Professional Documents
Culture Documents
A R T I C L E I N F O A B S T R A C T
Article history:
Received 20 April 2016 Physical properties of hydrogen isotopes, hydrogen (H2), hydrogen-deuterium (HD), hydrogen-tritium
Received in revised form 13 July 2016 (HT), deuterium (D2), deuterium-tritium (DT), and tritium (T2) were estimated through vapor pressure
Accepted 30 July 2016 prediction, and validated by steady-state simulation of ITER isotope separation system (ISS). Peng–
Available online 10 August 2016 Robinson (PR) equation of state with Twu alpha function was selected for modelling which showed
favorable prediction from the experimental vapor pressures of each hydrogen isotopes. The steady-state
Keywords: simulation of ITER ISS using Aspen Plus consists of four distillation columns and seven equilibrium
Hydrogen isotopes reactors with four purified products: D2, T2, HD, and DT. Converged solution from simulation produced
Isotope separation system
potential scenario for actual ITER ISS process.
Physical property estimation
© 2016 Published by Elsevier B.V. on behalf of The Korean Society of Industrial and Engineering
Process simulation
Peng–Robinson equation of state Chemistry.
http://dx.doi.org/10.1016/j.jiec.2016.07.053
1226-086X/© 2016 Published by Elsevier B.V. on behalf of The Korean Society of Industrial and Engineering Chemistry.
2 J. Noh et al. / Journal of Industrial and Engineering Chemistry 46 (2017) 1–8
are the 6 hydrogen isotopes: H2, HD, HT, D2, DT, and T2. In the ISS H2, HD, HT, D2, DT, and T2 components were obtained using the
process, out of the six mixed gas, D2, DT, and T2, which are used as a equation of state which allows the simulation of the ISS process.
raw material for ITER, should be purified with the desired
compositions. To produce the desired composition of D2, DT, Estimation of thermodynamic properties
and T2 from ISS, a total of four distillation columns and several
equilibrium reactors (Equilibrator) should be installed. In the Estimation of fixed properties of pure components
equilibrium reactor, the following reversible reactions take place:
2HD $ H2 + D2, 2HT $ H2 + T2, and 2DT $ D2 + T2. As shown in Table 2, the physical properties of the pure
Table 1, adapted from Song et al. [8], shows the equilibrium component for hydrogen isotopes, such as H2, HD, D2, HT, DT and
constant of isotopes based on temperature for each equilibrium T2, must be obtained in order to estimate the phase equilibrium
reaction. Generally, these equilibrium reactions have slow reaction properties using the equation of state such as Peng–Robinson [10]
rate at very low temperature and fast reaction rate at room and Soave–Redlich–Kwong [11]. However, from the compatibility
temperature in terms of D2 and T2 generation. The desired of chemical process simulator such as Aspen Tech Corporation’s
molecular bond form can lead to a reaction when the supplied Aspen Plus, Invensys’ PRO/II with PROVISION, etc., only few
composition was set in the equilibrium reactor by optimizing the properties are built-in for H2, HD and D2, and no properties are
position of the equilibrium reactor [9]. available for HT, DT, and T2. Thus, in this study, to estimate the
ITER ISS process uses a cryogenic distillation and catalytic thermodynamic properties of H2, HD, D2, HT, DT, and T2 required
reaction from temperatures 14 to 30 K. Catalytic reaction takes for the separation process, related experimental data were
place in equilibrium in which the reactor should be installed in an collected which were used for regression analysis to obtain model
optimum position depending on the concentration profile of the equation parameters, and, finally, to estimate thermodynamic
column. In addition, with WDS linked to the ISS process, a highly properties (Table 3).
complex optimization technique will be required. Therefore, in The pure component properties for the six hydrogen isotopes
order to design and optimize this process, process simulation based on comprehensive study of experimental data available in
should be performed. However, in Korea, there are few to none literature [12–14] are shown in Tables 3 and 4. The Gibbs free
organizations that have performed process simulation of the energy for each of the component is presented in which the value
cryogenic distillation process for the ITER ISS. One of the reasons is for pure elements, H2, D2 and T2, are defined as zero (0), while the
because domestic technology for the ISS process is still in the values for all the other compounds are defined by a specific value.
concept establishing phase and international technology also The value of Gibbs free energy of formation, DG f,i, for the
failed to facilitate the tracking of related data which is critically compounds are estimated from the standard Gibbs free energy
low. Another reason is the lack of technology and the physical change for a given chemical reaction. With these physical property
property data to perform the cryogenic distillation process values, the calculation for the equilibrium reactor is carried out
simulation. In this study, the necessary physical properties for (Table 4).
Table 1
Equilibrium constant for isotopes.
Table 2
Reaction formula Equilibrium constant Fixed properties for using equation of state model.
8e+5
Exp_H2(0.2% o-H2 and 99.8% p-H2)
Exp_HD
Exp_D2
Vapor pressure (Pa)
Exp_HT
6e+5 Exp_DT
Exp._T2
4e+5
2e+5
0
16 18 20 22 24 26 28 30
Temperature (K)
Fig. 2. Vapor pressure prediction for H2, HD, D2, HT, DT and T2.
4 J. Noh et al. / Journal of Industrial and Engineering Chemistry 46 (2017) 1–8
However, the coefficients of the acentric factor in Eq. (5) were Table 5
Coefficients for the Twu alpha function.
selected by regression analysis of the experimental data which is
limited components of hydrocarbon. Thus, it is unsuitable to apply Component L M N
the temperature-dependent steam pressure to other than those of 1. H2 0.7189 2.5411 10.2000
hydrocarbon since the vapor pressure predictions are inaccurate as 2. HD 0.1009 1.0204 1.9102
shown in Fig. 3. To make up for such flaw, the Twu alpha function 3. D2 1.2584 6.1846 0.0759
4. HT 0.6820 1.2469 0.2000
[17] is used as presented in Eq. (6).
5. DT 0.9783 1.6726 0.2000
aðTÞ ¼ T r NðM1Þ exp½Lð1 T NM
r Þ ð6Þ 6. T2 1.0943 1.6009 0.2000
The values of the L,M, and N are unique for each component and
are determined from the regression of pure component vapor Simulation of isotope separation system
pressure data. The coefficients, L, M, and N, for the new alpha
function of the components of hydrogen isotopes are shown in The estimated thermodynamic properties are evaluated
Table 5 which were obtained from the regression analysis using through process simulation of the ITER isotope separation. The
Aspen Plus. Positive vapor pressure prediction is achieved using concept of isotope separation is based on the difference of 20–25 K
the new alpha function as shown in Fig. 4. in boiling points at atmospheric pressure for the six combinations
8e+5
Exp_H2(0.2% o-H2 and 99.8% p-H2)
Exp_HD
Exp_D2
Vapor pressure (Pa)
Exp_HT
6e+5 Exp_DT
Exp._T2
Calc_H2 using PR EOS(Original a)
Calc._HD using PR EOS(Original a)
Calc._D2 using PR EOS(Original a)
4e+5 Calc._HT using PR EOS(Original a)
Calc._DT using PR EOS(Original a)
Calc._T2 using PR EOS(Original a)
2e+5
0
16 18 20 22 24 26 28
Temperature (K)
Fig. 3. Vapor pressure prediction for H2, HD, D2, HT, DT and T2 using Peng–Robinson (original alpha function).
Fig. 4. Vapor pressures prediction H2, HD, D2, HT, DT and T2 using Peng–Robinson (Twu alpha function).
J. Noh et al. / Journal of Industrial and Engineering Chemistry 46 (2017) 1–8 5
Fig. 6. ITER ISS process flow diagram with feed & product stages.
J. Noh et al. / Journal of Industrial and Engineering Chemistry 46 (2017) 1–8 7
Table 6 Table 7b
Isotopic compositions for feed streams (Ref. to stream numbers in Fig. 2). Isotopic compositions for product streams.
of H, D, and T isotopes. The differences in volatility allow the isotope equilibrators are first warmed up to ambient temperature
separation of isotopes through distillation with the tritium at the by exchanging heat with the returning equilibrated stream for a
bottom, and protium, the most volatile species in the system, at the more effective operation. The flow rates and composition of the
top of the distillation cascade. As shown in Fig. 5, the ISS design feed streams to the ISS are known to be fluctuating with the Torus
setup for ITER comprises of four cryogenic distillation columns pulses at normal operation. Thus, the column cascade setup needs
which supplies specific tritium, deuterium, and hydrogen compo- to be specifically designed with respect to thermal and isotopic
sitions. The system has two feed gas streams, the stream from the fluctuation [21]. Fig. 6 exhibits the equilibrator locations which is
Torus exhaust and another from the WDS blended with the return critical in obtaining the necessary products. Further analysis on
stream from NBI [20]. Column 1 separates the tritium from the various locations of the equilibrator is recommended for optimi-
hydrogen-rich stream to obtain pure hydrogen as the overhead zation design of the ISS.
product. This stream is required to be almost tritium-free before The ITER ISS process illustrated in Fig. 5 was simulated using the
discharging to the environment since this is one of the major safety Peng–Robinson equation of state with Twu alpha function. Table 6
and environmental concerns of this study. The overhead product is shows the feed streams entering the ISS process while Tables 7a
recommended to be returned to the WDS. The tritium bottom and 7b show the product stream. As shown in Table 7a, the top
product from Column 1 proceeds to Column 2 and is treated to (stream no. 5) of Column 1 (CD1) has trace T2 content, and
produce D2 stream which is then used as fuel and feed for Column <20 ppm T2 content in fuel grade deuterium product (stream no.
3. The tritium bottom product from Column 1 to Column 2 is a D2- 27). The D2 content for NB injector product (stream no. 26) is
DT rich stream which is passed through equilibrator 2 to convert >99 mol%. In Table 7b, the DT product (stream no. 30) is separated
the traces of tritium from HT into DT by the reaction: HT + D2 $ in a ratio of 50% T/50%D while the T product (stream no. 21) can be
DT + HD. Column 2 separates H and D from the overhead stream of separated in a ratio of 90%T/10%D. The resulting column design
Column 4, and generates a partially purified D2 which is then taken parameters are given in Table 8. The design data for the columns in
to Column 3. The final purification occurs in Column 3 in which ND the ISS collected from the process simulation can be used as a basis
quality deuterium is extracted as the top product. The exhaust for detailed internal column design and dynamic simulation.
stream from TEP is treated in Column 4 in which two products are Souers [13] explains the influence of the ortho–para composi-
extracted at the mid-column, and at the bottom of the column, DT tion of a hydrogen molecule in the physical properties which
with 50%T:50%D, and DT with 90%T:10%D, respectively. usually only is a few percent and diminishes as the molecular
Equilibrators are considered to be used in the ISS to promote weight increases. Despite the minimal influence in the physical
exchange reactions between the isotopomer molecules. The properties, the amount of energy within the metastable rotational
Table 7a
Isotopic compositions for product streams.
Stream no. 5 27 26
Description CD1 overhead hydrogen reject Fuel grade deuterium product Deuterium for NB injector product
Table 8 Acknowledgment
Column summary of ISS design parameters.
Column name Column 1 Column 2 Column 3 Column 4 This work was supported by the Ministry of Science, ICT and
Number of stages 123 98 58 68 Future Planning of the Republic of Korea under the ITER technology
Feed stage 15, 75 10, 19, 50, 93 10 25, 65 R&D program.
Reflux ratio 3.12 78.41 14.86 15.00
Reflux rate (mol/min) 15.10 288.31 5.45 90.75
References
Top distillate rate (mol/min) 4.83 3.68 0.37 6.05
Column top press. (atm) 1.09 1.09 1.09 1.38
[1] M. Glugla, ITER IDM Document (ITER_D_2X6K67), Plant Description (PD), Fuel
Stage pressure drop (atm) 1.38 1.14 1.25 1.57
Cycle and Radiological Monitoring, 2009 (Chapter 10).
Condenser temp. (K) 20.66 22.09 23.94 24.79
[2] D. Babineau, S. Maruyama, R. Pearce, M. Glugla, Li Bo, B. Rogers, S. Willms, G.
Reboiler temp. (K) 23.61 24.72 24.48 26.59 Piazza, T. Yamanishi, S.H. Yun, L. Worth, W. Shu, Proceedings of the 23rd IAEA
Condenser heat duty (W) 305.04 5196.97 118.78 1830.83 Fusion Energy Conference, Daejeon, Republic of Korea, 2010.
Reboiler heat duty (W) 307.04 5078.83 119.08 1932.38 [3] M. Glugla, D. Babineau, L. Bo, S. Maruyama, R. Pearce, G. Piazza, B. Rogers, S.
Willms, T. Yamanishi, S.-H. Yun, Proceedings of the 9th International Conference
on Tritium Science and Technology, Tritium 2010, Nara, Japan, 2010.
[4] S. Maruyama, Y. Yang, R.A. Pitts, M. Sugihara, S. Putvinski, S. Carpentier-
Chouchana, B. Li, W. Li, L.R. Baylor, S.J. Meitner, C. Day, B. LaBombard, M.
Reinke, Proceedings of the 23rd IAEA Fusion Energy Conference, Daejeon,
states can be extensive for up to hundreds of J/mol. Though this is Republic of Korea, 2010.
[5] S. Konishi, M. Glugla, T. Hayashi, Fusion Eng. Des. 83 (2008) 954.
the case, Younglove [22] and Leachman et al. [23] has pointed out [6] M. Glugla, A. Antipenkov, S. Beloglazov, C. Caldwell-Nichols, I.R. Cristescu, I.
that the maximum deviations or uncertainties of thermophysical Cristescu, C. Day, L. Doerr, J.-P. Girard, E. Tada, Fusion Eng. Des. 82 (2007) 472.
data for o-H2 and p-H2 are substantially small in which the results [7] H. Yoshida, O. Kveton, J. Koonce, D. Holland, R. Haange, Fusion Eng. Des. 39–40
(875) (1998).
of the simulation can be well-founded. [8] K.M. Song, S.H. Sohn, K.S. Kim, Proceedings of the Korean Nuclear Society
Autumn Meeting, 1997, p. 343.
[9] D.H. Ahn, S.W. Paek, K.R. Kim, H.S. Jeong, H.J. Choi, J.K. Kim, H.S. Kang, H.S. Lee,
Conclusion W.S. Kim, K.M. Song, Proceedings of the Korean Nuclear Society Spring
Meeting, 2001.
[10] D.Y. Peng, D.B. Robinson, Ind. Eng. Chem. Fundam. 15 (8) (1976).
The property data for the 3 hydrogen isotopes, namely HT, DT,
[11] G. Soave, Chem. Eng. Sci. 27 (1972) 1197.
and T2, are not presently incorporated in the built-in database of [12] E.F. Hammel, J. Chem. Phys. 18 (1950) 228.
the compatible chemical process simulator. Therefore, the [13] P.C. Souers, Hydrogen Properties for Fusion Energy, University of California
thermodynamic properties of the 6 hydrogen isotopes, H2, HD, Press, Los Angeles, California, 1986.
[14] H.M. Mittelhauser, G. Thodos, Cryogenics 4 (1964) 368.
D2, HT, DT and T2, were estimated through vapor pressure [15] R.M. Bliesner, J.W. Leachman, P.M. Adam, J. Thermophys. Heat Transf. 28 (4)
prediction using Peng–Robinson equation of state with Twu alpha (2014) 717.
function. The estimated physical properties were evaluated [16] T.M. Flynn, Cryogenic Engineering: Revised and Expanded, second ed., Marcel
Dekker, New York, 2005.
through steady-state process simulation using Aspen Plus. The [17] A. Zuttel, A. Borgschulte, L. Schlapbach (Eds.), Hydrogen as a Future Energy
converged solution of the steady-state simulation using the Carrier, vol. 45, Wiley-VCH Verlag GmbH & Co. KGaA Weinheim, 2012.
estimated properties of hydrogen isotopes defines a possible [18] H.W. Wolley, R.B. Scott, F.G. Brickwedde, J. Res. Nat. Bur. Stand. 41 (1948) 379.
[20] M. Glugla, L. Doerr, R. Laesser, D.K. Murdoch, H. Yoshida, Fusion Eng. Des. 61–62
scenario for an actual ITER ISS process. (2002) 569.
Basic design data is gained from the process simulation in [21] I.R. Cristescu, M. Glugla, A. Antipenkov, S. Beloglazov, C. Caldwell-Nichols, I.
which the column internal design can be further specified using Cristescu, C. Day, D. Demange, A. Mack, Proceedings of the First IAEA Technical
meeting on First Generation of Fusion Power Plants—Design and Technology,
the column loading data to model the packing internals.
Vienna, Austria, 2005.
Additional research is being undertaken on the design of the [22] B.A. Younglove, Thermophysical Properties of Fluids. I. Argon, Ethylene,
column internal packing. Moreover, analysis on the dynamic Parahydrogen, Nitrogen, Nitrogen Trifluoride, and Oxygen, vol. 11, American
Chemical Society and American Institute of Physics, New York, 1982.
simulation is being developed based on the steady state
[23] J.W. Leachman, R.T. Jacobsen, S.G. Penoncello, E.W. Lemmon, J. Phys. Chem. Ref.
simulation of the ITER ISS. Data 38 (3) (2009) 721.