Thin Film

3 May 2018

Electrical Conductivity Measurement of Carbon Nanotubes Film using Conductive Probe Atomic Force Microscopy (C-AFM)

John Paul Pineda, Gerald Pascual, Byong Kim, and Keibock Lee Park Systems Inc., Santa Clara, CA USA ABSTRACT Electrical conductivity measurement is an effective approach to describe how a material behaves for certain applications, ranging from energy storage and energy conversion devices, to interconnections in molecular electronics and nanometer-sized semi-conductor devices. A technique known as Conductive Probe Atomic Force Microscopy (C-AFM) is a powerful technique that provides accurate nanoscale measurement and mapping of relative difference in electrical conductivity of advanced materials such as CNTs film. Several characterization techniques were introduced in the past decade to study these materials, however, majority of these can only measure a limited electrical properties range. In this study, Park NX20 equipped with C-AFM was used to investigate 3 different materials with wide range of electrical conductivity. The data acquired in this experiment clearly demonstrate the ability of this technique in measuring wide range of electrical conductivity and differentiating surfaces of materials covered with various types of conductive materials, with the use of a logarithmic current amplifier integrated in the system. INTRODUCTION Carbon Nanotubes (CNTs) have attracted a great deal of scientific attention and industrial interest world-wide due to its unique electrical behavior [1, 2]. A number of different applications ranging from energy storage and energy conversion devices, to interconnections in molecular electronics and nanometer-sized semi-conductor devices were demonstrated over the past decade [3, 4]. CNTs can behave as metallic or semi-conducting materials depending on the arrangement of their atoms, their chirality (degree of twist), as well as their sizes (diameter and length) [1, 2]. The electrical conductivity of CNTs plays a major role in describing its behavior and its implication to science and electronics. Therefore, it’s critically important to utilize a technique that can effectively measure the electrical conductive properties of these new materials. However, due to their fragile characteristics and nano-scale dimensions, measuring their local properties have become a great challenge to many researchers and device engineers [5, 6]. There are several methods that were introduced to characterize these materials, and the most common are scanning tunneling microscopy (STM), transmission electron microscopy (TEM) and focused ion beams (FIBs). [5, 7, 8] However, some of these techniques are destructive, some have a limited measurement and property characterization modes, others require high vacuum environment. One of the most powerful tools that was designed to overcome these problems is Conductive Probe Atomic Force Microscopy (C-AFM). This technique provides both electrical properties and topography at the same time, first by monitoring the current flowing between the conductive tip and sample, and the latter by monitoring the cantilever deflection as the tip scans over the sample surface. In this study, Park NX20 equipped with C-AFM was used to investigate 3 different materials namely: 1) glass, 2) silver and 3) CNTs film. The results acquired in this experiment clearly demonstrate the ability of this technique to accurately measure wide range of electrical conductivity of advanced materials such as CNTs film. In addition, this technique can be used effectively to differentiate regions covered with various types of conductive materials. EXPERIMENTAL Sample and Probe The sample that was investigated in this experiment is a CNTs film made of soot of carbon nanotubes that were deposited on a glass substrate to form a thin film layer. Subsequently, an Ag electrode is patterned into the surface to make an electrical contact. A conductive diamond coated probe (NANOSENSORS™ CDT-CONTR) with a nominal force constant of k = 0.5 N/m and resonance frequency of f = 20 kHz was utilized in the entire test. C-AFM Experimental Conditions The CNT film sample was investigated using Park NX20 AFM system under ambient air conditions to perform a 45 x 45 μm scan outputting to an image size of 256 x 256 pixels.Figure 1 shows the principles of C-AFM. The topography and electrical properties of the sample can be acquired simultaneously during operation. The topography data is acquired by monitoring the deflection signal of the cantilever as the conductive tip scans the sample surface in contact. On the other hand, electric conductivity is acquired by measuring the electrical current passing in between the conductive probe and the sample, produced by applying a bias voltage in between the two. The electrical conductivity is measured through an electric current amplifier. In this experiment, the bias voltage that was used is positive 0.3V, since the optimum current distribution was observed using this value. Generally, the current flow is acquired by a current amplifier and then processed into an image. The Park NX20 is equipped with internal current amplifier with variable gain of 106~1012 V/A, however, since the sample consists of different materials with wide range of electrical conductivity from nearly non conducting 0μAtorelatively well conducting 23.18μA, a logarithmic amplifier adaptable to this range was utilized in the experiment.   Fig. 1.Principles of Conductive Probe AFM. This diagram demonstrates that feedback control from the AFM controller allows this technique to acquire both surface topography and conductivity data simultaneously. The topography data is acquired by monitoring the deflection signal of the cantilever, while electric conductivity is acquired by measuring the tunneling current using current amplifier. RESULT AND DISCUSSION The acquired images were analyzed using XEI software developed by Park Systems which mapped the acquired signals to a color table. For topography image, the intensity of the shading correlates to the surface height variation with extremely bright and dark areas having the highest and lowest height regions. Figure 2shows the topography and current image of the CNTs film sample. The topography data clearly shows that the 45um by 45um scanned surface of the sample is composed of areas with variousheights which can be divided into 3 regions. It can be observed that the region with the lowest height has arelatively smooth surface, suggesting that this region is glass substrate. On the other, the two other regions have relatively rough surfaces, suggesting that these regions are the CNTs and Ag materials. The surface roughness of each regions were calculated using XEI software. The measured surface roughness of the glass substrate was approximately1.48 nm, while the other two regions have around 14.25 and 14.71 nm. Since the CNTs and Ag has almost the same features, differentiating these two materials will be difficult by simply looking at the topography data. The current image clearly differentiates the material composition of the sample and it also shows that the surfaces of the sample is divided into 3 regions. The areas with highest conductivity are represented by the red color map, while areas with lower conductivity are represented by the green color map, and for the areas that are non-conductive are represented by brown color map. Among the 3 materials, Ag has the highest conductivity, while CNT has 2nd to the highest, and the glass has the lowest conductivity [9]. Fig. 2.Topography (top-left) and current image (top-right) acquired from CNTs film sample. Scan size: 45 x 45 μm, image size: 256 x 256 pixels.Topography line profile (red line, y-axis on left) and current line profile (green, y-axis on right). If one were to analyze the corresponding line profile of topography and current image, the region with the highest height in topography image is the region with the highest conductivity in current image which is the Ag materials, while the region with 2nd to the highest height and conductivity is the CNTs film, and the lowest is the glass substrate. The quantitative results of conductivity in terms of measured current values are shown in the Table. The average measured conductivity of Ag is approximately 23.56 μA, while for CNT is 0.98 μA, and for glass is 0 μA. The results were plotted in Figure 3 to better analyze the data. The standard deviations of the electrical conductivity results were also calculated. The glass sample has the lowest standard deviation with Std = 0 A, while Ag sample has the highest standard deviation with Std = 8. Table. 1.Measuredmean and standard deviation (Std) of current.   Figure. 3. Plotted value of measured mean standard deviation (Std) of current. SUMMARY The topography and electrical conductivity of a CNTs film have been characterized using C-AFM with a Park NX20 AFM system. The data collected in this experiment reveals that this technique can provide qualitative and quantitative information for electrical characterization of advanced materials. Moreover, the results demonstrate that this technique is an effective mean in measuring wide range of electrical conductivity and differentiating surfaces of materials covered with various types of conductive materials, with the use of a logarithmic current amplifier integrated in the system. Overall, the technique described in this study will successfully help researchers and device engineers with key electrical parameters information to better understand the behavior of certain material with unique properties such as CNTs RESULT AND DISCUSSION 1.Y. Xue, Experimental Study of Electrical Conductivity of Carbon Nanotube, Nanofiber Buckypapers, and Their Composites. 2. P. Collins, et al., The Electronic Properties of Carbon Nanotubes. 3. J. Brito, et al., Nanostructured films from phthalocyanine and carbon nanotubes: Surface morphology and electrical characterization. Journal of Colloid and Interface Science, Volume 367, Issue 1, 1 February 2012, Pages 467-471. 4. J. Maklin, ELECTRICAL AND THERMAL APPLICATIONS OF CARBON NANOTUBE FILMS 5. Z. Wang, et al., Measuring physical and mechanical properties of individual carbon nanotubes by in situ TEM. Journal of Physics and Chemistry of Solids 61 (2000) 1025–1030. 6. M. Li, et al., Analysis of variance on thickness and electrical conductivity measurements of carbon nanotube thin films. Published 19 July 2016 • © 2016 IOP Publishing Ltd, Measurement Science and Technology, Volume 27, Number 9. 7. K. Lee, Failure analysis and the innovative PinPoint™ conductive AFM 8. G. Homyak, et.al, TEM, STM and AFM as tools to study clusters and colloids. Micron, Volume 29, Issues 2–3, April–June 1998, Pages 183-190. https://doi.org/10.1016/S0968-4328(97)00058-9 9. A. Helmenstine, Table of Electrical Resistivity and Conductivity.

26 Dec 2013

Patterned Arrays of Magnetic Nanostructures

Patterned arrays of magnetic nanostructures have become one of the key issues in recent years because of its potential application to information technology with respect to magnetic storage devices. Particular attention has been given to the magnetic properties of periodic arrays of micron and submicron elements of very different shapes, especially dots and stripes. The study of these arrays of submicron elements is the basis for the development of magnetic memories and nanopatterned recording media, since the inter-element interaction determines the integration limits of magnetic memories and nanopatterned recording media.  The periodic magnetic domain structure was observed by using a magnetic force microscopy (MFM). The method was the first scanning probe technique and turned out to be a powerful tool for submicron-scale magnetic image mapping. In order to measure the magnetic properties of periodic structures based on magnetic/nonmagnetic films, the MFM technique is better suited that possess the high sensitivity required to monitor magnetic domain structure in thin films and very small elements of different shapes.  We fabricated two-dimensional (2D) Co magnetic arrays by the photolithography with KrF laser source and the wet etching (Transene Inc., TFB) process. A Co film, 40 nm thick, was deposited on a 6-inch Si(100) wafer by using a sputtering system.  The magnetic domain structure has been investigated using a MFM system, which is basically a scanning probe microscopy (Park SYSTEMS, XE-100) equipped with a magnetic tip (Nanosensors, PPP-MFMR). Also, the MFM system was equipped with an electromagnet, which allows us to investigate the evolution of the magnetic domains in an in-plane applied magnetic field. FIG. 1: (a) AFM and (b) MFM images of 2D patterned arrays of Co antidot structures. The scan area is 10 μm x 10 μm.   Figure 1 shows the atomic force microscopy (AFM) and the MFM images obtained from 2D patterned arrays of Co antidote  structures. The MFM image in Fig. 1(b) has three different domains. Two are wedge-shaped domains connecting the holes and the other is a diamond-shaped domain. For the wedge-shaped domains, the magnetization vector is parallel to the connecting boundaries at the holes and points from dark to bright inside the domain. The magnetization vector in the diamond-shaped domain is oriented along the diagonal. This configuration minimizes the magnetostatic energy.  FIG. 2: (a) AFM and (b) MFM images of 2D patterned arrays of Co dot structure. The scan area is 10 μm x 10 μm.   As shown in Fig. 2(a), the AFM image study reveal that a 2D patterned arrays of submicron square dot structures was fabricated. The AFM image clearly shows a square-lattice pattern with a periodicity of about 1.6 μm. By comparing the AFM and the MFM images [see Fig.2(a) and Fig. 2(b), respectively], it is clear that the topography and the magnetic contrast are well separated. The MFM image of the square pattern shown in Fig. 2(a) demonstrates a Landau-Lifshitz-type flux closure domain structure with 90° domain walls and a central vortex. The magnetic charges are mainly concentrated along the domain walls. In systems with a negligible magnetocrystalline anisotropy, the balance of demagnetization and exchange energies results in flux-closure structures such as four fold-closure-domain pattern in thin squares. It can be seen that, apart from the contrast associated with domain walls, the one related to the vortex core is also clearly detected.  As seen in Fig. 3, two kinds of vortex chirality appear in all the patterned arrays, namely clockwise and counter-clockwise, with an approximately equal probability due to the two fold energy degeneracy. This finding reveals the bistability of magnetic vortices and the statistically random distribution of vortex chirality. Additionally, the vortices of different chiralities present clearly the corresponding MFM images under an applied field. FIG. 3: Field-dependent MFM images of square Co dots. The scan area is 5 μm x 5 μm .   The field-dependent MFM images show that the application of the field (indicated by a black arrow) induces the domain-wall motions (see Fig. 3). The intersection moves, first, perpendicular to the applied field direction and gradually approaches the corner. At H = 300 Oe, the field increases the size at the expense of the right (or left) domain. As the strength of applied field keeps increasing, the magnetization vectors are antiparallel to the external field, and eventually all the magnetic moments are reversed and become parallel to the field. As seen in Fig. 3(d), the magnetic domains gradually collapse with each other. These characteristics are closely related to the distribution of vortex annihilation (nucleation) fields.

23 Dec 2013

Thin Films Nanolithography (XEL)

Thin Films Nanolithography (XEL) Scanning Probe Nanolithography is a very promising technology for nanofabrication. It uses the probe tip of scanning probe microscope to manipulate materials at nanometer scale. There are many ways to to make patterns, such as mechanical scratching and electrical field oxidation. Park System provides a very powerful software, XEL, for Nanolithography. It supports a variety of lithography modes such as Z scanner movement mode, Set point mode, and Bias mode. It supports both Vector and Raster scan modes. Figure 1.  A Big Buck pattern made on a silicon substrate using bias mode XEL Nanolithography. The pattern is made on a silicon surface using the Raster scan mode.   Publications Using XE-AFM and Nanolithography   AA Milner, K Zhang, and Y Prior; Floating Tip Nanolithography; NanoLetters; 2008 Vol. 8, No. 7; 2017-2022 Equipment: Park Systems XE-120   Figure 2. (a) Hot floating tip writing WIS (for the Weizmann Institute of Science) on AZ4620 photoresist film. The writing speed was 50 nm/sec at tip-sample gap 3 ( 2 nm and the average laser power was ∼3 × 105 W/cm2. (b) Direct mechanical scratching of the same polymer with an identical cold tip. Profiles c and e correspond to the red cross-section lines in panels a and b; images d and f are the magnified line intersections in panels a and b, respectively. Abstract: We demonstrate noncontact, high quality surface modification of soft and hard materials with spatial resolution of ∼20 nm. The nanowriting is based on the interaction between the surface and the tip of a standard atomic force microscope illuminated by a focused femtosecond laser beam and hovering (at ambient conditions) 1-4 nanometers above the surface without touching it. Field enhancement at the tip-sample gap or high tip temperature are identified as the causes of material ablation.  

23 Dec 2013

Thin Film - ZnO

Electrical Property (EFM/KPFM) The material and electrical properties of Zinc Oxide (ZnO) have made the compound attractive in the area of thin film transistors and light emitting diodes.  Due to it non-toxic, wide bandgap and electrical conductivity, it is thought to be ideal for use in solar cell devices.   Publication on ZnO Using KPFM and the Park AFM DR Doutt C Zgrabik, HL Mosbacker, LJ Brillson; Impact of near-surface native point defects, chemical reactions, and surface morphology on ZnO interfaces; J. Vac. Sci. Technol. B 26,4, Jul/Aug 2008 1477 Equipment: Park Systems XE-70 Abstract: The authors used a complement of depth-resolved cathodoluminescence spectroscopy _DRCLS_, atomic force microscopy _AFM_, and Kelvin probe force microscopy _KPFM_ to correlate the formation of native point defects with interface chemical reactions as well as surface morphology. A wide array of ZnO crystals grown by both melt and hydrothermal growth methods display orders-of-magnitude variation in 2.1, 2.5, and 3.0 eV native point defect optical transitions at their free surface and as a function of depth on a nanometer scale. AFM surface morphology scans taken simultaneously with KPFM potential maps reveal large variations in surface morphology related to the growth method and subsequent processing. Notably, when DRCLS defect emissions are low, the surface roughness is low and the morphology matches its respective KPFM potential map. When DRCLS emissions vary with depth, the morphology and potential maps do not correlate. Indeed, the latter can vary by hundreds of meV across micron square areas. These subsurface electrical changes are consistent with DRCLS features and emphasize the contribution of surface morphology to electrically active interface defects. The relative strength of near band edge to deep level defect emissions exhibit a threshold dependence on surface roughness.   FIG. 3. _a_ DRCLS spectra for hydrothermally grown single crystal ZnO substrate MTI 15 alongside its respective _b_ AFM topography and _c_ KPFM potential map. DRCLS spectra show some of the highest bulk and surface NBE intensities and deep levels at 2.1 and 2.4 eV. The dashed curves represent the Gaussian deconvolution. Its KPFM potential map shows large _hundreds of mV_ potential changes across its surface.  

23 Dec 2013

Characterization of Epitaxially Grown MnAs Films Using AFM and MFM

Introduction and Sample Description Because of its ferromagnetic properties with well-oriented interfaces at room temperature, epitaxially grown MnAs film on GaAs substrate is one of the most promising systems for future spintronic applications despite the large lattice mismatch between the epitaxial film and its substrate material. The bulk MnAs shows a phase transition from the ferromagnetic (FM) α-MnAs to the paramagnetic orthorhombic β-MnAs at the critical temperature of about 45 °C.3 In the MnAs thin film systems, the two phases coexist in the form of self-organized periodic stripes at room temperature, occurring via strain stabilization. The periodic length of the two coexisting phase stripes depends strongly on the film thickness. In this note, MnAs films with varying thickness are observed in various magnetization states.  The MnAs films investigated herein were grown on GaAs(001) substrates at 270 °C by molecular beam epitaxy with varying thickness of 10, 20, 50, 100, 250, and 500 nm, respectively. At room temperature the strain-stabilized coexistence of the two phases was confirmed by X-ray diffraction experiments. By using a superconducting quantum interface device magnetometer (SQUID) the magnetic properties were also investigated. Also, Atomic Force Microscopy (AFM) and Magnetic Force Microscopy (MFM) data show that the α-MnAs and β-MnAs stripes are aligned along the MnAs[0001] and perpendicular to the magnetic easy axis MnAs[1120].  The MFM images were acquired by using a Park AFM with MFM option and a magnetic cantilever (Nanosensors). A magnetic field generator was used to apply magnetic field to the samples, which generates magnetic fields up to 600 Oe. Experimental Details For the characterization of magnetic samples such as epitaxially grown MnAs films, measurement of their local magnetic property is crucial. Though SQUID gives quantitative magnetic data about the sample under investigation, the lack of spatial resolution does not give structural analysis regarding the material. MFM data, on the contrary, reveals local magnetic properties of the samples along with well-defined topographic data.  A vital requirement for these precise measurements by MFM is the explicit separation of the magnetic signal from the topographic data. While the MFM signal is collected from the instrument, the cantilever is positioned at a certain distance from the sample surface, so that the van-der-Waals force do not interfere the magnetic force. To maintain this condition, the position of the cantilever is controlled very accurately. In conventional piezoelectric tube based AFM systems, the Z direction movement (which corresponds to the height data) is coupled with the XY direction scanner movement due to its geometrical design limitation. The crosstalk between XY and Z movement introduces topographic error signal to the magnetic signal, and it needs accurate and stable calibration to obtain realistic results. On the contrary, Park AFM introduces a perfectly decoupled XY and Z-scanners to prevent the interference of the movement crosstalk. The separated movement of Park AFM secures the real MFM data which is decoupled from the topographic signal.   Data Analysis Figure 1. MFM images of epitaxially grown MnAs films with various film thicknesses of 500 nm, 250 nm, 100 nm, 50 nm, and 25 nm, respectively at (a) the saturated state and (b) the demagnetized state. Observed area is 5 μm in each direction, and the magnetic easy axis of each film is perpendicular to the FM stripes.   Figure 1 shows the MFM images of the MnAs films with various thicknesses of 25 nm, 50 nm, 100 nm, 250 nm, and 500 nm, respectively, at (a) saturated state and (b) demagnetized state. In these images, the easy axis of each film is perpendicular to the stripe direction, and the observed area is 5 μm in each direction. As shown in Figure 1(a), all the ferromagnetic (FM) stripes in each saturated film are in a single domain state, where MFM contrast in each stripe is ascribed to the stray field from the boundary between the neighboring stripes. Also it is observed that the periodic length and the widths of the α and β phase stripes decrease as the film thickness decreases, which correlates well with strain relaxation effects.  In Figure 1(b), change in the domain structure of the demagnetized films is observed, as the film thickness decreases. In the image of the 500-nm-thick sample, the domain structure within the FM stripes can be described as head-on domains, which are expressed as the spin configurations in which the magnetization directions of two domains meet head-on as demonstrated in the rectangular box of the MFM images. In the image of 250-nm-thick sample, two kinds of domain structures are observed, head-on domains and simple domains. The simple domains are defined as the spin configurations having a single spin direction along the width of the FM stripe, as demonstrated in the rectangular box of the MFM image. The existence of head-on domains in 250 and 500-nm-thick samples indicates the formation of a closed domain configuration inside the film, which is expected considering that the domain wall length of the MnAs film is about 100 nm.9 Also, the simple domains in the films whose thicknesses are less than 100 nm appear because the film thickness is less than the domain wall length.  When the film thickness is less than 100 nm, the correlation between the spin directions in the domains of the neighboring FM stripes becomes significant. In the MFM image of the demagnetized 100-nm-thick sample, the stripes in the rectangular box indicate that the spin directions of the stripes are aligned in the same direction due to the dipolar interactions between the neighboring stripes.   Figure 2. (a) AFM and (b) MFM image of the 250-nm-thick MnAs film on an area with a large fluctuation in the FM stripe width.   Especially, in the image of the demagnetized 250-nm-thick sample, the two types of domain structures are found despite of its uniform thickness. This indicates that there may be another major origin for the change of the domain structure in the FM stripe in addition to the film thickness. For further understanding, the AFM and MFM images of the 250-nm-thick MnAs film were investigated on the area with a large fluctuation in the stripe width, as in Figures 2(a) and (b). In these images, simple domains are observed where the width of stripes is relatively narrow, whereas the head-on domains appear where the width is relatively wide. This can be explained by the difference in the demagnetizing factor, which can be estimated by width over thickness (w/t) stripes ratio. From the images shown in Figure 1(b), the w/t ratio of the 500-nm-thick sample is 3.1±0.1, whereas 5.8±2.0 is measured for the 100-nm-thick sample. Considering this, the change of the domain structure from the head-on domain to the simple domain can be ascribed to the increase in the demagnetizing factor which is due to the change in the stripe width as well as in the film thickness. In Figure 2(b), the simple domain structure is observed where the width of the stripe is relatively wide. On the other hand, the head-on domain structure is observed where the width is relatively narrow.  The demagnetizing factor also can be influenced by the change in the gap distance between the stripes. As shown in the circular box of the Figure 2(b), the estimated demagnetizing factor of this area is 4.0, which is lower than the average demagnetizing factor of the FM stripes in the film, 4.7±1.3. This indicates that the increased dipolar interaction between neighboring stripes due to the narrow gap induces a simple domain structure despite the small w/t ratio in this particular area.   References 1. G. A. Prinz, Science 250, 1094 (1990). 2. M. Tanaka et al., Appl. Phys. Lett. 65, 1964 (1994); A. Trampert et al., ibid. 79, 2461 (2001). 3. N. Menyuk et al, Phys. Rev. 177, 942 (1969). 4. V. M. Kaganer et al, Phys. Rev. Lett. 85, 241 (2000); Phys. Rev. B 66, 045305 (2002). 5. T. Plake et al, Appl. Phys. Lett. 80, 2523 (2002). 6. L. Däweritz et al, J. Appl. Phys. 96, 5056 (2004). 7. K. Ryu et al., Phys. Rev. B 71, 155308 (2005). 8. K. Ryu et al., Appl. Phys. Lett. 89, 232506 (2006). 9. F. Schippan et al, J. Appl. Phys. 88, 2766 (2000).   Courtesy This note is based on the data kindly provided by Prof. Young Pak Lee, Quantum photonic science research center, Hanyang University, which is published in the article, Applied Physics Letters 89, 232506 (2006).

23 Dec 2013

Production and Measurement of Nanodot Array

Methods of Exploiting Anodic Aluminum Oxide and Block Copolymer as Templates One method of producing nanostructures is by using nanoscale templates. In this note, PS-b-PMMA copolymer and anodic aluminum oxide are used as templates. During the formation process, a Park Atomic Force Microscope (AFM) is used for verifying the size and the shape of the produced templates for nanostructures, and thus produced nano-dot arrays. Overview Like crystals, by making periodic structures, devices such as optoelectric devices, information storage, and optic sensors can be created. In this note, to achieve such periodic structures (or templates), two kinds of methods are introduced. Template Design Methods Templates from Copolymer It is known that covalently bonded copolymers form microphases spontaneously above the glass transition temperature, whose shape and size varies depending on the composition of the copolymer.1, 2 When the microphase formed is of cylindrical shape, the less abundant composite forms the cylinder, which also lies horizontally to the substrate. To use the cylindrical structures as templates for nanodots, the cylinders should lie vertically to the substrate surface, which can be achieved by neutralizing the substrate surface. The neutralization is done by substituting the surface of silicon substrate with hydrogen. On the neutralized substrate, PS(46.1 k)-b-PMMA(21.0 k) the copolymer is spin-coated and baked at 170 °C and under vacuum for 2 days. By applying UV (254 nm) for 3 hours, PS is hardened and PMMA is decomposed. PMMA is washed by acetic acid, and the hardened PS remains as a template. Figures 1 shows the PS template, where the void is the washed PMMA.   Figure 1. AFM images of copolymer template made from PS-b-PMMA. The voids (black features) are the washed PMMA which formed cylinder. The scan size is (a) 2 μm and (b) 0.5 μm, respectively. The holes are 20 nm (± 2nm) in diameter, and are around 40 nm (±1 nm) from one another.   Templates from Anodic Aluminum Oxide A second method for producing a template is using anodic aluminum oxide membrane. When voltage is applied to an aluminum surface in oxalic acid, the aluminum surface is anodized in hexagonal pillar shape. The oxide layer is lifted off by applying HgCl solution and NaOH solution sequentially, and observed with AFM after being put down on silicon substrates. Figure 2 shows the AFM image of the membrane.   Measurement   Figure 2. AFM images of anodic aluminum oxide membrane whose shape is hexagonal. The scan size is (a) 500 nm, and (b) is the 3D rendered image. The holes of cylinder are 50 nm (±2 nm) in diameter, and are around 100 nm (±5 nm) from one another.   It is crucial to control the size and the shape of the templates for the nanodot array applications. The dimension and the form of the units comprising the array are the key parameters which determine the properties of the array. Therefore, measuring and analyzing the status– size, shape, height, and distribution – of the templates and the produced nanodot array enables to control the production parameters. For the measurement and analysis, Scanning Electron Microscope (SEM) is used conventionally, which normally needs gold coating on the sample surface, which destructively affects the sample. On the contrary, AFM measures the sample without giving any deformation, and provides finer, more detailed data than what is provided by SEM. Park AFM which utilizes True Non-Contact mode for measurement, provides superior depth data which is one of the most crucial parameter of the nanodot array. Park AFM’s perfectly decoupled XY and Z-scanner enables the fast Z-scanner feedback which gives accurate height data and the symmetrical XY movement of the flexure guided XY-scanner which secures the linearity of the scanning. Figure 3. AFM images of Co/Pt nanodot array. The image shows that the array is arranged in hexagonal pattern locally. The scan sizes are (a) 2 μm and (b) 1 μm, respectively. The dots are 18 nm (±2 nm) in diameter, and are around 45 nm from one another. Figure 3 shows the AFM images of Co/Pt film deposited on the PS copolymer template by pulsed laser deposition (PLD) method. Co/Pt target is ablated by the laser and fills the voids of the PS copolymer template. By measuring the template and so produced Co/Pt nanodot array, the optimal parameter for the template and PLD can be acquired. In Figure 3, the acquired AFM images show that the diameter and the distance between dots are 18 nm and 45 nm, respectively, which correspond to those of PS copolymer template.   References 1. F. S. Bates et. al., Annu. Rev. Phys. Chem. 41, 525 (1990). 2. G. M. Whiteside et. al, Science 254, 1312 (1991).   Acknowledgement This note is based on the article, Journal of the Korean Vacuum Society 15, 427 (2006). Samples and Images courtesy of Prof. Jeong Sook Ha, Korea University.

23 Dec 2013

Characterization of Organic Photovoltaic Cells

Thin Film of MDMO-PPV: PCBM for Organic Solar Cells Used as an Electro-Optical Device Figure 1. Microstructures of spin-coated 1:4 MDMO-PPV:PCBM thin films from a chlorobezene solution (7 μm scan size). Courtesy of Dr. Jun Seok Lee and Prof. Changmo Sung, Univ. of Mass., Lowell.   The electro-optical industry is pursuing the benefits of low weight flexible polymers in the development of solar cells. One of the most reliable photovoltaic structures is the bulk hetero-junction organic solar cell. Current organic optical devices are based on a soluble fullerene derivative (PCBM) as an electron acceptor and an alkoxy-poly (p-phenylenevinylene) derivative (MDMO-PPV) as a donor. The microstructure of MDMO-PPV:PCBM blends, used in bulk hetero-junction organic solar cells, was studied using the Park AFM to image the surface morphology and to disclose the nanostructure of the film. Typical thin films consist of a 1:4 ratio by weight of MDMO-PPV and PCBM, a soluble electron accepting C60 derivative. A two-phase system was observed that consisted of PCBM-rich domains embedded in a matrix consisting of a mixture of MDMO-PPV and PCBM. We can clearly compare the different PCBM-rich particles between Figures 1 and 2. It has been proven that the change of solvent influences the size of the phase separated PCBM-rich domains. Based on topography information using NC-AFM, the size of particles spin-coated from chlorobenzene was determined to be around 15 nm, in contrast to roughly 100 nm when toluene was used as the solvent. Figure 2. Microstructures of spin-coated 1:4 MDMO-PPV:PCBM thin films from a toluene solution (20 μm scan size).