FFT Analysis of Poorly Organized Nanopores
FFT Analysis of Poorly Organized Nanopores
FFT Analysis of Poorly Organized Nanopores
Materials Letters
journal homepage: www.elsevier.com/locate/matlet
art ic l e i nf o a b s t r a c t
Article history: Poorly arranged nanopores focus attention of the researchers due to the many fundamental aspects.
Received 5 September 2013 Thus, quantitative arrangement analysis of non-ideally arranged pores is demanded. In the paper, radial
Accepted 24 November 2013 average fast Fourier transform based arrangement analysis of the nanopores has been reported. This
Available online 2 December 2013
arrangement analysis tool has been applied in the quantication of arrangement of anodic alumina
Keywords: formed in 0.3 M chromic acid at voltage ranging from 20 to 50 V and temperature ranging from 20 to
Anodic aluminum oxide 50 1C. Presented approach, in contrary to typically applied methods, is based on the radial average of the
Self-organization fast Fourier transform (FFT). Averaged regularity ratio, estimated with the use of the reported tool, gives
Nanopores more repeatable results with smaller spread. Thus, reliable dependencies between operating conditions
Fast Fourier transform
and the averaged regularity ratio can be also obtained for poorly arranged pores. It was found that the
Circularity
greater the voltage and temperature, the better nanopores arrangement.
Anodization
Arrangement analysis & 2013 Elsevier B.V. All rights reserved.
0167-577X/$ - see front matter & 2013 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.matlet.2013.11.099
70 W.J. Stpniowski et al. / Materials Letters 117 (2014) 6973
Fig. 1. An exemplary top-view FE-SEM image (a) and its fast Fourier transform (b) with three intensity proles (df) as well as the inuence of operating conditions on the
regularity ratio (c).
Fig. 2. An exemplary computer model of ideally arranged pores (a) and its fast Fourier transform (b) with radial average (c) as well as the inuence of surface area (d) and
number of pores (e) on intensity and width of the radial average peak evaluated for ideally arranged computer models.
W.J. Stpniowski et al. / Materials Letters 117 (2014) 6973 71
with WSxM software [12,13]. To estimate regularity ratio, previously 3. Results and discussion
reported method was applied [14]. To estimate averaged regularity
ratio a radial average of the FFT was performed. For every set of the Typical fast Fourier transform quantitative arrangement analysis
experimental conditions three images were taken into account [14] was performed for AAO obtained by anodization in chromic acid
(magnication 100,000 ). Circularity was measured using NIS- (Fig. 1). The main issue is selection of three major hexagonal
Elements software from six images taken for every set of operating directions of fast Fourier transform along which the intensity proles
conditions (anodizing voltage and temperature). have to be conducted (Fig. 1b). For highly-ordered nanoporous array,
Fig. 3. FE-SEM images (a, d, g, and j), their fast Fourier transforms (b, e, h, and k) and radial averages (c, f, i, and l) of the anodic aluminum oxide formed in 0.3 M chromic acid
at 30 1C at voltage of 20 (ac), 30 (df), 40 (gi) and 50 V (jl).
72 W.J. Stpniowski et al. / Materials Letters 117 (2014) 6973
three major directions are represented on the FFT by six distinct dots
at the corners of the hexagon. Here, the FFT image is a blurred ring,
therefore conducted intensity proles (Fig. 1df) may be affected by
the selection procedure. As a result, values of obtained regularity
ratio are widely spread and average values of regularity ratio are
accompanied by relatively high values of standard deviations, reach-
ing up to 80% (Fig. 1c). Thus, typical arrangement analysis dedicated
to the highly-ordered AAO fails in the case of poorly arranged pores
analysis. It implies a demand of application of a new tool where
issues like selection of hexagonal directions are avoided.
In this paper, we propose a new method based on the radial
average of the fast Fourier transform. Due to the application of the
analysis of whole FFT image, instead of three major directions,
problems described above are solved.
Firstly, to obtain trustworthy results, a series of models were
generated to investigate the inuence of number of pores and
Fig. 4. Averaged regularity ratio, estimated from radial averages of the fast Fourier
surface area on the major values (intensity of radial average and transforms, versus anodizing voltage.
width at half of its height) derived from FFT (Fig. 2ac). It was
found that the intensity and width at half of height of FFT radial
average are strongly inuenced by surface area and number of Table 1
pores (Fig. 2d and e). To remove these inuences, equations of Circularity of nanoporous alumina formed by self-organized anodization in 0.3 M
tted curves have to be taken into further calculations. Thus, chromic acid for various anodizing voltages and temperatures.
averaged regularity ratio (R), derived from the FFT radial average
Temperature [1C] Anodizing voltage [V]
can be estimated with the following
p 20 30 40 50
I n
R 1
W S32 20 0.58 7 0.22 0.30 7 0.05 0.247 0.02 0.25 7 0.02
30 0.32 7 0.02 0.34 7 0.03 0.337 0.02 0.30 7 0.01
where I is intensity of the radial average, W is the width of the
40 0.517 0.08 0.517 0.02 0.50 7 0.01 0.42 7 0.03
radial average at half of its height, n is the number of pores on the 50 0.417 0.04 0.54 7 0.01 0.55 7 0.02 0.447 0.01
analyzed image and S is analyzed surface area.
In the Fig. 3 top-view FE-SEM images of anodic alumina formed
in 0.3 M chromic acid at 30 1C at various voltages (Fig. 3a, d, g, and with operating conditions can be observed, which enhance opti-
j) are shown. The pores are irregular, however the fast Fourier mization of the experiments. Moreover, obtained tendencies can
transforms (Fig. 3b, e, h, and k) and their radial averages (Fig. 3c, f, be compared to the ones obtained for highly-ordered alumina.
i, and l) enable quantication of the arrangement. The radial The result obtained with the use of the FFT-based averaged
averages form smooth, distinct peaks which makes the estimated regularity ratio and circularity analysis shows that the greater
results trustworthy. The physical meaning of radial average is a voltage and electrolyte's temperature the better arrangement of
distribution of the interpore distances in the inversed space. Thus, the nanopores formed via self-organized two step anodization in
the worse arrangement, the more blurred FFT image and the lower chromic acid.
intensity of the radial average should be. Moreover, the abscissa of
the radial average peak corresponds to the average interpore
distance [2] and width of the peak at half of its width is strongly
4. Conclusions
related to the dispersion of the interpore distance values. Quanti-
tative arrangement analysis shows that for all of the electrolyte's
Fast Fourier transform based quantitative arrangement analysis
temperatures, averaged regularity ratio increases with the voltage
of poorly ordered nanopores was performed. In the presented
(Fig. 4). In this case, the best arrangement is for the nanopores
approach, whole FFT is being taken into account. Such a method
formed at 50 V. For AAO formed at the same range of voltages and
has the following advantages:
temperatures, but in oxalic acid, the best nanopores arrangement
is for 40 V [2,6,14]. Here, a well-organized hexagonally arranged
1. Quantitative arrangement analysis of poorly arranged nano-
array growth is disturbed by incorporation of chromate anions,
pores is enabled.
having relatively large hydrodynamic radius. These species are
2. Obtained values are repeatable and have small spread even for
being incorporated into the AAO walls during anodization, what
strongly disordered and deformed pores.
signicantly spoils the nanopores arrangement, in comparison to
3. Distinct tendencies in evolution of regularity with operating
the AAO formed in oxalic acid at the same voltage range [11]. The
conditions are obtained with the use of the presented method.
greatest values of averaged regularity ratio were obtained for AAO
formed at 50 1C. It is a result of better pores circularity (Table 1).
The greatest values of circularity are implied by the greatest values Moreover, presented method was tested for AAO formed in
of the chemical reaction rate of the grown anodic oxide with chromic acid, which provided poorly arranged nanopores, where
chromic acid. The greater uniformity of the nanopores' diameters typical approach of quantitative arrangement analysis, dedicated for
in all the directions, causes more uniform distribution of the highly-ordered pores is useless. It was found that regularity
interpore distances (geometrical centers of the pores are related (represented by averaged regularity ratio) of the anodic alumina
with them). nanopores formed in chromic acid increases with anodizing voltage.
Application of a fast Fourier transform quantitative arrange- The best nanopores arrangement was found for anodic aluminum
ment analysis tool dedicated to poorly arranged nanopores allows oxide formed at the greatest temperature, what is acknowledged to
obtaining repeatable values of averaged regularity ratio with small the enhanced chemical reaction of the electrolyte with the grown
spread. Clear tendencies of averaged regularity ratio evolution oxide.
W.J. Stpniowski et al. / Materials Letters 117 (2014) 6973 73
Acknowledgments [7] Salerno M, Patra N, Losso R, Cingolani R. Increased growth rate of anodic
porous alumina by use of ionic liquid as electrolyte additive. Mater Lett 2009;
63: 18261829.
The research was nanced by National Science Centre (Decision [8] Stpniowski WJ, Norek M, Michalska-Domaska M, Nowak-Stpniowska A,
no.: DEC-2012/05/B/ST8/02915). Scanning Electron Microscope has Bombalska A, Wodarski M, et al. Incorporation of copper chelate ions into
been purchased with the nancial support from the Innovative anodic alumina walls. Mater Lett 2013;106:2425.
[9] Shingubara S, Morimoto K, Sakaue H, Takahagi T. Self-organization of a porous
Economy Programme (POIG.02.01.00-14-071/08).
alumina nanohole array using a sulfuric/oxalic acid mixture as electrolyte.
Electrochem Solid State Lett 2004;7:E157.
References [10] Almasi Kashi M, Ramazani A, Noormohammadi M, Zarei M, Marashi P.
Optimum self-ordered nanopore arrays with 130270 nm interpore distances
formed by hard anodization in sulfuric/oxalic acid mixtures. J Phys Appl Phys
[1] Lillo M, Losic D. Ion-beam pore opening of porous anodic alumina: the
2007;40:703240.
formation of single nanopore and nanopore arrays. Mater Lett 2009;63:45760.
[11] Stpniowski WJ, Norek M, Michalska-Domaska M, Bombalska A, Nowak-
[2] Stpniowski WJ, Bojar Z. Synthesis of anodic aluminum oxide (AAO) at
Stpniowska A, Kwany M, et al. Fabrication of anodic aluminum oxide with
relatively high temperatures. Study of the inuence of anodization conditions
incorporated chromate ions. Appl Surf Sci 2012;259:32430.
on the alumina structural features. Surf Coat Technol 2011;206:26572.
[12] Horcas I, Fernndez R, Gmez-Rodrguez JM, Colchero J, Gmez-Herrero J, Baro
[3] Le Coz F, Arurault L, Datas L. Chemical analysis of a single basic cell of porous
anodic aluminium oxide templates. Mater Charact 2010;61:2838. AM. WSXM: a software for scanning probe microscopy and a tool for
[4] Zhang R, Jiang K, Ding G. Surface morphology control on porous anodic nanotechnology. Rev Sci Instrum 2007;78:013705.
alumina in phosphoric acid. Thin Solid Films 2010;518:3797800. [13] WSxM, http://www.nanotec.es.
[5] Ding GQ, Yang R, Ding JN, Yuan NY, Zhu YY. Fabrication of porous anodic [14] Stpniowski WJ, Nowak-Stpniowska A, Bojar Z. Quantitative arrangement
alumina with ultrasmall nanopores. Nanscale Res Lett 2010;5:125763. analysis of anodic alumina formed by short anodizations in oxalic acid. Mater
[6] Nielsch K, Choi J, Schwirn K, Wehrspohn RB, Gsele U. Self-ordering regimes of Charact 2013;78:7986.
porous alumina: 10%-porosity rule. Nano Lett 2002;7:67780.