ieices intellectual exchange and innovation conference on ... · you to the intellectual exchange...

IEICES

Intellectual Exchange and Innovation Conference on

Engineering & Sciences (IEICES)

15 October, 2015

Kyushu University, Fukuoka, Japan

CONFERENCE PROGRAM AND PROCEEDINGS

I

IEICES

CONFERENCE PROGRAM

AND PROCEEDINGS

WELCOME MESSAGE

Welcome to IEICES!

On behalf of the IEICES 2015 organizing

committee, it is my great pleasure to welcome

you to the Intellectual Exchange and Innovation

Conference on Engineering & Sciences (IEICES),

which is being held on 15th October, 2015 at

Kyushu University, Japan. The conference

brings together academics, researchers and

students to share the latest developments and

research in the fields of materials, energy and

environmental engineering & sciences. The

organizing committee has developed an exciting

program to give participants an opportunity to

present their academic works, concepts and

new discoveries. Moreover, participants will

have a chance to get to know one another

closely.

Before I close, I would like to thank each of your

for participating our conference and bringing

your expertise to our gathering.

Osama Eljamal

IEICES Chairman

GENERAL INFO

The first session of

Intellectual Exchange and

Innovation Conference on

Engineering & Sciences

(IEICES) is organized by

Interdisciplinary Graduate

School of Engineering

Sciences (IGSES) and the

PhD students of IEI Program

of Kyushu University. The

participants work on

different subjects including

applied science for

electronics & materials,

molecular & material

sciences, advanced energy

engineering science, energy

& environmental

engineering, and earth

system science technology,

etc. The participants will

have an opportunity to

present their academic

works, concepts and new

discoveries as well as they

will have a chance to

exchange their ideas and

develop their works.

Venue

3rd Floor

Administrative Building

Chikushi Campus

Kyushu University

6-1 Kasuga-koen, Kasuga,

Fukuoka, 816-8580, Japan

II

COMMITTEES

Conference Honorary Chairman Prof. Akira Harata

Conference Chairman Associate Prof. Osama Eljamal

Conference Coordinator Ahmad Syahrin Idris

Conference Organizing Committee Ahmed Mohamed Elsayed Khalil

Ding Dong

Mahmoud Ramadan Abusrea

Moses Kibunja Kamita

Muhammad Hamid Mahmood

Nik Mohd Izual Nik Ibrahim

Qi Shi-Chao

Ruan Hongcheng

Tarek Naem Dief

Wang Zhengxing

Zhang Lu

III

Technical Program

Time Thursday, 15 October 2015

9:00-10:00 Registration

10:00-10:10 Opening Remarks, Prof. Akira Harata

10:10-10:40 Invited Speaker (1) Prof. Kazuo Arakawa

10:40-11:10 Invited Speaker (2) Prof. Shigeo Yoshida

11:10-11:20 Photo Session & Break

11:20-11:35 1 Mahmoud Abusrea

11:35-11:50 2 Moses Kamita

11:50-12:05 3 Hongcheng Ruan

12:05-12:20 4 Tarek Dief

12:20-12:35 5 Ahmed Khalil

12:35-12:50 6 Nik Mohd Izual Nik Ibrahim

12:50-13:05 7 Muhammad Hamid Mahmood

13:05-14:00 Lunch

14:00-14:15 8 Zhengxing Wang

14:15-14:30 9 Awang Idris

14:30-14:45 10 Waled Daoud

14:45-15:00 11 Abdelrahman Zkria Mohamed

15:00-15:15 12 Mohamed Egiza

15:15-15:30 Break

15:30-15:45 13 Ahmed Lotfy Elrefai

15:45-16:00 14 Ahmad Syahrin Idris

16:00-16:15 15 Ding Dong

16:15-16:30 16 Shi-Chao Qi

16:30-16:45 17 Lu Zhang

16:45-16:55 Closing Remarks, Prof. Seigi Mizuno

16:55-17:00 Best Presenter & Awards

IV

INVITED SPEAKERS

Prof. Kazuo Arakawa Professor / Renewable Energy Center, Research Institute for Applied Mechanics, Kyushu University

Biography 2001-present: Associate professor at Research Institute for Applied Mechanics RIAM, Kyushu

University 1988-1989: Visiting researcher, University of Washington, USA.

1982: Research Associate, RIAM, Kyushu University

1982: D.Eng. from Osaka University

Membership in Academic Society

The Japanese Society for Experimental Mechanics (JSEM)

The Society of Materials Science, Japan

The Japan Society for Composite Materials

The Japanese Society for Non-Destructive Inspection

The Japan Society of Mechanical Engineers

Society for Experimental Mechanics

Presentation title Effect of time derivative of contact area on dynamic friction

Presentation Abstract: This study investigated dynamic friction during oblique impact of a golf ball by evaluating the ball’s angular velocity, contact force, and the contact area between the ball and target. The effect of the contact area on the angular velocities was evaluated, and the results indicated that the contact area plays an important role in dynamic friction. In this study, the dynamic friction force F was given by F= μN+μη.dA/dt, where μ is the coefficient of friction, N is the contact force, dA/dt is the time derivative of the contact area A, and η is a coefficient associated with the contact area.

Prof. Shigeo Yoshida Professor / Renewable Energy Center, Research Institute for Applied Mechanics, Kyushu University

Biography

Shigeo Yoshida, born in Fukushima, in 1967.

Working at Kyushu University since 2013, as a professor in Research Institute for Applied Mechanics

University Received B.E. Degree in Engineering Faculty of Kyoto University in 1990

Doctor of Engineering in Ashikaga Institute of Technology in 2007.

Industry Worked in Fuji Heavy Industries (SUBARU) since 1990, as an aerodynamic engineer for aircrafts and

wind turbines and transferred to Hitachi in 2012. .

Member of JSME, JSFM, JSES (Director 2008-2010),

JWEA (Director 2010-)

TSJ and a registered expert of IEC.

Best Technology Award (JSES, 2007)

Best Paper Award (JSES 2007, JWEA 2009)

Best Poster Award (Renewable Energy 2006)

Poster Award (JWEA, 2007, 2007, 2008, 2010)

Research Fundamental design, aerodynamics, control and safety of wind turbines Wind farm performance, fatigue mitigation, layout optimization Downwind rotor and tower shadow model

V

CONTENTS

CFRP ADHESIVE JOINTS AND STRUCTURES FOR OFFSHORE WIND-LENS TURBINE M R Abusrea, Shiyi Jiang, Dingding Chen, Kazuo Arakawa

1

IMMUNOSUPPRESSION BY COLON CANCER CELLS, MEDIATED BY TUMOR-SECRETED SOLUBLE FACTORS Moses K. Kamita, Arihiro Kano, Shindo Mitsuru

3

CATALYTIC PERFORMANCE OF AG-COATED ZEOLITE FOR SOOT OXIDATION Honcheng Ruan, Maiko Nishibori

5

ATTITUDE AND ALTITUDE STABILIZATION OF OUTDOOR TETHERED QUAD-ROTOR Tarek N. Dief, Shigeo Yoshida

7

DIFFERENT NANOSCALE ZERO VALENT IRONS FOR NITRATE-POLLUTED WATER REMEDIATION Ahmed M. E. Khalil, Osama Eljamal, Nobuhiro Matsunaga

9

WAVE-STRUCTURE INTERACTION USING FREE SURFACE LATTICE BOLTZMANN METHOD (FSLBM) Nik Mohd, Changhong Hu, Xuhui Li

11

STUDY ON DESICCANT AIR-CONDITIONING SYSTEM FOR AGRICULTURAL PRODUCT STORAGE IN PAKISTAN Muhammad H. Mahmood, Muhammad Sultan, Takahiko Miyazaki, Shigeru Koyama

13

THE INVESTIGATION OF THE COLORIMETRY TO MEASURE THE DEPOSITION THICKNESS ON THE PLASMA-FACING WALL IN QUEST Zhengxing Wang, Takahiro Shimoji, Kazuaki Hanada, Naoaki Yoshida, etc

15

NUMERICAL STUDIES OF NATURAL VENTILATION OF BUILDING WITH LEEWARD OPENINGS A. Idris, B. P. Huynh, Z. Abdullah

17

STATISTICAL EVALUATION OF COMPRESSION INDEX CORRELATIONS Waled A. Daoud, Kiyonobu Kasama, Naser M. Saleh, Abdelazim M. Negm

19

HETROJUNCTION DIODE OF NITROGEN-DOPED ULTRANANOCRYSTALLINE DIAMOND FILMS PREPARED BY COAXIAL ARC PLASMA DEPOSITION Abdelrahman Zkria, Tsuyoshi Yoshitake

21

ULTRANANOCRYSTALLINE DIAMOND/AMORPHOUS CARBON COMPOSITE FILMS SYNTHESIS ON CEMENTED CARBIDE SUBSTRATE BY COAXIAL ARC PLASMA DEPOSITION Mohamed Egiza, Hiroshi Naragino, Aki Tominaga, Kouki Murasawa, Hidenobu Gonda, etc

23

ORTHOGONAL FLUXGATE GRADIOMETER SENSOR AND ITS APPLICATION IN PARTICLE DETECTION Ahmed Lotfy Elrefai, Ichiro Sasada

25

DRY ETCHING OF GERMANIUM WAVEGUIDES BY USING CHF3 INDUCTIVELY COUPLED PLASMA Ahmad Syahrin Idris, Haisong Jiang, Kiichi Hamamoto

27

SUPPRESSING NEW GRAPHENE NUCLEI GENERATION BY HYDROGEN SWEEPING Ding Dong, Hiroki Ago

29

A HIGHLY ACTIVE Ni/ZSM-5 CATALYST FOR DEEP HYDROGENATION OF ARENES Shi-Chao Qi, Lu Zhang, Jun-ichiro Hayashi

31

THE ACCELERATED SOLVENT EXTRACTION OF XINYU COKING COAL AND ITS EXTRACTION MECHANISM Lu Zhang, Shi-Chao Qi, Koyo Norinaga

33

Intellectual Exchange and Innovation Conference on Engineering & Sciences (IEICES) Oct 15th, 2015

1

CFRP ADHESIVE JOINTS AND STRUCTURES FOR OFFSHORE

WIND-LENS TURBINE

M R Abusrea1, Shiyi Jiang1 , Dingding Chen2, and Kazuo Arakawa3. 1 Interdisciplinary Graduate School of Engineering Science, Kyushu University, Fukuoka, Japan

([email protected]) 2 National University of Defense Technology, China,

3 Research Institute for Applied Mechanics, Kyushu University, Fukuoka, Japan

Abstract: Novel wind-lens turbine designs can augment power output. Vacuum-assisted resin transfer molding (VARTM) is

used to form large and complex structures from a carbon fiber reinforced polymer (CFRP) composite. Typically, wind-lens

turbine structures are fabricated in segments, and then bonded to form the final structure. This paper introduces five new

adhesive joints, divided into two groups: one is constructed between dry carbon and CFRP fabrics, and the other is

constructed with two dry carbon fibers. All joints and CFRP fabrics were made in our laboratory using VARTM

manufacturing techniques. Specimens were prepared for tensile testing to measure joint performance. The results showed

that the second group of joints achieved a higher tensile strength than the first group. On the other hand, the tensile fracture

behavior of the two groups showed the same pattern of crack originating near the joint ends followed by crack propagation.

1. Introduction

Composite materials have high stiffness-to-weight

and strength-to-weight ratios, and have been used

for many applications including aerospace,

automotive, and wind turbine structures [1]. The

wind-lens, a curved ring around the turbine blades,

is manufactured from six identical parts joined

together to form the final structure. Consequently,

its performance depends not only on material

properties but also on the joining technique.

Bonded joints have mechanical advantages over

bolted joints because fibers are not cut, and stresses

are transmitted more homogenously [2].

This paper introduces various adhesive bonded

joints, made of carbon fiber reinforced polymer

(CFRP), for use in offshore wind-lens structures.

The main objective of this work was to develop

high-strength joint applicable to offshore wind-lens

structures. The strengths of five joints were

assessed. All joints and CFRP material tested in

this study were made using a technique developed

from the vacuum-assisted resin transfer molding

(VARTM) process.

2. Experimental work

The composite material was CFRP, consisting of

a carbon fabric (TENAX STS; 504 g m–2)

hardened with a resin (XNR6815/XNH6815). Four

unidirectional carbon fabric sheets were stacked

and molded together to form plates with an average

thickness of 2 mm [1]. All CFRP fabrics were

produced using VARTM, a variation of resin

transfer molding (RTM) in which a solid mold with

a flexible tape-sealed vacuum bag is used to replace

the closed mold. In the VARTM process,

reinforcements are stacked on a solid mold, which

is treated with mold releasing agent and covered

with a peel ply and distribution medium. They are

enclosed together with an inlet and a vent in a

vacuum bag and sealed with gum tape (see Fig. 1).

Joint strengths were evaluated via tensile testing

using standardized test specimens [1]. Fig. 2 shows

the dimensions of the specimens; the total length

was 250 mm and the width was 10 mm. Pairs of

GFRP tabs were used to reduce the stress when

holding each specimen.

The strength of the original CFRP was measured,

and used as a reference for the strength of

subsequent joints. Five joint types were tested,

divided into two groups. One was constructed using

dry carbon fabrics and CFRP. In this group, the

CFRP half of the joint was manufactured first, and

then re-molded again with dry carbon fabric. Fig. 3

shows types 1 and 2 of the first joint group. The left

half of both joints was a stepped CFRP portion that

had been molded, and the right side represents a

dry carbon fabric. We used these to investigate the

effects of the number of steps connected to dry

carbon fabrics, and therefore the major difference

between the two joints is the number of CFRP steps.

The second group was constructed with two dry

carbon fiber halves; thus, the whole joint was made

in a single step. Fig. 4 shows types 3, 4, and 5 of

the second joint group. These joints were named

laminated joint-1, laminated joint-2, and multi-

overlapped joint, respectively.

Fig. 1. A schematic view of the VARTM process

used in this work


2

Joint

Tab

35 mm 50 mm

250 mm

40 or 80 mm

2 mm2 mm

Fig. 2. The standard specimen dimensions used for

tensile testing

Mold

Distribution medium Vacuum bag

Carbon fiber

80 mm

(a)Peel Ply

(b)

CFRP

CFRP

Fig. 3. Joint group 1: (a) staircase joint-1, and (b)

staircase joint-2.

40 mm

(c)

(b)

(a)

Fig. 4. Joint group 2: (a) laminated joint-1, (b)

laminated joint-2, and (d) multi-overlapped joint.

3. Results and discussions

Figure 5 shows the tensile strengths of the five

joints, and the original CFRP. A tensile load of

34 kN was recorded for the CFRP. The lowest joint

tensile strength recorded was for staircase joint-1,

with a measured strength of 8.8 kN (26% joining

efficiency). However, the strength of the other

staircase joint, staircase joint-2, was significantly

higher (14.3 kN; 42% joining efficiency).

This behavior can be attributed to two factors.

First, resin residue on the CFRP surface before

joining can act as an insulator. Second, the absence

of overlap contact in these joints reduces the

contact area, resulting in a weaker joint [3]. On the

other hand, staircase joint-2 achieved a much

higher strength than joint-1. This is attributed to the

three fiber layers in that joint type that contact the

CFRP part, three times more than the number of

layers in joint-1. Hence, joining dry carbon fabrics

together resulted in generally higher strengths. The

laminated joint-1 had a tensile strength similar to

that of staircase joint-2, with a measured strength

of 14.7 kN. This can be attributed to the absence of

overlap contact between the two halves in

laminated joint-1, which reduces the strength and

promotes crack propagation near the joint ends.

The introduction of overlap areas not only

increases the contact area, but also increases joint

thickness. On the other hand, the remaining two

joints were much stronger than the previous four

joints. Laminated joint-2 and the multi-overlapped

joint had tensile strengths of 26.8 kN (79%) and

28.8 kN (85%), respectively. These two joints

performed similarly, with the major difference

being the greater thickness of the multi-overlapped

joint-2, which could be the reason for the higher

observed strength. Löbel et al. [9] constructed

CFRP joints based on stainless pins, which resulted

in a high joining efficiency of 83%. However, the

metal-to-carbon fiber contact caused galvanic

corrosion of the carbon fabrics, weakening the

structure over time [4].

0

5

10

15

20

25

30

35

40

Ten

sile

str

ength

, kN

8.8

14.3 14.7

26.828.8

34

Sta

ircas

e

join

t-1

Sta

ircas

e join

t-2

Lam

inat

ed

join

t-1

Lam

inat

ed

join

t-2

Multi-

ove

rlap

ped

join

t

Join

tless

CFR

P

Fig. 5. Tensile strengths of the joints and the

original CFRP material.

4. Conclusion

Five adhesive joints were designed using

manufacturing process developed from the

VARTM process. The tensile test results showed

low strength when one half of the joint is CFRP

fabrics, which was the case for the first two

developed joints. On the other hand, the last two

joints, laminated joint-2 and multi-overlapped joint,

showed higher tensile strength. However, joining

techniques that use dry carbon fibers are still

limited for simple shapes, so there are some

difficulties for applying these techniques for

complex curved shapes like wind blades and lens as

well.

References [1] Chen DD, Arakawa K, Jiang SY. Novel joints

developed from partially un-moulded carbon-fibre-

reinforced laminates. Journal of Composite

Materials, 2015, Vol. 49(14) 1777–1786

[2] Banea MD, Da Silva LF. Adhesively bonded joints

in composite material: an overview. J Mater Des

Appl 2009;223:1–18.

[3] LÖbel T., Kolesnikov B, Scheffler S, et al.

Enhanced tensile strength of composite joints by

using staple-like pins: working principles and

experimental validation. Compos Struct 2013; 106:

453–460.

[4] Wen-Xue W., Yoshiro T., Terutake M. Galvanic

corrosion-resistant carbon fiber metal laminates.

16th Int. conf. on composite materials.


3

IMMUNOSUPPRESSION BY COLON CANCER CELLS, MEDIATED

BY TUMOR-SECRETED SOLUBLE FACTORS

Moses K. Kamita1, Arihiro Kano2, Shindo Mitsuru2

1Molecular and Material Science, IGSES, Kyushu University - 2Institute for Material Chemistry and Engineering, Kyushu University

Abstract: In this study, soluble factors secreted by colon cancer cells (CT26) are being investigated. Conditioned medium

from cultured CT26 cancer cells was collected and analyzed for the presence of immunosuppressive compound. The

presence of the active compound in the samples was constantly monitored by measuring the suppression of IFN-γ expression

by ELISA using splenocytes from a normal mouse. Suppressive activity in CT26 conditioned medium was shown to be

mediated by a soluble factor(s) that is less than 10kDa. After fractionation with HPLC, suppression of IFN-γ expression was

detected in the two different fractions. Further processing of the fractions is needed for the identification of the specific

compounds that are responsible for the suppression of tumor immunity.

1. Introduction

The immune system works by distinguishing

self from non-self and eliminates the non-self. IFN-

γ is a pleiotropic cytokine that is produced

primarily by T-cells upon antigen stimulation, and

works as an antiviral, as well as an antiparasitic

agent in the body. It also inhibits the proliferation

of several normal and transformed cells. The

production of interferon-gamma (IFN-γ) in

splenocyte culture and its suppression by the

conditioned medium of tumor cells have previously

been reported1.

Effective tumor immunotherapy is hindered by

a number of obstacles such as the ability of tumor

cells to create a tolerant microenvironment,

activation of negative regulatory checkpoints in the

tumor microenvironment and the secretion of

immunosuppressive cytokines and soluble

inhibitory factors2. Additionally, tumor-associated

changes in myelopoiesis that lead to the excessive

accumulation of immature myeloid cells in the

tumor are thought to play a critical role tumor-

associated immunosuppression3,4. Although this

tumor-associated immunosuppression is thought to

be driven by soluble factors that are expressed by

cancer cells, their identity and mode of action

remain largely unknown.

Overexpression of different pro-inflammatory

cytokines in tumor conditions have been shown to

result in an elevated growth of tumors, as well as

other metastatic diseases. In addition, reports have

indicated that blocking these cytokines results in

significant reduction in the rate of tumor growth5,6

Tumor cells also secrete high amounts of pro-

inflammatory cytokines that have the potential to

induce a local pro-inflammatory microenvironment.

Although these observations are in agreement with

the reports that chronic inflammation plays a role in

tumor development and progression, there are no

much details on the link between these two

observations.

CT26 cells are murine colon tumor cells that

were developed through exposure of BALB/c mice

to N-nitroso-N-methylurethane resulting in a cell

line that is easily implanted and readily metastasize

7. The CT26 model in BALB/c mice has provided a

syngeneic test system for developing, as well as

testing different immunotherapeutic concepts8.

Using CT26 cells, this study aimed to identify the

soluble factor(s) that suppress the immune response

against tumor development. Suppression of IFN-γ

production in a splenocyte assay was used as an

indicator of the presence of active factors.

2. Material and Methods

Cell Culture. Murine colon tumor CT26 cells

were purchased from American Type Culture

Collection (ATCC) (Manassas, VA, USA), and

maintained with D-MEM (Wako Pure Chemical

Industries, Ltd., Osaka, Japan) supplemented with

5% heat- inactivated horse serum in a 5% CO2

atmosphere at 37°C. Balb/c mice were purchased

from Kyudo Co. Ltd. (Tosu, Japan) and all animal

experiments were carried out according to the

guidelines for the proper conduct of animal

experiments published by the Science Council of

Japan. All experimental protocols were approved

by the Ethics Committee and the Animal Care and

Use Committee of Kyushu University. Splenocytes

were prepared from a spleen of a normal mouse

and seeded at one million cells per well in a 96-

well culture plate (Thermo Fisher Science Inc.,

USA), stimulated with lipopolysaccharide and

cultured for 24 hours with or without 10% of the

test samples.

Conditioned Media Collection and

Processing: The tumor cells were cultured to sub-

confluence, and the medium changed to serum free

media. After two days, the medium was recovered

and filtered through a 10kDa filter. The filtrate was

separated using a C18 Sepak column and the

resulting sample fractionated using HPLC. The

presence of the active compound in the samples

was constantly monitored by measuring the

suppression of IFN-γ expression in a splenocyte

assay.

Enzyme-linked Immunosorbent Assay

(ELISA): Quantification of IFN-γ in conditioned

media and fractionated samples was performed by


4

ELISA (R&D Systems, USA), according to the

manufacturer’s instructions.

3. Results

In this study, the splenocyte assay was used to

study the soluble factors that are secreted by CT26

cells for their immunosuppressive activity. The

immunosuppressive activity of the conditioned

media was compared with that of 10kDa filtrate

and the fraction obtained from Sepak column

separation (Fig. 1). Higher activity was recorded in

the Sepak column sample compared with whole

media and 10K filtrate probably due to the

concentration of the active compounds after

separation.

Fig. 1: IFN-γ suppression in the presence of 10% of the

indicated samples. Splenocytes from a normal mouse were

incubated for 24hrs stimulated by LPS. Splenocytes stimulated

with LPS in the absence of the samples acted as the positive control (PC) while splenocytes with no stimulation and no

sample acted as the negative control (NC).

Further processing of the fraction after Sepak

column through fractionation by HPLC resulted in

the separation of the active compound into 11

fractions collected after every 5 minutes. A

splenocyte assay on the activity of the fractions

indicated the active compound(s) to be in fraction

number 2, and 5 eluted between 15-20 and 35-40

minutes respectively (Fig.2).

Fig. 2: IFN-γ suppression in the presence of 10% of the

indicated fractions. Splenocytes from a normal mouse were

incubated for 24hrs stimulated by LPS. Splenocytes stimulated with LPS in the absence of the samples acted as the positive

control (PC) while splenocytes with no stimulation and no

sample acted as the negative control (NC).

4. Discussion

Various soluble factors have been reported in

association with their role in suppressing the

immune response against the tumor. These factors

include IL-10, TGF-β, galectin-1, gangliosides,

PGE2, G-CSF, and M-CSF among others2. Here,

the immunosuppressive factor(s) from CT26 cells

that is responsible for this activity is a 10kDa

molecule. The presence of activity in two separate

fractions is an indication that more than one factor

is involved in this activity. Although different

factors use different mechanisms in different cancer

models to suppress the immune system, most of the

immunosuppressive factors suppress the immunity

through the myeloid-derived suppressor cells

(MDSC) by enhancing their proliferation. Small

factors such as PGE2 that can pass through the 10K

filter could be involved in the observed activity.

However, further analysis is needed to confirm the

identity of the compounds.

5. Conclusion

Use of splenocytes isolated from normal mouse

facilitated the detection of immunosuppressive

factors secreted by tumor cells. Further processing

of the fractions is needed for the identification of

the specific compounds that are responsible for the

suppression of tumor immunity. This identification

will pave way for the elucidation of the

mechanisms through which tumors evade the

immune response. In addition, the information may

be help in the development of tumor therapy.

6. References

1. Kano, A. Tumor cell secretion of soluble

factor(s) for specific immunosuppression. Sci.

Rep. 5, 8913 (2015).

2. Rabinovich, G. A., Gabrilovich, D. &

Sotomayor, E. M. Immunosuppressive

strategies that are mediated by tumor cells.

Annu. Rev. Immunol. 25, 267–96 (2007).

3. Gabrilovich, D. I. & Nagaraj, S. Myeloid-

derived suppressor cells as regulators of the

immune system. Nat. Rev. Immunol. 9, 162–74

(2009).

4. Ugel, S. et al. Immune tolerance to tumor

antigens occurs in a specialized environment of

the spleen. Cell Rep. 2, 628–39 (2012).

5. Voronov, E. et al. IL-1 is required for tumor

invasiveness and angiogenesis. Proc. Natl. Acad.

Sci. U. S. A. 100, 2645–50 (2003).

6. Apte, R. N. & Voronov, E. Interleukin-1--a

major pleiotropic cytokine in tumor-host

interactions. Semin. Cancer Biol. 12, 277–90

(2002).

7. Griswold, D. P. & Corbett, T. H. A colon tumor

model for anticancer agent evaluation. Cancer

36, 2441–4 (1975).

8. Castle, J. C. et al. Immunomic, genomic and

transcriptomic characterization of CT26

colorectal carcinoma. BMC Genomics 15, 190 (2014).


5

CATALYTIC PERFORMANCE OF AG-COATED ZEOLITE

FOR SOOT OXIDATION

Honcheng Ruan1, Maiko Nishibori1

1Department of Molecular and Material Sciences, Interdisciplinary Graduate School of Engineering Sciences, Kyushu

University, Kasuka, Fukuoka 816-8580, Japan; E-Mail：ruanhongcheng＠126.com

Abstract: Ag-coated zeolite catalyst (Ag/HSZ890HOA) mixed with candle soot (CS) for 1, 10, 30 and 60 min, exhibited T

max (temperature according to highest peak of DTA curves) value of 494, 485,425 and 376/492 oC for T0-1, T0-10, T0-30, T0-60,

respectively. Interesting, catalyst mixing with CS for 60 min showed two T max peaks on differential thermal analysis (DTA)

curves.

1. Introduction

Environmental issues are harmful aspects on

human activity on the biophysical environment.

Some environmental issues are belong to air

pollution, such as Ozone depletion, Global

warming, Acid rain, Photochemical smog and PM

2.5 issue. Diesel particulate matter (PM) consists

principally of combustion generated carbonaceous

materials (soot) on which some organic compounds

have become adsorbed [1]. The research of

corresponding catalyst on PM oxidation become

more and more important.

2. Experimental

Ag-doped catalysts (4.5%) were prepared by

ordinary impregnation of HSZ891HOA, with an

aqueous solution of the nitrate salt, AgNO3. Ag-

coated zeolite catalyst (Ag/HSZ890HOA) and

candle soot (CS) were mixed by mixing machine

instead of mortar for 1, 10, 30 and 60 min, as

showed in Fig. 1. The products mixed with CS for

1, 10, 30 and 60 min were denoted as T0-1, T0-10, T0-

30, T 0-60.

Thermogravimetry (TG) and differential thermal

analysis (DTA) measurement were carried out for

catalytic oxidation of CS. The morphologies of the

products were analyzed by scanning electron

microscope (SEM).

3. Results and Discussion

Surface morphologies of catalysts mixing with CS

for different time (1, 10, 30 and 60 min) were

showed in Fig. 2. There just be one exothermic

peak corresponding to a weight loss step on TG

curves with mix time 1 and 10 min, another one

peak appeared with the increasing mix time

gradually. Therefore, an obvious double-shoulder

peak appeared on catalyst DTA curves after mixing

60 min. Tmax of DTA curves for catalyst

Ag/HSZ890HOA are listed in Table 1.

SEM images (Fig. 2) showed that the number

of broken catalyst particles increased obviously as

the increasing mix time from 1 min to 60 min.

Table 1

T max of DTA curves for Ag/HSZ890HOA catalyst with

different mix time.

Mix time

/ min

1

10

30

60

T max / oC

494

485

425

376/492

Fig. 1. TG and DTA curves of candle soot with

Ag/HSZ890HOA catalysts in different mix time with mix

machine (a-d) 1-60min.


6

Fig. 2. SEM images of candle soot with Ag/HSZ890HOA

catalysts in different mix time with mix machine (a-d) 1-60min.

4. Conclusion

Ag-coated zeolite catalyst (Ag/HSZ890HOA)

shows a good catalytic performance. However, a

further study about the principle based on two

peaks on DTA curves need to be done.

5. References

[1] Cheol-B Lima, Hisahiro Einagab, "Yoshihiko Sadaokac, Yasutake Teraokab, Preliminary study on catalytic combustion-type sensor for the detection of diesel particulate matter", Sens. Actuators, B: Chemical, 160, 463-470 (2011).


7

ATTITUDE AND ALTITUDE STABILIZATION OF

OUTDOOR TETHERED QUAD-ROTOR

Tarek N.Dief1, Shigeo Yoshida2

1 Department of Earth System Science and Engineering

Kyushu University,[email protected] 2 Research Institute for Applied Mechanics

Kyushu University, [email protected]

Abstract: This paper presents an overview of the most effective ideas for the Quad-rotor project. The concept of

modelling using different methods is presented. The modelling part discussed the nonlinear model, and the

concept of linearization using small disturbance theory. Parameter identifications part explained the most

important parameters that affect the system stability and tried to get suitable solutions for these problems and

identify some parameters experimentally. The control part incorporates different classical schemes such as PD

and PID controllers to stabilize the Quad-rotor. The difference between the indoor and outdoor controller is

presented from the mathematical and the experimental techniques.

1. Introduction

Nowadays, automatic flying of intelligent vehicles

moving in space represents a huge field of

applications. The rapid development of the

microcontroller affects all control’s applications

which mean new control theory, new researches,

and new challenges. The applications of the

hovering aerial vehicles open the door for

researchers to present better controllers to achieve

more aggressive missions. Nonlinear control

theories and applications are the target of the

researchers to present or develop more controllers

which were very difficult to be implemented using

the old microcontrollers. Using the nonlinear model

will cancel the assumptions which we make during

the control design process; so we will deal directly

with the nonlinear model.

A lot of application for the Quad-rotor has been

applied especially for the outdoor flights. Civilian

application for the UAV includes atmospheric

analysis, mapping, maintenance inspection,

photography for natural disaster, etc…

2. System Identification and Control Design:

a. System Identification:

In the Quad-rotor model, we were dealing with a

fixed model and now we found a problem with the

parameters variation, so we need model can make

the update of the parameters real time by knowing

the history of the plant input also from the system

output data.

Placket’s model: This way depends on the least-

squares fitting. The concept of this model depends

on minimizing the square difference between the

predicted output from the calculation and

between the real data from the sensor data as

shown in equation (1):

Mean square error = (1)

And the following curves show the system open

loop transfer function parameters and plotting them

online.

Where, the system parameters are shown in

equation (2):

21

21

*2*11

*1*0

)(

)(

ZaZa

ZbZb

zU

zY (2)

Fig.1 system parameter b0 with time

Fig.2 system parameter a1, a2 with time

Also neural network is applied to make the system

identification for the system parameters (time series

prediction); also this algorithm is better from the

Placket’s model. Also the initial values have no

effect on the system parameters unlike the

Placket’s model.


8

Also applying the neural network is very important

in our application as the outdoor flights will have a

high disturbance due to the disturbance flow from

different directions. So we can make estimation for

the disturbance by modelling the system.

Fig.3 system response for the actual and estimated

data with time

b. Control design:

A PID controller was implemented because its

proportional gain aspect decreases the system time

to obtain desired state; its integral gain reduces the

steady state error, and the derivative gain will

increase the stability of the system. For our

problem, it becomes a SISO system and we have

the controller equation:

t

idpc edtKeKeKG

0

. (3)

After using (SISOTOOL) package in Matlab and

choosing the best gains to our system, it can be

used in the real model. These gains are for the

linear model so it is needed to use these gains with

the nonlinear model and compare between the

response between the linear and non-linear model

and tune it if necessary.

Fig.4 Block diagram of the Quad-rotor model

and controller.

Fig.5 Time response of the roll angle.

3. Conclusion

This paper presents the control design of outdoor

Quad-rotor. This controller is different for the

indoor Quad-rotors. The effect of wind and the

interaction between the wind and rotor flow

generate forces can disturb the system which mean

instability. Using different kind of system

identifications has advantages and disadvantages,

so due to your application and the hardware which

you are using will make you chose the suitable

algorithm. The classical control is used to due to its

simplicity in implementation and easier in tuning.

4. References

[1] Deif, T., Kassem, A., El Baioumi, G., Modeling and

Attitude Stabilization of Indoor Quad Rotor, (2014) International Review of Aerospace Engineering (IREASE), 7(2), pp. 43-47.

[2] Seul Jung, "A Position-Based Force Control Approach to a Quad-rotor System," Ubiquitous Robots and Ambient Intelligence (URAI), 2012 9th International Conference on , vol., no., pp.373,377, 26-28 Nov. 2012

[3] Etkin, B.: Dynamics of Flight. Stability and Control,

2nd edn. Wiley, New York (1982).

[4] Tarek N. Dief, Mohamed Abdelhady, Shigeo

Yoshida,”Attitude and altitude stabilization of quad

rotor using paramter estimation and self-tuning

controller”, AIAA Atmospheric Flight Mechanics

conference, doi: 10.2514/6.2015-2392.


9

DIFFERENT NANOSCALE ZERO VALENT IRONS FOR NITRATE-

POLLUTED WATER REMEDIATION

Ahmed M. E. Khalil, Osama Eljamal, Nobuhiro Matsunaga

Earth System Science and Technology (ESST), Interdisciplinary Graduate School of Engineering Sciences (IGSES), Kyushu

University, Co-author’s email: [email protected]

Abstract: In this study, four nanoscale zero valent iron (NZVI) types were characterized and compared for nitrate removal

from water. Through batch experiments, it was observed that old-purchased iron (OP-NZVI) had very low nitrate removal

efficiency (10%) for more than 8 hours. Treated iron (T-NZVI) removed approximately half of nitrate concentration within 3

hours. Synthesized iron (S-NZVI) successfully reduced the whole amount of nitrate in one hour. Meanwhile, the improved

iron (I-NZVI) removed the same amount within 20 minutes, which indicated the highest performance among other NZVIs.

1. Introduction

Nitrate is a well-known hazardous contaminant in

groundwater resulting from agricultural runoff,

domestic wastewaters, etc. It can be reduced to

nitrite causing methemoglobinemia, liver damage,

cancer, and so forth [1]. Attributed to high cost

effectiveness and reactivity, NZVI had proven its

high efficacy and efficiency in nitrate

decontamination from wastewater [2, 3]. However,

NZVI’s reactivity decreases through aging [4],

which requires a certain treatment to decrease the

thick passive shell layer of iron (I, II)

oxides/hydroxides. This study investigates the

effect of optimized treatment process on old-

purchased NZVI and compares nitrate removal

kinetics between OP-NZVI, T-NZVI, S-NZVI and

I-NZVI.

2. Materials and Methods

Ferric chloride, sodium borohydride and ethanol

were purchased for NZVI synthesis. Sodium nitrate

was used to prepare a reactant stock solution, while

pH buffer solution, hydrochloric acid and sodium

hydroxide were used for pH adjustment. Nano iron

powder was purchased a year ago. Anhydrous

copper chloride was used to enhance NZVI.

To produce T-NZVI, The surface of OP-NZVI

was washed using 0.1 N HCl solution (10 g iron

per 250 mL acidic solution) in 300 mL conical

glass flask [5]. The resulting slurry was washed

with ethanol then filtered using vacuum filter.

The S-NZVI was prepared according to the

following reduction method:

.

In this study, the synthesis conditions were

optimized based on previous research work [6].

Sodium borohydride (NaBH4, 98%, 1.1472 M)

was introduced into anhydrous ferric chloride

(FeCl3, 0.1434 M) using a roller pump (flow rate 1

L/h) with a volumetric ratio of 1:1 in 500 mL four-

neck glass flask. Anoxic condition was kept by

continuous flow of nitrogen gas. The reaction

mixture was stirred at 250 rpm and maintained at

25 ± 0.5 oC. After reduction, the jet-black iron

nanoparticles were vacuum-filtered and washed

with deionized (DI) water (>100 ml/g) and

anhydrous ethanol three times each. Finally, the

slurry was vacuum-filtered and used immediately

in batch experiments.

The batch experiments were anaerobically

carried out in 500 mL four-neck glass flask using

100 mg/L of nitrate solution and 2 g/L of NZVI as

shown in Fig. 1. The resulting mixture was kept at

25 ± 0.5 oC using water bath. At specific given time

intervals, 5 mL of solution sample were withdrawn

and filtered through a 0.22 μm membrane for

analysis of nitrate, nitrite, ammonium, ferrous, total

iron and total nitrogen. The off-gas was absorbed

by 100 mL acidic solution for analysis of ammonia

gas. 50 mg of CuCl2 was added in one of

experiments to enhance nitrate removal (the used

iron is denoted as I-NZVI).

Fig. 1. Schematic layout of batch experiment.

Concentrations of nitrogen compounds in

solution samples were analysed by UV–vis

spectrophotometer. To characterize NZVI, a

transmission electron microscopy (TEM), particle

size analyser, Brunauer–Emmett–Teller (BET)

specific surface area (SSA) analyser, X-ray

diffractometer (XRD) were used to determine

morphology, particle size, SSA and crystallinity,

respectively.

3. Results TABLE I

CHARACTERIZATION OF NZVIS

Property BET-SSA

(m2/g)

Mean aggregate size

(nm)

Particle size range (nm)

OP-NZVI 16.3 1388 70-100

TNZVI 15.2 950 50-70

SNZVI 61.1 50 20-50


10

Fig. 2. TEM images of (A) OP-NZVI and (B) S-

NZVI at resolutions of 100 and 50 nm.

Fig. 3. XRD pattern of (A) OP-NZVI, (B) S-NZVI

and (C) spent I-NZVI.

Fig. 4. Nitrate removal kinetics of NZVIs.

4. Illustrations

Fig. 5. Effect of CuCl2 addition on nitrate removal

by NZVI.

5. Conclusion

This study distinguished the differences among

four types of NZVI, and succeeded to increase

nitrate removal efficiency by treatment of OP-

NZVI, using freshly-prepared NZVI synthesized

under optimized conditions, or enhancing nitrate

removal using NZVI with addition of other

contaminant via electrocatalytic reactions.

6. References [1] A. Kapoor, T. Viraraghavan, “Nitrate removal from

drinking water-review,” J. Environ. Eng., 123(4),

371-380 (1997).

[2] Y.-H. Hwang, D.-G. Kim, H.-S. Shin, “Mechanism

study of nitrate reduction by nano zero valent iron,” J.

Hazard. Mat., 185(2), 1513-1521 (2011).

[3] D. O’Carroll, B. Sleep, M. Krol, H. Boparai, C.

Kocur, “Nanoscale zero valent iron and bimetallic

particles for contaminated site remediation,” Advan.

Wat. Res., 51, 104-122 (2013).

[4] http://www.nanoiron.cz/en/news/136-nzvi-slurry-

ageing-behavior, viewed on 1/5/2015.

[5] J. Luo, G. Song,J. Liu, G. Qian, Z. P. Xu,

“Mechanism of enhanced nitrate reduction via micro-

electrolysis at the powdered zero-valent

iron/activated carbon interface,” J. Colloid. Inter. Sc.,

435, 21-25 (2014).

[6] Y.-H. Hwang, D.-G. Kim, H.-S. Shin, “Effects of

synthesis conditions on the characteristics and

reactivity of nano scale zero valent iron,” App. Cat.

B: Environmental, 105 (1), 144-150 (2011).

A B

0

500

1000

1500

2000

2500

3000

3500

4000

4500

0 20 40 60 80

Inte

nsi

ty (

Co

unts

per

sec

ond

)

2 Theta (degrees)

0

200

400

600

800

1000

1200

0 20 40 60 80

Inte

nsi

ty (C

PS)

2 Theta (degrees)

Fe0

100 nm

Fe0

Fe0

CuFe2O4.Fe3

O4

Fe0

FeOOH

Stimulated

Corrosion

Electrocatalytic

effect

Reduction

Cu2+

50 nm

Fe0

CuFe2O4.Fe3

O4

Fe0

Fe3O

4

Fe3O4

A

C

B

100 nm

50 nm


11

WAVE-STRUCTURE INTERACTION USING FREE SURFACE

LATTICE BOLTZMANN METHOD (FSLBM)

Nik Mohd1, Changhong Hu2, Xuhui Li3

1,3Interdisciplinary Graduate School of Engineering Science, Kyushu University 2Research Institute for Applied Mechanics (RIAM), Kyushu University

E-mail: [email protected], [email protected], [email protected]

Abstract: In the present study, we have developed the free surface LBM (FSLBM) algorithm for wave-structure interaction

flow problem. Standard Single Relaxation Time (SRT) approximation with the Bhatnagar-Gross-Krook (BGK) collision

model were used. In addition, a Smagorinsky Large Eddie Simulation (LES) Model was implemented in order to capture

turbulent structure in the flow. From the results, a good agreement was yielded with the published results and it provided a

validation benchmark to qualitatively verify the proposed approach.

1. Introduction

The lattice Boltzmann method (LBM) has become

an efficient method for modelling and simulating

complex fluid flows. Several advantages have

shown to be promising with algorithm operations,

data locality and computational parallelism [1]. On

the other hand, free surface flow problems have

been widely used in interdisciplinary research such

as in civil engineering and ocean engineering. For

instance, in the wave impact on offshore structure,

dam breaking, flood waves and others[2]. In this

paper, we present the progress of development of

interaction water wave by implementing the

FSLBM approach.

2. Lattice Boltzmann Method

2.1 LBM Basic

The evolution of the LBM method is described by

lattice Boltzmann equation (LBE) with the standard

lattice BGK approximation model. This is given by

[3]

(1)

(2)

. At each time , the particle

distribution function is located at the

lattice node, x at respective lattice velocity, ,

dimensionless relaxation time. Equilibrium

particle distribution function, is given by

(3)

which consider a lattice-dependent constant,

The macroscopic values for density and velocity of

particle distribution function is defined by

(4)

2.2 LES Smagorinsky Model

Naturally, free surface flows usually occur at very

high Reynolds numbers. In the lattice Boltzmann

framework, the dimensionless relaxation time is

defined by [2]

(5)

the quadratic equation is obtained as

(6)

Simplification modified relaxation rate, as

(7)

3. Free surface LBM algorithm

By using FSLBM algorithm proposed by [4], two

values will considered to calculate the fluid fraction

as given by

(8)

For an interface cell at x the mass balance with a

neighbor at is given by [5]

(9)


12

the mass for next step can be calculated by adding

to the current mass for interface cells from the mass

exchange for all directions is defined as

(10)

The corresponding free surface boundary

conditions will be constructed by [4]

(11)

4. Numerical Results

Several numerical simulations were conducted for

wave-structure interaction at a high Reynolds

number which implemented the D2Q9 LBM model.

The present numerical results were compared with

an observed experiment and published numerical

simulation by [1]. This study used the original

configuration of tank size with 45cm x 58cm which

was reported by [6].The computation with domain

mesh 464 x 360 was referred to the parameter

setting as indicated by [1]. As can be seen from the

results, in the early stage, the jet evolved slowly

and started to increase gradually to achieve a

similar pattern with the previous study in [1]. A

limitation to the present study is that it did not

consider the gate movement and surface tension

effect. Nevertheless, the overall results indicated a

qualitatively comparable result with the previous

study.

5.Conclusion

In this study, a free surface lattice Boltzmann

method (FSLBM) model has been developed to

study wave-structure interaction problems.

Extension of this work should be considered with

Multi-Relaxation Time (MRT) in order to reduce

the spurious oscillation. This is a preliminary study

using obstacle block. In the future study, we plan to

model wave maker algorithm to generate effective

water wave.

6. References [1] Christian F.Janssen, Stephan T.Grilli, Manfred

Krafczyk, “On enhanced non-linear free surface flow

simulations with a hybrid LBM-VOF model,”

Computers and Mathematics with Applications, 65,

211-229 (2013).

[2] Christian F.Janssen, Manfred Krafczyk, “Free

surface flow simulations on GPGPUs using the

LBM”, Computers and Mathematics with

Applications, 61, 3549-3563 (2011).

[3] Z. Guo, C. Shu, “Lattice Boltzmann Method and its

Applications in Engineering”, ISBN 978-981-4508-

29-2, (World Scientific Publishing, 2013).

[4] N. Thurey, “Physical based Animation of Free

Surface Flows with the Lattice Boltzmann Method”

Phd thesis, Universitat Erlangen-Nurnberg, 2013.

[5] C. van Trigt, “Free surface with lattice Boltzmann

with enhanced bubble model”, Computers and

Mathematics with Applications, 67, 331-339 (2014).

[6] A. Kolke, E.Walhorn, B.Hubner, D.Dinkler, “Strong

Couple analysis of fluid-Structure Interaction with

Free Surface Flow, “Proc.Appl.Math.Mech” 4, 338-

339(2004).

Fig.1. Breaking dam with an obstacle: experimental results of [1], hybrid LBM-VOF-PLIC Model

[1] compared to present numerical results LBM-LES

(a) t=0.16s (b) t=0.24s (c) t=0.32s (d)

t=0.50s

(e) t=0.16s (f) t=0.24s (g) t=0.32s (h) t=0.50s

(i) t=0.16s (j) t=0.24s (k) t=0.32s (l) t=0.50s


13

STUDY ON DESICCANT AIR-CONDITIONING SYSTEM FOR

AGRICULTURAL PRODUCT STORAGE IN PAKISTAN

Muhammad H. Mahmood1,*, Muhammad Sultan1, Takahiko Miyazaki2, Shigeru Koyama2

1Interdisciplinary Graduate School of Engineering Sciences, Kyushu University, Japan 2Faculty of Engineering Sciences, Kyushu University, Japan

*Email: [email protected]

Abstract: The present study provides the basic understanding of desiccant air-conditioning (AC) system for storage of

agricultural products of Pakistan. In this regard, the study provides the ideal growth zones for various agricultural products.

The desiccant air-conditioning applicability has been determined for three different climatic conditions (A, B, C) in order to

achieve the sensible and latent load of AC for the studied products. It is determined that desiccant wheel perform higher

dehumidification under climatic condition ‘A; because of the maximum ambient air relative humidity. Consequently, higher

heat energy is required in climatic condition ‘A’ to regenerate the desiccant wheel. It has been concluded that the desiccant

AC system can be effectively used for agricultural product storage in many regions of Pakistan.

1. Introduction

1.1 Background of products storage

Pakistan is an agriculture based country, blessed

with fertile land and four seasons. A huge amount

of fruits and/or vegetables in Pakistan is affected

due to post-harvest losses, which are about 30-35%

and sometimes even more. These loses are mainly

due to the mechanical damage, physiological

deterioration, and attack of insects and pests

The agricultural products after the harvesting

act like living organism and perform respiration,

transpiration, ripening processes. Furthermore,

these products contain high moisture contents and

may spoil within few days after the harvest.

Therefore, their shelf or storage life can be

increased by retarding the physiological, bio- and

chemical changes through control of temperature

and relative humidity in the cold storage structure.

1.2 Motivation of the study

The temperature of the cold storage is the most

important factor to maintain the quality of the

product because the physiological and biological

reactions in products which are directly dependant

on it. The relative humidity of the storage space is

also crucial in controlling the loss of moisture

contents from the products and to keep them in

good physical and biological appearance [1].

The mechanical refrigeration and/or air-

conditioning systems are being used in cold

storages for preservation/storage of agricultural

products. Besides from other disadvantages of

mechanical refrigeration system like environmental

degradation, high energy requirements etc., it

cannot be used for on farm storage of many tropical

fruits and vegetables such as banana, tomatoes,

oranges, mangoes, and other leafy vegetables

because of chilling injury and discoloration [2].

The standalone evaporative coolers also cannot be

used in tropical climatic conditions because of

higher relative humidity. On the other hand, the

desiccant air conditioning (DAC) system has ability

to deal the sensible and latent load of air

conditioning distinctly which gives opportunity to

use this system for storage of agricultural products

efficiently. Therefore, a particular DAC system is

proposed in the present study for on-farm pre-

cooling or short term storage of tomatoes, sweet-

potatoes, watermelon and pears.

2. Ideal storage zone

The tomatoes, sweet-potatoes, watermelon and

pears belong to same compatible storage

temperature and relative humidity group. The

recommended temperature and relative humidity is

18-21°C and 85-90% respectively. The

approximate transit and storage life of these

products are given in Table I [4]. The ideal storage

zone for preservation of tomatoes, sweetpotatoes,

watermelons and pears is shown in Fig. 1. The

ideal DAC cycle is also drawn on the figure in

order to study the system compatibility for the

recommended storage conditions.

3. Proposed DAC system

A typical desiccant wheel based DAC system is

proposed for the preservation of fruits and

vegetables as shown in Fig. 2. Psychrometric

model [5] is used to analyse the desiccant wheel

performance under three different climatic

conditions (A, B, C) of the Pakistan with low

regeneration temperature (~40°C). The

representative values of temperature and relative

humidity of the climatic condition (A, B, C) are

given in Table II [4]. The indirect evaporative

cooler (IEC) is used to provide the conditioned

supply air to the cold storage structure

TABLE I

STORAGE LIFE OF THE STUDIED AGRICULTURAL PRODUCTS [4].

Products Storage life [weeks]

Tomatoes 1-3

Sweetpotatoes 16-28

Watermelons 2-3

Pears (ripening) 1-3

mailto:[email protected]


14

TABLE II

CLIMATIC CONDITIONS FOR DESIGNING THE DAC SYSTEM [4]

Climatic

conditions type

Temperature [°C] Relative humidity

[%]

A 24 80

B 28 60

C 32 50

Ideal storage zone for storage of tomatoes, sweetpotatoes, watermelons and pears

,

15

Dry bulb temeprature [ C]

20 25 30 35 40

5

0

Hu

mid

ity r

atio

[g/k

gD

A]

15

10

Fig. 1. The ideal storage zone and DAC cycle.

IEC

Fig. 2. The proposed DAC system [3].


The ideal DAC cycle for the climatic condition (A)

is drawn on the psychrometric chart as shown in

Fig. I. The inlet air at relative humidity 80% and

temperature 24°C passes through the desiccant

wheel. The desiccant wheel dehumidifies the air up

to 11.96 g/kgDA and increases its temperature as

35.08°C. This air is then entered to the heat

exchanger for sensible cooling.

0

1

2

3

4

5

6

7

8

40 50 60 70 80

Deh

um

idif

icat

ion [

g/k

gD

A]

Regeneration Temperature [ C]

A

B

C

Fig. 3. Effect of regeneration temperature on

desiccant dehumidification.

0

5

10

15

20

25

30

35

40 50 60 70 80

Hea

t in

pu

t [k

W]

Regeneration temperature [ C]

A

B

C

Fig. 4. Energy required to dehumidify the desiccant

wheel under different climatic conditions.

Finally, the air is passed through IEC to supply the

sensibly conditioned supply air to the products. The

DAC system operates in similar fashion for other

climatic conditions (B, C). Therefore DAC cycle

for only climatic condition (A) is represented in Fig.

1 to avoid the misunderstanding.

Analysis showed that the desiccant

dehumidification capacity increases with increasing

regeneration temperature as shown in Fig. 3.

Furthermore, desiccant wheel under climatic

condition A performs higher dehumidification than

other climatic conditions. It is because of higher

outdoor air relative humidity. On the other hand,

higher heat energy is required to regenerate the

desiccant wheel under climatic condition A than

the other studied conditions as shown Fig. 4.

5. Conclusion

The DAC system can achieve the thermal

conditions required for agricultural products

storage in Pakistan at different prevailing climatic

conditions. However, its applicability in dry

climatic areas is limited.

5. References [1] lal Basediya, Amrat, D. V. K. Samuel, and Vimala

Beera. “Evaporative cooling system for storage of

fruits and vegetables-a review.” Journal of food

science and technology 50.3 (2013): 429-442.

[2] Vala, K. V., F. Saiyed, and D. C. Joshi. “Evaporative

Cooled Storage Structures: An Indian Scenario.”

Trends in post harvest technology (2014).

[3] Miyazaki T, Oda T, Ito M, Kawasaki N, and Nikai I.

“The possibility of the energy cost savings by the

electricity driven desiccant system with a high

performance evaporative cooler.” Int symp innov

mater process energy syst (2010).

[4] Kitinoja, Lisa, and Adel A. Kader. “Small-scale

postharvest handling practices: a manual for

horticultural crops.” UC, Davis, postharvest

technology research and information center, (2002).

[5] Beccali, M, Butera, F, Guanella, R, and Adhikari, R.

S. “Simplified models for the performance evaluation

of desiccant wheel dehumidification.” International

Journal of Energy Research, 27.1 (2003): 17-29.


15

THE INVESTIGATION OF THE COLORIMETRY TO MEASURE THE

DEPOSITION THICKNESS ON THE PLASMA-FACING WALL IN

QUEST

Zhengxing Wang1, Takahiro Shimoji1, Kazuaki Hanada2, Naoaki Yoshida2, Mitsuki Miyamoto3, Hideki Zushi2, Hiroshi Idei2,

Kazuo Nakamura2, Akihide Fujisawa2, Yoshihiko Nagashima2, Makoto Hasegawa2, Shoji Kawasaki2, Aki Higashijima2,

Hisatoshi Nakashima2, Akira Kawaguchi2, Tadashi Fujiwara2, Kuniaki Araki2, Osamu Mitarai4, Atsushi Fukuyama5, Yuichi

Takase6, Kenji Matsumoto7, and Quest Group.

1Department of Advanced Energy Engineering Sciences, Interdisciplinary Graduate School of Engineering Sciences, Kyushu

University, kasuga, Fukuoka 816-8580, Japan [email protected] 2Research Institute for Applied Mechanics, Kyushu University, Kasuga, Fukuoka 816-8580, Japan, 3Department of Material

Science, Shimane University, Matsue, Shimane 690-8504, Japan, 4Institute of Industrial Science and Technology Research,

Tokai University, Kumamoto 862-8652, Japan, 5Department of Nuclear Engineering, Kyoto University, Kyoto 606-8501,

Japan, 6Graduate School of Frontier Sciences, University of Tokyo, Kashiwa, Chiba 277-8561, Japan, 7HONDA R&D Co.,

Ltd. Automobile R&D Center, Haga, Tochigi 321-3393, Japan

Abstract: A convenient innovative method named colorimetry is tried to be applied to the Kyushu

University Experiment with Steady-state Spherical Tokamak (QUEST) tokamak to measure the film thickness of deposition

on the plasma-facing wall. The colorimeter can measure the reflectivity of R-color, G-color and B-color in the same time.

The reflectivity is related with the film thickness and complex refractive index of deposition. In this paper, the result of

thickness of the deposition measured with colorimeter is compared with that from the reflectivity measured with ellipsometer.

They agree quite well with each other. The result shows that it is feasible to apply colorimeter to the measurement of film

thickness of deposition. This lays the foundation for the further study of application of colorimetry to the QUEST tokamak.

1. Introduction

For a future nuclear fusion reactor, the

characteristics of the plasma-facing wall including

the information of the deposition are studied for

understanding those such as tritium storage or the

hydrogen recycling [1]. The deposition thickness is

usually measured with TEM and ellipsometer. But

these two heavy devices are very difficult to be

applied to the actual plasma-facing wall directly.

Because of the convenience, the colorimetry can

measure the deposition thickness on the actual

plasma-facing wall easily. About the measurement of deposition thickness

the colorimetry is in good agreement with the

ellipsometry for a-C:H (amorphous hydrogenated

carbon) dominant deposition in TEXTOR [2]. But

for the metal first wall and divertor in ITER, the

deposition will contain metals and will not a-C:H

dominant any more. And for QUEST the first wall

is stainless steel and the divertor and limiter are

tungsten. After several campaigns the deposition

has been formed on the plasma-facing wall. The

main ingredient is the mixture of carbon and metal.

It is necessary to study the feasibility of the

colorimetry on measuring the metal-containing

deposition thickness.

2. The colorimeter

1. Measuring data RGB (Red, Green, Blue)

reflectivity.

2. Response range: R (590~720 nm), G

(480~600 nm), B (400~540nm).

Maximum sensitivity: R (615nm), G (540nm), B

(465nm). The sensitivity versus wavelength is

Gaussian distribution.

3. Sensor diameter: 8.1 mm.

4. Diameter of the integrating sphere: 47 mm.

5. Light source: white LED.

6. The reflectivity of light is output as RGB

values (0~1023).

(1)

(2)

(3)

3. The principle of measurement with the

colorimeter

Fig. 1. The principle of measurement of deposition

thickness with colorimeter

2

~

21

~

10

~

0 sinsinsin ininin (4)

1

~

111 cos2

ind

(5)

1

~

10

~

0

1

~

10

~

0s1

coscos

coscos

inin

ininr

1

~

00

~

1

1

~

00

~

1p1

coscos

coscos

inin

ininr

(6)

2

~

21

~

1

2

~

21

~

12

coscos

coscos

inin

ininr s

2

~

11

~

2

2

~

11

~

22

coscos

coscos

inin

ininr p

(7)

1

1

2

,2,1

2

,2,1

,s1

i

psps

i

psps

perr

errr

(8)

2

2

352

)615(

0RRI

R

eI

2

2

202

)540(

0GGI

G

eI

2

2

222

)465(

0BBI

B

eI


16

pspsp rrR ,,,s (9)

2

R ps RR

(10)

According to the equations from (4) to (10), if

the incident angle i0 and the complex refractive

index of support n2 have been known (For QUEST,

the support is stainless steel) and it is in the air, so

there is

),(R 11

_

dnf (11)

The colorimeter can measure the reflectivity R.

If the complex refractive index of deposition en1

has been known, then the deposition thickness d1

could be derived.

4. The measurement of complex refractive index

of deposition on the QUEST wall with samples

Fig. 2. The samples on the first wall of QUEST

Fig. 3. The T1, T2, T3, T4 and B15 measured with

TEM.

These samples could be measured with

ellipsometer and TEM so that the complex

refractive index of deposition could be studied.

Fig. 4. The measurement principle of the complex

refractive index of deposition with ellipsometry.

ipisi

ipip

i

isis eEEeEE

~~

, (12)

rprsi

rprp

i

rsrs eEEeEE

~~

, (13)

~

~

))()((

/

/Etan

s

pi

isip

rsrpi

r

re

EE

Ee isiprsrp

(14)

According to the equations from (4) to (9), and then

there is

),()(tan 1

~

12

~

1 dnfnfe e

i (15)

The Ψ and Δ could be measured with

ellipsometer directly. The n1 is the deposition’s

complex refractive index which contains two

parameters, refractive n1 and extinction coefficient

k1. The ne is the sample’s complex refractive index.

If the d1 has been measured with TEM, then the n1

could be derived.

5. The complex refractive index of the samples

and the deposition of T1, T2, T3, T4 and B15

The three maximum sensitive wavelengths of

RGB colorimeter have been chosen to calculate the

complex refractive index of deposition.

Fig. 5. The complex refractive index of the 21

samples and the deposition of T1, T2, T3, T4 and

B15

6. Conclusion

From the distribution of complex refractive

index of 21 samples, it is presented that on the

plasma-facing wall of QUEST there are three

regions in which the optical characteristic of

deposition is different. In the upper side and lower

side, we need do more experiment of TEM to get

more complex refractive index of deposition, so

that we can get right complex refractive index

which could be applied to the colorimetry.

7. References [1] K. Hanada, “Particle balance in long duration RF

driven plasmas on QUEST”,Journal of Nuclear

Materials, 463, 1084-1086 (2015).

[2] P. Wienhold, “Colorimetry of interference colours to

investigate thickness changes of protective coatings

in TEXTOR”, Nuclear Instruments and Methods in

Physics Research Section B: Beam Interactions with

Materials and Atoms, 94, 503 (1994).


17

NUMERICAL STUDIES OF NATURAL VENTILATION OF BUILDING

WITH LEEWARD OPENINGS

A. Idris1, B. P. Huynh2, Z. Abdullah3

1Universiti Kuala Lumpur Malaysian Spanish Institute, Kulim Hi-Tech Park, 09000 Kulim, Kedah Malaysia

[email protected] 2, 3Faculty of Engineering & IT, University of Technology, Sydney NSW 2007Australia

Abstract: Ventilation is a process of changing air in an enclosed space. Air should continuously be withdrawn and replaced

by fresh air from a clean external source to maintain internal good air quality, which may referred to air quality within and

around the building structures. In this work, computational simulation is performed on a real-sized box-room with

dimensions 5 m x 5 m x 5 m. Two opening of the total area 4 m2 are differently arranged, resulting in 4 configurations to be

investigated. A logarithmic wind profile upwind of the building is employed. A commercial Computational Fluid Dynamics

(CFD) software package CFD-ACE of ESI group is used. A Reynolds Average Navier Stokes (RANS) turbulence model &

LES turbulence model are used to predict the air’s flow rate and air flow pattern. The governing equations for large eddy

motion were obtained by filtering the Navier-Stokes and continuity equations.

1. Introduction

Natural ventilation is the process of supplying and

removing air through an indoor space without using

mechanical systems. In developed countries, most

of the buildings are responsible for 1/3 of all energy

consumption e.g. in US, about 30% of total energy

consumption is used in non-domestic buildings,

and of that fraction about 30% is used in heating

and cooling [1], [2], [3].

Single-sided ventilation has fewer adaptive

comfort hours than two-sided ventilation and

produces much less ventilation volume [4] is used

when there is non-availability of other choice.

The reliable methods of getting information

about the air flow and pressure distribution around

and inside the building are through full scale

measurements [5], scale model testing using wind

tunnel [6] and computational fluid dynamics (CFD)

[7, 8].

The common CFD techniques are direct

numerical simulation (DNS), large-eddy simulation

(LES) and Reynolds averaged Navier-Stokes

(RANS) equation with turbulence models. Each

technique handles turbulence in different ways [9,

10]. Among those techniques, RANS is widely

used by most CFD software [11], however, LES

should be more suitable than the RANS approach

to study highly three-dimensional or separated

flows, especially those in which the gradient

transport hypothesis, and consequently one and

two-equation models of turbulence, fails [12]. It

was introduced by Dearsdoff in the early 1970s

[13] for meteorological applications.

The present investigation is focused on the

application of three dimensional RANS and LES

modeling on wind driven natural ventilation of

double openings at single-sided buildings.

2. Methodology

This study involves 4 configurations models of

leeward double opening (Figure 1). The dimensions

of the building-like models are 5 m x 5 m x 5 m.

The computational domain constructed had a height

of 4H, width of 9H and length of 13H (H=5m),

sufficiently large to avoid disturbance of air flow

around the building [14]. A logarithmic wind

profile upwind of the building is employed.

The commercial software package CFD-ACE

from the ESI group is used for the computation.

2. Result & Discussion

The present results is about the study of the flow

patterns and air flow rate using RANS and LES

scheme when modelling wind driven natural

ventilation in buildings. The velocity components

have been determined, along streamwise and

vertical direction, respectively. LES is used to

compare with RANS as LES is produced more

precise results but more time-consuming method

[15].

Figure 1 shows the air flow pattern inside the

building. The opening is marked as A and B, where

the air enters the building through the lower

opening A and comes out through the upper

opening B. If the level of the opening is at the same

height, the air will enter through the lower area and

come out through the upper area of both openings.

Table 1 shows numerical values of air flow rate

of simulation that being performed at University of

Technology, Sydney. Each configuration gives

different flow rate and it is found that the highest

flow rate occurs at configuration a(iv) for RANS

and a(ii) for LES, whereas the lowest flow rate

occurred at configuration a(i) for both RANS and

LES.

4. Conclusion

In this recent study, RANS and LES approaches

have been applied to wind driven natural

ventilation in a cubic building. Both RANS & LES

scheme give almost the same air flow rate with the

difference between 4-8%. The location &

arrangement of opening influences the air flow rate

and air flow pattern.


18

Table 1. Numerical values of airflow rate

Opening

Configuration

Air flow rate (m3/s)

RANS LES

a(i) 6.63 x 10-1 6.09 x 10-1

a(ii) 6.91 x 10-3 7.46 x 10-3

a(iii) 6.52 x 10-1 7.10 x 10-1

a(iv) 7.45 x 10-1 7.15 x 10-1

Configuration 2D-Airflow Pattern

a(i)

a(ii)

a(iii)

a(iv)

Figure 1 Double Opening on Leeward Wall

5. References [1] P. F. Linden, "The fluid mechanics of natural

ventilation," Annual Review of Fluid

Mechanics, vol. 31, pp. 201-238, 1999.

[2] QUT High‐Density Liveability Guide (2009,

28 March).

Thermal Comfort and Natural Ventilation.

Available:

http://www.highdensityliveability.org.au/pdf/1

_Thermal%20Comfort_23sep09.pdf

[3] J. Morrissey, T. Moore, and R. E. Horne,

"Affordable passive solar design in a temperate

climate: An experiment in residential building

orientation," Renewable Energy, vol. 36, pp.

568-577, 2// 2011.

[4] Y. Wei, Z. Guo-qiang, W. Xiao, L. Jing, and X.

San-xian, "Potential model for single-sided

naturally ventilated buildings in China," Solar

Energy, vol. 84, pp. 1595-1600, 9// 2010.

[5] M. M. Eftekhari, L. D. Marjanovic, and D. J.

Pinnock, "Air flow distribution in and around a

single-sided naturally ventilated room,"

Building and Environment, vol. 38, pp. 389-

397, 3// 2003.

[6] T. S. Larsen and P. Heiselberg, "Single-sided

natural ventilation driven by wind pressure and

temperature difference," Energy and Buildings,

vol. 40, pp. 1031-1040, // 2008.

[7] G. Gan, "Effective depth of fresh air

distribution in rooms with single-sided natural

ventilation," Energy and Buildings, vol. 31, pp.

65-73, 1// 2000.

[8] K. Visagavel and P. S. S. Srinivasan, "Analysis

of single side ventilated and cross ventilated

rooms by varying the width of the window

opening using CFD," Solar Energy, vol. 83, pp.

2-5, 1// 2009.

[9] Q. Chen, "Ventilation performance prediction

for buildings: A method overview and recent

applications," Building and Environment, vol.

44, pp. 848-858, 4// 2009.

[10] Q. Chen and L. Glicksman. Application of

computational fluid dynamics for indoor air

quality studies. Available:

http://www.accessengineeringlibrary.com/mgh

pdf/0071450076_ar059.pdf.

[11] V. Yakhot, S. A. Orszag, S. Thangam, T. B.

Gatski, and C. G. Speziale, "Development of

turbulence models for shear flows by a double

expansion technique," Physics of Fluids A:

Fluid Dynamics (1989-1993), vol. 4, pp. 1510-

1520, 1992.

[12] U. Piomelli, "Large-eddy simulation:

achievements and challenges," Progress in

Aerospace Sciences, vol. 35, pp. 335-362, 5//

1999.

[13] J. W. Deardorff, "A numerical study of three-

dimensional turbulent channel flow at large

Reynolds numbers," Journal of Fluid

Mechanics, vol. 41, pp. 453-480, 1970.

[14] Y. Jiang, D. Alexander, H. Jenkins, R. Arthur,

and Q. Chen, "Natural ventilation in buildings:

measurement in a wind tunnel and numerical

simulation with large-eddy simulation,"

Journal of Wind Engineering and Industrial

Aerodynamics, vol. 91, pp. 331-353, 2// 2003.

[15] G. Evola and V. Popov, "Computational

analysis of wind driven natural ventilation in

buildings," Energy and Buildings, vol. 38, pp.

491-501, 5// 2006.


19

STATISTICAL EVALUATION OF COMPRESSION INDEX

CORRELATIONS

Waled A. Daoud1, Kiyonobu Kasama2, Naser M. Saleh3, Abdelazim M. Negm4

1 PhD Student, Egypt-Japan University of Science and Technology, Alexandria, Egypt (email: [email protected]) 2 Associate Professor, Civil and Structural Engineering Division, Faculty of Engineering, Kyushu University, Japan

3 Associate Professor, Civil Engineering Department, Faculty of Engineering at Shoubra, Cairo, Egypt 4 Environmental Engineering Department Head, Egypt-Japan University of Science and Technology, Alexandria, Egypt

Abstract: Primary consolidation of cohesive soil can significantly affect the serviceability of overlying structures and its

amount is calculated using the compression index (Cc). Determination of Cc is complex and time consuming that raised the

need for using empirical correlations with simpler tests. Development of these correlations started as early as the 1940s and

new correlations still being developed. Most of these correlations were derived using data fitting with site-specific measured

values and was evaluated using simple statistics. Better statistical evaluation may reduce the correlation deviation. ATIC

method in-line with other evaluation measures was used to evaluate 92 compression index correlations using measured data

from Egypt, UAE, Iraq, and Indonesia. For the studied data, each statistical measure ranks the correlations differently,

especially for the best correlation. The advantages and shortcoming of each statistical measure were briefly introduced.

1. Introduction

Soil compressibility can significantly affect the

serviceability of the overlying structures [1]. For

cohesive soil; primary consolidation has the most

significant effect that is estimated using

compression index (Cc). Determination of Cc is

complex and time consuming that made empirical

correlations more important. Many correlations

were developed to estimate Cc for different soil

conditions as early as the 1940s and new

correlations still being developed [2].

Most of these correlations were derived from

data fitting of measurements at specific site

conditions that may cause large deviation if used

for other sites [3]. Better statistical evaluation may

reduce the overall deviation of the geotechnical

parameter and enhance the overall assessment.

Many researchers attempt to evaluate Cc

correlations using statistical measures for decades.

Giasi et al. [4] evaluated 32 correlations using both

ranking distance (RD) and Ranking Index (RI).

Yoon et al. [5] evaluated 15 correlations using

correlation coefficient (R) and K-factor. Rani and

Rao [6] evaluated 12 correlations using The mean

absolute difference (MAD). Onyejekwe et al. [7]

evaluate 18 correlations using the root mean square

of error (RMSE), K-factor, RD, and RI. Lee et al.

[2] used determination coefficient (R2) and mean

absolute difference (MAD) to evaluate 29

correlation.

Most of the commonly used statistical

evaluation measures had shortcomings that it

considers position conformity or trend conformity

separately. This shortcoming may cause

misjudgement of the correlation. This paper

evaluates 92 Cc correlations using commonly used

statistical measures and Amended Theil Inequality

Coefficient (ATIC) method as presented by Song et

al. [8]. ATIC method has the advantage that it takes

into account both position and trend conformities in

the overall evaluation process.

2. Used Data

Subsurface investigation reports were collected

from Egypt, UAE, Iraq, and Indonesia and entered

into customized geotechnical database. Data for

this study was collected with the condition that the

sample has all needed parameters to insure

consistency and accuracy of the evaluation process.

Table 1 shows descriptive statistical measures for

the used soil properties. TABLE I

DESCRIPTIVE STATISTICS OF THE USED SOIL PROPERTIES

Property Range Mean Std. Dev.

Initial Voids Ratio 0.32 - 4.35 1.58 0.84 Bulk Density (t/m3) 1.04 - 2.29 1.61 0.31

Water Content (%) 11.9 - 168.12 57.15 31.09

Liquid Limit (%) 17.1 - 166.2 62.68 25.22 Plasticity Index (%) 2.48 - 113.9 30.71 17.84

Compression Index 0.07 - 1.66 0.57 0.3

3. Used Correlations

Several correlations were developed to correlate the

Cc with field state and intrinsic properties as in

Table II. Total of 92 correlations were considered

in this study. TABLE II

CORRELATIONS WITH SINGLE SOIL PROPERTY

Cor.

ID Formula

Cor

. ID Formula

C01 C47

C02 C48

C03 C49

C04 C50

C05 C51

C06 C52

C07 C53

C08 C54

C09 C55

C10 C56

C11 C57

C12 C58

C13 C59

C14 C60

C15 C61

C16 C62


20

C17 C63

C18 C64

C19 C65

C20 C66 C21 C67

C22 C68

C23 C69

C24 C70

C25 C71

C26 C72

C27 C73

C28 C74

C29 [9] C75

C30 C76

C31 C77

C32 C78

C33 C79

C34 C80

C35 C81

C36 C82

C37 C83

C38 C84

C39 C85

C40 C86

C41 C87

C42 C88

C43 C89

C44 C90

C45 C91

C46 C92

Where e: voids ratio, n: porosity, : dry density (g/cm3), LL: liquid limit,

IP: plasticity index, WC: natural water content, and GS: specific gravity

4. Correlations Evaluation

Current correlation evaluation measures consider

position and trend conformities separately that may

lead to misjudgment and wrong selection. The

ATIC method as proposed by Song et al. [8] has

the advantage that it considers both position and

trend conformities.

Table IV shows the 5 top-most and bottom-most

ranked correlations based on ATIC method using

the procedure given in [8]; R2 and MAD values

based on the equations given in [2]; RMSE value

based on the equation given in [7]; and RI and RD

values based on the procedures given in [4]. TABLE V

THE 5 TOP-MOST AND BOTTOM-MOST RANKED CORRELATIONS

Rank ATIC R2 MAD RMSE RI RD

5 T

op

-

Most

1 C22 C29 C60 C60 C67 C02 2 C80 C79 C29 C67 C20 C34 3 C43 C10 C33 C17 C23 C35

4 C10 C02 C10 C33 C64 C49 5 C50 C04 C17 C64 C92 C27

5 B

ott

om

-

Mo

st

88 C03 C71 C62 C03 C11 C45

89 C14 C55 C83 C14 C03 C14

90 C13 C56 C14 C62 C14 C13 91 C62 C53 C13 C13 C62 C83 92 C15 C54 C15 C15 C13 C15

5. Conclusion

Total of 92 compression index correlations were

evaluated using commonly used statistical

measures in-line with ATIC method. The results

show that each statistical measure gave different

ranking for the correlations. ATIC method has the

advantage of considering position and trend

conformity in the evaluation process. R2 has the

shortcoming that it considers only the trend of the

values without considering their relative position.

MAD and RMSE values have the shortcoming that

they evaluate only the values’ position around the

average. The RI considers only the position of the

data. The RD is biased for the odd ratios between

the correlated and observed values.

Acknowledgments

Waled A. Daoud would is thankful to the Egyptian

Ministry of Higher Education (MoHE) and EJUST

for funding his PhD studies and special thanks to

Kyushu University for offering the tools and

equipment needed for the research.

References [1] J. E. Bowles, Foundation Analysis and Design,

5th ed., vol. 20, no. 3. McGraw-Hill, 1997.

[2] C. Lee, S.-J. Hong, D. Kim, and W. Lee,

“Assessment of compression index of Busan

and Incheon Clays with sedimentation state,”

Mar. Georesources Geotechnol., vol. 33, no. 1,

pp. 23–32, 2015.

[3] B. A. McCabe, B. B. Sheil, M. M. Long, F. J.

Buggy, and E. R. Farrell, “Empirical

correlations for the compression index of Irish

soft soils,” Proc. ICE-Geotechnical Eng., vol.

167, no. 6, pp. 510–517, 2014.

[4] C. I. Giasi, C. Cherubini, and F. Paccapelo,

“Evaluation of compression index of remoulded

clays by means of Atterberg limits,” Bull. Eng.

Geol. Environ., vol. 62, no. 4, pp. 333–340,

2003.

[5] G. L. Yoon, B. T. Kim, and S. S. Jeon,

“Empirical correlations of compression index

for marine clay from regression analysis,” Can.

Geotech. J., vol. 41, no. 6, pp. 1213–1221, 2004.

[6] C. S. Rani and K. M. Rao, “Statistical

evaluation of compression index equations,” Int.

J. Civ. Eng. Technol., vol. 4, no. 2, pp. 104–117,

2013.

[7] S. Onyejekwe, X. Kang, and L. Ge,

“Assessment of empirical equations for the

compression index of fine-grained soils in

Missouri,” Bull. Eng. Geol. Environ., 2014.

[8] J. Song, L. Wei, and Y. Ming, “A method for

simulation model validation based on Theil’s

inequality coefficient and principal component

analysis,” in AsiaSim 2013, Springer, 2013, pp.

126–135.

[9] B. Tiwari and B. Ajmera, “New correlation

equations for compression index of remolded

clays,” J. Geotech. Geoenvironmental Eng., vol.

138, no. 6, pp. 757–762, 2011.


21

HETROJUNCTION DIODE OF NITROGEN-DOPED

ULTRANANOCRYSTALLINE DIAMOND FILMS PREPARED BY

COAXIAL ARC PLASMA DEPOSITION

Abdelrahman Zkria1,2, Tsuyoshi Yoshitake1

1Department of Applied Science for Electronics and Materials, Kyushu University, Kasuga, Fukuoka 816-8580, Japan

E-mail: [email protected] 2Physics Department, Aswan University, Aswan 81528, Egypt,

Abstract: Nitrogenated ultrananocrystalline diamond/hydrogenated amorphous carbon composite films were prepared in

atmospheres of nitrogen and hydrogen mixed gases, by coaxial arc plasma deposition method. Effect of nitrogen

incorporated to the films was electrically investigated. The nitrogen-doped films possesses n-type semiconductor, it was

confirmed by studying the heterojunction of the film with p-Si. I-V curve of the junction exhibited a high rectifying action in

room temperature. The obtained results confirm that the Nitrogen-doped ultrananocrystalline diamond/hydrogenated

amorphous carbon composite film is a good candidate material for the electronic device applications.

1. Introduction

Diamond like carbon (DLC) materials have been

of considerable interest to many researchers in

recent years, mainly due to their amorphous nature

which opens up a possibility of incorporating of

other elements such Si, F, P, Ag and N1,2) which

upgrades its application in the field of

semiconductor technology. However, the high-

density defects and complex structure negatively

effect on the doping process; accordingly limit their

applications on the electronic devices.

On the other hand, semiconducting

ultrananocrystalline diamond/hydrogenated

amorphous carbon composite (UNCD/a-C:H) films

have a specific structure, wherein a large number of

diamond crystallites of less than 10 nm diameter

are embedded into an a-C:H matrix. These films

are applicable to biomedical, mechanical and

electronics applications3,4) because of their

advantageous features including their unique

optical and electrical properties. It has been

experimentally proved that the UNCD/a-C:H films,

prepared by pulsed laser deposition (PLD)5) and

coaxial arc plasma deposition6) possess large

optical absorption coefficients. This property

mainly attributed to the UNCD grain boundaries7).

UNCD/a-C:H films attract many researchers due

to the possibility of realizing n-type and p-type

conduction by Nitrogen and Boron doping,

respectively8,9). Concerning nitrogen-incorporated

UNCD, Bhattacharyya etal10) have reported that

UNCD films can be doped with nitrogen to achieve

a conductivity of 143 S/cm by chemical vapor

deposition (CVD) method. In our group, we

succeed to grow nitrogen-doped UNCD by PLD

method11), and it is experimentally proved that the

electrical conductivity of the film is enhanced with

increasing nitrogen contents, the enhancement of

conductivity with nitrogen doping was attributed to

the change of structure properties of films for both,

CVD12), and PLD8,11) methods.

In this letter, we report heterojunction diodes

comprising nitrogen-doped UNCD/a-C:H and p-

type Si and fabricated by coaxial arc plasma

deposition method. The heterojunction performance was evaluated based on current–

voltage (I–V) measured in dark at room

temperature.

2. Experimental details

UNCD/a-C:H films were deposited on p-type Si

(100) substrates at a substrate temperature of 550

ºC in a nitrogen and hydrogen mixed gas

atmosphere of 53.3 Pa by coaxial arc plasma

deposition (CAPD) with a coaxial arc plasma gun

equipped with a graphite target6). The CAPD

apparatus was evacuated down to a base pressure

of less than 10Pa using a turbo molecular pump,

and then hydrogen and nitrogen gas was fed into

the apparatus at a total inflow rate of 10 sccm.

Heterojunction was formed by depositing nitrogen

doped UNCD and p-type Si by radio frequency

(RF) sputtering. Pd electrode was deposited on

both sides for ohmic contact.


Formation of diamond grains by CAPD compared

with that by CVD has the following major

differences: (i) lower substrate temperatures, (ii)

much higher deposition rates, and (iii) no

requirement of substrate pretreatment using

diamond powder

Fig. 1. Schematic of coaxial arc plasma

deposition (CAPD) system.

mailto:[email protected]


22

From these differences, the formation mechanism

by CAPD should be greatly different from that by

CVD. Schematic of coaxial arc plasma deposition

(CAPD) system is shown in Fig.1. UNCD/a-C:H

films have electrically insulating properties In

contrast, 3 at. % nitrogen content UNCD/a-C: H

films deposited of hydrogen and nitrogen gases are

electrically conductive.

Fig.2. (a) Top view microscope image of the diode

(b). Schematic diagram of heterojunction

In order to illustrate the conduction type of

nitrogen-doped UNCD/a-C:H films, heterojunction

diodes have been manufactured by depositing films

on p-type Si substrates, as shown in Fig. 2. To

provide better understanding of these

heterojunction properties, dark IV curves of the

diodes measured in dark at room temperature, as

shown in Fig.3. I–V curve exhibited a rectifying

action similar to that of conventional p-n

heterojunctions with a rectification ratio of more

than 104 for bias voltage of ± 1 V.

−2 −1 0 1 2 3

0

1

2

3

4

5

Cu

rre

nt (m

A)

Voltage (V)

−5 −4 −3 −2 −1 0 1 2 3 4 510

−9

10−8

10−7

10−6

10−5

10−4

10−3

10−2

10−6

10−5

10−4

10−3

10−2

10−1

100

101

Curr

ent (A

)

Curr

ent density (

A/c

m2)

Voltage (V)

Fig. 3. Current-Voltage characteristics of the

heterojunction diode. Inset is linear scale.

These results evidently confirm that nitrogen-doped

UNCD/a-C:H possesses n-type conduction and

easily forms junction with p-type semiconductor

4. Conclusion

3 at. % nitrogen-doped UNCD/a-C:H film

deposited by CAPD method possessed n-type

conduction, Heterojunction diodes composed of n-

type UNCD/a-C:H and p-type Si showed a typical

rectifying action with a high rectification ratio of

~104 at bias voltage of ± 1 V. The obtained results

suggest n-UNCD/a-C:H as a promising candidate

to be applied in electronic and optoelectronic

devices

5. References 1. T. Yokota, T. Terai, T. Kobayashi, T. Meguro, M.

Iwaki, Surf. & Coatings Tech.201, 8048–8051

(2007).

2. S.F.Ahmed, D.Banerjee, K. Chattopadhyay, Vac.

84, 837–842 (2010).

3. S. Srinivasan, J. Hiller, B. Kabius, O. Auciello,

Appl. Phys. Lett. 90, 134,101, (2007).

4. Williams, Semicond. Sci. Technol. 21, R49

(2006).

5. K. Hanada, T. Nishiyama, T. Yoshitake, and K.

Nagayama: J. Nanomater, 901241 (2009).

6. K. Hanada, T. Yoshida, Y. Nakagawa, and T.

Yoshitake: Jpn. J. Appl. Phys. 49, 125503 (2010).

7. T. Yoshitake, A. Nagano, M. Itakura, N. Kuwano,

T. Hara, and K.Nagayama: Jpn. J. Appl. Phys. 46,

L936 (2007).

8. S. Al-Riyami, S. Ohmagari, and T. Yoshitake,

Appl. Phys. Express 3, 115102 (2010).

9. S. Ohmagari and T. Yoshitake, Jpn. J. Appl. Phys.

51, 090123 (2012).

10. Bhattacharyya, O. Auciello, J. Birell, J.A. Carlisle,

L.A. Curtiss, A.N. Goyette, D.M. Gruen, A.R.

Krauss, J. Schlueter, T. Sumant, P. Zapol, Appl.

Phys. Lett. 79, 1441 (2001).

11. S. Al-Riyami, S. Ohmagari, T. Yoshitake:

Diamond Relat. Mater.19, 510–513 (2010)

12. J. Birrell, J.E. Gerbi, O. Auciello, J.M. Gibson,

D.M. Gruen, J.A. Carlisle, Appl. Phys.Lett. 93, 9

(2003).

13. Williams, M. Nesladek, M. Daenen, S.

Michaelson, A. Hoffman, E. Osawa, K.Haenen,

R.B. Jackman, Diamond Relat. Mater. 17, 1080

(2008).


23

ULTRANANOCRYSTALLINE DIAMOND/AMORPHOUS CARBON

COMPOSITE FILMS SYNTHESIS ON CEMENTED CARBIDE

SUBSTRATE BY COAXIAL ARC PLASMA DEPOSITION

Mohamed Egiza1, 2, Hiroshi Naragino2, Aki Tominaga2, Kouki Murasawa3, Hidenobu Gonda3, Masatoshi Sakurai3, and

Tsuyoshi Yoshitake2 1 Mechanical Engineering Department, Kafrelsheikh University, Kafrelsheikh, 33516, Egypt, [email protected]

2 Department of Applied Science for Electronics and Materials, Kyushu University, Japan, 3 OSG Corp., Japan

Abstract: Ultrananocrystalline diamond/amorphous carbon composite films, wherein a larger number of diamond grains

with diameters of less than 10 nm are embedded in an amorphous carbon matrix, were deposited on cemented carbide (WC-

6wt.% Co) substrates by using coaxial arc plasma deposition at different repetition rates and deposition temperatures. The

hardness and Young’s modulus were measured by nanoindentation. The film deposited at a repetition rate of 1 Hz and room

temperature exhibited the maximum hardness of 51 GPa and modulus of 520 GPa. This implies that catalytic effects of Co in

the WC-Co substrates, which induce the graphitization of the films, can be suppressed by the low deposition temperature,

since the catalytic effects are enhanced with increasing temperature.

Key words: Nanodiamond, Hard coating, Coaxial arc plasma deposition, CAPD, Nanoindentation, Cutting tools

1. Introduction

Hard coating can give functions such as enhanced

hardness, chemically stable surface, and thermal

protection to mechanical tools, particularly

lengthen the lifetime of the tools. Cathodic arc

discharge has ever been employed for the

deposition of sp3-rich amorphous carbon film

coatings [1], since this method can provide highly

energetic carbon ions, which is indispensable for

the formation of sp3 bonds, onto substrates [2- 4].

In addition to the merits of cathodic arc

discharge, coaxial arc plasma deposition (CAPD)

has the following distinctive feature: a coaxial arc

plasma gun is equipped with an anodic cylinder

that can bunch ions ejected from a cathodic

graphite rod located inside the cylinder. Owing to

this structure, a supersaturated condition with

highly energetic ions can be realized, which is

desired for the growth of ultrananocrystalline

diamond (UNCD) crystallites [5]. CAPD can form

ultrananocrystalline diamond/amorphous carbon

composite (UNCD/a-C) films without the pre-

treatment of substrates with diamond powder [6]. A

high hardness of UNCD/a-C films are attributed to

the coexistence UNCD grains and sp3-rich a-C [7].

Cemented carbide (WC-6wt.% Co), which is a

typical cutting tool substance, contains Co that

induce graphitization of diamond. Since the

substrate temperature is increased, the catalytic

effects of Co is enhanced [8-10], polycrystalline

diamond coating by chemical vapour deposition is

made after the preferential etching of the Co.

CAPD make possible the deposition of UNCD/a-

C:H films at low temperature without the pre-

treatment of substrates. The process and condition

of deposition are completely different between

CAPD and CVD.

In this study, the validity of employing CAPD

for UNCD/a-C coating is investigated, and the

influences of cemented carbide substrates on the

film growth and mechanical properties were mainly

discussed.

2. Experimental procedures

UNCD/a-C films with thicknesses of 2-3 µm were

deposited on cemented carbide plate substrates with

dimensions of 10 mm diameter and 4 mm thickness

by CAPD with a graphite target. As shown in Fig. 1,

the film was not peeled off. The main conditions of

samples are summarized in Table 1. The films

surfaces were observed by scanning electron

microscopy. The hardness and Young’s modulus of

the films were measured by nanoindentation.

Fig.1. Optical images of (a) uncoated and (b)

UNCD/a-C-coated cemented carbide plate

substrates.

TABLE I

Typical preparation conditions.

sample 1 sample 2 sample 3

Substrate Temp. [℃] 550 RT RT

Repetition Rate [Hz] 5 5 1

3. Results and discussion

Fig.2 shows the X-ray diffraction pattern of the

films. Diffraction rings due to diamond-111 and

220 are observed, which evidently indicates the

formation of UNCD grains in the films.

Whereas the film (sample 1) deposited at 550

C and 5 Hz exhibited 17.3 GPa hardness, films

deposited on Si substrates at the same conditions

exhibited approximately 30 GPa. This reduction in

hardness should be attributed to the catalytic effects

of Co containing in the cemented carbide [8-10].

(a) (b)


24

The film deposited at room temperature and 5 Hz

(sample 2) exhibited 31.8 GPa hardness and 294

GPa Young’s modulus. At the low substrate

temperature, as shown in Fig. 3, the hardness and

Young’s modulus are obviously enhanced, which

implies that graphitization due to the Co catalytic

effects are suppressed at the low substrate

temperature.

Figure 4 shows typical SEM images of the film

surfaces. The films deposited at 550 C (sample 1)

obviously comprises large grains as compared with

those of sample 2, which might come from the

graphitization.

Fig. 2. XRD pattern of typical UNCD/a-C film.

Fig.3. Comparison in hardness and Young’s

modules among samples.

Fig. 4. SEM images of film surface of (a) sample 1

and (b) sample 2.

The film deposited at room temperature and 1

Hz (sample 3) exhibits 51.3 GPa hardness and 520

GPa Young’s modulus. Both hardness and modules

are evidently enhanced with decreasing repetition

rate of discharge from 5 to 1 Hz. The long interval

time 1 s at 1 Hz is advantageous from the

viewpoints of cooling. In other words, an increase

in the effective substrate temperature might be

suppressed at 1 Hz owing to the long interval time,

since highly energetic species are arrived at the

substrate and the effective temperature on the

substrate surface should temporally be increased.

4. Conclusion

UNCD/a-C films were deposited on cemented

carbide substrates containing Co by CAPD. It was

found that the low temperature growth is effective

for suppressing the graphitization due to the Co

catalytic effects and CAPD, that makes the low

temperature growth possible, is a promising

method for the hard coating applications of

UNCD/a-C on cemented carbide.

Acknowledgments

This work was partially supported by Osawa

Scientific Studies and Grants Foundation.

References [1] M.Chhowalla et al., “Stationary carbon cathodic arc:

Plasma and film characterization,”J. Appl. Phys., 79,

2237-2244 (1996).

[2] D.Liu et al., “Filtered pulsed carbon cathodic arc:

plasma and amorphous carbon properties,”J. Appl.

Phys., 95, 7624-7631 (2004).

[3] J.J.Cuomo et al., “Vapor deposition processes for

amorphous carbon films with sp3 fractions

approaching diamond,”J. Appl. Phys., 70, 1706-1711

(1991).

[4] M.Chhowalla et al., “Influence of ion energy and

substrate temperature on the optical and electronic

properties of tetrahedral amorphous carbon (ta-C)

films,”J. Appl. Phys., 81, 139-145 (1997).

[5] K. Hanada et al., “Time-Resolved Spectroscopic

Observation of Deposition Processes of

Ultrananocrystalline Diamond/Amorphous Carbon

Composite Films by Using a Coaxial Arc Plasma

Gun,” Jpn. J. Appl. Phys., 49, 08JF09 (2010).

[6] K. Hanada et al., “Formation of

Ultarananocrystalline Diamond/Amorphous Carbon

Composite Films in Vacuum Using Coaxial Plasma

Gun,” Jpn. J. Appl. Phys., 49, 125503 (2010).

[7] Nevin N. Naguib et al., “Enhanced nucleation

smoothness and conformality of ultrananocrystalline

diamond (UNCD) ultrathin films via tungsten

interlayers,” Chem. Phys. Lett., 430, 345-350 (2006).

[8] Mario Lessiak et al., “Diamond coatings on hardmetal

substrates with CVD coatings as intermediate

layers,” Surface& Coatings Tech., 230,119-123

(2013).

[9] R. Haubner et al., “On the formation of diamond

coatings on WC/Co hard metal tools,” Ref. Metal.

&Hard Matt., 14, 111-118 (1996).

[10] R. Haubner et al., “Behaviour of Co binder

phase during diamond deposition on WC-Co

substrate,” Diamond Rel. Mat., 2, 910-917

(1993).

(a) (b)

1µm 1µm

https://scholar.google.co.jp/citations?view_op=view_citation&hl=ja&user=_DMQoxkAAAAJ&citation_for_view=_DMQoxkAAAAJ:YsMSGLbcyi4C

https://scholar.google.co.jp/citations?view_op=view_citation&hl=ja&user=_DMQoxkAAAAJ&citation_for_view=_DMQoxkAAAAJ:YsMSGLbcyi4C

http://scitation.aip.org/content/contributor/AU0328754

http://scitation.aip.org/content/aip/journal/jap/81/1/10.1063/1.364000




http://www.sciencedirect.com/science/article/pii/S0257897213006087



http://www.sciencedirect.com/science/article/pii/026343689683424X

http://www.sciencedirect.com/science/article/pii/026343689683424X


25

ORTHOGONAL FLUXGATE GRADIOMETER SENSOR AND ITS

APPLICATION IN PARTICLE DETECTION

Ahmed Lotfy Elrefai1, Ichiro Sasada2

1 Applied Science for Electronics and Materials, kyushu University, Kasuga, Japan, [email protected] 2 Applied Science for Electronics and Materials, kyushu University, Kasuga, Japan

Abstract: A novel method for constructing an integrated gradiometer and magnetometer is proposed based on Fundamental

Mode Orthogonal Fluxgate (FM-OFG) operation. In the integrated sensor construction, the summation of both sensor head

outputs is used for the magnetometer operation and the subtraction of them is used for the gradiometer operation. The

baseline separation distance between the sensor heads is adjustable. The sensor heads pair can be configured either axially

or in parallel. Experiments were conducted with axial configuration of 5 cm baseline to show the selective detection

capability to gradient and homogeneous magnetic field. The sensing system presented can be used to measure the average

and the gradient of input magnetic field and its interesting applications are the gradiometric sensing of magnetic field

distribution anomalies in relatively large homogenous magnetic field interference.

1. Introduction

The gradiometer is a sensor to detect the gradient

of the magnetic field. There are two methods to

make the gradiometer. One method is electronic;

where two identical magnetometers are used and

the gradient is obtained by subtracting one

magnetometer’s output from the other’s. However,

an obvious drawback with the electronic

gradiometer is that two complete magnetometers

are needed and that a wide dynamic range with a

good linearity is needed for the high resolution

measurement. The other method uses a single

sensor head and is able to obtain the difference of

magnetic fields at two specific points. A common

structure of this type of gradiometer heads uses two

pickup coils wound on a single core [1]. The

system becomes compact and can be built with a

single driving and detection electronic set, however

the base line is fixed and making a good balance

between two pickup coils to suppress uniform

magnetic field is not easy.

So in order to build a gradiometer having

positive aspects of both methods, we propose a new

method for building a gradiometer based on the

fundamental mode orthogonal fluxgate (FM-OFG)

mechanism as shown in Fig. 1 [2]. The proposed

FM-OFG gradiometer is composed of two identical

sensor heads having hairpin-shaped amorphous

wire core with a pickup coil wound around it. The

pickup coils of the two heads are connected in

counter series; thus allowing the differentiation of

the signal on the sensor head level. The proposed

configuration allows the gradiometer to have the

advantage of a flexible baseline while minimizing

the used electronics set and also allowing more

amplification of the measured signal. The

developed gradiometer heads show high

suppression ratio of uniform magnetic field noise

and has high sensitivity to magnetic field gradients;

which demonstrates the sensor capability of being

implemented in various applications for

measurement of small magnetic field gradients.

One of the applications for the FM-OFG

gradiometer is the detection of magnetic particles

in unshielded environment.[3] The developed

prototype particle detection system offers size, cost

and weight reduction as compared to existing

particle detection systems. The system has been

investigated numerically to optimize various design

parameters of the system. Experimental setup has

been developed to evaluate some of the

numerically predicted results. Steel balls were

successfully detected down to the diameter of 50

µm.

Fig. 1. Schematic sensor circuit

2. Experimental Setup

An illustration of the sensor construction is shown

in Fig. 1. The gradiometer head is composed of two

FM-OFG heads of length 30 mm. Each sensor head

has amorphous wire core of 120 µm diameter. A

pick up coil wound on each core, where the pickup

coils of both heads are connected in counter series.

The excitation current is applied to the amorphous

wire core of both heads in series. Figure 2 shows

the fabricated gradiometer head. The excitation

current has 40 mA dc bias current and 100 kHz ac

current of 12 mA root-mean-square, which

establishes the fundamental mode operation.

Fig. 2. FM-OFG gradiometer with plastic

protection cover


26

The experimental setup of the magnetic particle

detection system is realized based on the

considerations taken from the results of the system

numerical analysis. The two heads of the

gradiometer are fixated in parallel configuration

with an 8 mm separation baseline, which is the

minimum baseline considering that each sensor

head is placed inside a plastic casing of radius 4

mm for protection of the pickup coil and the

amorphous wire core. The plastic casing of the

sensor head also limits the minimum lift-off to 5

mm with a 1 mm clearance between the moving

particle and the sensor case. Magnetic particle

samples are prepared using steel balls of diameters

160 μm, 120 μm, 70 μm and 50 μm in order to

evaluate the detection system performance. The

particles are pre-magnetized using a neodymium

magnet and then placed on the rotating table for the

gradiometer to measure the remnant magnetization.

For a particle to be detectable, the magnitude of the

output signal should be higher the sensor noise

level taken to be 0.5 nT/m at 1 Hz from Fig. 3. To

maximize the amplitude of the measured flux

gradient, the particles are placed on the rotating

table such that they pass at 2.5 mm to the inner side

from the tip of each sensor head. The output signal

of the sensor electronic circuit is amplified 400

times and passed on low pass filter of 20 Hz cutoff

frequency and High pass filter of 2 Hz cutoff

frequency.


The application of magnetic particle detection

using FM-OFG gradiometer is affected by various

design parameters. Hence, numerical analysis of

the system using finite element method (FEM) has

been conducted to determine the optimized values

to be used in the construction of the experimental

setup. In Fig. 3, the picked up magnetic field

gradient by the gradiometer is shown for the

variation of the particle position for x, y, and z

dipole moment directions. For the x direction

dipole moment, the maximum amplitude of the

measured signal is around a quarter of that of the y

and z direction dipole moments. Hence, to improve

detection capability of the sensor, it will be more

convenient to pre-magnetize the sample particles in

Fig. 3 Gradiometer output [T/m] vs. particle

position

y or z directions. At midway of particle movement,

i.e. the particle is situated at midpoint of the

baseline separation distance; the y direction dipole

achieves the maximum signal amplitude, while the

z dipole induces equal axial flux in the same

direction to both heads; which leads to omission of

the axial flux density gradient.

Experimental results from the detection system

prototype are shown in Fig. 4. The 160 μm and 120

μm particles show the y direction magnetization

dipole waveform pattern, while the 70 μm and 50

μm particles are showing the z direction one. At the

specified lift-off distance, i.e. 5 mm, the 50 μm

particle was detectable at a peak output value of 4

nT/m and a signal to noise ratio of 5, which is

comparable to recently reported results for a

superconducting quantum interference device

(SQUID) detection system with magnetic shielding,

where a 50 μm particle was detected with signal to

noise ratio of 10 at 3 mm lift-off distance.

Fig. 4. Gradiometer output [T/m] for steel balls of

diameters 160 µm, 120 μm, 70 μm and 50 μm

4. Conclusions

New magnetic particle detection system in

unshielded environment was proposed using FM-

OFG gradiometer in which differentiation is taken

at sensor head level. Numerical analysis was

conducted to demonstrate the system output and

determine the optimum values for the system

design parameters. The developed experimental

system successfully detected a 50 μm steel ball in

unshielded environment with signal to noise ratio

of 5. The proposed detection system has potential

of detecting smaller particles. In addition, the

developed particle detection system can also be

used for magnetic nanoparticles (MNP) detection

for liquid phase Immunoassay applications.

5. References [1] M. Janosek et al, “Single-core fluxgate gradiometer

with simultaneous gradient and homogeneous feedback

operation,” J. Appl. Phys., 111, 07E328 (2012).

[2] I. Sasada and S. Harada, “Fundamental mode

orthogonal fluxgate gradiometer,” IEEE Trans. On

Magn., 50, 4007404 (2014).

[3] A. L. Elrefai and I. Sasada, “Magnetic particle

detection in unshielded environment using orthogonal

fluxgate gradiometer,” J. Appl. Phys., 117, 17C114

(2015).


27

DRY ETCHING OF GERMANIUM WAVEGUIDES BY USING

CHF3 INDUCTIVELY COUPLED PLASMA

A. S. Idris*, H. Jiang, and K. Hamamoto

Interdisciplinary Graduate School of Engineering Sciences, Kyushu University, 6-1, Kasuga-Koen, Kasuga, Fukuoka

816-8580, Japan.

*E-mail: [email protected]

Abstract: A dry etching procedure to etch germanium in a CHF3 inductively coupled plasma (ICP) using a polymer based

photoresist mask was developed to obtain a high selectivity ratio as well as to obtain a near vertical anisotropic sidewall etch profile.

In this study, a sidewall angle of 85° with an etch rate of 190 nm/min was obtained through optimization of the ICP bias power to

fabricate germanium waveguide structures with no under-cut.

1. Introduction

The integration of optoelectronics and electronics that

are compatible with existing Si based complementary

metal-oxide-semiconductor (CMOS) technology is a

much highly desired objective. Numerous concepts

have been investigated such as in hybrid photonic

integration of III-V devices and monolithic Si photonic

integration. Although hybrid III-V devices integration

has made tremendous progress in recent years, there

are still difficulties integrating it into the standard

CMOS flow such as incompatibility in the fabrication

and bonding procedures and temperatures. Monolithic

Ge-based photonics has been touted as one of the most

promising options in order to realize active and passive

optical functionalities in Si CMOS-compatible

photonic circuitry [1]. Reports have been published

concerning the implementation of Ge-based devices

showing efficient optical modulation and photo

detection within the telecommunication wavelength

range and although despite being an indirect bandgap

semiconductor, the direct bandgap optical properties of

Ge have also been extensively reported for use in

active optical devices [1].

Majority of the Ge devices reported however, tend

to make use of a oxide or metal hard mask during the

fabrication process due to the faster etching rate of a

polymeric photoresist mask compared to Ge [2].

Removal of the hard mask can be done through plasma

processing. However, exposure to further plasma

processing may lead to additional damage to the

structures surface and sidewall and can lead to greater

loss [3]. In this study, a novel dry etching fabrication

process for Ge was developed and optimized using a

polymeric mask to produce a selective etching recipe

and near vertical sidewall.

2. Experiment

N-type Ge wafers with a (110) crystal orientation were

used throughout the development process. Firstly, the

wafers were cleaned in warm butanol followed by

warm isopropyl alcohol and blow dried in N2 gas. A

polymer based photoresist; 23CP by Tokyo Ohka

Kogyo Co. Ltd was used as the photoresist mask and

spin coated onto the Ge surface. The samples were then

baked on a hotplate at 90°C for 90 seconds. The

waveguide structures were defined by lithography

technique, developed using a photoresist developer and

post exposure baked again on a hot plate at 90°C for 60

seconds before being etched in the ICP for 60 seconds

with the same operating conditions (chamber pressure

of 1.5x10-4 Torr, RF bias power of 50 W, CHF3 flow

rate of 10 sccm), except for the ICP bias power which

was varied from 800 W to 1400 W. Finally, the

residual photoresist was removed using the photoresist

remover and rinsed in warm isopropanol.

3. Results and discussion

Two figures of merit were considered in the fabrication

of the Ge waveguide structure. The first figure of merit

is the selectivity of the etching ratio between Ge versus

the photoresist. This first figure of merit is interwoven

with the etch rate for Ge and photoresist. The second

figure of merit is that the sidewall angle should be

vertical with minimal under etching of the waveguide

structure. An overview of the etching process is given

in Figure 1.

Fig. 1. Schematic overview of the ICP process,

(a) photoresist on Ge (b) CF2 passivation layer

(c) etching process (d) final etch profile and

sidewall angle Ɵ definition.

There are two competing phases that occur during

the etch process. One is where a passivation layer is

deposited onto the photoresist and Ge surface and

another is where the photoresist and Ge material is

etched. CHF3 is reported to dissociate into a number of


28

different neutral and ionic radicals by electron impact

dissociation during the ICP process and as seen in

Table 1 [4].

Table 1. Electron impact dissociation of CHF3 and

corresponding threshold energies

Reaction Threshold

energy (eV)

CHF3 + e CF2 + HF (1) 2.43

CHF3 + e CF3 + H (2) 4.52

CHF3 + e CHF2 + F (3) 4.90

CF3 + e CF2 + F (4) 3.83

CF2 + e CF + F (5) 5.35

As seen in Table 1, the lowest threshold energy is

exhibited by the dissociation of CHF3 into CF2 gas

which is a polymeric precursor. Clustering of CF2 then

subsequently leads to its deposition onto the etching

surface as a passivation layer. The passivation layer

helps to protect the photoresist and enables greater

selectivity.

Using the optimized dry etching conditions, Ge is

etched at a rate of 190 nm/min while the photoresist is

etched at a rate of 35 nm/min giving a selectivity ratio

of 5:1 (Ge: photoresist).

In order to produce a vertical sidewall angle with

no under etching, both physical bombarding as well as

chemical reactions must be optimized during the

etching process. The main components for physical

etching are the F radicals produced by the further

dissociation of CHF3 which reacts with Ge to produce

GeF4 gas that can be pumped out from the ICP

chamber.

HF has been reported to readily etch Ge [5] and the

HF produced by the CHF3 dissociation provides the

chemical reaction component of the etching mechanism.

Fig. 2. SEM images of Ge fabricated with ICP bias

power and corresponding sidewall angle Ɵ

(a) 800 W (50°) (b) 1000 W (55°)

(c) 1200 W (85°) (d) 1400 W (70°).

The vertical angle of the sidewall can be optimized

by varying the ICP bias power as shown in Figure 2.

Fig. 3. Change in sidewall angle, Ɵ due to

applied ICP bias power.

The sidewall angle increases to near vertical when

applying an ICP bias power of 1200 W leading to an

optimized equilibrium between the physical and

chemical etching components as shown in Figure 3.

Below 1200 W of ICP bias power, the sidewall angle is

bevelled while at ICP bias power of 1400 W, under

etching is evident leading to narrowing of the

waveguide structure and a sloping sidewall.

The photoresist mask can then be easily removed

after etching using the photoresist remover and cleaned

in warm isopropanol.

4. Conclusion

Dry etching conditions to fabricate germanium

waveguide using a polymer photoresist mask has been

developed and optimized to produce a highly selective

etch and near vertical sidewall angle. The results

obtained may provide a simpler low damage dry

etching procedure for fabrication of germanium

waveguides and optical devices.

Acknowledgements

The authors would like to acknowledge Profs. D. Wang

and H. Nakashima of Kyushu University for their

useful discussion and advice.

References

[1] J. Michel, et al "An Electrically Pumped Ge-on-Si Laser",

OFC/NFOEC, PDP5A.6 (2012)

[2] A. Malik, et . al, "Germanium-on-Silicon Mid-Infrared

Arrayed Waveguide Grating Multiplexers", Photonics

Tech. Lett. IEEE, 25, 1805-1808 (2013)

[3] J. Cardenas, et al, "Low loss etchless silicon photonic

waveguides", Optics Express, 17, 4752-4757 (2009)

[4] M. Goto, et al "Cross Section Measurements for Electron-

Impact Dissociation of CHF3 into Neutral and Ionic

Radicals", Jpn. J. Appl. Phys., 33, 3602-3607 (1994).

[5] B. Schwartz, H. Robbins "Chemical Etching of

Germanium in Solutions of HF, HNO3, H2O, and

HC2H3O2", J. Electrochem. Soc., 111, 196-201 (1964).


29

SUPPRESSING NEW GRAPHENE NUCLEI GENERATION BY

HYDROGEN SWEEPING

Ding Dong1, Hiroki Ago2

1 Graduate School of Engineering Sciences, Kyushu University 2 Institute for Materials Chemistry and Engineering (IMCE) and Graduate School of Engineering Sciences,

Kyushu University, Fukuoka 816-8580, Japan

Abstract: New graphene nuclei continually appear during CVD process when using Cu as catalyst could be the main issue

to get large domain graphene. Here we reported a facile way to suppress new graphene nuclei generation by introducing H2

by interval time. Finally, we found new graphene nuclei could be effectively suppressed by this method.

1. Introduction

Recently, there are many reports about decreasing

the graphene density methods when using copper

catalyst, such as mild oxidation copper in Argon

gas [1-3], copper pre-treatment [4], liquid

Copper[5], removing carbon impurities on copper

[6, 7],Argon pulse[8] and “Vapour Trapping”

structure[9-11].All of above methods were

demonstrated to be useful to suppress the graphene

density.

Actually, in order to get millimetre, even centimetre

size single graphene domain, at least several hours

needed for CVD reaction. However, even people

could get much lower graphene density in the

beginning, we will show that, there is still high

possibility that new graphene domain will appear

as reaction time grows. The new appearance

graphene domains not only increase the graphene

density, but lower or even broken the possibility to

grow large graphene single domain, especially

when the new graphene domain appeared near the

previous existing graphene domain. Thus how to

suppress or avoiding new graphene nuclei

generation and still maintain graphene density as

low as beginning during the graphene growing is

extremely significate and crucial to get large

enough graphene single crystalline domain.

Unfortunately, this is usually neglected in most of

papers, which mainly focusing on the graphene

density on fixing time. There is paper noticed about

this but without any resolution [12].In this paper,

we first show the relationship between the

graphene density and reaction time. Then we

propose a facile method to suppress new graphene

nuclei appear when prolong the reaction time.

2. Main text

Copper mild oxidation in Argon gas (including

microscale Oxygen) [1-3] or directly heating in air

was verified very helpful to suppress graphene

nucleation density [8]. After oxidation and

reduction, the Cu atoms of active Cu crystal

boundaries rearrange into less active structure, and

thus the graphene nucleation density can be well

decreased after Cu surface oxidation [13, 14]. Thus

before our CVD experiment, we heat our copper at

200℃ to oxidize copper surface in air. After pre-

oxidation in air, we load Cu foil to our CVD

chamber to grow graphene. Before introducing

methane to CVD system, we anneal copper in 5%

hydrogen that diluted in Argon gas as long as 20

min to remove the copper oxide layer at 1070℃.

Then we pump in 2% hydrogen and 15 ppm

methane (all diluted in Argon gas) for graphene

growth at the same temperature. After reaction, we

oxidize the copper again at 200℃ for 30s in order

to visualize the graphene domain directly [15]

For growth times of 30min (Fig.1a), 60min (Fig.1b),

and 90min (Fig.1c), the maximum diagonal

distances for the graphene domains are∼556,

∼1056, and∼1600 um, respectively. This largely

linear time dependence indicates that the growth

velocity of the edge is constant and the increase in

a linear dimension of a growing crystallite is

simply proportional to the diffusion rate of the

active carbon which is constant at a fixed

temperature [16, 17]. That means the graphene

domain diameter is proportional to reaction time. If

we assume that the largest graphene domain

generates at the beginning we could get single

graphene domain occurrence time distribution. Fig

1(d-f) shows that there is always new graphene

domain generation during CVD reaction period. If

we look into Fig.1d, we could found that more than

half of the graphene nucleate in the first 10 min,

which means the main graphene nucleation process

prefer gathering at the beginning. Nevertheless,

new graphene domain still will appear continuously

in the following reaction time. That indicate even

people could get satisfied graphene density at

particular time, there is extremely high possibility

that new graphene domain will come out if needing

increasing reaction time to make graphene bigger.

The biggest disadvantage is the new occurrence

smaller graphene domain may ruin the previous

larger single crystalline graphene domain, making

it poly-crystallization.

The overall growth processes for Cu-based

graphene could be divided into three main steps

that described in Fig.2a step 1-3. (1) CH4

dissociates and is chemically adsorbed on the Cu

surface to form the active carbon species (CHx<4) s,

where “s” signifies “ surface-adsorbed” to


30

distinguish it from a gaseous molecule. The exact

nature of the active carbon species has not been

well-defined and we used carbon monomers (C) to

represent all types of active carbon species in

Figure 2a. (2) From recent research, the carbon Cu

interaction is weak and the desorption rate of active

carbon species is comparable to its mobility on the

Cu surface above 870 ℃ .Under the growth

conditions described in this paper, the temperature

is 1070 ℃, suggesting that the movement of active

carbon species on the Cu surface is dominated by

diffusion and desorption. (3) When active carbon

reaches a critical supersaturation point, graphene

domains nucleate. Once the graphene nuclei are

formed, most of the active carbon species will be

captured and consumed in the growth of graphene,

reducing the probability that new graphene nuclei

will be formed in the nearby area of the Cu catalyst.

Because graphene nuclei won’t takes place until the

concentration of active carbon species reaches to

supersaturation level [20, 21], we could put off or

even avoiding new nuclei generation if these far

away from the existing graphene domain active

carbon atoms could be swept away. Hydrogen had

demonstrated to a dual role during graphene

synthesis [22]: an activator of the surface bound

carbon that is necessary for monolayer growth and

an etching reagent that controls the size and

morphology of the graphene domains. Thus we

could utilize hydrogen to etch the free active

carbon atoms away (see Fig.2a step 4).After H2

“sweeping” for short time, CH4 gas will be re-

introduced and the previous graphene could regrow.

In order to verify above idea, we only shut off CH4

gas for 10 minutes every 30 min interval during

graphene synthesis section, only keeping Hydrogen

and Argon gas on. Fig.2 b-d shows the SEM

images when total reaction time is 60 min, 90 min,

and 120 min respectively (excluding hydrogen

sweeping time),and hydrogen sweeping time is one,

two and three respectively.Fig.2e-g are the

corresponding graphene domains occurrence

distribution. It is obviously that most of graphene

domains only generate at the first 30 min. This

means by H2 sweeping, new graphene nuclei is

successfully suppressed.

Fig.1

Fig.2

4. Conclusion

In this paper we utilize H2 to etching the

redundancy active carbon in Cu surface and

successfully supressed new graphene nuclei

generation. This method is easy to complete.

5. References [1] |Zhou H, Yu WJ, Liu L, Cheng R, Chen Y, Huang X, et al. Chemical vapour deposition growth

of large single crystals of monolayer and bilayer graphene. Nature communications.

2013;4:2096.

[2] Li J, Wang X-Y, Liu X-R, Jin Z, Wang D, Wan L-J. Facile growth of centimeter-sized single-

crystal graphene on copper foil at atmospheric pressure. J Mater Chem C. 2015;3(15):3530-5.

[3] Gan L, Luo Z. Turning off Hydrogen To Realize Seeded Growth of Subcentimeter Single-

Crystal Graphene Grains on Copper. Acs Nano. 2013;7(10):9480-8.

[4] ZhengYan, Jian Lin, Peng Z, Sun Z, YuZhu, Li L, et al. Toward the Synthesis of Wafer-Scale

Single-Crystal Graphene on Copper Foils. Acs Nano. 2012;6(10):9110–7.

[5] Wu YA, Fan Y, Speller S, Creeth GL, Sadowski JT, He K, et al. Large Single Crystals of

Graphene on Melted Copper Using Chemical Vapor Deposition. Acs Nano. 2012;6(6): 5010–7.

[6] Magnuson CW, Kong X, Ji H, Tan C, Li H, Piner R, et al. Copper oxide as a “self-cleaning” substrate for graphene growth. J Mater Res. 2014;29(03):403-9.

[7] Strudwick AJ, Weber NE, Schwab MG, Kettner M, Weitz RT, Wünsch JR, et al. Chemical

Vapor Deposition of High Quality Graphene Films from Carbon Dioxide Atmospheres. Acs

Nano. 2014;9(1):31-42.

[8] Eres G, Regmi M, Rouleau CM, Chen J, Ivanov IN, Puretzky AA, et al. Cooperative Island

Growth of Large-Area Single-Crystal Graphene on Copper Using Chemical Vapor Deposition.

ASC NANO. 2014;8(6):5657–69.

[9] Wang C, Chen W, Han C, Wang G, Tang B, Tang C, et al. Growth of millimeter-size single

crystal graphene on Cu foils by circumfluence chemical vapor deposition. Sci Rep. 2014;4:4537.

[10] Shi YG, Wang D, Zhang JC, Zhang P, Shi XF, Hao Y. Fabrication of single-crystal few-layer

graphene domains on copper by modified low-pressure chemical vapor deposition.

Crystengcomm. 2014;16(32):7558. [11] Zhang Y, Zhang L, Kim P, Ge M, Li Z, Zhou C. Vapor trapping growth of single-crystalline

graphene flowers: synthesis, morphology, and electronic properties. Nano Lett.

2012;12(6):2810-6.

[12] Wang Z-J, Weinberg G, Zhang Q, Lunkenbein T, Klein-Hoffmann A, Kurnatowska M, et al.

Direct Observation of Graphene Growth and Associated Copper Substrate Dynamics by in Situ

Scanning Electron Microscopy. Acs Nano. 2015;9(2):1506-19.

[13] Hao Y, Bharathi MS, Wang L, Liu Y, Chen H, Nie S, et al. The role of surface oxygen in the

growth of large single-crystal graphene on copper. Science. 2013;342(6159):720-3.

[14] Miley HA. Copper Oxide Films. J Am Chem Soc. 1937;59(12):2626-9.

[15] Wang H, Wang G, Bao P, Yang S, Zhu W, Xie X, et al. Controllable synthesis of submillimeter

single-crystal monolayer graphene domains on copper foils by suppressing nucleation. J Am Chem Soc. 2012;134(8):3627-30.

[16] Kidambi PR, Bayer BC, Blume R, Wang ZJ, Baehtz C, Weatherup RS, et al. Observing

graphene grow: catalyst-graphene interactions during scalable graphene growth on

polycrystalline copper. Nano Lett. 2013;13(10):4769-78.

[17] Mohsin A, Liu L, Liu P, Deng W, Ivanov IN, Li G, et al. Synthesis of Millimeter-Size Hexagon-

Shaped Graphene Single Crystals on Resolidified Copper. Acs Nano. 2013;7(10):8924-31.

[18] Guo W, Wu B, Li Y, Wang L, Chen J, Chen B, et al. Governing Rule for Dynamic Formation of

Grain Boundaries in Grown Graphene. Acs Nano. 2015.

[19] Nguyen VL, Lee YH. Towards Wafer-Scale Monocrystalline Graphene Growth and

Characterization. Small. 2015.

[20] Kim H, Mattevi C, Calvo MR, Oberg JC, Artiglia L, Agnoli S, et al. Activation Energy Paths for

Graphene Nucleation and Growth on Cu. Acs Nano. 2012;6(4):3614-23. [21] Zhang X, Li H, Ding F. Self-assembly of carbon atoms on transition metal surfaces--chemical

vapor deposition growth mechanism of graphene. Adv Mater. 2014;26(31):5488-95.

[22] Vlassiouk I, Regmi M, Fulvio P, Dai S, Datskos P, Eres G, et al. Role of Hydrogen in

Chemical Vapor Deposition Growth of Large Single-Crystal Graphene. Acs Nano. 2011;5(7):6069–76.

A HIGHLY ACTIVE Ni/ZSM-5 CATALYST FOR DEEP

HYDROGENATION OF ARENES


31

Shi-Chao Qi, Lu Zhang, Jun-ichiro Hayashi

Applied Science for Electronics and Materials, Kyushu University. [email protected]

Abstract: A highly dispersive supported nickel catalyst was in situ prepared by decomposing nickel tetracarbonyl onto ZSM-

5 zeolite, respectively. Unexpectedly, amorphous nickel was formed onto ZSM-5. Strong interaction between supported

nickel and the support is proved. Catalytic hydrogenations of arenes, i.e. durene, anthracene, phenanthrene and the heavy

mixture as by-products derived from methanol-to-gasoline conversion, over the catalysts were examined. Ni/ZSM-5 shows a

dramatic activity for deep hydrogenation of arenes and complete saturation of arenes can be achieved over the catalyst.

1. Introduction

Risk of petroleum shortage has been stimulating

development of coal to oil technologies, including

direct coal liquefaction (DCL) and indirect coal

liquefaction (IDCL), but the high concentration of

arenes in the liquids from DCL and IDCL confines

the application of them as clean fuels [1-3]. For

instance, methanol-to-gasoline conversion (MTG)

is an important IDCL process, but it normally

produces substantial amount of a heavy mixture

(HM), which mainly consists of unsaturated species,

especially polymethylbenzenes (PMBs). Deep

hydrogenation of such arenes, either PMBs or

condensed arenes (CAs), has thus been a major

technical subject [4,5]. None of the catalytic hydrogenation of arenes,

reported so far, has reached full hydrogenation of

CAs or PMBs. Conventional methods of catalyst

preparation, such as impregnation, co-precipitation,

sol-gel, and in situ reduction, were investigated

extensively. However, those methods are relatively

tedious and neither noble nor non-noble metallic

catalysts developed so far could completely

hydrogenate CAs and PMBs with high yields. A

key to preparation to a highly active catalyst is to

perform metals loading and reduction below a

threshold temperature for avoiding the decrease in

specific surface area of metals. Metal carbonyls

(MCs) usually decompose at relatively low

temperature into metals directly and are therefore

ideal precursors of highly active metal particles,

which can be used to catalyst preparation processes

much simpler than conventional ones. Supported

catalysts were prepared through the decomposition

of MCs in 1980s. However, preparation of

supported metallic catalysts using MCs was not

developed since most MCs volatilize rapidly long

before decomposition. In the present work, using

nickel tetracarbonyl (NTC) for nickel has

successfully developed to in situ prepare supported

nickel catalysts for deep hydrogenation of arenes.

2. Main text

NTC was synthesized by reacting active nickel

powder with CO under 8.0 MPa at 100 oC in a 100

mL stainless steel and magnetically stirred

autoclave. Diethyl ether, ZSM-5 zeolite, and NTC

were put into the autoclave. After replacing air

inside the autoclave with nitrogen, the mixture in

the autoclave was slowly stirred for 1 h at room

temperature to sufficiently impregnate NTC into

the support. Then the autoclave was heated to 100 oC and kept at the temperature for 1 h with rapid

agitation to allow in situ decomposition of NTC

over the support. After cooling the autoclave, CO

in the autoclave was released followed by

subsequent heating and cooling process mentioned

above to decompose NTC as exhaustively as

possible. Then the reaction mixture was taken out

from the autoclave and filtrated under nitrogen

protection to obtain the supported catalyst.

Catalysis procedures were carried out as

follows: THF (20 mL), catalyst (0.5 g), and

substrate (arene 5 mmol or HM 1.0 g) were fed into

the autoclave. After replacing air in the autoclave

and being pressurized with hydrogen to 5.0 MPa at

room temperature, the autoclave was heated up to

an indicated temperature within 20 min and kept at

the temperature for a prescribed period of time

followed by cooling the autoclave rapidly. The

reaction mixture was taken out from the autoclave

and filtrated. The filtrate was analyzed with gas

chromatograph/mass spectrometry (GC/MS).

As Fig. 1 displays, a number of spherical metal

particles with diameter less than 100 nm adhere to

the surface of three supports. In Fig. 2, nickel

species are supported onto ZSM-5. The nickel

supported onto ZSM-5 does not show as the shape

of crystals as the former two, but equably and

amorphously disperses onto ZSM-5, which results

from the restriction of the inner channels of ZSM-5

and the interaction between nickel and the support.

The SAED of Ni/ZSM-5 shows a diffraction

pattern of amorphous state as well.

In order to further confirm the catalytic activity

of Ni/ZSM-5 towards PMBs, durene was

hydrogenated over Ni/ZSM-5 at different

temperatures for 24 h to analyze the conversion of

Fig. 1. SEM and HREM image of the Ni/ZSM-5.


32

Fig. 2. HREM image of the Ni/ZSM-5.

durene (Fig. 3). The complete conversion is

observed during the temperature interval from 120 oC to 180 oC, and a conversion jump, from 15% to

96%, could be observed from 80 oC to 100 oC,

which indicates the activation temperature for

durene is relatively low due to the high activity of

Ni/ZSM-5. Nuclear magnetic resonance was used

to ascertain the molecular structure of the

hydrogenation product: 1H, δ= 0.838, 0.854; 2H,

δ= 1.272, 1.284, 1.333, 1.363 and 3H, δ= 1.598,

1.603, which indicate the protons of methyl,

methylene and methylidyne, respectively.

Corresponding with the ratio of three kinds of

protons in 1,2,4,5-tetramethylcyclohexane, the area

integrals of these peaks are respectively 3.08, 1.06

and 1.00.

60 80 100 120 140 160 1800

20

40

60

80

100

Reaction temperature (oC)

Co

nv

ers

ion

(%

)

Fig. 3. Reaction temperature profile of conversion

from catalytic hydrogenation of durene over

Ni/ZSM-5 and 1H NMR spectrum of hydrogenated

durene.

Anthracene and penanthrene were hydrogenated

over the catalyst for 24 h at 180 oC. The anthracene

perhydride almost accounts for the whole

compositions of products. Deep hydrogenation of

phenanthrene is difficult due to its steric

configuration. A high catalytic activity of Ni/ZSM-

5 is further certified by perhydrophenanthrene yield

of 25.8 mol%, and the yield of two-ring

hydrogenation products also reaches 63%. The

outstanding catalytic activity of Ni/ZSM-5 is

attributed to the following two reasons, the

amorphous state of nickel loaded inside ZSM-5 and

the strong interaction between nickel and the

support. The amorphous dispersion of nickel

greatly increases the opportunity of interaction

between substrates and the catalyst, meanwhile the

synergistic effect between nickel and acidic support

creates an electron-deficient surface, which

substantially increases the activity of metal-loaded

and promotes the π bond cracking of arenes.

As shown in Fig. 4, main components in HM are

arenes, especially PMBs. The PMBs are extremely

difficult to be saturated; meanwhile the blended

arenes would restrain mutual activity to be

hydrogenated due to the competitive adsorption

over active sites. The extremely high activity of

Ni/ZSM-5 for hydrogenation successfully

overcomes the difficulties. As Fig. 4 (HHM

denotes hydrogenated HM) exhibit, all the arenes in

HM were catalytically hydrogenated to saturated

species over Ni/ZSM-5 at 180 oC for 20 h.

6.5 7.0 7.5 8.0 8.5 9.0 9.5 10.0 10.5 11.0

12 3

56 7 9 11

1415

16 17 19

20

2223 26

29

30

31

3233 37

38 40 41 4243

0

17

34

51

68

85

0

20

40

60

80

100

46 47

48

50

11.5 12.0

61

12.5

67

13.0 13.5 14.0 14.5 15.0 15.5 16.0 16.5

5960

63

64

65

6870

71 72 75 77 78 80 81 83 86 89 91 93 94 95 9899

101103

104 106 109 11158 69 85

17.0 17.5

113 123

5653

Retention time (min)19.5 20.0 20.5 21.0 21.5 22.0

197

145

0

0.04

0.08

0.12

0.16

0.20

144148149

151

158

160

162163

166

167

170171

173

176179

181184

187

189 192193

195 198

141

18.0 18.5

22.5 23.0 23.5 24.0 24.519.0

19.5 20.0 20.5 21.0 21.5 22.0 22.5 23.0 23.5 24.0 24.519.00

16

32

48

54

80

0

20

40

60

80

100

HHM

HHM

HHM

12.5 13.0 13.5 14.0 14.5 15.0 15.5 16.0 16.5 17.0 17.5 18.0 18.5

6.5 7.0 7.5 8.0 8.5 9.0 9.5 10.0 10.5 11.0 11.5 12.0

HM

HM

HM

0

6

12

18

24

30

Rela

tive a

bu

nd

an

ce (

%)

14

5 8 10 12 13 15 16 18 21 24 25 27

28

30 31 3234

35

3536

36 37 38 39 424341

44

45 49 5051

52

52

54

55

57 62 63 64

66

67 68 73 74 75 76 77 798284 8788 90 92 96 97

100

102 105

105

107108 110 112

114

115

116

117

119

118120

120

121

122 124 127

125

126128

129

130

131

132

133

134135

136 137

138

139

140

142

143

146

147 150

152

153154

155

156

157

159

161

164

165168

169 172

174 175177

178

180

182

183185186 188

190191194

196

199200

201

202

203

204

205

206

207

208

209211

210212

213

215

214

218216

217219

220221

222

223 226

224225

227

228229

230

231234

231233

232

236

235

237238

240239

241

243242

244

Fig. 4. Total chromatograms of the HM and HHM

(the numbers in red and blue color denote arenes

and alkanes respectively).

3. Conclusion

In summary, with adopting nickel tetracarbonyl as

precursor, a highly dispersive and dramatically

active nickel-loaded catalyst Ni/ZSM-5 for the

deep hydrogenation of arenes could be in situ

prepared. The nickel supported onto ZSM-5 does

not show as the shape of crystal, but equably and

amorphously disperses onto ZSM-5. The catalyst

distinctly facilitate the complete hydrogenation of

condensed arenes and polymethylbenzenes. Heavy

by-products derived from MTG are completely

saturated as well.

4. References [1] Longwell, J. P.; Rubin, E. S.; Wilson, J., Coal:

Energy for the future. Prog. Energy Combust. Sci.

1995, 21, 269-360.

[2] Cayan, F. N.; Zhi, M.; Pakalapati, S. R.; Celik, I.; Wu,

N. Q.; Gemmen, R., Effects of coal syngas impurities

on anodes of solid oxide fuel cells. J. Power Sources

2008, 185, 595-602.

[3] Breysse, M.; Djega, M. G.; Pessayre, S., Deep

desulfurization: reactions, catalysts and technological

challenges. Catal. Today 2003, 4, 129-138.

[4] Schulz, H.; Böhringer, W.; Waller, P.; Ousmanov. F.,

Gas oil deep hydrodesulfurization: refractory

compounds and retarded kinetics. Catal. Today 1999,

49, 87-97.

[5] Tsoskounoglou, M.; Ayerides, G.; Tritopoulou, E.,

The end of cheap oil: current status and prospects.

Energy Policy 1990, 88, 109.


33

THE ACCELERATED SOLVENT EXTRACTION OF

XINYU COKING COAL AND ITS EXTRACTION MECHANISM

Lu Zhang, Shi-Chao Qi, Koyo Norinaga

Applied Science for Electronics and Materials, Kyushu University. [email protected]

Abstract: Coking coal from Xinyu of Shanxi Province is extracted under elevated temperature and pressure via Accelerated

Solvent and Soxhlet Extraction. Analyzing their GC/MS results, we explore the two extraction methods’ impact on the

dissolution behavior of small molecules in coal and investigate the mechanism of the extraction.

1. Introduction

Destroying the hydrogen bonds amid coal

molecules, VDW and weak complexing forces,

solvent extraction frees the solvable molecues in

coal [1]. Traditional methods of extraction under

ordinary pressure were time-wasted, solvents-

exhausted and tedious but with bad efficiency. The

Accelerated Solvent Extraction improves the speed

and effeciency of the extraction, and decreases the

solvent cost, meanwhile separates the solid from

liquid phase because the reduced surface tension of

solvent, enhanced infiltration capacity and

accelerated equilibrium process are generated from

solvent heated and forced. In this article, coking

coal from Xinyu of Shanxi Province is extracted

under elevated temperature and pressure via

Accelerated Solvent and Soxhlet Extraction.

Analyzing their GC/MS results, we explore the two

extraction methods’ impact on the dissolution

behavior of small molecules in coal and investigate

the mechanism of the extraction.

2. Main text

Coal sample is coking coal after taken off the

minerals , comes from Xinyu Preparation Plant of

Shanxi Province. It was then pulverized to around

150 mesh. TABLE I shows the proximate and

ultimate analyses of the coal sample.

TABLE I THE PROXIMATE, ULTIMATE ANALYSIS OF COAL

SAMPLE

Proximate analysis/ w%

Mad Ad Vdaf FCdaf

0.69 10.13 22.18 77.82

Ultimate analysis/ w%

Cdaf Hdaf Odaf Ndaf St,d

89.08 4.63 1.99 1.72 2.30

Weigh air dried basis coal sample about 2g. Put

it in the soxhlet extractor after parceled with filter

paper. Use 1-propanol solvent to extract it 90

minutes at the boiling point temperature. The

extracts were refined by rotary evaporation and

analyzed by GC/MS. The residues were washed,

suction filtered, dried in turn and analyzed by FTIR.

Accelerated solvent extraction: Packed by 450 nm

PTFE and filter paper respectively, the samples (2

g) were extracted in a 22 mL extraction pool under

the pressure of 80 bar and 100 oC. The extraction

was cycled for 2 times after static extraction for 15

min. With referring to the extract, the extraction

rates (daf, w%) are calculated as follows: 3.85% for

accelerated solvent extraction and 3.41% for

Soxhlet extraction, respectively.

As shown in Fig. 1 and 2, there are 37 compound

categories detected in the soluble substances via

accelerated solvent extraction, including 17 alkanes,

13 aromatics, 3 oxy-compounds and elemental

surfur. There are also 37 compound categories

detected in the soluble substances via Soxhlet

extraction, including 14 alkanes, 22 compounds

containing oxygen or nitrogen, 1 thioether, and no

aromatics and sulfides detected in it.

2.1 Alkanes

There are 17 alkanes from C11 to C27 in the solub

le substances via accelerated solvent extraction and

14 alkanes from C20 to C33 in the soluble substance

s via Soxhlet extraction distributing continuously.

Despite similar number of compounds, the two solu

ble substances belong to two different series obviou

sly due to the different numbers of carbon atoms co

ntaining in the compounds. The former belongs to a

lkanes of low carbon and the second one is defined

as high carbon. As one kind of small molecules in

coal, n-alkanes exist as three states, free state,

micropores and network embedded states [2-4]. As

for micropores embedded states, mid- and high-

carbon n-alkanes usually embedded into relative

large micropores due to their big sizes, and low-

carbon n-alkanes usually embedded into relative

small micropores. Based on their own permeability,

solvents are easy to enter the relative large

micropores under ordinary pressure of Soxhlet

extraction, whereas the solvents also enter the small

micopores under the pressure above 80 bar. The

diffusing impetus of soluble substances comes from

the concentration gradient, which is provided by

the circulation flow of solvent in Soxhlet extraction.

As a result, accelerated solvent extraction only

improves the rate of solvent penetration, but does

not benefit the diffusing rate of soluble substances.

2.2 Aromatics

There are 13 aromatics, including derivatives of

benzene, naphthalene and anthracene, in the soluble

substances via accelerated solvent extraction,

whereas no aromatics are detected in the soluble


34

substances via Soxhlet extraction. Aromatics are

always embedded in the relative small micropores

[5], thus such derivatives are not dissolved via

Soxhlet extraction. The mechanism of dissolving

benzothiophene series is similar with mentioned

above.

2.3 Oxy- and nitrogen-compounds

Three oxy-compounds, including only one ester,

are detected in the soluble substances via

accelerated solvent extraction. However, there are

22 oxy- or nitrogen-compounds, including 6 acids

and 9 esters, detected after Soxhlet extraction. Most

of them are long chain aliphatic acids or esters,

which are obviously embedded in the relative large

micropores.

According to the FTIR spectra in Fig. 3, there is

no obvious difference about residues' structure

characteristics of functional group of two kinds

extraction. The reason is that both the extraction

rate are low, which is not enough to affect the

functional group distribution of the residues.

10. 00 15. 00 20. 00 25. 00 30. 00 35. 000

100

20

40

60

80

Ret ent i on Ti me/ mi n

Relative Abundence/%

Fig. 1. TIC of soluble substances of Accelerated

Solvent Extraction.

10. 00 15. 00 20. 00 25. 00 30. 00 35. 000

100

20

40

60

80

Ret ent i on Ti me/ mi n

Relative Abundence/%

Fig. 2. TIC of soluble substances of Soxhlet

Extraction.

1000 4000 2000 3000

1

2

Fig. 3 FTIR spectra of residues.

3. Conclusion

The extractants from ASE mainly contain materials

of molecular weight from 142 to 296, i.e. mid- or

high-carbon n-alkanes, aromatics and thiophene

series, while the extractants from SE contain

materials of molecular weight from 228 to 426, i.e.

mid- or high-carbon n-alkanes, and aliphatic

hydrocarbons (or esters) of long carbon chains.

Because of the limited time, the solvent in SE only

enters the relative large micropores. However, the

solvent in ASE can enter both relative large and

small micropores simultaneously. The advantage of

ASE is to improve the permeability and permeation

rate of solvents, but it makes little contribution to

diffusion rate of solvents. The intrinsic reasons for

the different results of the two extraction methods

are the different molecular weights of the materials

embedded in the two kinds of micropores.

4. References [1] LARSEN J W, MOHAMMAD I M, Energy Fuel,

1990, 4: 107

[2] Qin Z H, Wei X Y, Jiang C, J. China Univers. Mining

Technol., 2005, 34: 579

[3] Qin Z H, Jiang C, Sun Hao, J. China Univers. Mining

Technol., 2005, 34: 707

[4] Qin Z H, Gong T, Li X S, J. China Univers. Mining

Technol., 2008, 37: 443

[5] Qin Z H, Hou C L, Zhang D, J. China Univers.

Mining Technol., 2007, 36: 586

ieices intellectual exchange and innovation conference on ... · you to the intellectual exchange...

Documents