Study on Matrix Homogeneity and Interfacial Zone of sodium-poly(sialate-siloxo) (Na-PSS) Geopolymers

The First International Seminar on Science and Technology
January 24, 2009 ISBN : 978 979 19201 0 0
Study on Matrix Homogeneity and Interfacial Zone of sodium-poly(sialate-siloxo) (Na-PSS) Geopolymers

Subaer1, Abdul Haris1, Arie van Riessen2
Physics Department, Faculty of Mathematics and Natural Sciences Universitas Negeri Makassar Jalan Daeng Tata Raya, Makassar, Indonesia, 90223 2 Material Research Group, Department of Applied Physics, Curtin University of Technology: PO BOX U 1978, Perth, 6845, Western Australia Abstract
Inorganic geopolymers or simply geopolymers based on silico-aluminate are relatively novel materials with a wide range of potential applications. The main purpose of the present study was to experimentally investigate the composition-microstructure-property relationship of these new materials. These must be understood in order to optimise the performance of geopolymers. Na-PSS geopolymers with different chemical compositions (Si:Al and Na:Al atomic ratios) were prepared by thermally assisted alkali-activation of metakaolinite at 70 oC. Metakaolinite was obtained by dehydroxylation of kaolinite at 750 oC for 6 hours. Measurements indicated that the compositions of geopolymers influence the microstructural character as well as the physical and mechanical properties of these materials. The microstructural characterisation of geopolymers by means of Scanning Electron microscopy (SEM) and Transmission Electron Microscopy (TEM) revealed that the morphology of geopolymers consists of aluminosilicate matrix, unreacted metakaolinite, pores and microcracks. The presence of microcracks observed by SEM and TEM are categorised as secondary microcracks formed during sample preparations. Computed Tomography Imaging (CT-Scan) results for as prepared geopolymers with and without the inclusion of aggregate did not revealed any resolvable cracks. It was also observed that the interfacial zone between geopolymer paste and aggregate has the same chemical composition as the rest of the geopolymer matrix. Keywords: Geopolymers, Kaolinite, metakaolinite, Na-PSS, SEM, TEM,
1
Introduction
Geopolymers produced by using alkaliactivation method of aluminosilicate (such as kaolinite, Al2Si2O5(OH)4) are relatively novel material with a wide range of potential application. The family of geopolymers based on aluminosilicate is called poly(sialates) comprising an amorphous network of AlO4 and SiO4 tetrahedara linked alternately by sharing all the oxygens. Sialate is an abbreviation for silicon-oxo-aluminate (-Si-O-Al-O-). The presence of positive ions, such as Na+, K+, Li+, and Ca++, in the framework is necessary to balance the negative charge of Al3+ in IV-fold coordination with oxygen. This polymeric model is similar to the formation processes of zeolites and zeolite precursors. The empirical formula of poly(sialates) is shown in equation 2.4 (Davidovits 1991): Mn[(-SiO2)z-AlO2]n.wH2O (1)
Where Mn is a cation (the alkaline element), n is a degree of polycondensation, w 3 and z is 1, 2 or 3. According to Davidovits (1991) geopolymers are comprised of several fundamental poly(sialates) units as shown in figure 1.
Figure 1. Geopolymeric molecular networks (after Davidovits 1991). Figure 1 Geopolymeric molecular networks (after Davidovits 1991) The physical and mechanical properties of the resulting geopolymers can be varied by changing the atomic ratio of Si:Al and Na:Al. Poly(sialate-siloxo)(Na-PSS) for example can be used as refractory, high quality cement as well as 1
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
solidification of toxic chemical waste (Davidovits, 1994; Rahier et al., 1996, Lyon, et al., 1994; Phair et al., 2003). Geopolymers produced from dissolution and polycondensation of alkali-aluminosilicate are essentially XRD amorphous in nature. Hos, McCormick and Byrne (2002) systematically analysed the microstructure of geopolymer. By using scanning electron microscopy (SEM) they observed the nanoporous microstructure of their materials. The nanoporous structure is a result of extensive dissolution of aluminosilicate species that occurs before polycondensation commences and consolidates the shapes of the specimen through a chaotic three-dimensional network of polysodium aluminosilicate. The presence of unreacted metakaolinite in the matrix creates a composite-like microstructure of geopolymers and act as in-situ reinforcement responsible for high strength materials. Barbosa, MacKenzie & Thaumaturgo (2000) as indicated by the band at about 1460 cm1 of FTIR spectra, discover the formation of needle like-crystals of sodium carbonate on the surface of the specimen as a result of the reaction between the residual sodium silicate with atmospheric CO2. This suggests that excess sodium is transported by the water to the surface during evaporation to form sodium carbonate. Experiment Geopolymers produced in this study (sodium-poly(sialate-siloxo)) were prepared by alkali-activation of metakaolinite. Metakaolinite was produced through dehydroxylation of kaolinite at 750oC for 6 hours. The kaolinite used in this study was paint/filler Grade Kaolin Clay (Kingwhite 65) supplied by Unimin Australia Limited. The particle size distribution of the material is: 0.05 % > 53 m screen residue, 8.0 % > 10 m, 24 % < 2 m and 68 % < 2 m. The pH (20% suspension) is 7.0 and specific gravity is 2.6. The mineralogical analysis, performed by the supplier, showed that the material contained 99 % kaolinite and a trace of quartz. Technical grade of sodium silicate solution (Na2O.3SiO2) and sodium hydroxide (NaOH) were supplied by Sigma Chemical Ltd. Australia. The chemical composition of these starting materials are presented in table 1.
Table 1 Chemical composition of kaolinite sodium silicate Component (wt%) Al2O3 SiO2 K 2O Na2O CaO MgO Fe2O3 TiO2 LOI (1000oC) H 2O Kaolinite 37.8 46.4 0.21 0.01 0.08 0.15 0.8 0.8 13.8 Sodium Silicate 30.1 9.4 60.5
The sodium silicate solution, sodium hydroxide pellets and water were mixed and hand-stirred until dissolution occurred. Initially the temperature of the mixture was quite high, so it was left until it reached room temperature before proceeding. Geopolymer resins were obtained by the addition of metakaolinite powder to this solution followed by hand mixing or using a shake mixer to ensure homogeneity. The pH of the sodium silicate solution and geopolymer resins was measured using a digital pH from Dick Smith Electronics (Q1416), with pH resolution of 0.01. The resulting geopolymer resin was poured into polycarbonate cylindrical moulds (30 ml container, 2.50 cm in diameter and about 6.00 cm in height), compacted manually using a spatula and vibrated few minutes to release air bubbles and allowed to mature at room temperature for 30 - 60 minutes. Curing was undertaken at 70 oC for 2 hours followed by drying at room temperature before demoulding. Figure 3.1 shows a photograph of the resulting geopolymers, with and without the addition of sand aggregate. Natural sand in the form of -quartz with different particle size and concentration was used as aggregate and poured directly into the geopolymer resin.
Figure 2 Geopolymer samples prepared in this study
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
Samples preparation for SEM investigation is as follows. The samples were cut using a diamond saw to a size of 2.00 to 4.00 mm in height and 10 to 25 mm in diameter. Samples were polished to a 1 m finish using diamond paste. The samples were then cleaned to remove any polishing residue using an ultrasonic bath. The samples were oven dried at about 60 oC for at least 1 hour. Following drying the samples were gold coated for imaging and carbon coated for EDS work. Samples were mounted on the SEM sample stage with conductive, double-sided carbon tape. Thin film specimens of geopolymers for TEM examination were preparedas follows. Sections about 0.5 mm thick and 2.5 cm in diameter were cut with a diamond saw. From these sections several 3 mm diameter discs were cut using an ultrasonic disc cutter (Gatan model 601). These sections (maximum 3) were then mounted on the sample holder of a disc grinder (Gatan model 623) and polished by using 1000, 2000 and 4000 grit silicon carbide paper down to about 40 m. A dimple grinder (Gatan 656) was then used to grind the centre of the disc down to approximately 20 m. The dimple grinding was accomplished by progressively grinding the sample with 1 and 0.25 m size diamond pastes. After dimple grinding the disc was glued 5 minutes araldite onto a 50 mesh copper grid. The sample was then mounted on the sample holder of an ion beam-milling machine (Precision Ion Polishing System (PIPS), Gatan Model 691). Thinning was achieved by impinging 5 6 keV Ar ions on both sides of the thinned region of the disc. The intensity of the beam was controlled by regulating the Ar flow rate to maintain a pressure of 5 x 10-4 torr. These conditions resulted in gun currents of 8 13 A. The thinning was terminated when some regions appeared to be perforated. Ion milling of dimpled samples normally took place between 4 5 hours depending on the initial thickness of the sample.
euhedral some kaolinite crystals are anhedral (no euhedral faces) with elongated, rounded or illdefined morphologies. Selected area electron diffraction (SAED) of kaolinite crystals typically produced diffraction patterns as shown in figure 3(b). Indexing the patterns indicates that the c-axis of the crystals is almost parallel to the electron beam.
(a)
(b)
Figure 3 (a) TEM image of Kingwhite 65 kaolinite, (b) A typical SAED of Kingwhite 65 kaolinite showing the regular hexagonal grid and symmetry of the intensity distribution possessed by the reciprocal lattice plane perpendicular to the c-axis. Dehydroxylation of kaolinite at 750oC causes the crystalline structure of kaolinite to collapse. The loss of water during dehydroxylation of kaolinite destroys dioctahedral kaolinite and tranforms the octahdedrally coordinated Al layer into the tetrahedral form but the Si O sheets of the kaolinite structure are largely conserved (Brindley & Lemaitre 1987). Microstructural examination of metakaolinite by TEM revealed that the morphology of the original kaolinite is maintained although SAED shows the material to be amorphous. The morphology of Na-PSS geopolymers The XRD patterns of all Na-PSS geopolymers show that the structure of these materials are amorphous with no evidence of zeolite. The physical and mechanical properties of these materials are significantly different to those of Na-PS geopolymers. The compressive strength of these materials is between 40 90 MPa with a Vickers hardness between 0.2 0.6 GPa depending on the chemical composition of the sample. Figure 4 show SEM images of a geopolymer sample prepared with Si:Al = 2.0, Na:Al = 1.0. The images show that the sample 3
Results
Microstructure of Kaolinite and Metakaolinite Figure 3(a) shows a representative TEM image of kaolinite. The image shows that the kaolinite is composed of layers of small platy flakes less that 1 m in width and with slightly distorted hexagonal edges. Both TEM and SEM images reveal that while most crystals are
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
still contains unreacted metakaolinite within a geopolymer matrix, pores and microcracks. The image clearly shows that the unreacted metakaolinite acts like aggregate in stopping or deflecting the propagation of microcracks. Figure 5 also shows that this sample is more homogeneous (less unreacted metakaolinite) than the sample prepared with Si:Al = 1.5, Na:Al = 0.6.
(a)
(b)
Figure 5 A representative of CT-Scan image of as prepared Na-PSS geopolymer (a) without aggregate, (b) with 20 wt% quartz aggregate (Si:Al = 1.5, Na:Al = 0.6). Magnification 22 times. TEM examination was performed for further microstructural characterisation of the Na-PSS geopolymer including the determination of the crystallinity of unreacted metakaolinite. Details of sample preparation for TEM examination has been mentioned. Figure 6(a) shows a representative TEM image for a geopolymer sample with Si:Al = 1.5, Na:Al = 0.6.
Figure 4. SEM image of geopolymer sample with Si:Al = 2.0, Na:Al = 1.0 showing unreacted metakaolinite, a pore (air bubble) and microcracks. The presence of microcracks shown on figure 4 was assumed to be due to cutting, polishing and vacuum evacuation in coating unit during sample preparation. In order to ensure that this assumption is correct, a non-destructive imaging technique which does not involving cutting, polishing and coating of the sample was performed to examine the microstructure of as prepared geopolymers. A CT-Scan, also known as Micro-CT (skyscan-1072 Desk-Top-X-Ray Microtomograph) was used for this purpose. The size of the samples was 10 mm in length and 4 mm in diameter. Figure 5 shows a representative of CT-Scan image for as prepared Na-PSS geopolymer (Si:Al = 1.5, Na:Al = 0.6). The image was taken at a magnification where the whole sample fitted into the field of view, which gives the voxel size of 6.477 m by 6.477 m by 6.477 m. This parameter was kept constant to capture and reconstruct the cross section for all samples so that the intensities of cross sections which are linearly related to x-ray absorption density can be compared qualitatively.
Figure 6 (a) TEM morphology of Na-PSS geopolymer (mk = metakaolinite), (b) SAED from the circled area of figure 5.14 showing some retained of crystal order of the unreacted metakaolinite The image reveals grains with the original morphology of metakaolinite bonded to a geopolymer matrix similar to that observed in SEM images. The boundary or interfacial zone between the unreacted metakaolinite and the matrix is continuous indicating a strong intimate bond. Some areas of the specimen appear brighter than the others indicating that the thickness of the specimen is not uniform which may be a result of differences in density. Selected area electron diffraction (SAED) (Figure 6(b)) from the circled 4
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
area reveals that some degree of crystallinity still exists. This suggests that unreacted metakaolinite retains the morphology of the original kaolinite and in some instances also maintains a degree of crystal order.
Discussion
SEM and TEM investigations in conjunction with EDS, SAED, CT-Scan and XRD revealed that Na-PSS geopolymers are amorphous in nature. It was found that the geopolymers consist of unreacted metakaolinite, matrix, and pores. The concentration of unreacted metakaolinite as well as the volume of pores were significantly influenced by the atomic ratio of Si:Al and Na:Al. These results are in good agreement with those reported by Hos, McCormick & Byrne (2002) dan Rowles & OConnor (2003). The presence of unreacted metakaolinite in the matrix creates a composite-like microstructure of geopolymers and act as in-situ reinforcement responsible for high strength materials. The role of unreacted metakaolinite during geopolymer synthesis was also observed by Wei et al.(2003) by means of Environmental Scanning Electron Microscope (ESEM).
Conclusion
The microstructural characterisations by SEM and TEM revealed that the morphology of geopolymers consists of geopolymer matrix (aluminosilicate paste), unreacted metakaolinite, pores and microcracks. The CT-Scan images revealed that the presence of microcracks were due to cutting, polishing and vacuum evacuation in coating unit during sample preparation. No cracks were found in as prepared Na-PSS geopolymers with and without aggregate.
Reference
Barbosa, V. F. F. & MacKenzie, K. J. D. (2003), "Thermal behaviour of inorganic geopolymers and composites derived from sodium polysialate", Materials Research Bulletin, 38, (2), 319 - 331. Barbosa, V. F. F., MacKenzie, K. J. D. & Thaumaturgo, C. (2000), "Synthesis and characterisation of materials based on inorganic polymers of alumina and silica: sodium polysialate polymers", International
Journal of Inorganic Materials, 2, (4), 309317. Campbell, D. H. & Folk, R. L.(1991) "The ancient Egyptian pyramids - concrete or rock?" Concrete International, August, 2944. Davidovits, J. (1982), "Mineral polymers and methods of making them", 4,349,386 United States Patent. Davidovits, J. (1988), "Geopolymers of the first generation:SILIFACE-Process," Geopolymer '88, First European Conference on Soft Mineralogy, Compiegne, France,49-67. Davidovits,J.(1991),"Geopolymers:Inorga-nic Polymeric New Materials", Journal of Thermal Analysis, 37, 1633-1656. Davidovits, J. (1994), "Properties of geopolymer cements", Alkaline Cements and Concretes,Kiev State Technical University, Ukraine,131-149. Davidovits, J.(2000), "The dependence of geopolymer on Si:Al", [online], http://www.geopolymer.org, [21 July 2001]. Davidovits, J. & Sawyer, J. L. (1985) "Earlyhigh strength mineral polymer" 4,509.958 United States Patent. Hos, J. P., McCormick, P. G. & Byrne, L. T. (2002), "Investigation of a synthetic aluminosilicate inorganic polymer", Journal of Materials Science, 37, 2311-2316. Lee, W. K. W. & van Deventer, J. S. J. (2002), "The effects of inorganic salt contamination on the strength and durability of geopolymers", Colloids and Surfaces A: Physicochem. Eng. Aspects, 211, 115-126. Phair, J. W., Smith, J. D. & van Deventer, J. S. J. (2003),"Characteristics of aluminosilicate hydrogels related to commercial "Geopolymers"", Materials Letters, 4525, 112. Rowles, M. & O'Connor, B. H. (2003),"Chemical optimisation of the compressive strength of aluminosilicate geopolymers synthesised by sodium silicate activation of metakaolinite", Journal of materials chemistry, 13(13), 1-6. Subaer, van Riessen, A., O'Connor, B. H. & Buckley, C. E. (2002),"Compressive strength and microstructural character of aluminosilicate geopolymers", Journal of the Australasian ceramic society, 38(1), 83-86. Temuujin, J., Okada, K., MacKenzie, K. J. D. & Jadambaa, T. (1998), "The effect of water vapour atmospheres on the thermal transformation of kaolinite investigated by XRD, FTIR and solid state MAS NMR", 5
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
Journal of the European Ceramic Society, 19, 105 -112. van Jaarsveld, J. G. S., van Deventer, J. S. J. & Lorenzen, L. (1997), "The potential use of geopolymeric materials to immobilise toxic metals: Part I. Theory and applications", Minerals Engineering, 10, (7), 659-669. van Jaarsveld, J. G. S., van Deventer, J. S. J. & Schwartzman, A. (1999), "The potential use of geopolymeric materials to immobilise toxic metals: Part II. Materials and leaching characteristics", Minerals Engineering, 12, (1), 75-91. Wei, S., Yun-sheng, Z., Wei, L. & Zhi-yong, L. (2003), "In situ monitoring of the hydration process of K-PS geopolymer cement with ESEM", Cement & Concrete Research, 34, (6), 935-940.
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
Joint Analysis Of Shallow-Subsurface Seismic Properties Beneath Extra-High-Voltage PLN Towers

Eddy Hartantyo
Laboratorium Geofisika, Fakultas Matematika dan Ilmu Pengetahuan Alam Universitas Gadjah Mada, Bulaksumur, Yogyakarta Email:hartantyo@ugm.ac.id Abstract
Joint analysis of two shallow subsurface seismic methods, i.e. refraction seismic and spectral surface wave has been conducted. Refraction seismic may serve compressional wave velocities (Vp) at several layers or nodes and spectral surface wave analysis result shear wave velocities (Vs) beneath survey lines. Vs slices beneath tower lines shows capabilities of this method to infer low velocity zones lay under high velocities interpret as basement whereas refraction seismic method is not. These two variables plus densities are meaningful variables to calculate all rocks properties, i.e. Poisson ratio, Bulk modulus, and lame constants. Beneath PLN towers, seismically shows the water level existences especially with Vs slice, both for very shallow and shallow water level. Otherwise for Vp slice we can infer depths of civil-basement which could be reached for relative tall building such as towers. It also concludes that refraction seismic has better depth resolution than MASW, but MASW could reach deeper. 2) Several rock properties can be calculated indirectly by measuring Vp and Vs of medium using MASW and refraction analysis. 3) Unconsolidated sands occur at both tower, but at kanaka we found harder sediments lay below 4m depth with groundwater infill at some layer parts. 4) There is no compaction beneath these towers from seismic properties point of view. Keywords: Shear and Compression velocity, shallow seismic, PLN tower.
Introduction
Soil and rocks in subsurface have elastic properties which can be divided to wave and rock properties. Both are interconnected and rearranged by several equations [1]. Usually rock properties were measured directly with several technical instruments such as zonder, penetration test, and shear modulus test and so on. Otherwise, wave properties were calculated with several seismological instruments, such as seismometer/geophone, data logger, timing instruments etc to collect waveforms for natural or artificial sources. In this paper, wave properties have been carried out with two kind of seismic data acquisition i.e. refraction and MASW (Multi channel Analysis of Surface Waves) seismic. Refraction seismic yield subsurface layers or blocks with their pressure/compression wave velocities (Vp) in adjacent depths, and in other hand, surface wave analysis yield secondary/shear wave velocities (Vs) also in several subsurface layers or blocks. Based on measured Vp and Vs at the same locations, several rock properties can be inferred, i.e. Vp/Vs ratio, Poissons ratio and Youngs Modulus. They are very informative constants
which correlated with soil strength and compaction. This technique was applied to analyze subsurface soil properties beneath PLN Extra High Voltage (SUTET) Tower to understand the compaction, water saturation and strength of soil/compacted soil for its standing. Refraction Seismic Refraction seismic assumed that velocity layers are increasing with depth. There will be pure refraction wave from incoming wave which; moving downward, reaching subsurface contrast, moving in critical refraction angle and creep again to surface. By creating a simple point source in surface and deploying some geophones on some offsets, we can record these shaking and then define the time travel from source to each geophone [2]. Recorded ground shaking will consist of direct, reflected and refracted waves. Only direct and refracted waves are used indicated from first shaking events from source. [3] There are several methods to convert wave time travel to layering velocities. In this paper, we were using a linear tomography inversion [4] which added-on the seisimager software. 1
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
On the travel time tomography [4], model area has been chopped in grid scheme with Cartesian coordinate (i,j). Each cells denoted with position i,j and has their respective slowness (S) and width. First we should create several four-side area with shape area are denoted as lij and their slowness Sj (see figure 1).
Multichannel Analysis of Surface Wave (MASW) Important use of seismic involving surface waves is clearly shows by MASW-Multichannel Analysis of Surface Waves [5] for determining shallow Vs. The active MASW uses active source to generate surface wave and their ground shaking creeps are recorded by several geophones for several offsets. MASW method is raise from fact that SASW-Spectral analysis of Surface Waves [6], which are only using two receivers, is failed to analyze subsurface Vs which with relatively high lateral heterogeneity. This heterogeneity will strongly affect wave dispersion characteristic. But, for several reason, i.e. effectively and lowcost instruments, this method is widely used for civil construction to investigate building layer thicknesses; either for wall or floor. The MASW methods have been widely used for several purposes. Shear velocities mapping beneath camp-ground has been done by [7]. They are succeeding for analyzing stiffness variations of soil in vertical and lateral direction by inverting fundamental mode of Rayleigh wave. This method also has been implemented to analyze bedrock depth which highly important for determining building ground-base depth, design, and geometry in Olathe area, Kansas [8]. On the near surface geological analysis, MASW has been successfully used to delineate clay, water saturated sand and gravel layers at Alabama area [9]. Also in geohazard monitoring, attributes of ground-soil base beneath several housings has been analyze at Tampa area, Florida [10]. This area is covered with thick limestone which has high possibility for ground subsidence. In his publications, he modify geophones which can be used without ground embedded, but only laying geophones down with small circular metal-plate. This method also used to analyze shallow subsurface ground-base of hundreds wind turbine in Kansas area [11] and analyze Vs section beneath PLN ultra high voltage towers [12].
Figure 1. travel time tomography and its ray path from source to receiver (Hartantyo dan Sismanto, 2007) Complete time travel from source to receiver (ti) could be calculated from adjacent rectangular shape X as [2];
ti =
dX = s ( X )dX v( X ) X
(1)
Eq (1) could be digitized as;
t i = s j lij
j =1
(2) and finally equation 2 could be rearranging to matrix formulation as;
l11 l12 t1 t s1 l 2 s 21 l 22 = 2 l31 l32 s N l t M M 1 lM 2
l1N l2 N l3 N l MN
(3)
[T ] = [S ][L]
with M is number of simultaneous equation and number of known travel times, and N is number of unknown slowness. T indicated as trace line vector, S is a slowness model vector and L is a ray path matrix. In this paper, solution of (3) is using a least square method. Equation on (3) can be solved only if M greater or equal with N.
Rock/Soil Properties Relation of rocks or soil and wave properties was derived from elastic analysis of materials. From their stress and strain relationship, we can
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
derive rock properties from wave properties with their respective densities, see equation 4:
3 2 4 2 E = 2 2 2 2 2 2 = 2( 2 2 )
(4)
simplicity of refraction seismology data acquisition, we can rearrange the MASW deployment for refraction purposes easily. Both data then analyzed on similar offsetdepth points for rocks properties calculation. MASW analysis follow of work by [12] and refraction analysis also follow of work of [13]. Only Poisson ratio and Young modulus have been calculated from several rock properties.
k = ( )
2 4 3 2
= = ( 2 2 2 )
2
Result and Discussion

Separated process had been conducted for both refraction seismic and MASW data. It seem from analysis that depth responses of MASW is deeper than refraction seismic. Other wise, refraction seismic had better spatial resolution. These phenomena were inline with theoretical basis that refraction seismic will give shallower resolution for shorter offset. Figure 3 shows sections of Vs and Vp wave velocity beneath Salakan Tower. Vs section had been taken from [12], and Vp section yielded from refraction analysis, with similar data used in Vs calculation. Black dot represents data points.
Offset (m) 0 100 2 4 6 8 10
with E = Youngs modulus, = Poissons ratio, k = Bulk modulus, = Shear modulus, = Lame constant, = density, = Vp and = Vs. Because of limitation procedure to measure subsurface rock densities, in this paper, densities are not including in calculations. Experimental All seismic data are similar with previous work of [12], using two location of SUTET tower which are shown in figure 2. First location is Kanaka Tower, laid around 30m in front of Kanaka Sejahtera Residence main entrance gate, at Ngaglik sub district, Sleman district. This tower is one of hundreds tower as a main electric power transmission trough Jogjakarta.
Offset (m) 0 100

200
10
1500
95
95
Elevasi (m)
150
90
1000
90
100
85
500
85
80
50
80
a)
b)
Figure 3 Secondary (a) and Primary (b) velocity cross sections at Salakan Tower Depth penetration of MASW is clearly higher than refraction analysis which can be seen from point distributions. Maximum depth of MASW reached nearly 20m but refraction seismic only reaches 7m (coincidence with a 12m maximum offset). This phenomena also inline with Poisson solid medium [14] which ratio Vp/Vs = Vp and Vs is ratio line only (shown by red
Figure 2 PLN Sutet tower location at North West (Tower Salakan) and North (Tower Kanaka) city of Jogjakarta (Hartantyo et al, 2008) The second tower is Salakan tower, located 200m west part of nearby west ring road. Seismic data were acquired based on the MASW deployment platform with several shot in between geophones. Otherwise, because of
3 (see figure 4). Ratio between

closely near with Poisson solid for 5 meter depth from surface dot). For deeper subsurface, the 3
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
relationship is nearly weird extrapolation of Vp section.

250
because
of
200
150
80-84.9m
Ratio of Vp and Vs also show blue colors near surface with Vp/Vs lower than 2 [15]. This values correspondence with sand value [16] at sand-shale separation in borehole log analysis. For subsurface deeper than 95m, unfortunately these Vp/Vs ratio data are not analyzable due to depth limitation of refraction seismic. Nearly similar geometry with Vs section at figure 3a, Youngs modulus section (see fig 5c) has shown lower value nearly surface. It also interpreted as unconsolidated sand although no densities are included. a)
0 100 5 10 15 20 450 95
Vs (m/s)
100
85-89.9m 90-94.9m 95-100m Poisson Solid
50 0 250 500 750 1000 1250 1500 1750 Vp (m/s)
Figure 4 Plot Vp (fig 6b) versus Vs (fig 6a) interpolated at similar points classified as four class: red dot (elevation 95-99m), yellow dot (elev 90-94m), green dot (elev 85-89m), blue dot (elev 80-84m) and theoretical Poisson solid medium (small circle). a)
0 100 2 4 6 8 10 100 0 2 0.5 95 95 10 95 11000
350 Elevation (m) 90
250 85 150 80 50
b)
0 5 10 15 20 100
b)
4 6 8 10 100 0 2 4
c)
6 8 10
1300
95
0.4 Elevasi (m) Elevasi (m) Elevasi (m) 8 90 6
Elevation (m)
1000
90
0.3
90
90
0.2 4 85 0.1 85 85
60000
700
85
80 Modulus Young 10000
80 Poisson Ratio
80 VP/VS Kanaka
Figure 5 Indirect elastic properties beneath Salakan Tower; (a) Poisson Ratio, (b) Vp/Vs Ratio and (c) Modulus Young (E/). Other rock attributes could be calculated with Vp and Vs data (see eq 4) especially attribute without density (). Poisson ratio section beneath Salakan could be seen in fig 5a. Relatively medium Poisson value occurs as spots shallower than 1m interpreted as unconsolidated sand contained with shallow ground water. Other reddish colors interpret as more consolidated sand with lower porosity. It could also contained small portion of ground water. This data inline with Vs section [12] which show relatively low Vs which interpret as ground water contained unconsolidated sand.
400
80
100
Figure 6. Secondary (a) and Primary (b) velocity cross sections at Kanaka Tower
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
500
80-84.9m 85-89.9m 90-94.9m 95-100m Poisson Solid
good analyzed data only occur at maximum 10m depth. a)

0 100 5 10 15 20 0.5 100 10 0.4 95 0.3 Elevation (m) 95 8 0 5 10 15
400
b)
20
Vs (m/s)
300
90
Elevation (m)
200
90
0.2
4 85 0.1 2 85
100
80 Poisson Ratio
80 VP/VS Kanaka 0
0 0 500 Vp (m/s) 1000 1500

0 100 5 10
c)
15 20 1000000 95
Figure 7 Plot Vp (fig 6b) versus Vs (fig 6a) interpolated at similar points classified as four class: red dot (elevation 95-99m), yellow dot (elev 90-94m), green dot (elev 85-89m), blue dot (elev 80-84m) and theoretical Poisson solid medium (small circle). From these points of view, hard rock zones which required by tall civilian building (tower) are not exist at interpreted depth (6m). Only unconsolidated sand occurs until 6m depth, and probably more consolidated sands occurs at deeper than 6m. Deeper data analysis were shown at seismic result, both from MASW and refraction seismic (see figure 6). These sections had been processed for line section beneath Kanaka tower with North-South azimuth. Maximum depth reached with MASW method also similar to Salakan tower with shorter offset (nearly half). Optimum depth for both method is 15m (reached elevation 85m), from data points distribution. Relatively low Vs occur at surface until 4m deep (figure 6a) then followed with high Vs at 8m thickness [12] interpreted as unconsolidated sand at near surface with water content on their pores and more consolidated sand with low ground water contents. It also seen in Vp section (fig 6b) which Vp lower than 500m/s interpreted as unconsolidated sand. Maximum depth of unconsolidated sand from this figure predicted only 4m deep from surface (at elevation 96m). Relatively deep analyses data also could be seen on Vp/Vs plot data (see fig 7). Red and yellow dots data were nearly coinciding with Poisson-solid medium. Otherwise, blue and green dots, which correspondences with depth lower than 89.9m, were relatively far and did not give significance ratio. From this point of view,
Elevation (m)
90 500000
85
80 Modulus Young
Figure 8 Indirect Indirect elastic properties beneath Kanaka Tower; (a) Poisson Ratio, (b) Vp/Vs Ratio and (c) Modulus Young. Beneath Kanaka Tower, Poisson ratio dominated middle to low value (figure 8a), under 0.3. It shown that upper medium dominated with sand. At depth 3m, there are thin (nearly 1m thick) blue Poisson values at offset 4-10m. It was interpret as very shallow sand contained with ground water. Other blue closures also occur on the center of section, laid at offset 10-15mat 69m depth. It probably consists of relatively high fluid content but has a similar lithology in lateral variation. Dominated un and consolidated sand, from Vp/Vs section (see figure 8b), occur from surface until 11m depth (91m elevation). Small portion of relatively high values of Vp/Vs (nearly 6) found at 5-7m depth at offset 0-3m which interpreted as tight medium (sandstone boulder size?). Also similar values of Vs section yield from Youngs modulus section (figure 8c). It shows medium values near center of section, between 3-15m offset and 7-10m depth. From these seismic properties analysis, hard rock zones beneath Kanaka Tower exist at interpreted depth nearly 4m, but there are groundwater infills at the center. Similar unconsolidated sands dominate section from surface until 4m depth.
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
Conclusions
Some conclusion could be risen from this research; 1) refraction seismic has better depth resolution than MASW, but MASW could reach deeper. 2) Several rock properties can be calculated indirectly by measuring Vp and Vs of medium using MASW and refraction analysis. 3) Unconsolidated sands occur at both tower, but at kanaka we found a harder sediments laid below 4m depth with groundwater infill at some layer parts. 4) There is no compaction beneath these towers from seismic properties point of view. [9]
[10]
Acknowledgements
We thanks LPPM UGM for funding this research under Young Researchers Grant 2008, Geophysics Laboratory for seismic instruments rent and, Wiwit, Afif, Joko and Rudi for field data acquisition help.
[11]
[12]
References
[1] Telford, W.M., Geldart, L.P., and Sheriff, R.E., 1990, Applied Geophysics Second Edition, Cambridge University Press, Melbourne. Grant, F.S. and West, G.F., 1965, Interpretation Theoryin Applied Geophysics, McGraw-Hill, NewYork Parasnis, D.S., 1997, Principles of Applied Geophysics, 5th ed., Chapman & Hall, London anonim, 2001, Text-tomography-analysis, manual theory attached on Seisimager software (licensed), OYO Corp, Japan. Hayashi, K. and Suzuki, H., 2004, CMP cross-correlation analysis of multi-channel surface-wave data; Exploration Geophysics, v 35, 713 Nazarian, S., Stoekoe, K.H. and Hudson, W.R., 1983, Use of Spectral analysis of Surface waves method for determination of moduli and thicknesses of pavemen systems: Transport. Res. Record, 930, 3845 Park, C.B. and Miller, R.D., 2004, MASW to Map Shear-Wave Velocity of Soil, Kansas Geological Survey Open File Report 2004-30 Miller, R.D. and Xia, J., 1999, Using MASW to Map Bedrock in Olathe, Kansas, Kansas Geological Survey Open File Report No 99-9. Final report to [13]
[2] [3] [4] [5]
[14]
[15]
[6]
[16]
Harding Lawson Associates, Lees summit, Missouri Xia, J., 2006, Delineating Subsurface Features with the MASW Method at Maxwell AFB in Montgomery, Alabama, Kansas Geological Survey Open File Report No 2006-1. Report to Andrew Weinberg, Bechtel-S Corp., Texas Miller, R.D., Xia, J. And Park, C.B., 1999, MASW to investigate Sunsidence in the Tampa, Florida Area. Kansas Geological Survey Open File Report 9933. Report to ELM Consulting LLC, Olathe, Kansas Park, C.B., and Miller, R.D., 2005, Seismic Characterization of Wind Turbine Sites in Kansas by the MASW Method, Kansas Geological Survey Open File Report 2005-23. Report to Barr Engineering Company, Minneapolis Hartantyo, E., Afif, R., and Wiwit, S., 2008, Active Multichannel Analysis of Surface Waves (MASW) Survey for SUTET tower base soil compaction imaging, Presented in ICMNS 2008, ITB, Bandung Sismanto and Hartantyo, E., 2007, Resistivity and Refraction Seismic Mapping in Sewu Temple Area, Proceeding of Joint Convention IAGI, HAGI, IATMI, Bali, October 2007. Aki, K., and Richard, P.G., 1980, A Quantitative Seismology: Theory and Methods, vol 1, W.H. Freemann, New York Hanitzsch, C., de Vincenzi, L., Michel, J.M. and Semond, D., 2006, Dual Inversion Applied to 2D MultiComponent Seismic Data Onshore Libya, EAGE 68th Conference & Exhibition Vienna, Austria, 12 - 15 June 2006 Schlumberger, 2008, Glossary on Oil Prospecting, www.schlumberger.com
[7]
[8]
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
Preparation and Characterization of TiO2 Nanoribbons , Nanowires and Nanotubes by Hydrothermal Method
Shahidan Radiman, Anamt Mohamed Noor
School of Applied Physics, Faculty of Science and Technology, Universiti Kebangsaan Malaysia, 43600 Bangi, Selangor, Malaysia. E-mail: shahidan@ukm.my Abstract
Preparation of TiO2 nanomaterials with various size and morphology such as nanosheets, nanowires and nanotubes has attracted prominent interests due to their unique physical and chemical properties including large specific surface area, photocatalytic potential, solar cell/batteries potential and electroluminescent hybrid devices. Several methods such as solvothermal, sol-gel and hydrothermal have been applied to prepare TiO2 nanomaterials and in this research approach, we have successfully synthesized TiO2 nanosheets, nanowires and nanotubes using a simple, low cost and template-free hydrothermal method and characterized by XRD, SEM, EDX and TEM. The width of the nanowires falls in the range of 5-10 nm and the length ranges from few micrometers to several tens of micrometers. Experimental results confirmed that this technique is reliable for preparation of narrow size distribution, controlled shape and morphology of TiO2 nanomaterials. Keywords: TiO2 nanosheets, nanowires, nanotubes and hydrothermal
Introduction
Titanium dioxide is a wide-band gap semiconductor. It is widely used in industrial applications such as in dye-sensitized solar cell [1], gas sensing [2], photocatalysts and self-cleaning coating [3-4], electroluminescent hybrid device [5], water-splitting catalysts for hydrogen generation [67], and energy storage technologies [8] because of its nontoxicity, low cost and chemical stability. with high One-dimensional TiO2 morphological specificity, such as nanotubes, nanosheets, nanowires, nanofibers, nanoribbon and nanorods have attracted particular interest because of their unique microstructures and promising functions [9-10]. In the past few years, morphology controllable synthesis of TiO2 has attracted a lot of attention. Many techniques, including microemulsion mediated hydrothermal method [11], sol-hydrothermal approach [12], reverse micelle method [13], lowtemperature process [14], ultrasonic approach [1516], and hydrothermal method [17] have been applied to synthesize nanostructured TiO2 with well defined morphologies. Experimental All the reactants and the solvents were of analytical grade and were used without further purification.
Synthesis Preparation of TiO2 Nanoribbons 30 ml of 10 M NaOH aqueous solution was mixed with 30 ml ethanol under constant magnetic stirring. 1 g of TiO2 nanopowder (P-25, spherical from SigmaAldrich) was added into the above mixture respectively. The mixture was then transferred to a Teflon-lined stainless steel autoclave and put into furnace at 180C for 24 h. The obtained products were collected and washed with 0.1 M HCl solutions and deionized water for several time. After that the product were dried at 60C for 3 h and calcined at 350C for 3 h. Preparation of TiO2 Nanotubes The nanotubes TiO2 were synthesized by using the same route of section 2.1.1 but by replacing the ethanol with triethanolamine. Preparation of Nanowires TiO2 The typical experimental procedures are depicted as follows: As a starting material, monodispersed TiO2 particulate were prepared. Titanium butoxide was mixed with the same mol of acetylacetone to slowdown the hydrolysis and condensation reactions. Subsequently, this mixed solution was added in the 40ml distilled water and the solution was stirred at room temperature for 5 min. After kept stirring, 30 ml ammonia aqueous solution 25% was added in the solution drop by drop. After 3 minute of stirring, the solution was put into a Teflon-lined stainless steel autoclave and heated at 180C for 20 and 24 hours. 1152
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
The obtained product was washed with distilled water, 2-propanol and HCl aqueous solution for several time, followed by drying to collect the powder form. TiO2 nanowires were synthesized by adding monodispersed TiO2 to a 10 M aqueous solution of KOH. A total of 0.2 g of monodispersed TiO2 and 20mL of 10 M KOH aqueous solution was mixed and after stirring for 1 h the resulting suspension was transferred to a Teflon-lined stainless steel autoclave. The autoclave was heated at 180C for 24h. After it was cooled down to room temperature, the product was repeatedly and ultrasonically washed by dilute HCl solution (0.1M) and distilled water. Characterization Transmission electron microscope, TEM (Philips CM 12 operated at 100kV) and SEM (Leo 1450 VP) was used to examine the size and morphology of the TiO2 nanostructures. For the EDX characteristic, INCA 2000 were used. The nanostructures crystalline phase was determine by Xray diffraction (XRD) using a Philips diffractometer employing a scanning rate of 0.02 s-1 in a 2 range from 20 to 60 with Cu K radiation ( = 1.5418 Angstrom).
Figure 1. XRD patterns of samples prepared at 180C for 24 h. (a) NanoribbonTiO2, (b) Nanowires TiO2 and (c) Nanotubes TiO2. The peaks are labelled with ( ) denotes the sodium titanate while ( ) denote the anatase peak and ( ) the rutile peak. The morphologies and microstructures of the prepared samples were investigated with TEM and SEM. Figure 2-4 shows that the morphologies and dimension of the samples were strongly dependent on hydrothermal treatment. During the synthetic process of one-dimensional TiO2 particles, the used organic solvent determined the morphologies of the product [18]. When the solvent was ethanol, sheets-like nanostructures were obtained and when the solvent was triethanolamine, TiO2 nanotubes were formed. Figure 2 shows the TEM images for the nanosheets samples, from which one can see the mixture of sheets and rod-like shape formed. The diameter of TiO2 nanoribbon was about 50 to 100 nm and its length was about 100 nm to several micrometres. Fig 3 shows a typical morphology of the asprepared TiO2 nanotubes hydrothermal treatment for 24 h. The diameter size is around 4-5 nm in inner diameter and 10-11 nm in outer diameter while the length of the nanotubes was up to 100 nm. The different solvent played a key role in the growth process. In the initial stage, the TiO2 powder reacted with NaOH to produce thin titanate nanoribbons quickly. The titanate nanoribbons were not stable and had tendency to decompose into TiO2 nanorods or other structures while for the nanotubes, hydrothermal treatment with high pressure and a high temperature in this case at 180C, the layered structure would roll up into nanotubes due to surface tension. The properties of solvents, such as dielectric constant were different due to the presence of ethanol and triethanolamine which would influence the morphologies of the final product. Figure 4 shows the nanowires obtained with hydrothermal treatment in KOH aqueous solution at

Figure 1 shows XRD patterns for the different samples prepared with hydrothermal method 180C for 24 h. For the nanosheets the peak positions and their intensities are consistent with the standard powder diffraction pattern of anatase and rutile (JCPDS no. 21-1272 and JCPDS no. 77-0441) indicating the formation of anatase and rutile crystal structure. The broadening of TiO2 peak is due to reduction in crystallinity.Comparing the nanosheets with nanowires, the rutile peak in nanowires appear low then nanoribbons while in the nanotubes the sodium titanate are found which indicated Na+ still not fully removed.
Proceeding Book
1153

January 24, 2009 ISBN : 978 979 19201 0 0
180C for 24 h. The exact mechanism of hydrothermal method in formation of nanowires is still a controversial topic debated extensively in contemporary literature. In this paper, monodispersed spheres TiO2 particulate have been transformed into non-hollow nanowires with a small distribution both in length and in diameter. From the Figure 4 and 5(c) the diameters are in the range of 5-10 nm and the lengths are between few micrometres to several ten of micrometers.
Figure 2. TEM images of nanoribbons TiO2 hydrothermal treatment at 180C for 24 h.
Figure
3.
TEM images of nanotubes TiO2 hydrothermal treatment at 180C for 24 h.
Figure
4.
TEM images of nanowires TiO2 hydrothermal treatment at 180C for 24 h.
Based on the TEM and SEM images and experimental result, the possible formation mechanisms of the TiO2 nanostructured with various morphologies were proposed. As illustrated in Figure 6. At the primary stage of experiment, TiO2 nanopowder and monodispersed particulate were prepared first as a starting material for the formation of TiO2 nuclei. The further growth (maintained at 180C for 24 h) of those nuclei is remarkably influenced by the crystallographic structure of TiO2, hydrothermal treatment and experimental conditions. When TiO2 reacts with highly alkaline NaOH/KOH solution, some of the Ti-O-Ti bonds of the raw material are broken and layered titanates with Na+/K+ metal ions are formed in the form of thin small sheets. The formation of nanowires may arise from the very fast growth rate and rapid splitting of the layered titanate nanoribbons in highly alkaline solvent [1921]. After washing with HCl and distilled water, hydrated titanate (H2Ti3O7) are formed by the substitution of Na+/K+ by H+. The final TiO2 product are obtained when dehydration of the hydrated titanates are occurred at the time of the calcinations. Previous study described that the nanotubes were formed in highly alkaline aqueous solutions at low temperature (110-150C) [22]. In this experiment the nanotubes formed at 180C in the presence of triethanolamine. During the reaction process, titanium powder react with NaOH mixed with triethanolamine forming layered alkaline titanate which can be found at the beginning of the reaction. These layered titanate are very thin and easily exfoliate into nanoribbons that are highly anisotropic in two dimensions. At a high temperature and pressure, the layered structure assembled to link together by sharing edges and would roll up into nanotubes due to electrostatic interaction and surface tension [23]. So the tubular structure is formed in the hydrothermal treatment. Further study on the growth and formation mechanisms is still in progress.
Figure 5. shows the summarized results SEM and EDX of the prepared TiO2 (nanoribbons, nanotubes and nanowires). As can be seen clearly from these SEM images, the results support the TEM images. Large-area EDX spectra analysis was run to reveals the existence of Ti, O, K and Na as to know the effectiveness of HCl/H2O washing treatment. Acid treatment and washing with water plays an important role in controlling the amount of residuals Na+ and K+ ions remaining in the sample solution. From the Figure 5 (d-f) we can say Na and K ions can be remove with washing treatment but a small amount of Na+ and Cl- ions remain in the sample due to washing process.
Proceeding Book
1154

January 24, 2009 ISBN : 978 979 19201 0 0
[3]
[4] Figure 5. SEM images of the prepared TiO2, (a) nanoribbonss TiO2, (b) nanotubes TiO2 and c) nanowires(rods) TiO2. (d,e,f); energy-dispersive X-ray spectra (EDX) of the above products indicate the existence of organic residue of washing process.
[5]
[6] [7]
Figure 6. Schematic illustration of the growth mechanisms of the products prepared under different experimental conditions.
[8]
Conclusions
Based on the results of the experiments, TiO2 nanostructures synthesized by hydrothermal method are briefly introduced. This method will be further exciting addition to the fast growth of nanostructures TiO2 by tailoring the experimental conditions. This well defined nanostructure might have other novel properties which are currently being investigated in our laboratory.
[9] [10]
[11]
Acknowledgement
The author gratefully acknowledges the financial support of National University of Malaysia through Grant No: UKM-OUP-NBT-27-138/2008.
[12]
References
[1] [2] Grtzel, M. 2003. Dye-sensitized solar cells. J. of Photochemistry and Photobiology C: Photochemistry Reviews. 4: 145-153. Duy, N.V., Hieu, N.V., Huy, P.T., Chien, N.D., Thamilselvan, M. & Yi, J. 2008. Mixed SnO2/TiO2 included with carbon nanotubes for gas-sensing application. Physica E: Lowdimensional Systems and Nanostructures. 41: 258-263.
[13]
[14]
Fujishima, A., Zhang X. & Tryk, D.A. 2008. TiO2 photocatalysts and related surface phenomena. Surface Science Reports. 63: 515582. Zhao, X., Zhao, Q., Yu, J. & Liu, B. 2008. Development of multifunctional photoactive self-cleaning glasses. J. Non-Crystalline Solids. 354: 1424-1430. Carpi, F. & Rossi, D.D. 2006. Colours from electroactive polymers: Electrochromic, electroluminescent and laser devices based on organic materials. Optics & Laser Technology. 38: 292-305. Holladay, J.D., Hu, J., King, D.L. & Wang, Y. 2009. An overview of hydrogen production technologies. Catalysis Today. 139: 244-260. Dholam, R., Patel, N., Adami, M. & Miotello, A. 2008. Physically and chemically synthesized TiO2 composite thin films for hydrogen production by photocatalytic water splitting. International Journal of Hydrogen Energy. 33: 6896-6903. Wu, M.S., Wang, M.J., Jow, J.J., Yang, W.D., Hsieh, C.Y. & Tsai, H.M. 2008. Electrochemical fabrication of anatase TiO2 nanostructure as an anode material for aqueous lithium-ion batteries. Journal of Power Sources. 185: 1420-1424. Tsakalakos, L. 2008. Nanostructures for photovoltaics. Materials Science and Engineering: R: Reports. 62: 175-189. Lim, J.I., Yu, B., Woo, K.M. & Lee, Y.K. 2008. Immobilization of TiO2 nanofibers on titanium plates for implant applications. Applied Surface Science. 255: 2456-2460. Lu, C.H., Wu, W.H. & Kale, R. B. 2008. Microemulsion-mediated hydrothermal synthesis of photocatalytic TiO2 powders. Journal of Hazardous Materials. 154: 649-654. Huang, D., Liao, S., Quan, S., Liu, L., He, Z., Wan, J. & Zhou, W. 2008. Synthesis and characterization of visible light responsive N TiO2 mixed crystal by a modified hydrothermal process. Journal of NonCrystalline Solids. 354: 3965-3972. Wang, Y., Zhou, A. & Yang. Z. 2008. Preparation of hollow TiO2 microspheres by the reverse microemulsions. Materials Letters. 62: 1930-1932. Ambrus, Z., Mogyorosi, K., Szalai, A., Alapi, T., Demeter, K., Dombi, A. & Sipos, P. 2008. Low temperature synthesis, characterization and substrate-dependent photocatalytic activity of nanocrystalline TiO2 with tailor-made rutile to anatase ratio. Applied Catalysis A: General. 340: 153-161. 1155
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
[15]
[16]
[17]
[18]
[19]
[20]
[21]
[22]
[23]
Halalay, I.C. & Balogh, M. P. 2008. Sonochemical method for producing titanium metal powder. Ultrasonics Sonochemistry. 15: 684-688. Latt, K.K. & Kobayashi, T. 2007. TiO2 nanosized powders controlling by ultrasound solgel reaction. Ultrasonics Sonochemistry. 15: 484-491. Jitputti, J., Rattanavoravipa, T., Chuangchote, S., Pavasupree, S., Suzuki, Y. & Yoshikawa, S. 2009. Low temperature hydrothermal synthesis of monodispersed flower-like titanate nanosheets. Catalysis Communications. 10: 378-382. Das, K., Panda, S.K. & Chaudhuri, S. 2008. Solvent-controlled synthesis of TiO2 1D nanostructures: Growth mechanism and characterization. Journal of Crystal Growth. 310: 3792-3799. Wei, M., Konishi, Y., Zhou, H., Sugihara, H. & Arakawa, H. 2004. A simple method to synthesize nanowires titanium dioxide from layered titanate particles. Chemicals Physics Letters. 400: 231-234. Yuan, Z.Y. & Su, B. L. 2004. Titanium oxide nanotubes, nanofibers and nanowires. Colloids and Surfaces A: Physicochemical and Engineering Aspects. 241: 173-183. Wang, B.L., Wang, R.H. & Peng, L.M. 2003. Synthesis and characterization of K2Ti6O13 nanowires. Chemical Physics Letters. 376: 726-731. Lo, S.L. & Ou, H.H. 2007. Review of titania nanotubes synthesized via the hydrothermal treatment: fabrication, modification and application. Separation and Purification Technology. 58: 179-191. Guo, Y., Lee, N.H., Oh, H.J., Yoon, C.R., Park, K.S., Lee, W.H., Li, Y., Lee, H.G., Lee, K.S. & Kim, S.J. 2008. Preparation of titanate nanotube thin film using hydrothermal method. Thin Solid Films. 516: 8363-8371.
Proceeding Book
1156

January 24, 2009 ISBN : 978 979 19201 0 0
Distribution of Sulphide Veins in Sangon, Kulon Progo, Yogyakarta using Very Low Frequency (VLF) Electromagnetic Method
Sismanto, E. Hartantyo, Adry S. Sembiring, M. Nukman, and Arih B. Utomo
Geophysics Laboratory, Physics Department, University Gadjah Mada, Indonesia sismanto@ugm.ac.id Abstract
The measurements of resistivity and conductivity have been carried out using electromagnetic very low frequency (VLF) method at Sangon, Kulon Progo, Yogyakarta on February 23th until March 3rd, 2008. Sangon is a public gold mine area; this area is still suitable for micro scale study of mining geophysical exploration. The aim of study is to map the gold distribution associated with sulphide veins. The VLF transmitter station is NWC with 19.8 kHz. Seven transverse lines with interval about 50 m were set up on the field and measurements were made on 10 m interval on the line. The qualitative and quantitative interpretation show that the body of igneous rocks which contains sulphide veins in stockwork structure is located at (396800, 9133450) to (396850, 9133500) by southwest-northeast orientation and (5-20) m in depth. Keywords: VLF, Resistivity, Sangon public gold mine.
Introduction
Sangon in Kulon Progo, Yogyakarta Province (Fig.1) has been known as the local mining area and used traditional methods to explore and exploit the mineral resources. In the exploration, they hunt about for outcrop of sulphide or quartz veins that associate with gold ore on the surface. The next step is to dig the ground using mattock, crowbar, hammer, and other simple devices. The excavation was usually performed in vertical-ward, but in special cases they did horizontally. Generally, the direction of excavation was determined by the diggers instinct to follow the veins (Gunawan et.al, 2001). This technique is high risk, cost, time, and ineffective due to lack of systematic information on quartz veins distribution map in the mine area. For recently, there is no geophysical method to detect the gold ore directly on the field. The geophysical techniques are able to indicate and locate the traps or the associated rock of metal ore as an anomaly against to the circumstance rocks as a normal rock. The trap of metal ore deposition in Sangon is generally associated to sulphide, phyrite, calcophyrite, and galena in sulphide vein form. One of the geophysical methods that is able to seek and to locate the sulphide vein position is electromagnetic method by using very low frequency, called VLF method.
Figure 1 Topography map of survey area (on the right side) in Sangon Kulon Progo (the red square in the left map) (Google.com; Bakosurtanal, 1999). The VLF method is an electromagnetic technique of geophysical prospecting that 1
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
responds to the variations of the resistivity in the underground until a depth of a hundred meters. The idea is to use radio signs as source of the primary field and works in a range of frequencies among 10-30 kHz coming from potent military radio stations. In presence of conductive bodies, the signs emitted by the antenna (primary fields) induce secondary currents in these bodies that in turn will generate a secondary field that is superimposed to the primary one. In this way, the measure of the total field (the addition of the primary and secondary field) allows us to locate conductive bodies in the investigation area. Some filters have been published that allow making estimations of the relative variations of the resistivity in function of the measures of the magnetic field. Among the advantages of the method there is the execution speed, since it only needs some minutes to carry out a measurement and with no need of extending complex arrays. Also, when not needing equipment originator it will only be necessary a field operator. All this implies that the investigations carried out by this method are quick and cheap. Among the disadvantages, however, it is necessary to point out the shortage of investigation depth, in occasions the difficulty of finding radio stations in an appropriate direction, or the necessity of controlling the changes that can take place in the primary field. Geological Setting Sangon is located in Kokap district (south west of Yogyakarta). It is dominated by andesit porfiritic and quaternary sediment. Some andesitic rocks are brecciaed, silificated, and altered from propilitic to fillic. Several quartz veinsthatfillacrackandafaultzoneshowsgold mineralization which associates with pyrite and essentialofmetalsulphide.Inthelocalgeological structure is found various joints and breccias on andesit rock. The normal fault is oriented to northwestsoutheast and northeastsouthwest. This normal fault is also observed in Kokap (Raharjo,et.al,1995). In Kokap area, the gold mineralization is associated in sulphide quartz vein, and sometime it associates with altered shale of fillic-argillic that the distribution is controlled by stock-work vein crack. The thickness of vein is variable in the range < 1 cm 50 cm to form mineralization strip that generally has N60E N110E direction and the dip is about 70 - 80. In some places, the metal mineralization occurs in the fine quartz vein
(1 - 50 mm) to form stockwork veinlets on he altered andesit. Gold and silver ores are randomly dispersed in quartz vein by 1 - 13,8 ppm for gold (Au) and 5,4 - 63,2 ppm for silver (Ag) (Gunawan, et.al., 2001). Experimental The laws that govern the electromagnetic wave propagation are the Maxwell equations. The electromagnetic wave propagation in homogenous and isotropic medium are,
2E = iE - 2E 2H = iH - 2H
(1) (2)
where is the permeability (H/m), is the capacity (F/m), is the conductivity (mho/m), and = 2f is the angular frequency. The propagation of electromagnetic wave will induce electric charge on conductor. The conductor create a secondary field and interfere with the primary field, consequently the resultant field will polarize in elliptic shape. In the VLF data acquisition we use T-IRIS instrument. The tilt angle and ellipse value are recorded and filtered by linier filter of Karous and Hjelt, 1983 such as
z x Ia = 0 . 115 H 1 0 . 1383 H 0 + 0 . 1383 H 1 0 . 115 H 2 2 2
(3) where
is the depth (m), I a is the current
density equivalent (%),
x is the space (m), and
is the tilt angle (%). The result of tilt angle data filtering by means of equation (3) for each depth shows a pseudodepthsection of the current density equivalent. The ordinate is the measurement points, and the absis is the depth. NWC-Australia transmitter station (19.8 kHz) was used for the survey in Sangon. The lines survey was made to be right angle to the subsurface of metal sulphide vein on the surface. We measured the position by using handy Garmin GPS CX-2500 with 5 m spacing. Tilt angel was measured by T-IRIS VLF instrument in tilt mode. Tilt angel is a vertical magnetic field intensity ratio to horizontal magnetic field intensity which is expressed in percentage. The greater tilt angle, the more conductive is the source of anomaly. The subsurface metal sulphide vein is a more conductive target than the surrounding rock. The 2
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
tilt angel was corrected by topographic correction of Baker and Myers, 1980 but only line1 was performed by topographic correction due to the fact that correction effect gave not so different interpretation. The tilt data were then smoothed by weight moving average filter. Finally, we used the Karous-Hjelt, 1983 linier filter to obtain a pseudodepthsection of current density equivalent map for each line survey. It is also possible to carry out measurement of the electric field, one of which will be able to determine the resistivity of the materials of the underground. The electric field radiated by the radio station is not totally vertical on the surface of the land, since the field primary interacts with the land of finite conductivity giving place to a component of the horizontal electric field (Fig.3). Thus, the electric field forms with the vertical one a certain angle, which increases, like the horizontal component, as the resistivity of the land does. Knowing the value of Ex, generally measured in the direction of the profile, and of the primary field Hy, the a can be determined by the equation of Cagniard:
respectively. These techniques have an important application both in hydrogeology and in civil engineering, and even in studies of contamination of ground water, where this implies a modification of the a values (Hjelt et al, 1985; Wright, 1988). An important aspect when carrying out a campaign by means of the VLF method is the orientation of the antenna radio station with regard to our profile, which will be anyway perpendicular to the predominant direction of the structure to be investigated. When the parameters related to the polarization ellipse are measured (Hz and Hx), the antenna will be approximately in the direction of the expected anomaly, while in the case of measuring a the antenna will be located next to the direction of the profile. Measured Parameters in VLF When a conductive body is immersed into an electromagnetic field, it an electric current will be induced in it, that will generate, in turn, a secondary magnetic field. For each point in space this will be oriented in a certain direction and it will have certain out phase with regard to the primary field. The main parameters measured in VLF are the inclination of the bigger axis of this polarization ellipse () and the ellipticty (), this is, the relationship among their bigger and smaller axis (b/a). Mathematically it is demonstrated that is approximately identical to the part in phase (real component) of the vertical component (HZR), while the ellipticity is approximately similar to the imaginary part (component in quadrature) of the vertical component (HZI). Both parameters are measured usually in percentages of the horizontal primary field in the following way (Ogilvy and Lee. 1991):
Figure 3 VLF-EM wave propagation (Sharma, 1997)

2
Ex a = o H y 1
(4)
where a is the apparent resistivity 0 = 4 10-7 and = 2 f. The equipments of VLF designed for measuring the a, provide at the same time the out- phase between the horizontal components of the magnetic field and the electric field, so besides obtaining information of the lateral variation of the a along the profile, we can carry out estimations on the variation of this in the vertical of each measure point. On a homogeneous ground, the out phase angle has a constant value of 45, and depending on the material overburden being more or less resistant to the rest, higher or lower phases out will be obtained than 45
Result and Discussions

The measurement of VLF can be difficult to interpret and anyway they are always less intuitive than, for example, those of apparent resistivity. Although the smoothing techniques improve the representation of the anomalies, making them more readable, they still do not provide a physical parameter that can be compared directly with the results of other electric or electromagnetic methods. In the last years filters that transform the measurements of inclination of field of VLF in conductivity or in resistivity, have appeared providing very similar profiles to those obtained 3
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
with the a measure in VLF. On the other hand, it softens the data reducing the noise and only keeping the main characteristic features of the profiles. In Fig. 4 and Fig. 5 we show the result of the measurement of apparent resistivity and phase difference carried out with VLF-R. There are a highway (the red curve line), river (the blue curve lines), and some points of measurement in. The results of these maps are easier for interpretation to identify the metal ore. There is a low resistivity (9-15) ohm-m anomalous that correlates to the phase distribution map at the north part of 16 - 21. The anomaly is estimated the conductive vein in inserting resistive rock, and the vertical contact among materials of different resistivity. The phase difference more than 45shows that the low resistivity body is deeper than the phase difference less than 45. Otherwise, if the phase difference less than 45shows that the high resistivity body is deeper than the phase difference more than 45 (Telford, et.al., 1990). According to geological information, the study area is an alteration and mineralization zone that it can be found out in some location sulphides and metals ore such as Pyrite, Calcophyrite, gold and silver (Gunawan, et.al. 2001; Rahardjo, et.al, 1995). The alteration and mineralization region is a low resistivity area; it shows as an anomaly in the resistivity map at Fig.4. The maps show that some excavation wells are located in the low resistivity area. In addition, based on the geological observation via the samples, the rock in anomaly area has 50 % to 80 % of alteration scale, whereas in the other area the alteration scale is small enough by 20 % 30 % . In conclusion, it means that the low resistivity region on the map of Fig 4 (the red zone) is an alteration and mineralization zone of metal ore. However, we should not forget that the method VLF-R can't respond effectively to the deep measurement of resistivity changes in resistivity environments and so it would be necessary in this cases the direct measurement of the resistivity using instruments of VLF-EM on tilt mode, or using electric and magnetic field methods in tilt mode. The other filter that used, usually known as filter Karous-Hjelt, provides modifications in the way from the similar anomalies to those which are obtained with the application of a filter Fraser, although, the result of the filtrate is equal to the distribution of the current density responsible for the secondary magnetic field, so that the isovalues maps and
pseudo sections obtained with this filter should be interpreted as maps of current density.
Figure 4 The apparent resistivity map in gold mining survey area. The red curve line is a road, the blue curve lines are a river, and the some points in line are the measurement points.
Figure 5 The phase map of the survey area shows that the phase anomaly has correlation to the apparent resistivity anomaly The results for each line are offered in Fig. 6. The red area is current density anomaly that shows as metal-suphide vein distribution until 20 m in depth. To find out the distribution scale on the surface, the Karous-Hjelt filtered data are plotted in Fig.7 for 5, 10, 15 and 20 meter. Fig. 7 shows that the strong anomaly apparently in the depth of 15 m. The other anomaly looks like a small or low metal ore content. Based on the resistivity, phase maps and the current density maps we make an interpretation of gold metal ore vein distribution such as in Fig. 8. The interpretation of anomaly as gold metal ore vein distribution by VLF-R mode data i.e. resistivity and phase map (Fig.8A) and 4
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
VLF-Tilt mode data in vertical and horizontal current density map (Fig. 8B) are shown in the same location by yellow color area.
Conclusion
The qualitative and quantitative interpretations show that the anomaly is metal ore sulphide veins with stock work structures by southwest - northeast orientation and 5 to 15 meter in depth. The position of anomaly is about (396903; 9133478) to (396917; 9133510).
Acknowledgements
Figure 6 The Karous- Hjelt filtering of tilt measurement in the Sangon area. The read area is current density anomaly that shows as metal-suphide vein distribution until 20 m in depth. We appreciate our colleague Dr. Arief Hermanto for some minor corrections and advice to publish this paper. We are also grateful for the survey facilities from Geophysics Laboratory of Physics department, University Gadjah Mada and permission of PEMDA Kulon Progo to conduct study in Sangon.
B Figure 8. The interpretation of gold metal ore vein distribution by VLF-R mode data (A) and VLF-Tilt mode data (B). The anomaly is shown in the same location by yellow color area.
Figure. 7 The Karous- Hjelt filtering of tilt measurement in lateral mapping every 5 meter. The strong anomaly apparently in the depth of 15 m. The other anomaly look like a small or low metal ore content.
References
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
Baker, H. A. dan Myers, J. O., 1980. A Topographic correction for VLF-EM profiles based on model studies. Geoexploration, vol 18: 135-144. Bakosurtanal, 1999, Peta Rupa Bumi Indonesia Lembar 1408-213 Bagelen, Bakosurtanal, Bogor. Gunawan, Kuswandani, Fauzan, Sofyan, A., Setiawan, L., Subarna, Juju, Ariyadi, W. dan Suhendi, E., 2001, Percontohan Penambangan Emas di Kecamatan Kokap, Kabupaten Kulon Progo, Daerah Istimewa Yogyakarta, Puslitbang Tekmira, Bandung. Google.com Hjelt, S. E., Kaikkonen, P. And Pietil, R., 1985. On the interpretation of VLF resistivity measurements. Geoexplorations, 23, 171-181. Korous, M. R., and Hjelt, S. E., 1983, Linear filtering of VLF dip-angle measurements, Geophysical Prospecting, 31, 782-794.
R. D., and Lee, A. C., 1991, Interpretation of VLF-EM inphase data using current density pseudosection, Geophysical Prospecting, 39, 567-580. Rahardjo, W., Sukandarrumidi, dan Rosidi, H.M.D., 1995, Peta Geologi Lembar Yogyakarta, Jawa, Pusat Penelitian dan Pengembangan Geologi, Bandung. Sharma, P.V., 1997, Environmental and Engineering Geophysics. Cambridge University Press, United Kingdom. Telford, W. M., Geldart, L. P., Sheriff, R. E., 1990, Applied Geophysics, 2 ed., Cambridge University Press. Wright, J. L., 1988. VLF interpretation manual. EDA Instruments, Toronto, Ontario, Canada,.
Ogilvy,
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
Dynamics and Complexity of FitzHugh-Nagumo Neuronal Systems

Mada Sanjaya W. S, Muhammad Yusuf, Agus Kartono
1
Theoretical Physics Division, Department of Physics, Faculty of Mathematical and Natural Sciences, Institut Pertanian Bogor, Bogor 16680, Indonesia. e-mail : madasws@gmail.com Abstract
System of signals propagation from one neuron to another represent event of very complex electrochemical mechanism This work adresses the dynamics and complexity of neuron mathematical models. The aim is first the understanding of the biological meaning of existing mathematical systems concerning neurons such as Hodgkin-Huxley or FitzHugh-Nagumo models. The local stability and the numerical asymptotic analysis of FitzHugh-Nagumo model are then developed in order to comprehend the dynamic evolution of a single FitzHugh-Nagumo neuron. This examination naturally comes to the study of neuron networks. The analysis of these networks uses the synchronization theory via connections between neurons and can give rise to emergent properties and selforganization. Our result leads to a classical law which describes many self-organized complex systems like earthquakes, linguistic or urban systems. This has been performed using numerical tools. Keywords: Hodgkin-Huxley models, FitzHugh-Nagumo, action potentials, synchronization, chaos.
Introduction
Understanding the mechanisms of the propagation of the nerve activity is one of the fundamental problems in biophysics. The first detailed quantitative measurements of the ionic currents were performed by Hodgkin and Huxley in the early 50-s (Hodgkin & Huxley, 1952). Using the voltage clamp technique, they were able to measure the kinetics of Na+ and K+ currents in the squid giant axon. This led them to a set of differential equations which describe the dynamics of the action potential. Furthermore, by combining these equations with the cable equation for spreading of current in the axon they were able to calculate the shape and velocity of the propagating action potentials (Hodgkin & Huxley, 1952). The predictions of their model turned out to be in a remarkably good agreement with the experimental observations. Hodgkin-Huxley model for the action potential of a space clamped squid axon is de_ned by the four dimensional vector field. with variables (v; m; n; h) that represent membrane potential, activation of a sodium current, activation of a potassium current, and inactivation of the sodium current and a parameter I that represents injected current into the space-clamped axon. Although there are improved models the Hodgkin-Huxley model remains the paradigm for conductance-based models of neural system. FitzHugh was the first investigator to apply qualitative phase-plane methods to understanding the Hodgkin-Huxley model. To make headway in gaining analytic insight, FitzHugh first considered the variables
that change most rapidly, viewing all others as slowly varying parameters of the system. In this way he derived a reduced two-dimensional system that could be viewed as a phase plane. From the Hodgkin-Huxley equations FitzHugh noticed that the variables V and m change more rapidly than h and n, at least during certain time intervals. By arbitrarily setting h and n to be constant we can isolate a set of two equations which describe a two-dimensional (V; m) phase plane. The elegance of applying phase plane methods and reduced systems of equations to this rather complicated problem should not be underestimated. Determining the dynamical behavior of an ensemble of coupled neurons is an important problem in computational neuroscience. Commonly used models for the study of individual neuron which display spiking behavior (FitzHugh, 1961; Nagumo, 1962; Hodgkin, 1952). From the very beginning of the research in the field of computational neuroscience, people deal with single neuron and its behavior. Present trends of research include investigation of the behavior of neurons considered in a network and their way to fire synchronously. It is assumed that the activities in the brain are synchronous and underlying interests for synchronization of nonlinear oscillators in physical and biological systems range from novel communication strategies to understand how large and small neural assemblies efficiently and sensitively achieve desired functional goal. In recent years, there has been tremendous interest for the study of the synchronization of chaotic systems. The phenomenon of synchronism gives rise to 1
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
different dynamical behaviors such as chaotic synchronization etc (Belykh, I & Shilnikov, A., 2008). In (Mishra, 2006), nonlinear dynamical analysis on single and coupled modified FitzHugh-Nagumo model under steady current stimulation is carried out. Also the effect of parameter variation on its behavior is investigated. Mathematical Models Basics Models of Neuron To describe the evolution of the membrane potential V in the squid giant axon (Hodgkin, A.L & Huxley,A.F., 1952) developed the following conductance-based model:
dv 1 = g Na (v)m 3 h(v v Na ) + g K n 4 (v)(v v k ) + g L (v v L ) + I eks dt C
Figure 1 Dinamical Systems in Hodgkin_huxley models at I = 0 ; (a) Action potentials at t = 5 ms, (b) variable m, h and n at t = 5 ms. FitzHugh-Nagumo Models of Single Neuron In 1961 FitzHugh proposed to demonstrate that the Hodgkin-Huxley model belongs to a more general class of systems that exhibit the properties of excitability and oscillations. As a fundamental prototype, the van der Pol oscillator was an example of this class, and FitzHugh therefore used it (after suitable modification). A similar approach was (1) developed independently by Nagumo in 1962. FitzHugh proposed the following equations:
dn = n (v)(1 n) n (v)n dt dm = m (v)(1 m) m (v) m dt dh = h (v)(1 h) h (v)h dt
(1)
The peak conductances g, reversal potentials E and rate coefficients and of the individual currents and gating variables m, n and h are given by :
dv 1 = v v3 w + I dt 3 dw = (v + a bw ) dt
(2)
m (v ) = 0.1(v + 40)(1 e ( v + 40 ) / 10 ) 1 ,
m (v ) = 4e ( v + 65 ) / 18 ,
h (v) = 0.07e ( v + 65) / 20 , h (v) = (e ( v +35) / 10 + 1) 1 n (v) = 0.01(v + 55)(1 e ( v +55) / 10 ) 1 , n (v) = 0.125e ( v + 65) / 80
For I = 0, the membrane potential settles into a resting potential of about V = 65 mV. And with the gNa = 120, gK = 36, gL = 0.3 mmho cm-2, vNa = 50 mV, vK = -77 mV and mV. vL = -54.4
In these equations the variable v represents the excitability of the system and could be identified with voltage (membrane potential in the axon); w is a recovery variable, representing combined forces that tend to return the state of the axonal membrane to rest. Finally I is the applied stimulus that leads to excitation (such as input current), or rectangular pulses. Local Dynamics and Bifurcation Let us consider the FitzHugh-Nagumo system (2), Equilibria are given by the following system :
dv / dt = 0, dw / dt = 0
(3)
The nature of this equilibria is given by the study of the eigenvalues of the jacobian matrix J of this system :
1 v 2 J =
(a) (b)
1 (4) b
We obtain the following polynomial equation :

2 + ( 1 + v 2 + b ) + ( b + bv 2 + ) = 0
(5) 2
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
Numerical Analysis Table 1 Numerical Analysis of Stability Equilibrium Point in FitzHugh-Nagumo Systems No 1 2 3 4 5 6 7 8 9 10 Variation Ieks 0.00 0.32 0.33 0.50 1.25 1.42 1.43 1.45 1.50 2.00 Equilibrium Point -1.1994,-0.6243 -0.9769,-0.3461 -0.9685,-0.3357 -0.1311,-0.8048 1.8810, 0.8048 2.0857, 0.9685 0.9769, 2.0961 0.9933, 2.1166 2.1656, 1.0325 1.3341, 2.5426 Eigen value -0.2513 0.211900 i -0.009176 0.2774 i -0.001045 0.2757 i +0.1441 0.191500 i +0.1441 0.191500 i -0.001045 0.2757 i -0.009176 0.2774 i -0.02532 0.28020 i -0.06501 0.282800i -0.6412, -0.2026 Stability Spiral sink Spiral sink Limit cycle Spiral source Spiral source Spiral sink Spiral sink Spiral sink Spiral sink Stable Node
Obviously, this figures show transitions from stable to periodic or quasiperiodic, then complex and finaly chaotic behaviour.
(a)
(b) (c) Figure 2. Phases portrait for; (a) I = 0, (b) I = 0.33, (c) I = 1.45, (d) I = 2
(d)
(e)
(f) (g) Figure 3. Time Series for ; (a) I = 0, (b) I = 0.33, (c) I = 1.45, (d) I = 2
(h)
Synchronization Generalities Synchronization is a phenomenon characteristic of many processes in natural systems and non linear science. It has remained an objective of intensive research and is today
considered as one of the basic nonlinear phenomena studied in mathematics, physics, engineering or life science. This word has a greek root, syn = common and chronos = time, which means to share common time or to occur at the same time, that is correlation or agreement in time of different processes.Thus, synchronization of two 3
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
dynamical systems generally means that one system somehow traces the motion of another. Chaotic oscillators are found in many dynamical systems of various origins. The behavior of such systems is characterized by instability and, as a result, limited predictability in time. Roughly speaking, a system is chaotic if it is deterministic, if it has a long-term aperiodic behavior, and if it exhibits sensitive dependence on initial conditions on a closed invariant set. Consequently, for a chaotic system, trajectories starting arbitrarily close to each other diverge exponentially with time, and quickly become uncorrelated. It follows that two identical chaotic systems cannot synchronize. This means that they cannot produce identical chaotic signals, unless they are initialized at exactly the same point, which is in general physically impossible. Thus, at first sight, synchronization of chaotic systems seems to be rather surprising because one may intuitively expect that the sensitive dependence on initial conditions would lead to an immediate breakdown of any synchronization of coupled chaotic systems, which led to the belief that chaos is uncontrollable and thus unusable. Despite this, in the last decades, the search for synchronization has moved to chaotic systems. A lot of research has been carried out and, as a result, showed that two chaotic systems could be synchronized by coupling them : synchronization of chaos is actual and chaos could then be exploited. Ever since, many researchers have discussed the theory, the design or applications of synchronized motion in coupled chaotic systems. A broad variety of applications have emerged, for example to increase the power of lasers, to synchronize the output of electronic circuits, to control oscillations in chemical reactions or to encode electronic messages for secure communications (Aziz-Alaoui, MA., 2006). Modified FitzHugh-Nagumo Models The modified FitzHugh-Nagumo equations are a set of two coupled differential equations which exhibit the qualitative behavior observed in neurons, viz quiescence, excitability and periodicity (Mishra D et al, 2006). The system can be represented as
dv 1 = v v 3 w + I (t ) dt 3 dw = ( v + a bw ) dt
Where
I (t ) = ( A / ) cos(t ) (5)
A represents the magnitude of the stimulus and refers to the frequency of given stimulus.
(a)
(b)
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
(c)
(d)
(e) (f) Figure 4. FitzHugh-Nagumo Models at A = 0.7 ; (a) Time series at = 0.4 (b) Phases portrait at = 0.4 , (c) Time series at = 0.8145 (d) Phases portrait at = 0.8145 , (e) Time series at = 0.93, (f) Phases portrait at = 0.93
The coupling function h is given by :

n
The stimulus frequency is varied while keeping the magnitude at a fixed value of A = 0.71, since at this particular value of A, modified FitzHugh-Nagumo neuron model gives periodic spiking. Simulation results at different stimulus frequencies are shown in Figure 4. It is observed that with the variation in stimulus frequency, the neuron shows complex chaotic behavior. Hence the stimulus frequency can be considered as a significant parameter that affects the behavior of neuron. Coupling FitzHugh-Nagumo neurons Let us consider a network composed by n FitzHugh-Nagumo neurons. These neurons are coupled by there first variable v. This network can be modeled by the system (Lange, E.et al, 2005; Belykh, I & Shilnikov A., 2008):
h(vi , v j ) = (vi Vs ) g s cij (v j ) (7)

j =1
in which the synaptic coupling is modeled by a sigmoid function with a threshold :
(v j ) =
1 (8) 1 + exp( (v j s ))
s is the threshold reached by every action potential for a neuron. Neurons are supposed to be identical and the synapses are fast and instantaneous. The parameter gs corresponds to the synaptic coupling strength. The synapse is exitatory, that is why the reversal potential Vs must be larger than xi(t) for all i and all t. Coupling two FitzHugh-Nagumo neurons The first step here is to adapt the previous method to two FitzHugh-Nagumo neurons. In this case, we use a bidirectional connection since 5
dv i 1 3 = v i v i w i + I (t ) h ( v i , v j ) 3 dt (6) dw i = ( v i + a bw i ) dt
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
the number of inputs (that is one) has to be the same for all neurons to have a synchronous solution. v1
dv1 1 3 1 = v1 v1 w1 + I (v1 Vs ) g s dt 3 1 + exp( (v 2 s )) dw1 = (v1 + a bw1 ) dt dv 2 1 3 1 = v 2 v 2 w2 + I (v 2 Vs ) g s dt 3 1 + exp( (v1 s )) dw2 = (v 2 + a bw2 ) dt
(9)
v2
0 1 = 1 0
The following system represents our two FitzHugh-Nagumo neurons bidirectionally coupled :
Parameters are fixed as follows : a = 0.7, b = 0.8, = 0.08, Vs = 1, s = -0.25, =1, A = 0.71 and
= 0.95.
(a)
(b)
(c) (d) Figure 5. v2 according to v1 for the coupling strength; (a) gs = 0.1, (b) gs = 0.125, (c) gs = 0.1255 , dan (d) gs = 0.15 We observe that the synchronization numerically appears for a coupling strength gs 0.15. This phenomenon is also observed for the variables w, for the same numerical values of parameters. Coupling three FitzHugh-Nagumo neurons In the case of three neurons, we also use bidirectional connections since the number of inputs (that are two) has to be the same for all neurons to have a synchronous solution.
1 0 0 C 3 = 0 1 0 0 0 1
The following system represents our three FitzHugh-Nagumo neurons bidirectionally coupled :
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
dv1 1 1 1 3 ) = v1 v1 w1 + I (v1 Vs )g s ( + 1 + exp((v2 s )) 1 + exp((v3 s )) 3 dt dw1 = (v1 + a bw1 ) dt dv2 1 3 1 1 ) = v2 v2 w2 + I (v2 Vs )g s ( + 3 1 + exp((v1 s )) 1 + exp((v3 s )) dt dw2 = (v2 + a bw2 ) dt dv3 1 3 1 1 = v3 v3 w3 + I (v3 Vs )g s ( + ) 3 1 + exp((v1 s )) 1 + exp((v2 s )) dt dw3 = (v3 + a bw3 ) dt
(10)
(a)
(b)
(c) (d) Figure 6. v2 according to v1 for the coupling strength; (a) gs = 0.01, (b) gs = 0.062, (c) gs = 0.064 , dan (d) gs = 0.069. Synchronization numerically appears for a coupling strength gs = 0.069. This phenomenon is also observed for the variables w, for the same numerical values of parameters. Coupling four FitzHugh-Nagumo neurons There are different possible ways for coupling four neurons with the same number of input from other neurons. Here, we decided to connect each neuron to all the others.
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
(a)
(b)
(c) (d) Figure 7. v2 according to v1 for the coupling strength; (a) gs = 0.015, (b) gs = 0.018, (c) gs = 0.019 , dan (d) gs = 0.0195. In this paper, the characteristics of two dimensional modified FitzHugh-Nagumo neuron Synchronization numerically appears for a model is studied. Dynamical behavior of the coupling strength gs = 0.0195. This phenomenon modified FitzHugh-Nagumo system under is also observed for the variables w, for the same external electrical stimulation is presented and it numerical values of parameters. is verified that the introduction of periodic This is an emergent property which comes stimulation modifies the dynamics of biological from the collective dynamics of n neurons. system by presenting the dynamical behavior for Moreover, as given in figure, as the number of the modified FitzHugh-Nagumo system under neurons n goes larger, the synchronization external electrical stimulation. threshold g gets smaller. Synchronization phenomenon is exhibited in a neuron network, the transmission of Conclusion information between these neurons is optimal and the network can develop consciousness. This Since the birth of Hodgkin-Huxley equation is an emergent property which comes from the in neurophysiological modelling, researchers collective dynamics of n neurons. Moreover as have made considerable efort to try to analyze the number of neurons n goes larger, the this system in diferent way. Extensive eforts synchronization threshold g gets smaller. have been made to discover chaos in many Consciousness is more important when the physical and biological systems including neural number of neurons is larger. This phenomenon systems. We have investigated bifurcations which can describe many self-organized observed in the FitzHugh-Nagumo neuron model complex systems, like earthquakes, linguistic, as a simplified models. By calculating urban systems. we have bifurcations with changing I identified the parameter regions in which the FitzHugh-Nagumo neuron exhibits properties of excitability and oscillation as an existence of a couple of stable equilibrium points and one stable limit cycle.
ext
References
[1]. Aziz-Alaoui, MA. (2006), Complex emergent properties and chaos (De) 8
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
synchronization, Emergent Properties in Natural and Artificial Dynamical Systems. Heidelberg : Springer p129-147. Belykh, I., Shilnikov A. (2008), When Weak Inhibition Synchronizes Strongly Desynchronizing Networks of Bursting Neurons. PRL 101, 078102. Edelstein-Keshet, L., (1988), Mathematical Models in Biology. Random House, New York. FitzHugh, R. (1961), Impulses and Physiological state in theoretical models of nerve membrane. Biophysics Journal.,I,445-466. Hodgkin, A. L., and Huxley, A. F. (1952), A quantitative description of membrane current and its application to conduction and excitation in nerve. J. Physiol.,117.500-544. Lange, E., Belykh, I.,Hasler, M. (2005), Synchronization of Bursting Neurons: What matters in the Network Topology. PRL 94, 188101. Mishra D, Yadav A, Ray S, Kalra PK. (2006), Controlling Synchronization of Modified FitzHugh-Nagumo Neurons Under External Electrical Stimulation. NeuroQuantology Issue 1 Page 50-67. Muratov, C. B. (2008), A quantitative approximation scheme for the traveling wave solutions in the Hodgkin-Huxley model. arXiv.org/abs/nlin/0209053v1.
[2].
[3]. [4].
[5].
[6].
[7].
[8].
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
Floating-point Unit Implementation in Altera FLEX10K FPGA using VHDL

Agfianto Eko Putra1, Isnan Nur Rifai2
Electronics and Instrumentation, Physics Department Faculty of Mathematics & Natural Sciences Gadjah Mada University, Yogyakarta Indonesia 55281 Abstract
Floating-point systems were developed to provide high resolution over a large dynamic range. Floating-point systems can often provide a solution when fixed-point systems, with their limited dynamic range, fail. Floating-point systems, however, bring a speed and complexity penalty. Most microprocessor floating-point systems comply with the published single- or double-precision IEEE floating-point standard; while in FPGA-based systems often employ custom formats. In this research, a 16-bit floating-point unit, which has addition, subtraction, multiplication and division operator, and based on floating-point system, has been implemented in Altera FLEX10K FPGA using VHDL (VHSIC Hardware Description Language). The design used structural and behavior model implementation to know which the best design is. The best results are structural model using ROM which require 438 logic elements with 7.99 MFLOPS (million floating-point operations per second), and the behavior model using ROM which require 526 logic elements with 7.51 MFLOPS. The floating-point unit has been designed for normal floating-point operation, thus it can not detect overflow or underflow conditions. Keywords: FPGA, VHDL, Floating Point
1,2
Introduction
Floating Point Unit is a numerical computing unit which is important in the computer system. Almost algorithms that use real numbers, depends on this operation, because the floating point unit is a system of numerical interpretation series of digits or bits that are presented as real number [1]. In other words, floating point unit is serving a fraction of the computer. Many algorithms rely on floating point arithmetic for the dynamic range of representations and require millions of calculations per second. The programmers are freely to code numbers in various applications [2]. This shows the number of computing with floating point played an important role in various applications in the fields of science, engineering and industry, especially in the meteorology, mechanical design and simulation. The ability to perform floating point operations is an essential measurement to the performance of computer applications [2]. Floating point arithmetic operations can be implemented into the FPGA by using the schematic method, but the algorithm used in the floating point unit is too complex, so it requires a very high accuracy in making the optimum schematic model and requires analysis when errors occur in a complex schematic. So this method is considered less efficient. However, using the language VHDL description, the problems are very capable. Floating Point Unit
Floating point unit is part of the computer system used to process the number of operating floating point arithmetic itself [3]. Arithmetic operations which conducted consists of 5 operations namely, addition, subtraction, multiplication, division and square root. However, in this research, will be implemented only 4 out of 5 arithmetic operations, without square root operation. Floating Point Representation Standard 754 Format using IEEE
Generally, the form of floating point is presented in d.ddd ... d E format. Which d.ddd... is called as significant and have p digits precision of floating point numbers and is the base, equal 10 for decimal, 2 for binary or 6 for hexadecimal numbers. For example, = 10 and p = 3, the 0.85 number is presented as 8.50 10-1, and if = 2 and p = 24, then the value 0.85 approaching 1.101100110 2-1. In 1985, the Institute of Electrical Engineering (IEEE) standardizes the floating point format called IEEE Standard 754. This standard aims to facilitate the exchange of data floating point between the computers and provide a different format floating point for the hardware designers [3]. IEEE standards specify the extended 754 floating point format using 80 bits and the basic format consists of a double precision that use 64 bit and single precision that use 32 bits [3].
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
Comparison of these two types of precision is
shown in Table 2.1.
Table 2.1 The Comparison of IEEE 754 single precision and double precision Format Single Double VHDL VHDL is the brevity of VHSIC Hardware Description Language. VHSIC itself is brevity of Very High Speed Integrated Circuit [4]. In the preparation of a digital electronics design using VHDL, the model design needs to be used, whether structural or behavioral models (behavior). The main difference between two models is the division of functions. In the structural model, each of the different functions will be grouped into small modules (declared as a component), which will be linked again using port map statement [5]. Floating Point Unit Design Precision 24 53 Emax +127 +1023 Emin -126 -1022 Bias 127 1023 Exponent width 8 11 Format width 32 64
Floating point unit is created to run floating point arithmetic operations, which is addition, subtraction, multiplication and division. The format used in these operations is 16 bit floating point, which consists of 1 signed bit, 6 exponent bits and 9 fraction bits. Basically, this 16 bits floating point format is the same with the others, only because the number of bits for the exponent is only 6, then the bias that must be added, which exponent is able to serve with either positive or negative number, is 31. To be more easily in the design and error analysis, this unit was divided into four blocks and each block was designed to run one of the arithmetic operations. These four blocks is the adder, subtrahend, multiplier and divider block. The relation of these blocks is shown in Figure 1.
Figure 1 Floating Point Unit block diagram Floating Point Addition and Subtraction Algorithm The algorithms of addition and subtraction floating-point number have similar, then, in the hardware design, these two blocks are implemented together (as one unit). Although for the operations these two blocks have the function of its own. To explain the algorithm, we use, for example, fp1 and fp2 as a two input floating point numbers each of which has s1, e1, f1 and s2, e2, f2 (each as sign, exponent and fraction). Meanwhile, fp is the output or results of operations that have been run. The flow diagram for the addition and subtraction floating point operations is shown in Figure 2. 2
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
Explanation of the flow diagram in Figure 2 as follows: 1. Both input and the operation which will be run (addition or division) is read, when the operation is subtraction, then the sign in fp2 (s2) is changed to not s2 (s2=s2'); 2. Both exponent e1 and e2 are compared using 6 bits subtraction, when e2 is greater than e1, then both inputs, fp1 and fp2, are swapped, fp1 become fp2 and vice versa; 3. The fraction of the fp2 (f2) is shifted to the right as many as of the result of absolute subtraction of e1 and e2 to get the same exponent for both numbers; 4. Signs, s1 and s2, is XORed to know whether the two fractions are added or subtracted (op = s1 XOR s2); 5. When op is equal 1, then both fractions are subtracted, otherwise, when op is equal 0, both fractions are added; 6. The results from addition or subtraction will pass through the Leading One Detector (LOD) to know whether the first bit is equal to 1, if not, then the value will be shifted to the left until the first bit is equal to 1; 7. Exponent value (e) is obtained by adding or subtracting e1 and e2 (depend on the operation), and the shift results from the LOD; 8. Sign value (s) is obtained from carry which is obtained from fraction subtraction, but if e1 e2, then s = s1. Using algorithm that have been described previously, we obtained the required design of the micro-architecture and its modules. The detail, the design for addition and subtraction, is shown in Figure 3. Floating Point Multiplication Algorithm Floating point multiplication algorithm is not too complex compare to addition or subtraction algorithm. Multiplication operation does not require exponential equation and shift fraction. For both floating point input, for example fp1 and fp2, each of which has a fraction f1 and f2, the flow diagram for the floating point multiplication is shown in Figure 4. Explanation of the flow diagram in Figure 4 as follows: 1. After both of floating point input is read, then both fraction, f1 and f2, is multiplied; 2. The multiplication result will pass through the Leading One Detector
(LOD), to set the output fraction. The implicit bit must be equal one; 3. The output exponential value is obtained by adding the first and second exponential. The shift result which is obtained from LOD also affected the exponential arrangement. 4. The sign value is obtained by XORing both of the signed inputs. Using multiplication algorithm that has been described, we obtained the required design of the micro-architecture and its modules. In detail, the design for addition and subtraction is shown in Figure 5. Floating Point Division Floating Point Division operation algorithm and the micro-architecture design are almost the same as the multiplication floating point operation, but the differences is in the handling of the value of a fraction. According to the operations, for floating point multiplication operation, the fraction is generated by multiplying both floating point inputs. While for the floating point division operation, the fraction is generated by dividing both floating point inputs.
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
Figure 2 Flow-diagram for Floating Point Addition and Subtraction operations
Figure 3 Floating Point Addition and Subtraction Micro-architecture Design
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
b.
c.
d.
Figure 4 Flow-diagram for Floating Point Multiplication operation
value of the second exponent is greater than the first, due to the operation of this addition and subtraction, the input with larger exponent is considered as the first entries. Right Shifter Module This module is used to shift the second fraction input to the right. The number of shift is obtained from the exponent difference module. Add or Sub module This module is used to add or subtract both fraction inputs. This depends on the XOR operating results of both signs. Both fraction will be added if the XOR result is equal to 0, and will be subtracted if the result is equal to 1. Leading One Detector (LOD) Module LOD module is used to set the first bit value always 1. Because this first bit, later, used as an implicit bit in the results of floating point operations.
2.
Figure 5 Floating Point Multiplication Microarchitecture Design
3.
Experimental
Hardware Implementation using VHDL To design VHDL behavior model, we can implement directly the algorithm, but for structural design of VHDL model, these algorithms should be divided into several small modules based on the difference function. Modules that are used for the floating point unit with structural model are as follows: 1. Addition and Subtraction Operation a. Exponent Difference Module The function of exponent difference module is to compare the exponent value of both floating point number inputs, this comparison is required for swapping of both inputs, if the
Multiplication Operation a. Module LODmul The function of this module is almost the same as the LOD in the addition and subtraction operation. Namely to set the first bit that will be used as implicit bit always equal to 1. Division Operation a. Kurang Module This module is used to perform subtraction functions that will be used for the division operations of the bagi22 module. b. Bagi22 Module This module is used to repeat the Kurang Modules process 11 times, this is due to the expected results of the division is 11 bits, while each of the subtraction process will produce the output of 1 bit. This bit will enter into a serial output as a result of division.

Functional Simulation Result Before the design is implemented in FPGA, we have used functional simulation to validate the
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
design, and functional simulation for each module was running properly. FPGA Implementation of the FLEX 10K-10 After compiling (implementation), it could be known the number of resources (Logic Element and Logic Array) that are needed. The design requires 883 Logic Element (LE) and 0 embedded arrays (EA) for the structural model and 949 LE and 0 EA for the behavior model, using FPGA Altera EPF10K20TC144-3 which has 1152 LE and 6 EA. Based on the number of resources that are needed, clearly is impossible to implement floating point units into the FPGA FLEX 10K10 without modification, because this chip has only 576 LE, and 3 EA. So it is necessary to make additional program to divert some of the LE functions to EA, for this case, we can use ROM. By using the ROM, the amount of resources needed to be 438 and 2 EA for structural model and 526 LE and 2 EA for behavior model. Speed Floating Point Unit Operational In MFLOPS The ability or the performance of computer systems, those running arithmetic floating point instructions, is generally measured in MFLOPS (million floating point operations per second). To know the speed of floating point unit operations that have been implemented into the Altera FPGA FLEX10K10, we used Timing Simulation. But first, we have to set the clock frequency according to the chip clock frequency capability. In this case, the clock frequency of Altera FPGA FLEX10K10 chip is 10 MHz, so the period for floating point unit operations is 100ns. Based on the timing simulation we have been done, the speed for each unit are as follows: Structural Model Floating Point Unit 1. With ROM a. Addition : 134.4 ns b. Subtraction : 116.7 ns c. Multiplication : 117.5 ns d. Division : 120.3 ns e. Average Speed :122.2 ns From this result, it can be known (in MFLOPS):
e. Average Speed : 125.1 ns From this result, it can be known (in MFLOPS):
Behavior Model Floating Point Unit 1. With ROM : 7.25 MFLOPS 2. Without ROM : 7.51 MFLOPS The complete result is shown in Table 2 and 3. Table 2 Complete analysis of Altera FPGA FLEX10K without ROM Structural Behavior Parameter Model Model Resource Speed device 883 LE 0 EA 8.18 MFLOPS EPF10K20 949 LE 0 EA 7.25 MFLOPS EPF10K20
Table 3 Complete analysis of Altera FPGA FLEX10K with ROM Structural Behavior Parameter Model Model Resource Speed device 438 LE 2 EA 7.99 MFLOPS EPF10K20 526 LE 2 EA 7.51 MFLOPS EPF10K20
Conclusions
A 16-bit floating-point unit, which has addition, subtraction, multiplication and division operator, and based on floating-point system, has been implemented in Altera FLEX10K FPGA using VHDL (VHSIC Hardware Description Language). The design used structural and behavior model implementation to know which the best design is. The best results are structural model using ROM which require 438 logic elements with 7.99 MFLOPS (million floating-point operations per second), and the behavior model using ROM which require 526 logic elements with 7.51 MFLOPS. The floating-point unit has been designed for normal floating-point operation, thus it can not detect overflow or underflow conditions.
2.
Using ROM a. Addition b. Subtraction c. Multiplication d. Division
: 141.3 ns : 120.0 ns : 119.5 ns : 119.6 ns
Acknowledgements
This research is supported by Altera FPGA Development Board that belongs to Electronics and 6
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
Instrumentation Laboratory, Physics Department, Faculty of Mathematics and Natural Sciences, Gadjah Mada University Yogyakarta.
References
[1] Hok, H.C., 2003, Automatic Synthesis and Optimization of Floating Point Hardware, Thesis of the Chinese University of Hongkong. [2] Shirazi, N., 1995, Quantitative Analysis of Floating Point Arithmetic on FPGA Based Custom Computing Machines, IEEE Symposium on FPGAs for Custom Computing Machines, Napa Valley, California. [3] Goldberg, D., 1991, What Every Computer Scientist Should Know About FloatingPoint Arithmetic, ACM Computing Surveys, 23(1):5-48 [4] Ashenden, P.J., 1990, the VHDL Cookbook 1st Edition, Dept. Computer Science, University of Adelaide, South Australia. [5] Hwang, E. O., 2004, Digital Logic and Microprocessor Design with VHDL, Brooks/Cole, California
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
Theoretical Study on the Structure of Br- in Water Cluster by using Ab Initio-PCM Method
M.A. Putri1, R. Lubis2, A. Napitupulu3 and H. Sudrajat1
Department of Chemistry, Faculty of Mathematics and Natural Sciences, Gadjah Mada University, Sekip Utara, Yogyakarta 55281 2 Physics Education Department, Faculty of Mathematics and Natural Sciences, State University of Yogyakarta, Karangmalang, Depok, Yogyakarta 55283 3 Department of Physical Engineering, Faculty of Engineering Gadjah Mada University, Jl. Grafika No.2 Yogyakarta 55281 Corresponding author: ita_kz@yahoo.com Abstract
For Br-(H2O)4 cluster, it has long believed that the C4 symmetry structure is most stable isomer at low temperature. However, calculated vibrational spectra values using ab initio method disagree with experimental values. Therefore, we calculated the structure of Br-(H2O)4 using extensive ab initio calculation including polarizable continuum model (PCM) with BONDI solvation model. We used CEP121G basis set for Br- with augmented diffuse spd and 6-31++G(2d,2p) basis set for water molecules at second order Mller-Plesset perturbation (MP2) level of theory. Hence, we found most stable isomer for Br-(H2O)4 at low temperature. The isomer has C2 symmetry. The C4 symmetry structure was transition structure. These findings are useful for solving the confusion between the experiment and ab initio results. Keywords: Structure, Br-(H2O)4, Ab initio-PCM method
1
Introduction
The hydration of ions is a subject of interest in solute and biochemistry [1]. An iodide anion is of particular interest in hydration, since it has large polarizability and large ionic radius [2]. The various structures of Br-(H2O)n cluster (where n = 16) have been examined using extensive ab initio calculations [2]. Also, experimental vibrational spectra was recorded at low temperature [3,4]. The ab initio results are in excellent agreement with the low temperature experimental data except for n = 4 [2]. In the case of n = 4, it has long believed that the C4 symmetry structure is most stable isomer [2,5]. However, the calculated vibrational spectra values disagree with the experimental values. Therefore, we calculated the structure of Br-(H2O)4 using extensive ab initio calculation. Hence, we found other more stable isomer. The purpose of this paper is to propose the most stable isomer at low temperature.
perturbation (MP2) level of theory. We optimized the structure near the C4 symmetry structure. We also performed vibrational analysis for all of the isomers having the optimized structures to confirm that all of the vibrational frequencies are real. The core electrons were frozen. Nonspecific solvent effects have been estimated using the BONDI/polarizable continuum model (BOND/PCM). We have used for all calculations the BONDI radii obtained by single point calculations at the Mller-Plesset perturbation (MP2) level of theory using optimized structures in gas phase. In the BOND/PCM approach the solute is placed in a polarizable cavity formed of spheres centered in the atomic groups. Inside the cavity, the dielectric constant is the same in a vacuum, and outside it takes the solvent value ( = 78.4 for water).
Results and discussion

The most two stable isomers are the structures (1) and (2) in Figure 1. The structure (1) has C4 symmetry. The structure (2) has C2 symmetry. The isomer (1) is transition state structure with one imaginary frequency. The isomer (2) is local minima. The isomer (2) is more stable than the isomer (1) by 0.2 kcal/mol. Table 1 shows the OH harmonic frequencies for Br-(H2O)4. The values in the isomer 1
Computational Methods
We used the Gaussian 98 [6] and Gaussian 03 [7] programs, operating on the SX-7 and VPP5000 computers. We used CEP-121G basis set for Br- with augmented diffuse spd and 6-31++G(2d,2p) basis set for water molecules at second order Mller-Plesset
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
(1) disagree with the experimental values. The values in the isomer (2) agree with the experimental values. The results suggest that the isomer (2) is most stable isomer at low temperature. The C4 structures for X-(H2O)4 (X = F, Cl) are local minima. When X is heavier element, the X-(H2O) distance for the C4 structures is longer: F- (1.714 ) [8]; Cl- (2.386 ) [9];. Therefore, it is not surprising that the C4 structure for Br-(H2O)4 is transition structure. We will calculate the structures of Br-(H2O)n using higher level ab initio molecular orbital methods.
Conclusion
We found the most stable isomer for Br-(H2O)4 at low temperature. The isomer has C2 symmetry. It has long believed that the C4 symmetry structure is most stable isomer at low temperature. However, this structure was transition structure. These findings are useful for solving the confusion between the experiment and ab initio results.
Acknowledgements
We would like to thank STMIK AMIKOM Yogyakarta for providing us with support and computer time. We acknowledge the interesting discussion of the results with Muhammad Saad Saefullah, M.Kom., M.Sc.
References
[1] Marcus, Y., Ion Solvation, Wiley-Interscience, New York, 1985. [2] Lee, H.M. & Kim, K.S. 2001. Structures and Spectra of BromideWater Clusters Br-(H2O)n=16: An Ab Initio Study. J. Chem. Phys. 114: 44614471. [3] Robertson, W.H. & Johnson, M.A. 2003. Molecular Aspects of Halide Ion Hydration: The Cluster Approach. Annu. Rev. Phys. Chem. 54: 173-213. [4] Weber, J.M. Kelley, J.A. Robertson, W.H. & Johnson, M.A. 2001. Hydration of a Structured Excess Charge Distribution: Infrared Spectroscopy of the O2-(H2O)n, (1 < N < 5) Clusters. J. Chem. Phys. 114: 2698-2706. [5] Vila, F.D. & Jordan, K.D. 2001. Theoretical Study of the Dipole-Bound Excited States of Br(H2O)4. J. Phys. Chem. A. 106: 1391 1397. [6] Frisch, M. J. Trucks, G. W. Schlegel, H. B. Scuseria, G. E. Robb, M. A. Cheeseman, J. R. Irzewski, V. G. Montgomery, J. A. Jr. Stratmann, R. E. Burant, J. C. Dapprich, S. Millam, J.M. Daniels, A. D. Kudin, K. N. Strain, M. C. Farkas,
O. Tomasi, J. Barone, V. Cossi, M. Cammi, R. Mennucci, B. Pomelli, C. Adamo, C. Clifford, S. Ochterski, J. Petersson, G.A. Ayala, P. Y. Cui, Q. Morokuma, K. Malick, D. K. Rabuck, A. D. Raghavachari, K. Foresman, J. B. Cioslowski, J. Ortiz, J.V. Baboul, A. G. Stefanov, B. B. Liu, G. Liashenko, A. Piskorz, P. Komaromi, I. Gomperts, R. Martin, R. L. Fox, D. J. Keith, T. Al-Laham, M. A. Peng, C. Y. Nanayakkara, A. Challacombe, M. Gill, P. M. W. Johnson, B. Chen, W. Wong, M. W. Andres, J. L. Gonzalez, C. Head-Gordon, M. Replogle, E. S. Pople, J. A. 1998. Gaussian 98. Revision A.11.3. Gaussian, Inc.: Pittsburgh, PA. [7] Frisch, M. J. Trucks, G. W. Schlegel, H. B. Scuseria, G. E. Robb, M. A. Cheeseman, J. R. Montgomery, J. A. Jr. Vreven, T. Kudin, K. N. Burant, J. C. Millam, J. M. Iyengar, S. S. Tomasi, J. Barone, V. Mennucci, B. Cossi, M. Scalmani, G. Rega, N. Petersson, G. A. Nakatsuji, H. Hada, M. Ehara, M. Toyota, K. Fukuda, R. Hasegawa, J. Ishida, M. Nakajima, T. Honda, Y. Kitao, O. Nakai, H. Klene, M. Li, X. Knox, J. E. Hratchian, H. P. Cross, J. B. Adamo, C. Jaramillo, J. Gomperts, R. Stratmann, R. E. Yazyev, O. Austin, A. J. Cammi, R. Pomelli, C. Ochterski, J. W. Ayala, P. Y. Morokuma, K. Voth, G. A. Salvador, P. Dannenberg, J. J. Zakrzewski, V. G. Dapprich, S. Daniels, A. D. Strain, M. C. Farkas, O. Malick, D. K. Rabuck, A. D. Raghavachari, K. Foresman, J. B. Ortiz, J. V. Cui, Q. Baboul, A. G. Clifford, S. Cioslowski, J. Stefanov, B. B. Liu, G. Liashenko, A. Piskorz, P. Komaromi, I. Martin, R. L. Fox, D. J. Keith, T. Al-Laham, M. A. Peng, C. Y. Nanayakkara, A. Challacombe, M. Gill, P. M. W. Johnson, B. Chen, W. Wong, M. W. Gonzalez, C. Pople, J. A. 2003. Gaussian 03. Revision B.05. Gaussian, Inc.: Pittsburgh, PA. [8] Frisch, E. & Frisch, M.J. Gaussian 98 User's Reference, 2nd Edition, Gaussian, Inc., Pittsburgh, PA, 1999. [9] Masamura, M. 2002. Structures, Energetics, and Spectra of Cl-(H2O)n Clusters, n = 1-6: Ab Initio Study. J. Phys. Chem. A. 106: 8925-8932.
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
Table 1. Shift of the OH harmonic frequencies for Br-(H2O)4 clustera Isomer (1) Isomer (2) Experimentb 378 (729) 370 297 (610) [2] 273 (587) 230 126 (73) [2] 207 (242) 180/160 109 (611) 199 (650) a The shifts are given with respect to the corresponding mean of the symmetric OH strech in water (3913 cm-1). The parentheses are IR intensities. The numbers in square brackets denote degeneracy. The frequency shifts are given in cm-1 and the IR intensities in km/mol. bBased on Ref. [2-4].
(1) (2) Figure 1 The optimized structures of Br-(H2O)4. The values are distances in .
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
Synthesis of Biosilica Nanoparticles from Rice Husk

E.Handayani, K. Dahlan, A. Maddu
Departement of Physics, Faculty of Mathematics and Natural Sciences Bogor Agricultural University Jl. Meranti, kampus IPB darmaga Bogor 16680 Phone: +62-251-625481 Fax: +62-251-625708 Abstract
Rice husk is one of the major residues after rice production. It contains 1329% inorganic components, of which 8797% is SiO2 (silica) in an amorphous state. The use of silica from the residual rice husk has been a major concern of many researchers. Rice husk ash (RHA) can be obtained by burning or heating rice husk above 6001000oC. The presence of amorphous silica, crystalline silica or both in RHA depends on the temperature of heating used in this process. In this research, the presence and characteristic of amorphous and crystalline silica in RHA samples treated with various concentrations of hydrochloric acid (HCl) and heated at various temperatures were examined quantitatively. X-ray diffraction (XRD characterizations shows that at heating temperature at 600oC yielded amorphous silica and at temperature 800oC and 1000oC yielded crystalline silica. Keywords: rice husk, amorphous, crystalline, diffraction
In the era of energy crisis the use of biomass for alternative energy source has become more interesting. Rice husk can be a raw material for making fuel briquettes as well as a biomass fuel for Indonesia as agricultural country produced heat and electricity generation [2]. However, many waste of agricultural product such as: risk massive burning of rice husk produces some toxic husk, straw, wood powder and skin of many seed. gases, plenty particulate mattersand ash, which has Most of husks are commonly used for agricultural a serious impact on environment. purposes such as animal feed and fertilizer. Rice husk ash (RHA) is mainly composed of Additionally, it is used as an additive compound in silica. RHA can be obtained by burning or heating building materials such as a component of bricks rice husk above 400500oC. The presence of and a reinforcing material in cement. amorphous silica, crystalline silica or both in RHA Rice husk (or rice hull) is one of the major depends on the temperature of heating used in this residues after rice production. It contains 1329% process. Many studies have shown that RHA inorganic components, of which 8797% is SiO2 obtained at around 1,000oC contains a certain (silica) in an amorphous state [1]. The use of silica amount of tridymite and cristobalite [1]. Although from the residual rice husk has been a major there are many studies dealing with the concern of many researchers. Some researches transformation of amorphous silica to tridymite and show that rice husk content of some silica about 90 cristobalite at around 1,000oC, quite a few studies - 98 % [1-5]. The applications of the extracted have examined the quantities and characteristics of silica have been published in many papers. the crystalline phase in heated silica. Table 1 Sample description of rice husk and rice husk ash used in this study
Introduction
The aim of this research is to get silica powder from rice husk and to know its characteristics. While benefit of this research is
getting silica powder amorphous and crystal with high purity that able to be used in its wide application and also more optimal of exploiting 1
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
chaff waste. In this research, the presence and characteristic of amorphous and crystalline silica in RHA samples treated with various concentrations of hydrochloric acid (HCl) and heated at various temperatures were examined quantitatively by analyzing the crystalline silica content based on Xray diffraction (XRD) and Atomic Absorption Spectophotommetry (AAS) method [6].
4 5
B2 B3
800oC 1000oC

Purification process in acid condition meant to eliminate metal oxides and non metal from rice husk ash because chloride acid that given will fasten metal oxide that is P2O5, K2O, MgO, Na2O,CaO and of Fe2O3 become its chloride and oxide non metal except silica turned into its acid. Result of composition analysis rice husk at Tables 1 indicating that element silica (Si) very dominant in rice husk. Tables 3 show result of AAS of rice husk at some treatments. From Tables 3 it can be seen that yielded silica mean of rice husk is 87.22%. It can be seen also that acid pretreatment have an effect on to yielded silica content. At sample A1, higher heating temperature resulted in content of silica progressively lower, while at A3, higher heating temperature resulted that content of silica increased. Table 3 Result of AAS of rice husk at some treatments No. Sample Silica (%) 1 2 3 4 5 6 A1B1 A1B2 A1B3 A3B1 A3B2 A3B3 99.77 88.10 51.12 88.25 96.30 99.82
Experimental
Synthesis Silica from rice husk was conducted step by step started from washing, draining, charcoal, incineration and purification. The aim of washing is to clean rice husk from impurity effect of dirt. 10 gram rice husk cleaned with aquadest 2-3 times. Draining process conducted by rice husk put into a crucible and heated in a muffle furnace at 190oC during 1 h. Rice husk charcoal conducted by heated in a muffle furnace at 300oC during 30 minute. Incineration process conducted to know dusty content of rice husk. Incineration conducted by heated with muffle furnace at 600oC during 1 h so yielded RHA. Purification process of sample was conducted to dissociate silica of RHA. Method that used for this purification is acid pretreatment that is by using HCL solutions. In this research, it was conducted 3 variations acid pretreatment which is 10%, 20% and 37% HCl. Its purification process was conducted by including RHA into backer glass and wetting it with hot aquadest. Additions, the mixture was added by 5 ml HCL and evaporated. Additions, 20 ml aquadest and 1ml HCL was poured in to backer glass and let heat for 5 minute. The mixture then filtered with paper filter and cleaned 4-5 times with hot aquadest. Result of screening in the form of solid residue along with its paper filter heated at 300oC during 30 minute till paper filter to become charcoal and then continued by heating at 600oC, 800oC and 1000oC till which remain only silica sediment (SiO2) white chromatic during 6 h. After obtaining silica then process continued to identify by using Atomic Absorption Spectophotommetry (AAS) method. The physical properties and morphology of synthesized samples were examined by various techniques. The study of crystallinity was conducted by an X-ray diffractometer (Shimadzu diffractometer). Table 2 Treatment No. Sample code 1 A1 2 A2 3 B1
Variations 10 % HCl 37% HCl 600oC
In this research it was, used two variations temperature (600, 800 and 1000oC) and acid pretreatment (10, 20 and 37% HCl). Representative XRD patterns of the RHA and heated RHA samples at 600oC are shown in Fig. 1. The XRD patterns of the RHA samples show a single diffuse band centered at around 21.6o. This indicates that the silica in these materials was in an amorphous state. According to bellow picture, both pictures were not show difference which is significant. Its meaning that acid pretreatment was not too having an effect on to crystallinity. The XRD patterns of the heated RHA samples at 1000oC are show at Fig.2. The XRD patterns of the RHA samples show any diffuse band. This indicates that the silica in these materials was a crystalline state. The heated RHA samples contained over 78.45 % crystalline silica at 1000 C after 6 h of heating. Both the picture did not show the difference very significant, this indicated that acid pretreatment did not have an effect on the degree of crystalinity.
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
a b Figure 1 The XRD pattern of silica extracted from rice husk ash at 600oC. a) 10% HCl b) 37% HCl. Symbols Q: peak position of Quartz.
C, T
C, T
C T T
(a) (b) Figure 2 The XRD pattern of silica extracted from rice husk ash at 1000oC. a) 10% HCl, b) 37% HCl. Symbols C and T: peak position of cristobalite and tridymite respectively. 2. Areerob, T; Chiarakorn, S. ; Grisdanurak, N. Influence of Functional Silanes on Hydrophobicity of MCM-41 Synthesized from Rice Husk. Division of Environmental Technology, School of Energy and Materials, King Mongkuts University of Technology Thonburi, Thaliland. Bharadwaj, A. ; Wang, Y. ; Sridhar, S. and Arunachalam, V.S. 2004. Pyrolysis of Rice Husk. Current Science, Vol. 87, NO. 7. Dahlan, I. ; Meei Mei, G. ; Kamaruddin, A.H. ; Mohamed, A.R. and Keat Teong Lee. 2008. Removal of SO2 and No Over Rice Husk Ash (RHA)/CaO-supported Metal Oxides. Journal of Engineering Science and Technology Vol. 3, No. 2, 109 116. Goyal, A. ; Kunio, H. ; Ogata, H. ; Garg, M. ; Anwar, A.M. ; Asrhaf, A. and Mandula. 2007. Synergic Effect of Wheat Straw Ash and Rice Husk on Strength 3
Conclusions
In this study we know that rice husk ash (RHA) content of silica. This silica was yielded by heating and acid pretreatment. Amorphous silica was yielded at heating temperature 600oC and crystalline silica was yielded at 1000oC. Acid pretreatment have effects in yielded silica content but did not have effects in crystallinity.
3.
Acknowledgements
This work was supported by the Beasiswa Unggulan DIKTI Depdiknas RI.
4.
Refferences
1. Shinohara, Y and Kohyama, N. 2004. Quantitative Analysis of Tridymite and Cristobalite Crystallized in Rice Husk Ash by Heating. Industrial Health. 42, 277285 5.
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
6.
Properties of Mortar. Journal of applied Sciences 7 (21) 3256-3261. Ahmed, A.E. and Adam, F. Effective and Selective Heterogeneous Catalysts from Rice Husk Ash for the Benzylation of Some Aromatics. School of Chemical Sciences Universiti Sains Malaysia Penang, Malaysia. 2008.
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
The Use of Eggshells as Calcium Sources for Synthesis of Bone Mineral

A.Nurlaela1, S.U. Dewi1, K. Dahlan1, D.S. Soejoko2
1
Departement of Physics, Faculty of Mathematics and Natural Sciences Bogor Agricultural University Jl. Meranti, kampus IPB darmaga Bogor 16680 Phone: +62-251-625481, Fax: +62-251-625708 2 Department of Physics, Faculty of Mathematics and Sciences University of Indonesia, Kampus UI Depok 16424 Phone: +62 21 7872609-10,7270160, Fax : +62 21 786 3441
Abstract
Eggshell is a potential source of calcium because it contains CaCO3 about 94-97%. In the present work, a precipitation method of apatite synthesis has been used using eggshell as calcium source and synthetic KH2PO4 as phosphate source. The eggshell was calcinated to decompose all the carbonate (CO3) phases. Variations of calcinations temperatures were 900 oC and 1000 oC. X-ray diffraction (XRD) and Fourier Transform-Infra Red (FT-IR) characterizations were employed to characterize the calcined eggshell and the apatite. It was found that most CO3 decomposed during calcinations at 1000 o C for 5 hours. The XRD profile illustrated specific diffraction angles showing the formation of apatite crystals. Keywords: Eggshell, apatite, calcinations, precipitation
Introduction
Bone is the substantial unit of human skeletal system that has an important role in human life, which supports the body and its movement. Bone structure consists of 30% organic, 55% inorganic and 15% water constituents, by weight [1,2]. Inorganic substance of bone is well known as mineral phase of bone. The major mineral component of bone and other calcified tissues is poorly crystalline hydroxyapatite (HAP) [3]. Stoichiometric HAP, Ca10(PO4)6(OH)2, occurs as hexagonally packed crystal and its Ca/P ratio is 1,67 [1,3]. There is most of carbonate (CO32-) on apatite from biological tissues. The CO32- ion can substitute for either the phosphate (PO4)3- or the hydroxyl (OH-) groups of HAP, leading to a nonstoichiometric apatite mineral [3]. Calcium phosphate compound which has same characteristic with bone mineral such as HAP can be synthesized using various types of calcium (Ca) and phosphor (P) sources from either synthetic or natural materials. This calcium phosphate compound was chemically similar to that found in bone, so that can use for bone substitution [4,5]. It is believed that HAP was synthesized from natural materials can be accepted better by the organism because of its similar physical-chemical characteristics [6]. Several methods are used to produce HAP, they are conventional wet synthesis, solid-state reaction, hydrothermal exchange process and calcinations of thermal skeletal bone [2,6,7]. In 1995 M. Sivakumar et al have converted calcium carbonate skeleton of Indian coral into hydroxyapatite with used exchange process method [6]. G. Felicio et al in 1999 have produced non stoichiometric HAP from CaCO3 of marine algae collected on the Braziian coast, using hydrothermal synthesis method [4]. Stoichiometric hydroxyapatite with Ca/P molar ratio 1,67 was synthesized using hens eggshell as calcium source and phosphoric acid using precipitation method by K. Prabakaran et al in 2005 [2]. The present work describes the synthesis and characterization of apatite using eggshell as calcium source and synthetic KH2PO4 as phosphor source using precipitation method.
Experimental
Calcination of Eggshell Eggshell was washed and its inner membrane was eliminated. Before that, it was dried in room temperature. Then it was calcinated to remove organic composition of its content and decompose calcium carbonate into calcium oxide. The variations of calcinations treatment were 900 o C for 5 hours and 1000 oC for 5, 10, and 15 hours. According to FTIR and XRD characterization, the best treatment of calcination has been used to obtain the best calcined eggshell which will used as starting material for synthesis of apatite. It was indicated by the largest amount of calcium oxide (CaO) or the least calcium carbonate 1
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
(CaCO3). Eggshells used in this experiment were hens eggshell and ducks eggshell. Synthesis of Apatite Apatite was synthesized by in situ precipitation method. KH2PO4 solution (0,5 M) was dropped into CaO solution (0,3 M). Stirred precipitation was performed by controlling temperature at 37 oC. After the precipitation, stirring process was maintained as long as 30 minutes. Then, sample solution was precipitated through aging process for 12 hours. Sample solution was filtered using filter paper to gain the sediment. Furthermore, sediment was heated in furnace at 110 oC for 3 hours. Finally, white dried powder was obtained. Then, it was characterized by X-ray diffraction (XRD) and Fourier TransformInfra Red (FT-IR) characterizations.
Results and Discussion

Characteristics of Calcined Eggshell Eggshell contains calcium carbonate (CaCO3) about 94-47% which releases CO2 in to the air to form calcium oxide (CaO), when it is heated. Figure 1 and 2 shows FT-IR spectra of calcined hens eggshell at 900 oC for 5 hours and 1000 oC for 5,10, and 15 hours. The spectral data indicate that carbonate ion is present in each calcined eggshell. Carbonate (CO32-) content in each calcined hens eggshell is indicated by the infra red carbonates group bands around wavenumber of 871 cm-1 and 1420 cm-1 for heating at 900 oC, while for heating at 1000 oC the bands are detected around wavenumber of 875 cm-1 and 1450 cm-1. According to figure 2, transmittance of CO32- at 1000 oC for 5 hours highest than the other. The fact indicates that most CaCO3 was decomposed to form CaO during calcinations, at 1000 oC for 5 hours. Therefore, 1000 oC for 5 hours is chosen for further process. With same assumption, the optimum process of calcination was used for heating ducks eggshell. Figure 2 shows XRD pattern of both hen and duck heated eggshell at 1000 oC within 5 hours. According to JCPDS data base (82-1691 and 471743), both calcined eggshell is dominated by CaO, but in calcined hens eggshell still contain CaCO3. Figure 2 FTIR spectra of heated hens eggshell at 1000 oC within (a) 5 hours (b) 10 hours, and (c) 15 hours
a
Figure 3 XRD patterns of (a) calcined hens eggshell and (b) calcined ducks eggshell at 1000 oC within 5 hours Characteristics of Synthesized powder XRD pattern of white dried fowder which was obtained from in situ precipitation method (figure 4) shows that the majority of powder correspond to HAP (JCPDS 09-0432), ACA (350180), and BCA (19-0272). Apatite which was resulted from calcined hens eggshell have highest peak as BCA, but in the apatite was resulted from calcined ducks eggshell it is as HAP. The fact is 2
Figure 1 FTIR spectra of heated hens eggshell at 900oC for 5 hours
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
caused the presence of CaCO3 in calcined hens eggshell. The CO3- ion can substitute for either the phosphate (PO4)3- or the hydroxyl (OH-) groups of HAP. A carbonated apatite (ACA) will be formed, if CO3- substitutes OH- group. While substitution of (PO4)3- with CO3- will produce B carbonated apatite (BCA)[1,3,8].
a HAP ACA BCA
Conclusions
The heating process at 1000 oC within 5 hours has been obtained both calcined hens and ducks eggshell. Pure CaO was resulted from calcined ducks eggshell. According to XRD pattern, the highest peak of apatite was synthesized from calcined ducks eggshell was HAP. While in the apatite synthesized from hens eggshell, it was BCA.
Acknowledgements
This was supported by Beasiswa Unggulan and Hibah Kompetensi DIKTI. Depdiknas R.I.
References
[1]. A Hidaki, Science and medical applications of Hydroxyapatite, Tokyo Medical and Dental University. 1991. [2]. Prabakaran K, Balamurugan A, and Rajeswari S. 2005. Development of Calcium Phosphate Based Apatite from Hens Eggshell. Bull. Mater. Sci ; 28 (2): 115-119. [3]. Pleshko N, Boskey A, and Mendelshon R. 1991. Novel Infrared Spectroscopic Method for the Determination of Crystallinity of Hydroxyapatite Minerals. Biophysical society; vol. 60:786-793. [4]. Felicio G, Laranjeira C.M. 2000. Calcium Phosphate Biomaterials from Marine Algae, Hydrothermal Synthesis and Characterisation. Quimica Nova;23 (4):441-446. [5]. K Deepak, Pattanayak, P Divya, Upadhyay Sujal, Prasad R.C., Rao B.T., and Rama Mohan T.R. 2005. Synthesis and Evaluation of Hydroxyapatite Ceramics. Trends Biomater. Artif. Organs; 18 (2):87-92. [6]. Sivakumar M, Sampath Kumar T.S., Shantha K.L., and Panduranga Rao K. 1996. Development of Hydroxiapatite Derived from Indian Coral. Elsevier; Biomaterials; 17 (17):1709-1714. [7]. Garreta E, Fernandez T, Esteve J, Colominas C, and Kempf L. 2002. Synthesis of Biocompatible Surfaces by Different Techniques. Mat. Res. Soc. Symp. Proc; vol. 724:N8.11.1-N8.11.6. [8]. Sari Y.W, Maddu A, Dahlan K, Fajriyah H.I, Dewi S.U, and Soejoko D.S. 2008. In Situ Synthesis of Composite of Calcium Phosphate Carbonate-Polyglycolide. J. Nanosaintek; Vol X No Y: 1-4.
Figure 4 XRD pattern of apatite (a) calcined hens eggshell as calcium source (b) calcined ducks eggshell as calcium source
Figure 5 FTIR spectra of both apatite, red line: hens eggshell as calcium source and blue line: ducks eggshell as calcium source Figure 5 shows FTIR spectra for both sample indicated the presence of OH- group around 3500 cm-1, CO32- group at 1418, 1455 and 1525 cm-1, while (PO4)3- at 138, 605, and 565 cm-1.
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
Hydraulic Analysis of Water Installation in Beringin Tiga Village, Curup Bengkulu

Halomoan P. Siregar, Sukirno
Appropriate Technology Development Centre LIPI Jl. Ks. Tubun 5 Subang 41213, Telp. (0260) 411478, Fax. 411239 E-mail : halomoan2001@yahoo.com Abstract
Clean water procurement for family needs is not only urban problems but also happening in rural area like in the hilly areas. In the urban area we often heard that the water distribution supply to the settlement is not enough or sometimes clog. The water problems in the hilly rural area is the location of the spring water usually far from the settlement, so that cause difficulties for the people to reach it. In this paper will be forwarded the work of piping installation of the clean water from the water spring located in the hilly area to the storage tank to be built at near village street the distance about 700 meters and still another 700 meters to the people settlement in Beringin Tiga village Curup Bengkulu from storage tank. The contour of ground surface which is the piping will be through is valleyed. Hydraulic analysis is will be done on the first piping installation work from water spring to the storage tank based on topography measurement. The hydraulic analysis resulted that the supply of the water piping installation system can still be done by the gravity system. Water piping installation system result about 65 liters/minute or 1,1 liters/sec. The system design of piping installation and pipeline characteristic of the work will be also described. Keywords : Hydraulic analysis, Installation system design, Pipeline characteristic.
Introduction
Water is the basic needs of the people that can not be avoided. In the urban area we often heard that the water distribution supply to the settlement is not enough capacity for fulfilling society needs or sometimes clog. In the other hand the water problems in the hilly rural area is the location of the water spring usually far from the settlement, so that cause difficulties for the people to reach it. The same problem in the Beringin Tiga village, Curup Bengkulu, faced by the people, where this village situated in the hilly area, they should be going to the valley long distance for getting some water especially for cooking and so as for taking a bath. Every day they spent of time for about one hour for taking water for they needs. Many villages in the Sumatera island or may be in Java and others island still be like as in the Beringin Tiga village. But by the government program that the condition continually improved by building the infra structure of basic needs of the people like water supply and etc. Appropriate Technology Development Centre as one of the non department government institution, having a rural society empowerment programs that can be directly implemented to the people in the village. These program is aimed
especially to the people in the rural area for fulfilling the basic needs of the people like clean water needs and etc., for getting better life. In this paper will be described the work for building water installation in Beringin Tiga village, Bengkulu where the clean water from the water spring located in the hilly area far about 0,7 kilometer from the storage tank to approach the settlerment. The contour of ground surface which is the piping will be through is valleyed. Hydraulic analysis is done before piping installation based on topography measurement. The result of the hydraulic analysis is that the supply of the water piping installation can still be done by the gravity system. At the end of the piping installation to be built water storage tank of 5000 liters capacity. Water piping installation result about 65 liters/minute. The system design of piping installation and pipeline characteristic of the work will be also described. The Work of Installation System Design Some works have to be done first to get system design of installation as follows : Topography Measurement Before building a piping installation system, measurement of ground topography is a must. The measurement was done by simple tools, like water pass, some small wood bar, simple 1186
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
telescope/camera and metering roll as instead of using theodolite device. Thats why it is necsessary many point to measure along distance of 700 meters from water spring to the storage tank for proper accurate measurement. Head of the elevation from the sea level of some point is measured by the altimeter device. Water Source Capacity Measurement Firstly the capacity of water spring was measured by making conduit at the output part of
water catching. This work also done by the simple way to fill a bucket of a certain capacity from the water conduit output and timing it. Installation System Model and Piping Selection The installation system design can be approached by the gravity system installation model figured as follows :
hf
Z1 Z2
datum Figure 1 Model of the installation system. For selection of the piping used to the installation system i.e., some combination method is done by the simple way to decide the water flow in the piping is assumed to be about 0,5 2,0 m/sec and so that the pipe diameter can be determined later and rechecked it by the flow diagram based on Williams-Hazens formula or Darcy-Weisbach Chart. Where Z is head elevation and Q is capacity of flow with suffix supply 1 and output 2. And, Some mathematic model for the solution can be expressed from the installation system above,
Q = 0.27853CD 2.63h 0.54l 0.54
Qo
Where V is speed of flow, g is gravitational acceleration and f is coefficient of friction The pipeline characteristic can be figured on which the required head is plotted as a function of the rate of discharge of the pipeline.
H req = Z + kQ m
(4)
Where H req is required head; if the required head has been given in advance, it will called the available head, Z is elevation difference head, the value coefficient k and the exponent m vary depending on the flow regime. The regime is assumed to be laminar where value of m = 1 and k will be, 128v(lrated ) (5) k= gD 4
(1)
h = 10.666C 1.85 D 4.87Q1.85l (2) This is empirical formula of the HazenWilliams where flow of water Q (m3/sec), diameter D (m), a coefficient of velocity C Results and Discussion (constant of the type of the pipe material), head loss h (m) and pipe length l (m). Some data as the result of the work is as Another approach can be solved also by applying follows (table 1) dan table 2, Bernoullis equation, V2 4 fl hf = (3) 1.5 + D 2g Table 1. Data installation parameters resulted from design calculation/or assumption result. No. Installation parameter Value 1 Type of pipe PVC 2 Diameter of pipe, mm (inchi) 40 75 or (1.5 3.0) 3 Head loss, m/1000 m pipe length 2.0 5.0
Proceeding Book
1187

January 24, 2009 ISBN : 978 979 19201 0 0
Output capacity installation, liter/sec or (m3/sec) (If use pipe diameter of 75 mm) Output capacity installation, liter/sec or (m3/sec) (If use pipe diameter of 40 mm) C, coefficient of velocity (for smooth pipe) Coefficient of friction, f Table 2. Data installation parameters resulted. Installation parameter Head elevation of water spring (altimeter measurement), m Head elevation of lowest part of the installation(altimeter measurement), m
5.7 or (00057)
5 6 7
1.2 or (0.0012) 130 140 0.05
No. 1 2 3 4 5 6 7 8 9 10 11 12
Value 1045 990
Head elevation of the peak part of piping installation(altimeter measurement), m Head elevation of the end part of piping installation, m The distance water spring to the lowest part piping, m The distance the lowest part piping to the peak part of piping, m The distance from the peak part to the end of piping, m Hydraulic head of water spring (base on the lowest part of pipng installation at the distance of 990 m from the water spring as a datum point), m. Hydraulic head of the peak part of the piping installation, m Hydraulic head of end piping installation, m Output capacity of the water spring, l/sec Out put capacity of piping installation, l/sec Pipe diameter used from water spring to point B is 75 mm, and from B to D is 40 mm. Topography measurement result (figure 2),
1033 1032 300 199 201 40.93 33.68 32.18 2.17 1.1
Height level, m 1200

1000 800 600 400 200 0
datum
0 100 200 300 350 400 500 600 700 Figure 2. Topography sketch of pipe installation on ground surface
Distance, m
The distance of point A B = 300 m where point A is source of water spring (40.93 m), B C = 199 m ; point B (datum line) is the lowest point and there are river cross through this point ; point C is the peak head of pipeline (38.68 m); C D = 201 m; point D is the end of pipeline where storage tank located. The head difference between point A and D only about 8.75 m and the distance A to D 700 m, in this case the data that used for design calculation
is the data measured by manual method, because the scale of altimeter device too rough. And also because of the head difference too small, it is risky if using pipe diameter of 75 mm for all length of the pipe installation, because inertia force of water in the part of climb pipe (from point B C) can be bigger or almost the same than the force from A to B. Hence, by the recalculate again it is decided to use pipe diameter of 40 mm from point B to D.
Proceeding Book
1188

January 24, 2009 ISBN : 978 979 19201 0 0
H 6
4 2 0 -2 -4 -6 -8 -10
pipeline curve
A1
Q
The piping installation work resulted is about 65 liters/sec capacity or 1.1 liters/sec. Hydraulic analysis is a tool to predict the installation system that will be built for getting success installation. It should be there are some solution approach to be used to reach some comparable result.
References
1. Stephenson D., 1984, Pipeflow Analysis, Amsterdam, Elsevier Science Publisher B.V. JICA, 1974, Water Distribution, Text Book Series No. 41. Nekrasov B., (.), Hydraulics, Moscow, Peace Publishers. Lewitt E.H., 1958, Hydraulics and Fluid Mechanics, ELBS and Sir Isaac Pitman & Sons Ltd.
Figure 3 Pipeline characteristic on H Q diagram Graphical solution for output capacity of the pipeline installation shown by figure 3. Point A1 is intersection pipeline curve to the absis line Q, means that the value on point A1 is the output capacity of the installation.
2. 3. 4.
Conclusions
Proceeding Book
1189

January 24, 2009 ISBN : 978 979 19201 0 0
Characterization Ni Nanocatalyst on Si (100) Substrate By Sputtering Growth Methods

Patricia Lubis 1,2, Altje Latununuwe1,3, Toto Winata1
1
Laboratory for Electronic Material Physics, Faculty of Mathematics and Natural Sciences, Institut Teknologi Bandung,Bandung 40132, Indonesia 2 Department of Physics Education, FKIP PGRI University, Palembang, Indonesia 3 Department of Physics Education, FKIP Pattimura University, Ambon, Indonesia patricia_lubis@yahoo.com
Abstract
Characterization Ni metal nanocatalyst had been analyzed on Si (100) substrate with Sputtering Growth Methods. The sputtering conditions used were 4x10-1 torr in pressure; 1500C in temperature; 77.3 sccm in Ar gas flow; and 90,30,20 minutes in time depositions. After growth deposition, metal nanocatalysts were annealed in temperature 6000C for 30 minutes. Characterizations of Ni metal nanocatalysts were used Scanning Electron Microscopy (SEM) and Energy Dispersive Analysis X-Ray (EDAX) and we found the tendency of smallest structure with decrease of time deposition. Keywords: nanocatalyst, sputtering, anneal.
Introduction
Since Iijima [1] was first discovered carbon nanotubes (CNT) in 1991 with arc discharge methods, its stimulate a worldwide research effort to improving their synthesis, structure determination and physical properties calculation and identification. Carbon nanotubes (CNTs) are considered as one of the most promising candidates to replace some of todays materials used in microelectronics manufacturing because the advantage of CNT where it can be metallic or semiconducting depend on diameter and chirality (the angle at which the grapheme layer is rolled up)[24]. However to control accurate diameter and chirality of CNT still in research [2], some scientist introduce a metal nanocatalysts to lead the growth of carbon nanotubes. The metal catalyst is needed are transition metal such as Fe, Co, Mo, Ni and Mn [5]. The diameter and thickness (island) of catalyst must in nanometer scales; because the catalyst leads to build the CNT, ones of the metals nanocatalysts that investigate is Nickel [6]. The morphology and size of nanocatalysts are key factor in determining the shape and dimension of carbon nanotubes.[6-7]. So, synthesis of metal nanocatalyst is necessary to investigate and analysis. Scientists introduces some methods, the methods are use laser ablation, mechanical ball milling, combustion flame, microwave plasma discharge and ion sputtering, and sol-gel methods.[xxx] In this experiment we use the ion sputtering methods where the reactor called dcUnbalanced Magnetron Sputtering (dc-UBMS)
reactor This method had advantages from others such as simple method, clean process, and the way to control the shape and size of nanocatalyst [8-9]. Ion sputtering methods are used ion beam to sputter the solid target. To generate ion beam a high voltage was applied between two electrodes that placed in chamber and fill it with Ar gas. The chamber pressure and sputtering voltage both influence the morphology and growth rate of resulting nanocatalyst film. The high chamber pressures made high frequent collisions of sputtered species (Ni pellets) with the chamber gas atom leads to a rapid decrease in temperature of the sputtered atoms, and the nucleation growth above substrate. The sputtering voltage affects the chemical reactions between colliding molecules before they nucleate into substrate.
Experiment
The preparation of substrate must in good concern, before deposition of the nickel film. The substrate that we use is silicon wafer, Si(100). The substrate was cut into pieces measuring 10 mm 10 mm and then cleaned in acetone and stirred it for 5 minutes after that we put it in DI water and stirred it for 5 minutes, and clean it again with methanol and DI water and stirred it for 5 minutes each, after that we stirred it in HF 20% for 3 minutes and dry it with N2 gas. The cleaned pieces were mounted on the surface of a stainless steel resistive heater of the sputtering chamber (figure1).
Proceeding Book
1190
The Firs Internation Seminar on Science and Technology st nal n

January 24, 2009 ISBN : 978 979 1920 0 0 01
Table 1. The G Growth Parame eters of Ni-catalysts on Si(100 substrates 0) tdeposition tannealing n Num mber V I of o sam mples Sam 1 mple Sam 2 mple Sam 3 mple (Am mpere) 0,0 025 0,0 026 0,0 028 (Vol lt) 660 0 670 0 700 0 (minute e) 90 30 20 (minute) 30 30 20
Result and D R Discussion

The char racterization of metal na anocatalyst th we used w Scanning Electron Microscopy hat was g M (S SEM) to analyze the effe of time deposition ect d pa arameters on surface e morphology of na anocatalysts. In addition t the measur to rement of en nergy dispersi analysis X ive X-ray (EDAX) was done to know the Ni composition The SEM results can o i n. r be seen in the figures 2-4.F e From the figu it can ure es stimate the m morphology qu uality of the sample 1 (N Ni/Si 100) wit its time de th eposition 90 minutes is m bi igger than the sample 2 and 3. Fig.2 sho the Ni e d ows ca atalyst cluster with the size around 50 nm. The rs 0 cl lusters of sam mple 2 of Ni/Si i(100) are stil rough as ll well as the sa w ample 3, but the cluster size had t te endency to sm mallest cluster This rough condition r. du to their ti ue ime depositio and annea on aling time [1 especially sample 3. A 10], y According to th growth he pa arameter of sa ample 3, the a annealing time was just 20 minutes. It can be expl 0 lained that th sample he ne eeds much tim to form clu me usters.
. Reactor Figure 1. Sputtering R e ts get ing The nickel pellet put on targ of sputteri chamber. The nickel pellet was mad from 10 gra de am owder and gr rained it for 10 hours. T The nickel po chamber was pumped down, wit the pressu th ure below 3 10 1 to before a 3.6 orr argon gas w was introduce into the cha ed amber. After the Ar gas flo ow we maintained a wor rking pressure of 4.0 10 1 e as 77.3 sccm. Af fter torr. The argon gas wa flowed in 7 meter stable w generated ion sputterin we ng, all param with the voltage arou und 700 Volt The substra t. ate ure in me temperatu was kept i 150oC.The deposition tim that we u used was varia in 90, 30 and 20 minut able tes. The grow parameters are listed in the table 1. wth After the nickel dep position, the substrate w was nace tube, to grow nickels o annealed in the furn lusters on th substrate. The anneali he ing islands/cl temperatu was 600 oC, for 30 min ure nutes (sample 1 e and 2) an 20 minutes (sample3). nd
Figure2. SEM surface morphology of Ni /Si (sample1) Figur SEM surf re3. face morpholo of Ni /Si ogy (sample2) )
Proc ceeding Bo ook
1191

January 24, 2009 ISBN : 978 979 1920 0 0 01
round 50 nm, and had tende ency to smallest cluster ar fo time depos or sition in 20 m minutes. The annealing tim affects b me both the sur rface morpho ology and di iameter size of catalysts It needs advanced s. ex xperiment to in ncrease the qu uality of nanocatalysts. References R S. .Ijima.(1991).Helical Mic crotubules of Graphitic Nature, 354(63 348). pp 56-58 8. Carbon, N Sa aito.R, Dresse elhaus.G, Dre esselhaus.M.S (1998). S. Physical P Properties of CNT. Imperia College al Press. Y. Tong, C. Liu P. X. Hou, H M. Cheng, N. S. Xu, Y u, H. , and J. Ch hen, (2002) Field Emiss sion from Aligned M Multi-walled Carbon Na anotubes, Physica B, 323(1-4), pp. 156-157. A. C. Dillon, K M. Jones, T A. Bekked A K. T. dahl, C. H. Kiang, D. S. Bethune and M. J. Heben, e, J orage of Hyd drogen in Sing gle-walled (1997).Sto Carbon Na anotubes, N Nature, 386(6 6623), pp. 377-379. N. M. Rodrigue (1993). J. M N ez, Mater. Res., 8, 3233 8 T. Guo, P. Nik . kolaev, A. Thess, D. T. Co olbert, and R. E. Sma alley, (1995).Catalytic Growth of G Single-wall led Nanotubes by Laser Vaporizatio Chem. Ph Lett., 243 on, hys. 3(1-2), pp. 49-54, Wei.Y.Y, W Eres.G, M Merkulov.V.I I, and D.H. (2001). E Effect of Cata alyst Film Lowndes.D Thickness on Carbon Nanotube Growth by G Area Chemic Vapor Deposition. cal D Selective A App.Phys.L Vol.78. N Lett. No.10. 1394-1396. Th homas A. E Edison, Altje Latununuwe Ahmad e, Ridwan, S Sukirno. (2007 Fabricatio of Al7). on Nanocataly ysts using Ev vaporation Method for M Carbon Na anotube Grow Proceedin of the wth, ngs 2007 Asian Physics Sy n ymposium (A APS 2007) November 29 30, 2007 Bandung, In 7, ndonesia. Altje Latununu A uwe, Andhy Se etiawan, Patri Lubis, icia Yulkifli, TotoWinata, Sukirno. (2008). , an Nanocata alyst Co-Fe dengan e Penumbuha Metode Sp puttering Proc ceeding of the National e Physics S Seminar 2008, August 7, 2008, Malang, Indonesia. Ren.Z.F, Huan ng.Z.P, Xu.J.W Wang.J.H Bush.P, W, H, Siegal.M.P, Provencio.P P.N, (1998), Synthesis d of Large Arrays of Well-aligned Carbon Nanotubes on Glass. S Science Vol.2 282.11051107.
Figure 4. SEM surface morphology of Ni/Si . e y (sample3) ( ble.2 and Fig shows th the atom g.5 hat mic Tab composit tions decrease with the in e ncreasing of t the time depo osition. position of Si a Ni and Table 3. Atomic comp Deposition minute) Time (m 20 0 30 0 90 0 At tomic compos sitions (%) Si Ni 97.50 2.50 97.52 98.81 2.48 1.19
100 Timedeposition (minute) 80 60 40 20 0 1 2 3 AtomiccompositionofNi 5. f ime and Atom mic Figure 5 Relation of deposition ti n composition of Ni 0
Conclusion
The growth of Ni e ickel nanocata alyst on Si (10 00) substrate with depos sition time 9 30 and 20 90, have been de eposited. Its d diameter size is minutes h
Proc ceeding Bo ook
1192

January 24, 2009 ISBN : 978 979 19201 0 0
An Fpga-Based Cellular Phone Missed Call Detector

Jazi Eko Istiyanto
Electronics and Instrumentation Laboratory Physics Department, FMIPA Gadjah Mada University, Yogyakarta, INDONESIA 55281 e-mail: jazi@ugm.ac.id Abstract
Missed calls have a potential for low-cost control and monitoring of a remote instrumentation system as compared to SMS and DTMF. Normally, engineers devise the system based on the Hayes AT commands running on a PC or a micro-controller. The decreasing cost of FPGAs (Field-Programmable Gate Arrays) has increased an interest in implementing such task on an FPGA platform. A circuit, based on the Altera EPF10K10LC84-4 FPGA, for the detection of a cellular phone missed call has been designed and implemented. The number of unanswered rings is displayed on the display board. Two cellular phones are used: one as the calling phone, and the other as the receiving end. The receiving phone is connected to the FPGA using a data cable via an RS-232 interface. The design and bitstream downloading processes have been done using the Max+Plus II 10.2 software. The describe-andsynthesize philosophy has been used in which designs are expressed in VHDL. The VHDL-based designs are then synthesised automatically using the software to produce a bit-stream necessary to reconfigure the FPGAs. The missed call detector requires 30 CLBs and a delay of 13 ns. The circuit runs at 28,32 Mhz. It is concluded that the circuit does the job in spite of the fact that the EPF10K10LC84-4 is not the latest technology. The trickiest part of the design process is matching baud rates between the cellular phone and the FPGA board. The paper presents the design process and its associated tricks. Keywords: missed call, cellular-phone, FPGA, RS232, Max + Plus II 10.2
Introduction
Missed calls have a potential for lowcost control and monitoring of a remote instrumentation system as compared to SMS (Short_message Services) and DTMF (DualTone Multiple Frequencies). Normally, engineers devise the system based on the Hayes AT commands running on a PC or a micro-controller. The decreasing cost of FPGAs (Field-Programmable Gate Arrays) has increased an interest in implementing such task on an FPGA platform. Istiyanto and Effendy [1] devised a system to switch on/switch off a remote set of LEDs (Light-Emitting Diodes) attached to a micro-controller based circuitry. To switch on/off a set of LEDs, the user send a command, written as an SMS, to a cellular phone attached to the microcontroller. The system in [1] has been modified to enable aremote monitoring of a temperature measurement in Istiyanto and Purwadi[2]. Basically, the LEDs in [1] has been replaced, with the necessary circuitry, by an LM555 temperature sensor, in [2]. There has also been an improvement in the design of the command language to be used in the SMS and
the required algorithm to interpret the commands. While in [1], the user has to send the command to switch on/off a set of LEDs, in [2], the user can specify at what time or every what intervals he/she needs a report on the status of the temperature readings. The work in [2] was then modified into Istiyanto and Alrosyid [3]. In [3], the moniored variables were not only the temperature but also some events that triggered an alarm. The alarm may be the result of a temperature reading which exceeds a certain pre-set value. The work in [1], [2], and [3] were all based on the SMS. Istiyanto and Hakim [4] devised a system for home automation based on the DTMF. The prototype allows the user to switch on/off a set of light bulbs, and to lock/unlock a door by sending a DTMFencoded command. Depending on the government regulation, sending a DTMF can be cheaper or more expensive than sending an SMS. An alternative to SMS or DTMF, which is free of charge is the missed call. Istiyanto and Wibowo[5] devised a system to control a DC motor based on the missed call. A missed call is defined as an unanswered telephone call. In principle, the 1
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
system does not need to use a cellular phone, a PSTN might as well dose the job. But for a reason of practicality, two cellular phones has been used. In [5], the control command is encoded in the number rings before a missed call. A Field-Programmable Gate Arrays (FPGAs) consists of a set of reconfigurable blocks allowing a designer to utilise them without the need to consult the services of a fabrication foundry [6]. An FPGA is different from a microntroller. While a microcontroller is similar to a microprocessor, it lacks the number of registers and suitable only for control purposes. Hence, the name microcontroller. Both microcontroller and microprocessor have fixed architectures. FPGAs do not have fixed architecture. In fact, it is the designer who determnine the architecture. Circuits based on the FPGAs can be designed using the describe-and-synthesise approach [7] using the VHDL (Very High Speed Integrated Circuit Hardware Description Language) which is an IEEE standard [6]. VHDL improves the designer productivity while at the same time minimises design errors. The VHDL platform are now quite mature as compared to previous systems, and obviously as compared to laboratory quality systems such as the work by Istiyanto[8]. FPGAs and VHDL systems are now getting cheaper, so much so that their cost is competitive to the cost of a microcontroller development system. This has open up opportunities to implement digital systems, not only in microcontrollers, but also in FPGAs [9]. The rest of the paper is organised into several parts. The first part describes the Hayes AT command. Then, the components of the design are described. A discussion on the resulting circuit is then carried out. Finally, the paper is closed with concluding remarks. The Hayes AT Command Cellular phones can be used as modems. Modems obey Hayes AT command (AT stands for attention). There are resources downloadable through the Internet explaining the details of how to configure and treat cellular phones as modems. One such resource is [10]. While it is claimed that the Hayes AT commands are universal and applicable to all cellular phone makes and series, differences do exist and programmers are to consult the
suitable resources. A software which can be used to test the Hayes AT Commands is the Windows Hyperterminal. Alternatively, software such as serial watcher or PDUSpy can as well be used. These software converts text into PDU(Protocol Data Unit) code, or vice versa. The software can be used to send/read SMS in the PDU code and the information completed about SMS received. A missed call is an incoming call which is ignored/un-answered until disconnected by the cellular telecommunication provider or stopped by the caller before this time. When a cellular phone attached to the serial port of a Windows-based PC and connected to enable AT command is called, it is possible to set the hyperterminal to display RING. DESIGNING THE SYSTEM The Overall Design Fig 1. illustrates a cellular phone connected to the evaluation board wizard of FLEX-A01 EPF10K10LC84-4 FPGA via RS232 which in turn is connected to the FPGA. The FPGA is then connected to the seven segment display.
Figure 1. Overall design of an FPGA based cellular phone missed call detector Figure 1 shows cellular phone 2 (HP-2, the receiver. Note: Indonesian calls a cellular phone a handphone) connected to wizard the FLEX A01 via data cable. In the Wizard FLEX A01, there is RS-232 that turns a TTL input into serial signal as an interface between the cellular phone, and the output of the RS232 is then fed into the FPGA. FPGA module design The FPGA module design consists of 3 sub-modules:RS-232 Receiver, the missed call detector- counter and the seven segment driver. The modules are depicted as schematic as shown at figure 2. The design has been actually written completely in VHDL. However, the paper presents only the schematic due to available space. 2
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
Figure 2. The modules of an FPGA based cellular phone missed call detector RS-232 Receiver An RS-232 receiver module consists of 2 parts: a baud rate generator and an rx signal receiver. The diagram of the RS-232 circuit is shown in figure 3. The baud rate generator generates a clock signal to count the number of rx data input, so that the cellular phone communicate with the FPGA, while rx signal receiver receives the serial RXI and converts it into an 8-bit data a1. An RST pin is used to reset the baud rate generator, a CLK pin gives a trigger pulse from a 9,6 MHz crystal, a BAUDRATE pin functions as the baud rate selector and a RXI is a serial data input.
Figure 4 The diagram of the missed call detector - counter The Seven Segment Driver The seven segment driver is used to display the a2 input. In this case, the input uses a 5-bit data as illustrated in figure 5. Pin A acts as the selector of the seven segment display connected to the CM seven segment activator. The driver of the seven segment outputs SEGS. The seven segment will displaying a hexadecimal number from 0 to F.
Figure 5 The diagram of seven segment driver Implementing the System Missed call detector display The setting of baud rate of the cellular phone can be done by an empirical trial and error starting from 9600 to 115200 bps until a communication is established. It is found that the baud rate is 19200 bps. The baud rate (speed) can be varied as shown by the serial watcher (figure 6). The RING can be displayed in the serial watcher as hexadecimal numbers (figure 6) or ASCII characters (figure 7).
Figure 3 The diagram of RS-232 receiver The Missed Call Detector Counter A missed call detector - counter compares an input signal a1 with the hexadecimal number 0D 0A 52 (which is the ASCII code for <CR><LF>R). If the pattern is found then the device will start an incremental counting. The pattern to look for is actually RING (or in ASCII codes OD OA 52 49 4E 47 OD OA, but there is no other pattern beginning with R other than RING. The output of the compare is then fed into a counter and the output of the counter is a 4-bit data a2.
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
Figure 8. The waveform diagram of the RS232 receiver circuit The performance of the missed call detector based on the EPF10K10LC84-4 FPGA can be estimated by using a software simulation as part of Max + Plus II 10.2 software. The circuit is found to have a propagation delay of 13 ns and a clock period of 35,3 ns or a frequency of 28,32 MHz. Figure 8 and 10 display, respectively, the waveform diagram of the RS-232 and the seven segment display. The floorplan of the circuit is illustrated in figure 9.
Figure 6 Serial Watcher displays hexadecimal numbers representing 'RING'
Figure 9 Floorplan of RS-232 receiver circuit
Figure 7 Serial Watcher displays RING as ASCII characters Performance
Figure 10. The waveform diagram of the seven segment driver circuit
Conclusion
4
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
Based on the experiment with the design, implementation and testing of the circuit, it is concluded that 1. FPGAs can be used to implement the cellular phone missed call detector. 2. The RS-232 receiver circuit has a propagation delay of about 13 ns and the clock period is about 35,3 ns or the frequency is about 28,32 MHz. 3. This FPGA-based cellular phone missed call detector requires 30 CLBs.
Phone, Proceedings of The 2nd Jogja International Physics Conference (JIPC), UGM. [6] ICTP-INFN, 2006, Advanced Training Course on FPGA Design and VHDL for Hardware Simulation and Synthesis, Italy: ICTP-INFN M-Lab. [7] Gajski, D.D, N.D. Dutt, A.C-H. Wu, and S.Y-L. Lin, 1992, High-Level Synthesis : Introduction to Chip and System Design, Kluwer Academic Publishers. [8] Istiyanto, J.E, 1995 : The Application of Architectural Synthesis to the Reconfiguration of FPGA-based Special Purpose Hardware, Ph.D Thesis, Department of Electronic Systems Engineering, University of Essex, UK. [9] Istiyanto J.E., 2005, Field-programmable Gate Arrays (FPGAs) untuk otomasi Industri, Kendali dan Robotika, Seminar Nasional Otomasi Kontrol dan Robotika 2005, HMEI FMIPA UGM. [10] Anonim, 2001 , AT Command Set Reference Manual, Siemens.
Acknowledgement
The support of FMIPA UGM for financing the research is highly appreciated.
Future Works
1. While a software timing simulation may be adequate to estimate the circuit performance, a physical measurement may be necessary to get the real picture of the circuit delay contributors. Different makes of cellular phone may have different features related to the Hayes AT commands. Therefore, experiments with cellular phones of different vendors is required to enable a clearer picture of the application settings.
2.
Reference
[1] Istiyanto J.E. dan Efendy Y., 2004, Rancangan dan Implementasi Pengendali Jarak Jauh Berbasis Mikrokontroler AT89C52 dan Layanan SMS, Jurnal Ilmu Dasar, FMIPA-UNJ, Juli. [2] Istiyanto J.E. dan Purwadi E., 2005, Alat Pemantau Suhu Jarak Jauh Berbasis SMS, SNI 2005, UAD. [3] Istiyanto J.E. dan Alrosyid, 2005, A Prototype of a Device Control and Alarm Monitor System Based on the GSM Short Message Service and the AVR AT90S2313 Microcontroller, TSSA 2005, ITB [4] Istiyanto J.E. and Hakim, A.R., 2007, A DTMF-based Remote Device Monitoring and Control, Proceedings of The 2nd Jogja International Physics Conference (JIPC), UGM. [5] Istiyanto J.E. and Wibowo, F.W., 2007, DC Motor Control Based on Cellular
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
Application of Distributed System in Neuroscience: A Case Study of BCI Framework

Indar Sugiarto1, Iwan Handoyo Putro2
1,2
Department of Electrical Engineering - Petra Christian University Jl. Siwalankerto 121-131, Surabaya 60236, Indonesia email: 1indi@petra.ac.id, 2iwanhp@petra.ac.id
Abstract A distributed system approach has been applied for development of electroencephalography based BCI (brain-computer interface) framework. In this method, the computing processes of BCIs sub systems are distributed over networked computers. The three most important tasks of a BCI system are signals acquisition, feature extraction and classification, and command translation or mapping. Distributing those three highly CPUs resources consuming processes in a distributed system will reduce computing complexity of BCI framework thus increasing reliability of overall system performance. In this paper, the BCI framework is developed using C++ programming language which incorporate CORBA (Common Object Request Broker Architecture) module as the communication protocol manager among BCIs sub systems. After writing the BCI framework programs, we measure the programs performance through standard object oriented analysis metrics. The result shows that the average complexity of the distributed system architecture sub module is only 2.8 where the maximum complexity of overall architecture is measured as 6.0. The proposed method can then be extended to incorporate the more human-readable visualization strategy such as 3D contour representation of brain activities. Keywords: Neuroscience, distributed system, brain-computer interface, object oriented design
Introduction
Distributed computing deals with all forms of computing, information access, and information exchange across multiple processing platforms connected by computer networks. One obvious application of this method is in the area of complex and concurrent computing, where a certain task must be accomplished using limited performance computer system. In the field of neuroscience, there is a quite new research interest which requires several computing processes to be performed simultaneously and is called brain-computer interface (or BCI in short). BCI system, as its name implies, tries to create a direct bridge between human CNS (central nervous system) with a computer or machine through neurophysiologic signals generated by the brain. It then creates a new pathway for the brain to carry its message [1]. At some extent, it is better to describe a BCI system as a hybrid system in which all of its components collaborate together to form an integrated system, as written in IEEE Signal Processing Magazine Volume 25 Number 1, January 2008: A brain computer interface is a system that includes a means for measuring neural signals from the brain, a method/algorithm for decoding these signals and a methodology for mapping this decoding to a
behavior or action [2]. This description of BCI system sounds more technical than the one described in [3] since it emphasizes the importance of measurement-decoding-mapping interaction within the system. Those three core components of any BCI system require specific amount of processing resource in order to work as ideal as possible as in dream BCI [4]. The following diagram shows how those three components work to construct a BCI system [3].
Figure 1 BCI system in general. Here the measurement task is performed by the Signal Acquisition sub-system, 1198
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
decoding and mapping tasks are performed by the Signal Processing sub-system. Accomplishing those three tasks in only one of todays general computer is unlikely to be realizable. Hence, distributing those three tasks on a well-defined distributed system sounds more logical. The following diagram shows a common system setup for brain-computer interface experiment in medical application [5].
Figure 2. System setup for common brain monitoring experiment in medical application. This paper aims to show how this mechanism could be well performed using general purpose computers arranged in a network. This paper is presented as follows. After giving an introduction, which explains the motivation of the research, the theoretical background about distributed system is explained. The detail of the proposed distributed design will be covered in section 3. Discussion about the implementation issue and its experiment result will be given in section 4. This paper will be closed with discussion and conclusion in section 5. Distributed System Using CORBA A distributed system is a collection of independent computers that appears to its users as a single coherent system [6], [7]. Furthermore, distributed system deals with collection of difference hardware and software systems which consist of more than one process or programs that running under a tight controlled rule. In distributed system, an application is split up into parts and distributed to clients who will run it locally. Unlike parallel processing, distributed application have to face the heterogeneous environments, network links of varying latencies, and unpredictable failures in the network or the computers [8]. There are many different types of distributed computing systems, such as clusters, grids, peer 2 peer [6], [9]. Each of them has their own advantage
and disadvantage; however it all has the same purpose to connect users and resources in a transparent, open, and scalable way. In reality, various types of distributed systems and applications have been developed and are being used extensively in the real world. Those systems and applications are used in the area of cracking security, particle physics, weather, climate and geography, economics and finance, graphic and visualization, and bioinformatics [7]. The Common Object Request Broker Architecture (CORBA) is an emerging open distributed object computing framework that is standardized by the Object Management Group [10]. CORBA help the developer to automate some of the common network programming tasks in distributed computing. Those include object registration, location, and activation; request demultiplexing; framing and error-handling; parameter marshalling and demarshalling; as well as operation dispatching [10]. The Common Object Request Broker Architecture (CORBA) is structured to allow integration of a wide variety of object systems. Figure 3 shows all of the components in the CORBA ORB architecture.
Figure 3. CORBA ORB Architecture [10] Object is a CORBA programming entity that consists of an identity, an interface, and an implementation [10]. That programming entity, Servant, defines the operations which support a CORBA IDL interface and can be written in a variety of languages, including C, C++ and Java. Other program entity, Client, has a responsibility to invoke an operation on an object implementation. In this scenario, accessing the services of a remote object should be transparent to the caller [10], [11]. Functionally, The ORB provides a mechanism for transparent communications request between client and targeted object implementations. By decoupling the client from the details of the method invocations, distributed programming is simplified as that client requests appear to be local procedure calls [10]. In detail, the ORB would find 1199
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
the related object implementation when clients invoke a certain task. Furthermore, the ORB should activate that object, deliver the request to it and pass any response to the correct destination. In the CORBA architecture, CORBA IDL stub and skeletons has an important role. It binds the client, server and ORB itself and also has a responsibility to automate the transformation between CORBA IDL definitions and the target programming language. The inconsistency problem that would be arisen between client stubs and server skeletons would be reduced by the implementation of compiler. Distributed Design of BCI Framework In a standard EEG-based medical experiment, there are at least three components of data abstraction which work simultaneously: data acquisition, signals database, and signals visualization. But in an applicable EEG-based BCI system, there are at least seven major components which are required to be synchronized: data acquisition, signals database/storage, feature processing (extraction and classification), visualization (temporal or spatial), command generation for actuator(s), command database, and feedback acquisition. The following diagram shows high level data abstraction used in our BCI Framework.
Figure 4 High level data abstraction of the BCI framework. The entire system works as follows. After EEG signals are sampled and acquired, these raw data are stored in the hard drive, either in standard ASCII text file or binary file. The same data are fed to two other sub-systems: signal processing (for feature extraction and classification) and visualization. In the signal processing part, the features being extracted are highly depended on the mental task given by the experiment protocol to the subject. In the visualization part, the EEG signals can be displayed spontaneously (in time domain) or being interpolated for spatial visualization. After the features are extracted, certain translation algorithm is required to decode the feature into executable commands for actuator. In order to record the activity of the actuator, a command database program is required to keep track and store the commands. After a certain mental task has been given, the subject brain will react in a certain manner which will produce specific rhythmic signals. These feedback signals are then being recorded and/or processed for further analysis. The following sequence diagram shows the early part of the work of BCI framework.
Proceeding Book
1200

January 24, 2009 ISBN : 978 979 1920 0 0 01
Figure 5 A part of seq quence diagram in a BCI fra m amework nchronize all works in t l the In order to syn rk, op nization protoc col framewor we develo a synchron which generate timing signals and a cknowledge signals for d data interchan nge request/ac between sub-systems. This synchron nization proce ess ed t nt ing is handle by object managemen system usi CORBA. The progr ram was de eveloped usi ing O brary. ACE/TAO CORBA lib The BCI framework is develope mainly I ed us sing C++ language programming which g in ncorporate se everal librarie from ope source es en re eferences. The following di e iagram shows the UML (U Unified Mode eling Languag representa ge) ation of a ne etworking p part of the distributed system e d im mplementation of the BCI framework (note that n I no all part of the BCI fr ot f ramework is displayed he ere).
Result and Discus ssion
Fi igure 6 Repre esentation of d distributed des sign implemen ntation in UM ML.
Proc ceeding Bo ook
1201

January 24, 2009 ISBN : 978 979 19201 0 0
There are several methods that can be used to test software in order to maintain such level of software quality. At some points, software testing can be considered as part of Software Quality Control. After finishing the design process (also completing verification procedure), the program should undergo validation process through several testing procedures, statically or dynamically. Static testing is a form of software testing where the softwares code is inspected without actually running the software. Dynamic testing, on the other hand, will execute the software and inspect the behavior of the software during run-time. In this paper, static code analysis is performed by measuring several metrics in order to collect information about software complexity and vulnerability. Not all of metrics for C++ will be used in this paper; only the following metrics are used: Lines of Code (LOC), Methods per Class, Complexity, and Block Depth. We use Source Monitor program developed by Campwood Software to asses our code. Figure 7 shows the metric measurement result for the distribution system part of the BCI framework program.
module and it is a very strong component of the Maintainability Index measurement of maintainability [12]. Another important metric is the block depth, which measures how many nested blocks are exists within one upper level block. Maximum Block Depth is calculated from the top level of the source code, but namespaces are not included in this block depth metrics. Average Block Depth is the average nested block depth weighted by depth. Figure 8 shows Kiviat Metrics Graph for the corresponding metrics shown in figure 7.
Figure 8 Kiviat Metrics Graph for distributed system part of the program. We also measure how much network bandwidth is utilized by the system. We implement the system in a Local Area Network (LAN) using standard Ethernet adapter for each computer in the network. About 25% of available bandwidth is utilized when all programs in the framework run. Hence, the system doesnt need network management enhancement to increase the throughput of the overall network. Figure 9 shows the network utilization graph captured when the BCI framework is in active state. Figure 7. Metric measurement result for the distributed system part of the program. We optimized the code so that all important aspects are confined within the acceptable and reasonable range. For example, we define the maximum of complexity as 8.0 according to cyclomatic complexity measurement based on McCabe algorithm. This complexity metric is a kind of directed acyclic graph which represents the flow of control within each function as a measure of the minimum number of test cases to ensure all parts of each function are exercised. We can use this metric as a measure for the detection of code which is likely to be error-prone and/or difficult to maintain. When measuring software complexity, a significant complexity measure increase during testing may be the sign of a brittle or high-risk
Proceeding Book
1202

January 24, 2009 ISBN : 978 979 19201 0 0
Figure 9 Network utilization graph showing BCI framework activity in a distributed system.
Conclusion
A distributed system approach has been applied for development of electroencephalography based BCI (brain-computer interface) framework. Using a distributed system mechanism, we can allocate more computing power to reduce system complexity, thus improving systems reliability. The BCI framework is developed using C++ programming language which incorporate CORBA (Common Object Request Broker Architecture) module as the communication protocol manager among BCIs sub systems. Using static code analysis, the experiment shows that the average complexity of the distributed system architecture sub module is only 2.8 where the maximum complexity of overall architecture is measured as 6.0. Network utilization is only 25% and it is not a critical value. The proposed method can then be extended to incorporate the more human-readable visualization strategy such as 3D contour representation of brain activities.
References
[1] Jonathan R. Wolpaw, Brain-Computer
Interfaces as New Brain Output Pathways, Journal of Physiology, January 25, 2007. [2] Indar Sugiarto, Display and Feedback Approaches for BCI Systems, Master Thesis, University of Bremen, 2008. [3] Jonathan R. Wolpaw, Niels Birbaumer, Dennis J. McFarland, Gert Pfurtscheller, Theresa M. Vaughan. "Brain-Computer Interface for Communication and Control," Clinical Neurophysiology 113, 2002: 770. [4] Julies Kronegg, "Goals, Problems and Solutions in Brain-Computer Interfaces: a Tutorial," BCI-Info, 2005, www.bciinfo.tugraz.at/Research_Info/documents/prese ntations/files/kronegg-2005__tutorial_bci.ppt (accessed April 2008). [5] S. G. Mason, M. M. Moore Jackson, and G. E. Birch, "A General Framework for Characterizing Studies of Brain Interface Technology", Annals of Biomedical Engineering, Vol.33, No.11, pp.1653-1670, November 2005. [6] Nadiminti, K., Assuncao, M., Buyya, R., Distributed Systems and Recent Innovations: Challenges and Benefits, http://www.gridbus.org/~raj/papers/InfoNetArticle06.pdf (accessed Nopember 28, 2008). [7] A. Tanenbaum and M. Van Steen, Distributed Systems: Principles and Paradigms, Prentice Hall, Pearson Education, USA, 2002. [8] Yuan, X., Parallel and Distributed Systems, www.cs.fsu.edu/~xyuan/cis5930/lect1_intro.p pt (accessed Nopember 28, 2008). [9] Martin Crane, Karl Podesta, Distributed Systems (A very basic introduction of real world examples), http://www.computing.dcu.ie/~kpodesta/distri buted/ (accessed Nopember 1, 2008). [10] Douglas C. Schmidt, Overview of CORBA, http://www.cs.wustl.edu/~schmidt/corbaoverview.html (accessed December 12, 2008). [11] OMG, Common Object Request Broker Architecture: Core Specification, http://www.omg.org/docs/formal/04-0312.pdf (accessed December 8, 2008). [12] VanDoren, Edmond. Halstead Complexity Measure. January 10, 1997. http://www.sei.cmu.edu/str/descriptions/halste ad.html (accessed May 8, 2008).
Proceeding Book
1203

January 24, 2009 ISBN : 978 979 19201 0 0
Performance Analysis of Different Codecs in VoIP using H.323

Iwan Handoyo Putro1, Indar Sugiarto2
1,2
Electrical Engineering Department, Faculty of Industrial Technology Petra Christian University Siwalankerto 121-131, Surabaya 60256 iwanhp@petra.ac.id, indi@petra.ac.id Abstract
There are two popular Voice over Internet Protocol (VoIP) standards, H.323 and Session Initiation Protocol. The first one has become the basis for the widespread implementation of Internet telephony systems although it was not specifically designed for VoIP. The H.323 system implements G.711 codecs as its primarily speech compression protocol. However, other advanced compression algorithms that comply with H.323 system has been designed and implemented to increase compression factor while still maintain minimum loss in perceived quality. In this paper, some of those compression algorithms will be compared and analyzed base on its performances in H.323 based VoIP system. The codecs that was used in this experiment are: SJ Lab GSM 6.10, SJ Lab iLBC-30ms, SJ Lab iLBC-20ms, Microsoft CCITT G.711 A-law and Microsoft CCITT G.711 u-law. These Codecs are tested in terms of its ability to deal with jitter buffer variations. The result shows that Microsoft CCITT G.711 u-law gives the best performance on both wired and wireless LAN environment Keywords : VoIP, Codecs, H.323, Jitter Buffer, Performance Analysis
Introduction
Voice over Internet Protocol (VoIP) is a terminology for the process of transmitting voice through IP networks. The other terminologies that have the same meaning are IP telephony or Internet telephony. Although ISDN connection is suitable enough for this purpose, faster internet connections would give better voice quality [1] [2]. Table 1 reveals that for adequate VoIP service, the communication technology should provide at least several Mbps data transfer. Table 1 Suitable Communication Technologies for VoIP [2] Technology Speed Use in VoIP Dial-up Up to 56 Not suitable kbps ISDN Up to 128 Suitable, for kbps fixed and dedicated service ADSL Up to One of the best several WAN Mbps technologies, but provides no mobility Wireless Up to Some
Technology Technology (WiFi, WiMAX)
Speed several Mbps
LAN
Up to thousands of Mbps (Gbps)
Cable
1 to 6 Mbps
Use in VoIP technologies are suitable, while some are limited by distance and signal quality. They are the mobile alternatives to ADSL. The best, but limited to the length of wires which can be short in most cases. High speed but limits mobility. Is suitable is you don't have to move.
Building a VoIP network on internal network would require almost no cost if the IP network has already existed. Figure 1 shows that when the LAN has already built, the minimum requirement is just a pair of software phone which would behave basically like a traditional PSTN phone, i.e.: dialing, calling and terminating the conversation. However, inline 1204
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
with the increasing demand of broader coverage of communication and the mobility, some network and communication equipment such as: router, phone adapter, modems and adequate a reasonably fast internet connection should be added.
ip.com/products/media-gateways.htmGateways, Gatekeepers and Multipoint Control Units (MCUs). The three components: Terminals, gateways and MCUs are widely known as endpoints and managed by a gatekeeper [6].
Figure 1 VoIP Network [1] Basically, transmitting voice via IP networks require conversion process from analog into digital signal and vice versa. The first conversion, analog to digital, is happened at the caller site. First of all the voice should be digitized, that signal then would be put on the IP packets with its source and destination address and then send it over the networks. At the callee site, the reversion process should be done to in order to get the original voice [3]. There are two popular VoIP standards, H.323 and SIP. Both standards are responsible for call establishment, maintain connection and translate IP addresses as well as telephone numbers, bandwidth management, authentications and registration [3] [4]. An Overview of H.323
Figure 2. H.323 Call Flow Diagram [5] CODEC A codec, coder/decoder, is an algorithm that used for encode and decode voice conversation. Human voice is analogue; therefore it should be changed into digital signal before entering the networks[8]. On the other side, that digital form should be converted back into analogue in order can be heard by the callee [8]. Another task for the Codecs is to reduce the packet size so it requires minimum bandwidth. In short, Codecs has two main responsibilities, deliver acceptable sound quality while maintain its compression algorithm as simple as possible. Table 2 shows the lists of Codecs that would be used in this experiment.
H.323 is the ITU-T standard and was originally created for multimedia purpose especially for LAN based conference. As the result of its broad applications, this standard was then implemented on both audio and video telephony [5] [7]. H.323 has four components, they are Terminals, http://www.voip-voice-overTable 2 Codec Comparison Codec SL Lab GSM 6.10 SJ Lab iLBC-30ms SJ Lab iLBC-20ms Microsoft CCITT G.711 a-law Microsoft CCITT G.711 u-law Sampling rate 8 kHz 8 kHz 8 kHz 8 kHz 8 kHz Bandwidth (kbps) 13.2 15 15 64 64
License Open Source Royalty Free Royalty Free Open Source Open Source
Bit Rate 13 kbps 33 kbps 15.2 kbps 64 kbps 64 kbps
MOS 4.14 4.14 4.45 4.45 1205
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
Experimental and Discussion

The aim of this experiment is to investigate which Codecs could give better performance on jitter buffer variation. Jitter buffer in the VoIP area is a shared data area where voice packets would be collected, stored, before then sent to the next voice processing level [10]. The voice packets should be stored first in order to minimize delay variations which could be caused by network congestion, timing drift, or route changes. There are two problems related to the jitter buffer size. If a jitter buffer is configured too small then an excessive number of packets may be discarded. On the other side, if more space is given it will create longer delay [9]. Two common types of jitter buffer are in the form of hardware based and software based. A hardware-based jitter buffer is configured by the manufacturer while a dynamic jitter buffer is software-based and therefore can be easily configured to adapt with the variation in the networks [9]. The networks components that used in this experiment are shown on Figure 3. Two PC are connected through LAN. One of them has
PCI wireless adapter and connected with a PDA through wireless adhoc configuration. The PCs and the PDA are installed with SJPhone as its software phone [11]. The VoIP traffic is captured by using Wireshark [12] to get the percentage of the packet drop according to the jitter buffer variations.
PC-1 LAN PC-2
Computer
Computer
192.168.1.3
192.168.1.8 192.168.1.5
PDA
Wireless Connection using AdHoc Configuration
192.168.1.6
Figure 3 Network Configuration Results on LAN Testing
Table 3. Codecs Performance in terms of Drop Packets as a result of Jitter Buffer Variation LAN Testing Jitter Buffer (mS) Codecs SJ Lab iLBC20ms (%) 83.1 80.6 77.8 72.2 69.9 68.1 67.1 64.8 58.2 50.0 44.4 32.9 30.7 28.6 29.7 28.1 24.5 27.1 26.4
SJ Lab GSM 6.10 (%) 85.0 80.7 78.4 73.1 70.6 70.6 70.2 65.8 58.1 55.3 51.2 43.8 40.6 39.1 39.0 38.5 35.3 34.6 30.3
SJ Lab iLBC30ms (%) 98.0 91.1 87.8 86.9 85.5 84.0 78.8 77.5 69.3 66.1 55.6 50.3 47.6 46.2 45.7 42.9 36.3 34.7 29.8
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19
Microsoft CCITT G.711 A-law (%) 91.1 77.2 70.7 65.4 62.2 55.1 50.5 44.6 34.6 20.7 13.4 16.0 12.5 8.6 5.4 8.0 7.3 11.9 7.5
Microsoft CCITT G.711 u-law (%) 90.2 74.4 66.6 61.2 57.8 55.4 52.3 45.5 23.1 15.4 9.1 5.2 4.3 3.3 2.9 2.9 2.7 2.5 2.5 1206
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
Jitter Buffer (mS)
SJ Lab GSM 6.10 (%) 29.8 23.2 18.0 22.5 21.3 18.2 15.8 15.5 13.3 9.7 6.5 9.9 4.3 3.8 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
SJ Lab iLBC30ms (%) 19.2 10.4 14.4 14.7 13.7 12.6 12.5 11.0 10.9 12.5 11.2 1.3 2.9 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
120 Percentage of Drop Packet
Codecs SJ Lab iLBC20ms (%) 3.7 0.9 11.3 0.2 0.1 0.1 0.1 0.1 0.1 0.1 0 0 0.8 0.6 0.1 0.1 0.1 0 0.1 0 0 0 0 0 0 0 0 0 0 0 0
20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50
Microsoft CCITT G.711 A-law (%) 4.0 2.6 2.5 2.5 2.5 2.5 2.3 2.1 1.1 0.2 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
Microsoft CCITT G.711 u-law (%) 2.4 1.1 0.2 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
SJ Lab GSM 6.10 100 80 60 40 20 0 1 6 11 16 21 26 31 36 41 46 Jitter Buffer SJ Lab iLBC30ms SJ Lab iLBC20ms Microsoft CCITT G.711 A-law Microsoft CCITT G.711 u-law
Figure 4 Packet Drop Percentage on LAN Testing
Results on Wireless Testing Table 4. Codecs Performance in terms of Drop Packets as a result of Jitter Buffer Variation Wireless Testing
Proceeding Book
1207

January 24, 2009 ISBN : 978 979 19201 0 0
Jitter Buffer (mS)
SJ Lab GSM 6.10 73.9 72.9 72.8 72.7 72.6 72.6 72.6 72.6 72.4 68.7 56.8 44.1 37.3 36.5 36.4 36.4 34.0 20.5 7.1 36.4 36.1 36.1 32.6 19.6 15.8 15.7 7.6 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1
SJ Lab iLBC30ms 88.9 85.4 85.1 84.9 84.8 84.8 84.8 84.3 79.2 74.5 63.3 52.9 52.5 52.4 52.4 52.4 51.9 48.6 46.5 27.3 25.3 24.2 24.8 24.8 24.7 24.7 24.3 23.6 16.2 23.5 21.9 21.6 24.4 24.4 24.4 24.5 24.5 24.0 22.9 19.3 10.5 8.1 8.8 8.8 8.8 8.8 8.8 8.6 0.0 0.0
Codecs SJ Lab iLBC20ms 99.9 99.9 99.9 99.8 99.8 99.7 99.5 99.0 97.5 84.2 61.8 51.7 50.6 50.2 50.1 49.8 48.5 42.3 33.8 50.0 47.7 40.0 38.4 36.2 27.3 22.8 19.4 18.8 12.8 6.4 0.4 0.7 0.5 0.2 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50
Microsoft CCITT G.711 A-law 99.4 96.2 77.0 75.5 74.8 63.7 49.5 61.8 30.7 27.5 20.7 15.0 17.7 14.7 24.5 21.6 18.5 6.9 4.0 1.1 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
Microsoft CCITT G.711 u-law 20.0 16.8 14.0 13.6 12.3 12.8 13.6 12.5 8.1 6.8 3.3 5.3 5.4 4.9 3.6 4.5 4.4 4.3 4.3 4.2 1.7 0.3 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 1208
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
120 Percentage of Drop Packet 100 80 60 40 20 0 1 7 13 19 25 31 37 43 49 Jitter Buffer
SJ Lab GSM 6.10 SJ Lab iLBC30ms SJ Lab iLBC20ms Microsoft CCITT G.711 Alaw Microsoft CCITT G.711 ulaw
Figure 5.Packet Drop Percentage on Wireless Testing 8. Introduction to Codecs, http://www.ozvoip.com/codecs.php Conclusions 9. Problem : Jitter Buffer, http://www.voiptroubleshooter.com/problem Various VoIP Codecs has been analyzed s/jitterbuffer.html in both wired and wireless environment. The 10. experiment result shows that Microsoft CCITT http://searchunifiedcommunications.techtarg G.711 u-law gives the best performance in terms et.com/sDefinition/ of jitter buffer variations both on wired and 0,,sid186_gci906844,00.html wireless LAN environment. These could be 11. http://www.sjlabs.com/sjp.html happened as a result of its higher bit rate 12. http://www.wireshark.org/ implementation, although it consumes more bandwidth than other Codecs.
References
1. 2. 3. http://www.networklab.co.uk/cmodem/voip. html http://voip.about.com/od/requirements/a/ban dwidth.htm Ramakhrishnan, R.S., Kumar, P.V., Performance Analysis of Different Codecs in VoIP Using H.323, http://tifac.velammal.org/CoMPC/articles/30 .pdf Schulzrinne, H., Rosenberg, J., A Comparison of SIP and H.323 for Internet Telephony,http://www.cs.columbia.edu/~hg s/papers/Schu9807_Comparison.pdf http://www.td.rutgers.edu/documentation/FA Q/H.323_Tutorial/
4.
5. 6.
Overview of H.323,http://www.cisco.com/univercd/cc/td/doc/ product/ software/ios122/rel_docs/gktmpv31/ h323intr.pdf. 7. H.323, http://www.iec.org/online/tutorials/acrobat/h 323.pdf
Proceeding Book
1209

January 24, 2009 ISBN : 978 979 19201 0 0
The use of Chitosan from Shrimp Shell as A Medium of Apatite Growth for Biomaterial Composite Formation
S.U. Dewi1, A. Nurlaela1, K. Dahlan1, D.S. Soejoko2, Y.W. Sari1
1
Departement of Physics, Faculty of Mathematics and Natural Sciences Bogor Agricultural University Jl. Meranti, kampus IPB Darmaga Bogor 16680 Phone: +62-251-625481 Fax: +62-251-625708 2 Department of Physics, Faculty of Mathematics and Sciences University of Indonesia, Kampus UI Depok 16424 Phone: +62 21 7872609-10,7270160 Fax : +62 21 786 3441 setiautamidewi@yahoo.com
Abstract
Formation of biomaterial composite consisting of organic matrix and inorganic mineral has been studied. The inorganic minerals were calcium in the form of CaO resulted from calcinations of chicken eggshell at 1000 oC, and the other was phosphor in the form of phosphates. Whereas the organic matrix was a commercial chitosan originated from shrimp shell. To produce the composite, calcium phosphates were grown on organic matrix of chitosan using ex situ method. The result samples were further dried at 50 C. Characteristics of the samples were performed using X-ray Diffraction (XRD), Scanning Electron Microscopy (SEM) and Fourier Transform Infrared Spectroscopy. The XRD profile illustrated specific diffraction angles at peaks of apatite crystals. This data were supported by FTIR spectra that showed transmittance peak of chitosan, phosphates and carbonates from apatites. The SEM micrograph also showed granules of apatite. Keywords: composite, apatite, calcination, chitosan.
Introduction The use of biomaterial has raised as well as the world population [1]. One indicator is the development of biomaterial as bone material substitution [2]. Bone is composite composed by weight, about 55 % inorganic mineral, 35 % organic matrix and 10 % water [3]. The major inorganic mineral bone compound is in the form of calcium phosphate called biological apatite. It presents in many phase of apatite such as carbonated apatite, hydroxyapatite and octa calcium phosphate [4]. One that has comparably good stability in bone is hydroxyapatite (HAp). Stoichiometric HAp, Ca10(PO4)6(OH)2 in its most common form, occurs as a hexagonal closed packed crystal [5]. HAp has used widely biomaterial in biomedical applications, such as dental implants, coating hip implants and other orthopedic implants [6]. HAp can be synthesized both from synthetic and natural calcium source. Commonly, calcium synthetic sources are Ca(OH)2 [7], Ca(NO3)24H2O [8], CaCl22H2O [9]. Several natural sources for calcium that have been studied are obtained from Indian coral [10], marine algae [11], bovine [12] and eggshell [13-14]. An in vivo study of HAp which is resulted from bovine bones in construction
of resected mandible of rabbits shows excellent results. Other observation, in vitro and in vivo studied of natural apatite from compact bone and synthetic HAp show that the natural apatite is well tolerated and has good osteoconductive than synthetic apatite [15]. Therefore, it is highly assumed that the use of natural source can increase biocompatibility of biomaterials. In this obsevation, apatite-chitosan composite synthesized from eggshell as calcium precursor and chitosan as inorganic matrix. The main content in eggshells is calcium carbonate that is about 95%. The remaining 5% includes calcium phosphate and magnesium carbonate and soluble and insoluble proteins. Chitosan is natural biopolimer, polysaccharides obtained by chemical deacetilation of chitin from shrimp, crabs [16]. The chitosan has been reactivity as chemical bond and structrural resembles organic matrix bone. The other adventeges of using chitosan are biodegradable, biocompatible, non antigenic, biofuncrional and osteoconductive [17]. These reasons based for development composite for bone subtitution.
Experimental
The materials used in this study are calcium which was obtained from calcinations of chicken 1
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
eggshell at 1000 C for 5 hours, phosphor from KH2PO4 pro analysis and chitosan obtained from CV. Safatama Chitosan, Jepara Centre Java. The calcium and phosphor precursors were used to produce inorganic minerals, apatite. The chitosan was used to organic matrix. To produce composite apatite-chitosan, calcium phosphate was grown on organic matrix of chitosan using ex situ method. Calcium phosphate in the form of apatite was prepared by precipitation method. Solution of 0.5 M KH2PO4 was dropped wisely into stirred 0.3 M CaO solution at 37 oC. Ex situ synthesized of composite was formed as chitosan solution, chitosan dissolved in 3 % (v/v) acetic acid solution was added slowly into precipitation. Following the final step, thus mixture was aged at room temperature for 12 hours. Then, the precipitate was filtered and dried at 50 oC. The influence of chitosan was observed by varying the amount of chitosan that was added. Variances of chitosan were 10, 20 and 35 % (Table 1). Character of samples was analysed using Xray Diffraction (XRD), Scanning Electron Microscopy (SEM) and Fourier Transform Infrared Spectroscopy (FTIR). XRD carried out using a Philips diffractometer using CuK radiation ( = 1.5405 ). The measurements performed with diffraction angle (2) in the range of 10 to 80. Further, evaluation of surface morphology for each sample was performed by JEOL SEM (scanning electron microscope). Next analysis, to identify chemical functional group in apatite compound was performed using FTIR spectrometer. Table 1 Samples code Sample code Chitosan amount (%) A B C D 0 10 20 35
carbonated apatite, CA (JCPDS 19-0272) and Acarbonated apatite, CA (JCPDS 44-0778). Synthesizing composite without chitosan used to compare with apatite-chitosan composites. In sample A, apatite without chitosan shows highest peak at 2 = 32.00 indicated as CA; however, the composite, sample B, C and D at 2 = 31.88, 31.78 and 31.78 successively (Figure 2), it is HAp which is in reference at 31.77. Shifting angular is caused by presented carbonate (CO32-) group which is still at calcined eggshell. The CO32can substitute groups hydroxyl (OH-) and phosphate (PO43-). CA appears when it substitutes (PO43-) and CA presents when it substitutes (OH) [18-22]. Due to presenting of CO32- can be to shift diffraction angular to higher and lattice parameter being bigger [23]. Lattice parameters of HAp were showed in Table 2. Another CA, CA also presents but increasing of added chitosan resulted decreasing of CA phase in sample. Rising of chitosan resulted composites that is one phase, HAp. It results from ionic interaction between CO32- as the negatively charged and NH2 from chitosan as the positively charged . In these patterns, chitosan peak cant identify. It is caused by molecules being homogeneously dispersed in the composite [24] and chitosan have amorphous phase (Figure 1) [25]. To identify of presented chitosan is showed by analyses FTIR. Table 2 Lattice parameters of samples Sample code Lattice parameter a () c () A 9.461 6.925 A1 9.446 6.915 A2 9.424 6.876 A3 9.425 6.878
RESULTS AND DISCUSSION Composite of apatite-chitosan has obtained in this research. It was indicated from several analyses. The XRD patterns show peaks that is correspond to HAp (JCPDS 09-0432), BFigure 1 X-Ray diffractrogram of pure chitosan [25].
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
HAp BCA ACA
B
I (Counts)
2 Figure 2 X-ray diffractogram of Sample A, B, C and D. The observation of FTIR shows presenting groups compound in the samples. To identify of presented chitosan is used spectrum of pure chitosan (Figure 3) [15] that shown overlapping NH and OH starching vibrations at wavenumber 3500-3100 cm-1 is the, -CH starching 2990-2850 cm-1. Major absorption band 1220-1020 cm-1 represents free primary amino group (-NH2). The other group is amide at 1651-1600 cm-1. The characteristic of apatite spectrum HAP is obtained from FTIR spectrum of sample A (Figure 4A). In this spectrum shows peak of OH, CO32-, Peak ofOH is around 3500 cm-1. and PO43-, Absorption bands at 1033, 604 and 565 cm-1 are assigned to vibration of 4 and 3 of PO43. Peak of CO32- group were indicated at 1461, 1422, and 873 cm-1. Presented chitosan is identified by comparison spectrums of pure chitosan and apatite with the composites (Figure 4B-4D). To identify of chitosan is showed at 1630 cm-1, 2958 and 2856 cm-1 respectively. Group of NH occurs at around 3500 cm-1 that overlap with OH. The rising of chitosan resulted area the peaks which are bigger. It shows increasing of amount the groups.
Figure 3 FTIR Spectrum of chitosan [15]
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
A B C D
Figure 4 FTIR spectrum of sample A, B, C and D.
Acknowledgements
A B This work is partially supported by the Hibah Kompetensi 2008 research project and Beasiswa Unggulan program from the Directorate of Higher Education, Ministry of National Education, The Republic of Indonesia.
References
C D [1] Laurencin, C.T., Ambrosio, A.M.A., Borden, M.D. & Cooper, J.A. 1999. Tissue engineering: Orthopedic applications. Ann Rev Biomed Eng;1:19-46. Pilliar, R.D., Filiaggi, M..J, Wells, J.D., Grynpas, M.D., & Kandel R.A. 2001. Porous calcium polyphosphate scaffolds for bone substitute applications- in vitro characterization. Biomaterials;22:963-972. Daftari, T.K., Whitesides T.R Jr., Heller, J.G. 1994. Nicotine on the revascularization of bone graft. An experimental study in rabbits. Spine;19:904-911. Nurmawati M. 2007. Analisis Derajat Kristalinitas, Ukuran Kristal dan Bentuk Partikel Mineral Tulang Manusia Berdasarkan Variasi Umur dan Jenis Tulang [Thesis]. Bogor: Bogor Agricultural University. Pleshko, N., Boskey, A. & Mendelsohn R. 1991. Novel Infrared Spectroscopic Method for The Determination of Crystallinity of Hydroxyapatite Minerals. Biophy. J. 60: 786-793. Wilson, O.C.Jr. & Hull, J.R. 2008. Surface modification on nanophase hydroxyapatite with chitosan. Mater. Sci. and Eng;28:434437. 4
[2] Figure 5 Scanning electron micrographs of Sample A, B, C and D. SEM micrographs showed granules of apatite. White spots indicated higher crystalinity of apatite. The micrographs supported XRD results that identified apatite crystal phase. [3]
Conclusions
It is to be hoped that the use of bio-based calcium source and biopolymer will increase the biocomptibility of calcium phosphate. Ex-situ method of apatite solution and chitosan solution yields composite apatite-chitosan. Added chitosan result in decreasing of carbonate which is into apatite crystall so it produces single phase composite, HAp. From this study found that forming of HAp can be at low temperature not only at high temperature.
[4]
[5]
[6]
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
[16]
Victoria, E.C. & Gnanam, F.D. 2002. Synthesis and characterization of biphasic calcium phosphate. Trends Biomater. Artif. Organs;16 (1):12-14. Kong, L., Gao, Y., Lu, G., Gong, Y., Zhao, N., & Zhang, X. 2006. A study on the bioactivity of chitosan/nanohydroxyapatite composite scaffolds for bone tissue engineering. Europ Polymer J.;42:3171-3179. Sari, Y.W., Maddu, A., Dahlan, K., Fajriyah, H.I., Dewi, S.U., & Soejoko, D.S. 2008. In situ synthesis of composite of calcium phosphate carbonatepolyglycolide. Jurnal Nanosains & Nanoteknologi;2 Sivakumar M., Kumar T.S.S., Shantha K.L., & Rao K.P. 1996. Development of Hydroxiapatite Derived from Indian Coral. Biomaterials; 17 (17):1709-1714. Felicio, G., & Laranjeira, C.M. 2000. Calcium Phosphate Biomaterials from Marine Algae, Hydrothermal Synthesis and Characterisation. Quimica Nova;23 (4):441-446. Ooi, C.Y., Hamdi, M., Ramesh, S. 2007. Properties of hydroxyapatite produce by annealing of bovine bone. Ceramics International;33:1171-1177. Prabakaran, K., Balamurugan, A., & Rajeswari, S. 2005. Development of Calcium Phosphate Based Apatite from Hens Eggshell. Bull. Mater. Sci; 28 (2): 115-119. Sari Y.W., Amrina Q. H., Maddu A., Dewi S.U., Dahlan K., Soejoko D.S.. (2008). Thermal Effect on Apatite Crystal Synthesized from Eggshells Calcium. 4th Annual meeting of ISACB Shangri-la, Ortigas Centre, Manila Philipines. September 10-13, 2008. Saraswathy, G., Pal, S., Rose, S. & Sastry, T.P. 2001. Novel bio-inorganic bone implant containing deglued bone, chitosan and gelatin. Bull. Mater. Sci.;24 (4):415420. Kong L. 2006. A study on the bioactivity of chitosan/nano-hydroxyapatite composite scaffolds for bone tissue engineering,
[17]
[18]
[19]
[20] [21]
[22]
[23]
[24]
[25]
European Polymer Journal; 42: 31713179. Liu H. Novel injectable calcium phosphate/chitosan composites for bone substitute materials. Acta Biomaterialia. 2006; 2: 557-565. Tomazic, B., Tomson, M., & Nancollas, G.H. 1976. The Growth oc Calcium Phosphates on Naturl Enamel. Calcif Tiss Ress.;19:263-273. Meyer ,J.L., Eanes, E.D. 1978.A Thermodynamic Analysis of the Amorphous to Crystalline Calcium Phosphate Transformation.Calcif Tiss Ress.; 25:59-68. Takagi, S., & Mathew, M. 2001. Structures of biological minerals in dental research. J Res Natl Stand Technol 106 : 1030-1044. Danilchenko, S.N., Pokrovskiy, V.A., Bogatyrov, V.M., Sukhodub, L.F., & Sulkio-Cleff, B. 2005. Carbonate location in bone tissue mineral by X-ray diffraction and temperature-programmed desorption mass spectrometry. Cryst. Res. Technol.; 40 (7): 692 697. Muller, L., Muller, F.A. 2006. Preparation of SBF with different HCO3 Content and Its Influence on the Composition of Biomimetic Apatites. Acta Bioamterialia. ; 2(2):181-189. LeGeros, R.Z., LeGeros, J.P., Trautz, O.R., & Shirra, W.P. 1971. Conversion of monetite, CaHPO4, to apatites: effect of carbonate on the crystallinity and morphology of apatite crysllites. Advances in X-Ray analysis. Plenum Press, New York; 57-66. Jin, HH., Lee, CH., Lee, WK., & Lee, JK. 2008. In situ formation of the hydroxyapatite/chitosan-alginate composite scaffolds. Mater. Latters;62:1630-1633. Mohd Nasir, N.F., Mohd Zain, N., Raha, M.G.,& Kadri, N.A. 2005. Characterization of Chitosan-poly (Ethylene Oxide) Blends as Haemodialysis Membrane. American Journal of Applied Sciences; 2 (12): 1578-1583.
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
An Amplified Single Channel Soliton Pulse Propagation in Optical Fibres at 3500 Km Distance.
Zakaria Z 1,2, M.A. Eberhard 1, K. J. Blow 1
1
Photonic Research Group, Electronic Engineering Dept. School of Engineering and Applied Science, Uni. of Aston, Birmingham, B4 7ET, UK 2 Pusat Pengajian Sains, UTHM, Parit Raja, Batu Pahat 86400 Johor Abstract
Using the derivation of nonlinear Schrdinger equation, NLSE, the combined effect of GVD and SPM in the anomalous dispersion region can lead to the formation of optical soliton. Loss-induced broadening in fibre is unfavourable especially when solitons are used in optical transmission. Solitons need to be amplified periodically to restore its energy. The development of the EDFA was a major breakthrough in the field of communication. The problem resulting from the introduction of EDFAs is the amplified spontaneous emission ASE added at the amplifiers which can severely degrade the signal-to-noise-ratio, SNR at the receiver. The ASE noise of the amplifiers results in the fluctuations in the arrival time of a soliton in the receiver, which is called the Gordon-Haus, GH timing jitter. Since fibre losses lead to soliton broadening, the concept of average soliton is introduced where the soliton power is increased by a factor of G ln G G 1 , where G is the gain of the amplifier which compensates for the losses of the fibre segment so that the average power in one amplification period coincides with the power of a fundamental soliton in the absence of fibre losses. Another limitation is the interaction between adjacent solitons, which in turn may lead to a considerable reduction of the velocity of information transmission. Finally the concept of design diagram is introduced to check that every condition for proper propagation of soliton is satisfied. In designing such a system, the required system length and the operating bit rate are the main considerations. Keywords: NLSE, Soliton, guiding centre soliton, GH timing jitter
to give Introduction The evolution of the slowly varying field amplitude over a nonlinear, dispersive fiber is described by the generalised nonlinear Schrdinger equation (GNLS), which is given by [1,2]
i U 1 2U 2 = sgn ( 2 ) N2U U Z 2 2
(1.4)
where N is defined as
LD P0T02 = . N = L NL 2
2
(1.5)
2 1 2 i A = 2 A A A A 2 z 2 T 2
Eliminating N from equation 1.4 by introducing (1.1)
u = NU = LD A to take the standard form of

NLSE:
where it includes the effects of group velocity dispersion 2 , nonlinearity and loss . Consider first the lossless case where is set to zero, equation 1.1 becomes the nonlinear Schrdinger equation (NLS),
u 1 2 u 2 + u u=0 (1.6) + i 2 Z 2 where sgn( 2 ) = 1 is taken for anomalous GVD;

for normal dispersion regime, the dispersive term (2nd term in equation 1.6) should be negative. The parameter N governs the relative importance of the effects of GVD and SPM on the pulse propagation: N << 1 : dispersive effects are dominant N >> 1 : nonlinear effects are dominant N 1 : both dispersive and nonlinear effects are important 1
1 2 2 i A = 2 A A A 2 z 2 T
Equation 1.2 can be normalized using
U= A P0 u = NU = N
(1.2)
A , z , = T (1.3) Z= T0 LD P0
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
N = 1: SPM and GVD in the anomalous and normal dispersion regimes Second case is when loss is considered. In practice loss cannot be disregarded and it has a major effect on soliton propagation since any decrease in the pulse power leads to a decrease in nonlinear effects and the balance between GVD and SPM is destroyed. This can lead to pulse broadening as the pulse propagates because GVD begins to dominate the pulse evolution. The implication of fibre losses as well is that the signal power in a communication system must be boosted in order to compensate for the loss in the fibre. The amplification is usually done by lumped erbium doped fibre amplifiers (EDFAs)[3,4,5]. The inclusion of EDFAs introduces some noise into the system as well as amplifying the signal. In a linear system this noise is undesirable as it degrades the signal to noise ratio (SNR) but in a nonlinear soliton system the effect is more drastic and the ability of the soliton to readjust to accommodate small changes leads to a random timing jitter known as Gordon-Haus (GH) jitter as the solitons propagate.[6,7] In order to increase the information carrying system, it is desirable to launch the pulses close together. Unfortunately the overlap of the closely spaced solitons leads to mutual interaction and therefore to serious performance degradation of the soliton transmission system. This is as a result of the small finite tails of the soliton that extend into neighboring solitons which in turn form a
superposition that causes them to propagate at different velocities [8].Another study showed that the inclusion of fibre loss also leads to dramatic increase in soliton interactions[9].To reduce the soliton interactions few studies have been conducted; the use of Gaussian-shaped pulses [10], introduction of phase difference between neighboring solitons[11], the use of third order dispersion [12,13] are among others. Numerical implementations In this study we look at a single channel soliton system where we are to compare the predictions of the analytical models with numerical simulations of the system performance. We look at the example of the design diagram and when certain effects are dominant. We describe the methodology for the numerical simulations and the system to be used for the simulations. Consider figure 1.1 below. We start with the design diagram which consists of a transmitter, a series of fibres with amplifiers in the channel and a receiver. Optical amplifiers are placed periodically along the fibre link to compensate losses between the amplifiers. The parameters we are interested to look at are the pulsewidth, of the Gaussian pulse from the transmitter, dispersion, D of fibre and the spacing L A between amplifiers.
Transmitter
Fibre
amplifier
Fibre
Receiver
LA
Figure 1.1 Fibre links with periodic loss compensation through optical N amplifiers In this work, we formulate analytically all the limitations imposed in the system using a design spreadsheet in excel. Analytically we look at the Q factor for a soliton system with a fixed amplifier spacing but different values of the input pulse width. The pulse width values should span the curves plotted in the design diagram. Next we vary the dispersion value and repeat for different pulse width. Later we carry out the similar process with different amplifier spacing. We then compare the results of the analytical and the numerical as described below. For the purpose of the numerical simulation, we use the mqocss optical communication systems 2
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
simulator written by Dr Marc Eberhard of Aston University, and a sample of a configuration file (the XML code) describing the communication system operating at 10 Gbit /s and employing soliton-like pulses is attached as an index A PRBS of 128 bits in a time window of 12.8 ns and a numerical resolutions of 213 bins, is generated and is modulated by a range of FWHM Gaussian pulses with respective peak power from the analytical model, with a rise time of 12.5 ps, 30 dB modulation depth and no insertion loss. It finally transmitted along the fibre link. The numerical simulation of the propagation is equivalent to the recirculating loop technique which consists of a fibre span and an amplifier. The actual transmission fibre in the loop is chosen to be 35, 50 and 70 km to give a total propagation distance of 3500 km i.e. a 100, 70 and 50 spans each. The loss was set to the standard value of 0.2 dB/ km. The nonlinearity in the system is assumed to be described by the pure Kerr effect, the effective area of the fibre was set to 72
propagation distance which is at every 3500 km. For an average soliton, the factor Q is defined as , We first describe the design diagrams for a fixed bit rate of 10 Gbit/s and allow the propagation distance to increase from 1000 km to 4000 km. Next we will keep the propagation distance fixed at 1000 km and allow the bit rate to increase from 5 Gbit/s to 20 Gbit/s. Second, we look at the variation of pulse width with dispersion at respective amplifier spacing.

Design diagram can help to identify possible operating ranges for amplified soliton systems so that conflicting restrictions stated are respected. For example, a design diagram to see the variation of pulse width (which refers to various effects of limitations) with amplifier spacing when propagation distance and bit rate are kept fixed, respectively. Figures 1.2-1.3 show the pulse width as a function of amplifier spacing necessary to achieve acceptable system for 10 Gb/s and 3500 km system length for average soliton. The plots show impairments arising from design limitations namely GH jitter, ASE, average soliton limit and soliton-soliton interaction. The region of a safe operation would be the intersection region below SC and ASE limits and above GH jitter. The shaded area in figures 1.2-1.3 denotes the permissible pulse width for propagation which is based purely on the analytical predictions. It can be concluded the permissible area of propagation is getting smaller as the propagation distance increases at a fixed bit rate . The inset in each figure shows the position of GH and ASL limitations (as it is too small a difference in pulse width). In figure 1.4, propagation fails as there is no intersection between the curves as the ASE is lower than GH limit. There is also a drop in permissible area of propagation as the bit rate increases from 5 to 10 Gbit/s as shown in figures 1.5-1.6 when the propagation distance is kept fixed.
m 2 . Dispersion was set from as low as 0.3 ps/ km

nm to as high as 2.5 ps/ km nm (as referred to a design diagram) and there is no dispersion slope (second order dispersion). In a typical configuration of 50 km amplifier span, the gain in amplifier would be 10 dB to compensate for the round trip loss of the loop of 10 dB and it would be 7 dB and 14 dB each for 35 km and 70 km length of fibre respectively. In the receiver an optical Gaussian filter with 20 GHz bandwidth is inserted before the detector. An optical signal passes through this detector then is filtered by an electrical Bessel filter with 8 GHz bandwidth. Noise is normally generated by the amplifiers which add noise to the transmitted signal during amplification. Each time the amplifier works, it generates a random noise signal that has different pulse format but same average noise power. Therefore the results should differ a little bit each time the simulation is run. Usually the simulation jobs are repeated many times to see how the value of Q changes. We look at Q = 6 border at the end of
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
Pulsewidth vs amplifier spacing

Pulsewidth, ps 8 6 4 2 0 0 10 20 30 40 50 60 Amplifier spacing, km GH ASL
250 200 150 100 50 0 0 10 20 30
Pulsewidth, ps
40
50
60
Amplifier spacing, km GH ASL ASE SC
Figure 1.2: propagation distance=2000 km

Pulsewidth, ps 8 6 4 2 0 0 10 20 30 40 50 60 Amplifier spacing, km
Pulsewidth, ps
150 100 50 0 0 10 20 30
GH
ASL
40
50
60
Figure 1.3: propagation distance=4000 km
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
Pulsewidth, ps
150 100 50 0 0 10 20 30 40 50 60 Amplifier spacing, km GH ASL ASE SC
Figure 1.4 Lt=10000 km- no intersection! There is where there is no combination of system parameters that satisfies all the criteria.

500 400 300 200 100 0 0 10 20
P ulsew idth, ps
30
40
50
60
Figure 1.5: Bitrate = 5 Gbit/s
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0

500 400 300 200 100 0 0 10 20
P ulsew idth, ps
30
40
50
60
Figure 1.6: Bitrate = 10 Gbit/s
A plot in figure 1.7 shows the propagation limitations for different amplifier spacings in analytical for respective amplifier spacing.The upper bound on the pulsewidth (i.e. dispersion below 0.17
ps/ km nm) is due to ASE and above that dispersion, is bounded by soliton collapse, SC. The lower bound is all bounded by GH jitter limitation.
90 80 70 60 ASE 50 40 30 20 10 0 0
Pulsewidth vs dispersion
(analytical)
Pulsewidth, ps
Soliton collapse
GH jitter
0.5 1 1.5 2
La=70 km
2.5
Dispersion,ps/km nm La=35 km La=50 km
Figure 1.7 Analytical plot for pulsewidth versus dispersion for different amplifier spacings in analytical method. Plots in figure 1.8 describing the pulse width versus dispersion with their respective amplifier spacing for a numerical method. For each plot the peak pulse width is at around dispersion 0.2 ps/km nm. It can be noted from the plot that the longer the amplifier spacing the smaller the value of its maximum dispersion. Also as the amplifier spacing increases the area under the curve decreases while maintaining the same limitations i.e. GH jitter, ASE and soliton collapse.
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
Pulsewidth vs dispersion
100
Pulsewidth,ps (numerical simulation)
80 60 40 20 0 0 0.5 1 1.5
La=70 km
Dispersion, ps/km nm La=35 km La=50 km
Figure 1.8 Numerical plot for pulsewidth versus dispersion for different amplifier spacings in numerical method.
Acknowledgements
The author wishes to thank RMIC of UTHM for the short grant available. Also to Prof K.J. Blow and Dr M.A. Eberhard of Aston University, UK. This article is taken from one chapter in PhD dissertation which is in writing-up progress. [7] D. Marcuse. 1986. An alternative derivation of the Gordon-Haus effect. IEEE J. Lightwave Technol., 10(2):273-278 [8] J.P.Gordon 1983. Interaction forces among solitons in optical fibers. Opt. Lett., 8(11): p. 596. [9] Blow K.J, D.N.J. 1983. Bandwidth limits of nonlinear (soliton) optical communication systems. Elec. Lett., 19(11): p. 429430. [10] Chu P.L, D.C. 1983. Gaussian pulse propagation in nonlinear optical fibre. Elec. Lett., 19(23): p. 956-957. [11] Anderson D, L.M. 1986. Bandwidth limits due to mutual pulse interaction in optical soliton communication systems. Opt. Lett, 11(3): p. 174-176. [12] Chu P.L, D.C. 1985. Effect of third-order dispersion of optical fibre on soliton interaction. Elec. Lett., 21(6): p. 228-229. [13] Chu P.L, D.C. 1987. Soliton interaction in the presence of loss and periodic amplification in optical fibers. Opt. Lett, . 12(5): p. 349-351.
References
[1] G.P.Agrawal. 2001. Nonlinear fiber optics. 3rd ed., Academic Press,London. [2] P.Agrawal. 2001. Fiber-optic communication systems. 3rd ed., New York: John Wiley and sons. [3] K.J. blow and N.J. Doran. 1991. Average soliton dynamics and the operation of soliton systems with lumped amplifiers. IEEE Photon. Technol. Lett., 3(4):369-371. [4] A. Hasegawa and Y. Kodama. 1990. Guiding center soliton in optical fibers. Opt. Lett.,15(24):1443-1445. [5] A. Hasegawa and Y. Kodama. 1991. Guiding center soliton. Phys. Rev. Lett.,66(1):161-164. [6] J. P. Gordon and H. A. Haus. 1986. Random walk of coherently amplified solitons in optical fibre transmission. Opt. Lett.,11(10):665-667
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
Spin-orbit Interaction Effect on the Energy Levels of Single-electron Quantum Dot

Kamsul Abraha1, Pekik Nurwantoro 1 , Moh.Adhib Ulil Absor 1 , Mohammad Darwis Umar 1 , Wahyu Tri Cahyanto2, and Sunardi 2
1 2
Department of physics, Gadjah Mada University Yogyakarta, 55281

Email : adhib_2004ugm@yahoo.com
Department of physics, Jendral Soederman University, Purwokerto, 53123
Abstract
In this paper we theoretically explore the effects of spin-orbit interaction (SOI) in a single-electron system confined in a two-dimensional parabolic quantum dot (QD). The SOI is described by the Rashba type spin-orbit coupling created by the external gate voltage. We find that the system is formally identical to the Jynes-Cummings model (JCM) in quantum optics, first derived in the context of the atom-light interaction, wherein the present case of QD the roles of such interactions are now played by the spin and orbital angular momenta of the electron. The resulst of the energy spectrum calculation shows that the SOI leads to the hybridisation of n + 1, and n , states
which result in the anti-crossing of the spectrum. Keywords: Quantum dot (QD), spin-orbit interactions (SOI), the Jynes-Cummings model (JCM).
Introduction
Semiconductor technology has come along in leaps and bounds since the development of the transistor in 1948. The major trend of the development of electronic components is to improve fabrication methods so that the size of the components becomes smaller and more circuits can be included on a chip. Smaller components also work with smaller currents and voltages and therefore more efficiently with less energy consumption. This progress in semiconductor material fabrication has given scientists the opportunity to explore a new world between the macroscopic world which obeys classical physics and the microscopic world which obeys quantum mechanics. This new world which has he range of nanometers (then called mesoscopics) offers a large variety of interesting phenomena. The growth of nanotechnology is started with the Thouless idea [1] about possibility to reduce the conductor dimension to the size of just few atoms. This question was replied immediately by the growth of microelectronics fabrication which has succeeded to make a controlled system which is only consisted of just few particles. The restriction
particle into two dimensions is known as a quantum well, then the advanced confinement in one dimension is referred to a quantum wire, and electron confinement into all three spatial directions is known as a quantum dot. Quantum dots are usually made by restricting the two dimensional electron gas (2DEG) in the semiconductor heterostructure laterally using a very small electrostatic gate or vertically by etching technique. Gallium Arsenide (GaAs) and Alumunium Gallium Arsenide (AlxGa1-xAs) are semiconductors with similar lattice constants ( 5.65 for GaAs) and thus can be brought together to form a heterostructure. At room temperature the bandgap for GaAs is 1.424 eV, while the bandgap in AlGaAs depends on the mixing factor x and can be approximated by [2] (1) Eg ( x ) = 1.424 + 1.429 x 0.14 x 2 [ eV ] 0 < x < 0.44 i.e., it varies between 1.424-2.026eV, leads to a discontinuity at their interface. Working with the GaAs-AlGaAs mentioned above, a twodimensional electron system is typically created by n-doping the AlGaAs in a heterostructure (Fig 1)
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
Original n-AlGaAs band
Energy
2DEG
Space Conduction band
donors
Original GaAs band GaAs
n-AlGaAs AlGaAs (spacer-layer)
Figure 1 Two-dimensional electron gas seen along the confining dimension

By doping AlxGa1-xAs-layer with Si-donors, the conduction electrons move to the GaAs-layer, since it has a lower bandedge. Some of the electrons added will eventually migrate into the GaAs. These electrons will still be attracted by the positive donors in the AlGaAs, but be unable to go back across the heterojunction because of the conduction band discontinuity. Squeezed in between the discontinuity and the donor potential, they are trapped in a narrow potential well (see Fig 1) and their energy component in this dimension will be quantized. At sufficiently low temperatures all electrons will be in the same (the lowest) energy state with respect to motion perpendicular to the interface. In other word, electrons are free to move in the plane parallel to the heterojunction, but restricted to the same (lowest) energy state in the third dimension. Control and manipulation of spin degree of freedom one of the most important topic both in spintronics and quantum computing processing. The SOI in semiconductor heterostructures provides the ways to couple to spin with spatial degrees of freedom, consequently has attracted more and more attention in recent years[14]. The spin properties of a few electron in nanostructure such as quantum dot have been studied. The result shows that the carrier spin properties are strongly affected in the presence of the SOI and novel features emerge in these nanostructures compared with the traditional ones without SOI. Now we notice that the SOI phenomena in nanostructure were investigated with help of the quantum optic method. In this paper, we propose and study a model of single-electron confined in parabolic quantum dot with Rashba type SOI. Under the rotating wave approximation (RWA), the SOI of the electron confined in the single quantum dot, is reduced to a Jaynes-Cummings model (JCM), which is very typical model in quantum optic[15]. This paper is organized follow. Sec. II presents general consideration of the model of single-electron quantum dot in presence the Rashba SOI.The elaboration of calculation the eigenvalue and eigenstate are given in sec.III. The result of this calculation are obtained and discussed in Sec. IV. In Sec.V we provide some concluding remarks. The Model This paper will consider a two-dimensional electron system in a semiconductor heterostructure created by n-doping the AlGaAs on GaAs. Fig 2 shows type of sample use in this analysis schematically.
Proceeding Book
Figure 2 Heterostructure AlGaAs-GaAs in the vertical cross-section. The 2DEG is localized in the substrate near the interface with the spacer. Bias voltage Vg is applied between the gate and the

January 24, 2009 ISBN : 978 979 19201 0 0
At negative gate voltages, the 2DEG is electrostatically squeezed from underneath the gate into the regions under the uncovered openings. As a result, the conduction band electrons will be confined in the xy plane, in addition to the vertical confinement at the heterointerface. By changing the lithography of the split gate, one can reduce the 2DEG to 1-D forming quantum wires and 0-D forming quantum dots. The single-electron spectra of a QDs are well described by means of the Fock-Darwin theory. This exactly solvable model assumes a parabolically confined QDs in two dimension with
decouples the system into eigenmodes of frequency % = (c / 2) . Adding the Zeeman term and rescalling energies by h0 we arrive at the Hamiltonian
+ + H = a + a+ + a a +
2
Ez z + H SOI 2
(7)
where E = g m l0 is the Zeeman energy with z 2me lb
perpendicular magnetic field ( B = Bz ),

2 1 r e r r 1 * 2 2 H= p A ( r ) + m 0 r 2 m* c 2
me the
bare
mass
of
the
electron
and
l0 = h /(m * 0 ) is the confinement length and

(1)
lB = h /(m * c ) is the magnetic length.

The Fock-Darwin spectrum and the Zeeman effect in absence the SOI since for fixed n+ these modes converge in to the same landau level at large B. Adjacent
where
r B=Bz
r r and vector potential A ( r ) is

2
r r r specified to be Gauge symmetric with A = 1 B r .
+ modes,
however,
Its mean that the symmetric gauge is also a r r Coulomb gauge, i.e. .A ( r ) = 0 , and hence r r r A ( r ) , p = 0 . Including the Rashba type SOI
converge towards different Landau levels, the Zeeman splitting is generally not large enough to match this energy sparation. Thus, to reach coincidence of spin split + modes an additional in-plane magnetic field may be applied, enabling the independent manipulation of landau level separation and Zeeman splitting. The magnetic field evolution of the low-lying Fock-Darwin spectrum can be see in figure 3[15].
H SOI
r e r = ( p A) , h c z
Hz = 1 g B B zz , 2
(2)
and the Zeeman term (3)
The total Hamiltonian reads:

2 1 r e r r 1 * 2 2 p A ( r ) + m 0 r + H SOI + H Z 2m* c 2 H = H 0 + H Z + H SOI
H=
(4)
Here, e,
B , and m* are the electron charge, Bohr
magneton, and effective mass of electron. Elaboration We start the elaboration of Hamiltonian (4) by using second quantized notation, we have Figure 3. The low-lying Fock-Darwin spectrum for clarity only n 4. The different parameterization in quantum numbers ( n , n+ ) and (n,l) are marked for the 10 lowest modes We now include the Rashba interaction of eq.(2), for which the coupling strength is related to the spin precession length and we introduce relation (6)
2 1 l% = 1 2 lB
% x + H 0 = h ( a +ax + a y a y + 1) +
with c = eB / mc and Introduction of operator
h c x + ( a y a + a x a y ) (5) 2i
% 2 = 02 + (c2 / 4) .
lSOI = h 2 /(2m * )
(8)
1 ( a x m ia y ) a = 2
Proceeding Book
101

January 24, 2009 ISBN : 978 979 19201 0 0
= ( x i y ) ,
1 2
(9)
where l% = h 2 /( m * ) . The total Hamiltonian of % QDs in presence the Rashba SOI can be written as
we have H SOI = + ( a+ + + a+ ) ( a + a + + ) (10) l%
+ + H = a + a+ + a a +
l2 Ez z + 0 + (a+ + + a+ ) (a + a + + ) % 2 2llSOI
(11)
We now derive an approximate form of this Hamiltonian by borrowing the observation from quantum optics and that the terms preceded by + in eq.(11) are counter-rotating, and thus negligible under the rotating-wave approximation[14,15] when the SO coupling is small compared the confinement. This decouples the + mode from the rest of the system, giving
identical to the JCM of quantum optics and provides systematic insight into effect SOI in QDS. Analogy to quantum optics we can understand the above Hamiltonian (12,13) in term of a spin which is coupled to two boson modes with energies . The presence of the spin leads to two effect: i) every state becomes spin split due to the Zeeman effect, ii) the additional coupling between boson modes and spin leads to anticrossings in the spectrum. In the representation in terms modes we see that the SOI leads to a coupling between adjacent modes with opposite spin due to the operators
H = + n+ + H JC
where
(12) (13)
H JC = n +
With
E z z + (a + + a + + ) 2
and
% = (l02 ) / 2llSOI
n = a + a . This is
a + , a in the last term of the Hamiltonian
the well-known Jaynes-Cumming model (JCM) of quantum optics. This model is completely integrable has ground state 0, with energy
(12,13). There are no direct transitions between + and modes. Figure 4 shows that resonance condition for Zeeman split and anti crossing in with presence magnetic field[15].
Eg =
Ez and exited energies 2
1 ( En , ) = n + n 2 2
with
(14)
n = 2 + 4 2 (n + 1) and detuning = Ez , corresponding to the eigenstates

( ( ( n , ) = cos n , ) n, + sin n , ) n + 1,
Figure 4.Resonance condition for Zeeman split The effect SOI on the spectrum is most strikingly seen in this anticrossing. From the last equation we see that the SOI has only non-zero matrix element for[15]
(15)
with tan
( n , )
= ( n ) /(2 n + 1) .

Our analysis differs in crusial respect: motivated by the analogy between parabilically confined SOI system and quantum optics that found in the previous section, we have derived an effectic model for the SOI in QDs. This model is formally
(n+ + 1, n , ) H SOI (n+ , n , ) , (n+ + 1, n , ) H SOI (n+ , n , ) (n+ , n + 1, ) H SOI (n+ , n , ) , (n+ , n + 1, ) H SOI (n+ , n , )
This marks different physical relevance of the terms in eq.(16,17). The term of the left
(16) (17)
correspond to SOI-induced transitions where an up spin is flipped downwards and the index of 102
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
modes
is exited by one additional quantum,
corresponding to a change of orbital state in QDs. Close to resonance magnetic field these transitions are energy conserving; the energy which gained from the spin flip is transferred into a orbital exitation. On the contrary, the terms on right correspond to transition where the spin and orbital are exited simultaneously, thus being energy nonconserving with respect to the total exitation.
[5]
Conclution
[6] In this paper we have considered a system single-electron QDs in the presence Rashba SOI. We showed that for parabolically confined nanostructures, it is helpful to map the underlying one-electron model onto a bosonic representation, which highlights the effects of spin-orbit coupling in con_ned systems and shows many similarities to models of quantum optics. We theretically demonstrate that the effective model is shown to be formally identical to the Jaynes.Cummings model (JCM) of atom-light interaction,and its integrability provides valuable insight into the coupling between spin and orbital degrees of freedom in the quantum dot. In comparison to the JCM of quantum optics, here the roles of atomic pseudo-spin and quantised light field are played by the spin and orbital angular momentum of the same electron.
[7] [8]
[9]
[10]
Acknowledgment
This work was financially supported by DP2M Directorate General of Higher Education with contract No. 015/SP2H/PP/DP2M/III/2008 Hibah Pekerti 2008.
[11]
[12]
References
[1] Thouless, D.J., 1977, Maximum Metallic Resistance in Thin Wires, Physical Review Letter 39 (18), 1167 D.W. Palmer, The semiconductorsinformation: www.semiconductors.co.uk, accessed 17 April 2003 Chakraborty, T. and Apalkov, V.M., 2003, Quantum cascade transition in nanostructures, Advanced in Physics 52 (5), 455 Tarucha, S., Honda, T., Austing, D.G., Tokura, Y., Muraki, K., [13]
[2]
[14] [15]
[3]
[4]
Oosterkamp, T.H., Janssen, J.W., and Kouwenhoven, L.P., 1998, Electronic states in quantum dot atoms and molecules, Physica E 3, 112 Tamura, H., Shiraishi, K., and Takayanagi, H., 2003, Quantum dot atoms, molecules, and superlattices, in Quantum Dots and Nanowires edited by Bandyopadhyay, S., dan Nalwa, H.S., American Scientific Publisher, USA Tarucha, S., Austing, D.G., Sasaki, S., Fujisawa, T., Tokura, Y., Elzerman, J.M., van der Wiel, W., de Franseschi, S., and Kouwenhoven, L.P., 2001, Novel Kondo anomaly in quantum dots, Materials Science and Engineering B84, 10 Ashoori, R.C., 1996, Electrons in artificial atoms, Nature 379, 413 Kouwenhoven, L.P., and Marcus, C., 1998, Quantum dots, Physics World June 1998 Abramowitz, M. and I. A. Stegun, I.A., 1972, Handbook of Mathematical Functions, Dover,New York Fock, V., 1928, Bemerkung zur Quantelung des harmonischen Oszillators im magnetfeld, Zeitschrift fur Physik 47, 446 Tarucha, S., Austing, D.G., Honda, T., van der Hage, R.J., and Kouwenhoven, L.P., 1996, Shell Filling and Spin Effect in a Few Electron Quantum Dot, Physical Review Letter 77 (17), 3613 Kouwenhoven, L.P., Austing, D.G., and Tarucha, S., 2001, Few-electron quantum dot, Report on Progress in Physics 64, 701 Datta, S., 1995, Electronic Transport in Mesoscopic Systems, Cambridge University Press, Cambridge, Chapter 1 Nan Zao., L.Zhong., jia-lin Zhu., and C.P. Sun.,2006, cond-mat/0603363 Stephan Deblad., 2005, Interaction and confinment in nanostructure: Spin orbit coupling and electron-phonon scattering, PhD dissertation Universitt Hamburg, Germany.
Proceeding Book
102

January 24, 2009 ISBN : 978 979 19201 0 0
Electronic Structures of Spherically Symmetric Wide-Band Quantum Dots in Direct Silicon Nanocrystals Confined by An Amorphous Insulator
Kamsul Abraha, Muhamad Darwis Umar, and Mirza Satriawan
Jurusan Fisika FMIPA UGM, Sekip Utara, Yogyakarta 55281 E-mail : kamsul@ugm.ac.id , kamsoel@yahoo.com , Mobile phone: 08122710165 Abstract
In this work we have calculated the electronic structure of spherical quantum dots (QDs) based on direct semiconductor of silicon nanocrystals. The weak conduction-valence band coupling in this class of direct semiconductors induces a weak mixing of the electronic states which requires a theoretical model to properly take into account this effect. This study was done within the framework of the k.p effective mass method. We have used a 6x6 Luttinger-Kohn (LK) Hamiltonian in order to include the weak admixture and to study sets of symmetry associated with the electronic states and their angular momenta in this central force problem which are orthogonal periodic functions (OPFs). The application of the k.p effective mass approximation on quantum dots with spherical symmetry was performed analytically and numerically by introducing the concept of two Hilbert space tensorial product or multiplication. This space was constructed by Kane bases and the Hilbert space constructed by OPF. The application of the k.p effective mass was successful in showing the dependence of the electron-hole energies on the size of the QDs. Keywords: wide-band QD, 6x6 LK Hamiltonian, OPF, silicon direct bandgap semiconductor
Introduction
Low-dimensional nanoscale systems, backed up by well-developed semiconductor technology, have been the subject of intensive study for the last 25 years [1]. Nanostructured materials may be defined as those materials whose structural elementsclusters, crystallites or moleculeshave dimensions in the 1 100 nm range. Much of the current academic and industrial interest in these materials over the past decade arises from the remarkable variations in electrical, optical and magnetic properties that occur as one progresses from an infinitely extended solid to a particle of material consisting of a countable number of atoms [2]. Modern semiconductor processing techniques allow the artificial creation of quantum confinement of only a few electrons. Such finite, quantum-size electronic systems have much in common with conventional atoms, yet they are man-made structures, designed and fabricated in the laboratory. They are usually called quantum dots (QDs), referring to their quantum confinement in all three spatial dimensions. Consisting of only a few hundred to a few hundred thousand atoms, QDs bridge the gap between the solid (bulk) state and single atoms, and hence these specks of matter
exhibit a mix of solid state and atomic properties with diameter ranging from 2 to 20 nm [3]. A common way to fabricate QDs is to restrict the two-dimensional electron gas in a semiconductor heterostructure laterally by electrostatic gates, or vertically by etching techniques. Nanoparticles can also be synthesized by colloidal chemistry, and therefore, they are often called colloidal or nanocrystal QDs (NQDs). Using this technique, their chemical composition, crystal structure and material form can be precisely controlled to suit either the needs of a certain scientific study or a specific technological application. Colloidal technique is usually related to the growth of monostructure nanocrystalline semiconductor in a certain form mixed with conductive polymers, semiconductors, sol-gels or even in porous thin films. Careful and precise control on its crystal growth has now paved the way for designing semiconductor structures with novelty playing pervasive roles in their fundamental physical as well as in other excellent properties which are closely related to their wide range of applications as optical and electrical devices [4] such as LEDs (light-emitting diodes), solar cells, lasers, and even biological fluorescence tags or labelling [3].
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
(a) Bulk Semiconductor (BS)
(b) Quantum Dot (QD)
Figure 1 (a) Continuous conduction band (CB) and valence band (VB) of a bulk semiconductor (BS) with a fixed energy gap [Eg(bulk)]. Electrons occupy all states up to the edge of the VB, whereas states in the CB are empty. (b) A quantum dot (QD) is characterized by discrete atomiclike states with energies determined by the QD radius R. Well-separated QD states are labeled with atomic-like notations (1S, 1P, and 1D). (c) The expression for the size dependence separation between the lowest electron and hole QD statesthe QD energy gap Eg(QD)obtained with the spherical quantum box model. (d) Continuous absorption spectrum of a BS (solid curved lines) and a discrete absorption spectrum of a QD (vertical bars). After Ref. [5]. One of the defining features of a semiconductor is the energy gap separating the conduction and valence energy bands (see Fig. 1). In semiconductors of macroscopic sizesbulk semiconductors (BSs)the gap width is a fixed characteristic parameter for the material itself (Fig. 1a). The situation changes, however, in the case of nanoscale semiconductor particles. This size range corresponds to the regime of quantum confinement, for which the spatial extent of the electronic wave function is comparable with the dot size. As a result of these geometrical constraints, electrons feel the presence of the particle boundaries and respond to changes in particle size by adjusting their energy. This phenomenon is known as the quantum-size effect which plays a very important role in QDs. In the first approximation, this effect can be described by a simple quantum box model [6], in which the electron motion is restricted in all three dimensions by impenetrable walls. For a spherical QD with radius R, this model predicts that a size-dependent contribution to the energy gap is simply
proportional to 1/R2, implying that the gap increases as the QD size decreases (Fig. 1c). In addition, quantum confinement leads to a collapse of the continuous energy bands of a BS into discrete, atomic-like energy levels (Fig. 1b). The discrete structure of energy states leads to a discrete absorption spectrum of QDs, which is in contrast to the continuous absorption spectrum of a bulk semiconductor (Fig.1d). Various physical theories and methods have been applied to monostructure as well as to heterostructure QD systems [4], some of which are tight-binding approximation [7], k.p effective mass approximation [8], ab-initio calculation [9], and empirical pseudopotensial method [10]. For a material of realistic dimension, k.p effective massempirical pseudopotensial method is one that has been widely used in QD modelling for determining its electronic structures (eigenfunctions and their corresponding eigenvalues) [4, 11-14]. The explosion in both academic and industrial interest in nanomaterials over the past decade arises from silicon as it has played considerable roles in microelectronics that has change the way people live and communicate [3]. Related to the novelty of QDs, silicon nanocrystals are of particular importance when it is present in a certain medium, for example, a silicon nanostructure sandwiched in an organic/anorganic insulator-semiconductor-conductor system. This mixing sets up future directions of silicon technology, such as mixing of silicon nanocrystals with some kind of polymers which leads to more flexible and durable solar cells of broader absorption spectrum. Much of the crystalline phenomena (opticelectronic, magnetic) found in semiconductors may be understood by partial inspection of their band structures, that is, their energy-momentum (E-k) relationships. Of particular importance is the region mostly occupied by charge carriers. This contains the lowest points in the conduction band (CB) and the highest points in the valence band (VB) known as -points k x = 0, k y = 0, k z = 0 in k-space
[15]. In most compound semiconductors, also called direct bandgap semiconductors, maxima of the VB and minima of the CB intersect at the same point in k-space or -point. For indirect bandgap semiconductors, on the contrary, minima of the CBs are located at some points in k-space and for this case, k.p theory allows computing the band structure E n (k ) near the band edge (at the CB lower edge and the VB upper edge) which may be applied for the cases of single as well as multiple degeneration. It should be noted that effective mass components will arise from the k.p formulation
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
which may be related to the experimental LuttingerKohn (LK) parameters [15]. In this paper we give a fuller theoretical account of the electronic structures of spherically symmetric wide-band QDs in direct silicon nanocrystals. Our main concern here is to determine the eigenfunctions and their corresponding eigenvalues for a more realistic model/case when the silicon nanocrystals are surrounded or potentially confined by an amorphous insulator. For simplicity, we may assume that the phase of the nanostructure resembles or is close to the bulk phase of its direct bandgap semiconductor. The main point of calculations is centered on the use of a 6x6 LK Hamiltonian in order to include the weak admixture of electronic states and the study of sets of symmetries associated with the electronic states and angular momentums which certainly involves orthogonal periodic functions (OPFs). In this case, the LK Hamiltonian treats the CB as a small perturbation. In the next section (Sec. II) we discuss the calculation methods and strategies used in this work including the potential model for spherically symmetric QDs, k.p representation and electronhole wavefunctions. In Sec. III we then present some calculation results, including the electronhole eigenfunctions and energy eigenvalues, interaction among electronic states and size dependence of the QD energy. Here, we would emphasize that the QD energy clearly decreases as the QD size increases. Calculations are presented and illustrated numerically using the experimental parameters of Si and GaAs crystals for use in the empirical pseudopotential method as well as in the Hamiltonian matrix elements of k.p effective mass [11, 16-17]. We finally present a general discussion and conclusions of this paper in Sec. IV.
Calculation Methods and Strategies

Mathematical model for the spherically symmetric QD In this work, confinement of spherically symmetric silicon nanocrystals in an amorphous insulator is simply modelled by introducing a potential of the form
V (r ) = 0 V (r ) =
rR r>R
(1)
Accuracy of this model depends on the size of the nanostructure. The larger the size the more accurate the assumption taken in a sense that the surface edge including its interface may be treated as a smooth interface like circles and straight lines. k.p Representation Within this approach, electronic structures (eigenfunctions and eigenvalues) of spherically symmetric wide-band QDs are obtained by using the 6x6 Luttinger-Kohn (LK) Hamiltonian, treating the CB as small perturbation, and the states close to valence and conduction bands are divided into two groups, say A and B. Of particular interest is the three uppermost doubly-degenerate valence bands: heavy hole (HH), light hole (LH) and split-off (SO) bands. These bands comprise group A whereas other bands including conduction bands comprise group B [4]. Therefore the LK Hamiltonian takes the form [15]
6 H LK6
PQ S R = 0 S 2 2R
S P+Q 0 R 2Q 3 2S
R 0 P+Q S 3 2S 2Q
0 R S PQ 2R S 2
2 2Q 3 2S 2R P 0
2 R HH 3 2 S LH 2Q LH S 2 HH SO 0 P SO
(2)
6 where the matrix elements of H LK6 are given by
h2 2 P= 1 k x2 + k y + k z2 2m0
Q=
(3a)
h2 2 2 k x2 + k y 2k z2 ( 2 m0
h2 2 3 2 k x2 k y + 2i 3 k x k y 2m0
R=
[ (
(3c)
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
S=
h2 2 3 3 (k x ik y ) k z 2 m0
(3d)
In Eq. (3), 1, 2, 3 are Luttinger parameters given by
1 =
2m0 ( A0 + 2 B0 ) 3h 2
1 sin = sin cos + cos cos r r sin x r 1 cos = sin sin + cos cos + y r r r sin 1 = cos sin z r r (6)
(4a) Electron and hole wavefunctions of spherically symmetric wide-band QDs in direct silicon nanocrystals As mentioned, nanostructure states confined in an insulator (with all possible interactions between the two) are plausibly perceived by the introduction of a macroscopic confining potential, such as given by Eq. (1), which is supposed to take part in the entire nanostructures. In view of the Helmholtz wave equation, it is clear that this potential would lead to a new particle quantization in addition to other quantization schemes arising from nuclear potentials, periodic crystal fields, internal spins and spin-orbit interactions. As a result, for instance, we now should deal with a new operator L = J + L for the new quantization and therefore the common eigenfunctions for both operator J and L are certainly as a product of eigenfuntion for each operator. What new in our work here is that, based on the noncommutative relation between the two operators, we have treated the aforementioned product of eigenfunctions as a tensorial produt of two Hilbert spaces. In doing this, bases for the resulting Hilbert space are considered as a tensorial product between bases related to the LK Hamiltonian and bases coming from the wave equation solutions which accommodate the confining spherical potential. Therefore, eigenfunctions for the QD nanocrystals can be expressed as a superposition of the combined bases. It is widely known that a set of orthogonal periodic functions (OPFs) nm"l " (r , , ) for a system with spherical symmetry are solutions of the Helmholtz wave equation
2 =
2m0 ( A0 B0 ) 3h 2
2 m0 C0 3h 2
(4b)
3 =
where
(4c)
h2 h2 A0 = + 2 2m0 2m0 B0 = h2 C0 = 2 m0
x p X pxX 0
E
y p X pyX 0
(5a)
h2 h2 + 2 2 m0 2 m0
(5b)
x y p X pxX + p X pyX
E 0 E
(5c)
Detailed derivation of the aforementioned Eqs. (2)(5) can be found in [15]. Transforming Hamiltonian from bulk to QD systems can be carried out by replacement of p h and wavenumber k with operator consequently the plane wave solution no longer exists in the system wave vector. In spatial representation, coordinate operator p h transforms into i [4, 11, 14]. Spherical coordinate system is suitable for spherically symmetric QDs in which the cartesian coordinates (x, y, z) would transform into the spherical coordinates (r, , ) as follows:
2 nm"l " + K 2 nm"l " = 0
(7)
nm"l "
For a sphere with boundary condition = 0 at r = R, OPF, nm"l " (r , , ) is
given by [14, 20]
nm"l " (r , , ) = nl " (r )Yl "m" ( , )
(8)
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
l" nl " (r ) = J l " n r , 0 r R R
(9)
1 d sin r l" J l " (r ) = ( 1) r l " r dr r

With
l"
(11)
Yl "m" ( , ) = l ",m" ( ) m" ( ) , 0 2 , and 0

(10)
l ",m" ( ) = ( 1)
m"
(2l"+1)(l" m")! P m" (cos ); l" 2(l"+ m")

ui
m Pl " " ( x ) = 1 x 2
m "/ 2
d m" Pl ( x ) dx m"
(12)
J, J Z
1 2 3 2 3 2 1 2 1 2 3 2
J ,J
Z +
and
1 im" m" ( ) = e 2
U1 (13) U3 U5 U7 U2 (14) U4 U6 U8
1 2
iS
2 3
,1 2 ,3 2 ,1 2
It may be noted that the boundary conditions on the Helmholtz equation (7) gives
1 3
(X + iY ) 1 (X + iY ) 2 (X + iY ) + 13 Z
1 6
l" K = n R
2
, 1 2 , 1 2
1 6
iS
In Eq. (11), J l " (r ) = ( 1) r

l" l"
1 d sin r r dr r
l"
3 2 1 2
, 3 2 , 1 2
(X iY ) + 2 Z 3 1 (X iY ) 2 1 (X - iY ) + 13 Z 3
l is a spherical Bessel function [18], where n" is the l nth root of the l-Bessel function (for J l " n" = 0 ). The number n is determined by the QD size in that the larger the size the bigger n. Each root of l" J l n r is associated with the quantum numbers R (n, l) for which each corresponding energy level would be (2m+1)-fold in m. Here,
( )
Results and discussions

Electron-hole eigenfunctions and eigenvalues In a case where electrons are outside of the nuclear potential and do not interact with those in the valence band, they can be treated as a system with effective masses due to the influence of periodic crystal fields and therefore the Schrdinger equation for those electrons should be in the form of
h2 2 r , , + V p (r ) (r , , ) , = E (r , , ) , 2m
m Pl " " (x ) = 1 x 2
m"/ 2
d m" Pl " ( x ) is a Legendre dx m"
function [19]. Finally, bases for the 6x6 LuttingerKohn Hamiltonian [4] are given as follows:
(15) where
(r , , ) =
n ,l ", m"
nm"l "
nm"l " (r , , )
(16)
It is straightforward to show that substitution of Eq. (16) into Eq. (15) then multiplied by the complex conjugate of
nm"l " ,
that is
n '(m")'(l ")' ,
and integration over the whole volume of the symmetric QD would give rise to the following matrix equation:
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
H nm"l "n '(m")'(l ")' EA nn ' m "( m " ) l "(l " )' = 0
If
(17) orthonormal,
the diagonal matrix the form of
H nm"l "n '(m")'(l ")' would then take
nm"l "
and
n '( m" )'(l " )'
are
A nn ' m"( m " ) l "(l " )' is an identity matrix, and elements of
h2 2 E nl " = n '(m")'(l ")' r , , nm"l " r + n '(m")'(l ")'V p (r ) nm"l " r sin dr d 2m QD h2 = K2 +0 2m
where QD is the QD volume. Therefore, the energy eigenvalues are given by
(18)
The time independent Schrdinger equation for holes in the spherically symmetric wide band quantum dots is given by
E nl "
h = 2m R
2
l" n
(H
(19)
6 x6 LK
+ V p (r ) lm (r , , ) = Elm lm (r , , )
(21)
where
with corresponding eigenfunctions given by
lm (r , , ) =
(r , , ) =
l" J l" n R n , m",l "
r Yl "m" ( , ) ,
n , j ,l ",l , m
Anjl "lm njl "lm (r , , )

(22)
(20)
As noted, the tensorial product of two Hilbert spaces gives rise to
which is known as the envelope spinor function as it comes as a eigenfunction for spin and total angular momentum of the system [19]. The
3,3 2 2 3 1 2,2 l" 3 , 1 njl "lm (r , , ) = C ( jl", m' m"| lm ) J l " n r Yl "m" ( , ) 2 2 (23) R 3 3 2 , 2 1,1 12 21 2 , 2 ( jl"; m' m"| lm ) in Eq. (23) are coefficients C
conventional Clebsch-Gordan coefficients [20]:
l j l" C ( jl"; m' m"| lm ) = ( 1) j l "+ m (2l + 1)1 / 2 m' m" m j l" l 12 m' m" m = [( j m')!(l" m")!(l m )!( j + m')!(l"+ m")!(l + m )!]
( j + l"l )!( j l"+l )!( j + l"+l )! m'+ m",m ( 1) ( j + l"+l + 1)! k ( 1) k!( j + l"l k )!( j m' k )!(l"+ m"k )!(l l"+ m'+ k )!(l j m"+ k )! k
j l "+ m 12
(24)
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
max (0, l"l m' , j l + m") k min ( j + l"l , j m' , l"+ m")
As usual, indices k in Eq. (24) should be in the range Therefore
(25)
It should be pointed out here that for the nondiagonal Hamiltonian, Anm"l "n '(m")(l ")' can be ' seen as a diagonalizing matrix for the following matrix
3,3 2 2 3 1 2,2 l" l " 3 , 1 lm (r , , ) = Anjl "lm C ( jl" , m' m"| lm ) J l " n r Yl "m" ( , ) 2 2 (26) 3 3 R n ,l " m = l " 2 , 2 1,1 12 21 2 , 2 66 n'(m")'(l")' H LK njl"lm (r , , ) r sin d d
QD
(27) In brief, a framework for determining the hole energy eigenvalues and eigenfunctions of the wide band quantum dots in direct silicon nanocrystals is summarized in form of a flowchart depicted in Fig. 2.
Start
n, l" , m" , H
QD
nl "m"
H nl "m" d
Matrix Diagonalization
A jl "m 'm"lm
C jl" , m' m" lm
lm , E nl
Stop
FIG. 2. Flow chart for determining eigenvalues and eigenfunctions of the wide band quantum dots in direct silicon nanocrystals
Parameters required in the numerical computation are given in Table 1.
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
Table 1. Experimental parameters of Si and GaAs crystals for use in the empirical pseudopotential method as well as in the Hamiltonian matrix elements of k.p effective mass 1 2 3 4 5 6
[100 mlh ] [100 mhh ] [111 mlh ] [111 mhh ]
11 =
22
1 3 1 1 1 1 J 1 (r ) cos( ) ' + J 1 (r ) cos( ) ' 2 2 2 2 2 2

3 1 1 1 1 J 1 (r ) cos( ) ' + J 1 (r ) cos( ) ' 2 2 2 2 2 2 10 1
0.167a 0.274a 0.097a 0.681a 0.046c 1.06b
31 =
22
1
3
1 2 2
1 3 1 1 1 1 = J 1 (r ) cos( ) ' + J 1 (r ) cos( ) ' 2 2 2 2 2 2

= 1 2 10 3 1 1 1 1 J 1 (r ) cos( ) ' + J 1 (r ) cos( ) ' 2 2 2 2 2
m 0 (E e )
E g (eV )
1 2 2
(direct) 7 8
a c
(mV )
0.33c -0.56c
(29) and, again, with the same energy eigenvalue of Elm = 3.724eV . Interactions among different bases or states
Ref. [17], b Ref. [16], Ref. [11] suitable for GaAs
We realize that the off-diagonal elements of the Hamiltonian matrix H LK as indicated in Eq. (2) are related to interactions among different bases or states/bands. It should be carefully remarked up to this point that the elements of the diagonal matrix obtained from Eq. (27) do not automatically vanish, but what we find in our calculation is the fact that its off-diagonal elements clearly have limiting null values which means that indicated interactions among different bands are relatively very weak compared with their counterparts in the diagonal elements. As a consequence, the coefficients A nn ' m "( m " ) l "( l " )' in the linear combination given by Eq. (17) automatically constitute an identity matrix. Electron and hole energies in spherically symmetric QDs Our calculation results show that energy eigenvalues for our spherically symmetric quantum dots (QDs) are all negative which can be explained as follows. We start the calculation by adopting a plausible assumption in that we can partly use the available experimental parameters of GaAs (28) indicating that the maximum energy of its valence band is zero. This means that in the absence of confinement, states of the outmost electrons (valence electrons) in spherically symmetric QDs have null energies, as indicated by the null superposition of energy components in the Hamiltonian. Conduction electrons, in contrast, which have translational momentums in periodic potentials, would give positive energies whereas 2
6 x6
The following is ten eigenfunctions obtained as a result of numerical computation for quantum numbers n = 1, l =1, m = 0, and QD radius of R = 2.5 nm (detailed calculations are presented in [15]). The first six is given by
3 3 3 J 1 (r ) cos( ) ' 2 2 40 22 3 3 3 3 3 = J 1 (r ) cos( ) ' 2 2 40 2 2 1 3 3 53 = J 1 (r ) cos( ) ' 2 2 10 22 1 3 3 5 3 = J 1 (r ) cos( ) ' 2 2 10 2 2
33 =
51 =
22
3 3 1 J 1 (r ) cos( ) ' 20 2 2 3 3 1 J 1 (r ) cos( ) ' 20 2 2

(28)
1 2 2
where all eigenfunctions have the same energy eigenvalue, that is, Elm = 1.697eV . The other four eigenfunctions are given by
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
electrons in the inner region of valence bands would give negative energies indicating the fact that these electrons are in bound states or are in the influence of nuclear potentials of the atoms. To conclude, negative QD energies obtained from this work indicate that electron-hole systems are in both nuclear and macroscopic confining potentials, and they cannot move out of the QD system. This result also shows that, due to the macroscopic confinement as modelled by Eq. (1), atomic conduction states in spherically symmetric QDs are truly bound states as expected. Size dependence As before, our calculation results in the fact that electron-hole energies for the spherically symmetric QDs increase with respect to the quantum numbers l and n. This occurs because increasing l is equivalent with increasing energy/wave function density whereas n is related to a distance inversely proportional to interaction energy. Moreover, we find that the system energy decreases as the system size increases because, as expected, increasing size leads to increasing n which determines the interaction distance.
References
[1] Reimann, S. M.and Manninen, M., 2002, Electronic structure of quantum dots, Rev. Mod. Phys. 74, 1283 [2] Li, S., 2004, Materials Design from ab initio Calculations, PhD thesis, Uppsala University, Sweden [3] Tews, M., 2004, Structure and Transport Properties of Quantum Dots, PhD dissertation, Universitt Hamburg, Germany [4] North, S. M., 2001, Electronic Structure of GaSb/GaAs and Si/Ge, PhD thesis, University of Newcastle, England [5] Klimov, V.I., 2003, Nanocrystal quantum dots, from fundamental photophysics to multicolor lasing, Los Alamos Science 28, 214 [6] Efros, Al. L. and Efros, A.L., 1982, Pioneering Effort I , Sov. Phys. Semicond. 16, 772 [7] Schulman, J. N. and Chang, Y. C., 1985, Band mixing in semiconductor superlattices, Phys. Rev. B 31, 2056 [8] Wang, L.W. and Zunger, A., 1996, Pseudopotential-based multiband kp method for 250000-atom nanostructure systems, Phys. Rev. B 54(16), 11417 [9] Jones, R.,1988, The phonon spectrum of diamond derived from ab initio local density functional calculations on atomic clusters, J. Phys. C 21, 5735 [10] Gell, M. A, Ninno, D., Jaros, M., and Herbert D. C.,1986, Zone folding, morphogenesis of charge densities, and the role of periodicity in GaAs-AlxGa1-xAs (001) superlattices, Phys. Rev. B 34, 2416 [11] Fonoberov, V., 2002, Electronic and Optical Properties of Semiconductor Quantum Wires and Dots, PhD thesis, University of Moldova, Moldova [12] Grigoryan, G. B, Kazaryan, E, Efros, Al. L, and Yazeva, T. V., 1990, Quantization of hole and absorption edge in spherical microcrystal of semiconductor with complex structure of valence band, Sov. Phys. Solid State 32, 1031 [13] Ekimov, A.I, Hache, F., Schanne-Klein, M. C., Ricard, D., Flitzanis, C., Kudryavtzev, I. A., Yazeva, T. V., Rodina, A. V., and Efros, Al. L., 1993, Absorption and IntensityDependent Photoluminescence Measurement on CdSe Quantum Dots: Assignment of the First Electronic Transitions, J. Opt. Soc. Am. B 10, 100 [14] Prado, S. J., Marques, G. E., and TralleroGiner, C., 1999, Electronic Structure in Narrow-Gap Quantum Dots, Brazilian Journal of Physics 29, 4 3
Conclusions
In this paper we have given a detailed theoretical discussion of the electronic structures for wide-band quantum dots of spherical symmetry in direct silicon nanocrystals. Our main concern here has been to determine the eigenfunctions and their corresponding eigenvalues for a more realistic case in which the silicon nanocrystals are modeled to be surrounded or potentially confined by an amorphous insulator. The following remarks are of particular importance: (a) it is clear in this work that the tensor product of two Hilbert spaces may be applied to determine an appropriate Hilbert space in which the symmetric quantum dot Hamiltonian should be properly defined and manipulated, (b) the effective mass k.p method may be used in gaining the electronic structures, that is eigenfunctions and energy eigenvalues, of wide band quantum dots with spherical symmetry in silicon nanocrystals. Some of the important findings related to the application of the aforementioned method are: first, electron-hole energies in spherically symmetric quantum dots show a clear dependence on the size of quantum dots as expected, second, the off-diagonal elements of the LK Hamiltonian with limiting null values lead to the conclusion that interactions among different bands are relatively very weak compared with their counterparts in the diagonal elements.
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
[15] Umar, M. D., 2007, Kajian Struktur Elektronik Quantum Dot Simetris dalam Nanokristal Silikon dengan Pendekatan Massa Efektif k.p dan Ikatan Kuat (Tight-Binding), Tesis S2, Jurusan Ilmu Fisika, Sekolah Pascasarjana, Fakultas Matematika dan Ilmu Pengetahuan Alam, Universitas Gadjah Mada, Yogyakarta. [16] Rio, S. R. and Iida, M., 1999, Fisika dan Teknologi Semikonduktor, PT Pradya Paramita, Jakarta. [17] Trani, F., 2004, Electronic and Optical Properties of Silicon Nanocrystals: a Tight Binding Study, PhD dissertation, Universitarion Monte S. Angelo, Napoly, Italy. [18] Arfken, G. B. and Weber, H. J., 2001, Mathematical Methods for Physicists: Fifth Edition, Harcourt Science and Technology Company, USA. [19] Johnson, W. R., 2006, Lectures on Atomic Physics, Department of Physics, University of Notre Dame, Notre Dame, Indiana, USA. [20] Cowan, R.D., 1981, Theory of Atomic Structure and Spectra, 4th ed., Clarendon Press, Oxford.
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
Theory of Far-Infrared Optics of the Giant-Dielectric-Constant Material CaCu3Ti4O12

Kamsul Abraha, Isom Hilmi, Sudarmono
Physics Department, Gadjah Mada University, Yogyakarta 55281, Indonesia .E-mail : kamsul@ugm.ac.id , kamsoel@yahoo.com , Mobile phone: 08122710165
Abstract
The recently discovered high-dielectric-constant material CaCu3Ti4O12 has shown numerous technological applications including the fabrications of microelectronic and optoelectronic devices, but the demanding level of device miniaturization critically depends on the physical origin of their unusual giant dielectric response which are still unsolvable. Benefitted from our previous experience in the far-infrared optics of functional magnetic materials, we thus present detailed theoretical study for the surface phonon-polaritons (SPPs) and attenuated total reflection (ATR) spectra in the semi-infinite CaCu3Ti4O12 sample. This study includes the derivation for the dispersion relations using Maxwell equations and corresponding boundary conditions applying for both tangential and normal components of the electric and magnetic fields across the air-sample interface. The SPP dispersion curves are then computed numerically using suitable parameters for the ceramic sample. It is carefully remarked from the computed dispersion curves that the SPPs only propagate in TM (Transverse Magnetic) modes and are reciprocal in the sense that the change of SPP propagating wave vector direction does not change its mode frequency. We also find that there is no SPP but bulk or volume phonon-polariton which propagates in the TE (Transverse Electric) mode. Moreover, the computed ATR spectra using Si prism in the Otto configuration indicates that the SPPs can be optimally detectable when the gap distance between the prism and the sample reaches an optimum value for the ceramic sample CaCu3Ti4O12. Finally, the invented surface modes are also clearly verified in the computed ATR spectra, which are in excellent agreement with the related SPP dispersion relations. Keywords: giant-dielectric-constant material, surface phonon-polaritons, attenuated total reflection
Introduction
High dielectric constant materials are of increasing interest because of their importance in the miniaturization of microelectronic devices, including dynamic random access memory (RAM). This has led to a quest among the materials science community to improve upon the dielectric properties of the existing ferroelectric or relaxor materials such as (Ba,Sr)TiO3 and PbTiO3 [1-3]. However, both types of material show a peak in the dielectric constant as a function of temperature, making it undesirable from the electronic device applications view point because a strong temperature dependence may cause the related device to fail with temperature variations in the environment. Recently there has been considerable interest in the dielectric behaviour of the cubic perovskite-related CaCu3Ti4O12 (CCTO). This compound in ceramic and single crystalline forms was discovered to have an extraordinarily high dielectric response with relative permittivity up to 105 at room temperature, which is practically frequency independent between DC and 106 Hz for the temperature range between 100 and 600 K. The dielectric permittivity abruptly drops down to a value ~ 100 upon lowering the temperature below
100 K. The dielectric constant ~ 100 is also seen at frequencies higher than 106 Hz [4-5]. Apart from the giant values of the dielectric constant, the independence of dielectric properties with frequency and temperature along with the fact that no ferroelectric lattice distortion has been observed certainly pose some interesting questions: why does the material can have such a gigantic dielectric constant and why does the constant decrease dramatically below 100 K. It is therefore of valuable interest both from a fundamental and an application point of view to understand the mechanism underlying the bahaviour of CCTO ceramics and single crystals. Numerous scenarios have been proposed to uncover the origin of the giant dielectric response in CCTO, including both intrinsic [6] and extrinsic mechanisms [7-9]. We will also present our results on these issues elsewhere in the Proceedings of this Seminar. The observation of the temperaturedependent reflectance of CCTO in the far-infrared (FIR) region has been reported by Homes et al. [6]. The observed spectra are typical of a nonmetallic system with their sharp features are due to phononactive modes. These FIR-active modes were well fitted and reproduced using the calculation based on the frequency( ' )-dependent complex dielectric % function ( ') for Lorentz oscillators 1
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
~ ( ) = +
j
2, j p 2 2 i j j
and
(1)
where
j, j,
p, j
are the frequency,
width, and effective plasma frequency of the jth vibration. is the core contribution to the dielectric function at high frequencies [6]. The phonon parameters required in the Lorentz oscillators are tabulated in Table 1. It is worth mentioning here that in Eq. (1) we assign
for the components of electric and magnetic fields across the material-air interface [21-22]. The dispersion curves are then obtained numerically using suitable parameters for the sample as depicted in Table 1. Section 3 is concerned specifically with the calculated ATR spectra for the material. The calculations are presented and illustrated numerically using the known parameters of CaCu3Ti4O12 dielectric material. We finally present a general discussion and conclusions of this paper in section 4. The surface phonon-polariton dispersion relations for CCTO With the system of coordinates shown in Fig.1, we first consider surface phonon-polaritons propagating along the interface y = 0 between vacuum (y < 0) and the CCTO material (y > 0) following the usual derivation [12, 21-22]. This means first finding solutions of Maxwells equations that decay away from the interface and second applying the electromagnetic boundary conditions at the interface y = 0 to find the required dispersion relation. In the CCTO crystal the dynamic field H satisfies
that can be expressed in units of cm-1. This function represents the response of the material to a driving electromagnetic field at FIR frequency [10-11]. However, the reported spectra [6] were not explicitly related to the contribution of phononpolaritons, neither by bulk nor surface modes. Therefore we will argue below that the detailed study particularly on their surface phononpolaritons is still desired for fundamental and practical research. Table 1. The phonon parameters for Lorentz oscillators of CCTO ceramic at 10 K
2c
as widely adopted in spectroscopy so
r r 1 2H H () 2 2 = 0 c t
2
(2)
and 295 K, where
j , j ,and p, j
are the frequency, width, and effective plasma frequency of the jth vibration. (All quantities are in cm-1 ). After Ref. [6]. As motivated and inspired by our previous experience in FIR spectroscopy of magnetic materials [12-19], we take the view here that the full theoretical discussion for the surface phononpolaritons (SPPs) and attenuated total reflection (ATR) spectra in the semi-infinite dielectric CCTO samples [20] is still required. In this work we therefore report our full theoretical studies of the material. The remainder of the paper is organized as follows. In the next section we derive and discuss the surface phonon-polariton dispersion curves. These include the derivation for the dispersion relations obtained through celebrated Maxwell equations and boundary conditions that are applied
Figure 1. Geometry, axes and notation for calculation of surface phononpolaritons in the dielectric CCTO. The propagation vector is along the x axis and perpendicular to the c (z) axis (Voigt geometry) The surface polariton fields, with amplitudes that decay with distance from the interface and propagating parallel to the y axis as a plane wave in the x direction, can be written
r r H = H1 exp(iqx + 0 y) exp(i t ) ; y < 0 (3)

r r H = H 1 exp(iqx y ) exp( i t )
; y > 0 (4) 2
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
where q is the wave vector of the surface phononpolaritons and 0 and are the spatial decay constants in vacuum and the medium of dielectric constant , respectively. By substituting Eq. (3) into Eq. (2) it can be shown that the decay constant 0 is simply given by
By applying the boundary conditions for normal and tangential components of (E, H) fields at the interface y = 0, we get the relation
( ) 0 + = 0 ,
(8)
02 = q 2
'2
c2
(5)
After substituting Eq. (4) into Eq. (2) we have a set of three homogeneous linear equations for the components H1x, H1y, and H1z in the medium, namely
which is simply the implicit dispersion relation relating the magnitude of the wave vector ( q = |q| ) with frequency ' of the surface phonon-polaritons (SPPs). As can be seen, Eq. (8) implies reciprocal propagation of the surface mode. That is, changing the sign of q will not change the frequency ' :
' ( q ) = ' ( q ) .
H1x 2 2 2 q + ( ) 2 H1 y = 0 c H1z
(6)
which means that the SPP is found to be TM or p polarized with the only nonvanishing component of magnetic field E along the z axis. The spatial decay constant in the medium is then given by
In Figs. 2(a-c) we plot dispersion curves for CCTO sample obtained numerically from Eq. (8), using the suitable parameters as quoted in Table 1. In order to show real ' versus real q, we plot the curves without damping. In all cases, the SSP starts from the vacuum light line. As |q| , all surface mode curves approach the electrostatic limit. Both bulk and surface modes are markedly reciprocal in q. We also plot the attenuated-total-reflection (ATR) scan lines explained in the caption of Fig. 2.
= q2
2 ( )
c2
(7)
(a)
(b)
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
(c) Figure 2. Computed dispersion curves for bulk (shaded regions) and surface p-polarized (solid lines, sp) phonon-polaritons on CaCu3Ti4O12 for frequency range (a) 50 s/d 200 cm-1, (b) 200 s/d 420 cm-1, and (c) 420 s/d 800 cm-1. Label L indicates the vacuum light line q = line is specified by q =
k , and the ATR scan = 2 2 c
k 1 = p 2 sin with p = 11,56 (Si) and = 30o, 45o, and 60o. 2 c

R= r =
2
Calculation of the attenuated total reflection (ATR) for CCTO material As indicated in Sec. 2, the CaCu3Ti4O12 (CCTO) surface polariton modes can only couple to p polarized light in an ATR experiment. Here, we consider an ATR arrangement in the Otto configuration, as already illustrated elsewhere [2122]. Thus in view of this polarization feature, we will consider a three-layer system consisting of prism, air gap and CCTO sample. We choose the z axis normal to the interfaces and the yz plane as the plane of incidence. The air gap extends from z = - d to z = 0. If the angle of incidence exceeds the critical angle for total internal reflection, the incident wave generates an evanescent wave in the air gap above the sample with the parallel wave vector
k (1 + f exp(2d )) i (1 f exp(2d ) k (1 + f exp(2d )) + i (1 f exp(2d )

g s ( )k y k y , f = ( )k g + k s y y
(9)
Where (10)
2 k 4 = ( ) , c k xp = k // = 1 / 2 sin , p c p 1/ 2 k y = k = p cos , c 1/ 2 , 0 = p sin 2 0

c
1/ 2
(11) (12) (13) (14) ,

1/ 2
k =
g ky =
p1/ 2 sin . If the wave vector and
( c
p sin 2
1/ 2
(15)
and
s ky
frequency of this wave are close to those of the surface polaritons, then coupling occurs and, as a result, there may be a sharp dip in reflectivity which can be related to the excitation of a surface polariton by the EM power from the evanescent wave. A reduction in the ATR reflectivity occurs at the mode frequency. The detailed calculation is given elsewhere [12, 21-22] and therefore the ATR reflectivity (R) is given by
2 2 = k 4 p sin 2 c
(16)
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
Figure 4 shows p-polarized theoretical ATR spectra of CCTO at varying angle of incidence in a Si prism, damping = 10-2 cm-1 and d = 10-5 cm. Again, three spectra were plotted for displaying the effect of varying with = 30o, 45o and 60o. As markedly displayed in Fig. 2, varying causes a slight shift in the resonance frequency of the mode.
Conclusions
As motivated and inspired by our previous experience on FIR magnetic spectroscopy, in this paper we have presented full theoretical discussions for the surface phonon-polaritons (SPPs) and attenuated total reflection (ATR) spectra in the semi-infinite dielectric CaCu3Ti4O12 (CCTO) sample. These include the derivation for the dispersion relations obtained through Maxwell equations and boundary conditions that are applied for the components of electric and magnetic fields across the sample-air interface. The dispersion curves have been obtained numerically using suitable parameters for the sample. We have presented the theoretical study of the polariton dispersion curves on the surface of CCTO with no external field when the direction of wave propagation is perpendicular to the easy axis. This gives a theoretical description of the sample surface and bulk modes at frequency range 0 800 cm-1. It is found that the SPP only propagates in TM (Transverse Magnetic) modes and is reciprocal in the sense that the change of wave vector direction does not change the frequency, namely ( q ) = ( q ) , whereas there is no SP
Figure 3. The computed ATR spectra in p polarization for surface polaritons of CCTO, given by Eq. (9), whose dispersion curves are obtained from Eq. (8). Parameters are given p = 11.56 (Si), damping = 10-2 cm-1. Three curves are drawn for three gap distance d = 10-3, 10-4 and 10-5 cm and angle of incidence = 60o.
Figure 4. The computed ATR spectra in p polarization for surface polaritons of CCTO, given by Eq. (9), whose dispersion curves are shown in Fig. 2. Parameters are given by d = 10-5 cm, p = 11.56 (Si), damping = 0.01 cm-1. Three curves are drawn for three angles of incidence = 30o, 45o and 60o (associated with three scan lines in Fig. 2) In Fig. 3 we present theoretical ATR spectra at an angle of incidence = 60o in a Si prism and damping = 10-2 cm-1. To display the effect of varying d, three spectra were plotted with d = 10=3, 10-4 and 10-5 cm. As d decreases, the ATR dip is sharpened and its depth increases approaching the value R 0.
that propagates in TE (Transverse Electric) modes. The computation of ATR reflectivity shows that the SPP can be detected when the gap/distance between the prism and the sample reaches an optimum value of 10-5 cm. The surface modes are also verified in the spectra, which are in agreement with the related dispersion relations.
References
[1] Kim, B. G., Cho, S. M., Kim, T. Y. and Jang, H. M. (2001) Giant dielectric permittivity observed in Pb-based perovskite ferroelecrics, Phys. Rev. Lett. 86, 3404 3406 Dressel, M. and Grner, G. (2002) Electrodynamics of solids: optical properties of electrons in matter, Cambridge University Press, Cambridge, UK Kittel C. (2005) Introduction to solid state physics, 8th ed., John Wiley & Sons, USA Ramirez, A.P., Subramanian, M. A., Gardel, M., Blumberg, G., Li, D., Vogt, T.and Shapiro, S.M. (2000) Giant dielectric
[2]
[3] [4]
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
[5]
[6]
[7]
[8]
[9]
[10] [11] [12]
[13]
[14]
[15]
[16]
[17]
constant response in a copper-titanate, Solid State Commun. 115, 217-220 Subramanian, M. A., Li, D., Duan, N., Reisner, B. A. and Sleight, A. W. (2000) High dielectric constant in ACu3Ti4O12 and ACu3Ti3FeO12 phases, Journal of Solid State Chemistry 151, 323 - 325 Homes, C. C., Vogt, T., Shapiro, S. M, Wakimoto, S., Subramanian, M. A. and Ramirez, A. P. (2003) Charge transfer in the high dielectric constant materials CaCu3Ti4O12 and CdCu3Ti4O12 , Phys. Rev. B 67, 092106 - 092109 He, L., Neaton, J. B., Cohen, M. H. and Vanderbilt, D. (2002) First-principles study of the structure and lattice dielectric response of CaCu3Ti4O12, Phys. Rev. B 65, 214112 - 214122 Sinclair, D. C., Adams, T. B., Morrison, F. D. and West, A. R. (2002) CaCu3Ti4O12: One-step internal barrier layer capacitor, Appl. Phys. Lett. B 80, 2153 - 2155 Lunkenheimer, P., Bobnar, V., Pronin, A. V., Ritus, A. J., Volkov, A. A. and Loidl, A. (2002) Origin of apparent colossal dielectric constants, Phys. Rev. B 66, 052105 052108 Landau L.D. and Lifshitz E.M. (1984) Electrodynamics of continuous media, 2nd ed., Pergamon Press, Oxford, UK Grosso, G and Parravicini, G.P. (2006) Solid state physics, Elsevier (Singapore) Kamsul Abraha (1995),Theory of Surface Polaritons and Far-Infrared Reflectivity of Antiferromagnets, Rare Earth Metals and Ferrimagnets PhD thesis, University of Essex, England, UK Kamsul Abraha and D.R.Tilley (1996),Theory of far infrared properties of magnetic surfaces, films and superlattices Surf. Sci. Rep. 24, 125-222 Kamsul Abraha and D.R.Tilley (1994),Theory of far-infrared reflectivity and surface magnetic polaritons of rare earth magnets Infrared Phys. Technol. 35, 681699 T.Dumelow, R.E.Camley, Kamsul Abraha, and D.R.Tilley (1998),Nonreciprocal phase behavior in reflection of electromagnetic waves from magnetic materials, Phys.Rev. B 58, 897908 Kamsul Abraha, D.E.Brown, T.Dumelow, T.J.Parker, and D.R. Tilley (1994), Oblique-incidence Far-infrared Reflectivity Study of the Uniaxial Antiferromagnet FeF2, Phys. Rev. B 50, 6808-6816 Kamsul Abraha and D.R.Tilley (1995),The theory of far-infrared optics of
[18]
[19]
[20]
[21]
[22]
layered antiferromagnets, J.Phys.Condens.Matter 7, 2717-2730 Kamsul Abraha, S.R.P.Smith, and D.R. Tilley (1995),Surface Polaritons and Attenuated Total Reflection Spectra of Layered Antiferromagnets in the Faraday Configuration, J. Phys.: Condens. Matter 7, 6423-6436 M.R.F.Jensen, T.J. Parker, Kamsul Abraha, and D.R. Tilley (1995), Experimental Observation of Magnetic Surface Polaritons in FeF2 by Attenuated Total Reflection, Phys. Rev. Lett. 75, 37563759 Sudarmono (2006) Kajian teoritis ragam polariton permukaan material baru CaCu3Ti4O12 (CCTO), Tesis S2, Sekolah Pasca Sarjana UGM M.G.Cottam and D.R.Tilley (1989), Introduction to Surface and Superlattice Excitations, Cambridge University Press, Cambridge G.C.Aers and A.D.Boardman (1980),Attenuated Total Reflection Analysis of Surface Polaritons in Physics Programs, editor: A.D. Boardman, John Willey & Sons, New York
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
Nixtric Oxide Detection using Wavelength Modulation Spectroscopy and Its Applications in Breath and Cell Death Diagnostics
Mitrayana1, T. Pierera2, S. Persijn2, S.M. Cristescu 2, F.J.M. Harren2, M.A.J. Wasono1, and W. Rochmah3
2
Atomic and Nuclear Physics Lab. Physics Dept. Gadjah Mada University Indonesia Molecular and Laser Physics Dept. Radboud University of Nijmegen, the Netherlands 3 Medical Faculty of Gadjah Mada University Indonesia mitrayana@ugm.ac.id Abstract
We used a thermoelectrically cooled, continuous-wave, quantum cascade laser operating between 1891.3 and 1907.6 cm1 in combination with wavelength modulation spectroscopy for the detection of nitric oxide at sub-part-per-billion volume (ppbv). The laser emission overlaps with the P7,5 doublet of NO centered around 1903.9 cm1. Using an astigmatic multiple pass absorption cell with an optical path length of 76 m we achieved a detection limit of 0.8 ppbv at 100 mbar, with a total acquisition time of 30 seconds. The corresponding minimal detectable absorption is 3.5 x 10 10 cm1Hz1/2. We used that detector to detect the concentration of NO gas from the sample air from the breath and algae cell-death. Keywords: wavelength modulation spectrosocopy, quantum cascade laser, Nitrix oxide
Introduction
The improvement of compact optical sensors for nitric oxide (NO) detection is of importance for a number of applications, such as environmental monitoring [1], atmospheric chemistry [2], industrial process control [3], combustion studies [4], and medical diagnostics [5,6]. NO is involved in many vital physiological processes in the human body. For example, an elevated level of NO in exhaled air is correlated with airway inflammation in asthmatic patients. Knowledge of NO concentrations in the exhaled breath of these patients may allow health-care providers to adjust therapeutic drug dosages [7,8]. For medical diagnostic purposes, it is essential to time resolve the NO concentration as a function of a breathcycle phase, because the corresponding air samples originate in different parts of the respiratory tract. This application requires a sensor response time of 1 s and a NO minimum detection sensitivity of < 1 ppbv. Such high sensitivity, rapid-response measurements are possible with laser absorption spectroscopy in the fundamental absorption band of NO. Distributed feedback quantum cascade lasers (DFB QCLs) operating in a pulsed or continuouswave (cw) mode are promising laser sources because of their narrow line widths, single-mode operation, tunability, power, reliability, and compactness. Only recently continuous wave (cw) QCLs able to operate single mode at room temperature are avaialbe [9]. Single mode operation is a very important step forward for molecular spectroscopy, because of the narrow absorption
lines down to tens of MHz. Since they are thermoelectrically cooled, these single mode cwQCLs do not require liquid nitrogen cooling and therefore, are very well suited to be utilized a compact and high-sensitivity trace gas sensor [10]. A number of approaches utilizing QCLs for the optical sensing of trace NO have been reported. Background-free be capable of measuring biogenic nitric oxide with a sensitivity of a few ppbv [11,12]. Mid-infrared spectrometers achieve a minimum NO detection limit 1 ppbv [13]. A technique based on cavity ring-down spectroscopy (CRDS) reaches a noise equivalent sensitivity at the sub-ppbv level in several seconds using a comparatively small sample volume because of a long optical path length obtained with ultra-high-reflectivity mirrors ( R ~ 99.99%) [14]. In this paper we show the capability of the single mode thermoelectrically cooled cw-QCL for molecular absorption spectroscopy. The laser is used in combination with wavelength modulation spectroscopy for the detection of nitric oxide (NO) at the part per billion volume (ppbv) mixing ratio.
Experimental set-up
A schematic arrangement of the experimental set-up is shown in Figure 1. The QCL has a thermoelectrically cooled housing (Alpes Lasers); the laser is supplied by an external current driver (Keithley 2420). The QCL operates in the wavelength region between 1891.3 and 1907.6 cm1 at an output power ranging from 0.1 mW at 253 K to 26.6 mW at 243 K. The laser is scanned via a function generator and additional home-made 1
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
electronics; the latter also served as an over-voltage protection for the laser.
the laser beam was focused by a CaF2 lens of 15 cm focal length onto an infrared photovoltaic detector (PVI-6, Vigo System S.A.). All optics equipment was positioned such that optical interference in the beam path was minimized. Even with optimal positioning of all optics, fringes were not completely eliminated. Further optimization was achieved by optimizing the phase sensitive detection. It was observed that the fringe signals exhibit a strong phase-dependence and can be minimized by manually adjusting the phase of the lock-in amplifier. To further diminish the influence of fringes on the absorption signal, a background frequency scan was recorded with N2 in the multiple pass cell. This background spectrum was then subtracted from the measurement spectra.
Figure 1 Schematic diagram of the experimental setup: A combination of a stable current with scanning and modulating function generators form the laser driver electronics. Their signals are added in a home made combiner, which also acts as a voltage protection for the QCL. A inch diameter ZnSe lens of inch focal length (L1) collimates the beam before it is redirected via a 1 inch diameter CaF2 lens (L2) of 50 cm focal length to the astigmatic multiple pass absorption cell. A Helium-Neon laser is used as a tracer beam and coupled in via a beam splitter. After leaving the absorption cell, the light is focused by a 1 inch diameter CaF2 lens (L3) of 15 cm focal length onto an infrared photovoltaic detector. The 2f signal from the lockin amplifier is sent to a computer via a fast 14 bit digitizer for data acquisition. For wavelength modulation spectroscopy the modulation depth is determined by optimizing the peak-to-trough values of the 2f-signal from the unresolved NO-doublet at 1903.9 cm1 [10]. The modulation frequency f was set to 425 kHz and the 2f signal was detected with lock-in amplifier (SRS 844, Stanford Research Systems). The integration time of the lock-in amplifier was set to 100 s, which allowed a scanning rate of 100 Hz over the selected wavelength range of the laser (0.5 cm1) at a single thermo-electrical temperature. The wavelength scans were stored using a fast data acquisition card (Gage Compuscope 14200). For the detection of NO the laser beam was collimated using a short focus ZnSe collimating lens of inch (1 inch = 2.54 cm) diameter and focal length and directed into a multiple pass optical cell (Aerodyne, AMAC-76) with an absorption path length of 76 m and a volume of 300 ml. During flow-through experiments the pressure inside the cell was kept at 100 mbar by a pressure controller and a pump (see Figure 1). After the absorption cell

Figure 2. shows the wavelength modulation signal obtained after averaging 3000 frequency scans (10 ms each) over the P7.5 doublet of NO at 1903.9 cm1, which has an absorption coefficient equal to 80% compared to the strongest NO doublet at 1903.13 cm1 [10]. Here, the gas sample consisted of 100 part per billion volume (ppbv, 1:109) NO in N2 at 100 mbar pressure. After background subtraction, the absorption signal was fit using a numerical approximation of the secondderivative of a Voigt profile. Comparing the area of the Voigt curve to the standard deviation in the Gaussian distributed residual; we obtain an instrumental sensitivity of 0.8 ppbv for a total acquisition time of 30 s.
Figure
2f signal of NO from various concentrations versus frequency laser.
Figure 3. shows concentration-time data from a dilution experiment carried out to verify the experimental sensitivity. The standard deviation of measured data points agrees with the 0.8 ppbv detection limit previously calculated.
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
1 l/h. The air sample from the algae is carried with the air to multipass cell via the KOH scrubber to remove the CO2 at the same flow rate. The result is showed at figure 5, 6, dan 7, for volunteer 1, 2 and algae respectively.
Figure 3 Dilution experiment where the NO concentration was set by mixing N2 and 100 ppmv of NO using two mass flow controllers. The solid line represents the concentration calculated from the different flow rates, the circles are the measured concentration-time points. Figure 4. shows the results from a dilution experiment where the area of the fitted Voigt profile was taken as a measure of the NO concentration in the absorption cell. To determine the detection limit of our instrument, the standard deviation in the measured NO concentration has been calculated while keeping the composition of the sample gas constant. This resulted in a detection limit for NO of 0.8 ppbv in N2 at a pressure of 100 mbar, and an acquisitions time of 30 seconds. The corresponding minimal detectable absorption is 3.5 x 1010 cm1 Hz1/2 for an absorption path length of 76 m.
Figure 5 NO concentration from volunteer 1.
Figure 6 NO concentration from volunteer 2.
Figure 4 Measured concentration versus calculated concentration, with R = 0.99982 .

2
We decided to test the capabilities of our setup by sampling air from breath and algae with the treatment (MP). The volunteer blow the exhale air to the balloon via the sampling device to sure that the air is from the deep of the lung. From the balloon the air is flowed to multi pass cell via the KOH scrubber to remove the CO2 at a flow rate of
Figure 7 NO concentration from cell death of algae.
Conclusion
We demonstrate that wavelength modulation spectroscopy in combination with a thermoelectrically cooled, cw QCL at 1903.9 cm1 3
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
is suitable for detection of nitric oxides down to 0.8 ppbv in N2 (30 s acquisition time), which is equivalent to a sensitivity of 3.5 x 1010 cm1 Hz1/2. We have developed a compact, convenient setup that allows measurements with high sensitivity, is relatively easy to align compared to cavity enhanced techniques, and can be configured for autonomous operation. Our detection limit for NO of 0.8 ppbv over 30 seconds can be compared with pervious results of 0.12 ppbv Hz1/2 using a 210 m path length and temporal gating of signal and reference laser-pulses on the detector [11], 41 ppbv at 25 mbar using Faraday modulation spectroscopy [12], 0.7 ppbv over 8 seconds at 80 mbar using CW cavity ring down spectroscopy [13], and 1 ppbv for CO-laser based photoacoustics [14]. Furthermore, trace gas detection using thermoelectrically cooled cw QCLs has proved its capability in environmental monitoring and shows promise for other applications in life sciences and medical diagnostics as well.
[7]
[8]
[9]
[10]
References
[1] P.A. Martin. Near-infrared diode laser spectroscopy in chemical process and environmental air monitoring. Chem. Soc. Rev., 31: 201-210, 2002. C.D. Mansfield, H.H. Mantsch, and H.N. Rutt. Application of infrared spectroscopy in the measurement of breath trace compounds: A review. Canadian Journal of Analytical Sciences and Spectroscopy, 47(1):14-18, 2002. K. Namjou, S. Cai, E.A. Whittaker, J. Faist, C. Gmachl, F. Capasso, D.L. Sivco, and A. Y. Cho. Sensitive absorption spectroscopy with a room-temperature distributed-feedback quantum-cascade laser. Optics Letters, 23(3):291-221, 1998. L. Menzel, A.A. Kosterev, R.F. Curl, F.K. Tittel, C. Gmachi, D.L. Sivco, J.N. Baillargeon, A.L. Hutchinson, and A. Y. Cho. Spectroscopic detection of biological NO with a quantum cascade laser. Applied Physics B, 72(7):859-863, 2001. B.A. Paldus, T.G. Spence, R.N. Zare, J. Oomens, F.J.M. Harren, D.H. Parker, C. Gmachl, F. Capasso, D.L. Sivco, J.N. Baillargeon, A.L. Hutchinson, and A.Y. Cho. Photoacoustic spectroscopy using quantum-cascade lasers. Optics Letters, 24(3):178-180, 1990. A.A. Kosterev, A.L. Malinovsky, F.K. Tittel, C. Gmachl, F. Capasso, D.L. Sivco, J.N. Baillargeon, A.L. Hutchinson, and A.Y. Cho. Cavity ringdown spectroscopic detection of nitric oxide with a continouswave quantum-cascade laser. Applied Optics, 40(30):5522-5529, 2001.
[2]
[3]
[11]
[4]
[12]
[5]
[13]
[14]
[6]
M.L. Silva, D.M. Sonnenfroh, D.I. Rosen, M.G. Allen, and A.O Keefe. Integrated cavity output spectroscopy measurements of NO levels in breath with a pulsed roomtemperature qcl. Applied Physics B, 5:705710, 2005. Y.A. Bakhirkin, A.A. Kosterev, R.F. Curl, F.K. Tittel, D.A. Yarekha, L. Hvozdara, M. Giovannini, and J. Faist. Sub-ppbv nitric oxide concentration measurements using cw thermoelectrically cooled quantum cascade laser-based integrated cavity output spectroscopy. Applied Physics B-lasers and Optics, 82:149-154, 2006. S. Blaser, D.A. Yarekha, L. Hvozdara, Y. Bonetti, A. Muller, M. Giovannini, and J. Faist. Room-temperature, continuouse-wave, single-mode quantum-cascade at 5.4 m. Applied Physics Letters, 86(4):041109, 2005. L.S. Rothman, A. Barbe, C.D. Brenner, L.R. Brown, C. Camy-Peyret, M.R. Carleer, K. Chance, C. Clebaux, V. Dana, V.M. Devi, A. Fayt, J.M. Flaudi, R.R. Gamache, A. Godman, D. Jacquemart, K.W. Jucks, W.J. Lafferty, J.Y. Mandin, S.T. Massie, V. Nemtchinov, D.A. Newnham, A. Perrin, C.P. Rinsland, J. Schroeder, K.M. Smith, M.A.H. Smith, K. Tang, R.A. Toth, Vander Auwera. J., P. Varanasi, and K. Yoshino. The HITRAN molecular spectroscopic database: edition of 2000 including updates through 2001. Journal of Quantitative Spectroscopy & Radiative Transfer, 82(1-4):5-44, 2003. D.D. Nelson, J.H. Shorter, J.B. McManus, and M.S. Zahniser. Sub-part-per-billion detection of nitric oxide in air using a thermoelectrically cooled mid-infrared quantum cascade laser spectrometer. Applied Physics B, 75:343-350, 2005. H. Ganser, W. Urban, and J.M. Brown. The sensitive detection of NO by faraday modulation spectroscopy with a quantum cascade laser. Molecular Physics, 101:545550, 2003. A.A. Kosterev, F.K. Tittel, D.V. Serebryakov, A.L. Malinovsky, and I.V. Morozov. Application of quart tuning forks in spectroscopic gas sensing. Review of Scientific Instruments, 76(4):043105, 2005. L.A.J. Mur, I.E. Santosa, L.J.J. Laarhoven, N.J. Holton, F.J.M. Harren, and A.R. Smith. Laser photoacoustic detection allows in plants detection of nitric oxide in tobacco following challenge with avirulent and virulent pseudomonas syringae pathovars. Plant Physiology, 138(3):1247-1258, 2005.
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
Study on the Transport Properties of Single-Electron-Transistor Quantum Dot Using the Master Equation Approach
Moh. Adhib Ulil Absor*, Muh.Darwis Umar, Kamsul Abraha
Departement of Physics, Gadjah Mada University Yogyakarta 55281, Indonesia * Corresponding author : adhib_2004ugm@yahoo.com Abstract
Theoretical investigation of transport properties of single electron transistor ( SET) based on quantum dot (QD) is given. We modeled the transport properties of SET QD using the circuit model that consists of configuration of small dot capacitively coupled to all three electrodes: the source, the drain, and the gate electrodes. This model has been performed with rate equation approach that discrete electrons flow onto and off the dot. The analytical calculation of I-V characteristics of SET QD shows properties of non-linearity of the I-V curves. Calculation of Coulomb oscillation of SET QD shows that the temperaturedependent coulomb oscillation peak is proportional to 1/T. Keywords: SET, QD, the rate equation, Coulomb oscillation.
Introduction
Semiconductor nanostructures are interesting from a standpoint of fundamental science and compelling for their promises in a wide variety of applications. It is expected to be a good hope that will break through the limitation of conventional devices. Semiconductor technology has come along in leaps and bounds since the development of the transistor in 1948. The major trend of the development of electronic components is to improve fabrication methods so that the size of the component becomes smaller so that more circuits can be included on a chip. During the rapid development of nanotechnology, the discoveries in the physics of small electronics structure have concurrently opened up new channels in this extremely active field of both theoretical and experimental researches. Quantum dots (QDs), which are fundamentally confined electron bunches, represent basic components of nanoelectronics [1]. A QD is sometimes called as an artificial atom is a small region of space defined in a semiconductor material with a size of order 100 nm [2]. Each dot is made out of roughly a million of atoms together with an appropriate number of tightly bound electrons. However, a dot can also contain a small number of free electrons, which we can vary at will. Just like in a real atom, these electrons are attracted to a central location being trapped in a bowl-like, parabolic potential well [3]. Because the wells dimensions are so small, at sufficiently low operating temperatures electrons occupy quantized energy levels, and thus have a discrete excitation spectrum [1-3]. Measurements of the orbital energies in these atoms allow us to probe directly many important laws of quantum mechanics [4]. A large number of experimental techniques have proved successful in such measurements. One of them
is based on the fact that when an electron enters or leaves the quantum dot, it leads to a noticeable change in dots capacitance, which can be detected using, among other options, a single-electron transistor (SET) [4]. The concept of SET was proposed by Averin and Likharev in 1986 [5] and it was observed experimentally for the firs time by Lafarge et.al in 1991 [6]. During 1990s, the technology developed drastically, and a lot of fabrication methods were devised. As the years pass by, operation at room temperature was realized. SET consists of an island in between a source and drain with a gate voltage exerted on the island. The so-called SET is due to its operation single electronic regime where only one electron can tunnel from the source to drain via the island [7]. The processes that take place in this nanostructure are coulomb blockade and single electron tunneling. In this mesoscopic area, the charge discretization and energy contribution from a single electron are not negligible and lead to non linear I-V characteristics for the device. In a number of recent experimental and theoretical papers, SET semiconductor quantum dot was explored [8]. Here we focus on theoretical modelling of SET semiconductor quantum dot in which the behavior of electronic properties of SET is explained by using master equation technique. The outline of paper is as follows. In Sec.II we present the theoretical model of SET-QD. Analytical calculation of transport properties of SET-QD using rate equation approach is given in Sec.III and the oscillation coulomb curve and its temperature dependence are given in Sec.IV. Finally we outline the concluding remarks in Sec.V. Theoretical Model 1
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
We describe the configuration of SET with all parameters where a small QD is surrounded by three electrodes: the source, the drain, and the gate electrodes. Figure 1 shows the circuit model of SET quantum dot with all parameters.
position of
relative to electrode Fermi level
S , D
changes according to V and VG . Therefore
one controls the electrochemical potential of the dot for an arbitrary charge state by charging the bias voltage V and/ or the gate voltage VG [9]. Consider three energy scales: the charging energy EC , the thermal energy k BT , and the
quantum exitation energy E . From this point, we will assume that EC > k BT and E > k BT . We also asumme that EC and
E are large enough that only
one additional charge state (N+1) is accesible and that no quantum exited states are accesible for the QD. Therefore we will dealing with only two charge states N and N+1 in their own ground states. If N +1 when Figure 1. Circuit model of SET and its parameters V= VG =0 is defined as E0 , the following The dot is capacitively coupled to all three electrodes and a potential charge in any electrode will modify the electrostatic potential of the dot. The tunnel barriers separate the dot from the source and the drain electrodes where the rate of electron tunneling between the dot and the source and the drain electrode are represented by the tunneling rates S and D . They are defined as the numbers of electrons that tunnel through one of the tunnel barriers per unit time, so the units of S and D is 1/s or Hz. Using constant interaction model (CI model) [8, 9], the electrochemical potential N can be written as
N +1
can be written as
N +1 = E0 + eVdot = E0 + e
C CS V + e G VG Ctotal Ctotal
(3)
The Master Equation Approach for Single Charge Transport in SET QD For SET with single charge transport based on the assumption explained above, there are only two states available to the dot: one with an empty level (state 0, N electron state) and the other with an occupied level (state 1, N+1 electron state). During electron transport measurements, currents will flow while the quantum dot fluctuates between the two states. It is a stochastic process and therefore should be approached using a statistical methods. We define P
1 0 1 0
1 ( N )e 2 2 + eVdot N = EN + 2C
with
(1)
and P where P corresponds to the probability that the dot is in state 0 (empty dot) and P corresponds to the state 1 (occupied dot). The time change rate of P and P can be readily written as [6],
1 0
Vdot =
Vdot
1 C C C ,CiVi = C s VS + C G VG + C D VD Ctotal i =S , D G total total total C C = S V + G VG (2) Ctotal Ctotal
The new term eVdot in the equation (1) describes the effect of capacitive coupling with individual electrodes, with Vdot as function of the gate bias VG and the source bias V. In the equation (1), a term related to the drain electrode is dropped because it is grounded in the circuit ( see Figure 1) and Ctotal = CS + CG + CD . Since Vdot does not depend on the number of electrons on the dot N , the charge addition energy EC + E does not change in this case. However, the
P0 = P0 (S fS + D fD ) + P1[S (1 fS ) + D(1 fD )] , t P1 = P0 (S fS + D fD ) P1[S (1 fS ) + D (1 fD )] t

(4) so that
P1 P 0 = t t
(5)
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
These are called the rate equations based on the master equation technique [10]. In Eqs.(4) and (5), f S and
following parameters:
f D are the Fermi functions calculated at N +1 for the

source and drain electrodes. The complete form of f S and f D is
S = D =10 GHz, T=1.5K, E0 = 0(VC = 0) and CD : CS : CG = 30 : 60 : 10.
fD = 1+ exp N+1 D = 1+ exp KBT

(6)
1
N+1 KBT
+ eV fS = 1+exp N+1 S = 1+exp N+1 KBT KBT

with the value of
(7)
N +1
is given by
N +1 = E0 + eVdot = E0 e N +1 = e
CS C V e G VG Ctotal Ctotal
(8)
Figure 2. The current I as function of V that is calculated using parameters quoted in the text with S = D =10 GHz. Five I-V curves are plotted at different VG 's. Figure 2 shows the characteristic of current I as a function of V at different gate voltages. Each curve in this figure shows a non-conducting with a currents e . This suppression of conductance at low bias is a direct result of the large addition energy and it is called a Coulomb blockade. The widh of the Coulomb blockade changes accordingly to VG and become zero at
CG (VG VC ) + CSV Ctotal
As a result, the electric currents flowing from the source electrode to the drain electrode in equilibrium can be obtained by setting
P1 P 0 = = 0 . Using t t P1 =1- P 0 , we can solve equation (5) for P 0 to get (1 f S ) + D (1 f D ) P0 = S (9) S + D
VG = VC .
From Eq.(10) we can calculate the differential conductance that can be obtained by differentiating I with respect to V, namely
( f f ) f dI = e D S N+1 S dV N+1 V V = CS C +C e2 +(1 fS ) fS D G (1 fD ) fD Ctotal KBT Ctotal (11)
Finally, the electric currents will be
I = e [ P S fs + PS (1 fS )] 0 1 =e S D ( f D fS ) S +D (10)
= e ( fD fS )
so that this equation we can plot differential conductance as function of V as can be seen on Fig.3.
The final Eq.(10) is rather simple and has two major parts. determines the maximum current amplitude, while ( f D - f S ) determines whether the current will flow or not. At low temperature the values of f S and f D are either 0 or 1 in current case. The current is zero when they have different values. Therefore, the conducting case correspond to these regions where the N +1 is located between the S and
D .
Due to Eq.(10) we can plot the current I as function of V at different gate voltages using the
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
Figure 3. Five dI/dV-V curves plotted for different VG 's. It is calculated using the same parameters used in Fig. 2 Since both
V=
CG (VG VC ) Ctotal CS
(aligned to source)
(13)
f S and f D are either 0 or 1 for most cases
CG (VG VC ) = CG + CD
Similarly, the alignment condition betwen the QD and the drain electrode leads to
at low temperature, dI/dV = 0 for most cases. The first term of right-hand side of Eq.(11) will be nonzero when value of f D is between 0 and 1, which corresponds to the case where
N +1
is aligned to
D .
V =
Likewise, the second term of the right-hand side will be non-zero only when f S is between 0 and 1, corresponding to the case where
CG (VG VC ) (aligned to drain) CS
(14)
N +1
is aligned
The Coulomb Oscillation Curve and Its Temperature Dependence Now, we discuss the conductance of a quantum dot as a function of VG at low bias. Using Eq.(10), we obtain the analytic form of dI/dV curve at V=0. The equation can be simple since f S = f D in this case, so we will get
to S . So, the non-zero region in Fig.3 signifies an event that the electron energy level of the QD ( N +1 ) is aligned to one of the measure
S and D . Therefore we can
N +1 by
measuring dI/dV of SET-QD while
changing the Fermi levels of the electrodes by varying V and VG . Lets us figure out the condition for the level alignment between the quantum dot level and the source electrode ( N +1 = S ). By using Eq.(8), we can rewrite this equality as follows:
e(VG VC ) dI e2 e2 cosh [(1 fD ) fD ] = = dV KBT 4KBT 2KBT

(15) Here, we have used
C C N +1 = E0 + e S V + e G VG Ctotal Ctotal C C = e S V + e G VG = eV = S Ctotal Ctotal

We can solve Eq. (12) for V to get
= CG / Ctotal ,
the gate
efficiency factor exponent of the right-hand side. From Eq.(15), we plot dI/dV as function VG that is shown (12) in Fig. 4.
(b) (a) Figure 4. (a) Temperature dependence of Coulomb peaks, and (b) the peak height decreases with an increasing temperature. The same parameters as in Figs.2 and 3 were used and S = D =1GHz .
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
The temperature dependence of a Coulomb oscillation peak is one of the signature behaviors of a quantum dot. As we see in Eq. (15) and Fig. 4, the height of a coulomb oscillation peak is proportional to 1/T [7].
Park, J., 2002, Coulomb blockade and the Kondo effect in single-atom transistors, Nature 417(6890): p. 722-725. [ 10 ] Beenakker, C.W.J., 1991, Theory of CoulombBlockade Oscillations in the Conductance of a Quantum Dot, Physical Review B 44(4): p. 1646-1656
[9]
Concluding Remarks
We have investigated the transport properties of SET-QD using the circuit model that consists of configuration of small dot capacitively coupled to all three electrodes: the source, the drain, and the gate electrodes. Most calculations have been performed with rate equation approach in which discrete electrons flow onto and off the dot. Moreover, Coulomb blockade charging phenomena play an important role in the SET-QD operation.
References
[1] Meirav, U. and E.B. Foxman, 1996 Singleelectron phenomena in semiconductors. Semiconductor Science and Technology 11(3): p. 255-284. Chakraborty, T. and Apalkov, V.M., 2003, Quantum cascade transition in nanostructures, Advanced in Physics 52 (5), 455 Tarucha, S., Honda, T., Austing, D.G., Tokura, Y., Muraki, K., Oosterkamp, T.H., Janssen, J.W., and Kouwenhoven, L.P., 1998, Electronic states in quantum dot atoms and molecules, Physica E 3, 112 Tamura, H., Shiraishi, K., and Takayanagi, H., 2003, Quantum dot atoms, molecules, and superlattices, in Quantum Dots and Nanowires, Eds: Bandyopadhyay, S., dan Nalwa, H.S., American Scientific Publisher, USA Pasquier, C., et al., 1993, Quantum Limitation on Coulomb Blockade Observed in a 2d Electron-System, Physical Review Letters 70(1): p. 69-72. Averin, D.V. and K.K. Likharev, 1986, Coulomb blockade of Single-Electron Tunneling, and Coherent Oscillations in Small Tunnel-Junctions, Journal of Low Temperature Physics, 62(3-4): p. 345- 373. P. Lafarge, H. Pothier, and E. R. Williams, et al., 1991, Z. Phys. B 85, 327. Kouwenhoven, L.P., et al., 1997, Electron transport in quantum dots, in Mesoscopic Electron Transport, Eds: L.L. Sohn, L.P. Kouwenhoven, and G. Schoen, Plenum: New York and London, p.105-214.
[2]
[3]
[4]
[5]
[6]
[7] [8]
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
Recent Progress in Solar Cell Technology Based on Quantum Dot Structure

Moh. Adhib Ulil Absor*, Muh Darwis Umar, Isom Hilmi, Kamsul Abraha
Department of Physics, Gadjah Mada University Sekip Utara PO BOX BLS 21 Yogyakarta 55281 * Corresponding author : adhib_2004ugm@yahoo.com Abstract
In this work we propose and review recent progress and current trends in solar cell technology based on quantum dot (QD) structure and discuss a short exposition of theoretical descriptions. Furthermore, despite extensive research, the achieved conversion efficiency is nearly saturated during the last decade. Nanotechnologies are the building blocks for next-generation solar cells because low-dimensional quantum structures with the quantized energy levels naturally become the driving force for next-generation solar cells due to the mitigation of the thermal scattering as well as the control of the light absorption range. Here, three QD solar cell configurations are reviewed: (1) photoelectrodes comprising QD arrays, (2) QD-sensitized nanocrystalline TiO2, and (3) QDs dispersed in a blend of electron- and hole-conducting polymers. Keyword : quantum dot (QD), solar cell, organic semiconductor
Introduction
The solar cell technologies using solar light are promising candidates among other renewable energy technologies, because the sun is our primary source of clean, abundant energy. Since 120,000 TW of solar radiation strikes on the earth surface, 10 % efficient solar conversion system covering 0.16 % of the land on the Earth would produce 20 TW electric power, nearly twice the annual global energy consumption [1]. The maximum thermodynamic efficiency for the conversion of unconcentrated solar irradiance into electrical free energy in the radiative limit, assuming detailed balance and a single threshold absorber, was calculated by Shockley and Queisser in 1961 [2] to be about 31%; this analysis is also valid for the conversion to chemical free energy [3]. Since conversion efficiency is one of the most important parameters to be optimized in implementing photovoltaic and photochemical cells on a truly large scale [4], several schemes for exceeding the Shockley Queissar (SQ) limit have been proposed and are under active investigation. These approaches include tandem cells [3,4], hot carrier solar cells [4], solar cells producing multiple electronhole pairs per photon through impact ionization [9,10], multiband and impurity solar cells [4,11], and thermophotovoltaic/thermophotonic cells [4]
In recent years it has been proposed [7,8], and experimentally verified in some cases [6], that the relaxation dynamics of photogenerated carriers may be markedly afected by quantization effects in the semiconductor (i.e., in semiconductor quantum wells, quantum wires, QDs, superlattices, and nanostructures). That is, when the carriers in the semiconductor are confined by potential barriers to regions of space that are smaller than or comparable to their de Broglie wavelength or to the Bohr radius of excitons in the semiconductor bulk, the relaxation dynamics can be dramatically altered; specifically the hot carrier cooling rates may be dramatically reduced, and the rate of impact ionization could become competitive with the rate of carrier cooling [6]. As shown in Fig.1, low-dimensional semiconductors can modify the density of states (DOS) by restricting their size to the other of an electron wavefunction. The quantized energy levels in quantum structures become the driving force for next-generation solar cells due to the mitigation of the thermal scattering as well as the control of the light absorption range [6].
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
Figure 1. Geometry and density of state (DOS) of low dimensional systems There are two fundamental ways for utilizing hot carriers in enhancing the efficiency of photon conversion: first, by producing an enhanced photovoltage, and second by producing an enhanced photocurrent. The former way requires that the carriers be extracted from the photoconverter before they cool [9,10], whereas the latter requires the energetic hot carriers to produce a second (or more) electron-hole pair through impact ionization [11,12]. This latter process is the inverse of an Auger process whereby two electron-hole pairs recombine to produce a single highly energetic electron-hole pair. In order to achieve the former, the rates of photogenerated carrier separation, transport, and interfacial transfer across the semiconductormolecule interface must all be fast enough compared with the rate of carrier cooling [13]. In this process, the promising candidate allowed for achieving high efficiency photon conversion is the so-called QD Solar Cell. In this paper we discuss recent progress and current trends in solar cells based on quantum structure especially QD solar cells and give a short view of theoretical descriptions. The outline of this paper is thus as follows. In Sec.II we present a novel description of solar cells based on quantum dot systems. Explaining carrier multiplications and hot carriers as the main factors in increasing eficiency conversions are the concern of Sec. III. This report ends up with concluding remarks on the current trends and future development of quantum dot solar cells and are given in Sec. IV. Solar Cells Based on Quantum Dots
Clearly, the great leap forward in solar cell technology that will dramatically improve conversion efficiency is to design cells that incorporate more than one band gap into a single device. This requires a device with several different materials or a single material whose band gap is tailorable. By including several different materials into the device, one could help solve the conversion efficiency problem, but in the process create many more problems. One such problem is known as lattice mismatching. When two different materials are forced to bond together, stresses can develop between the two layers if the unit cell spacing of the materials is not identical. The stress comes out as a result of the forced stretching and compression of bonds at the interface. This lattice mismatching creates a non-equilibrium state for the atoms involved, making them more susceptible to change. They act as defects so that electrons and holes created within the cell are often drawn to these lattice mismatched sites and trapped. This effect serves to reduce the overall efficiency of the solar cell. Another potential negative effect on conversion efficiency due to multiple materials is the issue of materials interaction. When a material is brought together with another material, there is always a likelihood that an alloy can form at the interface, where essentially a mixing process occurs. However, as is often the case, this resulting material may not possess the kind of optical or electrical properties needed. As a consequence, either the incident light reflected or scattered, or carrier trap sites are created. Whichever may be the case, both result in large carrier losses and poor conversion efficiency. QD offers an interesting option since they can be fabricated in different sizes and can exhibit a range of band gap energies. Since their band gap is tailorable, based on size, different layers could act together to reduce spectral mismatch and increase the conversion efficiency. However, quantum dots are usually formed in a matrix of different composition, and so there may be additional concerns of lattice mismatch and materials interaction. In addition, since a process of quantum confinement alters the band gap, there may also be barriers to the injection of these carriers into the matrix, limiting carrier transport in the device. If these problems can be resolved, quantum dots have the potential to improve the present state of solar cell technology because these unique structures have the ability to change their band gap. The two fundamental pathways for enhancing the conversion eficiency (increased 2
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
photovoltage [7,8] or increased photocurrent [9,10] ) can be accessed, in principle, in three different QD solar cell configurations which are described below. (QD superlattice in pin cells In this configuration, the QDs are formed into an ordered 3-D array with inter-QD spacing sufficiently small so that strong electronic coupling occurs and minibands are formed to allow long-range electron transport; the QD array is placed in the intrinsic region of a pin structure [6, 7]. The delocalized quantized 3-D miniband states could be expected to slow down the carrier cooling and permit the transport and collection of hot carriers at the respective p and n contacts to produce a higher photopotential in a PV cell or in a photoelectrochemical cell where the 3-D QD array is the photoelectrode and also, impact ionization might be expected to occur in the QD arrays, enhancing the photocurrent. The schematic configuration of QD superlattice in p in cells is shown in Figure 2 [6, 11].
sensitized solar cells could produce much higher conversion effiencies than are possible with dyesensitized solar cells. The schematic configuration of QD-sensitized nanocrystalline TiO2 solar cells is depicted in Figure 3 [6].
(a)
(b)
Figure 3. (a) QD-sensitized nanocrystalline TiO2 solar cell configuration. (b) Schematic process of solar cell conversion based on QD-sensitized nanocrystalline TiO2. QD solar cells based on organic semiconductor polymer matrices In this configuration as shown in Figure 4, a disordered array of CdSe QDs is formed in a hole-conducting polymerMEH-PPV(poly(2methoxy, 5-(2-ethyl) hexyloxyp henylenevinylene) [38]. Upon photoexcitation of the QDs, the photogenerated holes are injected into the MEH-PPV polymer phase, and are collected via an electrical contact to the polymer phase. The electrons remain in the CdSe QDs and are collected through diffusion and percolation in the nanocrystalline phase to an electrical contact to the QD network. Initial results show relatively low conversion efficiencies but improvements have been reported with rod-like CdSe QD shapes embedded in poly (3-hexylthiophene) (the rodlike shape enhances electron transport through the nanocrystalline QD phase). In another configuration, a polycrystalline TiO2 layer is used as the electron conducting phase, and MEH-PPV is used to conduct the holes; the electron and holes are injected into their respective transport media upon photoexcitation of the QDs. A variation of these configurations is to disperse the QDs into a blend of electron and hole-conducting polymers. This scheme is the inverse of lightemitting diode structures based on QDs. In the PV cell, each type of carrier-transporting polymer would have a selective electrical contact to remove the respective charge carriers. A critical factor for success is to prevent electronhole recombination at the interfaces of the twopolymer blends; prevention of electronhole recombination is also critical for the other QD configurations mentioned above. All of the 3
(a)
(b)
Figure 2. (a) The schematic configuration of quantum dot superlattice in pin cells. (b) Band diagram of QD superlattice in pin cells configurations. QD-sensitized nanocrystalline TiO2 solar cells For the QD-sensitized cell, QDs are substituted for the dye molecules; they can be adsorbed from a colloidal QD solution [12] or produced in situ [13, 14]. Successful PV effects in such cells have been reported for several semiconductor QDs including InP, CdSe, CdS, and PbS [13, 14]. Possible advantages of QDs over dye molecules are the tunability of optical properties with size and better heterojunction formation with solid hole conductors. Also, as discussed here, a unique potential capability of the QD-sensitized solar cell is the production of quantum yields >1 by impact ionization (inverse Auger effect). Dye molecules cannot undergo this process. Efficient inverse Auger effects in QD-
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
possible QD - organic polymer photovoltaic cell configurations would benefit greatly if the QDs can be coaxed into producing multiple electron hole pairs by the inverse Auger/impact ionization process. This is also true for all the QD solar cell systems described above. The most important process in all the QD solar cells for reaching very high conversion eficiency is the multiple electron hole pair production in the photoexcited QDs; the various cell configurations simply represent different modes of collecting and transporting the photogenerated carriers produced in the QDs [6].
before the thermal scattering of excess energy. Thus the low dimensional systems of materials such as QDs are promising as selective energy contacts. Although there have been intense attempts, both the concepts of carrier multiplication and hot-carrier solar cells have not been materialized in device yet. Most of related technologies have only been published in academic journals [13]. Concluding Remarks on Current and Future Developments The recent progress in solar cell based on quantum dot structures has been reviewed here. Incorporation of QD superlattice in pin cells, QD-sensitized nanocrystalline TiO2 solar cells, and QD solar cell based on organic semiconductor polymer matrices into photovoltaic device leads to a spectacular improvement of the theoretical maximum efficiency, compared to the conventional bulk-semiconductor-based solar cell. In point of fact, the currently achieved efficiency from the solar cells using the quantum structures cannot come up with the theoretical maximum value, but it is rather lower than the efficiency of the conventional bulk-semiconductor-based solar cells. Therefore, grand challenges to reduce the gap between the ideal and real values of the conversion efficiency are requested. Besides, large-area, costeffective, and highly reproducible fabrication processes should be developed for mass production [13].
Figure 4. Schematic conversion process in QD solar cell based on organic semiconductor polymer matrices Carrier Multiplication and Hot Carrier The efficiency barrier of the conventional solar cells caused by thermalization may be overcome by direct carrier multiplication [13]. In bulk semiconductor crystals, impact ionization produces carrier multiplication, however, the quantum efficiency is low (1.3 electron-hole pairs per absorbed photon) [12, 13]. In contrast, recent experimental results have reported significantly higher quantum efficiency of carrier multiplication (or multiple exciton generation) in QD, i.e., 7 excitons per absorbed high energy photon in PbSe colloidal QD nanocrystals [13]. Quantum confinement in the nano-sized QD quantizes the energy levels for electrons and holes. These discrete energy levels retard hot exciton cooling and improve multiple exition generation [13]. The calculated maximum efficiency for quantum dot solar cell in the presence of carrier multiplication is 42 % [12, 13]. The hot-carrier solar cell using selective energy contacts has also been suggested in order to reduce the energy loss due to thermalization. The selective contacts must extract photogenerated hot-carriers from the absorber
References
[1] Nault R. M. 2005 in Basic research needs for solar energy utilization: Report on the basic energy sciences workshop on solar energy utilization . Argonne, Aronne National Labolatory USA. Shockley W, Quisser H. J, 1961 Detailed balance limit of efficiency of p-n junction solar cells. J. Appl Phys 32: 510-519. Green, M.A., 2004, Third generation Photovoltaics: Advanced Solar Energy Conversion, Springer-Verlac, BerlinHeidelberg. Nozik, A. J., 2005, Exciton Multiplication and Relaxation Dynamic in Quantum Dots: Application to Ultrahight-Efficiency Solar Photon Convertion, Inorg. Chem. 44 , 6893-6899 Ross, R. T. dan Nozik, A.J., 1982, 4
[2]
[3]
[4]
[5]
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
[6] [7]
Efficiency of Hot Carriers Solar Energy Converter, J. Appl. Phys. 53, 3813 Nozik, A. J., 2002, Quantum dot solar cells. Physica E 14: 115-120 Fangsuwannarak, T., 2007, Electronic and Optical Characterisations of Silicon Quantum Dots and its Application in Solar Cells, Phd Dissertation, ARC Centre of Excellence for Advanced Silicon Photovoltaics and Photonics, University of New South Wales, Sydney, Australia. Luque A, Marti A, Lopez N, Antolin E, Canovas E, Stanley C, Farmer C, Caballero LJ, Cuandra L, Balenzategui JL., 2005, Experimental analysis of the quasi-fermi level split in quantum dot intermediate-band solar cells. Appl Phys Lett 87: 083505. Luque A, Marti A, lopez N,. 2006, Operation of the intermediate band solar cell under nonideal space charge region conditions and half filling on the intermediet band, J. Appl Phys 99: 094503.
[8]
[9]
[ 10 ] Jiang C, Green MA., 2006, Silicon quantum dot superlattices: Modeling of energy bands, densities of states, and mobilities for silicon tandem solar cell applications, J. Appl Phys 99: 114902 [ 11 ] Rafffaellei RP, Castroz SL, Hepp AF, Bailev SG., 2002, Quantum dot solar cells, Prog. Photovolt: Re. Appl.; 10: 433-439 (DOI:10.1002/pip.452). [ 12 ] Schaller R.D, Klimov V.I., 2004, High efficiency carrier multiplication in PbSe nanocrystals: Implication for solar energy conversion, Phys Rev Lett 92: 186601 [ 13 ] Myong AY., 2007, Recent Progress in inorganic solar cells using quantum structrures, Recent Patents on Nanotechnology I, 67-73
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
A Proof of Dependency of Organic Photovoltaic Device Diode Ideality Factor on Light Intensity Using Linear Approximation near Break-down Voltage Method
Sholihun, Kuwat Triyana, Timothy Siahaan and Budi Prabowo Soewondo
Physics Department, Faculty of Mathematics and Natural Sciences, Gadjah Mada University Sekip Utara BLS 21, Yogyakarta 55281, Indonesia e-mail: triyana@ugm.ac.id, lion_avs@yahoo.com
Abstract
In determining photovoltaic device internal parameters, we have compared the result of two assumptions about diode ideality factors (n). First assumption is that n is a generic property of the device, which is independent on light intensity, so that the value of the parameter determined from the data obtained on darkness (without illumination). Second assumption is that n is an intensity-dependent parameter. The calculation of internal parameters is done based on the reverse bias characteristic and used the assumption that there is a linear profile near the breakdown voltage on the current-voltage (JV) curve. We called this method as Linear Approximation Near Breakdown Voltage (LANBV) method from now on. To find out whether n is independent or dependent on light intensity, we have done curve fitting between the J-V curves obtained from experiment with the curve obtained from each assumption. For the first assumption there is significance deviation between the curves obtained from both conditions. Surprisingly, the second assumption gives a very good fitting to the current density-bias voltage curve in that we cannot distinguish the curve from our approximation with the curve from the experimental data. Finally, we conclude that there is dependency between organic photovoltaic device diode ideality factor and intensity of illumination. Keywords : photovoltaic device, reverse bias characteristics, LANBV method
Introduction
Researches on organic photovoltaic device were conducted to gain information that can be used to increase its energy conversion efficiency. This information was found from the device internal parameters, in which the performance of the device depends on it and also from its physical values and behavior upon the changes of illumination intensity. By knowing this information, hopefully we can get a deeper insight about the device performance (Triyana et al, 2005). In the end, the knowledge about the device parameters values and its behavior upon illumination in certain intensity will gives valuable information in order to fabricate a better device in the future. Intensive studies on heterojunction organic photovoltaic devices have been carried out in recent years because it is possible to fabricate them on lightweight conformable substrate using low-temperature processing techniques potentially at low cost (Pettersson et al, 1999). Nowadays, the power conversion efficiency of such devices has achieved a few percent while their inorganic materials counterpart has achieved more than 25% (Gupta et al, 2008). One of the difficulties to increase the efficiency of these devices is the lack of knowledge
needed to identify internal parameters of such photovoltaic devices. Knowledge on internal parameters affecting the performance of such devices will enable one to enhance its power conversion efficiency (Triyana et al, 2008). Therefore, in this article we introduce two assumptions related to the ideality factor of diode based on single diode equivalent circuit to determine the internal parameters of our organic photovoltaic device. Hopefully from these steps we can eventually conclude which one of the cell performance that may result in the highest energy conversion efficiency.
Experiment
In this research we study the heterojunction organic photovoltaic device with the active layers of CuPC/PTCBI. Schematic structure of the device shown in Fig. 1.
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
The characteristic equation of current-voltage density for the equivalent circuit above is given by
Figure 1 Schematic structure of the device Indium Thin Oxide (ITO) and Ag layers are the electrodes that attract charges created from the exciton wich is ussually assumed appears on the junctions of active materials (CuPC/PTCBI). The layer PEDOT:PSS is used as an insulator in order to prevent short circuits that might have happened because of the unsmooth surface of ITO layer. The total active layers are 100 nm thick wich are equally devided into the p-type and the n-type (each have 50 nm thickness). The absorption spectra of the active layers are the same with the one presented in (Triyana et al, 2005) without the contribution from PTCDI layer. The top metal electrode was deposited through a shadow mask, so that the active area of the device is about 4 mm2. The J V characterization was carried out using Keithley 238 source measure unit in vacum for varied intensity of illumination using Xenon lamp where the white light from a 1.5 AM solar simulator was used. Forward bias is defined as positive voltage applied to the ITO electrode. The Model The model used in this research is a circuit that has one diode attached parallel with a resistance R p and both are attached series with a resistance Rs . A photo-current flows in the diode with the density J ph that varies linearly in the ideal condition on intensity of illumination. The photo-current that flows on the device is parallel to the electric fields present in the diode, which is parallel with the direction of the saturation current on the diode in the reverse bias condition. The equivalent one diode circuit model is given schematically in Fig. 2.
V JRs A J s exp e 1 nkT 1 J= . Rs V 1+ J + R p ph AR p
(1)
where A is the active area of the device, e is the absolute value of electrons charge, k is the Boltzman constant, and T is the temperature of the device, which is assumed to be the same with the temperature of the room in which the experiment was done. There are two important points in the J V curve that give important information about the device we are studying, they are short-circuit current J sc and open-circuit voltage Voc , which obey the following equations
Voc =
and
nkT J ph ln 1 + e Js
1 Voc , J R A ph p
(2)
1 J sc = R 1+ s Rp
J sc Rs A exp e nkT . J ph J s
(3)
With algebraic manipulation, from Eqs. (2) and (3) we get
J 1 eV J ph = s 1 exp oc + R p A Voc nkT J s

eVoc J ph J s + 1 + Rs A1 exp nkT J s Voc =0 J sc Rs A 1 + 1 J J s exp J sc ph nkT
(4)
(5)
The J V characteristic curve of a diode can be divided into four segments, the breakdown-voltage segment, the reverse bias voltage segment, the photocurrent segment, and the exponential segment in which the exponential term dominates the shape of the curve. In this situation V is negative and JRs A is small. So
V JRs A
is
negative.
Because
e 100 , the value of exp e V JRs A decays nkT nkT

Figure 2 One diode equivalent circuit model rapidly producing a linear profile in the curve which can be approached by a linear equation
Proceeding Book
113

January 24, 2009 ISBN : 978 979 19201 0 0
J=
J s + J ph V (Rs + R p )A Rs + R p R p , 1 , ( Rs + Rp ) A
J s + J ph
(6)
Equation (4) giving us two additional equations
directly predicted from the J V curve because theoretically the voltage values of an open circuits and also the short circuits current density must vanish when there is no illumination on the device. Using these assumptions, the value of Rs and R p can be determined by solving eqs. (7) and (8) after considering the linear part of the J V cuve. By minimizing the function on eq. (9), we can determine the value of diode ideality of the device, wich is 1,98. On the second assumption, n presumed to be dependent on the illumination intensity. In other words, on each intensity, correspond different n. parameters to be found are s ph p s . From all of these five parameters, n is the one choosen as the free parameters, and with algebraic manipulation we can determine the other four parameters as a function of
m=
and
(7)
(8) Rp , Rs + R p where m, c are the slope of the linear part of the curve and the intersection of it with the vertical axis in the graph. This is intuitively correct since until the breakdown voltage is met, when the device is given a reverse bias voltage, the current density produced will be a constant (i.e the sum of saturation current density and the photocurrent density) and an amount of current from the leakage path which is influenced by the resistivity of the bulk material, too. Using this analysis and from Fig. 2, we will arrive at (4). This procces we called LANBV (Linear Approximation near Break-down Voltage) method. The parameters to be found are . To find parameters, we minimize function
c=
J , J , R , R and n
J ph
the effect is eq. (1)
will also become a function of
J ph
. The value of
J ph
was found by solving eq.(1) numerically
J s , J ph , R p , Rs and n
2
the
expected
ph considering the varied values of n. If is determined so too the other four parameters. To find the values of the parameters wanted, it can be done by minimizing S in eq. (9).
S = ( Kl J l )
l

(9) The focus of our study in this paper is to compare the results obtained from the parameters calculation using the LANBV method for the first assumptions and the second assumptions, which finally will be proved that there exists dependency between the diode ideality factors (n) with the illumination intensity. The parameters obtained from each assumption presented in Table 1. Table 1 Values of parameters based on first and second assumption for intensity 600 W/m2 First Second Assumption Assumption N 1.98 3.57 Js (A/m2) Jph (A/m2) Rp () Rs () 0.002 20.1 23376 3627 0.082 18.1 25888 1116
where l covers the experimental data from J V characterization and
Vl J l Rs A 1 J s exp e nkT 1 Kl = R V 1 + s J ph + l Rp Rp A
(10)
In calculation of these parameters, we have used two assumptions to be compared to decide wich one is more accurate. Both assumptions were 1. Diode ideality factor (n) is independent from illumination intensity. 2. Diode ideality factor is dependent to illumination intensity. In first assumption, n presumed to be a generic property of the device and it is independent from illumination intensity, so the value of this parameter is define from the data obtained in the dark (without illumination). To define the value of n and other parameters that also have roles when there is no illumination on the device, we can use eqs. (7) and (8) with neglecting the terms that contain the currentvoltage density. With only four parameterss and three equations to connect those parameters, and also considering the form of eqs. (4), (7), and (8), so the value of the saturation current density J s , can be
Then we compare the J-V curve from our model data (each assumption) with the one from experimental data. Curve fitting for first and second assumption respectively, shown in Fig. 3 and Fig. 4.
Proceeding Book
114

January 24, 2009 ISBN : 978 979 19201 0 0
J-V Curve under Intensity 600 W/m2

10
600 W/m2 (experiment) 600 W/m2 (model)
-5
from the J-V curve fitting from the first assumption and very good fitting on the second assumption gives a very clear justification on the validity of the assumption that n is a intensity dependent parameter. So, we have successfully proved that there is dependency between the diode ideality factor and intensity of illumination by means of LANBV method.
-10
Acknowledgement
The authors would like to thank the Ministry of National Education of the Republic of Indonesia for sponsoring this work through Hibah Fundamental Grant 2008. under Contract of LPPM-UGM /573/ 2007, M. A. K. 2560. 4256.573119.
-15
-20
-25 -0.1
0.1
0.2
0.3
0.4
0.5
0.6
Bias Voltage (volt)
Figure 3
Curve fitting based on first assumption under intensity 600 W/m2.

J-V Curve under Intensity 600 W/m2
References
Gupta, D., Mukhopadhyay, S. dan Narayan, K.S., 2008, Fill factor in organic solar cells, Sol. Energy Mater. Sol. Cells, doi:10.1016/j.solmat.2008.06.001. Pettersson, L. A. A., Roman, L. S., and Inganas, O., 1999, Modeling photocurrent action spectra of photovoltaic devices based on organic thin films, Journal of Applied Physics, vol. 86, no. 1, 487-96. Triyana, K., Siahaan, T., Sholihun, Abraha, K., and Salleh, M.M., 2008, On the Dependency of Equivalent Circuit Parameters of Heterojunction Bilayer Copper Phthalocyanine/Perylene Photo-voltaic Device on Light Intensity based on Reverse Bias Characteristic, ICSE Proc. 2008. Triyana, K., Yasuda, T., Fujita, K., and Tsutsui, T., 2005, Improvement of Heterojunction Donor/Acceptor Organic Photovoltaic Devices by Employing Additional Active Layer, Japanese Journal of Applied Physics, vol. 44, no. 4A, 1974-7.
5 600 W/m2 (model) 600 W/m2 (experiment)
-5
-10
-15
-20
-25 -0.1
0.1
0.2 Bias Voltage (volt)
0.3
0.4
0.5
Figure 4
Curve fitting based on first assumption under intensity 600 W/m2.
From table 1, we cannot yet conclude which one of the assumptions is accurate to describe the characteristic of the photovoltaic device. However we can get clear information by seeing on the J-V curve in fig. 3 and fig. 4. In fig. 3 we can see a clear difference between the curve obtained from the model and the curve obtained from the experimental data. It means that the result obtained contrary to the first assumption. In Fig. 4, the result of the curve fitting is very good, this indicates agreement between the result with the assumption before.
Conclusion
The calculation of parameters using the LANBV has been accurately done for each assumption. It is done to prove the dependency between the diode ideality factor (n) and the illumination intensity. A significant difference appear
Proceeding Book
115

January 24, 2009 ISBN : 978 979 19201 0 0
Problems in Determining Two Diodes Equivalent Circuit Parameters of Organic Solar Cell Using Reverse Bias Characteristic
Kuwat Triyana, Sholihun, Budi Prabowo Soewondo, Timothy Siahaan and Kamsul Abraha
Physics Department, Faculty of Mathematics and Natural Sciences, Gadjah Mada University Sekip Utara BLS 21, Yogyakarta 55281, Indonesia e-mail: triyana@ugm.ac.id, lion_avs@yahoo.com Abstract
An equivalent circuit parameters of heterojunction bilayer copper phthalocyanine/perylene organic solar cell for one diode model have been accurately calculated in our previous work. The calculation has been performed based on reverse bias characteristic using an assumption that there will be a linear profile near the breakdown voltage on the current-voltage (J-V) curve. The confirmation has been clarified by comparing the curve from this model with the curve obtained from the experiment. The result gives a very good fitting to the current density-bias voltage curve in that we cannot distinguish the curve from our model with the one from the experimental data. However, since we still have problem in describing the physics mechanism based in our previous work, we are going to continue our investigation through the recent work. For instant, it is still difficult for us to determine whether the shunt resistance is a result from the bulk or from the interface. In the recent study, we found that problems came up when we applied this approach to the two diodes model such as the negative value of shunt resistance (Rp), while the value of series resistance (Rs) exceed the value of Rp. These problems contrary to the reality that Rp must be positive and the value of Rs is always smaller than Rp. Hopefully, the problems presented in this paper can give useful information to the development of the research to increase the efficiency of organic solar cell in the future. Keywords: two diodes equivalent circuit model, photovoltaic device, reverse bias characteristic
Introduction
Research on organic photovoltaic device has becoming a trend to researchers around the world lately. This research aimed to increase the efficiency of performance of organic photovoltaic device. However, up to now, the goal of this research hasnt been reached yet. All of the experiments still result in much lower in efficiency compared to that of the conventional inorganic solar cells (Triyana et al, 2005). The numbers of problem that occur in the experiment have vexed the researchers. The source of these problems is the lack of theory or model describing them (Marsh et al, 2007). So that the researchers can only improve the cell performance by trial and errors fashion (Gruber et al, 2005). The step that should be done to minimize the time and cost spent on the research is by using the computational method to make models and then to compute their internal parameters (Siahaan et al, 2008). The results of the computation will be used to analyze the problems that occurred as mentioned earlier. In this article we introduce some problems in applying two diode models to determine the internal parameters of organic solar cell. Hopefully from these steps we can eventually conclude which one of the cell performance that may result in the highest energy conversion efficiency.
The organic solar cell and its current-voltage (J-V) characteristic used in this study are the result of our previous work (Triyana et al, 2008; Siahaan et al, 2008). In the previous research, the cell parameters was calculated based on the one diode equivalent circuits using the reverse bias characteristic and assumed that there is a linear profile on the J-V curve near the breakdown voltage. With only using the basic knowledge of circuits and the Ohms law one can find, from this model, the equation relating to the current density J and bias voltage V as Eq. 1 (Triyana et al, 2008).
V JRs A 1 J s exp e nkT 1 (1) J= Rs V 1+ J + R p ph R p A

Now, we are trying to apply the assumption of the one diode model into the two diodes model that is schematically depicted in Fig.1.
Method Proceeding Book

1

January 24, 2009 ISBN : 978 979 19201 0 0
value
of
V JRs A
is
negative.
Because
e 100 resulting the value of nkT V JR s A V JRs A exp e and exp e n kT decay nkT l
rapidly, producing a linear profile in the curve which can be approached by a linear equation (Eq.3). Figure 1 Two equivalent circuit model Parameters
J=
Rs , R p , J s , J l , n and nl are total of
J s + J l + J ph V (Rs + R p )A (Rs + R p ) R p ,
(3)
series resistances, total of shunt resistances, saturation current density without illumination, saturation current density under illumination, diode ideality factor without in the dark light condition and diode ideality factor under illumination, respectively. Based on the Fig.1, the relation between current density J and bias voltage V is changed to be Eq.2.
The Eq.3 results in two additional equations as follows (Eqs. 4 and 5).
m=
and
1 , ( Rs + Rp ) A J s + J l + J ph
(4)
c=
(R
+ Rp )
Rp ,
(5)
V JRs A 1 J s exp e nkT V JRs A 1 J= + J exp e 1 Rs l nl kT 1+ Rp V J + ph R A p
(2)
where m and c are the slope of the linear part of the curve and the intersection of curve formed by Eq.3. From these equations, we may obtain most of cell parameters. We should be careful that when in the one diode model the chosen free parameter is n while in the two diode model the chosen free parameter is nl,, meanwhile n is to be obtained from dark and J s being calculated from equation of our previous report (Triyana et al, 2008) as Eq.6,
where A is the active area of the device, e is the absolute value of electrons charge, k is the Boltzman constant, and T is the temperature of the device, which is assumed to be the same with the temperature of the room in which the experiment was carried out. In this case, the shunt resistance describes the leakage currents caused by paths that may be created during the fabrication while the series resistance describes the resistance of the device as a whole, or in other words it is a description of the bulk resistance of the device. The recombination current is not considered here since in organic photovoltaic devices, theoretically, the photo-current production in the device is not a direct process but through the formation of exciton (Ghosh and Feng, 1978; Moeller, 1993; Pettersson et al, 1999). The role of the recombination effect is taken by the quantum efficiency of the current production by molecular semiconductors and is beyond the scope of this study. Based on our model, illumination to the cell will produce additional current density, Jph in the same direction with Js and Jl. There are some segments of the J V characteristic curve. One of them is reverse bias voltage segment which is going to be the main focus in this paper. In this situation the value of V is negative and the value of JRs A is small, so that the
J s = C exp( n )
where
(6)
and = 2.2 . By using mathematical manipulation, the other four parameter can be given as functions of Jph. As a consequence of this approach is that Eq.(2) is also a function of Jph. The value of Jph can be found by solving Eq.(1) numerically with the input parameter nl being varied. If the value Jph has been found then the other four parameters can be found either. To find the expected parameters, just like on the one diode model, it can be done by minimizing a function that we choose as a measure of the deviation of our calculation from the experimental data Eq.7.
C = 3.0 10 Am
S = ( Kl J l )
l
(7)
where l covers the experimental data from J-V characterization and
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
Vl J l R s A 1 J s exp e nkT 1 V J l R s A 1 Kl = + J l exp e l R nl kT 1+ s Rp Vl J ph + R A p
(8)
value of Rs exceeds the value of |Rp|. In reality, Rp must positive and Rs must be much smaller than Rp. Under illumination with intensity of 600 W/m2, for example, there are minimum deviations more than unity, while it is supposed to be unity. The calculated parameters for the two diodes model of our cell under illumination with intensity of 50 W/m2 and 600 W/m2 presented on Table 1 (the value of parameters presented for 600 W/m2 intensity are the values obtained from one of the minimum deviations).

In our previous work, the calculation of cell parameters had been accurately carried out. The parameters obtained from numerical computation is said to be valid if there is an argument match between the result obtained from the model with the result obtained from experiment. The computation accuracy can be checked by comparing the J-V curve obtained from the model with the curve obtained from experiment. The result we obtained was amazing (Siahaan et al, 2008), the curve from the model is almost identical to the curve from experiment. In other words, our model makes good fitting with the experiment. Since we still have problem in describing the physics mechanism based in our previous work, we are going to continue our investigation through the recent work. For instant, it is still difficult for us to determine whether the shunt resistance is a result from the bulk or from the interface. Therefore, in this study we are trying to develop our model using the two diodes model. However, we found a problem in this case, that is the value of cell parameters turns out to be different from what we have expected. The scenario to implement our new model is executed from using the mathematical manipulation (to obtain eq. (4), (5), and (6)) until the using of numerical computation (to calculate the parameters). However, the computation result indicates a contra to reality (it is clear to understand this problem without plotting J-V curve. This problem occurred is not caused by computational errors in the process, because the same model has been succed in the previous research (for one diode model). So, the next possible steps that can be taken are to analyze equation (2) and find the representatives with the model in fig. 3, or using a new equivalent circuit for the two diodes model. Table 1 Values of parameters for two diodes model Intensity 50 W/m2 600 W/m2 nl Jph (A/m ) Jl (A/m2) Rp () Rs ()
2
Conclusion
We have been designing a scenario to search for the organic solar cell parameters for the two diodes model using the approach that we have been used on the one diode model. Further predictions on what might cause these problems are not yet studied in detail, but at least we have calculated the parameters and shown the problems which emerge in determining the ones for two diodes model. Hopefully, the problems presented in this paper can give useful information to the next investigation to analyze such problems. Finally, we hope to be able to give contribution in the developing of organic solar cell in the future.
Acknowledgement
The authors would like to thank the Ministry of National Education of the Republic of Indonesia for sponsoring this work through Hibah Fundamental Grant 2008. under Contract of LPPM-UGM /573/ 2007, M. A. K. 2560. 4256.573119. The authors also would like to thank C. Nur Yusuf Arif for some technical support.
References
Ghosh, A. K., and Feng, T., 1978, Merocyanine organic solar cells, Journal of Applied Physics, vol. 49, no. 12, 5982-9. Gruber, D.P, Meinhardt G., dan Papousek W., 2005, Spatial distribution of light absorption in organic photovoltaic devices, Solar Energy 79, 697704. Marsh, R.A., Groves, C., dan Greenham, N. C., 2007, A microscopic model for the behavior of nanostructured organik photovoltaic devices, J. Appl. Phys. 101, 083509. Moeller, H. J., 1993, Semiconductors for Solar Cells, Artech House, Boston, USA. Pettersson, L. A. A., Roman, L. S., and Inganas, O., 1999, Modeling photocurrent action spectra of photovoltaic devices based on organic thin films, Journal of Applied Physics, vol. 86, no. 1, 487-96. Siahaan, T., Triyana, K., Sholihun, 2008, A method of determination of equivalent circuit
1.45 3.03 -5.99 10-6 -90613 219945
0.96 40.79 0 -11528 38532
Some of the parameters are even contrary to the reality, like the negative value of Rp and the obtained
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
parameters based on reverse bias characteristic of photovoltaic devices, ICMNS Proc. 2008. Triyana, K., Siahaan, T., Sholihun, Abraha, K., and Salleh, M.M., 2008, On the Dependency of Equivalent Circuit Parameters of Heterojunction Bilayer Copper Phthalocyanine/ Perylene Photovoltaic Device on Light Intensity based on Reverse Bias Characteristic, ICSE Proc. 2008. Triyana, K., Yasuda, T., Fujita, K., and Tsutsui, T., 2005, Improvement of Heterojunction Donor/Acceptor Organic Photovoltaic Devices by Employing Additional Active Layer, Jpn. Journal of Applied Physics, vol. 44, no. 4A, 1974-7.
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
What is the Worst News in Physics?

Fianti1,2 dan Kamsul Abraha1
2
Jurusan Fisika FMIPA UGM, Sekip Utara, Yogyakarta 55281 Jurusan Fisika FMIPA UNNES, Jl Sekaran, Gunungpati, Semarang 50229 e-mail : fieanty@yahoo.com, fianewvita@yahoo.com Abstract
For centuries, physicist have satisfied by the results of applying optic laws such as discovery and observation of the universe that make the entire perspectives about the universe appearance can be assumed has been established. Because of that, astronomers have strong confident that the mapping of objects in the universe are accurate and can represent the real situation of the universe. The observation aspects is included position, age and distance of the objects such as stars, planets, galaxies in the universe based on electromagnet wave that be detected. The whole perception from interpreting the astronomical data will be dramatically changed by the discovery of new phenomenon in optic material called metamaterial. The optical phenomenon emerges from a radical idea of Russian physicist about negative index material in 1968. In line with this issue, our research is a attempt to understand the phenomenon from classical electromagnet point of view. Our study shows that negative index material can be designed by SRR and thin wire in which depend on plasma and operating frequency. Further analysis can also show that several optical laws must be modified. In line with astronomical data, if there are systems in the universe have negative refractive index then astronomical data should have new interpretation. Keywords: metamaterial, refractive index, disperse, plasma frequency
Introduction
Astronomers like thinking that they know absolutely where the stars are. They can point to them in the night sky: there is Polaris, there is Vega, there is Adhara, and so on. The night sky had been mapped for millennia. But the rightness can be proved, because no one has been out to check that the stars really are, like where they are. Our celestial maps are based on the assumption that photons of light almost always travel from the stars to our telescopes on Earth in a straight line. The universe is roughly 13 billion years old: a lot of things could have happened to photon trajectories in that time. Light is traveled in the different way, not like in usual way, when through near a twisted black hole. It will give a wrong interpretation about the star position. The cause is a new phenomenon discovered, called negative refraction, a new thing that attemted to be understood. So far, astronomers have been observate to search how the light can be refracted by giant objects, like black holes. If the negative refraction is a real thing, the astronomers measurement be receck, espesialy black hole and giant objects involve in it. The negative refraction is a new thing in astronomy,
but it can rise much changing in science[1]. That is very bad thing in science. That is a big phenomenon in cosmology that can not be observe in a labotary, but the good news is there are a new concept for understood a negative refraction through metamaterial, that can show a negative refraction. A metamaterial (or meta material) is a material that gains its properties from its structure rather than directly from its composition. To distinguish metamaterials from other composite materials, the metamaterial label is usually used for a material that has unusual properties. The term was coined in 1999 by Rodger M. Walser of the University of Texas at Austin. Metamaterial can be understood as macroscopic composites having a manmade, threedimensional, periodic cellular architecture designed to produce an optimized combination, not available in nature, of two or more responses to specific excitation. Among electromagnetics researchers, the term is often used, quite narrowly, for materials that exhibit negative refraction [2]. Metamaterial has negative refraction in light traveling becouse of having negative refraction index in its medium, but every negative refraction does not have to have a negative refractive index. So, it is important to distinguish between negative refraction and negative refraction index. An example for negative refraction is a procces getting image of an 1
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
invetebrate that has a superposition eyes-stuctures that behave like metamaterial [3].
differ from the electrodynamics of materials with positive and . There are three possible answers to this rhetoric question: (1) There are no difference, i.e., electrodynamics is invariant with regard to be simultaneous
Figure 1 Invertebrate superposition eyes-structures that behave like metamaterial with negative refraction (not a negative refraction index) [3]. Metamaterials, sometimes called as double negative metamaterials, are materials having negative permittivity and permeability simultaneously that give a negative refractive index. Metamaterial electric is a material having a negatif permittivity and a positive permeability. Metamaterila magnetic is a material having a negative permeability and a positive permittivity. This paper just gives a view throug electomagnetic windows to explain a negative refractive material (metamaterial), the wave propagation, and energy conservation and momentum consevation in it. The idea of metamaterial Metamaterial had been a technology of our ancient in the past. First metamaterial is known from a Lycurgus Cup (4th century AD), an ancient random metamaterial with gold. Also found out, example of meta-atom that found out from Proceeding of Royal Soc. London in 1898 [4]. After that time, discussion about metamaterial death. 1968, metamaterials discussion was rebuilt from its long sleep by Veselago with his issue that announced in Soviet Physics Uspekhi. He said metamaterial in electromagnetic way [5]. The choice of a negative sign for both (relative permittivity) and (relative permeability) does not cause mathematical contradictions; in particular this does not change the classical expression for n :
and . change the signs of (2) Simultaneously negative values of and are in principle impossible because this conflicts with some basic principles. (3) Simultaneously negative values of and are possible, but the electrodynamics for the case of positive and . It is clear, for having material with negative and simultaneously, we must choose number three and make new formulas for the material that different with the ordinary material (material with positive and ). Wave propagation in metamaterial The complex refractive index of a given medium is defined as the ratio between the speed of an electromagnetic wave through that medium and that in vacuum and can thus be written as n = , where is complex relative dielectric permittivity and is complex relative magnetic permeability. If both and are negative in a given wavelength range, this means that we may write
2
= + i = exp(i )
(2) (3) (4)
= + i = exp(i ) .
It follows that
n = exp(2i ) =
exp(2i ) =
n = .
(1)
i.e. the refractive index of a medium with simultaneously negative and must be negative [6].
The main question now is how the electrodynamics of the materials with negative and
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
< 0, > 0
electrical plasma evanescent decaying waves many metals (UV- optical) thin wires structures (GHz)
> 0, > 0 v fase > 0

vgrup > 0
r S
ordinary optical materials right handed propagating waves
< 0, < 0 r v fase < 0 S v grup > 0

artificial meta-material negative refractive index left handed propagating waves
> 0, < 0
magnetic plasma structured materials evanescent decaying waves some natural magnetic materials (upto GHZ)
Figure 2 A schematic showing the classification of materials based on the dielectric and magnetic properties. Dispersion relation, from Maxwell equations, gives Some time ago it was shown that wire structures with lattice spacings of the order of a few millimeters behave like a plasma with a resonant frequency, ep , in the GHz region. The ideal dielectric response of a plasma is given by
r k = kk (the wave vektor) i k= 1 + + , c 2
(9)
= 1 ep22
and takes negative values for
(5)
from Eq. (2) and (3). It will be understood, when both and positive simultaneously, k > 0 , and vice versa. Thus, when and are simultaneously positive, E , H , and k form right-handed set, while for double-negative materials, it is left handed. It can be seen in Fig. 2.
r E
r k r k
< ep .
More
recently, a structure containing loops of conducting wire has properties mimicking a magnetic plasma [7],
1
If
< ep ,
and
mp 2
(6)
r E
will be negative and so does

r H
n by Eq. (4). Usually ep = mp = p called

plasma frequency. Maxwell equations in ordinary medium are the shame used in metamaterial.
r H
r r r k E = H , r r r k H = E .
(a)
(b)
(7) (8)
FIQ. 2 KEH sistem, (a) with k > 0 , and (b) with k < 0. Energy Conservation
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
Conservation law for energy in electromagnetic is
r r r W (10) + S = E J * t r which W is total energy density and S is a Poynting

vector. Usually, W has a formula
conventional right handed material (left) and in left handed material (right). It can be see that the wave vectorand Poynting vector are in opposite directions in LHM [10].
W =
Since and are simultaneously negative, it will give negative energy by Eq. (11), and in order the light can travel in metamaterial, the energy must positive. Thats why a disperse relation is needed in this medium to make the energy having a positive value. So, the expression of Eq. (11) becomes
1 r2 1 r 2 E + H . 2 2
Momentum conservation
(11)
Conservation law for momentum in electromagnetic is
1 ( ) 2 1 ( ) 2 W = E + H , 2 2
(12)
r r t G + T = f , t and its momentum density vector is r r r G = D B, t 1 r r r r t rr rr T = D E + B H I DE BH 2
(16)
(17) (18)
( ) ( ) > 0 and > 0 [8].

Poynting vector in ordinary material is
in ordinary material. Because of energy in metamaterial must be positive, and must in disperse function of frequency, it gives and
r r r r r r S (r , t ) = E (r , t ) H (r , t ) ,
(13)
( ) > 0 . Thus the momentum density
( ) >0
but in metamaterial, all vektor must be in harmonic time dependent. It gives the Poynting vector becomes
vector in metamaterial must change to be

r r r r r r r r r r G (r ) = D (r ) B * (r ) = 1 ( ) E (r ) ( ) H * (r ) , 2 c (19) and the Maxwell stress tensor is
r r r r r r S (r ) = E (r ) H * (r ) ,
(14)
in order to have a positive part in Poynting vector that induce a positive energy in negative refractive material [9]. By a time averaging process, in a period
t 1 r r r r t rr rr T = D E * + B H * I DE * BH * , (20) 2
r r 1 S (r , t ) = 2
r r r r 1 d ( t )S (r , t ) = Re S (r ) . (15)
0
{ }
r E
and all of fields are in harmonic condition. The average value in a period are [11]
r E
r k r S
RHM
LHM
r 1 1 ( ) r ( ) r * G = Re 2 E H 2 c
(21)
t rr rr 1 1 1 ( ) 2 1 ( ) 2 t 1 T = Re E + H I Re DE * Re BH * 2 2 2 2 2
r k
r S
(22)
Conclusions
r H
r H
Figure 3 A diagram showing the electric, magnetic, Poynting and wave vector direction in a
Metamaterials have a negative refractive index that cause a negative refraction. Because of its properties, Metamaterial are left handed system. Light travel in metamaterial gives Poynting and wave vector direction in opposite directions. In energy conservation, to make a the light travel through
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
metamaterial and must beyond in disperse law, and in momentum conservation too.
Reference
[1]
Brooks, M., 2005, Illusions of a Stary, Stary Night, http://www.newscientist.com/article/mg18625 021.200-illusions-of-astarry-starry-night.html Metamaterial, http://en.wikipedia.org/wiki/Metamaterial Stavenga, D.G., 2006, Journal of European Optical Soc. 1, 06010 Bose, J.C., 1898, On The Rotation of Plane of Polarization of Electric Wave by a Twisted Structure, Proceeding of Royal Soc. London Veselago, V.G., 1968, The Electrodynamics of Substances with Simultaneously Negative Values of and , Soviet Physics Uspekhi 10 (4), 509 Jaksic, Z., Dalarsson, N., dan Maksimovic, M., 2006, Negative Refractive Index Metamaterials: Principles and Applications, Microwave Review, 36 Pendry, J.B., 2000, Negative Refraction Makes a Perfect Lens, Physical Review Letters 85 (18), 3966 Veselago, V.G., 2005, Some Remarks Regarding Electrodynamics of Materials with Negative Refraction, Applied Physics B 81, 403. Kong, J.A., 1986, Electromagnetic Wave Theory, John Wiley and Sons, Inc., Singapore Sudhakaran, S., 2006, Negative Refraction from Electromagnetic Periodic Structures and Its Applications, Ph.D Thesis, University of Londons, UK Orfanidis, S.J., 2008, Electromagnetic Waves and Antennas, www.ece.rutgers.edu/~orfanidi/ewa
[2] [3] [4]
[5]
[6]
[7] [8]
[9] [10]
[11]
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
Giant Dielectric Response of Cubic Perovskite-Related CaCu3Ti4O12 Ceramics under Extrinsic-Lorentz Oscillator Model
Isom Hilmi, Kamsul Abraha, Kuwat Triyana, Muh Darwis Umar, Muh Adhib Ulil Abshor
Physics Department, Gadjah Mada University, Yogyakarta, Indonesia e-mail : isomhilmy@yahoo.com Abstract
A theoretical study on the dielectric response of new giant dielectric material CaCu3Ti4O12 under extrinsic-oscillator model has been done. The gigantic static dielectric response observed in the material has been discussed by adopting some extrinsic mechanisms, such as contributed by point, line, or planar defects, or more generally, from sample microstructure morphology, or boundary layers, although we simply cannot abandon the corresponding intrinsic mechanism that may also cause its giant dielectric phenomena. For doing so, we have therefore suggested to take into account some considerable contribution from Lorentz oscillators in the context of the extrinsic model dielectric constant arising from the microstructures of the material. The various responses have been calculated for some morphology as implications of the mechanisms. Keywords: giant dielectric CaCu3Ti4O12 , dielectric constant, Lorentz oscillator
Introduction
The large dielectric response of a material has usually been observed at relaxor and ferroelectric materials. As a matter of fact, the complex perovskite compound CaCu3Ti4O12 (CCTO) has been reported of having a colossal quasistatic dielectric constant for a wide range of temperature [1, 2] showing remarkably characteristics different from ferroelectric and relaxor materials, with no ferroelectric phase transition observed in the material. The giant dielectric phenomena arising from extrinsic mechanisms such as point, line, and planar defects, and more generally, with sample morphology, or boundary layers, have been proposed by Cohen et al [3] to explain the phenomena. They argue that extrinsic behaviour should arises from two-dimensional defects. There are two obvious candidates for such internal interfaces or boundaries even in single-crystal samples: antiphase boundaries separating regions in which the identity of the Ca sublattice is shifted by a primitive-cubic lattice vector, and twin boundaries separating domains in which the sense of rotation of the oxygen octahedral reverses [4]. Lorentzian oscillator model of dielectric constant has also been proposed by fitting it to the measured conductivity of the material from which phonon parameters have been obtained at temperatures of 10 K and 295 K [2]. The remainder of the paper is organized as follows. In the following section we briefly discuss some important characteristics of the morphologies used in describing the properties of the dielectric response of CCTO. In Sect. 2 we then derive and discuss the main role of the morphologies
specifically in the context of the dielectric response giving rise to various distinct optical response characteristics of the material next discussed in Sect. 3. In the last section we present a general discussion and conclusions of this paper. Sample Morphologies By exploring each morphology of the extrinsic model and taking into account the lattice mode contribution in the Lorentzian oscillator model, the optical responses of the material have been determined. That is, it has been assumed that the macroscopic dielectric constant in the morphologies is in the form of Lorentzian oscillator within the existence of surface blocked and percolating conducting regions. In the nonpercolating cases, the macroscopic relaxation time has been derived from Lorentz oscillator model resulting in the Debye form of the dielectric response. In morphologies without interface blocking layers, bulk dielectric constant has been assumed to be in the form of Lorentz oscillator dielectric constant. Taking into account the assumptions mentioned above, and associating them with extrinsic model, we can determine the model to six morphologies that have been classified into four classifications found in [3, 5]. Blocked Morphologies Conducting Region with Percolating
In the blocking cases, the contribution for the effective dielectric constant comes from the blocking layer and also from associated macroscopic dielectric constant of the material. In the blocked morphologies with percolating 1
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
conducting region, dielectric constant of the blocking layer plays a significant role in enhancing the static dielectric constant [3]. From any observation, it can be seen that the optical dielectric constant is small enough compared to its static value [1] so that we can eventually abandon its resulting constant relaxation time. For higher frequency response (far-infrared region), the relaxation time is assumed to be such given by the Lorentzian oscillator model [5]
( ) =
where
0 Mr Mim
Here, the material can be considered as a composite dielectric made out of a heterogeneous system of multiconstituent components of the material. In such system, the static dielectric constant has to obey the exact Wiener bound [7]. In the morphologies with conducting internal boundaries, the conducting internal boundaries associated with the domains can be considered as having ellipsoids form that are identical except for their random orientation, with major axes
a = b >> c occupying volume fraction . The
(1)
Mr = +
j
2 , j ( 2 '2 ) p j
( 2 '2 ) 2 4 4 j j
2 p, j
(2)
effective macroscopic polarization is contributed from the isolating domains and also from the conducting boundaries. Through effective medium approximation dielectric response can be written as [3]
and
Mim =
j
j , ( ' 2 ) 2 4 4 j
2 j
2 B i 3 , 1 1 B i 3
b
(6)
(3)
with j, j , and p,j the frequency, width, and effective plasma frequency of the jth-vibration. the core contribution to the dielectric function at high frequencies. The assumption inherent in the model is that the dielectric response is related with the phonon contribution by using phonon parameters counted for at 293 K [6]. Blocked Morphologies Conducting Region with Nonpercolating
is
The real part of the frequency-dependent macroscopic dielectric constant is [5]

Mr = M + 0 M M 2 1 + 2 M ( )
correction, and b is dielectric constant of the bulk material. The static dielectric constant of the material can have high value as has been observed to obey Wiener bound. With small enough volume fraction (fi ~10-3 ), the collossal dielectric constant can be achieved. For morphologies with isolating internal boundaries, various geometrical forms of domains and their associated boundaries are possible. The less unrealistic one is spherical conducting domains included within spherical insulating shell. Through effective medium approximation the dielectric constant can be expressed as [3]
( ) =
3 c ( ) 2 ( c ( ) b ) f i b. 3 b + ( c ( ) b ) f i
where Bi can be considered as converse depolarization factor with additional dynamical
(4)
(7)
so that the macroscopic relaxation time is given by

0 M M 1 M ( ) = ( ) 2 2 M Mr 1/ 2
By including some phonon contribution, we can consider the complex dielectric constant of the conducting domain c() of having Lorentz oscillator form, showing that it is contributed by the bulk material.
(5)
where the macroscopic dielectric constant is assumed in the form of Lorentzian oscillator. For doing so, M and M are taken as Lorentz oscillator dielectric constants with static and optical (infinite approximation) frequencies respectively.
0

For the first classification of morphologies [3, 5], although the optical properties of the material can be approximated by the assumption as mentioned above, the response is still high temperature dependent as can be concluded form the log of the relaxation time versus 1/T curve, which shows linear relation [1], indicating that the greater the temperature of the system, the smaller
Unblocked Morphologies
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
the associated relaxation time of the material. It then also indicates that the greater temperature then will imply the slower reduction of dielectric constant.
the dielectric constant, there are several low peaks just before reduction occurs on it. At the same time, abrupt rise of the imaginary part is seen indicating loss-tangent peaks. These peaks are in distinct frequency regions showing values of relaxation time of associated polarization that has been occurred.
(a) Figure 1 Dielectric response of CCTO under blocked morphologies with (a) percolating and (b) nonpercolating conducting region. The real and imaginary parts of dielectric constant are shown respectively the upper and lower curves. In the infrared region, the most significant role for the dielectric constant (see Fig.1 (a)) comes from phonon parameters. The peaks formed show that many kinds of polarization have shown up. The dielectric constant of the blocking layer is
Figure 2 Dielectric (b) response of CCTO under unblocked morphologies with isolating internal boundaries. The real and imaginary parts of dielectric constant are shown respectively by the curves initially started on far left at the upper and lower position. In the unblocked case, we have assumed that the material is the compound dielectric because of its multiconstituent nature. The dielectric response is highly dependent on its geometrical domains we consider. The effective medium approximation has been used to count for the effective dielectric constant. In the morphologies with isolating internal layer, we have taken the complex bulk dielectric constant in the form of the Lorentzian oscillator. The dielectric response can be seen in Fig. 2. Similar to the previous case, the reduction of the real part of the dielectric constant indicates peak of its imaginary component showing the maximum loss.
L = 50
in
2
which
volume
fraction
has been used. The static dielectric constant is found around 1000. In the morphologies with nonpercolating conducting region, we look for the dielectric response by taking the dielectric constant and volume fraction of the blocking layer to be 102 and f L =10-3, respectively. The real-(upper-) and imaginary-(lower curve) parts of the dielectric constant can be seen in Fig. 1(b). In the real part of
f L = 5 x10
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
(a) (b) Figure 3. Real (a) and imaginary (b) parts of the CCTO dielectric constant under unblocked morphologies with conducting internal boundaries. Three curves are respectively plotted for three values of the a and c
2 4 6
part. We have used a conductivity of ellipsoid i = 2x106 S m-1 in the a/c variation. The most significant difference seen in the dielectric constant curves is in the low frequency region until 110 cm-1.
ratio: 10 , 10 , and 10 . There are significant roles played by the ratio of their semi-major and semi-minor axes (a/c ratios). In Fig. 3 we can see that the smaller the ratio the bigger the volume fraction of conducting internal boundaries, implying the ascent of the real part of dielectric constant, so do for the imaginary
National Education under the contract Surat Perjanjian Pelaksanaan Penelitian No. LPPMUGM/643/2008 on 5 March 2008.
References
[1] C. C. Homes, T. Vogt, S. M. Shapiro, S. Wakimoto, and A. P. Ramirez 2001, Optical Response of HighDielectric-Constant PerovskiteRelated Oxide, Science 293, 673676. M. A. Subramanian, D. Li, N. Duan, B. A. Reisner, and A. W. Sleight 2000, High Dielectric Constant in and Acu3Ti3FeO12 Acu3Ti4O12 Phases, J. of Solid State Chem. 151, 323-325 M. H. Cohen, J. B. Neaton, L. He, and D. Vanderbilt 2003, Extrinsic models for the dielectric response of CaCu3Ti4O12, J. of Appl. Phys. 94(5), 3299-3306 L. He 2003, First-Principles Study of Complex Material Systems, PhD dissertation, University of New Brunswick, New Jersey. I. Hilmi, and Kamsul Abraha 2008 Theoretical and computational study of surface phonon-polariton modes on new giant dielectric under material CaCu3Ti4O12 extrinsic-Lorentz oscillator model, Proceedings of The 2nd International Conference on Mathematics and Natural Sciences, ITB Bandung, 28 30 October 2008 C. C. Homes, T. Vogt, S. M. Shapiro, S. Wakimoto, M. A. Subramanian, and A. P. Ramirez 2003, Charge transfer in the high dielectric constant materials CaCu3Ti4O12 and CdCu3Ti4O12, Phys. Rev. B 67, 092106 P. S. Neelakanta 1995, Handbook of electromagnetic materials, CRC Press, Florida, US.
'
[2]
Conclusions
We have discussed the morphologies that are possibly responsible in the CCTO giant dielectric phenomena. In the blocked morphologies, the giant dielectric response is guaranteed by its blocking layer associated with its conducting regions. The difference in the dielectric constant curve in the infrared frequency region for both blocked cases is caused by whether it becomes percolating or nonpercolating, hence the macroscopic dielectric constant. The peaks formed at the curves show the phonon contribution at certain temperature. In the unblocked morphologies, the dielectric response is found to be highly dependent on the materials domains and their associated boundaries. This shows how special treatment has to be done at the fabrication of the material that can determine the geometrical microscopic structure of the material. In the conducting internal boundaries case the dielectric response is highly dependent on the microstructure of the internal blocking layer, and also on their eccentricity and their conductivity. In the morphologies with insulating internal boundaries, the effective medium approximation has been used. In addition, the phonon contribution has been found coming out from the dielectric constant of the bulk material because of its dominating volume fraction in the material. Further empirical study is required for each morphology in order to find the most appropriate morphologies under the extrinsic-Lorentz oscillator model especially at the nano-size probe. Hopefully this will provide evidence what mechanism is really responsible for the giant dielectric phenomenan, namely, by considering the most appropriate morphologies. Finally, we greatly appreciate financial support from the DP2M-DIKTI Fundamental Research Grant, Department of
[3]
[4]
[5]
[6]
[7]
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
Fabrication and Characterization of Membrane Based on Taste Sensor to Classify Five Types of Basic Taste
Ummi Kaltsum1, Kuwat Triyana1, Dwi Siswanta2
Physics Department, Gadjah Mada University Chemistry Department, Gadjah Mada University Sekip Utara BLS 21, Yogyakarta 55281, Indonesia e-mail: um_mik@yahoo.co.id
2 1
Abstract
A lipid/polymer membrane based on taste sensor has been fabricated by a mixture of chemical substances. The membrane was made to classify five types of basic taste quality. In addition, it is designed to provide an alternative method for chemical analyses. For this purpose, it was fabricated by employing the mixture of oleic acid, dioctyl phthalate (DOP), polyvinyl chloride (PVC), and tetrahydrofuran (THF) with a particular ratio. The oleic acid as a lipid determines the selectivity to ion or molecule of samples. In this case, the lipid is a transducer that changes the concentration difference to the electrical signal (electrical current) and subsequently results in the output as electrical potential which can be measured in ion meter. The samples were used in this research are glucose, KCl, quinine hydrochloride, HCl (E. Merck) and monosodium glutamate (Ajinomoto). These samples were prepared at different concentration, where all substances were dissolved in 1 mM KCl solution. As a result, the membrane gave potential responses to all samples at the concentration of 1 mM. This value is lower than that of recognition threshold of humans tongue. It means that the membrane is more sensitive and superior compared to that of humans tongue. In addition, the membrane also response to all samples which produces five types of taste characteristic by showing different potential values. It implies that the membrane has global selectivity as in biology. However, due to the response time of 1 mM KCl as a reference solution, the response times of the membrane is still too long about six minutes Keywords: membrane, lipid/polimer, taste sensor
Introduction
Human always create the new equipments to make their life easier. Many types of equipment are made based on the biology systems, for example high frequency sonar system at dolphin gives inspiration for dive ship, machine or robot is inspired by muscle, and odor and taste sensor are developed from nose and tongue. So that appear biomimetic, it is a branch of science which try to catch ideas from living organism then develop it to be technology products which is inspired by natural phenomenon. Nevertheless, for odor and taste sensor are still new, and the difficulties to make these sensor are selectivity and sensitivity. It is caused by difficulties in mechanism of selection chemical substances that gives specific taste information. For example, in tea or coffee, there are over 1000 chemical substances. Human can not distinguish each chemical substance. Instead, human decompose the taste of foodstuffs into five types of basis taste quality: sweet, bitter, salty, sour, and umami (Toko, 2000). Taste is constructed from five basic taste qualities as above: sweetness produced by sucrose,
glucose, and fructose, KCl; bitterness by quinine, caffeine and MgCl2; saltiness produced by NaCl, KCl, and KBr; sourness by H+ of HCl, acetic acid, and citric acid; and umami produced by MSG, IMP, and GMP (Iiyama et al, 1998). The main methods of measurements of taste quality are sensory evaluations by humans (tester) and conventional chemical analyses. However, human sensory is not objective, it depend on subjectivity, emotion and healthy of the tester. This conventional chemical analysis has low selectivity, difficulty to make selective sensor to all chemical substances, and only chemist who can use it. Therefore, in this article we report our preliminary result in developing membrane based on taste sensor that can classify of five basic taste qualities by means of output signal for each taste quality. The sensor uses lipid membrane as transducer of taste substances (Nagamori, et al, 1999). The applications of taste sensor including to classify wine (Riul, A., 2004), to analyst tomato (Katrien, et al, 2007) and to evaluate the astringency of black tea (Hayashi, et al, 2007).
Materials and Methods
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
To fabricate the membrane, the oleic acid as a lipid was mixed in the test tube containing polyvinyl chloride and plasticizer (DOP, dioctyl phthalate) dissolved in tetrahydrofuran. The ratio of the mixture is 64 mg : 28 mg : 8 mg for DOP, PVC, and oleic acid, respectively. The mixture is then stirred and stored it in the room temperature. The lipid membrane was transparent and soft film with the thickness of about 300 m (Kadidae, 2000). For the testing material, we employed the substances of the commercial products consisting of glucose, KCl, quinine hydrochloride, HCl from E. Merck and MSG from Ajinomoto, Japanese product. These samples were made at different concentration.
electrode consists of Hg wire where its surface was plated with Hg2Cl, while its internal cavity is filled with 3 M KCl solution. To measure the performance of membrane, these electrodes are connected to ion meter.

Figure 3 shows the response of membrane for all samples. It is clear that the response of membrane is influenced by different concentration of each sample indicated by difference of its electrical potential. Therefore, the taste quality can be distinguished based on its potential patterns. In this preliminary research, this membrane could work properly. As the concentration samples increases, the electrical potentials of membrane also increase. It is because the cations (X+) of samples are bounded by the negative charge of membrane (RCO2-). The scheme of the binding process is shown in Fig. 4. The cations of KCl, quinine hydrochloride, HCl, and MSG are brought by K+, quinine ion, H+ and Na+, respectively. However, since glucose is nonelectrolyte, the response potential of membrane to the glucose is low. Compared to the humans tongue this membrane is more sensitive and superior. The recognition threshold of human is 170 mM for sucrose (sweet), 30 mM for NaCl (salty), and 2 mM for MSG, while this membrane is able to response at very low concentration of 1 mM for all samples. This sensitivity of this membrane is caused by ion samples that interfered very strongly to the membrane and bounded to the active material of membrane results in stable complex.
Figure 1 Schematic of membrane characterization To characterize the membrane, all substances were dissolved in 1 mM KCl solution. Moreover, this fresh 1 mM KCl solution was also used as a reference sample and also as a rinsing fluid for the electrode after each measurement. The electrical potential response of the sample was measured as the difference between the potentials of the sample and 1 mM KCl reference solution. Meanwhile, each measurement time was set to be 6 minutes. In this case, the measurement was performed using ion meter (TOA Electronic Ltd IM-20E model) with the reference electrode and working electrode made by lipid/polymer membrane is shown in Fig 1.
Wire
Electrode body
Internal solution (100 mM KCl) Ag/AgCl
Membrane spacer
Membrane
Fig.ure 3 Potential responses to all samples Based on Fig. 3, this membrane is not selective to specific sample. This implies that the membrane has global selectivity, as in the taste reception in the biological systems. The taste quality is produced by interactions between sample molecules and membrane, namely taste is not 2
Figure 2 Working electrode The working electrode consists of an Ag wire where its surface was plated with AgCl, while its internal cavity filled with 100 mM KCl solution. The lipid membrane was adhered to the opening of the tube (Fig.2). In other hand, the reference
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
specific of characteristic of each molecule. The five types of taste quality reflect the differences in the above interactions. The response time of all samples is referred to response time of 1 mM KCl as a reference solution. The response time is the time taken of membrane to give stable potential response. In the case of our membrane, the response time is still too long, that is about six minutes. It may be because there is interfering other unexpected ions and also because the diffusion process of cations into the membrane takes place very long time.
Figure 4 Interaction between cation and membrane (Evan, 1987)
Scope and Limitation, Biosci. Bioetanol. Biochem., 71 (2), 587-589 Iiyama, S., Yasuji, I., and Kiyoshi, T., 1998, Measurement of Umami Subtances Using Multichannel Taste Sensor with Lipid Membranes, Sensors and Materials, vol. 10, no. 8,475-485 Kadidae, L.O., 2000, Sintesis Benzileugenol dan Pemanfaatanya sebagai Komponen Membran Selektif Ion, Tesis, Universitas gadjah Mada, Yogyakarta. Nagamori, T., Kiyoshi, T., Yukiko, K., Takehiko, W., and Kenichi, E. , 1999, Detection of The Suppression of Saltiness by Umami Substances Using a Taste Sensor, Sensors and Materials, vol. 11, no. 8,475-485. Riul, A. , Humberto C. de Sousa b, Roger R. M., David S., Andr C., Fernando J., Osvaldo N., and Luiz H., 2004, Wine Classification by Taste Sensors Made from Ultra-Thin Films and Using Neural Network, Sensors and Actuators B 98, 77-82 Toko, K. 2000. Biomimetic Sensor Technology. Cambridge University Press, United Kingdom
Conclusion
We have investigated the performance and physics mechanism of membrane based on taste sensor. The membrane is able to distinguish five types of basic taste qualities (sweet, salty, sour, bitter and umami). In addition, this membrane has global selectivity as biological systems. However, the response time is still too long and needed further improvement.
Acknowledgement
The authors would like to thank Material Physics Laboratory for sponsoring this research.
References
Baullens, K., Peter M., , Steven, V., Dmitriy K., Andrey, L., Saskia, B., Nathalie, C, Bart, M., and Jeroen, L., 2008, Analysis of Tomato Taste Using Two Types of Electronic Tongues, Sensors and Actuators, B 131, 10-17 Evans, A., 1987, Potentiometry and Ion Selective Electrode, Analytical Chemistry by Opening Learning, London Hayashi, N., Ronggang, C., Hidekazu, I. Tomomi, U., Hideyuki, K., and Yuzo, M., 2007, Evaluation of the Astringency of Bleack Tea by a Taste Sensor System:
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
Determination of Theoretical Dispersion Curves on the Surface Phonon Polaritons in PTCDA

Asih Melati1 dan Kamsul Abraha1 Theoritical Group of Material Physics Physics Department Gadjah Mada University Abstract
Theoretical and computational studies of the polariton dispersion curves on the surface of the PTCDA organic material have been done. By applying the Modified Lorentz Oscillator Model (MLOM) of PTCDA (3,4,9,10 Perylene tetracarboxylic dianhydride/ C24H8O6 ) has the dispersive dielectric function
( )
depending on the of wave frequency
. The direction of wave propagation is perpendicular
to the easy axis. This study gives theoretical description of the PTCDA surface and bulk modes at frequency range =1.5 6 cm-1.
2 c
The dispersion relation is obtained through Maxwell equation. Subtitution of
magnetic ( H ) fields into the wave differential equation and application of boundary conditions will lead to the dispertion relations. The dynamics of internal or system parameters is included in the material permittivity. It has been found that the surface polaritons only propagate in TM (Transverse Magnetic) mode and prove are reciprocal in the sense that the change of wave vector r r direction does not cause the change of frequency ( q ) = q . There is no surface polariton propagating in the TE (Transverse Electric) mode, but only bulk polaritons can propagate in this mode. The computed of ATR (Attenuated Total Reflection) spectra show that the surface polaritons can be detected optimally when the gap distance between the prism and the sample reaches an optimum value of 0.8 m . Not Only variating the gap of distance between the prism, but also the effect of damp on the methods. In this methods we are also trying to compare using of silicon prism and the other prism which different permittivity to reaches an optimum value of p = 11 . 56 . The surface polariton
uu r
electric ( E ) and
ur
modes are clearly verified in their reflectivity spectra, which are in agreement with the related dispersion relations. Keywords: PTCDA, Surface Polariton, ATR, Reflectivity
Introduction
For more than 40 years the phonon polariton problem originally formulated by Jansen9 and Pekar9 has been of permanent interest for intensive investigations with a strong revival in recent years due to novel theoretical and experimental studies. In this research examines the surface polariton modes in organic material PTCDA by showing dispersion curves and ATR spectrum. This organic material used in semiconductor manufacture,OFET, OLED, photovoltaik cells, fotodetektor, etc. The properties of PTCDA phenomenon is still quite strange and need to be studied further. Maxwell equations :
r .D = f r .B = 0
(1) (2)
r r D (4) H = J f + t PTCDA (3,4,9,10 perylene tetracarboxylic dianhydride) are organic conductive material which has their semiconductor. Dielectric function of PTCDA can be obtained with the method LOM (Lorentz Oscillator Model) and MLOM (Modified Lorentz Oscillator Model). The MLOM more effective than the LOM mode due to the MLOM is suitable for a curve-fitting multi-wavelength approach to determine the optical constants of evaporated dye layers from the normal incidence R and T measurements [6]. The Applied MLOM model for the optical dielectric function, the real and imaginary parts of the complex dielectric function.
1
u r ur B E = t r
(3)
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
1 ( ) =1 ( ) + i 2 ( )
1 ( ) =1 +
j=i k
Fj
2 j
2 j
2 j 2
) + ( )
2 j
j 2 j
(5)
2
condition of H and E , suitable with attenuated constanta in medium and vacuum space [1]. Wave equation there are :
uu r
ur
2 ( ) =
p
j=i
F j
2
(6) (7)
) + ( )
2
h E j
(8)
Attenuated constanta in vacuum and medium space there are :
u u r r E0 ' = E0 exp( iqx +0 y) exp( i't ) r r E = E1 exp(iqx y) exp(it ) r 2 r 2 E + 22 E = 0 r r c 2 E + 2 ( ) E = 0 c
y<0 y>0 y<0 y>0
(9) (10) (11) (12)
Calculating Method
Dispersion curves of the surface phonon polariton reviemed in the interface (y = 0), between the vacuum (y < 0) and in the medium (y > 0). Wave propagates in the direction of x and vanish exponentially in the y axis with the constant shedding. q
x Z
02 = q2
2
c
2
and
2 = q2
2
c
2
()
(13)
boundary condition electric field is E0 x = E1x s v and magnetic field H 0 z = H 1 z gives this disperstion relation : (14)
( ') 0 + = 0
The appearance of linear term q in the form dispersion relation show that surface polariton is reciprocal
Figure 1 Propagation geometry of phonon polariton with wave vectors q The surface polariton dispersion derived from Maxwell equation. This is parameters table used to calculates dispersion curve and ATR spectra. Table 1 Parameters PTCDA [6] J 1 2 3 4 5 6
q = q
( ) ( ).
EE
2.221 2.449 2.641 3.356 3.943 5.379
j
2.212 2.480 2.720 3.464 3.967 3.620
Fj
0.571 3.645 1.313 1.301 0.762 66.316
j
0.118 0.367 0.400 0.339 0.542 0.605
Ej
1.943 0.637 0.243 1.500 1.044 0.913
q (cm1 ) 2

Dispersion Relation Dispersion relation can be obtained from applied Maxwell equation. Its considering boundary
Figure 2 Dispersion curve of bulk polariton modes (hatched) and surface polariton modes (thick lines) on the frequency of 1.5 - 6 cm-1 [12] Fig 2 shows dispersion curves on the frequency of 1.5 - 6 cm-1, there are 5 surface polariton modes be signed in thick line and this is located in the around the frequency = 2.212 cm-1, 2.480 cm-1, 2.720 cm-1 , 3.464 cm-1 and 3.943 cm-1. In this frequency scale we 2
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
also show 6 bulk polariton which be signed in hatching For a big size figure can be shown above
k = kC =
p x
1 2 p
'
c
1 2 p
sin
cos
2
(17) (18)
k yp = k =
'
c
0
v y
=
=
'
'
c
(
0
sin
p
1 2
(19) (20)
sin
1 2
k
q (cm1 ) 2
s y
2 ' = k 42 c
sin
1 2
(21)
Fig 2. Dispersion curve of bulk polariton modes (hatched) and surface polariton modes (thick lines) on the frequency of 3.2 4.2 cm-1 [12] There are simetry in their opposite frequency so we called its reciprocal. Attenuated Total Reflection If we are applied boundary condition in the boundary surface prism-vacuum, we can obtain nisbah of incidence angle to refraction angle. This last result of it called ATR. If refraction angle smaller than incidence angle, it is become imaginer, and there is no energy flow in boundary surface. Because of it, evanescent waves will arise. Evanescent wave in material can be detected using high refraction index medium which located in the gap distance d from the sample, so there is no total refraction. The familiar ATR configuration is Otto configuration, the other name is called PAM system (prism-air-medium) [2]. The critical angle is defined as the smallest incidence angle for which an optical ray, in the absence of absorption, will be totally reflected at the interface. It is equal to.
The descending curves occurs if photons and fonon coupled. In the same condition, there are abbreviation between ATR line and surface polariton modes. The deference of low high level curve depend on the variety of modes. By variating the gap distance, incidence angle, damping and dielectric constant can be detected existence polariton.
( j ) 1 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 SP2 0 1.5 2 2.5 SP3 3 3.5 SP4 4 4.5 5 5.5 6 SP5 SP1 R
2 c
Figure 3 Reflectivity of ATR by variating the gap distance:(blue:0,8 m; red:8m and green:80m) [12] Fig 3 shows that the distance between the prism-sample influence the formation polariton on the sample surface. By variating the distance, in which incidence angle and damping constant, obtained optimum value of distance is 0,8 m . We are also variating the incidence angle, in which the gap distance and damping contant. Its obtained the optimum value of incidence angle is 600 suitable fig.4 above
c = sin 1
R= r =
(15) where n1 and n2 represent the refraction index of the analyte and the ATR element, respectively. The calculation of the material PTCDA using TM modes. Prism used have a high index of refraction (silikon prism). Reflektivity of ATR and the other parameter provided by Abraha and Tilley [1] 2 (16) k (1 + f exp( 2d )) i (1 f exp( 2d ) 2
k (1 + f exp( 2d )) + i (1 f exp( 2d )
np nu
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
ATR(variasi theta) 1 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 SP2 0 1.5 2 2.5 SP3 3 3.5 SP4 4 4.5 5 5.5 6 SP5 SP1 R
In the last variating, we are trying to compute adjustable the using of prism, although, in this result mention that silicon prism dielectric contant can not reach the optimum value, in the lab works not yet obtained the effective prism except silicon prism which has high dielectric contant p = 11,56 . From the third graph of reflectivity can be seen that there is existence phonon polariton in frequency = 2.212 cm-1, 2.480 cm-1, 2.720 cm-1, 3.464 cm-1and 3.943 cm-1. There are accordance between relation dispersion curves and ATR curves, since the dispersion relation exceeded bulk from the vacum in the ATR curve appearance shows an polariton, its called bulk phonon polariton. An the other hand if the dispersion curve exceeded the vacuum, ATR curve appearance in the the value 1.
2 c
Figure 4 Reflectivity of ATR by variating incidence :450;blue angle (green :300;red 0 :60 )[12] Next, we are variating the damping to reach the optimum value, its obtained the optimum value of damping is 0.02 suitable plotted in Fig.5.
ATR(variasi damp) 1 0.9 0.8 0.7 0.6 R 0.5 0.4 0.3 0.2 0.1 SP2 0 1.5 2 2.5 SP3 3 3.5 SP4 4 4.5 5 5.5 6 SP5 SP1
Conclusion
We have modeled dispersion curve on surface polariton in organic molecul PTCDA, based on the research can be can be summed up as follows: 1. 1.Propagation modes of organic molecul PTCDA is TM (Transverse Magnetic) mode that electric field ( H ) perpendicular to waveguide direction. 2.Direction of surface polariton propagation on material PTCDA is reciprocal because vektor propagation direction does not change
uu r
2.
2 c
Figure 5 Reflectivity of ATR by variating damping (green:0.05;red:0.02) [12] And the last, we are try to experiment suitable prism with reflectivity of ATR
1 0.8 0.6 R 0.4
3. 4.
q = q and the frequency reflectance does not depend on incidence angle reversible R = R . ATR calculation can be used to prove the existence of surface phonon polariton in materials PTCDA. 4.Variating the sample distance to the prism, incidence angle and damping reaches optimum value of
() ( )
( ) (
d = 0.8 m , = 600 and damping 0.02.

Acknowledgement
0.2 0 1.5
2.5
3.5
4.5
5.5
2 c
We gratefully acknowledge supported by Kamsul Abraha for stimulating discussion and Department of physics Gadjah Mada university.
Figure 6 reflectivity of ATR by variating prism dielectric contants
References
[1] Abraha, K., Brown, D.E., Dumelow, T., Parker. TJ, and Tilley, DR, 1994, Oblique-
Proceeding Book

January 24, 2009 ISBN : 978 979 19201 0 0
incidence Far-Infrared Reflectivity Study of the Uniaxial Antiferromagnet FeF2, Phys. Rev. B 50, p. 6808-6816. [2] Aers, G.C. and Boardman, AD, 1980, attenuated Total Reflection Analysis of Surface Polaritons, Programs in Physics, editors: AD Boardman, John Willey & Sons. [3] Barnas, J., 1986 Elektromagnetic Modes in Magnetic system with linear MagnetoElectric Interactions, Journal of Magnetism Materials 62, p. 381-391 [4] Burstein, 1972, Introductory Remarks On Polaritons, Polaritons in the case 1-3, [5] Cottam, MG and Tilley, DR, 1989, Introduction to Surface and Superlattice Excitations, Cambridge University Press, New York. [6] Djurisic, 2000, Modeling the Optical Constant of thin Organic Films: Application to 3,4,9,10Perylenetetracarboxylic Dianhydride (PTCDA), the Institute fur Angewandte photophisik, TU Dresden, Mommsenstr.13D-01062 Dresden, Germany [7] Forest, SR, 2003, Organic-inorganic semiconductor devices, and 3,4,9,10 perylene tetracarboxylic dianhydride: an early history of organic electronics, Journal of Condensed Matter Physics, 15 (2003) 52599-52610. [8] Griffith, J. 1989, Introduction Classical Electrodynamics, John Wiley & Sons, New York. [9] Jensen, M.R.F., Feiven, S.A, Parker, T.J. and Camley, RE, 1996, Experimental Determination of Magnetic Polariton Dispersion Curves in FeF2, Physical Review B 55, p.. 2745-2748. [10] Mills, D.L. and Burstein, E., 1973, Polaritons: The Electromagnetic Modes of Media, Department of Physics, University of California, Irvine, California, USA, and Department of Physics and Laboratory for Research on the Structure of Matter, University of Pennsylvania, Philadelpia, Pennsylvania , USA. Pergamon Press, Inc., New York. [11] Otto A, 1976, Spectroscopy of Surface Polaritons attenuated by Total Reflection: Optical Properties of Solid, New Development, edited: B. O. Seraphin, North Holland Co.. Ltd., Oxford, 679-729 .. [12] Melati. A., 2008, Theoretical And Computational Studies phonon polariton in Organic Material PTCDA, Department of Physics, FMIPA UGM Yogyakarta
Proceeding Book

Study on Matrix Homogeneity and Interfacial Zone of sodium-poly(sialate-siloxo) (Na-PSS) Geopolymers

Transféré par

Informations du document

Description originale:

Titre original

Copyright

Formats disponibles

Partager ce document

Partager ou intégrer le document

Options de partage

Avez-vous trouvé ce document utile ?

Ce contenu est-il inapproprié ?

Droits d'auteur :

Formats disponibles

Study on Matrix Homogeneity and Interfacial Zone of sodium-poly(sialate-siloxo) (Na-PSS) Geopolymers

Transféré par

Droits d'auteur :

Formats disponibles

The First International Seminar on Science and Technology

January 24, 2009 ISBN : 978 979 19201 0 0