Ammonia-Water mixture Property Code (AWProC) development,

verification and Kalina cycle design for nuclear power plant

Mingjun Wang1,2, Annalisa Manera2*, Suizheng Qiu1*, G.H. Su1

(1. Department of Nuclear Science and Technology, Xi’an Jiaotong University, Xi’an 710049, China;
2. Department of Nuclear Engineering and Radiological Sciences, University of Michigan, Ann Arbor, MI
48109, USA)
Prof. Manera: [email protected]; Prof. Qiu: [email protected]

Abstract: The improvement of thermodynamic efficiency of power plants is of great interest for
the whole energy industry. The use of Kalina cycle has a great potential to improve the thermal
efficiency of a nuclear power plant. This cycle uses a mixture of ammonia and water as working
fluid. In this paper, we discuss the development of an Ammonia-Water mixture Property Code
(AWProC). The estimation of the mixture properties are based on the Gibbs free energy
functions. The code is verified and validated against experimental data available in the literature
and REFPROP code. It is shown that AWProC can accurately estimate the thermodynamic
properties of ammonia-water mixtures over a wide range of conditions, including high
temperature and pressure region. The code is then used to investigate the feasibility of applying
the Kalina cycle to a typical Pressurizer Water Reactor (PWR) plant as an effective way to
improve the plant efficiency. The fundamental of Basic-Kalina (B-K) cycle is described in detail
firstly. Then, two modified configurations, Recuperation-Kalina (R-K) and Flash-Kalina (F-K)
cycles respectively, are proposed for a typical 1000 MWe PWR. The simulation results indicate
that the R-K type cycle can reach about 31.2% efficiency with simple equipment requirements,
while the F-K type cycle can reach efficiencies up to about 34.8%, but at the expenses of a
slightly more complex design. The present work demonstrates the applicability of the Kalina
cycle as a way to improve the thermal efficiency of a nuclear power plant. This concept is
meaningful for improving nuclear power plants economic and competitiveness.

Key words: Ammonia-water mixture, Property code, Kalina cycle design, Efficiency calculation

1
© 2016. This manuscript version is made available under the Elsevier user license
http://www.elsevier.com/open-access/userlicense/1.0/
 NOMENCLATURE

Variables
v Specific volume
Cp Specific heat at constant pressure
G Gibbs free energy function
R Universal gas constant
P Pressure
T Temperature
h Enthalpy
s Entropy
x Ammonia mass fraction
A, B, C, D, E, m, n, a Coefficients

Scripts
l Liquid phase
g Vapor phase
o Ideal gas state
r Ratio between actual value and the reference point value
E Gibbs excess energy
I Indicator
B Reference state

1. Introduction
The thermodynamic cycle employed for the conversion of thermal energy into electricity is an
important design aspect of a power plant. Achieving high thermodynamic efficiency is very
beneficial, since it implies either an increase of the electric power for a given reactor core
thermal power, or a reduction of the total thermal power to be produced in the reactor core for a
given output electric power. In the latter case, a lower decay heat would be produced with a
consequent smaller load on the safety systems. The environmental footprint associated to the
plant is another significant reason to reduce the total thermal power. Current Light Water reactor
(LWR) designs utilize the Rankine cycle for the conversion of thermal power. A novel
thermodynamic cycle, Kalina, has been recently proposed and successfully employed in
geothermal plants [1, 2, 3] to achieve very high (up to 50%) thermal efficiencies. Current
applications are characterized by low operating pressures and temperatures, and by relatively
small power outputs. The working fluid of Kalina cycle is a binary mixture of water-ammonia.
Generally, binary mixtures boil and condense at increasing and decreasing temperatures
respectively, while pure component fluids (e.g. pure water) boil and condense at constant

2
temperatures [4]. So the advantage of the Kalina cycle is that the heat transfer takes place at a
variable temperature because the mixture bubble temperature and dew temperature depend on the
fraction of ammonia within the mixture, as shown in Figure 1. The reason for the higher
efficiency of the Kalina cycle with respect to a Rankine cycle can be easily understood by
comparing the two cycles in a T-S diagram, as illustrated in Figure 2 and Figure 3. Obviously,
the Kalina cycle is more efficient than Rankine cycle when the same boundary conditions are
applied. This is because for a binary mixture boiling and condensation occur at a variable
temperature.

Fig. 1 Variation of bubble temperature and dew temperature with the ammonia mass fraction

Fig. 2 Ammonia-water mixture working T-S diagram

3
 Fig. 3 T-S diagram comparison of Kalina cycle and Rankine cycle [5]

In order to analyze the efficiency of the Kalina cycle, it is necessary to estimate the
thermodynamic properties of the ammonia-water mixture at different operating conditions.
Several correlations for the ammonia-water mixture properties have been developed and
published in the literature. The theoretical framework at the basis of the available correlations
can be classified into nine groups: cubic equations of state [6], virial equations of state [7], Gibbs
excess energy [8], corresponding states method [9], perturbation theory [10], group contribution
theory [11], Leung-Griffiths model [12], Helmoltz free energy [13] and polynomial functions
[14]. Ziegler and Trepp [15] described an equation for the thermodynamic properties of
ammonia-water mixture in absorption units. In their work, the Gibbs excess energy equation was
utilized for determining the specific enthalpy, specific entropy and specific volume. They
developed the properties up to a pressure of 50 bars and temperature of 500 K.

Also, for the Kalina cycle, several research efforts have been dedicated to its application. The
original design concept known as Kalina Cycle System 1 (KCS 1) was published in 1984. The
original Kalina concept was then modified in order to better match the working fluid and the
heat-source heat-temperature curves in the boiler. This variant, known as KCS 6, achieves a 10%
efficiency improvement over the initial KCS 1. KCS 1 is preferable for small units (below 20
MW total power output; about 8 MW as bottoming cycle), while the more sophisticated KCS 6
variant is preferable for larger units [16]. Hettiarachchi et al. [17] analyzed the performance of
the Kalina cycle system 11 (KCS 11) with low-temperature geothermal heat sources, and studied
the effect of the ammonia fraction and turbine inlet pressure on the overall cycle performance.

4
The KCS 11 design has recuperator in the turbine exhaust steam before flowing to condenser.
They found that, for a given turbine inlet pressure, an optimum ammonia fraction exists, which
yields the maximum cycle efficiency.

Later a Kalina cycle with ejector (E-Kalina cycle) was proposed by Xinguo and collaborators
[18]. In the E-Kalina cycle, an ejector is used to replace the throttle valve and the absorber used
in the Kalina cycle system 11 (KCS 11). Results show that the net power output and thermal
efficiency of the EKalina cycle are higher than that of the KCS 11. Zhang et al. [19] presented a
detailed review of the research performed on the Kalina cycle, including the different variants of
the Kalina cycle, the comparison with the Rankine cycle, and energy and exergy analyses.
Rogdakis and Antonopoulos [20] proposed a triple stage power cycle which is similar to the
Kalina cycle. However, they replaced the distillation condensation of the Kalina cycle with an
absorption condensation process. Kouremenos et al. [21] analyzed this absorption type of power
cycle as a bottoming cycle with a gas turbine topping cycle. Since this cycle still uses water-
ammonia vapor mixtures going through the turbine, the exit temperature must be relatively high
in order to avoid condensation in the turbine.

Above all, for ammonia water mixture property, most papers describe it in a limited pressure
range and generally at relatively low pressures. However, a very broad range of pressures is
required, since we want to apply Kalina to a PWR. And the most previous Kalina studies were
focused on the low powers and low pressures and the work in this paper investigates the
feasibility of the Kalina cycles at high powers, pressures, and temperatures typical for nuclear
power plants. Based on the nuclear power plant development requirements, an approach based
on the Gibbs free energy functions is described in this paper. The equations reported in chapter 2
are then implemented in the AWProC (Ammonia-Water mixture Property Code) in-house code.
In chapter 3 comparisons between the code calculations and published experimental data and
REFPROP code calculations [22] are presented. Then different Kalina cycle configurations,
including basic and modified versions, are proposed and their applicability to a typical 1000
MWe PWR plant is investigated. The calculations of the cycle thermal efficiency of proposed
cycle types are carried out using AWProC in chapter 4.

5
2. AWProC development
2.1 Gibbs Free Energy Function
The Gibbs free energy equation for a pure component is shown as:
T P T
G  h0  Ts0   C p dT   vdP  T  (C p / T )dT (1)
T0 P0 T0

Based on the empirical relations of v and Cp, the following equations can be derived for the
liquid phase and vapor phase, respectively.
For the liquid phase:
GrL  hrL,0  Tr srL,0  B1 (Tr -Tr，0 )+(B2 / 2) Tr2  Tr2,0  
 T 
(B3 / 3) Tr3  Tr3,0   B1Tr ln  r  
T 
 r ,0  (2)
B2Tr (Tr -Tr，0 )  (B3 / 2)Tr T  T r
2 2
r ,0 
A  AT
1 3 r  A4Tr2   Pr  Pr ,0    A2 / 2   Pr2  Pr2,0 

For the vapor phase:

Grg  hrg,0  Tr srg,0  D1 (Tr -Tr，0 )+(D2 / 2) Tr2  Tr2,0  
T 
(D3 / 3) Tr3  Tr3,0   D1Tr ln  r   D2Tr (Tr -Tr，0 ) 
T 
 r ,0 
 P 
(D3 / 2)Tr Tr2  Tr2,0   Tr ln  r   C1  Pr  Pr ,0  
P  (3)
 r ,0 
P P TP   P P TP 
C2  r3  4 r3,0  3 r 4r ,0   C3  11r  12 r11,0  11 r 12r ,0  
T Tr ,0 Tr ,0    Tr ,0 
 r  Tr Tr ,0

 C4   Pr Pr3,0 Pr3,0Tr 
3

   11  12  11 
 3   Tr Tr11,0 Tr12,0 

Where the subscript o means the ideal gas state, the Tr, Pr and Gr are the ratio between a
parameter actual value and the reference point value, and are expressed as:
Tr  T / TB (4)
Pr  P / PB (5)
Gr  G / RTB (6)
The reference values for the above equations are R=8.314 J/mol·K, TB=100 K and PB= 10 bar.
The coefficients in the equations are listed in Table 1 [23].
Table 1 Coefficients for the equations of the pure components

6
 Coefficient Ammonia Water
A1 3.971432e-2 2.748796e-2
A2 -1.790557e-5 -1.016665e-5
A3 -1.308905e-2 -4.452025e-3
A4 3.752836e-3 8.389246e-4
B1 1.634519e1 1.214557e1
B2 -6.508119 -1.898065
B3 1.448937 2.911966e-1
C1 -1.049377e-2 2.136131e-2
C2 -8.288224 -3.169291e1
C3 -6.647257e2 -4.634611e4
C4 -3.045352e3 0.0
D1 3.673647 4.019170
D2 9.989629e-2 -5.175550e-2
D3 3.617622e-2 1.951939e-2
hl 4.878573 21.821141
hg 26.468879 60.965058
sl 1.644773 5.733498
sg 8.339026 13.453430
T0 3.2252 5.0705
P0 2.000 3.000

The pure component molar specific enthalpy, entropy and volume equations can be derived from
the general Gibbs free energy functions of the pure component, as:

h   RTBTr2 [ (Gr / Tr )]Pr (7)
Tr
Gr
s   R[ ]P (8)
Tr r

RTB Gr
v [ ]T (9)
PB Tr r

The derivation of the ammonia-water mixture properties on the basis of the Gibbs free energy
functions is described in the following chapters.

2.2 Ammonia Water Mixture Property Calculation

The mixture properties equations of vapor phase and liquid phase derived from the pure
component property functions are:
For the vapor phase,
hmg  xg hag  1  xg  hwg (10)

7
 smg  xg sag  1  xg  swg  s mix (11)

vmg  xg vag  1  xg  vwg (12)

For the liquid phase,

hmL  x f haL  1  x f  hwL  h E (13)

smL  x f saL  1  x f  swL  s E  s mix (14)

vmL  x f vaL  1  x f  vwL  v E (15)

Where the hE, sE, vE and smix are derived from:


h E   RTBTr2 [ (G E r / Tr )]Pr ,x (16)
Tr

G E r
s   R[
E
]P (17)
Tr r ,x

RTB G E r
vE  [ ]T (18)
PB Tr r ,x

s mix   R  x f ln  x f   1  x f  ln 1  x f  (19)

The GrE equation is:

E  E P   E  E P T  E / T  E / T 2 
 1 2 r 3 4 r r 5 r 6 r

GrE     2 x  1  E4  E8 Pr   E9  E10 Pr  Tr  E11 / Tr  E12 / Tr2   1  x  x (20)
 
   2 x  12  E  E P  E / T  E / T 2  
 13 14 r 15 r 16 r 
The coefficients in the above equations are shown in Table 2 [23].

Table 2 Coefficients for the Gibbs excess energy function

Coefficients Value
E1 -4.626129e1
E2 2.060225e-2
E3 7.292369
E4 -1.032613e-2
E5 8.074842e1
E6 -8.461214e1
E7 2.452882e1
E8 9.598767e-3

8
 E9 -1.475383
E10 -5.038107e-3
E11 -9.640398e1
E12 1.226973e2
E13 -7.582637
E14 6.012445e-4
E15 5.487018e1
E16 -7.667596e1

In order to estimate the thermodynamic properties for the two phase mixture, it is necessary to
define the dryness first:
x  xl
dryness  (21)
xg  xl
Then the two phase mixture properties are calculated using the following equations,
htwo phase  hl  dryness  hg (22)

stwo phase  sl  dryness  sg (23)

vtwo phase  vl  dryness  vg (24)

At this point, the equations for the ammonia-water mixture properties from single liquid phase to
single vapor phase have been obtained. To complete the set of equations, it is necessary to have
expressions for the mixture bubble and dew point temperatures as function of pressure and
ammonia mass fraction. These equations are addressed in the following section.

2.3 Ammonia-water mixture bubble temperature and dew temperature

The ammonia-water mixture bubble temperature and dew temperature are calculated using the
empirical equations (25) and (26) developed by Patek and Klomfar [13]. These empirical
equations make the code avoid an iterative procedure.
ni
 p0 
T  p, x   T0  ai 1  x 
mi
ln p  (25)
i  
ni
 p0 
T  p, y   T0  ai 1  y 
mi /4
ln p  (26)
i  
The equations (25) and (26) are used to calculate the mixture bubble temperature and dew
temperature, respectively. In the above equations, the reference pressures P0 is equal to 2 MPa,

9
while the reference temperature T0 is 100 K. The coefficients used in the equations are shown in
Table 3 and Table 4.

Table 3 Exponents and coefficients of the bubble equation

i mi ni ai
1 0 0 0.322302e1
2 0 1 -0.384206
3 0 2 0.460965e-1
4 0 3 -0.378945e-2
5 0 4 0.135610e-3
6 1 0 0.487755
7 1 1 -0.120108
8 1 2 0.106154e-1
9 2 3 -0.533589e-3
10 4 0 0.785041e1
11 5 0 -0.115941e2
12 5 1 -0.523150e-1
13 6 0 0.489596e1
14 13 1 0.421059e-1

Table 4 Exponents and coefficients of the dew equation

i mi ni ai
1 0 0 0.324004e1
2 0 1 -0.395920
3 0 2 0.435624e-1
4 0 3 -0.218943e-2
5 1 0 -0.143526e1
6 1 1 0.105256e1
7 1 2 -0.719281e-1
8 2 0 0.122362e2
9 2 1 -0.224368e1
10 3 0 -0.201780e2
11 3 1 0.110834e1
12 4 0 0.145399e2
13 4 2 0.644312
14 5 0 -0.221246e1
15 5 2 -0.756266
16 6 0 -0.135529e1
17 7 2 0.183541

10
The equations described in this and previous chapters have then implemented in the AWProC
code. A flow chart of the code is presented in Figure 4.
Start

Mole fraction Mass fraction

Ammonia fraction type

x=xw/17/(xw/17+
(1-xw)*18)

State
Certification
Unknown Two Phase

Input Selection
Factor K
Input T, x, P
K=1 K=2
K=3 K=4 K=5

X, T-> Y, T- X, P-> Y, P-> T, P->

Y, P >X, P Y, T X, T X, Y
Tbuble=f(P, x)
Tdew=f(P, x)

Single
Yes Liquid
T<Tbuble End
Phase,vl,
hl, sl

No

Single
No Vapor
Tbuble<T<Tdew End
Phase,vg,
hg, sg

Yes

x0=f(P, T)
y0=f(P, T)

Dryness=f(x0, y0, x)

vl, hl, sl and vg, hg, sg

vmix, hmix and smix

End

Fig. 4 Flow chart of AWProC

11
The AWProC is developed using FORTRAN language and it can be used to calculate ammonia-
water mixture property continuously from single phase liquid to single phase vapor. The bubble
temperature and dew temperature can be calculated as well In addition, when the mixture is in
two phase state, the different ammonia mass fraction in the liquid phase and vapor phase can be
obtained at the given conditions.

3. Code verification and validation

The AWProC code has been verified and validated by comparing the code results against
published data, both experimental and computational.

3.1 Bubble Temperature and Dew Temperature

The bubble temperature and dew temperature calculated using AWProC are compared with the
IGT experimental data [24], and presented in Figure 5. As shown in the figure, the code
calculation results are in very good agreement with the IGT data. The bubble point and dew
point temperatures decrease with increasing ammonia mass fraction. When the ammonia mass
fraction is not equal to 0.0 or 1.0 (pure water or pure ammonia respectively), the mixture bubble
temperature and dew temperature are different, leading to different ammonia mass fractions in
vapor and liquid phase respectively when the mixture is in two phase. It is clear to see the
difference between the bubble temperature and dew temperature, which is a main factor leading
to a high efficiency for Kalina cycle.

Fig. 5 Bubble and dew temperatures at 3.447 MPa vs. ammonia mass fraction

12
3.2 Specific Enthalpy and Entropy
The specific enthalpy and entropy are significant and necessary parameters to perform the
thermodynamic cycle calculation. Figures 6 (a) and (b) show the comparisons between AWProC
calculated results and the IGT of the saturated liquid at 3.447 MPa and vapor enthalpy at 1.8
MPa. As shown in the figures, the code calculation results in this work agree with the
experimental data very well. It demonstrates that the ammonia-water mixture specific enthalpy
can be simulated accurately using AWProC. Also, seen from the figures, the mixture liquid
specific enthalpy declines with the ammonia mass fraction and it reaches a lowest value when
the ammonia mass fraction is between 0.6 and 0.7. Then the specific enthalpy begins to increase
until the ammonia mass fraction reaches 1.0. The vapor specific enthalpy decreases with an
increase of the ammonia mass fraction as shown in the Figure 6 (b).

(a) Saturated liquid at 3.447 MPa (b) Saturated vapor at 1.8 MPa
Fig. 6 Specific enthalpy vs. ammonia mass fraction

The saturated liquid and vapor entropy of the AWProC calculation results is compared with the
data from [25], [26] and [25] respectively, which are shown in Figure 7 (a) and Figure 7 (b). The
code calculation results in this work are in good agreement with the experimental data when the
saturated liquid is at 3.447 MPa and the saturated vapor is at 1.8 MPa. However, the entropy
calculation results between different researchers are different and at an acceptable range. As can
be seen in the figures, the mixture liquid specific enthalpy declines with the ammonia mass
fraction and it reaches a lowest value when the ammonia mass fraction is around 0.5 and then the
specific enthalpy begins to increase until the ammonia mass fraction reaches 1.0. In comparison,
the vapor specific enthalpy decreases with an increase of the ammonia mass fraction as shown in
the Figure 7 (b).

13
(a) Saturated liquid at 3.447 MPa (b) Saturated vapor at 1.8 MPa
Fig. 7 Specific entropy vs. ammonia mass fraction

(a) Specific enthalpy (b) Specific entropy

Fig. 8 Properties comparison at 570 K vs. pressure

(a) Specific enthalpy (b) Specific entropy

Fig. 9 Properties comparison at 15 MPa vs. temperature

14
 (a) Specific enthalpy (b) Specific entropy
Fig. 10 Properties comparison at 15 MPa and 570 K vs. ammonia mole fraction

The property estimation comparisons between AWProC and REFPROP code are also studied at
the high temperature and pressure region in this section. Figure 8 shows the results comparison
between AWProC and REFPROP code at 570 K with an increase of pressure from 10 MPa to
16MPa. The specific enthalpy and entropy variations with temperature at 15 MPa are shown in
Figure 9. Figure 10 illustrates the specific enthalpy and entropy variations with ammonia mole
fraction at the specific state, 15 MPa and 570 K. The results show that AWProC estimation
results are in good agreement with REFPROP code and the AWProC can be used to estimate the
ammonia water property at high temperature and high pressure region.

4. Kalina cycle design

4.1 Kalina cycle description
The simplified Kalina cycle is basically a ―modified‖ Rankine cycle which uses a mixture of
water-ammonia as working fluid. Its configuration is depicted in Figure 11. As shown in the
diagram, the secondary cycle is heated by the primary side of the steam generator (5, 6) through
the boiler. The ammonia-water mixture is heated to superheated vapor (1), and expands in the
turbine to generate electricity. The turbine exhaust (2) is cooled and condensed in the condenser
by cooling water (7, 8). Then the saturated liquid ammonia-water mixture (3) is compressed to a
high pressure through the pump (4). At last the mixture flows back to the boiler to remove heat
from the primary side of the steam generator. The most significant characteristic is the variation

15
of the boiling temperature and dew temperature and higher efficiency associated to absorbing
and rejecting heat.
Heat
inlet 1

5
Turbine

6 4 2

Heat
outlet Pump

absorber

8
7

Coolant Coolant
inlet outlet

Fig. 11 Basic Kalina (B-K) type cycle configuration

1

Turbine

Heat
inlet
2
Heat
7 exchanger

8 6 5

Heat
outlet
Pump

Condenser
4
9 10

Environment

16
 Fig. 12 Recuperation Kalina (R-K) type cycle configuration

To take advantage of the Kalina cycle, two modified configurations, Recuperation Kalina (R-K)
and Flash Kalina (F-K) types respectively, are proposed and applied to a typical 1000e MW
PWR plant. The two cycles are shown in Figure 12 and Figure 13 respectively.
1

Turbine

Heat
inlet
Pressure- 2
relief
valve

Flash 4
Tank

7
10
Condenser
9 8

Compressor 5
Heat
outlet 6
Pump

Fig. 13 Flash Kalina (F-K) type cycle configuration

The R-K type includes a heat exchanger, in which the single phase liquid is heated by the turbine
exhaust. The ammonia mole fraction of mixture used in this study is 0.8. The pressure is
15.5MPa and the corresponding temperature is 560.0 K. Through the turbine expansion, the
mixture pressure and temperature decrease and then flows to condenser. After pumping, the
mixture flows to the heat exchanger. Then the heated ammonia-water mixture goes back to the
boiler. The R-K type can utilize the exhaust heat and reduce the heat loss, leading to increase of
the cycle efficiency.

For the F-K type shown in Figure 13, an incubator, pressure-relief valve, flash tank and
compressor are adopted to further improve the cycle efficiency. The pressure-relief valve is

17
adopted in order to decrease the pressure of the mixture at the exit of the turbine to achieve two-
phase flows. The two phase ammonia-water mixture is then separated in the flash tank, in order
to get a mixture vapor rich in ammonia and a mixture liquid weak in ammonia. The mixture
liquid weak in ammonia is later mixed with the turbine exhaust, decreasing the turbine back
pressure for a higher efficiency. In order to keep the mixture flowing through the pressure-relief
valve at a constant temperature, a heat insulated box is designed to utilize the compressor heat
release.

4.2 Efficiency Analysis

In this section, a three-loop 1000MWe PWR is taken as reference plant. The heat transferred
from the primary to the secondary side in each loop is calculated based on the energy
conservation reported in equations (27) and (28). The boundary and initial conditions assumed
for the Kalina cycle are listed in Table 5. The in-house code AWProC is used to calculate the
ammonia-water mixture properties in every node state of the Kalina cycle.

Q12  c p1T1m1 (27)

h2  Q12 / m2 (28)

Table 5 Boundary conditions for the Kalina cycle

Parameters Value
Primary side flowrate 23790 m3/h
Primary side inlet temperature 327 ℃
Primary side outlet temperature 293 ℃
Primary side pressure 15.5 MPa
Primary side coolant specific volume 0.00144 m3/kg
Secondary side pressure 6.89 MPa
Secondary side flowrate 400 kg/s

Based on the first law of thermodynamics, the thermal efficiency of the cycle is defined as:
Wnet
 (29)
Qin
where the Qin is the heat gain in the boiler and Wnet is the work output of the cycle. Based on the
thermal balance, the different node state mixture properties of B-K, R-K and F-K cycles are
calculated. The results are shown in Table 6, 7 and 8, respectively.

18
 Table 6 Node state of working medium in B-K configuration

Node P T Ammonia mole h (kJ/kg) s

(MPa) (K) fraction x (kJ/kg·K)

1 15.5 560.0 0.8 1960.035 4.81

2 0.4 355.0 0.8 1353 4.85
3 0.4 278.15 0.8 -104.7 4.1
4 15.5 278.15 0.8 -92.448 0.99

Table 7 Node state of working medium in R-K configuration

Node P T Ammonia h (kJ/kg) s

(MPa) (K) mole fraction x (kJ/kg·K)
1 15.5 560.0 0.8 1960.035 4.81
2 0.4 355.0 0.8 1353 4.85
3 0.4 330 0.8 956.1 3.69
4 0.4 278.15 0.8 -104.7 4.1
5 15.5 278.15 0.8 -92.448 0.011
6 15.5 350.0 0.8 213.05 0.99

Table 8 Node state of working medium in F-K configuration

Node P T Ammonia mole h (kJ/kg) s

(MPa) (K) fraction x (kJ/kg·K)
1 15.5 575.0 0.81 2009.419 4.903
2 0.2 335.0 0.81 1279.6 4.926
3 0.2 332.66 0.49 581.6 2.49
4 0.08 330.0 0.49 912.1 3.68
5 0.08 330.0 0.13 147.74 0.79
6 0.2 330.0 0.13 147.83 0.79
7 0.08 330.0 0.81 1654.38 6.45
8 1.8 330.0 0.81 781.1 2.88
9 1.8 328.0 0.81 117.13 0.75
10 15.5 328.0 0.81 121.74 0.75

For the B-K and R-K types, the temperatures of node 1 could not increase to a higher value due
to the turbine isentropic increase process limitation. Also, the pressures of node 2 could not be
lower because the environmental temperature limit. The lower pressure in R-K type with the
ammonia mole fraction 0.8 makes the mixture unrealistic condensation. However, for the F-K
type, because of the lower ammonia mole fraction with the lean ammonia mixture flowing back,

19
the mixture can condense in lower pressure condition. Then from the node state parameters from
above tables, the B-K, R-K and F-K type cycles efficiencies are obtained based on the first law
of thermodynamics. The turbine efficiency is assumed as 90% in this study to make it more
realistic. Then he computed efficiencies are 26.6%, 31.2% and 34.8% respectively. As shown in
tables, unpractical low temperatures are required in the first two cycle condensers. Therefore, the
F-K type has therefore the potential to improve the thermal efficiency of a nuclear power plant.

5. Conclusion
The thermal efficiency is a significant factor to be considered during the design of a nuclear
power plant, in order to enhance the competitiveness of nuclear energy.

In this work, the AWProC, which is derived from the Gibbs free energy functions and can be
used to calculate the ammonia-water mixture properties in both single phase and two phases, is
developed in this work. And through comparison with experimental data published in the
literature and the REFPROP code, it demonstrates that the AWProC is available and can be used
in future ammonia water mixture property calculation at a broad range of pressures and
temperatures.

Also, it is demonstrated that the innovative concept of a Kalina cycle can be applied to the
operating conditions of a typical light water reactor in order to increase the overall plant
thermodynamic efficiency.

In particular three variants of the Kalina cycles are analyzed (B-K, R-K and F-K types
respectively) and applied to a typical 1000MWe PWR plants. A thermal efficiency up to 34.8%
has been achieved. Even though additional studies are necessary to further increase the cycle
efficiency and make Kalina cycle apply in the PWR plants, including the ammonia water mixture
heat transfer and flow characteristic, HX heat transfer and turbine expansion process, the present
study has shown that the Kalina has very good potentials for the nuclear industry.

Acknowledgement

20
This research has been supported by the Chinese Scholarship Council and by the DOE Office of
Nuclear Energy’s Nuclear Energy University Programs (NEUP).

References
[1] Kalina AI. Generation of energy by means of a working fluid, and regenera- tion of a
working fluid. United States Patent 4346561. Filed date: Aug. 31, 1982.
[2] Kalina AI. Combined cycle and waste heat recovery power systems based on a novel
thermodynamic energy cycle utilizing low-temperature heat for power generation. American
Society of Mechanical Engineers, 1983, New York, Paper No. 83-JPGC-GT-3, 1–5.
[3] Kalina AI. Combined-cycle system with novel bottoming cycle. ASME Journal of
Engineering for Gas Turbines and Power 1984; 106: 737-742.
[4] Mlcak HA. An introduction to the Kalina Cycle. PWR-Vol. 30, Proceedings of the
International Joint Power Generation Conference, 1996, Houston, Texas.
[5 ]Yutaka Takeuchi, Hirozo Shiomi, Nobuhiko Miyamae, Miki Hiramatsu, Mistuto Montani
and Iwao Ohshima. Feasibility study for application of mixture working fluid cycle to nuclear
reactor power plant. 7th International Conference on Nuclear Engineering. Tokyo, Japan, April
19-23, 1999: 1-8.
[6] Zhao E, Sugie H, Lu BCY. Calculation of vapor–liquid equilibria and saturated liquid
volumes for water–ammonia mixtures. Chemical Engineering Com- munication 1994; 129: 99-
108.
[7] Duan Z, Møller N, Weare JH. Equation of state for the NH3–H2O system. Journal of
Solution Chemistry 1996; 25: 43-50.
[8] Xu F, Goswami YD. Thermodynamic properties of ammonia–water mixtures for power-cycle
applications. Energy 1999; 24(6): 525-536.
[9] Nowarski A, Friend DG. Application of the extended corresponding state method to the
calculation of the ammonia–water mixture thermodynamic surface. International Journal of
Thermophysics 1998; 19(4): 1133-1142.
[10] Abovsky V. Thermodynamics of ammonia–water mixture. Fluid Phase Equilibria 1996;
116: 170-176.
[11] Enick RM, Mcllvried HG, Gale TK, Klara JM. The modeling of LEBS Kalina power cycles.
Proceedings of Joint Power Generation conference, ASME 1997; 2: 55-67.

21
[12] Rainwater JC, Tillner-Roth R. Critical region vapor–liquid equilibrium model of ammonia–
water.13th International Conference on the Properties of Water and Steam, 1999, Toronto,
Canada.
[13] Tillner-Roth R, Friend DGA. Helmholtz free energy formulation of the thermodynamic
properties of the mixture (water+ammonia). Journal of Physical and Chemical Reference Data
1998; 27(1b): 63-69.
[14] Pate`k J, Klomfar J. Simple functions for fast calculations of selected thermodynamic
properties of the ammonia–water system. International Journal of Refrigeration. 1995; 18(4):
228-234.
[15] Ziegler B, Trepp CH (1984). Equation of state for ammonia-water mixtures, Refrig., 7: 101-
106.
[16] Kalina AI. Novel power cycle for combined-cycle system and utility power plants(ESL-IE-
86-06-39). Proceedings from the eighth annual industrial energy technology conference,
Houston, TX, June 17-19, 1986.
[17] Hettiarachchi HDM, Golubovic M, Worek WM. The performance of the Kalina cycle
system 11 (KCS-11) with low-temperature heat sources. J Energy Resour Technol 2007; 129(3):
243-247.
[18] X. Li, Q. Zhang, X. Li. A Kalina cycle with ejector. Energy, 2013, 54: 212-219
[19] Zhang XX, He MG, Zhang Y. A review of research on the Kalina cycle. Renew Sust Energ
Rev 2012; 16(7): 5309-5318.
[20] Rogdakis ED, Antonopoulos KA. A high efficiency NH3/H2O absorption power cycle.
Heat Recovery Systems 1991; II: 263-275.
[21] Koremenos DA, Rogdakis ED, Antonopoulos KA. Cogeneration with combined gas and
aqua-ammonia absorption cycles. In: Krane RJ, editor. Thermodynamics and The Design
Analysis, and Improvement of Energy Systems, vol. 33. New York: ASME AES, 1994: 231-238.
[22] Eric W. Lemmon, Mark O. McLinden, Marcia L. Huber. NIST Reference Fluid
Thermodynamic and Transport Properties—REFPROP. NIST Standard Reference Database 23.
Aug. 2002.
[23] N. Shankar Ganesh, T. Srinivas. Evaluation of thermodynamic properties of ammonia-water
mixture up to 100 bar for power application systems. Journal of Mechanical Engineering
Research 2011; 3(1): 25-39.

22
[24] Macriss, R.A., Eakine, B.E., Ellington, R.T. and Huebler J. 1964. Physical and
thermodynamic properties of ammonia-water mixtures. Research Bulletin No. 34. Institute of
Gas Technology, Chicago (IL), USA.
[25] Scatchard, G., Epstein, L.F., Warburton, J. and Cody, P.J. 1947. Thermodynamic properties
saturated liquid and vapor of ammonia-water mixtures, Refrigeration Engineering. Vol. 53: 413-
452.
[26] Park YM, Sonntag RE. ASHRAE Trans 1992; 97: 150.

23