核融合炉固体増殖ブランケットからのトリチウム回 収に関する研究

九州大学学術情報リポジトリ

Kyushu University Institutional Repository

核融合炉固体増殖ブランケットからのトリチウム回 収に関する研究

河村, 繕範

九州大学工学エネルギー量子応用原子核

https://doi.org/10.11501/3075442

出版情報：Kyushu University, 1993, 博士（工学）, 課程博士 バージョン：

権利関係：

4. Study on Mass Transfer Rate of Water on Lithium Ceramics

4.1 Introduction

The diffusivity of tritium in the crystal grain of lithium ceramic breeder materials has been investigated by measurement of tritiated water release rate from the crystal grain[l-5]. Understanding of the process of diffusion of tritium in the crystal grain is important to understand the release behavior of tritium bred in the blanket. However, the release process of tritium bred in the blanket is complex and difficult to understand, because it is not only constituted with the process of diffusion in the grain but also that of diffusion in pore and that of surface reaction.

At present, it is impossible to estimate the release behavior of tritium bred in the blanket because of shortage of data about tritium diffusion in pore and surface reaction and others. Therefore, the release behavior of tritium bred in the blanket are studied with the in-situ experiments based on the idea that the investigation of release behavior of tritium from the pellet under the condition of blanket for use is more realistic. Almost results of the in-situ experiments are analyzed assuming that the release process of tritium bred in the blanket is controlled with the process of tritium diffusion in the crystal grain. However, it is also pointed out from some results that the effect of surface reaction can not be ignored [ 6-18].

Present author has quantified the amount of water captured In lithium ceramic breeder material and compared the tritium inventory by sorption with that by diffusion at the chapter 3 of this study. Tritium inventory by sorption is much larger than that by diffusion in the crystal grain. Accordingly, it can not be considered that the process of tritium diffusion in crystal grain controls the release process of tritium bred in the blanket. And then, it have been confirmed that the tritium release rate was enhanced when hydrogen was added to the purge gas at the ln·situ experiments[6,13,15,18]. These results suggest that tritium release process is Possibly controlled by the process of surface reaction and that of pore diffusion.

- 127 -

As mentioned in the opening of this section, the diffusivity of tritium in the crystal grain have been obtained by measurement of release rate of tritium in the form of tritiated water. Therefore, it can be considered that values of diffusivity obtained from the release experiments also include the resistance due to surface reaction.

Accordingly, the adsorption and desorption rate of water at the surface of various breeder materials are measured in this work, and the effective diffusivity of water in the particle of breeder material is estimated using data obtained as adsorption and desorption rate and is compared with the so to speak tritium diffusivity in the crystal grain.

4.2 Theoretical consideration

The mass transfer process of adsorption is constituted with the process of diffusion of adsorbate in the fluid film, that of diffusion in pore of adsorbent and that of adsorption on the surface of adsorbent[19,20]. When it is assumed that the solid film exists at the surface of particle and the difference of concentration of adsorbate at both end of film acts as the driving force of adsorption, the rate of adsorption per unit volume of packed bed is expressed as

(4-1)

Y~~ =

kjO,v ( C - C; )

=

ksav ( q; - q ) ' where y : density of packed bed [g/cm3]

q : the amount of adsorption [mol/g]

t : time [sec]

k1 : mass transfer coefficient of fluid film [em/sec]

~ : surface area of particle per unit volume of packed bed [cm²/cm^3]

c : concentration of adsorbate in gas phase [mol/cm3]

ci : concentration of adsorbate at the surface of particle [mol/cm3]

- 128-

ks : mass transfer coefficient of solid film [g/cm²•sec]

qi : the amount of adsorption at the surface of particle [moVg].

The surface area of particle per unit volume of packed bed is approximately expressed as

- 6(1 - £) av - ---'---'-

dp

(4-2)

where ^£ : void fraction of packed bed [-]

dp : diameter of particle [em].

And then, the product with mass transfer coefficient (kf or k s) and the surface area of particle per unit volume of packed bed (av) is called the mass transfer capacity coefficient (kflv or k sav). However, the concentration of adsorbate at the surface of

particle (ci) and the amount of adsorption at the surface of particle (qi) are impossible to quantify. And then, when the amount of adsorption equilibrated to the concentration of adsorbate in gas phase (q*) and the equilibrium concentration of adsorbate given by the amount of adsorption (c*) are introduced, eq.(4-l) is changed as

yaq

=

KJGv ( c- c* )

=

Ksav ( q* - q) dt

where K f : over-all mass transfer coefficient of fluid film [em/sec]

(4-3)

Ks :over-all mass transfer coefficient of solid film [g/cm²•sec].

The over-all mass transfer coefficient is expressed as j_= _l_+ _1_

Kj kt {3 ks _1 =f3+_L Ks kr ks ,

Where

f3

means the adsorption coefficient and expressed as f3=q

c.

- 129 -

(4-4)

(4-5)

If adsorption progresses following Henry's low, the adsorption coefficient {3 is to be constant.

The mass transfer capacity coefficient of solid film is also expressed as

(4-6) k _ 60D;y

sClv - _ __:__.<_

dp2

where D 't : the effective diffusivity of adsorbate in the particle at the standard of the amount of adsorption [cm²/sec].

The correlative equations of mass transfer coefficient of fluid film have been proposed by many authors. Therefore, if Kflv or K₅av is obtained by curve fitting of breakthrough curve or release curve obtained from experiments, it is possible to estimate k

5av. And then, D 'i is estimated using eq.(4-6).

4.3 Experimentals

4.3.1 Sample

The samples used in this work were LiA10₂, Li₄Si0 _{4 ,} Li₂Zr0₃ made in Mitsubishi Atomic Power Industries Inc. and their specifications are shown in tables 4.1, 4.2 and 4.3, respectively. The chemical composition of LiA10₂ used in this work was analyzed with the ICP method and is shown in table 4.4. To apply the ICP method for LiA10_2, the solid sample is to be dissolved into aqua regia and it takes about several days to dissolve crystal powders of LiA10₂. In case of analysis for the sintered pellet, of which the result is shown in table 4.4, the ICP analysis is made for solution when about 90% of the sintered pellet is dissolved because the main purpose is to check the amount of impurities. The activation analysis using Kyoto University Reactor(KUR) gives similar results to the ICP analysis.

The result of spectrographic analysis of Li₂Zr0₃ used in this work is also

- 130-

shown in table 4.5.

4.3

.2

Experimental method

Fig.4.1 shows a schematic diagram of experimental apparatus. Sample of breeder material was charged in a reaction tube. The reaction tube was a quartz tube of 22.8mm in inner diameter and 500mm in length, in which a filter plate with several lmm holes was mounted at 220mm from the bottom end.

Helium was used as the carrier gas and it's flow rate was kept at 0.2-0.4 l/min by a flow rate adjuster. The water concentration in He was measured with a hygrometer continuously. The hygrometer used in this work was a moisture analyzer type MAH-50 from SHIMADZU Co. A cold trap was attached at the dry gas channel of the hygrometer that was packed with molecular sieves and kept at liquid nitrogen temperature. The calibration of sensitivity of the hygrometer was carried out before and after each experiment by introducing H₂0/He gas of which H₂0 concentration was known, and it was confirmed that the span was not changed during the experiment. The water concentration in He purge gas at the inlet of sample packed bed was controlled by a hydrogen oxidizing method with a spongy CuO bed which was heated to 350°C. The hydrogen was confirmed to be completely oxidized because gas chromatography detected no hydrogen in the outlet gas of the CuO bed. The pressure gradients along the packed bed and that bet~een

the bed and the exit end were determined to be negligible by a pressure gauge, so the pressure of the gas flowing over the experimental sample could be taken as 1 atm with negligible error.

The measurement was carried out as follows.

(1) The sample was heated up very slowly to 1073K for more than several hours Under moisture-free helium gas flow to desorb residual water before each experimental measurement. The moisture in the helium gas from a gas cylinder Was removed by a cold trap, that is a packed column of molecular sieves 5A at the

- 131 -

liquid nitrogen temperature, and the moisture content in the outlet gas of the cold trap was determined to be lower than 0.3ppm by a hygrometer.

(2) Then helium gas containing water vapor with a certain partial pressure was passed through the experimental apparatus without the sample bed until the concentration of water in the inlet gas of hygrometer was equal to that in the outlet gas of CuO bed because of elimination of system effect, that is caused by adsorption and desorption of water to the surface of piping materials, from the breakthrough curve. And then, H₂0/He gas was passed through the sample bed after setting it to the experimental temperature, and change of the water vapor concentration in the outlet gas was traced with time by the hygrometer. Helium gas containing water vapor was passed through the sample bed until the equilibrium state was attained.

(3) And then, moisture-free helium gas was passed through the sample bed again to release the adsorbed water changing the bed temperature from the experimental temperature to 1073K stepwisely.

The experimental conditions for LiA10_2, Li₄Si0₄ and Li₂Zr0₃ are shown in tables 4.6, 4.7 and 4.8, respectively. It have been known that such breakthrough curve as described by Johnson et al. can not be obtained under the experimental conditions as mentioned in chapter 3. Because the mass flow rate and the water concentration in the carrier gas were too large to obtain the desirable breakthrough curve for curve fitting in the case of the experimental conditions as mentioned in chapter 3. Accordingly, experiments were carried out below lOOppm of water concentration in the carrier gas and below 0.4//min of mass flow rate at the sample bed. However, 0.4//min of mass flow rate is necessary to operate the hygrometer normally. And then, He gas channel was attached to the outlet of sample bed to keep the flow rate of 0.4//min at the inlet of hygrometer.

4_·4 Results and discussion

Figs.4.2-4.4 show the breakthrough curves of water on LiA10₂, Li₄Si0 ₄ and

-132-

LizZr0

3, respectively. The tendency of these breakthrough curves suggests that the early step of adsorption with a fast adsorption rate and the late step of adsorption with a slow adsorption rate coexist. The changes of water concentration from the start point to the breakthrough point at each temperature agree with each other.

This is due to the so to speak system effect; the water captured on the piping surface is released. In this work, H₂0 !He gas was passed through the piping system without the reaction tube before each experiments to saturate the water on the p1p1ng surface. Accordingly, the influence of the system effect that gives to the breakthrough curve is restrained. But if H₂0/He gas is not passed through the piping system without the reaction tube before each experiment, the influence of the system effect that gives to the breakthrough curve must be large.

The over-all mass transfer coefficient clearly changes with the progress of adsorption as shown in figs.4.2,...,4.4. Generally, the step of diffusion in fluid film is the rate controlling step at the early step of adsorption, and the rate controlling step changes to the step of diffusion in pore with increase of the amount of adsorption.

However, it can not be considered that the step of diffusion in fluid film is the rate controlling step at the early step of adsorption in these cases, because the experimental temperature is high enough. Furthermore, the difference of over-all mass transfer coefficient between at early step and at late step is too large to consider that the water concentration dependency of over-all mass transfer coefficient is the basic reason of the change of over-all mass transfer coefficient.

And then, present author assumes that two kinds of adsorption having different mass transfer rate (adsorption(l) with a fast rate and adsorption(2) with a slow rate) progress at the same time. Then, the mass balance equation for water in the sample bed is expressed as

(4-7) u

ac

^{+ £}

ac

^{+ y}

^oq

^{= DL}

o

²

c

az

ot ot

^()z2

The mass flow rate is large enough to ignore the influence of axial diffusion under

- 133 -

the experimental conditions in this work. Accordingly, eq.( 4-7) is expressed as

U dC + £ dC + ydQ = 0

dZ dt dt ( 4-8)

It

can

be considered that the rate controlling step is any steps progress in the sample particle. Accordingly, it is assumed that the driving force of adsorption is the difference of the amount of adsorption. The mass transfer equation is expressed as

raq

⁼

raq1

⁺

raq2

dt dt dt

where

dQl K ( * ) dQ2 ^I *

Yat = sav

ql - q1

Yat = Ksav

(q2 - q2)

(4-9)

u c z

£

q ql 2

'

*

q 1 2 '

: superficial gas velocity [em/sec]

: concentration of water [mol/cm3]

: axial distance [em]

: void fraction of packed bed [-]

: the amount of adsorbed water [mol/g]

: the amount of adsorbed water by adsorption( I) or (2) [mol/g]

: the amount of water adsorption equilibrate to the concentration of water in gas phase for adsorption( I) or(2) [mol/g]

K₅av : the over-all mass transfer capacity coefficient of adsorption(!) at the standard of the amount of adsorbed water [g/cm3•sec]

K ₅^' av : the over-all mass transfer capacity coefficient of adsorption(2) at the standard of the amount of adsorbed water [g/cm3•sec].

The adsorption isotherms of water vapor on LiA10_2, Li₄Si0₄ and Li₂Zr03 are expressed as eqs.(3-3), (3-7) and (3-10), respectively. These equations are the Freundlich equation of which exponent is 1/2;

qo = a co112 (a

=

constant) ( 4-1 0)

And then, the relation of adsorption(!) and (2) are expressed as

-134 -

where : inlet concentration of water [mol/cm3]

q_1,0 :capacity of adsorption for adsorption(!) [mol/g]

q_{2 0} : capacity of adsorption for adsorption(2) [mol/g]

'

(4-11)

a : ratio of capacity of adsorption for adsorption( 1) to total capacity [-]

c* _{1 2 :} equilibrium concentration of water correlated to adsorption(!)

'

or (2) [mol/cm3].

The initial and boundary conditions are expressed as {

t=O,z>0-7q=O

z

=

0 ' t ~ 0 ^-7 c

=

^co. _{( 4-12)}

The curve fittings of breakthrough curves were carried out solving eqs.( 4- 8-12) numerically with the over-all mass transfer capacity coefficients (K sav, Ks'~) and the ratio of adsorption capacity (a) as the parameters. Figs.4.5-4.7 show examples of the curve fitting for the breakthrough curve of water on various breeder materials. The calculation values almost agree with the data obtained.

Figs.4.8-4.1 0 show the over-all mass transfer capacity coefficients for adsorption(!) and (2) and the ratio of the amount of adsorption( I) to the total amount of adsorption for LiA10₂, Li₄Si0 ₄ and Li₂Zr03, respectively. The over- all mass transfer capacity coefficient for adsorption( 1) is about one order magnitude larger than that for adsorption(2) for each material. But the

- 135 -

- - - -

temperature dependence of the over-all mass transfer capacity coefficient for adsorption(2) is larger than that for adsorption( I). Then, the ratio of the amount of adsorption( I) to the total amount of adsorption without Li₂Zr0₃ increases with the rise of temperature. Accordingly, the influence of adsorption(2) decreases with the rise of temperature. Following equations were obtained from figs.4.8-4.10.

Ksav

=

^{0.208 exp (-} 1jfi0) (LiAl02)

=

^{5.22x10-2exp (-}

4J~O)

^(L4Si04)

=

^8.39xi0-2exp (-

.lllQ_)

^(LhZr03)

RT , (4-13)

K'sav = 0.102 exp (-

2~i0)

^(LiA102)

= 1.32xi0-²exp (-

7~~0)

^(Li4Si04)

= 4.44xi0-²exp (- 10580) (Li2Zr03)

RT ,

a= 1.93 exp (-

5~~)

^(LiAl02)

= 1.18 exp (-

1~}0)

(Li4Si04)

= 0.60 exp (1⁴60) (LhZr03)

RT .

( 4-14)

( 4-15)

However, the ratio of the amount of adsorption( I) to the total amount of adsorption for Li₄Si0₄ or Li₂Zr0₃ may be considered to be constant. Figs.4.11 and 4.12 show the comparison of the over-all mass transfer capacity coefficient for adsorption(!) and (2) among various breeder materials, respectively. The over-all mass transfer capacity coefficients of LiA10₂ are the smallest among various breeder materials. Accordingly, the mass transfer rate of water on LiA10₂ is the smallest among various breeder materials. The estimation of the breakthrough

- 136-

curve of water adsorption for LiAI0_2, Li₄Si0₄ or Li₂Zr0₃ is possible using eqs.( 4-13 )-( 4-15).

The evaluation of the mass transfer rate of water at desorption step is important to estimate the tritium release rate at the blanket. The release curves of water on LiA10_2, Li₄Si0₄ and Li₂Zr0₃ are shown in figs.4.13-4.15, respectively.

The release rate of water at early step of desorption is fairly large, but that at late step is very small. To release all water that is captured on the sample, the sample bed must be heated up from the temperature at the adsorption experiment as shown in figs.3.5-3.7. When the release curves are estimated, the initial and boundary conditions are expressed as

{

t

=

^{0 , z > 0} ~ q

=

^qo

z=O,t~O~c=O. ( 4-16)

The curve fitting of the release curves of water were carried out solving eqs.( 4- 8)-(4-11) and ( 4-16) with the over-all mass transfer capacity coefficients and the ratio of the amount of desorption(!) to the total amount of desorption as the parameters. Figs.4.16-4.18 show examples of the curve fitting of the water release curves for LiA10_2, Li₄Si0₄ and Li₂Zr0_3, respectively. The amount of water captured on the piping surface was ignored when the estimation of the water release curves were carried out. However, the water release curves obtained in this work include the effect of water captured by the piping surface. Because the water captured by piping surface without the reaction tube was not purged away before the water release experiment. But it is impossible to eliminate the system effect from the release curves even if the water captured by piping surface without the reaction tube is purged away before the water release experiment, because the Water released from the sample bed moves downward with the repetition of adsorption on and desorption from the piping surface. Therefore, the over-all mass transfer capacity coefficients and the ratio of the amount of desorption(l)to the

total amount of desorption obtained by the curve fittings include not only the

-137-

influence of the sample but also the system effect. Figs.4.19-4.21 show the over-all rnass transfer capacity coefficients and the ratio of the amount of desorption( 1) to the total amount of desorption. The values of K sav are about two order of magnitude larger than those of K s' av at each material. The order of K sav at desorption step almost agree with that at adsorption step. The values of

a

at desorption step is smaller than those at adsorption step. Disagreement of values of

a

between at adsorption step and at desorption step is due to the system effect and using adsorption isotherm for estimation of the release curve. Whole water captured at adsorption step is not released at desorption step when the bed temperature is kept at the adsorption step. Accordingly, the desorption isotherm must be used for estimation of the real release curve. Following equations are obtained from figs.4.19-4.21.

Ksav = 0.235exp (-

1~6~)

^(LiAl02)

= 0.293exp (-

8~~0)

(L4Si04)

= 0.290exp (- 7640) (Li2Zr03)

RT ,

K~v

^{= 2.77xl0-2exp (-}

212~0)

^(LiAl02)

= 8.99x1

o-

^{3 exp (-}

1:~7~0)

^(Li4Si04)

= 4.63x10-³exp (-

2.800_)

(Li2Zr03)

RT ,

a= 5.96exp (-

1~~)

^(LiAl02)

= 0.426exp (-

1~~0)

^(Li4Si04)

= 0.86exp

(4200)

^(LhZr03)

RT .

( 4-17)

( 4-18)

( 4-19)

Figs.4.22 and 4.23 show the comparison of Ksav or Ks' av at desorption step among

- 138 -

various breeder materials, respectively. K s~ and K s' av for LiA10₂ are the smallest amount various breeder materials. Accordingly, the mass transfer rate of water on LiA10

2 is the smallest among various breeder materials.

The relation between the over-all mass transfer coefficient and the mass transfer coefficient is expressed as eq.( 4-4 ). The mass transfer coefficient in fluid film (kf) can be considered to be much larger than ks under the experimental conditions in this work. Therefore, it can be considered that the over-all mass transfer coefficient (K s) is equal to the film mass transfer coefficient (k s). It is possible to estimate the effective diffusivity of water in the particle at the standard of the amount of adsorption using eq.( 4-6). Figs.4.24-4.26 show the effective diffusivity of water in the particle of LiA10_2, Li₄Si0 ₄ and Li₂Zr0₃ at adsorption step, respectively. And figs.4.27 ... 4.29 show the effective diffusivity of water in the particle of LiA10_2, Li₄Si0 ₄ and Li₂Zr0₃ at desorption step, respectively.

Following equations are obtained from these figures.

(adsorption step)

D; 1 = 4.45x10-⁵exp (-

1~~0)

^(LiAl02)

=

^l.llx1

o-

^{5 exp (-}

4~~0)

^(Li4Si04)

= 6.26xl0-⁶exp {- 3000) (Li2Zr03)

RT ,

D; 2

=

2.18xl0-⁵exp (-

2~i0)

(LiAl02)

= 2.81x10-⁶exp (-

~)

^(Li4Si04)

= 4.75xl0-⁶exp (- 12100) (LhZr03)

RT ,

-139-

( 4-20)

(4-21)

(desorption step)

o;

^{I =} 5.03x10"⁵exp (-

1~6~)

^(LiA102)

= 6.23x10-⁵exp (-

8~~)

^(Li4Si04)

= 2.32xl0·⁵ exp (-

1S2Q_)

^(LhZr03)

RT ,

o;

^{2 = 5.92x10-6exp (-}

2~2Ff)

^(LiAl02)

= 1.91x10-⁶exp {-

1~~)

^(LliSi04)

= 3.98x10-⁶exp (- 10300) (LhZr03)

RT ,

(4-22)

(4-23)

where D / _{1 2 :} the effective diffusivity of water in the particle correlated '

adsorption(l) or (2) (desorption(l) or(2)) [cm²/sec].

Disagreement between the temperature dependence of diffusivity and that of the over-all mass transfer capacity coefficients at Li₂Zr0₃ is due to the difference of the density of packed bed.

Tritium diffusivity In the crystal grain of breeder material have been investigated by Kudo and Okuno[l-5] using measurement of tritiated water release rate. However, it can be considered that the tritium diffusivity obtained by Kudo and Okuno includes the resistance of surface reaction and the system effect.

Tritium diffusivity was estimated using eq.(4-6) and the over-all mass transfer capacity coefficient for desorption(2). When the tritium diffusivity was estimated, the grain size was used as dp and the density of pellet was used as y, because single crystals were used in their investigation of the tritium diffusivity. Fig.4.30 shows the comparison of the so to speak tritium diffusivity in the crystal grain.

Temperature dependence of diffusivity obtained in this work disagree with that Investigated by Kudo and Okuno. But, from the view point of order of values, the

data obtained in this work almost agree with that investigated by them. Therefore,

-140-

it can be considered that tritium diffusivity in the crystal grain obtained by Kudo and Okuno is possible to include the resistance of surface reaction and the system effect and that real tritium diffusivity in the crystal grain is much larger. So much difference in diffusivity by Kudo and Okuno and that by FINESSE for each breeder material may be explained using the same reasons.

4.5

Conclusion

The over-all mass transfer capacity coefficients of water on LiA10₂, Li₄Si0 ₄ and Li₂Zr0₃ were obtained. When it was considered that two kinds of adsorption whose rate are different with each other progress at same time, the estimation of the breakthrough curve and the release curve of water were to be possible.

The effective diffusivity of water in the particle of various breeder materials at the standard of the amount of adsorption were obtained from the over-all mass transfer capacity coefficient.

Tritium diffusivity was estimated when it was assumed that the release rate of water adsorbed on the breeder material is equal to the diffusion rate of tritium in the crystal grain of the breeder material, and was compared with the so to speak tritium diffusivity in the crystal grain measured by other investigators. Tritium diffusivity have been obtained is possible to include the resistance of surface reaction and the system effect.

When the water or the tritiated water release experiment from breeder material is carried out, it is impossible to eliminate the system effect without heating the piping system of an experimental apparatus. Accordingly, it is necessary that the amount of water captured on the piping surface and the mass transfer rate of water on the piping surface are quantified.

- 141 -

Nomenclature

c

C· _l

: surface area of particle per unit volume of packed bed [cm²/cm3]

: concentration of water [mol/cm3]

: concentration of adsorbate in gas phase [mol/cm3]

: concentration of adsorbate at the surface of particle [mol/cm3]

: inlet concentration of water [mol/cm3]

c* _{1 2 :} equilibrium concentration of water correlated to

'

adsorption(!) or (2) [mol/cm3].

D 'i : the effective diffusivity of adsorbate in the particle at the standard of the amount of adsorption [ cm² /sec].

D / _{1,2 :} the effective diffusivity of water in the particle correlated adsorption(!) or (2) [cm²/sec].

dp : diameter of particle [em].

K1 : over-all mass transfer coefficient of fluid film [em/sec]

K s : over-all mass transfer coefficient of solid film [g/cm²•sec].

K sav : the over-all mass transfer capacity coefficient of adsorption(!)

at the standard of the amount of adsorbed water [g/cm3•sec]

Ks'

0v :

the over-all mass transfer capacity coefficient of adsorption(2) at the standard of the amount of adsorbed water [g/cm3•sec].

k1 :

mass transfer coefficient of fluid film [em/sec]

k s : mass transfer coefficient of solid film [g/cm²•sec]

q : the amount of adsorbed water [mol/g) : the amount of adsorption [mol/g)

qi : the amount of adsorption at the surface of particle [mol/g).

- 142-

q_1,2 : the amount of adsorbed water by adsorption(!) or (2) [mol/g) q ₁ ,O : capacity of adsorption for adsorption(!) [mol/g)

q_2,0 : capacity of adsorption for adsorption(2) [mol/g)

q* _{1,2 :} the amount of water adsorption equilibrate to the concentration of water in gas phase for adsorption( I) or(2) [mol/g)

t : time [sec]

u : superficial gas velocity [em/sec]

z : axial distance [em]

a :

ratio of capacity of adsorption for adsorption( 1) to total capacity [-]

f3 :

adsorption coefficient [ cm3/g]

y : density of packed bed [g/cm3]

£ : void fraction of packed bed [-]

- 143 -

References

[1] H.Kudo, K.Okuno, J.Nucl.Mater. 101 (1981) 38.

[2] H.Kudo, K.Okuno, J.Nucl.Mater. 133-134 (1985) 192.

[3] K.Okuno, H.Kudo, J.Nucl.Mater. 116 (1983) 82.

[4] K.Okuno, H.Kudo, J.Nucl.Mater. 138 (1986) 210.

[5] H.Kudo, K.Okuno, J.Nucl.Mater. 155-157 (1988) 524.

[6] T.Kurasawa et al., JAERI-M86-152, (1986).

[7] H.Watanabe, T.Katsuta, E.Roth and D.Vollath, Tritium Recovery Experiments in JRR-2NOM, Proc.Int. Symp.Fusion Reactor Blanket and Fuel Cycle Technology, Tokai, (1986).

[8] T.Kurasawa, G.W.Hollenberg and H.Watanabe, Tritium Recovery and Behavior of Li₂0 and LiA10₂ Spheres, Proc.Int. Symp.Fusion Reactor Blanket and Fuel Cycle Technology, Tokai, (1986).

[9] T.Kurasawa, H.Watanabe, G.W.Hollenberg et al., J.Nucl.Mater.

141-143 (1986) 265.

[10] T.Kurasawa, H.Watanabe, E.Roth and D.Vollath, J.Nucl.Mater.

155-157 (1988) 544-548.

[11] R.G.Clemmer et al., J.Nucl.Mater. 133&134 (1988) 171.

[12] J.P.Kopasz, S.W.Tam and C.E.Johnson, J.Nucl.Mater.

155-157 (1988) 500-506.

[13] H.Werle et al., J.Nucl.Mater. 155-157 (1988) 538-543.

[14] H.Werle et al., J.Nucl.Mater. 141-143 (1986) 321.

[15] M.Briec et al., J.Nucl.Mater. 155-157 (1988) 549-552.

[16] J.M.Miller et al., 3rd Topical Meeting on Tritium Technology in Fission, Fusion and Isotopic Applications, Trant, (1988).

[17] M.Briec et al., SOFT Conference (1988).

[18] C.Alvani et al., SOFT Conference (1988).

- 144-

[19] H.Yanai, Kyuutyaku Kougaku Youron, 57 (Kyoritsu Press, Tokyo, 1978).

[20] H.Kitagawa and K.Suzuki, Kyuutyaku no Kiso to Sekkei, 43 (Maki Press, Tokyo, 1978).

-145 -

,_.

+:>.

0\

I

Table 4.1 The sample specification.

Sample Name LiA10₂

Theoretical Density 2.6g/cm³

Density 2.2lg/cm³ (85%T.D.)

Grain Size 20J.Lm (diameter)

Pellet Size 12~16mesh

BET Surface Area 0.29m²/g

...

.+:>..

-...]

Table 4.2 The sample specification.

Sample Name Li₄Si0₄

Theoretical Density 2.2lg/cm³

Density 1.88g/cm³ (85%T.D.)

Grain Size 20fJ.m (diameter)

Pellet Size 12---16mesh

BET Surface Area 0.73m²/g

...

+>-.

00

I

Table 4.3 The sample specification.

Sample Name Li₂Zr0₃

Theoretical Density 4.15g/cm³

Density 3.57g/cm³ (86%T.D.)

Grain Size 13~m (diameter)

Pellet Size l.Omm<f>

BET Surface Area 0.09m²/g

- - - - - - - - - --- - - - - - - - - - --- - - -

...

~

\0 ,,

Table 4.4 The chemical composition of LiA10₂ used in this work.

Results of chemical analysis (ICP)

element sintering pellet (1934.7mg)

(detectable limit) measurement ^I content

(ppm) (ppm) ^{I I} _I (mg)

I

AI (<0.018) 7094 ^I I 709.4

I

Li ( <0.013) 1777.6 ^I I 177.76

I

Mo (<1.072) 0.779 ^I I 0.078

I

Cr (<0.009) 0.0101 ^I I 0.001

I

Fe (<0.008) 0.4580 ^I I 0.046

I

Ni ( <0.064) <0.064 ^I I <0.006

I

total I ^I

- - - - - -·- - - -

*

ICP analysis is made for LiA10₂ solution in aqua regia when about 90% of sintered pellet is dissolved.

content (o/o) 36.66718

9.18799 0.00403 0.00005 0.00237

<0.00033 45.86631

-

lJt 0

I

Table 4.5 Spectrographic analysis of Li₂Zr0₃ used in this work.

AI Ca

Cr Cu

RESULTS OF SPECTROGRAPHIC ANALYSIS

<0.01 o/o

<0.01

<0.01

<0.01

Fe Mg Ti

0.04o/o

<0.01 0.04

...

V\ ...

Table 4.6 Experimental conditions.

Sample Name LiAI0₂

Packed bed weight of sample 18.02g

Height of packed bed 3.9cm

Carrier gas He

Volumetric gas velocity 0.2-0.4//min

Concentration of H₂0 36.0-48.2ppm in atmospheric pressure (3.6--A.9Pa)

Temperature of sample bed 150-600°C

(423-873K)

I - ' U\

t0

I

Table 4.7 Experimental conditions.

Sample Name Li₄Si0₄

Packed bed weight of sample 11.62g

Height of packed bed 2.5cm

Carrier gas He

Volumetric gas velocity 0.2~0.4//min

Concentration of H₂0 48.2-59 .4ppm in atmospheric pressure

(4.9~6.0Pa) .

Temperature of sample bed ^150~3oooc

(423~573K)

...

V\ w

Table 4.8 Experimental conditions.

Sample Name Li₂Zr0₃

Packed bed weight of sample 70.0g 99.92g

Height of packed bed 7.7cm 11.4cm

Carrier gas He

Volumetric gas velocity 0.2~0.4!/min

Concentration of H₂0 36.0----59 .4ppm in atmospheric pressure

(3.6~6.0Pa)

Temperature of sample bed 150-400°C

(423-673K)

Fig.4.1

He-H2

Liquid N2 Cold Trap (glass beads)

Spongy CuO Bed

~

0

z

0 ^{0 0} I 0

0

Sample Bed

Hygrometer

He

Liquid N2 Cold Trap (MS-5A)

GasChromatography

A schematic diagram of experimental apparatus.

- 154-

r---1 I L-...J

c

~

u

Ul

Ul I ~

::l

u

0

1.0•

^{I - · - · I I I I}

0.8

0.6

0.4

0.2

I

LiAI02 (85o/aT.D.) Inlet Cone. of H20 Flow Rate

---

-o-

----

: 18.02 g : 36.0 ppm : 0.4 //min

200°C 300

oc

400

oc

0.0~--~~--~~----~----~----~----~

0 20 40 60 80 100 . 120

Time [min]

Fig.4.2 Examples of breakthrough curves of water for LiAI0₂.

1.0 I I ___... • I • • I

⁰

e ^il

^u

^{a I}

^{0 <}

0.8

,.---, ^I

0.6

'----J

...

·-

c

lJI

u

0\

I ~

:::::3

0

0.4

u

_{I I I}

Li4Si04 (85%T.D.) : 11.62 g Inlet Cone. of H20 : 59.4 ppm Flow Rate : 0.4 //min

0.2

^~

I I

^{-o- 150 °C}

---

³⁰⁰

^oc

0.0~~--~----~----~----~----~----~

0 10 20 30 40 50 60

Time [min]

Fig.4.3 Examples of breakthrough curves of water for Li₄ SiO ₄.

...

Lll -..l

I

1.0T ~ ^{•• •} •!••••• ., •••• ^••1

0.8

~

0.6

u

c:

~ ::J

u

0

0.4

0.2

Li2Zr03 (86%T.D.) Inlet Cone. of H20 Flow Rate

_.._

-o-

--11-

:70.0 g : 36.0 ppm : 0.4 //min

150

oc

250°C 350

oc

0.0~~--~----~----~----~----~----~

0 ^{20 40 60} 80 100 120

Time [min]

Fig.4.4 Examples of breakthrough curves of water for Li₂Zr0_3.

r--1 I 1...-.J

c

...

Ul

c5

00 ~

I ::l

uo

1. 0 }----,.---,.---,----,---.---:---,

0.8 I

0.6 r

0.4

0.2

0

ooo

f

vestimated

LiAI02 (85o/aT.D.) Inlet Cone. of H20 Temp.

Flow Rate

~av

~·Clv

: 18.02 g : 36.0 ppm :300 oc : 0.4 //min.

0^{-2 3} : l.Ox 1 g/cm •sec

-4 3

: 6.0X10 g/cm •sec

0.0~--~~----~---~----~--~---~

0

Fig.4.5

20 40 60 80 100 .

Time [min.]

An example of curve fitting for the breakthrough curve of water for LiA10_2.

120

~

I L...-....J

c

... 0~

lJI

\0 ~

I :J

u

0

1. 0

r---....,---r---r----~-r-r---<

0.8

0.6

0.4.

0.2

^~

0 0

0.0 0

Fig.4.6

~estimated

Li4Si04 (85%T.D.) : 11.62 g Inlet Cone. of H20 : 48.2 ppm

Temp. : 200

oc

Flow Rate :0.18//min

I

_~av : 0.017 g/cm •sec ³

!\,'ely : 0. 002 g/cm •sec ³

I

10 20 30 40 so .

Time [min]

An example of curve fitting for the breakthrough curve of water for Li₄Si0₄.

60

...

8

I

1 . 0

^{1 1 I I I} n ____f"\...1 , , c ) I

0.8

~

0.6

c

u

~ :J

u

0

0.4

0.2

0 0

Li2Zr03 (86%T.D.) Inlet Cone. of I-120 Temp.

Flow Rate Ksav

~·av

: 70.0 g : 36.0 ppm : 250 °C :0.19//min : 0.04 g/cm •sec 3

: 0.0035 g/cm •sec 3

0.0~----~--~~----~----~----~----~

0

Fig.4.7

10 20 30 40 so .

Time [min]

An example of curve fitting for the breakthrough curve of water for Li₂Zr0₃.

60

Fig.4.8

Temp. [°C]

900 600 400 300 200

I I

Adsorption Step LiA102 (85o/o.T.D.) Water Vapor Pressure

o 3.6 Pa

o 4.9 Pa

0.8 1.0 1.2

o:

=

1.93 exp (-

5~~0)

K5av

=

0.208 exp (-

1~6i0)

o/

1.4

1.6 1.8 2.0 2.2

The over-all mass transfer capacity coefficients for adsorption(!) and (2) and the ratio of the amount of adsorption(!) to total amount of adsorption for LiA10 2.

- 161 -

>

C'd

Te1np. [°C]

800 500 300 200

10°

I I

Fig.4.9

Adsorption Step Li4Si04 (85o/oT.D.) Water Vapor Pressure

o 4.9 Pa

o 6.0 Pa

a =

1.18 exp (- 1

~ ~0)

0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4

The over-all mass transfer capacity coefficients for adsorption( 1) and (2) and the ratio of the amount of adsorption( I) to total amount of adsorption for Li4Si0_4.

- 162-

Tetnp. [°C]

800 500 300

a= 0.60 exp

(1~~0)

Adsorption Step Li2Zr03 (86o/oT.D.) Water Vapor Pressure

o 3.6 Pa o 6.0 Pa

200

10-

⁴ ~----~---~---~---~

1.0 1.5 2.0 2.5

Fig.4.1 0 The over-all mass transfer capacity coefficients for adsorption(l) and (2) and the ratio of the amount of adsorption(l) to total amount of adsorption for Li₂Zr0_3.

- 163 -

,..,

(.) (L) {/)

•

C1')

E

(.) ...

01)

L-....1

>

ro _(/)

~

1 o-

¹

10-

²

600

Te1np. [°C]

400 300

Ksav of Adsorption Step _._ LiAl02 -o- Li4Si04 --o- Li2Zr03

10-

³ ~--~--~--~----~--~--~--~--~

0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4

Fig.4.11 The comparison of over-all mass transfer cap~city

coefficient for adsorption( 1) among various breeder materials.

- 164-

,....,

u ~ r./)

•

('(')

s

u

...

b1)

L...J

>

-

ro ^en

~

Te1np. [°C]

600 400 300 200

10-1

rr-r-r-.---.---.----~---~----~

Ks' av of Adsorption Step -e- LiAl02

-11-

Li4Si04

-o- Li2Zr03

10-

²

10-

³

10-

⁴ ~--~--~--~----~--~---L--~--~

0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4

Fig.4.12 The comparison of over-all mass transfer capacity coefficient for adsorption(2) an1ong various breeder materials.

- 165 -

~

~

I

r---1 I L...J

... c

-d ro

u

~ ::::l

u

0

1. 0

---~----,---,.---.---r---,

0.8

0.6

0.4

0.2

~I l I

l \ l ^I l \

l I

l I

: ^\

\e ~ \

\a

l \ l \

l \

\

\.

.,

^{~ \}

LiAI02 (85%T.D.)

Cone. of H20 (adsorption step) Flow Rate

---

^-0-

···•···

: 18.02 g : 36.0 ppm : 0.4 l/min

200°C 400°C 600°C

\ 0,

^\ _\ ^'

j

'

..

\• o,,-o.

^{·. ...0}

~ • • • • • • • • • • •

·-... - o- -

D--6- - - -

···-•····-•···•· ··-··•--·~···~···~···~-····~---·~---~·-··~·r·~·--·~-···

^{.. -···}

0.0~--~--~--~--~--~----~

0 10 20 30 40 50 . 60

Time [min]

Fig.4.13 Examples of desorption curves of water for LiAI0₂.

.-..

0\

-..) I

1.0

0.8

~

_c:

^0.6

...

"'d

u

C'd

~ ::l

u

0

^0.4

0.2

Li4Si04 (85%T.D.)

Cone. of H20 (adsorption step) Flow Rate

---

^-0-

: 11.62 g : 59.4 ppm : 0.4 //min

150

oc

300

oc

0.0~----~----~----~----~----~----~

0 10 20 30 40 50 . 60

Time [min]

Fig.4.14 Examples of desorption curves of water for Li₄Si0₄.

,.--,

I L.-.1

... c

-o Clj

~

u

0\ 00

~ :J

u

0

1.0

0.8

0.6

~'I

ⁱ _i ^\ _\

0.4 ~ \

' ^\

' .

^'. _'.

.

^\

^,

^\

0.2

I

Li2Zr03 (86%T.D.)

Cone. of H20( adsorption step) Flow Rate

-o-

-·-

^-0-

: 70.0 g : 36.0 ppm : 0.4 //min

200°C 300 °C 400

oc

0 0

-.-. =-

~-~-~ ~~ ~-.-.-.

0.0~----~----~----~----~----~----~

0 10 20 30 40 so 60

Time [min]

Fig.4.15 Examples of desorption curves of water for Li₂Zr0₃.

1.0

Ft

LiAI02 (85%.T.D.) : 18.02 g

0.8

^~ Cone. of H20 (adsorption step) : 36.0 ppm

Temp. :600

oc

Flow Rate : 0.4 //min

Ks~ ^{-2 3}

r---1

0.6 L\

^{: 3 .Sx} 10 g/cm •sec

I -4 3

L-...J

Ks'av : 5.0X10 g/cm •sec

c

~ '"d

0\ ro

\0 ^I

u

~ :::l

0

0.4

u

I \

estimated

0.2

^~

'h

I

^{I '-'~ I}

I

0.0 0 10 20 30 40 50 60

Time [min]

Fig.4.16 An example of curve fitting for desorption curve of water for LiA10₂.

,..--,

'

L,..__J

c

... "'d

-..J ^C'U

0 ^I

u

~ :::1

u

0

1.0

0.8

0.6

0.4

0.2

Li4Si04 (85%T.D.)

Cone. of H20 (adsorption step) Temp.

Flow Rate Ksa_,

Ks'av

estimated

I

0 0 0 0 0

: 11.62 g : 48.2 ppm : 200

oc

: 0.4 //min

-2 3

: 4.0X10 g/cm •sec

-4 I 3

: 4.5X10 gem •sec

0 0 0 0 0

o.o

0 0

0.0~----~----~----~----~----~----~

0 10

Fig.4.17

20 30 40 50 ..

Time [min]

An example of curve fitting for desorption curve of water for Li₄Si0 ₄.

60

r---1 I L...-.J

--

c

~ '"d

-.) ro

~

u

~ :::1

u

0

1.0

0.8

0.6

0.4

0.2

Li2Zr03 (86o/oT.D.)

Cone. of H20( adsorption step) Temp.

Flow Rate

~a.,

~'~

estimated

~

:70.0 g : 36.0 ppm : 150

oc

: 0.4 //min

-2 3

: 3 .5x 1 0 g/cm •sec

-4 3

: 3.0X10 g/cm •sec

0 0 0 0 0 0 0

I I

0. 0

L - - - " - - - ' - - - ' - -

0 10

Fig.4.18

20 30 40 50

Time [min]

An example of curve fitting for desorption curve of water for Li₂Zr0₃.

60

>

ro

Temp. [°C]

900 600 400 300 200

K a

=

^0.235 exp (- l2600)

s v RT

K 'a

=

^2.77x10-2exp (- 25200)

s v RT

0

Desorption Step LiA102 (85o/oT.D.) Water Vapor Pressure

o 3.6 Pa

o 4.9 Pa

I

0

10-

⁵ L--L----~--~--~----~--~----~--~

0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2

Fig.4.19 The over-all mass transfer capacity coefficients for desorption(!) and (2) and the ratio of the amount of desorption(!) to total amount of desorption for LiA10₂.

- 172-

,...,

I I,...J

Temp. [°C]

800 500 300 200

o: ^:= 0.426 exp (-

1~}0) 7

K a _sv

=

^{0.293 exp (- 8300)} _RT

Desorption Step Li4Si04 (85o/oT.D.) Water Vapor Pressure

K'a

=

8.99x10-³exp (- 12700)

sv RT

o 4.9 Pa o 6.0 Pa

0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4

Fig.4.20 The over-all mass transfer capacity coefficients for desorption( 1) and (2) and the ratio of the amount of desorption(!) to total amount of desorption for Li₄Si0 _{4 .}

- 173 -

Temp. [°C]

800 500 300 200

a= 0.86exp (- ⁴~~⁰)

K _{s v} a

=

⁰ _• ^{29exp (- 7640 )} _RT

Desorption Step Li2Zr03 (86%T.D.) Water Vapor Pressure

o 3.6 Pa o 6.0 Pa

K'a _sv

=

4 _. 63xl0-³exp (- 9800) RT

10-

⁵ ~----~---~---~---~

Fig.4.21

1.0 1.5 2.0 2.5

The over-all mass transfer capacity coefficients for desorption(!) and (2) and the ratio of the amount of desorption(!) to total amount of desorption for Li

2Zr0

3.

- 174-

,..--, () 0 (/:)

•

('()

J2. s

1...-J b1J

>

~ en

~

1 o-

¹

10-

²

600

Temp. [°C]

400 300

Ksav of Desorption Step

-e- LiAl02

-o- Li4Si04

-u- Li2Zr03

200

10-

³ L---~---L--~----~~~~--~--~--~

0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4

Fig.4.22 The con1parison of over-all mass transfer capacity coefficient for desorption( 1) a1nong various breeder materials.

-175 -

,...,

()

~ r:/:J

•

('()

< 8

b1)

L...J

>

C':j

-

^en

~

10-

³

10-

⁴

600

Te1np. [°C]

400 300

Ks' av of Desorption Step __._ LiAl02

-II-.

Li4Si04

-o- Li2Zr03

200

•

10-

⁵ ~--~--~----~--~--~----~--~--~

0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4

Fig.4.23 The comparison of over-all mass transfer capacity coefficient for desorption(2) an1ong various breeder materials.

- 176-

Temp. [°C]

800 500 300 200

D;' 1 = 4.45x10-⁵exp (-

1~6~0)

• /e

Adsorption Step LiA102 (85o/oT.D.)

--- Di'l

-o- Di'2

0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2

Fig.4.24 The effective diffusivity of water in the particle of LiAI0₂ at adsorption step.

- 177-

Temp. [°C]

400 300 200

Adsorption Step Li4Si04 (85%T.D.)

--- Di'I

-o- Di'2

0

10-

⁷

~----~---~----~----~---~~

1.4 1.6 1.8 2.0 2.2 2.4

Fig.4.25 The effective diffusivity of water in the particle of Li₄ SiO ₄ at adsorption step.

- 178 -