M. Vazquez, Europhys. Lett. 102, 17009 (2013).19D. Kaur, D. K. Pandya, and S. Chaudhary, J. Electrochem. Soc. 159, D713

(2012).20D. Kaur, S. Chaudhary, D. K. Pandya, R. Gupta, and R. K. Kotnala,

J. Magn. Magn. Mater. 344, 72 (2013).21J. Escrig, R. Lavın, J. L. Palma, J. C. Denardin, D. Altbir, A. Cortes, and

H. Gomez, Nanotechnology 19, 075713 (2008).22A. Encinas-Oropesa, M. Demand, L. Piraux, I. Huynen, and U. Ebels,

Phys. Rev. B 63, 104415 (2001).

043909-5 Kaur, Chaudhary, and Pandya J. Appl. Phys. 114, 043909 (2013)

[This article is copyrighted as indicated in the abstract. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to ] IP:

180.149.52.49 On: Thu, 14 Nov 2013 04:37:32