Applied Thermal Engineering 31 (2011) 119e128

Contents lists available at ScienceDirect

Applied Thermal Engineering

journal homepage: www.elsevier.com/locate/apthermeng

Improvement of chemical control in the water-steam cycle of thermal power

plants
Vladana N. Rajakovi

c-Ognjanovi

c a, *, Dragana Z. Zivojinovic b
, Branimir N. Grgur b, Ljubinka V. Rajakovi

cb
a
Faculty of Civil Engineering, University of Belgrade, Bulevar Kralja Aleksandra73, Belgrade, Serbia
b
Faculty of Technology and Metallurgy, University of Belgrade, Karnegijeva 4, Belgrade, Serbia

a r t i c l e i n f o a b s t r a c t

Article history: A more effective chemical control in the water-steam cycle (WSC) of thermal power plants (TPP) is
Received 26 July 2010 proposed in this paper. Minimization of corrosion effects by the production of ultra pure water and its
Accepted 30 August 2010 strict control is the basis of all the investigated processes. The research involved the analysis of water
samples in the WSC through key water quality parameters and by the most convenient analytical tools.
The necessity for the stricter chemical control is demonstrated through a concrete example of the TPP
Keywords:
Nikola Tesla, Serbia. After a thorough analysis of the chemical control system of the WSC, diagnostic and
Chemical control
control parameters were chosen for continuous systematic measurements. Sodium and chloride ions
Analytical measurements
Thermal power plant
were recognized as the ions which indicate the corrosion potential of the water and give insight into the
Water-steam cycle proper production and maintenance of water within the WSC. Chemical transformations of crucial
corrosion elements, iron and silica, were considered and related to their quantitative values.
Ó 2010 Elsevier Ltd. All rights reserved.

1. Introduction control in TPPs has been studied partially in some papers. In

Chemistry and energy sciences pervade many topics, nowadays
they are invited to give new contributions in the task of energy
efficiency, water quality and rationalization of water consumption.
The efficient production of electric energy, the effective produc-
tion of ultra pure water and the minimization of corrosion effects
are at least three aspects which can not be considered without the
expertise of analytical chemists [1]. Assembling the scientific and
professional work which in the last ten years was devoted to the
quality of the analytical tools and methods applied in thermal
power plants (TPP) gave rise to the results presented in this work.
In order to maintain high reliability of TPPs and to realize their
expected lifetime, high water quality in all the water systems
within a TPP have to be achieved which was elaborated in
applicative guidelines and scientific reports [2e9]. The main water
systems in TPPs for steam generation can be recognized as: raw
water, demineralised (demi) water, condensate, boiler feed water
and steam. The measurement of impurities throughout the power
generation process can provide valuable information regarding the
source of contamination, the rates of contaminant build-up and
the probable rates of corrosion, as well as the data collected
during the start-up and shut-down of power plants. Chemical
advising operators some useful tools were proposed for the
identification of a chemical problem, which was particularly
valuable when laboratory staff is not available at the power plant
[10]. In some papers wastewater reuse was proposed as potential
improvement in a fossil unit [11]. A very important issue con-
cerning energy policy [12] and large water consumption [13] was
also considered. The origin, composition and rate of built-up were
analyzed [14] and a review of the common corrosion phenomena
were described taking in account the predominant corrosion
mechanisms in high temperature and supercritical water [15]. The
research about long-term impact of common ions (especially
sodium and chlorine ions) in a real thermal power plant which
demonstrate the corrosion effect in wateresteam system are still
neglected.
The presence of corrosive ionic species in water/steam even at
low mg L
1 levels can make the components of a TPP susceptible to
corrosion [6,8,16]. The production of electrical energy, mechanical
problems and economy are usually the main concern in TPPs.
However, the water chemical control is incorporated in and has an
influence on all these aspects; it is also a high priority issue.
2. Chemical control in wateresteam cycle (WSC)

A chemical water treatment plant (CWTP) for demi water

* Corresponding author. production and a polishing plant (PP) for condensate purification
E-mail address: vladana@grf.rs (V.N. Rajakovi

c-Ognjanovi

c). are the systems for ultra pure water production. The PP is a system

1359-4311/$ e see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.applthermaleng.2010.08.028
120 V.N. Rajakovi
c-Ognjanovi
c et al. / Applied Thermal Engineering 31 (2011) 119e128
for the make-up of feed water and it is a part of a WSC, as presented
in Fig. 1.
The WSC consists of: a PP, a feed water tank (FWT), a boiler (B),
a turbine (T) and a condenser (C). Conditions within the WSC are
extreme, with high pressures, high temperatures and various other
features of steam. Due to the loss of the condensate (caused by
evaporation or leakage) in the WSC, demineralised water from the
FWT is continuously added. Deaerated water in the FWT is the
source for steam generation, produced in boiler. Steam, with an
adequate temperature and pressure, enters the turbine. A more
detailed scheme of the WSC in the TPP Nikola Tesla, which was the
subject of this work, is presented in Fig. 2. In addition to the main
parts of the WSC, which are explained in the legend, the sampling
points, which were the basis for the experimental work, are also
indicated and listed in the legend.
Water for the WSC is provided either from groundwater sources
(wells) or surface water (rivers). Water is chemically prepared in
the CWTP. Ion-exchange (IE) is no longer the only reliable method
for the production of high-purity water. Other techniques, such as
reverse osmosis (RO) and electrodialysis, are available for this
process. Often a combination of these techniques, such as RO and IE,
may be the most economic arrangement [17]. The final result
should be ultra pure water convenient for the WSC.
The path and conditions of the water and steam in a WSC could
be explained in a simplified way using Fig. 2 and the Tes diagram
for steam transformation in the WSC of the TPP Nikola Tesla B,
which is presented in Fig. 3.
Demineralised water from the CWTP, which is the starting point
in Fig. 2, goes to the DWT situated in the main building (MB). Water
from the DWT is used as a substitute for the loss of condensate in
the system and together with condensate is treated in the PP.
The make-up water polished with ammonia (NH3) and hydra-
zine (N2H4) goes to the LPH and FWT, where the water tempera-
ture, t, is 180  C and pressure, p, is 12 bar. In the FWT, the
deaeration process is accomplished. After a second polishing with
ammonia and hydrazine, the feed water goes to the FWP. The
further path of fluid can be described with the process conditions
on the Tes diagram presented in Fig. 3. After the FWP (the line
STEAM
B cost. Action level 1 implies minor disturbances requiring investi-
T
UNREGULATED
STEAM
REDUCTION
FEED WATER
C bances requiring substantial interventions, such as load reduction,
COOLING WATER
DEMI WATER
CONDENSATE
FWT PP
Legend
Chemical Water Treatment Plant CWTP
Polishing Plant
Feed water tank FWT
Boiler
Turbine
Condenser
Fig. 1. Water Steam Cycle (WSC).
between 5 and 6 in Fig. 3.), the pressure reaches 250 bar. The water
goes to the HPH (6e7 in Fig. 3), where it is heated to a temperature
of 250  C. From the HPH, the water goes to B and to Ec, where it is
heated to 330  C, close to the boiling temperature. From EC, the
water goes to EV, which is used for the transformation of water from
liquid to vapour state (point 7 in Fig. 3). From EV, the wateresteam
goes to SS, where isentropic expansion of the steam occurs and
water is separated from steam. The steam goes to SH (8e1 in Fig. 3)
and to HPT (1e2 in Fig. 3). Steam after HPT is reheated to
t ¼ 543  5  C, p ¼ 37 bar (2e3 in Fig. 3) and it goes to MPT, where it
expands (3e4 in Fig. 3) to t ¼ 256  C, p ¼ 3.6 bar. Steam continues to
expand to LPT (up to point 4 in Fig. 3) as wet steam. In LTP, the
steam expands up to t ¼ 34  C, p ¼ 0.05 bar and it goes to condenser
(4e5 in Fig. 3).
In order to monitor the quality of ultra pure water, two groups of
parameters were introduced: control and diagnostic parameters.
The control parameters are: pH, conductivity, oxygen, sodium,
chloride and silica. The diagnostic parameters are: iron, copper,
organic matter and oil (traditionally water hardness is also
controlled). These water quality parameters need to be measured
precisely and quickly in order to estimate the water quality at
critical working and operational points.
The main characteristics and the reason for their measurements
as control and diagnostic parameters of wateresteam quality in the
WSC are highlighted in Tables 1 and 2.
In Table 3 the characteristic and limiting values for the control
and diagnostic parameters for the proper functioning of the WSC
are listed in Table 3, together with sensitive analytical tools which
could be applied for the measurements. Control parameters are
measured on line, continuously (C), and they give information on
the current status of the WSC. Diagnostic parameters are measured
daily (D) and weekly (W) and they give wide and basic information
on the water quality in all segments of the cycle. Analytical tools
and techniques for the measurement of the control and diagnostic
parameters are on line analyzers and/or sophisticated instruments
of high precision and sensitivity.
The water and steam are of high purity but in the case of failures,
a TPP should operate with adequate actions and improvement of
water quality. Usually three levels of action are defined based on
the experience and failures resulting from inadequate chemical
control. A target range (TR) and three levels of action are usually
defined as guidelines in TPP [2]. The TR covers the optimal values
which plant managers will normally achieve without excessive
gations, diagnosis and optimization, and correction within two
weeks. Action level 2 implies serious disturbances in chemical
control requiring not only diagnosis but also action to eliminate the
cause within two days. Action level 3 implies very serious distur-
within 8 h, or plant shut-down.
Impurities in water and steam are closely related to deposits
CWTP on tubes and all segments of a WSC. Deposits not only impair heat
transfer, they also increase the risk of corrosion processes. Puri-
fication of water and consumption and the wastage corrosion
mechanism in the WSC provide an incentive to improve water
chemistry control, which has to be in balance with mechanical
and electrical operating conditions [13]. Chemical control is a very
important factor for the daily operation and for the future life of
PP a power plant. Chemical control should guarantee the maximum
purity of the water. Any deviation of a parameter must be
B detected and explained, in order to prevent or to correct its cau-
T ses. Although in the last decades chemical control has received
C certain attention [18], there is still a need to monitor a set of
parameters reflecting the chemical status of the water at key
sampling points.
 V.N. Rajakovi
c-Ognjanovi
c et al. / Applied Thermal Engineering 31 (2011) 119e128 121

7 8 HPT
SS LPT
MPT
6 SH
9
Ev CW
C
10
5

Ec
FWT CP

11
RH DWT 1 CWTP
B

PP
LPH
HPH

4 3
FWP
NH3
NH3
N 2H 4
N2H4
Legend
The main parts of the WSC
Chemical water treatment plant CWTP Evaporator Ev
Demineralisation water tank DWT Steam separator SS
Condensate pump CP High pressure turbine HPT
Polishing plant PP Superheater SH
Low pressure heater LPH Reheater RH
Feed-water tank FWT Middle pressure turbine MPT
Feed-water pump FWP Low pressure turbine LPT
Boiler B Condenser C
High pressure heater HPH Cooling water CW
Economiser Ec
Samples of water/steam
Demineralised water 1 Superheated steam (67 m) 7
Condensate I 2 Superheated steam (19 m) 8
Condensate II 3 Wet steam (96.5 m) 9
Feed water 4 Reheated steam (72.5) 10
Steam, inlet to evaporator (19.5) 5 Reheated steam (19.0 m) 11
Saturated steam (103.8 m) 6
Fig. 2. A scheme of the WSC of the TPP Nikola Tesla.

3. Experimental analysis instruments. For pH measurements, two types of instruments were

The object of experimental analysis in this work were water and
steam samples of the WSC in TPP Nikola Tesla, the main parts of
which are presented in Fig. 2.
The experiments were focused on:
I. The chemical analysis of a set of parameters reflecting the
chemical status of the water at key sampling points of the
WSC in order to monitor the quality of the water and steam,
II A comprehensive analysis of the results of the chemical control
in order to make an expert appraisal with the priority to
explain, prevent or correct any causes of deviations in the WSC.
3.1. Instrumentation
Control and diagnostic parameters of the water quality in
a TPP demands the employment of highly sensitive and reliable
used: an on line (Polymetron 8271) and a laboratory pH-meter
(Metrohm 827 pH Lab). The conductivity was measured with an on
line (Polymetron 8246) and a laboratory (HQ 40d multi HACH)
conductometer. Dissolved oxygen was measured with an on line
(T.Dr.Thiedig K2-DigoxK2) and a laboratory (HQ 40d multi HACH)
electrode. The contents of iron, copper and silicon were determined
by induced coupled plasma with mass spectrometry (ICP-MS
7500 ce, Agilent Technologies 7500), according to standard methods
with a sensitivity up to 0.001 mg L
1 [19]. Standard solutions were
prepared with a Fluka multi-element standard solution 70006. The
water hardness and oil content were measured by classical
analytical methods (titrimetric, gravimetric and FTIR). The content
of cations (NHþ þ
4 and Na ) was measured by ion chromatography
(IC) using Metrohm, type 861, Advanced Compact IC. For the anion
concentration, a Dionex DX 100 IC system (Dionex, Sunnyvale, CA,
USA) consisting of an IP20 isocratic pump and a CD20 conductivity
detector was used. The PeakNet 5.1 chromatography workstation
was used for instrument control, data collection, and processing.
122 V.N. Rajakovi
c-Ognjanovi
c et al. / Applied Thermal Engineering 31 (2011) 119e128

Dionex IonPac AS14 (250  4 mm I.D.) and AG14 (50  4 mm I.D.)

T columns were used for anion determination and separation. An
Anion Auto Self Regenerating Suppressor (ASRS, 4 mm I.D.) oper-
ated in the recycle mode was used as the suppressors. A sample
loop with a volume of 1000 mL was made from poly(ether ether
ketone) (PEEK) tubing of length 210 cm and I.D. 0.750 mm. Its
volume was verified by measuring the mass difference between the
sample loop filled with deionized water and the empty loop.

3.2. Chemicals and samples

s
Legend:
1–2 isoentropic expansion of steam (HPT)
2–3 isobaric heating of steam (SH)
3–4 isoentropic expansion of steam (MPT and LPT)
4–5 isobaric-isothermal heat reduction during total condensation of steam (C)
5–6 isoentropic compression of feed water (FWP)
6–7 isobaric heating and warming-up of feed water (HPH, Ec and Ev)
7–8 isoentropic steam expansion (SS)
8–1 isobaric heating for steam warming up (RH)
Fig. 3. Tes diagram for steam transformation in the WSC of the TPP Nikola Tesla B.
The chemicals and reference materials used in this work for
water quality measurements are listed in Table 4.
3.2.1. Samples from the WSC of the TPP Nikola Tesla
A scheme of the wateresteam cycle of the TPP Nikola Tesla,
Serbia, is shown in Fig. 2 with the sampling points indicated. Each
sampling point and type of water is described in the legend given
below Fig. 2. Some parameters, i.e., pH values, conductivity and
oxygen, were measured on line. The other parameters were
measured in the laboratory of the Faculty of Technology and

Table 1
The main characteristics of the control parameters and their specific advantages for an estimation of wateresteam quality and the functioning of the main parts of the WSC in
TPP.

Control parameter The characteristics and influence to corrosion process

pH Optimal pH value for corrosion prevention of the WSC is in the range of 8.8e9.2.
Simple indication of all acidic and basic compounds and products of their reactions in the water and steam. It has the feature of a cumulative
parameter. The pH value has to be in agreement with the values of conductivity and ammonia content. Lower pH values cause higher corrosion
of tubes in the WSC and even dissolution of already formed deposits in the system.
Specific advantages:
 an estimation of the quality of the demi water in the CWTP,
 to control the PP (dosage of NH3 and N2H4).
Conductivity The recommended values are <0.2 mS cm
1. Conductivity measurements are reliable and simple. It is a cumulative parameter, it detect all ionic
forms (for example inorganic impurities: Naþ, Kþ, Cl
, SO2


4 , NO3 ) dissolved in the water.
Specific advantages:
 control of the CWTP,
 an indication of possible leakage from condenser pipes,
 an estimation of the salt content in the water and the solubility of deposits in the boiler,
 an estimation of the steam quality.
Oxygen, O2 The recommended values are < 20 mg dm
3. This value can be obtained with thermal treatment and conditioning, and by prevention of oxygen
breakthrough into the system. Oxygen corrosion is the most common corrosion that appears in boilers. It occurs when oxygen interacts with
metal surfaces in the boiler. If alkaline conditions are dominant and if there are no other impurities, the protective layer of magnetite, Fe3O4,
protects steel from further corrosion. The content of iron, copper and pH value are directly connected with oxygen content and vice versa.
Specific advantages:
 the control of leakage from the condenser pipes,
 an estimation of conditioning quality.
Sodium, Naþ The recommended values are <10 mg dm
3. In ultra pure water, the presence of Naþ ion is an indication of the presence of the corresponding
soluble salts, which are hardly recognized as impurities in water, but if present, they contribute to deposit formation (for example:
Na2O$Fe2O3$4SiO2, Na2O$4CaO$6SiO2$H2O).
Specific advantages for:
 control of the PP,
 leakage detection of condensate (Naþ ion concentration is first indication of leakage and particularly it is useful when the conductivity is too
low to be measured),
 steam quality between boiler, separator and preheater,
 the transport of impurities,
 an estimation of the amount of deposited solids (through their solubility),
 an indication of quality of demi water after the mixed-bed ion-exchanger.
Chloride, Cll
The recommended values are <10 mg dm
3. In a high pressure atmosphere, excellent indicator for steam purity. Cl
ion is extremely reactive
and increases the corrosion processes.
Specific advantages:
 monitoring wateresteam quality (retained HCl from the regeneration of the ion-exchange resins),
 monitoring possible breakthrough of FW in the boiler.
Silica, SiO2 The recommended values are <20 mg dm
3. Silica combines with a wide variety of elements to produce silicates or it may form deposits on
its own (for example: SiO2, Na2SiO3$9H2O, NaFeSi2O6, Mg6[(OH)8Si4O10]). Silicates form tenacious deposits in cooling water systems, boiler tubes
and on turbine blades. The scales are inert to most chemical cleaning solutions, with exception of HF. This is an extremely dangerous compound
and makes prevention of silica deposition even more important.
Specific advantages:
 monitoring of the CWTP. Silica is the first impurity which passes through an ion-exchanger if it requires regeneration.
 FW monitoring,
 PP monitoring.
 V.N. Rajakovi
c-Ognjanovi
c et al. / Applied Thermal Engineering 31 (2011) 119e128 123

Table 2
The main characteristics of the diagnostic parameters of the wateresteam quality in the WSC.

Diagnostic parameter The characteristics and influence on the corrosion process

Iron, Fe The recommended values in the feed water are <20 mg dm
3. If this value is exceeded (when the temperature in feed water increases)
or if oxygen from water reacts, several possible reactions will occur:
4Fe2þ(aq) þ 2HCO
3 (aq) þ O2(g) þ H2O ¼ 4Fe(OH)3(s) þ 4CO2(g)
Fe2þ(aq) þ HCO
3 (aq) þ heat ¼ Fe(OH)2(s) þ 4CO2(g)
4Fe(OH)2(s) þ O2(g) þ H2O ¼ 4Fe(OH)3(s)
2Fe(OH)3(s) ¼ Fe2O3(s) þ 3H2O
In the boiler deposits, such as iron-silicates, are also present. These deposits are adherent, robust solids and hard to remove and they cause
breakage of boiler pipes. The sources of boiler deposits are corrosion products, such as magnetite (Fe3O4) and hematite (Fe2O3). A layer of
magnetite is desirable protection from further corrosion.
Copper, Cu The recommended values are <3 mg dm
3. Copper is more resistant toward corrosion then steel, but if ammonia is present in the water,
copper intensively corrodes forming oxides and corresponding complex compounds. Copper products deposit in form of oxides: Cu2O, CuO,
or in forms of fine crystals between layers of magnetite. The best protection method of copper parts from corrosion is removal of oxygen
from the system. If this cannot be achieved, the ammonia concentration needs to be lower than 0.5 mg dm
3.
Organic matter The recommended values are <3 mg dm
3. Organics break down in the boiler to form organic acids. The resultant low pH can be quite
deleterious. Organic acid and carbon dioxide produced by decomposition can carry over to the steam, turbines and corrode the blades.
Organics may also be found in the condensate return. These organics are much shorter chained than surface water organics and may
require different treatments.
Oil content The recommended values are <100 mg dm
3. Oil is toxic and it decreases the efficiency of the demineralisation process and condensate
purification. Once oil comes in contact with ion-exchange resins in the PP, it decreases the ion-exchange process. Oils decompose at
higher temperatures and produce impurities and particles in the water and steam. Even very low oil concentrations are toxic and, therefore,
removal of an oil phase is essential before it enters the WSC.

Metallurgy, University of Belgrade. Samples for measurement of
these parameters were handled carefully to avoid contamination
and they were analyzed as quickly as possible. These samples were
analyzed without any pre-treatment, except for filtering for the IC
measurements. The sampling was performed according to a stan-
dard procedure; the water was placed in polypropylene bottles and
transported in a portable refrigerator to the laboratory for
measurement.
3.3. Analytical methods and tools for ultra-pure water quality
The following methods were applied:
parallel: on line and in the laboratory. The reliable and accurate
determination of corrosive ions in power plant water samples at
sub to low mg dm
3 levels is a challenging analytical problem
[20e28]. Over the past 20 years, ion chromatography (IC) has
become an indispensable technique for addressing this problem
because IC offers the unique capability of separating and identifying
individual ionic species at sub to low mg dm
3 levels. Anions in the
ppm (mg dm
3) and ppb (mg dm
3) range can be analyzed easily
with direct injection. For the determination of low levels of ionic
impurities in the low ppb or even ppt (ng dm
3) levels, sample
preconcentration is usually required. In a previous study, an ion
chromatographic method was developed for the determination of
inorganic anions at trace levels in power plant water samples [8].

 Electroanalytical for pH, conductivity and the dissolved oxygen

content,
 IC for the ion content (Naþ, NHþ

4 and Cl ),
 Optical, ICP-MS for iron, copper and silica and FTIR for the oil
content, and
 Classical analytical (titrimetric) for the water hardness.
For all electroanalytical measurements, standard procedures
were applied from Standard Methods for water and wastewater
determination [19]. This group of parameters was measured in
4. Results and discussion
4.1. Chemical analysis of the control and diagnostic parameters
Quality control of representative samples of the WSC in the TPP
Nikola Tesla was accomplished in June 2009 and in June 2010 for
the two blocks of the TPP. The results for the control parameters of
water quality are presented in Table 5. The results for the diagnostic
parameters of water quality are presented in Table 6. Samples of

Table 3
Control and diagnostic water quality parameters within the WSC of TPP [2].

Water quality parameter Characteristic value Monitoring frequency Analytical tool Sensitivity
Control parameters
 pH 8.8e9.2 C pH-meter, on line 0.01 pH unit
 Conductivity, k  0.2 mS cm
1 C Conductivity-meter, on line 0.01 mS cm
1
 Oxygen, O2 10e20 mg dm
3 C Oxygen-meter, on line 1 mg dm
3
 Sodium, Na 10 mg dm
3 C Sodium-meter, on line 0.1 mg dm
3
 Chloride, Cl 20 mg dm
3 C/D Chloride-meter, on line 1 mg dm
3  0.1 mg dm
3
IC in the laboratory
 Silicon, Si 20 mg dm
3 C/D Sodium-meter, on line ICP-MS in lab 1 mg L
1  0.001 mg L
1

Diagnostic parameters
 Iron, Fe 20 mg dm
3 D/W ICP-MS 0.001 mg L
1
 Copper, Cu 3 mg dm
3 D/W ICP-MS 0.001 mg L
1
 Organic matter 3 mg dm
3 D TOC 1 mg dm
3
 Oil content 100 mg dm
3 W FTIR 0.1 mg dm
3

Legend: C-Continuously, D-Daily, W-Weekly; IC-Ion Chromatography, ICP-MS-Induced Coupled Plasma-Mass Spectrometry, FTIR-Fourier Transformer Infra Red. TOC-Total
Organic Carbon.
124 V.N. Rajakovi
c-Ognjanovi
c et al. / Applied Thermal Engineering 31 (2011) 119e128

Table 4
Chemicals used for water quality parameters measurements.

Chemicals Reference materials

 H2SO4 (96%, r ¼ 1.84 g cm
3, p.a., Aldrich)
 CCl4 (p.a., Aldrich)
 n-hexane (p.a., Riedel de Haen)
 anhydrous Na2SO4 (p.a., Aldrich)
 HCl (1:1, Merck)
 HNO3 (1:1, Merck)
Technical Buffer pH 7.00 (WTW; certified: traceable to NIST/PTB).
Control Standard for Conductivity Cells 0.01 mol KCl (WTW; E/SET 300 572;
according to DIN 38404/ISO 7888)
Hach Dissolved Oxygen Standard Solution, 10 mg L
1, bottle 500 mL, Product number: 40149, 10 mg dm
3 as DO, APHA
(for Winkler titrations only) 0.00125 mol dm
3 KIO3.
Fluka multi-element standard solution 70006
Fluka eCertified Multianion Standard Solution PRIMUS for ion chromatography 10 mg kg
1  0.2% F, Cl, Br, NO3,
SO4, PO4 each anion. Produced with high purity materials (NaF, NaCl, NaBr, NaNO3, Na2HPO4, Na2SO4), ultrapure water
(18 MU, 0.22 mm filtered). Certified by EMPA/BAM.
Merck-CertiPUR Multieelement Standard for cation chromatography e 9 cations (NH4, Ba, Ca, K, Li, Mg, Mn, Na, Sr)
with an accuracy 95.0e105.0 mg dm
3

demineralised water (#1), condensate (#3), feed water (#4) and
wet steam (#9) are highlighted in bold as representative types of
water which were thoroughly analyzed. The numerical values that
were above the recommended values are presented in bold.
Obtained results presented in Tables 5 and 6 can be discussed
from two aspects, based on: a) the quality of type of water and b) the
quality of a particular parameter.
a) Concerning the type of water, results indicate the following.
 Demineralised water, water which enters from the CWTP into
the WSC, samples 1 in Tables 5 and 6, for both blocks, has
a quality which is close to the recommended range. The pH of
the demi water ranged from 6.0 to 6.6, which was expected
after the mixed IE in the CWTP. The pH was in the acid range
due to the adsorption of CO2 (g-gas from air) and the formation
of H2CO3 (aq-aqua in water) according to the chemical equi-
librium at the gas/liquid interface and the chemical equilibrium
(1) in the water:
CO2 ðgÞ þ H2 O#H2 CO3 ðaqÞ#Hþ þ HCO
þ 2

3 #2H þ CO3
Conditioning of the demi water and achieving the target range of
pH values (8.8e9.2.) for a WSC can be achieved by the addition of
a base reagent and an oxygen scavenger; in the studied WSC,
ammonia and hydrazine are added, as indicated in Fig. 2.
 Condensate, samples 2 and 3, in block 1, were not satisfactory
because of the high concentration of ions that promote corro-
sion activity (chloride and sodium). The condensate is a source
of feed water and it has to be water of high quality. The iron and
copper content were higher than recommended: 26 mg L
1 for
iron (the recommended value is 20 mg L
1) and 3.7 mg L
1 for
copper (the recommended value is 3 mg L
1), which is
a consequence of corrosion of pipes due to reaction with the
chloride ions present in the water. The total conductivity is
much higher than recommended: 61 and 29 mS cm
1 (the
recommended value is 0.2 and maximal value that could be
tolerated is 0.3 mS cm
1). The conductivity was higher due to
the presence of chloride and sodium, which implies that
cooling water from the condenser, had made a breakthrough.
The conductivity can be improved by more stringent control of
the number sodium and chloride ions in the CWTP and the PP.
 Feed water, sample 4, is a mixture of return water (from the
condenser) and additional demi water (used when there is
water loss from the system). The feed water quality has to be of
high quality. The Feed water in block 1 had higher concentra-
tions of chloride and sodium which indicates to contamination
with NaCl, the reagent used for the regeneration of the mixed
ion exchanger in the CWTP. The functionality of CWTP and PP
should be improved by more stringent control of sodium and
chloride ion in demi water, condensate and particularly in the
condensate return.
(1)
 Steam, samples 5e11 in block 1 and 2 had relatively good qual-
ities. The steam samples represent wet steam (sample 5) and
reheated steam (samples 6e11). Wet steam is the steam at the
temperature which is the same as the temperature of saturation
at the pressure in the system. Sample 5 presents the sample at
the evaporator entrance. Sample 8 presents super heated steam
which presents steam at the temperature which is higher than
the saturation temperature at the pressure in the system.
In some samples, higher concentrations of sodium (sample 7, in
block 1) and chloride (samples: 5, 6, 8, 9 and 11, in block 1) were
detected, which is a consequence of the inadequate quality of the
feed water. The silica content was slightly higher in sample 8 in

Table 5
The results obtained for control parameters in the WSC of the TPP Nikola Tesla.

# Sample Control water quality parameter

pH k mS cm
1 O2 Na Cl
Si

mg dm
3
1 Demineralised water 6.60 6.00 0.1 0.1 14 11 9 10 8 10 14.0 19.0
2 Condensate I 9.15 9.10 61 <1 6.0 4.0 46 18 91 20 26.0 14.0
3 Condensate II 9.20 9.20 29 <1 4.8 5.1 61 10 37 7 24.8 14.1
4 Feed water 9.12 9.10 <1 <1 4.6 5.2 29 9 43 3 24.6 14.5
5 Steam, inlet to evaporator (19.5 m) 9.24 9.20 <1 <1 4.7 6.1 <1 <1 21 1 44.7 14.9
6 Saturated steam (103.8 m) 9.20 9.20 31 <1 5.0 5.7 <1 <1 27 7 25.0 14.7
7 Superheated steam (67 m) 9.31 9.10 <1 <1 4.7 4.6 31 10 14 4 24.0 17.2
8 Superheated steam (19 m) 9.22 9.20 <1 <1 5.0 4.5 <1 <1 78 7 15.0 15.8
9 Wet steam (96.5 m) 9.20 9.00 <1 <1 4.9 5.6 <1 <1 25 5 4.9 25.8
10 Reheated steam (72.5 m) 9.35 9.10 <1 <1 5.0 4.7 <1 <1 17 7 25.0 15.2
11 Reheated steam (19.0 m) 9.27 9.15 <1 <1 4.9 4.7 <1 <1 39 10 5.9 15.9

Notification: In the first column are the results from 2009, block 1 and in the second the results are from 2010, block 2.
The results given in the table are as the mean of five measurements (n ¼ 5).
 V.N. Rajakovi
c-Ognjanovi
c et al. / Applied Thermal Engineering 31 (2011) 119e128 125

Table 6 CO2 ðgÞ þ H2 O#H2 CO3 ðaqÞ (3)

The results obtained for the diagnostic parameters in the WSC of the TPP Nikola
Tesla. The absorption of CO2 into the condensate drives the pH down-
# Sample Diagnostic water quality parameter ward. The acid solution directly attacks the pipe walls via the
formation of ferrous bicarbonate (equation (4)):
Fe Cu Organic Oil
matter content
FeðsÞ þ 2H2 CO3 ðaqÞ#FeðHCO3 Þ2 ðaqÞ þ H2 ðgÞ (4)
mg dm
3
Any oxygen present in the feed water will oxidize the ferrous
1 Demineralised water 18 7.00 1.50 0.2 <3000* <100*
2 Condensate I 26.0 16.40 2.71 0.5 bicarbonate to ferric oxide and liberate CO2 to repeat the process, as
3 Condensate II 3.94 28.30 3.50 3.0 presented in equation (5):
4 Feed water 4.30 7.30 0.62 1.4
5 Steam, inlet to evaporator 4.12 10.80 3.57 1.5 4FeðHCO3 Þ2 ðaqÞ þ 2O2 ðgÞ#2Fe2 O3 ðsÞ þ 4H2 O þ 8CO2 ðgÞ (5)
(19.5 m)
6 Saturated steam (103.8 m) 4.10 12.80 0.41 1.9
 The conductivity: the conductivity of most of the samples was
7 Superheated steam (67 m) 3.80 6.70 0.40 0.7 between 1 and 10 mS cm
3, which is satisfactory, according to the
8 Superheated steam (19 m) 4.00 6.85 0.52 2.2 values indicated in Table 3 [2], except for the samples 2, 3 and 6 in
9 Wet steam (96.5 m) 4.60 5.80 0.64 2.8 block 1. These values were also tested under laboratory conditions.
10 Reheated steam (72.5 m) 24.1 8.40 0.29 1.2
In the laboratory it was possible to measure just the total
11 Reheated steam (19.0 m) 5.10 9.50 0.60 0.7
conductivity and not the cationic conductivity. The conductivity
Notification: In the first column are the results from 2009, block 1 and in the second (total and cationic) is the critical parameter for the quality control
the results are from 2010, block 2.
The results given in the table are as the mean of five measurements (n ¼ 5).
of ultra pure water. According to the guidelines for ultra pure
*
The same values for each year were obtained. water, the value of cationic conductivity has to be lower than
0.3 mS cm
3 [2]. Cationic conductivity represents the quality of
water that has passed through an acid cation exchange column.
block 1 (this was also the case previously and it is probably because Passing through an acid cation exchange column, all the cations
of the presence of deposits in this part of the system). are attached to the resin while Hþ ions are released. Cation
The copper content was similar in all the samples, which indi- conductivity is always lower than the total conductivity. When
cates that the concentration of ammonia was in the optimal range, analyzing conductivity measurements, it is important to empha-
in harmony with the pH. size that measurements in the laboratory also do not present the
The pH and conductivity of the steam samples were in the real water quality. The values of conductivity measured in labo-
vicinity of the recommended values. ratory are little bit higher than in the WSC, which is due to
The oil content was satisfactory in all the samples. absorption of CO2 from air and formation of carbonic acid, which
Higher amounts of cations and anions in the analyzed samples is a source of Hþ ions that increases the conductivity.
indicate there is contamination of the condensate which affects the  Sodium content: the sodium content in water samples: 2, 3, 4
steam quality at the high temperatures and pressures. and steam sample 7 in block 1 were above the recommended
Almost all types of water in block 2 were in better condition value of 10 mg dm
3 [2]. The recommendation is that sodium
than the water in block 1, which means that the quality of the feed should be measured on line and that a sodium-meter should be
water and the chemical control in block 2 was more efficient than in installed in order to prevent corrosive conditions, unacceptable
block 1. In samples 2 and 9, some excess of sodium and silica was contamination levels and other deposition conditions. An on-
observed, which can be the consequence of partial dissolution of line conductometer is useful but an on line sodium-meter is
silica scales formed in the tubes of the WSC. even more important and precise when breakthrough in the
cation exchangers occurs. Unified conductometer and sodium-
b) Concerning the quality of particular parameters meter represent the optimal control of the proper functioning
I Control parameters of the cation exchanger.
 pH values: The pH values of all the analyzed samples were  Dissolved oxygen: the dissolved oxygen content was measured
within the recommended range. The pH value of the steam on line, only a few samples were controlled in the laboratory.
samples were within 8.8e9.2, except for sample 7 and 10 in These values were not taken into consideration because during
block 1, which can be explained by inappropriate polishing of sampling, transport and measurements in the laboratory, the
the water with ammonia and hydrazine. In the demi water, as saturation with oxygen from the air was reached. For water
was explained, the pH was in the acid range due to adsorption of quality analysis, the on line measurements are the ones that
CO2 and the formation of H2CO3 according to the chemical should be considered. The oxygen content is expected to be
equilibrium. CO2 also has a great influence on the pH within 10e20 mg dm
3 and lower [2]. Oxygen and CO2 are the most
a WSC if it is not removed in a deaerator. It causes corrosion of common corrosive species in a steam generation system.
the condensate lines, which is recognized as a serious problem. Hydrazine, the most common and effective scavenger for many
CO2 can form in the boiler. CO2 formation primarily occurs when years is now listed as a hazardous chemical [29]. Due to the
partial demineralisation (softening) to remove the hardness handling procedure hydrazine has to be applied with more
ions, calcium and magnesium, is employed for the production of precautions. The use of hydrazine is common as it does not add
make-up water. The make-up system does not remove bicar- any dissolved solids (s-solid) to the feed water, as presented by
bonates, which decompose in the boiler to produce CO2, as chemical equation (6), and it also passivates oxidized areas of
presented by the equation (2): piping and tube materials, presented by chemical equation (7):

HCO


3 ðaqÞ þ heat#CO2 ðgÞ þ OH ðaqÞ (2) N2 H4 ðaqÞ þ O2 ðgÞ#N2 ðgÞ þ H2 O (6)
CO2 carries over with the steam and then dissolves again in the
N2 H4 ðaqÞ þ 6Fe2 O3 ðsÞ#4Fe3 O4 ðsÞ þ N2 ðgÞ þ 2H2 O (7)
condensate return lines. Chemists debate over whether CO2 actually
forms an acid or whether it exists as a discrete hydrated molecule Alternative oxygen scavengers are: hydroquinone (C6H4O2), car-
but the net effect of this is shown in the following equation (3): bohydrazide (N4H6CO) and methyl ethyl ketoxime (C4H4NOH).
126 V.N. Rajakovi
c-Ognjanovi
c et al. / Applied Thermal Engineering 31 (2011) 119e128
Fig. 4. The corrosion potential of silica compounds in the WSC.
However, the alternative organic scavengers and the pH-controlling
amines can break down in the boiler system to produce organic
acids and CO2, which in turn can cause corrosion of after boiler
components including turbine blades.
 Chloride content: the chloride content was too high in almost all
of the analyzed samples from block 1. In sample 1, the content
of chloride ion was 8 mg dm
3, the value which is recom-
mended. In the other samples, the content of chloride was 2e9
times higher. A major part of the water costs is the treatment of
water to attain boiler feed water quality [11,13]. Higher
concentration of chloride implies that the feed water was
contaminated with NaCl and HCl, reagents used for the
regeneration of the resins in the CWTP.
 Silica: the obtained values for the content of silica were around
5 mg dm
3, except for samples 1 and 2, in block 1 and sample 9
in block 2. It is important to stress that a silicometer was added
in block 2 after the first measurements in June 2009. Silica is
important for corrosion analysis of a WSC system because it
forms thick and dense deposits on the boiler and turbine
surfaces which lead to thermodynamic and mechanical loss in
plant efficiency. Silica is present in water in the form of sus-
pended solids and as dissolved salts. There is a balance
between these compounds, which can be presented as (8):
SiO2 ðsÞ þ H2 O#H2 SiO3 ðaqÞ#Hþ ðaqÞ þ HSiO
þ
3 ðaqÞþ2H ðaqÞ
þ SiO2

3 ðaqÞ
(8)
Dissolved silica can be indirectly measured by conductivity and
removed by an anionic ion exchanger. Solid silica can be measured
Fig. 5. The corrosion potential of iron compounds in WSC The abbreviations are explained in Fig. 1.
using a silicometer and removed in pre-treatment by coagulation.
The cycle of the silica corrosion potential is presented in Fig. 4. In
the same figure, the recommended values are also indicated, as
well as the mechanism for silica removal (in the PP).
II Diagnostic parameters
 Iron: the obtained values for iron were below 5 mg dm
3, except
for sample 2 in block 1 and for sample 3 in block 2. The rec-
ommended value of the iron content for all types of water in
the WSC is lower then 20 mg dm
3. If this value is exceeded,
several possible deposits will form (for example: Fe(OH)3, Fe
(OH)2 and/or Fe2O3). In the boiler, the deposits are magnetite
Fe3O4 and hematite Fe2O3. The cycle of the iron corrosion
potential is presented in Fig. 5. In the same figure the recom-
mended values are also indicated, as well as the mechanism for
iron removal (in the PP).
 Copper: the obtained values for the content of copper were
below 3 mg dm
3, except for sample 2 in block 1 and sample 3
in block 2. Copper product deposit in forms of oxides: Cu2O,
CuO, or in forms of fine crystals between layers of magnetite.
The mechanism of ammonia corrosion can be described in two
stages, equations (9) and (10):
2Cu2 OðsÞ þ O2 ðgÞ ¼ 4CuOðsÞ (9)
 
CuO s þ 4NHþ
2þ
4 ðaqÞ þ 2OH ðaqÞ ¼ CuðNH3 Þ4 ðaqÞ þ 3H2 O
(10)
The best protection method of copper parts from corrosion is the
removal of oxygen from the system. If this cannot be achieved,
ammonia concentration needs to be lower then 0.5 mg dm
3. In
 V.N. Rajakovi
c-Ognjanovi
c et al. / Applied Thermal Engineering 31 (2011) 119e128 127

Fig. 6. Water cycle system-a review of the water quality parameters with values recommended for the target ranges (TR) and adequate actions levels.

addition to oxygen and carbon dioxide, ammonia is corrosive in papers, for water recovery in steam injected gas turbines [30], for
a steam generation system when the tubes are made of cooper. mixed gasesteam cycles [31] and also some alternative trends in
development of corrosion protection in thermal power plants [32].
 Organic matter and oil content: in all samples, the content of One of the aspects how to solve the corrosion phenomenon is to
organic matter was below 3 mg dm
3 and the oil content was design new materials resistant to corrosion [33]. The production of
below 100 mg dm
3, which confirms the proper functioning of ultra pure water for the WSC and permanent chemical control is
the CWTP. Organics break down at higher temperatures and connected with significant cost. One line in the CWTP of the TPP
high pressure and form organic acids and CO2. The resultant Nikola Tesla produces 400,000 m3 year
1 of demi water (z2 euro
low pH can result in intensive corrosion. m
3). The annual cost for chemicals used in the CWTP for one line is
100,000 euros. Each on line parameter in the WSC should be labo-
ratory tested on expensive and sophisticated instruments. Strict
4.2. Comprehensive analysis of the results of chemical control in the chemical control has a great influence on: the consumption of
WSC chemicals, number of regenerations of the resins in the CWTP. Con-
cerning leakage, losses of ultra pure water, contamination of water
In order to be concise, the main control and diagnostic parame- and the number of damages and outages due to deposits in the tubes
ters for the main type of water and steam in the WSC are presented in of the WSC are costly, which means that the benefits of chemical
Fig. 6. It shows the conclusive improvement of the chemical control control are highly valuable. A better understanding of system ener-
which comprehends and connects the target range of parameters, getics and monetary loss should also involve chemical control.
frequency of measurements, analytical tools and action levels. In this Chemical control and measuring equipment in a thermal power
way, chemical control can play an important role in the reliable plant should be improved in order to improve quality of feed water.
maintenance of a TPP. A line for demineralisation should be equipped with on line
Almost all countries with similar power plants in Europe, as analyzers for sodium, chloride and silica. Those cost and their
presented in Table 7, suggested the introduction of silica measure- maintenance is app. 50,000 euro per analyser.
ments in the WSC of TPPs. On line measurements of sodium and
chloride as control parameters of water and steam in the WSC were 5. Conclusion
suggested by only a few countries.
The losses and damages due to corrosion are very hard to estimate, The necessity for consequential chemical control of the WSC of
their economic and ecological aspects were the subject of some recent a TPP was demonstrated through the concrete example of the TPP
Nikola Tesla, Serbia. The improvement of water quality was
considered and connected with chemical processes which affect
Table 7 corrosion processes. High quality of ultra pure water assures more
Comparison of the control parameters in similar power plants in Europe [2]. reliable WSC operational capabilities, which lead to a more reli-
able TPP operation. Each point inside the WSC and each water
Control parameter Country
quality parameter have to be analyzed permanently. It is
GB FR ES NL IT RS concluded that expertise of the WSC from demi water to steam
pH 8.8e9.2 9.1e9.3 9.0e10.0 9.0e9.2 9.1e9.2 8.8e9.2 and condensate can be accomplished by analysis of an optimal
Conductivity, <0.1 0.1 <0.2 <0.2 <0.5 <0.1
number of diagnostic and control parameters. Sodium and chlo-
mS cm
1
Sodium, mg dm
3 <2 e <3 <1 <3 <10 ride ions were recognized as ions which indicate the corrosion
Chloride, mg dm
3 e e <3 <3 <3 <10 potential of water and give insight into the proper functioning of
Silica, mg dm
3 <20 <20 <10 <20 <20 <20 the WSC. An insight into the chemical transformations of crucial
GB-Great Britain, FR-France, ES-Spain, NL-the Netherlands, IT-Italy and RS-Republic corrosion elements, iron and silica, were considered and related
of Serbia. to their quantitative values.
128 V.N. Rajakovi
c-Ognjanovi
c et al. / Applied Thermal Engineering 31 (2011) 119e128
Acknowledgements
Support from the Research Fund of Serbia is gratefully
acknowledged.
References
[1] Lj.V. Rajakovi
 
c, V. Sija 
cki-Zerav 
ci

c, P. Stefanovi
c, Corrosion Potential of Water,
Book 2. Codex, Belgrade, 2002.
[2] Pressurized Water Reactors Secondary Water Chemistry Guidelines, Revision
6 (EPRI Report 1008224), December, 2004.
[3] N. Rachid, Method and device for operating the water/steam cycle of
a thermal power station, Applied Thermal Engineering 18 (1998) XXI.
[4] K.R. Cooper, R.G. Kelly, Sources of variability in cation analysis for a quanti-
tative exfoliation corrosion resistance test, Journal of Chromatography A 739
(1996) 183e190.
[5] D.H. Thomas, M. Rey, P.E. Jackson, Determination of inorganic cations and
ammonium in environmental waters by ion chromatography with a high-
capacity cation-exchange column, Journal of Chromatography A 956 (2002)
181e186.
[6] B. De Borba, M. Laikhtman, J. Rohrer, Determination of sodium at low ng/l
concentrations in simulated power plant waters by ion chromatography,
Journal of Chromatography A 995 (2003) 143e152.
[7] D. Ci
ckari

c, J. Markovi
c, Lj. Rajakovi
c, GF-AAS determination of iron traces in
the ultra pure water, Water Quality 2 (2004) 14e16.

[8] D. Ci ckari

c, I. Dersek-Timoti
c, A. Onjia, Lj. Rajakovi
c, Development of ion chro-
matography methods for determination of trace anions in ultra pure water in
the power plant, Journal of the Serbian Chemical Society 70 (2005) 995e1003.
[9] S. Mladenovi
c, Corrosion of Materials. TMF, Belgrade, 1990.
[10] M.A. Sanz-Bobi, I.J. Perez-Arriaga, J.L. Serrano-Carbayo, M.E. Ortiz-Alfaro,
J.J. Alba, A. Domenech, M.J. Villamediana, J. Gonzales-Huerta, J.J. Fernandez-
Matrinez, Control and diagnosis of water chemistry in the wateresteam and
water make-up in a fossil fuelled power plant, Electric Power and Energy
Systems 16 (1994) 251e258.
[11] M.S. Mohsen, Treatment and reuse of industrial effluents: case study of
a thermal power plant, Desalination 167 (2004) 75e86.
[12] H.H. Erdem, A. Dagdas, S.H. Sevilgen, B. Cetin, A.V. Akkaya, B. Sahin, I. Teke,
C. Gungor, S. Atas, Thermodynamic analysis of an existing coal-fired power
plant for district heating/cooling application, Applied Thermal Engineering 30
(2010) 181e187.
[13] M. De Paepe, E. Dick, Technological and economical analysis of water recovery
in steam injected gas turbine, Applied Thermal Engineering 21 (2010)
135e156.
[14] M. Siddartha Bhatt, Effect of water side deposits on the energy performance of
coal fired thermal power plants, Energy Conversion and Management 47
(2006) 1247e1263.
[15] P. Kritzer, Corrosion in high-temperature and supercritical water and aqueous
solutions: a review, The Journal of Supercritical Fluids 29 (2004) 1e29.
[16] E. Kaiser, J. Riviello, M. Rey, J. Statler, S. Heberling, Determination of trace level
ions by high-volume direct-injection ion chromatography, Journal of Chro-
matography A 739 (1996) 71e79.
[17] V.V. Slesarenko, Electrodialysis and reverse osmosis membrane plants at
power stations, Desalination 158 (2003) 303e311.
[18] M. Xu, S. Li, W. Cai, Cascade generalized predictive control strategy for boiler
drum level, ISA Transactions 44 (2005) 399e411.
[19] 21st Edition of Standard Methods for the Examination of Water & Waste-
water: Centennial Edition, 2005.
[20] H. Lu, Sh. Mou, R. Deng, J. Riviello, Studies of trace anions analysis by ion
chromatography, Microchemical Journal 64 (2000) 1e7.
[21] R. Slingsby, R. Kiser, Sample treatment techniques and methodologies for ion
chromatography, Trends in Analytical, Chemistry 20 (2001) 288e295.
[22] Z. Lu, Y. Liu, V. Barreto, C. Pohl, N. Avdalovic, R. Joyce, B. Newton, Determi-
nation of anions at trace levels in power plant water samples by ion chro-
matography with electrolytic eluent generation and suppression, Journal of
Chromatography A 956 (2002) 129e138.
[23] Y. Liu, E. Kaiser, N. Avdalovic, Determination of trace-level anions in high-
purity water samples by ion chromatography with an automated on-line
eluent generation system, Microchemical Journal 62 (1999) 164e173.
[24] E. Santoyo, S.P. Verma, F. Sandoval, A. Aparicio, R. Garcia, Suppressed ion
chromatography for monitoring chemical impurities in steam for geothermal
power plants, Journal of Chromatography A 949 (2002) 281e289.
[25] J. Weiss, Ion Chromatography, second ed. VCH, Weinheim, Germany, 1995.
[26] W.W. Buchberger, Detection techniques in ion chromatography of inorganic
ions, Trends in Analytical Chemistry 20 (2001) 296e303.
[27] R. Nordhaus, J. Anderson, Recent advances in ion chromatography suppressor
improve anion separation and detection, Journal of Chromatography A 956
(2002) 15e22.
[28] D. Douglas, R. Saari-Nordhaus, P. Despres, J. Anderson Jr., New suppressor
technology improves trace level anion analysis with carbonate-hydro-
gencarbonate mobile phases, Journal of Chromatography A 956 (2002) 47e51.
[29] T. Schwarz, Heat transfer and fouling behaviour of Siemens PWR steam
generators e long-term operating experience, Experimental Thermal and
Fluid Science 25 (2001) 319e327.
[30] M. De Paepe, E. Dick, Technological and economical analysis of water recovery
in steam injected gas turbines, Applied Thermal Engineering 21 (2001)
135e156.
[31] A. Traverso, F. Massardo, Thermoeconomic analysis of mixed gas-steam
cycles, Applied Thermal Engineering 22 (2002) 1e21.
[32] V.A. Prisyazhniuk, Alternative trends in development of thermal power plants,
Applied Thermal Engineering 28 (2008) 190e194.
[33] B.N. Grgur, M.M. Gvozdenovi
c, V.B. Miskovi

c-Stankovi
c, Z. Ka
carevi

c-Popovi

c,
Corrosion behavior and thermal stability of electrodeposited PANI/epoxy
coating system on mild steel in sodium chloride solution, Progress in Organic
Coatings 56 (2006) 214e219.