Controlled growth of Co nanoparticle assembly on nanostructured template Al

(1)

Controlled growth of Co nanoparticle assembly on nanostructured template Al

₂

O

₃

/ NiAl „100…

Wen-Chin Lin

Department of Physics, National Taiwan University, 10617 Taipei, Taiwan and Institute of Atomic and Molecular Sciences, Academia Sinica, 10617 Taipei, Taiwan

Shen-Shing Wong, Po-Chun Huang, Chii-Bin Wu, Bin-Rui Xu, Cheng-Tien Chiang, and Hong-Yu Yen

Department of Physics, National Taiwan University, 10617 Taipei, Taiwan Minn-Tsong Lin^a兲

Department of Physics, National Taiwan University, 10617 Taipei, Taiwan and Institute of Atomic and Molecular Sciences, Academia Sinica, 10617 Taipei, Taiwan

共Received 18 May 2006; accepted 16 August 2006; published online 10 October 2006兲

Based on the systematic studies of the growth temperature, deposition rate, and annealing effects, the control of Co nanoparticle density, size, and alignment is demonstrated to be feasible on a nanostructured template Al₂O₃/ NiAl共100兲. At 140–170 K, a slow deposition rate 共0.027 ML/min兲 promises both the linear alignment and the high particle density. 1.5 ML Co nanoparticle assembly sustains the density of⬃260/10⁴nm²even after 800– 1090 K annealing. This study also indicates the possibilities of the controlled growth for nanoparticles of different materials. © 2006 American Institute of Physics. 关DOI:10.1063/1.2358926兴

Due to the potential applications in electronic and magnetic nanodevices as well as for catalysts in growing further various nanomaterials such as nanotubes or nanowires, many methods are developed in recent years for the fabrication of nanoparticle assembly. Because of the high cost and resolu- tion limit of conventional lithography, self-organized ap- proach of growing metallic materials on insulating substrates has been realized to be a promising way for growing nan- odots with uniform size.^1–12 As reported in previous studies,^6–8,13 the single-crystalline Al₂O₃ layer grown on NiAl共100兲 through high temperature oxidation provides a su- perior template for the nanoparticle assembly. Co nanoparticles can be prepared by thermal deposition on the single- crystalline Al₂O₃/ NiAl共100兲 with such features as uniform size distribution, well-ordered alignment 共interdistance

⬃4 nm兲, and high thermal stability. Since the nanoparticles are self-organized, it is interesting to ask if there are any possibilities to control the growth of nanoparticles for the various intentions. For example, high particle density with proper particle size might be useful for catalyzing the oxidation of carbon monoxide.³ High particle density with good alignment might lead to the formation of nanowires, which are expected to reveal the electronic and magnetic properties different from bulk and thin film materials. In this letter, we take Co as an exemplification to study the various effects of growth temperature, deposited coverage, deposition rate, and thermal annealing. Since not only Co but also Fe, Cu, etc., have been shown to reveal the similar growth mode in the particle size and alignment, on the Al₂O₃/ NiAl共100兲,⁶ the detailed understanding of the controlled growth of Co nanoparticles in this letter might be a good example for other nanoparticle assembly composed of different materials in the further research and application.

The experimental apparatus and the preparation of Al₂O₃/ NiAl共100兲 template are described in our previous

report.⁶ Since the deposition rates were calibrated from the epitaxial growth on Cu共100兲,⁷1 ML was defined as the atom density on Cu共100兲 surface: 1.54⫻10¹⁵at./ cm². Due to the limitation of instruments, the sample is prepared in a preparation chamber connected to scanning tunneling microscope 共STM兲 chamber. Thus it is unavoidable to recover the room temperature共RT兲 during the sample transportation. The mor- phology of the nanoparticle assembly was investigated by STM at RT, with the bias voltage of 1.6– 2.0 V and the tunneling current of 0.8– 1.0 nA.

Figures 1共a兲 and1共b兲 exhibit the STM images of 0.08

a兲Electronic mail: [email protected]

FIG. 1.共Color online兲共a兲 and 共b兲 are STM images of 0.08 and 0.15 ML Co on Al₂O₃/ NiAl共100兲 grown at various temperatures with the deposition rate= 0.027 ML/ min.共c兲 The statistics of nanoparticle density as functions of growth temperature共TG兲 with different coverages and deposition rates.

共F兲 and 共S兲 denote the fast and low deposition rates of 0.27 and 0.027 ML/ min, respectively.共d兲 Logarithm of 0.15 ML Co particle density as a function of 10³/ T with the Arrhenius fitting lines.

APPLIED PHYSICS LETTERS 89, 153111共2006兲

Downloaded 20 Oct 2006 to 140.112.101.80. Redistribution subject to AIP license or copyright, see http://apl.aip.org/apl/copyright.jsp

(2)

and 0.15 ML Co nanoparticles at different growth temperatures 共TG兲共deposition rate=0.027 ML/min兲. Figure 1共c兲 summarizes the statistics of particle density as functions of TGwith different coverages共0.08 and 0.15 ML兲 and different deposition rates共0.27 and 0.027 ML/min兲. For 300–400 K growth, the statistics of the four different growth conditions are very similar. When艋200 K, the curves can be separated into two groups: 0.15 and 0.08 ML Co. For 0.15 ML Co, the deposition rate effect is shown to be minor and the particle density increases monotonically with reducing TG. But for 0.08 ML Co, the particle density is nearly invariant with reducing T_G. In general, the mobility of nanoparticles strongly depends on their sizes. The small ones can move more easily than large ones due to the less binding to the substrate. Therefore high TG, which means high thermal energy, might activate small particles moving more easily than low T_G, resulting in the reduce of particle density. Thus in- tuitively, the particle density should decrease monotonically with increasing TG, as the case of 0.15 ML Co. Apparently 0.08 ML Co does not follow the expected trend. The reason might be that even the high particle density can be prepared at low T_G, warming up to RT will trigger the diffusion of the small particles to form large ones. The 0.15 ML deposition at low TGprovides more Co atoms to enlarge and stabilize the particles with high density, so that the high density at low TG

survives after warming up to RT for 0.15 ML deposition and routs for 0.08 ML.

For the case of irreversible nucleation, both the mean field theory and Monte Carlo studies predict the particle den- sity N to be of the Arrhenius form.^14,15 N⬀exp共Em/ kT兲, where Emis the activation energy. The proportional factor is a function of the coverage, deposition rate, and diffusion coefficient. The Arrhenius form of N can be extended for two-dimensional and three-dimensional island growths, with different proportional factors.^14,15 In Fig. 1共d兲, for 0.15 ML Co, the particle density between 300– 140 K indeed follows the Arrhenius form. There exists a transition between 300 and 400 K关Fig.1共d兲兴. Similar phenomena are also reported in other systems.^14–16It is usually attributed to the transition between different growth mechanisms, for example, from the single atom diffusion to the moving and coalescence of small particles.¹⁶From the fitting in Fig.1共d兲, E_mis deduced to be 10.1± 1.8 and 12.9± 2.0 meV for the slow共0.027 ML/s兲 and fast 共0.27 ML/s兲 deposition rates, respectively. The Em

⬃10 meV is quite small as compared to the studies of metal islands on metal,^14–16in which Em⬃1–0.4 eV. Nevertheless, similar results of low E_mare reported in the studies of metal on oxide^17–19 or nanostructured surface.²⁰ There are two common features in these studies, as well as in this work.共1兲 The Emis relatively small as compared with metal on metal.

It also means that the particle density is weakly affected within one order of magnitude as T_Gis changed between 300 and 100 K.^17–20In the cases of metal on metal, particle den- sity N can be easily enhanced by one to two orders of mag- nitude when TG varies from 100 to 300 K.^14–16 共2兲 The island density depends weakly on the deposition rate when T_G⬍300–400 K. In the previous studies of Cu/TiO2,¹⁷Au/

armorphous Al₂O₃,¹⁸ as well as this work, for an order of magnitude increase in deposition rate, the island density increases by⬍10%. In the studies of metal/oxide^17–20or nanostructured surface,²⁰the surface defects are shown to play an important role in the adatom diffusion and particle density.

For the single-crystalline Al₂O₃/ NiAl共100兲, the periodical

stripes with the interdistance⬇4 nm, which reveal signifi- cant attraction for Co atoms, can also confine the diffusion of adatoms or small particles. It prevents or delays the very low particle density with increasing TG.

Although in Fig.1the deposition rate reveals very minor effects on the particle density than coverage, it is shown to play a critical role in the linear alignment of nanoparticles in Fig.2. For 0.15 ML Co deposited with a slow deposition rate 共0.027 ML/min兲 at 170 K, the nanoparticles are well aligned, following the stripes of the Al₂O₃/ NiAl共100兲 template. In contrast, a fast deposition rate 共0.27 ML/min兲 results in disorder of the particle alignment. Although the same amount of Co atoms共0.15 ML兲 is deposited, the disorder of particle alignment, higher particle density, as well as the en- larging effect from STM tip make Fig. 2共b兲 seems to have higher Co coverage than Fig.2共a兲, which is actually not true.

At lower TG, the diffusion rate of deposited atoms is slower.

Thus the deposited atoms might meet each other and create nucleations before arriving the Al₂O₃ stripes, incurring disorder of particle arrangement, while a slow deposition rate prevents the deposited atoms from creating nucleations before arriving the stripes of oxide layer. Figure2共c兲summa- rizes the density of disordered particles as functions of TG. The “disordered particles” here are defined as the nanoparticles which are not aligned by the Al₂O₃stripes. Apparently 0.15 ML Co deposited at 160 K with 0.27 ML/ min reveals more disorder than that grown at 140– 170 K with 0.027 ML/ min. However, there are no observable differ- ences for 0.08 ML Co grown at 140– 170 K with different deposition rates 关Fig.2共c兲兴. As mentioned in the discussion of Fig. 1, 0.08 ML is not enough to stabilize the nanoparticles during warming up to RT. In other words, small particles in 0.08 ML Co grown at 140– 170 K are capable of moving toward the oxide stripes while warming to RT. Thus no observable difference exists for low temperature 共LT兲- grown 0.08 ML Co with different deposition rates.

Figures3共a兲and3共b兲show the STM images of 0.15 and 1.5 ML Co nanoparticle assembles grown at RT共deposition rate= 0.5 ML/ min兲 after annealing at various temperatures for 30 min. Figure3共c兲summarizes the statistics of particle density as a function of annealing temperature. For 0.15 ML

FIG. 2. 共Color online兲共a兲 and 共b兲 are STM images of 0.15 ML Co/ Al₂O₃/ NiAl共100兲 prepared with different deposition rates at 170 and 160 K, respectively.共c兲 The statistics of disordered particle density as functions of growth temperature.

153111-2 Lin et al. Appl. Phys. Lett. 89, 153111共2006兲

(3)

Co, 627 K annealing induces the diffusion along the stripes, resulting in particle sintering and reducing the density to 80%, without losing the alignment 关middle image in Fig.

3共a兲兴. After 695 and 756 K annealing, the density signifi- cantly drops to only 26%. For 1.5 ML, the density gradually reduces to 50% during annealing at 400– 800 K. For further annealing at 800– 1090 K, the particle density sustains nearly the same value:共⬃300–260兲/共10⁴nm²兲, about 50%–

40% of the as-grown density. From Fig. 3共b兲, the particle size is apparently enlarged due to the sintering of smaller particles, but the width of size distribution does not spread out. Even after 1090 K annealing, no particles of very large size共⬎⬃6–8 nm in diameter兲 can be observed. It also sug- gests that such size of supported Co nanoparticles on Al₂O₃/ NiAl共100兲 should be stable up to as high as 1090 K.

With the understanding of the growth temperature, deposition rate, and thermal annealing effects, the controlled growth of Co nanoparticles can be demonstrated, as shown in Fig.4. The particle density is analyzed in each process as a function of total coverage. At first, 0.15 ML Co particles are prepared at 170 K with low deposition rate 共0.027 ML/min兲, revealing high density and well alignment in Fig.4共a兲. Afterward 0.08 ML Co is deposited at 300 K with low deposition rate for three times. Based on the idea discussed in the section of growth-temperature effect, growth at higher temperature with low deposition rate promises the deposited atoms to arrive the nucleations, which is already prepared in advance, without meeting other atoms. Thus the nanoparticles in Fig.4共b兲can be enlarged without change of density. Statistical analysis in Fig.4共a兲indicates that the particle density is kept at⬃800/共10⁴nm²兲 with the total coverage up to 0.39 ML. After 0.39 ML, the nanoparticle assembly is annealed at 500 K for 30 min, resulting in notable reduce of density. Sequential Co deposition at RT also en- larges the particle size both in height and diameter with the density kept at⬃600/共10⁴nm²兲. Figure4reveals thus a pro- cedure of controlled growth for Co nanoparticles.

In summary, a slow deposition rate 共0.027 ML/min兲 at LT promises the good linear alignment and high particle density. Relative to 0.15 ML Co, 1.5 ML Co nanoparticle assembly reveals the high thermal stability of sustaining the density ⬃260/共10⁴nm²兲 after 800–1090 K annealing. The detailed understanding of the controlled growth of Co nanoparticle might be a good exemplification for other nanoparticle assembly composed of different materials.

This work was supported by the National Science Coun- cil of Taiwan under Grant Nos. NSC 94-2112-M-002-005, 95-2120-M-002-015, and 95-2112-M-002-MY3.

1J. P. Pierce, M. A. Torija, Z. Gai, Junren Shi, T. C. Schulthess, G. A.

Farnan, J. F. Wendelken, E. W. Plummer, and J. Shen, Phys. Rev. Lett. 92, 237201共2004兲.

2Z. Gai, B. Wu, J. P. Pierce, G. A. Farnan, D. Shu, M. Wang, Z. Zhang, and J. Shen, Phys. Rev. Lett. 89, 235502共2002兲.

3M. Valden, X. Lai, and D. W. Goodman, Science 281, 1647共1998兲.

4S. Gwo, C.-P. Chou, C.-L. Wu, Y.-J. Ye, S.-J. Tsai, W.-C. Lin, and M.-T.

Lin, Phys. Rev. Lett. 90, 185506共2003兲.

5H.-J. Freund, Surf. Sci. 500, 271共2002兲.

6W. C. Lin, C. C. Kuo, M.-F. Luo, Ker-Jar Song, and Minn-Tsong Lin, Appl. Phys. Lett. 86, 043105共2005兲.

7W. C. Lin, P. C. Huang, Ker-Jar Song, and Minn-Tsong Lin, Appl. Phys.

Lett. 88, 153117共2006兲.

8M. F. Luo, C. I. Chiang, H. W. Shiu, S. D. Sartale, and C. C. Kuo, Nanotechnology 17, 360共2006兲.

9M. Bäumer, M. Frank, M. Heemeier, R. Kühnemuth, S. Stempel, and H.-J.

Freund, Surf. Sci. 454–456, 957共2000兲.

10M. Heemeier, S. Stempel, Sh. K. Shaikhutdinov, J. Libuda, M. Bäumer, R.

J. Oldman, S. D. Jackson, and H.-J. Freund, Surf. Sci. 523, 103共2003兲.

11W. Chen, K. P. Loh, H. Xu, and A. T. S. Wee, Langmuir 20, 10779共2004兲.

12W. Chen, K. P. Loh, H. Xu, and A. T. S. Wee, Appl. Phys. Lett. 84, 281 共2004兲.

13M. F. Luo, C. I. Chiang, H. W. Shiu, S. D. Sartale, T. Y. Wang, P. L. Chen, C. C. Kuo, J. Chem. Phys. 124, 164709共2006兲.

14Harald Brune, Surf. Sci. Rep. 31, 121共1998兲.

15J. A. Venables, G. D. T. Spiller, and M. Hanbücken, Rep. Prog. Phys. 47, 399共1984兲.

16A. Steltenpohl and N. Memmel, Surf. Sci. 454–456, 558共2000兲.

17J. Zhou and D. A. Chen, Surf. Sci. 527, 183共2003兲.

18J. Carrey, J.-L. Maurice, F. Petroff, and A. Vaurès, Phys. Rev. Lett. 86, 4600共2001兲.

19G. Hass, A. Menck, H. Brune, J. V. Barth, J. A. Venables, and K. Kern, Phys. Rev. B 61, 11105共2000兲.

20B. Fischer, H. Brune, J. V. Barth, A. Fricke, and K. Kern, Phys. Rev. Lett.

82, 1732共1999兲.

FIG. 3. 共Color online兲 STM images of 共a兲 0.15 and 共b兲 1.5 ML Co after grown at 300 K and annealed at different temperatures.共c兲 Summary of the particle density as functions of annealing temperature.

FIG. 4.共Color online兲共a兲 Sequential particle density with variation of deposited coverage. 共b兲–共e兲 are STM images with different total coverages during the processes exhibited in共a兲.

153111-3 Lin et al. Appl. Phys. Lett. 89, 153111共2006兲