fast learning algorithm for fuzzy inference systems using ... · this time, we apply multiple...

International MultiConference of Engineers and Computer Scientists 2016 (IMECS 2016), Hong Kong, 16–18 March, 2016.

Fast Learning Algorithm for Fuzzy Inference Systems using Vector Quantization

Hirofumi Miyajima *, Noritaka Shigei**, and Hiromi Miyajima***

Abstract Many studies on modeling of fuzzy inference systems have been made. Their aim is to construct automatically fuzzy systems from learning data based on steepest descent method [1]. Further, there is difficulty for learning with high dimensional spaces [2]. In order to overcome them, learning methods using VQ and SDM are proposed and they are superior in the number of rules to other methods, but they need a great deal of learning time [3]. The cause seems to be both local searches. On the other hand, it is known that learning method of RBF (Radial Basis Function) networks using Vector Quantization and Generalize Inverse Method is much fast compared to other learning methods.

In this paper, we propose a new learning method composed of iterating three stages. It starts by breaking the method into three stages: learning in the first stage, intermediate stage of adjusting the center and width parameters, and the next stage of updating the weight parameters using the generalized inverse method. As for final stage, three parameters are updated by learning based on SDM. In order to demonstrate the validity of the proposed method, numerical simulations for function approximation and pattern classification problems are performed. References [1] M.M. Gupta, L. Jin and N. Homma, Static and Dynamic Neural Networks, IEEE Press, 2003. [2] B. Liu, Theory and Practice of Uncertain Programming, Studies in Fuzziness and Soft Computing, Vol. 239,

Springer, 2009. [3] Hirofumi Miyajima, Noritaka Shigei, Kazuya Kishida, Yusuke Akiyoshi and Hiromi Miyajima, “An Improved

Learning Algorithm of Fuzzy Inference Systems using Vector Quantization”, Advanced in Fuzzy Sets and Systems, vol.21, no.1, pp.59–77, 2016.

* Graduate Student, Department of Electrical and Electronics Engineering ** Associate Professor, Department of Electrical and Electronics Engineering *** Professor, Department of Electrical and Electronics Engineering

- 58 -

IEICE Thailand-Japan MicroWave 2015, Bangkok, Thailand, 6–8 Aug., 2015.

Technique to Expand Power Transfer Area on Short-range WPT system

Tatsuro KUWAHARA*, Ryo TAKAMORI*,

and Kenjiro NISHIKAWA** Abstract This paper investigated and clarified the contribution of a Multi-Input-Multi-Output (MIMO) wireless power transfer (WPT) system to the energy transmission efficiency on short distance WPT system. To evaluate the improvement of the energy transmission efficiency, a 4x4 MIMO WPT system was employed. The MIMO system effectively enhances the power transfer efficiency and expands the power transfer coverage in our investigation. The effective coverage of the MIMO wireless power transfer is 1.6 times wider than that of the 4x4 array system without MIMO control. References [1] Ken Hiraga, Tomohiro Seki, Kentaro Nishimori, Kenjiro Nishikawa, and Kazuhiro Uehara, “Ultra High Speed Short

Range Parallel Transmission Using Millimeter Wave Frequency Band,” 2009 Asia Pacific Microwave Conference Proceedings, pp.1212–1215, Dec. 2009.

[2] Minh Wuoc Nguyan, Dakota Plesa, Smitha Tao, and J.C.Chiao, “A Multi-Input and Multi-Output Wireless Energy Transfer System,” Microwave Symposium (IMS), 2014 IEEE MTT-S International, pp.1–3, June 2014.

[3] R.Zhang and C.K.Ho, “MIMO Broadcastiong for Simultaneous Wireless Information and Power Transfer,” IEEE Transaction wireless Communications, vol.12 No.5, pp.1989–2001, May 2013.

* Graduate Student, Department of Electrical and Electronics Engineering ** Professor, Department of Electrical and Electronics Engineering

- 59 -

IEICE Thailand-Japan MicroWave 2015, Bangkok, Thailand, 6–8 Aug., 2015.

Modeling of Grounded CPW Line with Anomalous Skin Effect in THz Band

Yuta SAKIYAMA*, Masahiro MURAGUCHI*, Hiroto SAKAKI*,

and Kenjiro NISHIKAWA** Abstract Recently, the terahertz (THz) frequency band has been focused and investigated for developing new applications, such as an ultra-high-speed wireless system, and so on. These new system have demanded a highly integrated and compact IC, such as CMOS ICs. Transmission loss on the ICs is the most important issue due to insufficient device performances in THz band. This paper models grounded CPW line with loss characteristics, including anomalous skin effect. The anomalous skin effect is characterized by the proposed donut-style conductor structure. The calculated loss of a grounded CPW line constructed by the donut-style conductor model is 2.5 times higher than those of previous models. References [1] 2013 IEEE MTT-S IMS Workshop, WMD: Technologies for THz Integrated System, Seattle WA, June 2013. [2] Akira Tsuchiya and Hidetoshi Onodera, “Gradient Resistivity Method for Numerical Evaluation of Anomalous Skin

Effect,” 2011 IEEE Workshop on Signal Propagation on Interconnects, pp.139–142, Nepal, May 2011. * Graduate Student, Department of Electrical and Electronics Engineering ** Professor, Department of Electrical and Electronics Engineering

- 60 -

11th International Conference on Cyber Warfare and Security (ICCWS 2016), Boston, USA, 17–18 March 2016.

Truncated Differential Attack on Block Cipher PRINCE

Satoshi Setoguchi*, Yasutaka Igarashi**, Toshinobu Kaneko**, Kenichi Arai***, and Seiji Fukushima****

Abstract Now that networking is advanced, a variety of information is transmitted through its information network all over the world. Therefore confidentiality of the information is very important, and a variety of security technology has been established. A block cipher algorithm is also one of them. In order to be used secure, it needs to be evaluated its security by a third party. In this study we focus on the block cipher PRINCE and evaluate its security. PRINCE is an SPN-type 64-bit block cipher with a 128-bit key proposed by Borghoff et al. in 2012. The number of round functions of PRINCE is designed as 12. Although the designers have stated that differential attack, linear attack, algebraic attack, and biclique attack would not be a threat to the security of PRINCE, we evaluate its security against truncated differential attack from a third party standpoint. Differential attack was proposed by Biham et al., and it is based on the stochastic event of differential path caused by the property of nonlinear function used for an encryption process. Truncated difference attack was proposed by Lars Knudsen, and it considers a difference that is only determined to a limited extent, e.g. zero and nonzero difference. In 2014 Anne et al. reported the truncated differential attack on 10-round PRINCE, which requires 2 to the 57.94th power pairs of chosen plaintext and ciphertext, and 2 to the 118.56th power times of encryption operation. This time, we apply multiple round-elimination method to the 1st and the 2nd rounds of PRINCE. From the 3rd round to the 9th round, we construct differential path. On the 10th round, we construct truncated differential path. As a result, we can attack 11-round PRINCE with 2 to the 62.81th power pairs of chosen plaintext and ciphertext, and 2 to the 106.82th power times of encryption operation. References 1) J. Borghoff et al., “PRINCE - A Low-Latency Block Cipher for Pervasive Computing Applications,”

http://eprint.iacr.org/2012/529.pdf 2) S. Setoguchi et al., “Truncated Differential Attack on Block Cipher PRINCE,” IEICE Tech. Report, vol. 115, no. 28,

ISEC2015-2, pp. 9–14, May 2015. (in Japanese) 3) E.Biham et al., “Differential Cryptanalysis of DES-like Cryptosystems,”

http://sota.gen.nz/crypt_blues/biham91differential.pdf 4) A. Canteaut et al., “Multiple differential cryptanalysis of round-reduced PRINCE,”

https://eprint.iacr.org/2014/089.pdf 5) Toshinobu Kaneko, “Security evaluation of symmetric-key cipher,”

https://www.jstage.jst.go.jp/article/essfr/7/1/7_14/_pdf (in Japanese) * Graduate Student, Department of Electrical and Electronics Engineering ** Lecturer and Professor, Department of Electrical Engineering, Tokyo University of Science *** Graduate School of Engineering, Nagasaki University **** Professor, Department of Electrical and Electronics Engineering

- 61 -

International Association for Shell and Spatial Structures (IASS), Symposium 2015 Amsterdam, 17 - 20 August 2015, Amsterdam, The Netherlands

Form-Finding Analysis for Membrane Structure with Cable Using Geometric Energy Minimization

Takuya Satonaka 1, Yohei Yokosuka 1 and Toshio Honma 1

1 Department of Architecture &Architectural Engineering, Kagoshima University, 890-0065, Kagoshima, Japan

Abstract In the membrane structure, a structural stiffness is given by introducing the geometric rigidity. Therefore, the

form-finding analysis is necessary to obtain the structural stiffness effectively. In the form-finding process of the membrane structure, the minimal surface is used for a design prefiguration shape. Because, the shape of the minimal surface is equivalent to the curved surface of the uniform stress field. The discretization procedures with finite element technique are generally adopted and are used for the minimal surface analysis (as shown in reference [1]). However, in some ill-conditioned cases, these procedures are difficult to obtain a solution of the minimal surface without the special approach such tanas reduction of variables.

In other hands, discrete differential geometry is enthusiastically researched in mathematics for applied digital geometry processing and physical solution. The Willmore energy is one of the functional of discrete differential geometry. This energy of a surface is defined as the mean and Gaussian curvature (as shown in reference [2]). It is shown that the minimal surface exists at the critical point of the Willmore energy. In addition, the Willmore energy can be described as a discrete curved surface preserving invariant of a continuum curved surface. In the computation using this energy, a numeric steady approximate solution is obtained with simple algorithm.

In this paper, at first, we introduced the Willmore energy by the continuum and discrete formulation, and evaluate this energy in comparison with the area functional. This functional is computed on the discretized catenoid model. Furthermore, we show the form-finding analysis method of the membrane structure with cable, and verify accuracy of solutions and effectiveness of this method.

References 1. T. Suzuki and Y. Hangai, Shape Analysis of Minimal Surface by the Finite Element Method, Spatial Structures at

the Turn of the Millennium, Vol.2, Structural Form, 1991. 2. Alexander I. Bobenko and P. Schroder, Discrete Willmore Flow, Eurographics Symposium on Geometry

Processing, 2005.

- 62 -

NDSU-KU Joint Symposium on Biotechnology, Nanomaterials and Polymers, North Dakota, 15-16 Oct, 2015

Systematic synthesis and interaction analysis of heparan sulfate partial structure containing glucuronic acid moiety

Rikiya Aramaki*, Nao Matsuyama*, Masahiro Wakao**, Yasuo Suda***

Abstract Heparan Sulfate (HS) belongs to glycosaminoglycan (GAG) superfamily is a liner sulfated polysaccharide and is widely distributed in various tissues as components of cell membrane or extracellular matrix. HS is often sulfated heterogeneously owing to the multiple and random enzymatic modifications in the biosynthesis. The specific microdomain structure in HS chain is considered to be responsible for their specific binding interaction. However, the elucidation of the structure-activity relations of HS with HS-binding protein is very difficult due to their naturally occurring structural diversity. For that reason, structurally defined partial structures are required to analyze the functions at the molecular level. In this study, we systematically synthesized HS disaccharide structures containing glucuronic acid from a common intermediate and analyzed structure-activity relations of HS with HS-binding proteins using surface plasmon resonance (SPR) imaging.

Figure 1. Syntheis of ligand conjugate containing HS partial disaccharide and flow of the interaction analysis using SPR imaging

References 1) Y. Suda, A. Arano, Y. Fukui, S. Koshida, M. Wakao, T. Nishimura, S. Kusumoto, and M. Sobel, Bioconjugate

Chem., 17, 1125 (2006). 2) M. Wakao, A. Saito, K. Ohishi, Y. Kishimoto. T. Nishimura, M. Sobel, and Y. Suda, Bioorg. Med. Chem. Lett., 18,

2499 (2008). 3) A. Saito, M. Wakao, H. Deguchi, A. Mawatari, M. Sobel, and Y. Suda, Tetrahedron, 22, 3951 (2010). * Graduate Student, Department of Chemistry, Biotechnology and Chemical Engineering ** Assistant Professor, Department of Chemistry, Biotechnology and Chemical Engineering *** Professor, Department of Chemistry, Biotechnology and Chemical Engineering

- 63 -


Synthetic study of keratan sulfate disaccharide library

Yuki Kinoue*, Masahiro Wakao**, Yasuo Suda***

Abstract

Keratan sulfate (KS) is a sulfated polysaccharide belonging to glycosaminoglycan (GAG) superfamily. KS is mainly distributed in cornea, cartilage, bone, and brain as components of cell membrane or extracellular matrix. Recently, KS was found on the surface of iPS and ES cells, and a role of KS for the maintenance of cells has attracted attention. KS chain is composed of repeating disaccharide consisting galactose (Gal) and N-acetylglucosamine (GlcNAc) and is sulfated heterogeneously owing to the multiple and random enzymatic modifications during the biosynthesis. The resultant microstructure of KS is related to the specific interaction with KS-binding proteins to regulate their activity. Therefore, the analysis of the structure-activity relations of KS with KS-binding protein at the molecular level is very important for clarifying their biofunction. In this study, we addressed systematic synthesis of KS disaccharide library to evaluate the binding property of KS against KS-binding proteins by surface plasmon resonance (SPR) bio-sensor. Gal donor 1 and GlcNAc acceptor 2 were synthesized from D-galactose and D-glucosamine, respectively. The glycosylation of 2 with 1 gave the disaccharide intermediate 3, which possesses orthogonally removable protecting group and can be converted to KS disaccharides with sulfation. The disaccharide 3 was then condensed with the glucose (Glc) moiety 4, which works as a hydrophilic spacer at the immobilization on SPR sensor chip1) to prevent unexpected non-specific interaction. Currently selective deprotection and sulfation of trisaccharide intermediate 5 are on-going.

Fig. Schematic image of synthesis of KS disaccharide library.

Reference 1) Suda Y, Arano A, Fukui Y, Koshida S, Wakao M, Nishimura T, Kusumoto, S. Sobel, M. Bioconj. Chem. 2006, 17,

1125-1135. * Graduate Student, Department of Chemistry, Biotechnology and Chemical Engineering ** Assistant Professor, Department of Chemistry, Biotechnology and Chemical Engineering *** Professor, Department of Chemistry, Biotechnology and Chemical Engineering

AllylOO

OBz

OO

Ph

OONAPO

PhthN

OTBDPS

SPh

OHONAPO

PhthN

OTBDPS

SPh

AllylOO

OBz

OO

OCCF3

Ph

NPh

AllylOO

OBz

OO

Ph

OONAPO

PhthN OBnOBnO

OBn

OBn

OTBDPS

O

OBnOBnO

OBn

OBn

HOD-galactose 1

D-glucosamine 2 dissaccharide intermediate 3

trisaccharide intermediate 5 R1 = H or a FucR2, R3 = H or SO3-

HOO

OH

HOOO

R1OAcHN OHO

HOOH

OH

OR2

O

OR3

glucose moiety 4

- 64 -


Development of single chain Fv antibody (scFv) which recognizes O-glycns on Adult T-cell Leukemia (ATL) cell surface for targeted therapy

Taro Todaka*, Kaname Muchima*, Masahiro Wakao**, Hikaru Matsumoto*, Yuji Ito***, Yasuo

Suda****

Abstract Adult T-cell leukemia (ATL) is a blood cancer caused by the infection of retrovirus human T-cell lymphotropic virus type-1 (HTLV-1). For an individual infected with HTLV-1, the risk of developing ATL is estimated about 5%. Approximately 10 to 20 million individuals are estimated to be infected with HTLV-1 worldwide. In Japan, about 1 million individuals are infected with HTLV-1, and ca. 1,000 patients of ATL are died each year. Mechanism of carcinogenesis has not yet been revealed, and antibodies for therapeutic use of ATL have not been established. In this study, we developed single chain Fv antibodies (scFv) which recognize O-glycans (O-linked sugar-chains, O-SCs) on ATL cell surface for an effective marker or its novel drug for targeted therapy of ATL using our nano-biotechonology. O-SCs were extracted from cell surface of S1T, which is HTLV-1 infected T-cell line established from an ATL patient. To prepare a fiber type Sugar Chip, obtained O-SCs were conjugated with our original fluorescent linker molecule (designated as f-mono) [1]. O-SC ligand conjugates were then immobilized on the terminal of optical fiber (diameter: 50 mm) via gold nano-particles. Human B-cell derived phage library was firstly negatively screened using peripheral blood mononuclear cells (PBMC) of the healthy volunteer to eliminate phages which bound to PMBC. The phages were then applied to the Sugar Chip to screen the phages specifically bound to O-SCs on S1T cell by the localized plasmon resonance (LPR) method. Obtained O-SC-binding phages were transfected to Escherichia coli, TG1 to prepare monoclonal phages according to the standard protocol. The monoclonal phages were transfected to E.coli, HB2151 to express scFv, and eight kinds of scFvs were isolated and purified. The all purified scFvs showed single band (about 27 kDa) in the SDS-PAGE stained with CBB. By flow cytometry analysis, one scFv, named TS1T-1, bound to S1T and other ATL cell lines, but did not to CEM or other leukemia cell lines, suggesting specific binding to ATL cells. The precise structure of the TS1T-1 binding sugar-chain is currently evaluated with our array-type Sugar Chip and surface plasmon resonance (SPR) imaging.

Fig. Schematic image of preparing O-SC binding scFv.

Reference [1] Sato M, Ito Y, Arima N, Baba M, Sobel M, Wakao M, Suda, Y., J. Biochem. 2009, 146, 33-41. * Graduate Student, Department of Chemistry, Biotechnology and Chemical Engineering ** Assistant Professor, Department of Chemistry, Biotechnology and Chemical Engineering *** Professor, Department of Chemistry and Bioscience **** Professor, Department of Chemistry, Biotechnology and Chemical Engineering

- 65 -

2015 Taiwan/Korea/Japan Joint Meeting on Chemical Engineering, November 5-7, 2015, E-DA Royal Hotel, Kaohsiung, Taiwan

Effects of gas properties on bubble behaviors in fluidized catalyst beds

Keita ETO, Takami KAI, Tsutomu NAKAZATO

Department of Chemical Engineering, Kagoshima University, Kagoshima 890-0065, Japan

Abstract

Bubble size is an important parameter for the reactor model of fluidized beds. Therefore, many researchers have measured bubble size and proposed the equations to predict bubble size (Karimipour and Pugsley, 2011). However, almost all the measurements were carried out using air at ambient temperature, and so the proposed equations ignore the influence of the gas properties such as density and viscosity. As the results, these equations cannot correctly predict the bubble size in fluidized catalyst beds. It has been reported that bubble size is affected by the apparent viscosity of the emulsion phase (Kai et al., 1987b). In addition, the emulsion phase voidage is greater than that at minimum fluidization for the fluidized bed with fine particles, and the apparent viscosity decreased with an increase of the voidage (Kai et al., 1991). Furthermore, because the emulsion phase voidage is affected by the gas properties (Kai et al., 1987a), hence the gas properties affect the bubble size in the fluidized catalyst beds. In this study, the expansion ratio of the emulsion phase and bubble size were measured in a two-dimensional fluidized bed. Five types of gases were used as the fluidizing gas; argon, helium, carbon dioxide, nitrogen and hydrogen. The measurement was carried out using two optical probes, and the bubble size was calculated from the signals from these probes.

The emulsion phase voidage was strongly affected by gas viscosity. The voidage was high for high viscosity gas. The bubble size was small for high viscosity gas, while was large for low viscosity gas such as hydrogen. The binarized images of the beds fluidized by argon gas and hydrogen gas are shown in Figure 1. This figure indicates that the bubble size were affected by the type of fluidizing gas. The relationship between the apparent viscosity of the emulsion phase and bubble size was obtained using the correlation of the apparent viscosity considering the emulsion phase voidage. The relationship agreed with the theoretical formula introduced by Kurooka et al. (2008). References Kai, T., A. Iwakiri, T. Takahashi, Emulsion phase expansion and sedimentation velocity in fluidized beds of fine

particles, J. Chem. Eng. Japan 20 (1987a) 282–286. Kai, T., Y. Shirakawa, T. Takahashi, S. Furusaki, Change in bubble behavior for different fluidizing gases in a

fluidized bed, Powder Technol., 51, 267–271 (1987b). Kai, T., M. Murakami, K. Yamasaki, T. Takahashi, Relationship between apparent viscosity and fluidization

quality in a fluidized bed with fine particles, J. Chem. Eng. Japan, 24, 494–500 (1991). Karimipour, S., T. Pugsley, A critical evaluation of literature correlations for predicting bubble size and velocity

in gas–solid fluidized beds, Powder Techonl. 205, 1–14 (2011). Kurooka, T., R. Yamazaki, G. Liu, Hydrodynamics of gas-solid fluidized bed of fine particles and two phase

theory, Kagaku Kogaku Ronbunshu, 34, 571–579 (2008).

(a) Ar (UG=3.16, 6.95, 10.5 cm s

−1) (b) H

2 (U

G=2.97, 6.67, 9.71 cm s

−1)

Fig. 1 Binarized images of 2D bed fluidized by (a) Ar and (b) H2.�

�

- 66 -

The 2015 International Chemical Congress of Pacific Basin Societies (PACIFICHEM 2015) Honolulu, Hawaii, December 15-20, 2015

Preparation of imidazolium-group-containing cyclic siloxane

indicating ionic liquid nature

1Takuya Kubo, 2Sayako Koge, 2Joji Ohshita, 1Yoshiro Kaneko*

1Graduate School of Science and Engineering, Kagoshima University, Kagoshima, Japan 2Graduate School of Engineering, Hiroshima University, Hiroshima, Japan

Abstract

Ionic liquids have been widely studied for their remarkable potential. However, little has been reported regarding the preparation of ionic liquids containing inorganic frameworks. Recently, cage-like oligosilsesquioxane (POSS) with an ionic liquid nature was developed by Chujo et al.1 More recently, we have also reported the preparation of ionic liquids containing random- and cage-structured oligosilsesquioxanes by the hydrolytic condensation of quaternary ammonium-group-containing and imidazolium-group-containing organotrialkoxysilanes using superacid catalysts, such as bis(trifluoromethanesulfonyl)imide (TFSI).2

In this study, as a new siloxane based ionic liquid, imidazolium-group-containing cyclic siloxane was prepared by the hydrolytic condensation of 1-[3-(dimethoxymethylsilyl)propyl]-3-methylimidazolium chloride (DSMIC) using superacids, such as TFSI and trifluoromethanesulfonic acid (TFSA), as catalysts.3

Imidazolium-group-containing cyclic siloxane with TFSI anion (Im-CyS-IL-TFSI) was prepared by the following procedures: DSMIC was stirred in water/methanol mixed solvent of TFSI at room temperature. The resulting solution was heated in an open system until the solvent completely evaporated. The resulting crude product was heated at 100 oC for 2 h, washed with water, and then dried to obtain Im-CyS-IL-TFSI. 1H NMR, 29Si NMR, and MALDI-TOF MS results of Im-CyS-IL-TFSI indicated that this product was a mixture of cyclic tetra- and pentasiloxanes. Im-CyS-IL-TFSI showed fluidity at 0 °C.

On the other hand, when the hydrolytic condensation of DSMIC was performed using aqueous TFSA, we also found that ionic liquid containing cyclic tetra-, penta-, and hexasiloxanes (Im-CyS-IL-TFSA) was obtained. Im-CyS-IL-TFSA showed obvious fluidity over 20 °C. References 1. Y. Chujo et al., J. Am. Chem. Soc., 132, 17649 (2010). 2. Y. Kaneko et al., Bull. Chem. Soc. Jpn., 87, 155 (2014).; RSC Adv., 2015, 5, 15226. 3. T. Kubo, S. Koge, J. Ohshita, and Y. Kaneko, Chem. Lett., 2015, 44, 1362.

- 67 -

2015 International Chemical Congress of Pacific Basin Societies December 15-20, 2015, HONOLULU, HAWAII

Sol-gel synthesis of amphiphilic silsesquioxane capable of forming reverse micelle

Akito Nagatomo and Yoshiro Kaneko*

Graduate School of Science and Engineering, Kagoshima University, Japan Abstract

Amphiphilic molecules can form reverse micelles in non-polar organic solvents. Most of these molecules are organic compounds. On the other hand, the amphiphilic molecules containing inorganic components may become the materials capable of forming reverse micelles with exotic functions.

So far, we have reported the preparation of water-soluble ammonium group-containing ladder-like polysilsesquioxanes (PSQs), as soluble inorganic materials, by the sol-gel reaction of 3-aminopropyltrimethoxysilane (APTMS) in aqueous hydrochloric acid.1

In this study, we prepared an amphiphilic silsesquioxane (Am-SQ) capable of forming a reverse micelle by the sol-gel reaction of a mixture of APTMS and triethoxysilane containing ether-chain (TES-EC) in aqueous hydrochloric acid.2

First, TES-EC was prepared by the reaction of 3-(triethoxysilyl)propyl isocyanate with triethyleneglycol monomethylether. Then, the sol-gel reaction of a mixture of TES-EC and APTMS was performed by stirring in aqueous hydrochloric acid at room temperature, followed by heating in an open system until the solvent was completely evaporated. After the resulting solid product was washed with acetone, chloroform was added into the product. The soluble-part was isolated by filtration and the solvent was removed. The resulting product was dried under reduced pressure to obtain Am-SQ. In the DLS and GPC measurements of Am-SQ in chloroform, there was a correlation between the particle sizes or excluded volumes (molecular weights) and the concentrations, i.e. larger particle sizes or higher molecular weights for higher concentrations. These results indicate that Am-SQ can form reverse micelle at relatively higher concentration in chloroform. The reverse micelle of Am-SQ can incorporate hydrophilic dye compounds. References 1. Y. Kaneko et al., Chem. Mater., 16, 3417 (2004).; Polymer, 46, 1828 (2005).; Z.

Kristallogr., 222, 656 (2007).; Int. J. Polym. Sci., 684278 (2012). 2. A. Nagatomo and Y. Kaneko, J. Nanosci. Nanotechnol., 2016, in press.

- 68 -

2015 International Chemical Congress of Pacific Basin Societies (Pacifichem 2015) Hawaii Convention Center, December 15-20, 2015, Honolulu, Hawaii

Preparation of Chitin Nanofiber-based Composite Materials by Surface

Modification Ryo Endo, Kazuya Yamamoto, Jun-ichi Kadokawa

Department of Chemistry, Biotechnology, and Chemical Engineering, Graduate School of Science and Engineering, Kagoshima University, Kagoshima 890-0065, Japan

Abstract

Chitin is widely distributed in nature and an important renewable resource. However, it has been difficult to provide various material applications from chitin, due to lack of solubility and processability. Derivatization of chitin is one of the efficient methods for its material applications. Recently, we found that an ionic liquid of 1-allyl-3-methylimidazolium bromide (AMIMBr) dissolved or swelled chitin, which was also used as a solvent for acetylation of chitin with acetic anhydride under mild conditions [1,2]. Furthermore, we found that self-assembled chitin nanofiber film was obtained by regeneration from a chitin ion gel with AMIMBr using methanol, followed by filtration [3]. In this study, we performed the preparation of self-assembled chitin nanofiber-based composite materials through surface modification (Scheme). First, a self-assembled chitin nanofiber dispersion was prepared according to our previous publication [3]. Then, surface acetylation of the product was performed by reaction with acetic anhydride for 12 h at rt in a dispersion with DMF to obtain partially acetylated chitin nanofibers, which was isolated by filtration to give a film. The film showed compatibility with hydrophobic polymers. For example, the composite film of chitin nanofiber with polyethylene was obtained. References 1) K. Prasad, M. Murakami, Y. Kaneko, A. Takada, Y. Nakamura, J. Kadokawa, Int. J. Biol. Macromol., 45, 221 (2009). 2) S. Mine, H. Izawa, Y. Kaneko, J. Kadokawa, Cabohydr. Res., 344, 2263 (2009). 3) J. Kadokawa, A. Takegawa, S. Mine, K. Prasad, Carbohydr. Polym., 84, 1408 (2011).

Scheme. Preparation of partially acetylated chitin nanofiber film by

surface modification

- 69 -

2015 International Chemical Congress of Pacific Basin Societies (Pacifichem 2015) Hawaii Convention Center, December 15-20, 2015, Honolulu, Hawaii

Preparation of Amylose-Polypeptide Inclusion Complexes

by Vine-Twining Polymerization

Ryuya Gotanda, Kazuya Yamamoto, Jun-ichi Kadokawa

Department of Chemistry, Biotechnology, and Chemical Engineering, Graduate School of Science and

Engineering, Kagoshima University, Kagoshima 890-0065, Japan

Abstract

Amylose is a well-known helical host compound that forms inclusion complexes with relatively lower

molecular weight guest compounds by noncovalent interaction. However, little has been reported regarding

the formation of inclusion complexes between amylose and polymeric compounds. The main difficulty in

incorporating polymeric compounds into the cavity of amylose is that the driving force for the binding is

only caused by hydrophobic interaction. Amylose, therefore, does not have sufficient ability to include the

long chains of guest polymers into its cavity. By means of the enzymatic method for direct construction of

amylose, we have developed a new methodology, named “vine-twining polymerization”, for the preparation

of inclusion complexes composed of amylose and synthetic polymers, which was achieved by the

phosphorylase-catalyzed enzymatic polymerization forming amylose in the presence of guest polymers [1-3].

In this study, we found that amylose-poly(D-alanine) inclusion complex was obtained by the vine-twining

polymerization using poly(D-alanine) as a new guest polymer [4] (Scheme). The XRD pattern of the product

showed the typical diffraction peaks due to inclusion complex composed of helical conformation between 71

and 61 of amylose. Although the 1H NMR spectrum of the product showed both the signals assignable to

amylose and poly(D-alanine), the integrated ratio of the poly(D-alanine) signal was lower than the theoretical

value. This result suggested that amylose partially included poly(D-alanine) in the product.

References

[1] J. Kadokawa, Polymers 2012, 4, 116. [2] J. Kadokawa, Biomolecules 2013, 3, 369. [3] J. Kadokawa,

Pure Appl. Chem. 2014, 86, 701. [4] R. Gotanda, K. Yamamoto, J. Kadokawa, Macromol. Chem. Phys. in press.

Scheme. Stereoselective inclusion by amylose in vine-twining polymerization.�

- 70 -

2015 International Chemical Congress of Pacific Basin Societies (Pacifichem 2015)

Hawaii Convention Center, 15-20 December 2015, Honolulu, Hawaii

Enzymatic Synthesis of Amphoteric Polysaccharide Materials

Yusei Takata, Kazuya Yamamoto, and Jun-ichi Kadokawa

Department of Chemistry, Biotechnology, and Chemical Engineering, Graduate School of Science and Engineering,

Kagoshima University, Kagoshima 890-0065, Japan

Abstract

Phosphorylase is the enzyme that catalyzes phosphorolysis of α-(1→4)-glucans at the nonreducing

ends, such as glycogen and amylose, giving α-D-glucose 1-phosphate (Glc-1-P). By means of the

reversibility of the reaction, α-(1→4)-glucans can be prepared by phosphorylase-catalyzed enzymatic

polymerization of Glc-1-P as a monomer using a maltooligosaccharide primer according to the

manner of successive glucosylations. Because of loose specificity for the recognition of substrates,

furthermore, thermostable phosphorylase recognizes α-D-glucuronic acid 1-phosphate (GlcA-1-P) and

α-D-glucosamine 1-phosphate (GlcN-1-P) as glycosyl donors, and accordingly, catalyzes

glucuronylation and glucosaminylation, respectively, to give oligosaccharides having glucuronic acid

(GlcA) and glucosamine (GlcN) residues at the nonreducing ends. On the other hand, glycogen, a

natural polysaccharide, acts as a multifunctional glycosyl acceptor for the phosphorylase catalysis

because of the presence of a number of nonreducing α-(1→4)-glucan ends interlinked by

α-(1→6)-glycosidic bonds. In this study, we performed the thermostable phosphorylase-catalyzed

subsequent enzymatic glucuronylation and glucosaminylation of glycogen to give an amphoteric

glycogen having GlcA/GlcN residues at nonreducing ends. Furthermore, cross-linking of amylose

chains elongated from the product was conducted by the phosphorylase-catalyzed enzymatic

polymerization of Glc-1-P to give a hydrogel (Scheme 1) [1].

Reference

1) Y. Takata, K. Yamamoto, J. Kadokawa, Macromol. Chem. Phys., 2015, 216, 1415

Scheme 1. Synthesis of amphoteric glycogen gel by thermostable phosphorylase catalysis

- 71 -


Hawaii Convention Center, 15-20 December 2015, Honolulu, Hawaii

Preparation of Amylose Supramolecular Gel Materials by Vine-Twining Polymerization

Kazuya Tanaka, Daisuke Hatanaka, Kazuya Yamamoto, Jun-ichi Kadokawa



Abstract

Amylose is a natural glucose polymer linked through α-(1→4)-glycosidic linkages. It acts as a host molecule and

forms polysaccharide supramolecules by inclusion complexation with various guest molecules of relatively low

molecular weight owing to the helical conformation. However, only limited studies have been reported regarding

the direct construction of inclusion complexes composed of amylose and polymeric molecules. We have already

reported the polymerization method for the architecture of amylose-polymer supramolecular inclusion complexes

be named “vine-twining polymerization”. In this method, amylose-polymer inclusion complexes were obtained

with the progress of phosphorylase-catalyzed enzymatic polymerization, which was conducted using

α-D-glucose1-phosphate (G-1-P) as a monomer from a maltoheptaose (G7) primer in the presence of the guest

polymers [1-3]. In this study, we performed the preparation of supramolecular gel materials by the vine-twining

polymerization using poly(γ-glutamic acid-graft-ε-caprolactone) (PGA-g-PCL) as a new guest polymer. When the

enzymatic polymerization of G-1-P from G7 primer was performed using the graft copolymer, a supramolecular

hydrogel was obtained (Scheme 1) [4]. The resulting hydrogel, purified by immersing in water, had a self-standing

property. The XRD result of a cryogel obtained by lyophilization of the hydrogel indicated the presence of

inclusion complexes of amylose with the PCL graft-chains between intermolecular (PGA-g-PCL)s, which acted as

supramolecular cross-linking points for the

hydrogelation. Macroscopic healing of two

hydrogel pieces was achieved by the

formation of inclusion complexes at their

interfaces through the enzymatic

polymerization. Furthermore, an ion gel was

fabricated by immersing the hydrogel into an

ionic liquid, 1-butyl-3-methylimidazolium

chloride.

References

1) J. Kadokawa, Polymers, 2012, 4, 116. 2) J. Kadokawa, Biomolecules, 2013, 3, 369. 3) J. Kadokawa, Pure Appl.

Chem., 2014, 86, 701. 4) J. Kadokawa, K. Tanaka, D. Hatanaka, K. Yamamoto., Polym. Chem., 2015, 6, 6402.

Scheme 1. Preparation of hydrogel by vine-twining polymerization using graft copolymer.

Phosphorylase0.2 M acetate buffer

(pH 6.2)

Inclusion complex (Cross-linking point) Amylose

OOH

HO OHHO

O P

O

O

O

OOH

HO OHO

O6

OOH

HO OHOH

H

+ +

G-1-PG7

Hydrophobic graft chain

PGA

hydrogel

- 72 -


Hawaii Convention Center, December 15-20, 2015, Honolulu, Hawaii

Synthesis of Chitin/Chitosan Stereoisomers by Phosphorylase-catalyzed Enzymatic Polymerization

Kento Yamashita, Riko Shimohigoshi, Kazuya Yamamoto, Jun-ichi Kadokawa



Abstract

Phosphorylase catalyzes enzymatic polymerization of α-D-glucose 1-phosphate (Glc-1-P) as a monomer using a maltooligosaccharide primer according to the manner of successive glucosylations to produce α-(1→4)-glucan with liberating inorganic phosphate [1]. Because of loose specificity for the recognition of substrates, we previously reported that α-D-glucosamine 1-phosphate (GlcN-1-P) could be used as a glycosyl donor in potato phosphorylase-catalyzed enzymatic glucosaminylation to give oligosaccharides having a glucosamine (GlcN) residue at a nonreducing end [2]. Because it is known that thermostable phosphorylase differs in recognition ability of substrates from potato phosphorylase, in this study, we performed the thermostable phosphorylase-catalyzed enzymatic polymerization of GlcN-1-P as a monomer using a maltotriose primer under the conditions with removal of inorganic phosphate from the reaction media to give α-(1→4)-linked glucosamine polymer, that is, chitosan stereoisomer (Scheme 1) [3]. To produce chitin stereoisomer, furthermore, we also carried out N-acetylation of the obtained chitosan stereoisomer by using acetic anhydride and sodium carbonate. The 1H NMR spectra of the products supported the structures of the chitin/chitosan stereoisomers.

References

1) G. Ziegast, B. Pfannemüller, Carbohydr. Res., 160, 185 (1987). 2) M. Nawaji, H. Izawa, Y. Kaneko, J. Kadokawa, Carbohydr. Res., 343, 2692 (2008). 3) J. Kadokawa, R. Shimohigoshi, K. Yamashita, K. Yamamoto, Org. Biomol. Chem., 13, 4336 (2015).

Scheme 1. Enzymatic Synthesis of Chitin/Chitosan Stereoisomers by Thermostable Phosphorylase-catalyzed Polymerization

- 73 -

Society for Information Display 2015

San Jose Convention Center, San Jose, America, June 2-5, 2015

Subjective Size of News Presenter Shrinking with Recent Enlargement of Display Size in Japan

Youhei Kumagai1, Ken Kihara1, Sakuichi Ohtsuka1

1Graduate School of Science and Engineering, Kagoshima University, 890-0065, Kagoshima, Japan

Abstract The average size of displays sold was stable at around 29.5-inch to 30.5-inch in period of 2011 to 2012, but rapidly increased in 2013, and the average display size reached 34.6-inch in Sep. 2014 as shown in Figure 1A [1]. We investigate the changes in figure size of news presenter(s) from two viewpoints; “impression” represents subjective size, i.e. the observer’s “impression” of figure size when viewing contents, and “area” represents objective size, i.e. judgment of physical area occupied by figure(s) in the whole screen. In Figure 1B, for example, the left panel is larger than the right panel in “area”. However, the right panel is larger than the left panel in “impression”. Result clearly indicates that “impression” was decreased whereas “area” was not changed in the period between 2011 and 2014 as shown in Figure 1C. The mean size estimation of “impression” was significantly lower than that for “area” in weather forecast scenes as shown in Figure 1D. Because the amount of information in the weather map is basically same in 2011 and 2014, it is considered that the difference in “impression” of figure size of news presenter is indicative of a change in the television receiver’s environment. We detected some indications of the composition changes that were considered to be induced by the enlargement of display size in the last three years, from 2011 to 2014.

A B C D

Figure 1. Figures used in the poster.

References 1. http://www.bcn.co.jp/press/press.html?no=302 (in Japanese)

- 74 -

fast learning algorithm for fuzzy inference systems using ... · this time, we apply multiple...

Documents