Studies of milling technology's approaches for establishing

the chemical recycles on some wasted glasses.

○Masataka KAMITANI1)2), Mitsunori KONDO1), Atsushi NAKAHIRA2)

1)Makino corporation (Aichi Tokoname,Japan)

2)Graduated School of Osaka Prefecture University.(Osaka Sakai,Japan)

Japan-Korea powder technology symposium 2015

Hold on Soul National University Sept.14,2015

1. The Objects and Targets of This study.

２．Some directions for wasted glass chemical recycle applications

①Silicate resources for minerals and ceramics synthesis

②Ion exchanging devices for environmental cleanup

③The dehydration condensation for like cement

Synthesized LTA in Bamboo inside pore

200nm

Condensed mass from LCDG

200μm

Abstracts of Spring Meeting of Japan Society of Powder and Powder Metallurgy,2013

Abstracts of Autumn Meeting of Japan Society of Powder and Powder Metallurgy,2013

IREP the first Meeting in Imabari,2013

2nd meeting on Environmental radioactive decontamination technology, Tokyo 2013

Materials Science Forum Vols,22-227 (1996) pp.587-592

Kinzoku,Vol.68 (1998) No.9

LTA

M.kamitani,A,Nakahira,T.Wakihara､et.al ISAC-5,Wuhan,Chaina 2014

M.Kamitani,M.Kondo.A.Nakahira, J. Jpn. Soc. Powder Powder Metallurgy Vol. 62, No. 6 ,2015

Fabrication and evaluation of hybrid materials from A-zeolite and ground glass powders for vitrified

radioactive, j. ceram. soc, japan 122 (2) 151-155,2014

M.kamitani,A,Nakahira,T.Wakihara､et.al ISAC-5,Wuhan,Chaina 2014

Abstracts of Autum Meeting of Japan Society of Powder and Powder Metallurgy,2014

3. Fundamental procedure of chemically activation for wasted glass by Ball milling

Energy consumption of milling with ball media

compressive impact shear

≧

Fracture mode of glass grain by impact

Rumpf, H, Struktur der Zerkleinerungswissenschaft

Aufbereitungs-Technik No. 8/1966 421-435

Fig.1 model of generating forces in ball mill

ref.

Fig.2 fracture model from reference

Optimization of ball milling condition by DEM simulation

j-th ball

i-th ball

usus

s

Ks

slider

i-th ballunun

n Kn

j-th ball

Normal force Shear force

j-th ball

i-th ball

usus

s

Ks

slider

i-th ballunun

n Kn

j-th ball

Normal force Shear force

Ball weight

Impact speed

Specimen weight

Number of impact

Fig.3 DEM simulation based on Voigt model

Mill dimension

L=725mm diameter=725mm

Ball: φ＝10-30mm steal

0.75 0.80 0.85 0.90 0.95

2500

3000

3500

4000

4500

5000

Impa

ct

energ

y (J

/s)

Relative rotational speed, N/Nc (-)

J=40 J=50 J=60

dB=10 mm

0.75 0.80 0.85 0.90 0.95

2500

3000

3500

4000

4500

5000

Impact

energ

y (

J/s)

Relative rotational speed, N/Nc (-)

J=40 J=50 J=60

dB=15 mm

0.75 0.80 0.85 0.90 0.95

2500

3000

3500

4000

4500

5000

Impact

energ

y (

J/s)

Relative rotational speed, N/Nc (-)

J=40 J=50 J=60

dB=30 mm

Fig.4 Impact energy estimation by

DEM simulation.

This point

Energy consumption

Mill dimension

L=725mm diameter=725mm

Ball filling ratio=0.5

Ball: φ＝15mm steal

Raw material:24-32Kg

8

①

②

③

Glass weight

（kg）

Ball milling condition

rotation Ball radius occupy

sampling（ｈｒ）

0.5 1 2 3 6

0.95

0.95

32

32

24

0.85

15

15

15

50

50

50

（N/Nc) (mm） （％）

○

○

○ ○ ○ ○

○ ○ ○ ー ー

ー ○ ○ ー

Ｎｏ． Glass species

LCDG1)

④

⑤

SLG2) 0.95 20

15

30

30 0.95

50

50

○ ○ ○ ○ ○

○ ○ ○ ○ ○

ー

ー

ー

○

○

Ｔａｂｌｅ.１ Test plan

1）①②③：LCDG alumino-silicate glass：ＳｉＯ２/Ａｌ２Ｏ３/ＣａＯ/Ｎａ２Ｏ＝58.8/17.1/9.5/0.3(ｗｔ％ratio）

2）④⑤：Soda-Lime Glass：ＳｉＯ２/ＣａＯ/Ｎａ２Ｏ＝73/5/17（ｗｔ％ratio）

4.Experiment and Results

⑥3) 12 0.75 15

50 ー ○ ○ ○ ○ ○

3) Test ⑥ is another experiment for in-line measuring system development.

Milling by small 50L mill on another glass:ＳｉＯ２/Ａｌ２Ｏ３/MgＯ/Ｎａ２Ｏ＝60.2/15.2/7.2/16.8(ｗｔ％）

Al2O3

①Milling test plan

2hr

3hr

2hr

4hr

5hr 6hr

6hr

Fre

qu

en

cy

Test② Test②

Fig. 5 Grain size distribution of test2 and test5 by Laser diffraction MT-3000Ⅱ

②Grain size distribution change

6hr

2hr

6hr 4hr

6hr

2hr 6hr

5hr

T-2

T-5

(μm) (μm)

(μm) (μm)

Fre

qu

nc

y (

vo

l.ba

sis)

Fre

qu

nc

y (

vo

l.ba

sis)

Cu

mu

lativ

e(%

) C

um

ula

tiv

e(%

)

5hr

3hr

4hr

③Morphology on SEM’s images

Fig.8 SEM’s images of T-2 and T-5 samples that 6hr milled.

Raw material

T-2

T-5 T-5

0

0.5

1

1.5

2

2.5

3

3.5

4

4.5

0 1 2 3 4 5 6 7 8 9

Sp

ec

ific

su

rfa

ce

are

a (

m2/g

)

Milling time (hr)

LCDG(T-2)

SLG(T-5)

④Specific surface area

Fig.9 BET specific surface area of T-2 and T-5 measured by Bellsorp mini Ⅱ.

0

0.2

0.4

0.6

0.8

1

0 0.1 0.2 0.3 0.4 0.5 0.60

0.2

0.4

0.6

0.8

1

0 0.1 0.2 0.3 0.4 0.5 0.6Atomic distance (nm)

Exis

ten

ce

ra

tio

Atomic distance (nm)

⑤XAFS analysis

T-2:LCDG T-5:SLG

Fig.10 XAFS profiles of T-2 and T-5

6hr

raw 2hr

6hr

raw

2hr

The monochromator serves soft X-rays in the

energy region from 585 to 4000 eV

KTP (KTiOPO4)

Si-K

Soft X-Ray Beamline for Photoabsorption Spectroscop

Thanks UVSOR

6hr

Fig.11 Titration curves of T-2 and T-5 samples. Solid concentration is 10% １N-NaOH used

⑥Titration for surface analysis

Measured by DT-1200

1N-

NaOH

blank

1.54 1.25

BET 2.8m2/g

BET 2.1m2/g

BET

1.8m2/g

2hr

4hr

6hr

blank

4.93

T-2

T-5

6hr

5hr

2hr

3hr

Zero-charge point

4hr

3hr 6hr 12hr

Test2-6hrmilling reacted in S-solution for 3-12hr at 95℃

Fig.12 SEM’s images of autoclaved grains

5. For Silica resource

LTA

0

10000

20000

30000

40000

50000

60000

0 10 20 30 40 50

0

5000

10000

15000

20000

25000

0 10 20 30 40 50

0

5000

10000

15000

20000

25000

30000

0 10 20 30 40 500

5000

10000

15000

20000

25000

30000

35000

40000

45000

0 10 20 30 40 50

T-2-3hr milling

T-2-6hr milling

3hr reaction 6hr reaction 12hr reaction

6hr reaction 12hr reaction

Fig.13 XRD profiles of autoclaved sample

0

10000

20000

30000

40000

50000

60000

70000

80000

0 10 20 30 40 50

2θ(degree) 2θ(degree)

2θ(degree)

2θ(degree)

2θ(degree)

Inte

nsi

ty (

cp

s)

Inte

nsi

ty (

cp

s)

Inte

nsi

ty (

cp

s)

Inte

nsi

ty (

cp

s)

Inte

nsi

ty (

cp

s)

0

0.015

0.03

0.045

1 10 100 1000

BJH-Plot

dp/nm

dVp/

dlogd

p

①aqueous washing treatment

T-2 6hr milled

Washed by water

Fig.14 BJH profiles of T-5-6hr milling samples that washed by pure water.

6. For Ion exchange device

BET=2.8m2/g

BET=9.6m2/g

NaI detector (NHC-7 fuji electric)

Sample in U8 cantor Pb shielding brock

②Decomtanimation of radio active material;137Cs

●Testing solution

Fig.15 photos and schematic view of radioactive Cs removing test.

Wash out solution from sewage ashes in Fukushima pref.

134Cs: 213±7.5 (Bq/kg)

137Cs: 383±14

Ge detector model GC-3020,canberra

(Bq/kg)

Concentrate to

around 137Cs:1200Bq/kg

material w/s shaking

time(min)

137Cs initial (Bq/kg)

137Cs remain

(Bq/Kg) removing(%)

washed milling Glass

powder T-2-6hr&washed

10 10 1262 902 18.9

fine Mordenite

10

10

1262

382.9 69.6

coarse Mordenite 10 786 37.7

A-type Zeolite 10 1219 3.3

Table.3 Test result of removing 137Cs

6hr milled LCD glass(30g)

10N NaOH 30ml

RT one night

Hardened sample

Casting

Mixing

7. For Dehydration Condensation

① Hardening procedure

Fig.16 SEM’s images of hardens samples

②Morphology

0

10

20

30

40

50

60

70

80

90

100

05001000150020002500300035004000

Tra

nsm

itta

nc

e(%

)

wavenumber(cm-1)

O-H Stretching

Si-O Stretching

Si-O Rocking

Si-O Bending

H-O-H Bending

6hr

milled

hardened

Fig.18 FT-IR spectrums of samples

④ Chemical bonding

8. Inline measuring for ball milling

-0.20

0.00

0.20

0.40

0.60

0.80

0

0.5

1

1.5

2

2.5

0 5 10

dS/d

t

BET

(m2/g

)

-30

-20

-10

0

0 5 10

Ze

ta p

ote

ntia

l

(mV

)

-10

-5

0

5

10

30

40

50

60

70

0 5 10

Si c

on

ce

ntr

ate

ra

te

(mg

/L・

hr)

Si c

on

ce

ntr

atio

n

(pp

m)

Flocks and

agglomerates

Agglomerates

scrap and build

Fine milling

1st St. (Start～4hr)

2nd St.(4～8hr)

Agglomerates

crush and formed

Particles down sizing

and agglomerates

formation

①The model of glass milling from T-⑥

②Specific surface area measurement by NMR

Rav:reciprocal of sample relaxation time

Rav=ѱ𝑝𝑆𝐿𝜌𝑝 𝑅𝑠 − 𝑅𝑏 +Rb

=KASѱp＋Rb

Ѱp：volume fraction of particles

S：surface area

Rs:reciprocal of adsorption relax.time

Rｂ:reciprocal of solvent

KA: coefficient of wettability

=R

avｰR

b

𝑅𝑏

Rsp

Pulse NMR

0

500

1000

0 1000 2000 3000

Ma

gn

etiza

tio

n

Time [milliseconds]

0.3T

T2 mode:Spin-Spin relaxation

A short RF Pulse B1

Rotates H atoms

N

S

N S

N S

N

S

N

S

N

S

-> Bo

->

Bo

B1 Temporary

magnetic field

When B1 disappears, H atoms realign with Bo

producing a signal

③Comparison with BET and R2sp

0

0.5

1

1.5

2

2.5

0 2 4 6 8 10

BE

T s

pec

ific

su

rfac

e (m

2/g

)

Milling time(hr)

0

0.5

1

1.5

2

2.5

3

3.5

4

4.5

5

0 2 4 6 8 10

Ca

lcu

late

d s

pe

cific

su

rfa

ce

are

a f

rom

R2

sp (

m2/g

)

Milling time (hr)

0

20

40

60

80

100

120

140

160

180

200

0 500 1000 1500 2000

Relaxation time (ms)

Fig.20 The relaxation time curves on T-⑥

Fig.21 Comparison with cal.surface area and BET

1hr

2hr

3hr

4hr 6hr

9. Summery of this presentation

１）We’ve confirmed structural changes of wasted glass materials by ball milling.

２）By ball milling , the higher impact energy makes disorder on grain surface or inner

structure, We consider they play as dissolving points, and/or to be meso scale

cavities on grain surface.

It is very important knowledge for the chemical recycling technology.

４）We would develop the operating methods for environmental cleanup with

these devices

3) We’ve recognized these phenomenon induced by “Mechanochemical effect”

5）We’ve already developed the optimized milling systems on inline surface area

that mainly consists with P-NMR. measuring