Professional Documents
Culture Documents
Carlos Brebbia
Jerome J. Connor Editors
J. W. Newkirk · A. A. Popov
A. S. Zhilin Associate Editors
Progress in
Materials Science
and Engineering
Innovation and Discovery in Russian Science
and Engineering
Series Editors
Carlos Brebbia
Wessex Institute of Technology, Southampton, United Kingdom
Jerome J. Connor
Massachusetts Institute of Technology, Cambridge, MA, USA
Associate Editors
J. W. Newkirk A. A. Popov
Materials Science and Engineering Ural Federal University
Department Yekaterinburg, Russia
Missouri University of Science
and Technology
Rolla, MO, USA
A. S. Zhilin
Institute of New Materials
and Technologies
Ural Federal University
Yekaterinburg, Russia
This Springer imprint is published by the registered company Springer International Publishing AG part
of Springer Nature.
The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland
Preface
This book is concerned with current issues related to progress in materials science
and technologies.
Progress in modern materials science determines the necessity for rapid develop-
ment in production methods for new materials. The techniques of production have to
be updated and adjusted for achieving new aims and objectives in materials science.
Specifically, the need for new composite materials determined the application of
innovative approaches.
The mechanisms of structure formation in complicated alloys have been investi-
gated. Interpreting the results of experimental studies into the properties of materials
presents considerable difficulty due to the inherent complexity of their chemical
composition. Nevertheless, the authors suggest multiple solutions to these problems,
for instance, modeling as a way of anticipating the properties of materials.
Novel approaches to multiple factor analysis of processing initial and secondary
raw material including nonferrous metals have been proposed. These new strategies
are aimed to increase the yield of nonferrous metals. Dealing with the problem of
using secondary raw materials will serve to decrease the manufacturing cost and
address the environmental issues.
The dynamically developing field of composite materials creates ample opportu-
nities for their application not only in machine manufacturing, but also in building
engineering. Improvement of some properties of materials often leads to deteriora-
tion of others. In attempts to solve this problem, the authors used not only theoret-
ical, but also experimental approaches. Thus, combining scientific and practical data
has ensured their actual implementation in the field.
New patterns of high-pressure treatment of metals have been identified. The
authors have suggested various ways of treatment for large products such as tubes,
which, undoubtedly, presents both scientific and practical interest. The production of
modern materials is inseparable from the methods of high-pressure treatment;
therefore, the outlined techniques present high significance for manufacturing
high-quality products.
v
vi Preface
The editors would like to express their gratitude to all the authors for their
contribution, and to the Editorial Board and other scientists who reviewed the papers
and thus ensured the quality of this book.
This book has been published with financial support of the key Centre of
Excellence “Industrial mining institute of scientific research and projects” by Act
211 Government of the Russian Federation, contract № 02.A03.21.0006, which is
part of the program aimed at enhancing competitiveness of Ural Federal University
named after the first president of Russia B.N. Yeltsin from 2013 to 2020.
The editors also gratefully acknowledge the support of Vice-Rector in Science of
Ural Federal University V.V. Kruzhaev and Director of Institute of Materials
Science and Metallurgy of Ural Federal University V.A. Maltsev.
vii
viii Contents
1.1 Introduction
There has been a great body of research work (e.g., [1–4]), which, based on some
physical conceptions, provides numerous equations to describe the kinetics of
undercooled austenite decomposition occurring by pearlitic and bainitic transforma-
tion mechanisms. In the majority of cases, these equations provide a rather inaccu-
rate description of the processes actually running in steels. That can be explained by
the multicomponent compositions of steels, featuring a diversity of shapes and sizes
of their constituent phases. On the other hand, an enormous pool of experimental
data has been accumulated on the kinetics of processes in steel upon cooling
[5, 6]. These data are typically reported in the form of various TTT diagrams of
undercooled austenite. The TTT diagrams plotted under isothermal conditions allow
a distinction to be made between pearlitic and bainitic transformations and explicitly
describe the undercooled austenite stability for a specific steel grade. However, such
isothermal TTT diagrams in their graphic representations are difficult to use for
quantitative calculations of temperature fields and structures occurring in the heat
treatment processes.
The present study proposes a phenomenological approach that, based on the
experimental TTT diagrams, makes it possible to determine a set of parameters
that are explicitly described with the kinetic curves.
where
f is the volume fraction of the transformed austenite
A and n are called A and n parameters
τ is the transformation time.
The temperature dependence of the A parameter is suggested to be described with
the Arrhenius equation as follows:
Q
A ¼ A0 exp : ð1:2Þ
RT
where
A0 is the preexponential factor
Q is the process activation energy
R is universal gas constant
T is the absolute temperature.
The process activation energy Q includes a thermodynamic factor (ΔG being the
energy for critical nucleation) and a kinetic factor (Qdif being the diffusion activation
factor):
Q ¼ ΔG þ Qdif : ð1:3Þ
While the latter factor is undoubtedly the constant value, the former factor
depends heavily on temperature. However, this dependence is of a complicated
nature as it comprises the processes of nucleation and growth of a new phase. To
describe this temperature dependence of the thermodynamic factor, phase equilib-
rium diagrams are used, and thermodynamic models are calculated. [7–9]. All these
instruments yield rather complicated expressions, which appear difficult to be
applied to real systems; however, all the referenced studies have demonstrated that
altering the free energy with a new phase formation is accomplished with a degree of
supercooling ΔT from the phase equilibrium temperature.
The present study reports about the process activation energy Q determined
through analyzing the TTT diagrams, and an effort is made to describe the temper-
ature dependence thereof.
1 Kinetic Parameters and Activation Energy of Pearlitic and Bainitic. . . 5
The existing TTT diagrams were digitized using standard programs. Figure 1.1
displays TTT diagrams plotted for the steels of the grades 90Kh2N2 M (content of
major alloying elements: C-0.93; Mn-0.71; Cr-1.70; Ni-2.10; Mo-0.20) and U10G
(content of major alloying elements: C-0.96; Si-0.32; Mn-0.55; Cr-0.11; Ni-0.08).
For each steel, the dependence of the degree of transformation f on τ was determined
for as a great number of temperatures as possible. Figure 1.2 displays the dependence
of f on ln τ.
From the curves describing the dependence of ln (1/1-f) on ln τ (Fig. 1.3), A and
n parameters were determined in the Avrami equation.
The values of n were averaged and are presented in Table 1.1; and the curves of
ln А are displayed in Fig. 1.4.
Apparently, the temperature dependence of ln А is to a great degree retracing the
С-curve diagrams.
Analyzing the curves in Fig. 1.4, the activation energy was determined for the
processes. To this end, a dependence of ln A on 1/T (where Т is the absolute
temperature) was established. For each temperature range in the curve, the value
of Q/R was determined by the curve’s slope. The calculated values as a function of
temperature are shown in Fig. 1.5.
It can be seen that, for a bainitic transformation, regardless of whether it is
distinguished from the pearlitic transformation on the C-curve or not, the activation
energy shows little dependence on temperature for both the 90Kh2N2 M steel and
the U10G steel, and in the first approximation, the activation energy can be assumed
constant. For the pearlitic transformation, the activation energy changes dramatically
6 M. Gervasyev et al.
f, %
(b) U10G 60
40 400
20 550
0 600
0 5 10 15
Ln τ 650
b 120 190
100 200
80
300
f, %
60
40 400
20 500
0 600
0 2 4 6 8 10
Ln τ 650
0
0 5 10 15 400
-2
550
-4
600
-6
650
-8
Ln τ
b 4 190
2 200
300
LnLn(1 /(1 -f))
0
0 2 4 6 8 10 400
-2
-4 500
600
-6
650
-8
Ln τ
Ln A
-20
-30
-40
temperature, °C
b 0
0 200 400 600 800
-10
Ln A
-20
-30
-40
temperature, °C
20
Q /R
10
0
0 200 400 600 800
-10
temperature,ᵒC
b
60
40
Q /R
20
0
0 200 400 600 800
-20
temperature,ᵒC
perlite bainite
1 Kinetic Parameters and Activation Energy of Pearlitic and Bainitic. . . 9
The approach suggested in this study makes it possible to analyze the kinetics of
pearlitic and bainitic transformations.
In the pearlitic transformation, the effective activation energy of the process
changes dramatically (see Fig. 1.5). The shape of this curve approaches the shape
of a hyperbola as in the dependence on the supercooling, i.e.:
K perl
ΔG ¼ : ð1:4Þ
ΔT m
where
Kperl is constant for the pearlitic transformation
ΔTm ¼ A1– T (A1 is the temperature of the onset of the pearlitic transformation).
The study demonstrates that the exponent m for the pearlitic transformation is
approaching 2, which is in agreement with the data known from the reference
literature [1].
The bainitic transformation, unlike the pearlitic transformation, is not developing
to its completion; therefore, the Avrami equation was used in this study to describe
the kinetics up to the maximum possible degree of the bainitic transformation
development fbmax, i.e.:
f b ¼ f bmax ð1 expðAτn ÞÞ: ð1:5Þ
where
Kb is the bainitic transformation constant
ΔТb ¼ Bs – T (Bs is the onset temperature of the bainitic transformation)
A1 and Bs were determined from the ТТТ diagrams.
The activation energy calculated with A parameter under the bainitic transforma-
tion is practically constant, i.e., Q ¼ Qdif. The parameters calculated for the specific
steels can be used in quantitative calculations for the structure formed through the
sections of a workpiece under various heat treatment regimes.
10 M. Gervasyev et al.
References
1. Christian, J. W. (1975). The theory of transformations in metals and alloys, part 1, equilibrium
and general kinetic theory. Oxford-New York-Toronto-Sydney-Paris-Braunschweig: Pergamon
Press.
2. Lyubov, B. Y. (1981). Diffusion processes in heterogeneous solid medium (p. 296). Moscow:
Nauka. (in Russian).
3. Hillert, M. (1975). Diffusion and interface control of reaction in alloys. Metallurgical Trans-
actions, 6A, 5–13.
4. Umemoto, M., Komatsubara, N., & Tamura, I. (1980). Prediction of hardenability effects from
isothermal transformation kinetics. Journal of Heat Treating, 1(3), 57–64.
5. Atkins, M. (1977). Atlas of continuous cooling transformation diagrams for engineering steels
(p. 260). Rotherham (Yorkshire): BSC.
6. Popova, L. A., & Popov, A. A. (1991). Diagrams for austenite transformation in steels and for
β-solution transformation in titanium alloys (p. 502). Moscow: Metallurgia. (in Russian).
7. Hillert, M. (1982). An improved model for discontinuous precipitation. Acta Metallurgica, 30,
1689–1696.
8. Bhadeshia, H. K. D. H. (2001). Bainite in steels. Transformations, microstructure and properties
(p. 454). London: IOM Communication Ltd.
9. Mirzaev, D. A., Schastlivcev, V. M., Yakovleva, I. L., & Okishev, K. Y. (2000). The kinetics of
isothermal ferrite formation in Fe-9%Cr alloys of different purity. Materialovedeniye, 1, 39–42.
(in Russian).
Chapter 2
Alloying Addition Effect on the Structure
and Behavior of a Cu-Ni-Zn Three-
Component System Alloy Under Lost-Wax
Casting
2.1 Introduction
Due to high corrosion resistance and endurance, nickel silver is generally used for
the production of molded pieces for precision instruments and medical equipment
[1]. Moreover, Cu-Ni-Zn ternary alloy is of an attractive silver color, which is the
reason why nickel silver is frequently used for manufacturing metal artwork such as
commemorative coins, medals, and award pins. The strength and wear resistance of
nickel silver (copper 65%, nickel 15%, zinc 20%) are higher than those of alloys
(silver 92.5%, copper 7.5%) [2].
Mechanical and technical properties of nickel silver are determined in sufficient
detail, as Cu-Ni-Zn alloys are most commonly used as wrought metals. Moderniza-
tion of precision instruments has determined the necessity of applying figurine-
shaped details, which are produced by lost-wax process. However, thin-walled
nickel silver casting can sometimes lead to short-run casting (Fig. 2.1). Therefore,
studying the characteristics of Cu-Ni-Zn alloy casting under the lost-wax process is
of particular interest.
The first stage of the research involved studying casting mold fillability by Cu-15Ni-
20Zn alloy depending on casting temperature. For this purpose the method of
estimating casting mold fillability was developed. It consists of the determination
Fig. 2.1 Short-run nickel silver (copper 65%, nickel 15%, zinc 20%) castings
of filled cell fraction in a perforated plate under the lost-wax process [3]. In order to
produce wax models, a perforated plate model was designed by means of Solid Edge
software (Fig. 2.2), and then a master model was constructed by a 3D printer, Objet
Connex 350, using the fast prototyping method. After that the rubber mold was
prepared, and the wax models were obtained and clustered.
Melting and casting were conducted by an INDUTHERM-MUVV700 machine.
Annealing of casting mold was carried out under the same stage heating conditions.
The temperature of all molds before casting was set at 560 C. Nickel silver (copper
65%, nickel 15%, zinc 20%) liquidus temperature is 1112 C; hence, the casting
temperature was set at Tc ¼ 1150 C, 1170 C, 1190 C, and 1210 C. The
perforated plates resulting from the experiments are presented in Fig. 2.3.
The analysis of cast perforated plates revealed that the increase of casting
temperature results in better alloy fillability of molds, which corresponds to the
general notion about the impact of casting temperature on alloy flowability.
The next stage of the research was studying dopant influence on casting mold
fillability, nickel silver (copper 65%, nickel 15%, zinc 20%) castings, surface
2 Alloying Addition Effect on the Structure and Behavior of a Cu-Ni-Zn. . . 13
Fig. 2.3 Nickel silver (copper 65%, nickel 15%, zinc 20%) castings perforated plate
2.3 Results
The results of the first stage of the experiment devoted to casting mold fillability with
Cu-15Ni-20Zn alloy depending on casting temperature are presented in Table 2.1.
As Table 2.1 shows, the best alloy fillability of investment molds is observed
when the casting temperature is 1190 C, i.e., at melting temperature above liquidus
line up to 75–80 C. In this case, fillability is 94%. When the casting temperature is
1210 C, alloy fillability of molds is lower, which is probably due to oxide film
(ZnO) formation on the metal bath surface. Moreover, the high casting temperature
causes intensive zinc evaporation, which prevents ingate mold filling in and leads to
pinhole formation on casting surface.
Alloying 4.0 mass % Sn and 0.4 mass % Al dopant allows reaching 100% casting
mold fillability, which is demonstrated in Table 2.2.
As Table 2.2 shows, the best alloy fillability of investment mold is observed when
the nickel silver is alloyed with aluminum and tin. Moreover, the surface of a casting
sample presented in Fig. 2.4a demonstrates surface irregularity and gritty scale.
More than likely it is also explained by intensive zinc evaporation under at high
2 Alloying Addition Effect on the Structure and Behavior of a Cu-Ni-Zn. . . 15
Fig. 2.6 Nickel silver sample microstructure: (a) without dopant; (b) with 4.0 mass % Sn and 0.4
mass % Al dopant
casting temperature. The casting sample surface shown in Fig. 2.4b is clean, without
any irregularities. It is probably connected with aluminum deoxidizing power on
copper alloy, as well as the formation of aluminum oxide layer, which prevents zinc
evaporation and decreases zinc loss upon melting.
Furthermore, the dopant alloyage makes it possible to obtain fine-grained micro-
structure of nickel silver casting samples (Fig. 2.6). Grain size of nickel silver sample
alloyed by aluminum and tin is 75–100 μm, which is 2.0–2.5 times smaller than
grain size of nickel silver without alloying.
16 I. A. Gruzdeva and S. S. Gerasimov
The nickel silver sample alloyed by aluminum and tin has more compact, close-
grained, and ordered microstructure as opposed to porous and random microstruc-
ture of nickel silver sample without dopant. It can be explained by the capacity of
aluminum to decrease the solidification range of copper-nickel-zinc alloys, and as a
result, the crystallization of liquid nickel silver melt proceeds more regularly.
In the final stage of the study, we attempted to decrease casting temperature of
Cu-15Ni-20Zn-4Al-0.4Sn alloy, since lower temperature means lower loss of fusible
elements and, therefore, less energy consumption. The results of the analysis are
presented in Table 2.3.
As Table 2.3 shows, 100% alloy fillability of molds is observed when the casting
temperature is 1170 С. Most probably, low-melting elements reduce nickel silver
liquidus temperature, which allows decreasing casting temperature. Moreover, cast-
ing at 1150 С provides sufficient quality casting.
2.4 Conclusions
References
1. Kurdumov, A. V., Pikunov, M. V., Chursin, V. M., & Bibikov, E. A. (1986). Castings
production from non-ferrous alloys (pp. 217–218). Moscow: Metallurgy.
2. Osintsev, O. E., & Fedorov, V. N. (2004). Copper and copper alloys (pp. 186–187). Moscow:
Mechanical engineering.
3. Gruzdeva I. A. (2013). Casting mold fillability studying by alloy МНЦ15-20 under the lost-wax
process. In S. S. Gerasimov, A.V. Sulitsin, (Eds.), Proceeding of the 13th Conference Russian
Founders, (pp. 254–259), Ekaterinburg.
Chapter 3
Structural Changes in Austenitic-Ferritic
Steel 03Kh14N10K5M2Yu2T During
Heating and Cooling
3.1 Introduction
Developed austenitic-ferritic steel has the necessary strength, hardness, high pro-
cessability and high corrosion resistance. It is intended for springs and elastic
elements. Earlier studies [1, 2] have shown solidification through δ-ferrite in the
investigated steel during crystallization. 03Kh14N10K5M2Yu2T steel’s structure
consists of two main phases – austenite and δ-ferrite in approximately equal amounts
(50/50) after hardening at 1000 C. It has been found [2] that δ-ferrite has an
abnormally high hardness ( 500 HV) in the investigated steel. The hardness of
the austenite is low ( 200 HV). A superstructure reflection of δ-ferrite areas is
forbidden to δ-ferrite. The best description of superstructure is type B2 lattice. The
darkfield image of superstructure reflections (001) indicates the presence of round
shape superfine particles. It is distributed in δ-ferrite crystals. The phase ratio of
austenite and δ-ferrite ( 50/50) is stable in the temperature range of 800–1100 C
(Fig. 3.1a, b). The amount of δ-ferrite increases above the temperature 1100 C.
Ferrite grains increase at a temperature of intensive γ ! δ transformation (above
1200 C). This is due to exceeding a recrystallization threshold. Diffusion mobility
of bcc ferrite lattice atoms is more than austenite fcc lattice atoms.
The Chemical composition of the investigated steel (wt.%) is: 0.02 C; 14.2 Cr;
9.8 Ni; 5.1 Co; 2.5 Mo; 2.4 Al; 0.36 Ti; total content of S, P and Cu 0.35; rest
Fe. The melting of investigated steel ingots was carried out by open-type induction
furnaces. A homogenizing annealing was carried out. Ingots were forged to the
necessary diameter. Samples for tensile tests and bars for microsections were
prepared. Microstructural studies were carried out on the microscope OLYMPUS
JX-51, with 500-fold magnification; also, by electron microscopy using the
Fig. 3.1 Hardened steel 03Kh14N10K5M2Yu2T microstructure: (a) 1000 C (in water);
(b) 1100 C (water); (c) 1300 C (water); (d–h) laser tempering melting, (d) darkfield image
superstructural reflex (001) in the central part of the specimen; (e), (f) bright-field image;
(g) electron diffraction pattern; (h) scheme of indexing
3 Structural Changes in Austenitic-Ferritic Steel 03Kh14N10K5M2Yu2T. . . 21
Table. 3.1 shows the effect of quenching temperature on physical properties of the
investigated steel.
There was a significant increase of magnetic saturation (Ms) in proportion to
δ-ferrite at temperatures above 1100 С. The main phase is δ-ferrite at 1300 С.
Precipitations of reversed austenite formed due to insufficiently high cooling rate
on the ferrite background. Colonies of reversed austenite form Widmanstatten
pattern with characteristic needle (plate) structure (Fig. 3.1c). Such form of the
precipitating phase usually is typical for shear transformation [3]. Suppressing of
reversed austenite precipitation is not possible during cooling at 1300 С to water.
The maximum value of the coercive force Hc can be explained by structural
changes after quenching at 800 C (the features of δ ! γ transformation [3], as well
as the influence of new paramagnetic impurities in the austenite). This is impeding
the process of magnetization reversal. Hc minimum has been observed after hard-
ening at 1200 C and above. The steel structure becomes almost completely ferrite
and grain growth occurs.
Table 3.1 The effect of quenching temperature on physical properties of the investigated steel
T quenching, С (water cooling) Нс, А/сm Br, Tl Ms, kА/m
800 22 0.12 400
900 15 0.10 380
1000 15 0.10 400
1100 7 0.50 950
1200 5 0.70 1000
22 L. Mal’tseva et al.
Fig. 3.2 The microstructure of the steel 03Kh14N10K5M2Yu2T: (a) after laser treatment with a
reflow; (b) after laser treatment with the rehardening at 1000 C
The surface layer of the investigated steel has δ-ferrite single-phase structure
(Figs. 3.1d and 3.2a). The δ-ferrite hardness reaches 600–650 HV. The investigated
steel structure formation after laser treatment has been shown in [1]. However, it
should be noted some of structural features of the transition from δ-ferrite phase to
the ferritic-austenitic structure. Figure 3.1d–g shows that identical orientation frag-
ments of austenite are arranged at the periphery of weakly oriented subgrains of
δ-ferrite with dispersed spherical particles (Fe, Ni, Al) (on the outside of the pictures,
these fragments merge into one γ-grain).
The relative orientation of δ and γ lattices is close to the Kurdjumov-Sachs
orientation relationship, and extensive flat areas of phase boundary correspond to
the trace of the plane {111}δ || {110} γ.
Reheating to 1000 C leads to the formation of a two-phase austenitic-ferritic
structure, with very fine grains of austenite, in the δ-ferrite single-phase zone. This
indicates a high cooling rate and a large number of nucleation centres (Fig. 3.2b).
Reheating does not lead to visible changes in the microstructure in the case of the
equilibrium structure up to temperatures of about 700 C after quenching and can
only note the etch increasing (Fig. 3.3b). The heterogeneous structure of the phases
3 Structural Changes in Austenitic-Ferritic Steel 03Kh14N10K5M2Yu2T. . . 23
is observed during ageing above 650 C. This is due to the disintegration of the solid
solution (δ-ferrite), a partial conversion δ ! γ and intermetallic phase of a complex
composition. The presence of this phase (Fe74Cr12Ni14)2.09 (Mo64.4 Ti35.6) was
confirmed by the physical-chemical analysis of the precipitate of steel hardened at
1200 C and aged at a temperature of 650 C. An austenitic component is increased
(70–80%) to temperatures 700–750 C (Fig. 3.3d).
The differential scanning calorimetry (DSC) heating curves (Fig. 3.4) show two
exothermic peaks. The first (400–530 C) is associated with ageing in δ-ferrite, and
24 L. Mal’tseva et al.
DDSC*10-3/(uv/mg/min)
TG/% DSC/(uv/mg)
Peak 1627 °C exc
0.04
101.2
0
0.03
[1.1]
100.2
-8
[1.1]00116_Ai203-H6 Zak [01] -0.02
DSC
100.0
Value 928.0 °C
-0.03 -10
Value 911.0 °C
99.8
200 400 600 800 1000 1200
Temperature/*c
Fig. 3.4 Curves of differential scanning calorimetry of the investigated steel at heating and cooling
The δ-ferrite quantity increases at 1200–1250 C. The process of hot plastic defor-
mation occurs in a temperature range; therefore the forging heating temperature
might be at a temperature of 1200–1250 C, but the finish rolling temperature must
be at a temperature of 900 C.
3.3 Conclusions
References
1. Nemirovskii, Y., & Khadyev, M. (2009). Structure of 03Kh14Ni10K5M2Yu2T steel after laser
treatment. Metal Science and Heat Treatment, 51(5–6), 297–300.
2. Mal’tseva, L., & Arkhangel’skaya, A. (2010). Special features of hardening of austenitic-ferritic
steel after thermoplastic treatment. Metal Science and Heat Treatment, 52, 39–45.
3. Apaev, B., & Voronenko, B. (1974). The kinetics of phase transformations in austenitic-ferritic
steels. Iron and Steel, 8, 90–95.
4. Hong, C., & Ding, T. (2011). A new economical sigma-free duplex stainless steel 19Cr-6Mn-
1.0Mo-0.5Ni-0.5W-0.5Cu-0.2N. Journal of Iron and Steel Research International, 18(4), 52–57.
5. Michalska, J., & Sozanska, M. (2006). Qualitative and quantitative analysis of σ and χ phases in
2205 duplex stainless steel. Materials Characterization, 56, 355–362.
Chapter 4
Correlation of Instrumented Charpy Load-
Displacement Curves with Crack Growth
in API 5L Grade X80 Pipeline Steels
4.1 Introduction
Charpy impact testing is required to provide the necessary pipeline life resource
especially for operation in low-temperature climate regions [1, 2]. X80 grade steels
are used for gas pipelines with 11.8 MPa air pressure. It is known that such steels
have extremely high Charpy energy more than 250 J/cm2 at –40оС. Special cracks
called separations are observed on the fracture surface. A high amount of separations
parallel to rolling plane were found as the main feature of the fracture surface in X80
pipes. Evidently, fewer separations were formed in pipes with similar Charpy energy
but high arrestability. Based on these findings, the question is raised of whether the
Charpy test is suitable for predicting fracture arrest in high strength material.
Through subsequent refinements in testing and certification standards, the Charpy
test has been developed in accordance with the requirements of the engineering
community. Due to its low cost and convenience, there is a growing interest in using
the instrumented Charpy test to obtain a wide range of information about fracture
initiation, crack propagation mechanisms, and impact energy properties [3–6]. Accu-
rate determination of load and/or energy versus displacement behavior during the
instrumented impact test is essential to obtain such information [7, 8].
The following impact strength properties and crack propagation resistance of
low-carbon ultrahigh ductile steel will be further discussed in detail, including the
analysis of both smoothed load-displacement curves and fracture surfaces.
The material used for the present research was API X80 grade low-carbon (LC) steel
with 0.05 wt.%C – 1.5%Mn – 0.15%Si – 0.04%V chemical composition. After
TMCP and subsequent accelerating, cooling LC steel had ultrafine (~3–5 μm)
ferrite-bainite structure with ultrahigh toughness level.
Charpy specimens were machined from the middle pipe wall. Specimen geomet-
rical dimensions were those of typical full size V-notch, that is, 10x10x55 mm,
according to specification ASTM E23.
Charpy impact tests were carried out in temperature range from +20 to –80 C on
an Instron Ceast 9350 instrumented droptower machine with maximum impact
energy of 757 J supplied by data acquisition system (DAS) Ceast 16000. The data
were acquired using a sampling frequency of 1.0 MHz and a point number of 8000
with the sampling time of 0.001 ms. Load vs. displacement curves were smoothed by
a hardware filter with 2 kHz frequency on the DAS to reduce the effects of proper
oscillations of the specimen-striker elastic system.
The fracture surfaces of Charpy impact specimens were investigated via visual
observation with fivefold magnification and were subjected to identification and
measurement with 0.1 mm accuracy of the next fracture zones [9]:
– Near to notch shear zone Θ
– Homogeneous ductile zone Lh
– Fiber ductile zone Lf
– Brittle zone of Lbr
– Ductile final break zone Lfb
– Shear tips λ
Impact load vs. displacement curves are shown in Fig. 4.1. The impact load behavior
on the elastic displacement stage has been described by the slope angle α1 (tg α1)
named as elastic rebound compliance (RCel) and the level of general yield point Fgy.
It can be seen that RCel is constant and does not depend on test temperature up to the
Fgy level, equal to 18.5 kN, at any test temperature.
It is evident that typical impact load points Fgy, Fm, Fiu, and Fa can be correctly
construed on these traces, but it is surprising that there are two maximums for
ultraductile steel. Appearance of the first peak like yield drop on the tensile test
curve was induced by crack initiation from point Fgy to point Fmax1 and crack
propagation in shear zone Θ from point Fmax1 to point Fmin. Typical for steels at
any test temperature, the shear zone is caused by severe plastic deformation near to
notch (Fig. 4.2). Shear lips λ are plastic deformation products as well, but their width
depends on test temperature – the lower the test temperature, the lower the shear lips’
width.
4 Correlation of Instrumented Charpy Load-Displacement Curves. . . 29
F,kN Fmax2
Fmax1 T= +20°C
T= -40°C
20 T= -60°C
Fgy T= -80°C
Fmin
Fiu Fiu
15 Fgy
α2
Fmax1
Fmin
10 α2 Fa Fa Fmax2
Fiu
α1 Fa
Fa
5
0
0 5 10 15 20 S, mm
The second peak occurrence is a result of the interaction between the crack and
the plastic deformation zones ahead the tip of a crack. This peak disappeared when
the plastic deformation zone could not resist crack propagation at a lower test
temperature. Curve stage size from points Fmin to Fiu is correlated with crack
propagation in homogeneous zone Lh due to a ductile stable fracture.
Valuable information can be obtained from the analysis of an unstable crack
growth stage from point Fiu to point Fa which is correlated with fiber zone Lf,
(Fig. 4.2). Rebound compliance (RCiu ¼ tg α2) of segment FiuFa led to a growing
portion of brittle fracture component at a lower test temperature.
Figure 4.3 shows the temperature dependences of the fracture zone length of API
X80 steel. It can be seen that the lengths of both shear zone Θ and ductile final break
zone Lfb are only slightly sensitive to the change of test temperature, whereas the
lengths of both the homogeneous ductile zone Lh and fiber ductile zone Lf change
noticeably. The decrease in test temperature from 20 to –60 C leads to a decrease
in the length of the Lh zone and to an increase in the length of the Lf zone.
The correlation between Charpy impact toughness and load-displacement curve
characteristic parameters Fmax2 and RCiu is shown in Fig. 4.4. It is well known that
steels that are subjected to a thermomechanical controlled process often show a
30 V. Khotinov et al.
6.0
4.0
Lh
Lf
3.0
2.0
Lfb
1.0
Q
0.0
-100 -80 -60 -40 -20 0
Test temperature, оС
KCV,
500 Fmax2,
25
J/cm2 kN,
RCiu
Fmax2 upper
400 20
plateau
300 15
rising
shelf
200 10
100 – KCV; 5
– Fmax2;
lower
RCiu – RCiu
plateau
0 0
-120 -100 -80 -60 -40 -20 0 20 Т, 40
°С
Fig. 4.4 Correlation of impact toughness KCV and instrumented impact parameters Fmax2 and RCiu
in brittle-ductile temperature transition of LC steel
“rising shelf” behavior when upper plateau, rising shelf, transition stage, and lower
plateau on transition curve are observed [10, 11].
It is seen that Fmax2 degeneration is well correlated with KCV on the upper plateau
and corresponds to a decrease of Lh zone on the Charpy fracture surface (Fig. 4.4).
Slope angle RCiu accounts for an increase in the brittle fracture component – the
4 Correlation of Instrumented Charpy Load-Displacement Curves. . . 31
more brittle the fracture, the bigger the RCiu value. At the same time, the fiber
fracture zone Lf expanded.
It is clear that the most complete information about material crack resistance can
be obtained by using both Fmax2 and RCiu values as the most sensitive and unique
characteristics for crack propagation. Analysis of these fracture parameters has
clearly shown that fine-grained ferrite-bainite LC steel has a very high crack
resistance.
4.4 Conclusions
References
1. Arabey, I. Y., Pyshmintsev, I. Y., Farber, V. M., Khotinov, V. A., & Struin, A. O. (2012).
Failure of pipe steel of X80 (K65) strength class. Steel in Translation, 42(3), 212–218.
2. Farber, V. M., Pyshmintsev, I. Y., Khotinov, V. A., et al. (2010). Proc. of XVIII Int. Pipeline
Conf. (pp. 108–117). Chelyabinsk: RosNITI.
3. Manahan, M. P., & Stonesifer, R. B. (2000). The difference between total absorbed energy
measured using an instrumented striker and that obtained using an optical encoder, ESIS/ASTM
STP1380 (pp. 181–197).
4. Tronskar, J. P., Mannan, M. A., & Lai, M. O. (2002). Measurement of fracture initiation
toughness and crack resistance in instrumented Charpy impact testing. Engineering Fracture
Mechanics, 69, 321–328.
5. Kobayashi, T., Inoue, N., Morita, S., & Toda, H. (2000). On the accuracy of measurement and
calibration of load signal in the instrumented Charpy impact test, ESIS/ASTM STP1380
(pp. 198–209).
6. Viehrig, H. W., Boehmert, J., Richter, H., & Valo, M. (2000). Use of instrumented Charpy test
for determination of crack initiation toughness, ESIS/ASTM STP1380 (pp. 354–364).
7. Steel Charpy, V. (2000) Notch pendulum impact test – Instrumented test method, ISO 14556.
8. Standard test method for instrumented impact testing of metallic materials, ASTM E2298
(2013).
9. Botvina, L. R. (2008). Fracture: Kinetics, mechanisms and total regularities. Moscow: Nauka
publishing house.
10. Mannucci, G., & Demofonti, G. (2011). Control of ductile fracture propagation in X80 gas
pipeline. Journal of Pipeline Engineering, 10(3), 133–145.
11. Wilkowski, G., Rudland, D., & Wolterman, R. (2005). Predicting the brittle-to-ductile fracture
initiation transition temperature for surface-cracked pipe from Charpy data. In Proc. of ASME
Pressure Vessels and Piping Conf. 6 (pp. 545–550). Denver, USA.
Chapter 5
Damping Behavior
of Aluminum Replicated Foam
5.1 Introduction
W. Riehemann · U. Arlic
Technical University Clausthal, Clausthal-Zellerfeld, Germany
A. Finkelstein (*) · D. Husnullin
Ural Federal University, Yekaterinburg, Russia
5.2 Review
Conversion of elastic vibration energy into internal friction heat is the main mech-
anism of damping. There are some values to characterize material damping:
• Damping capacity, which is a ratio of converted vibration energy to full energy of
load-unload cycle [5].
• Loss tangent tan ϕ, where ϕ is the phase lag between stress and strain. The
technique was described in [6], for example.
• Logarithmic decrement of decaying vibration, see Eq. (5.1):
An
δ ¼ ln : ð5:1Þ
Anþ1
where δ is logarithmic decrement of decaying vibration, and An and An+1 are adjacent
amplitudes of decaying vibration.
For different porous metals wide experimental investigations were carried out by
Golovin, Riehemann et al. in Clausthal Technical University. They based their
measurements of damping on the logarithmic decrement technique [7], which has
also been applied for the present research.
5 Damping Behavior of Aluminum Replicated Foam 35
Samples for dumping tests were prepared by the replication casting technology. This
is similar to a chill casting process, but in addition to that, a salt preparation branch is
implemented in the technological cycle. The technological scheme of production is
shown below (Fig. 5.2).
Commercial aluminum alloy AlSi7 (AK7 GOST 1583–93 grade in Russia) was
used for producing samples. Apart from AK7, self-production scrap from feeder
heads was used (25–30 wt. % of charge). The melting procedure was conducted in
resistance-heated pot-type furnace. The melt was overheated up to 720–750 C.
Sodium chloride particles were sieved by fractions using DIN standard sieves
(0.1–0.2 mm, 0.2–0.315 mm, 0.315–0.63 mm, 1.0–1.6 mm) and then heated in
ceramic cells in a chamber furnace up to 600–650 C. After that a portion of
preheated NaCl was placed into a metal mold and compacted by vibration. The
temperature of the metal mold was in the range 350–450 C to provide mold
fillability. At the next stage, molten aluminum was poured over the filler into the
metal form. And then the infiltration of NaCl layer was carried out under pressure
drop between the vacuum receiver (300 hPa) and the atmosphere (1024 hPa).
Finally, finished solidified castings were removed from the mold. After that they
were machined to obtain samples for mechanical tests according to Standard Test
Methods for Tension Testing of Metallic Materials (ASTM E8/E8M) and for
damping tests, Fig. 5.3. Samples for damping measurements are thin plates of
3 mm thickness with cylindrical tip.
To get the metal porous structure, NaCl was removed from the composite sample
by dissolving salt particles in water with a forced convective stream over the samples
surface. The porosity of finished samples was determined indirectly by weight.
It is worth mentioning that the porous structure of replicated aluminum samples
obtained by vacuum impregnation technique may be characterized as homogenous
and open cell.
For more information on replication casting technology, see Refs. [3, 4, 8].
As a rule, the measurement of damping (δ) of porous materials like foams is not
trivial. Subjecting porous materials to stable vibrations might cause difficulties.
Extended quantity of samples is needed to perform representative measurements
for the whole material. Moreover careful machining and clamping without plastic
deformation is necessary. Due to the extended surface of the porous material, surface
layers of oil or other liquids have to be avoided. In addition, damping is often
confused with acoustical silencer processes: reflection, refraction, and dispersion of
sound. In this study solely material damping of porous metal AlSi7 was measured
and discussed.
Material or solid state damping (δ) was measured in the air of normal pressure as
the logarithmic decrement of free bending beam vibrations. The single cantilever
bending beams fixed at one end were excited into the first bending resonance mode
by means of a permanent magnet fixed at the free end and an exciting coil.
Logarithmic decrement of decaying vibrations was obtained from time dependency
of voltage induced in coil by a permanent magnet. Further details about the exper-
imental setup are described elsewhere [5, 7].
In addition to the abovementioned capability, the experimental setup is able to
measure the elastic modulus of materials [9].
5 Damping Behavior of Aluminum Replicated Foam 37
The results of the damping measurements are presented in Fig. 5.4, where the
damping is plotted versus maximum strain amplitude of the vibrating bending
beams for samples with various pore sizes. The numbers of successive measure-
ments after excitation to vibration are also presented.
The damping curves for the samples with pore sizes up to 1.6 mm show typical
metallic behavior with practically amplitude-independent damping value of 0.002 up
to strain amplitudes of about 104. For higher values of strain amplitude strong
monotonously increasing damping is observed. This behavior can be accounted for
progressive dislocation break away from weak pinning points with increasing strain
amplitude, as it has already been explained by the theory of Granato and Lücke
[10]. It is amazing that for maximum pore sizes in the range of 0.63–1.6 mm no
systematic dependency of damping on porosity was noticed. In earlier investigations
concerning the damping of aluminum foams stronger dependencies on foam size
were found [5, 11]. For our samples the statistical fluctuation of strain amplitude-
dependent damping established between different samples of the same mean pore
size (1.5 mm) is even higher than the differences between samples of different pore
sizes (0.315–1.6 mm max).
Contrary to this stable and invariant damping behavior, the whole strain
amplitude-dependent damping curve changes dramatically for samples with bigger
pores (>3.0 mm), as can be noted for measurement #1 and the directly following
measurement #2, shown in Fig. 5.4, at the same sample with >3.0 mm pore size.
Fig. 5.4 Damping versus maximum strain amplitude of porous bending beams for various max-
imum pore sizes, samples, and numbers of successive measurements after excitation to vibration
38 W. Riehemann et al.
Table 5.1 Young’s modulus of AlSi7 porous material with different pore sizes
Pore size range Porosity, % Young’s modulus, MPa
Nonporous AlSi7 alloy 0 71,000
0.35–0.63 mm 52 4410
1.00–1.60 mm sample #1 52 3230
1.00–1.60 mm sample #2 53 2960
1.60–3.00 mm measurement #1 54 1910
1.60–3.00 mm measurement #2 54 1900
The damping of these samples proved to be much higher, especially for strain
amplitudes in the medium strain amplitude range. Moreover, it does not necessarily
go up monotonously, as the strain amplitude increases, and it changes during the
measurement itself. This damping behavior is also found in metal foams [5, 11] or
other materials with cracks [12], and it demonstrates time- and temperature-
independent hysteretic behavior instead of time- and temperature-dependent inelas-
tic behavior determined for dislocation damping. The hysteretic damping curves can
also be explained by a simple rheological model [13].
While both the damping and the damping behavior change erratically for mate-
rials with pore sizes between 1.6 mm and 3 mm, the Young’s modulus decreases
continuously with increasing pore size in the whole investigated pore size range.
This is evident from Table 5.1, where the elastic moduli of the materials with various
maximum pore sizes are listed and compared with the bulk material as well.
The damage of the material with the coarse pores (1.6–3.0 mm) by the vibrations
of the damping measurements can be detected by the small drop of the modulus due
the vibration (1910–1900 MPa), but it is obvious that the change in damping due to
fatigue is much higher and therefore much more sensitive to the detection of cracks.
To compare damping of replicated porous aluminum, we took into account some
information on damping behavior of the main competitive material, that is, foamed
aluminum Alp-85 [9] and current results for coarse pore (1.6–3 mm) replicated
aluminum. For full understanding, pure aluminum data were added (see Table 5.2).
Obviously, Alp-85 has a higher damping due to higher porosity but low level of
mechanical properties. That means the higher loadings on damper, the more volume
of material demanded. In this respect, replicated aluminum appears to be more
convenient material to use for damping and impact absorbing.
5.6 Conclusions
Table 5.2 Comparison of mechanical properties for porous and compact aluminum
Replicated porous
Material Pure aluminum Alp-85 aluminum
Type of structure – Close cell Open cell
Pore size, mm – 4.0–5.0 1.6–3.0
Damping at strain amplitude 105 0.002 0.040 0.005
104 0.002 0.050 0.01–0.03
Density, kg/m 2700 400 1460
Porosity, % 0 85 54
Specific Young’s modulus, (GPa*m3)/kg 0.0259 0.0055 0.0013
Young’s modulus, GPa 70 2.2 1.9
Ultimate tensile strength, MPa 70 2.5 7.7
Acknowledgment The authors would like to acknowledge Composite Materials Ltd. (Kirovgrad,
Russia) for the kind assistance in sample production and machining.
References
1. Kuchek, H. A. (1964). Patent US 3138856 Method of producing clad porous metal articles.
2. Despois, J.-F. (2005). Replicated aluminium foam, processing and properties, Ecole
Politechnique Federale de Lausanne, p. 265.
3. Furman, E. L., Finkelstein, A. B., & Cherny, M. L. (2013). Permeability of aluminium foams
produced by replication casting. Metals, 3(1), 49–57.
4. Furman, E. L., Finkelstein, А. B., & Cherny, M. L. (2014). The anisotropy of replicated
aluminum foams. Advances in Materials Science and Engineering, 1–6.
5. Golovin, I. S., Sinning, H. R., Göken, J., & Riehemann, W. (2003). Amplitude dependent
damping of some metallic foams. Solid State Phenomena, 89, 267–272.
6. Zhang, Y., Ma, N., & Wang, H. (2007). Effect of particulate/Al interface on the damping
behavior of in situ TiB2. Materials Letters, 61, 3273–3275.
7. Göken, J., & Riehemann, W. (2002). Thermoelastic damping of the low density metals AZ91
and DISPAL. Materials Science and Engineering A, 324(1–2), 134–140.
8. Kazantsev, S. P., & Husnullin, D. V. Technological processes of obtaining of replicated
aluminium foam. Contemporary Engineering Sciences, 8(16), 723–727.
9. Golovin, I. S., & Sinning, H.-R. (2003). Damping in some cellular metallic materials. Journal of
Alloys and Compounds, 355(1–2), 2–9.
10. Granato, A., & Lücke, K. (1956). Theory of mechanical damping due to dislocations. Journal of
Applied Physics, 27(6), 583.
40 W. Riehemann et al.
11. Golovin, I. S., Sinning, H.-R., Göken, J., & Riehemann, W. (2001). Mechanical damping of
some Al foams under cyclic deformation. Proc. of MetFoam, Int. Conf. on Cellular Metals and
Metal Foaming Technology, 323–328.
12. Golovin, I. S., Sinning, H.-R., Göken, J., & Riehemann, W. (2004). Fatigue related damping in
some cellular metallic materials. Materials Science and Engineering A, 370(1–2), 537–541.
13. Göken, J., & Riehemann, W. (2004). Damping behaviour of AZ91 magnesium alloy with
cracks. Materials Science and Engineering A, 370(1–2), 417–421.
Part II
Non-ferrous Metals Metallurgy due to the
Refining of Secondary Raw Materials
Chapter 6
Recovery of Lead and Zinc from Fine Dust
of the Copper Smelting Industry Using
a Chelating Agent
6.1 Introduction
Small-scale tests of fine dust leaching were carried out in thermostated setup
provided with immersion stirrer, pH meter, and temperature compensator system
for pH automatic maintenance (Fig. 6.3). Calculated amount of dried, milled, and
2
250 2 2
3
Intensity (pps)
1 2 6
200
4
150
1 4 1 3
1 5 5 2
100
50
0
10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90
2θ
Fig. 6.1 Diffractogram of fine dust: 1 – ZnO, 2 – PbSO4, 3 – ZnFe2O4, 4 – Cu2O, 5 – As2O3, 6 –
FeAsO42H2O
screened out (fraction 0.1) dust was leached by water solution of OEDP
(1.5 mol/l). A value of pH ¼ 11 was chosen due to conditions of OEDP selective
complexation with lead. The pH value was held constant during tests by dosing
necessary alkali volume.
After the liquid was tested and solid phases were separated, the received cake was
dried at 383 K and weighed. The cake was triturated and averaged. Achieved
material was pelleted and analyzed by X-ray diffractometer XRD 7000C (Shimadzu)
for lead, zinc, copper, and iron concentration. Filter liquor was analyzed by the
atomic absorption method (spectrometer novAA 300).
On the basis of the cake and filter liquor analysis, we calculated metal extraction.
The resultant extraction value was taken up as the average extraction value within
the accuracy of 5%.
By means of an experiment design, the array of dust leaching parameters was
formed. The following parameters were varied: liquid-to-solid ratio (X1) from 3 to
7 and process temperature (X2) from 298 to 353 K. The extraction of lead into
solution as main component and extraction of zinc, copper, and iron were controlled.
6 Recovery of Lead and Zinc from Fine Dust of the Copper Smelting. . . 45
Fig. 6.3 Setup for dust leaching. 1 – Thermostat, 2 – thermostat stirrer, 3 – thermostat heater, 4 –
contact thermometer, 5 – control module and driver thermostat stirrer, 6 – control module and driver
stirrer, 7 – volume burette, 8 – upright, 9 – pH meter, 10 – combination electrode, 11 – temperature
compensator, 12 – stirrer, 13 – pulp, 14 – reacting glass
With the aim of examination of complex existence, there were received residuals
after leaching of pure components of dust. The IR spectra of separated compounds
and pure OEDP were recorded on FTIR (Fourier transform infrared) spectrometer
ALPHA-T.
Fig. 6.4 Influences of temperature and pulp density on extraction: (a) lead, (b) zinc, (c) copper,
(d) iron
zinc (Fig. 6.4b), copper (Fig. 6.4c), and iron (Fig. 6.4d) extractions on temperature
(X1) and pulp density (X2).
The relations of metal extraction into solution to temperature have extremal
character. Maximum magnitude of extraction was observed in the area of elevated
temperatures. It is possible that leaching process is limited by chemical interaction
step. It should be pointed out that lead extraction reached 96%, whereas a maximum
zinc extraction under the same conditions is 20%, and impurities extraction is equal
to 0.1–3.0%. This phenomenon allows one to assume lead complexation selectivity
at the alkaline conditions.
Maximum lead, zinc, and iron extractions were received at the conditions of
middle diluted pulps (l:s ¼ 5:1), while copper extraction is increased along with pulp
density reduction up to 7:1.
6 Recovery of Lead and Zinc from Fine Dust of the Copper Smelting. . . 47
Probably, the value of copper extraction depends not only on pulp density but
also on pH area of complexation with OEDP and MeHL complex indissolubility in
broad pH area [9].
0.995
0.985
0.975
0.965
6.4 Conclusions
In this paper, features of metallurgical fine dust leaching was studied. Optimal
parameters of lead and zinc leaching were established by using chelating reagent;
these parameters are initial OEDP concentration 1.5 mol/l, pH ¼ 11–12, pulp density
5:1, and 303–323 K. Under these conditions, metal extractions in solution are Pb
96–98%, Zn 15–20%, Fe 1–3%, and Cu 0.05–0.1%. By means of IR spectrometry,
OEDP complexing with metals at leaching was determined. The reported study was
funded by State Assignment No. 10.7347.2017/8.9.
6 Recovery of Lead and Zinc from Fine Dust of the Copper Smelting. . . 49
References
7.1 Introduction
Lead, zinc, and arsenic are unavoidable impurities in copper ores. Due to the
concentrates quality deteriorating and being involved into secondary raw material
processing, these impurities are increasingly circulated in copper smelting produc-
tion and accumulated in the dusts.
According to Makhov et al. [1], most of the arsenic (up to 95–96%) goes to the
fumes and fine dusts during smelting of polymetallic raw materials to produce poor
mattes. Fine dusts of OJSC Sredneuralsk Copper Smelter (SUMZ) contain, %: Cu
10.68, Zn 10.67, Pb 14.31, Fe 15.81, S 8.09, and As 7.51. Return of these dusts to
copper pyrometallurgical processing will result in contamination of blister copper
with arsenic and lead [2].
It is reasonable to take out the fine dusts and similar negotiable intermediate
products from recycling and subject them to smelting reduction to produce
polymetallic matte.
Such a technology was implemented at PSTSM “Uralelectromed”
(Top-Neyvinsky plant) where dust reduction smelting was running in the shaft
furnace (SF) and ore-smelting furnace (OSF) [2]. Obtained mattes contained,%:
Cu 30–40, Zn 2–5, Pb 14–23, Fe 7–20, S 15, and As 4–5.
Return of these mattes to pyrometallurgical processing will again result in the
accumulation of arsenic and lead.
Therefore, obtained mattes must be exposed by separate technology.
This chapter presents the results for the selective extraction of copper with
oxidative sulfuric acid pressure leaching and pressure leaching with preliminary
treatment of mattes with copper sulfate followed by oxidative leaching.
7.2 Material
Tests were carried out in titanium autoclave having a volume of one dm3 [3] under
the following conditions: pulp density L: S ¼ 6: 1; oxygen pressure 0.4 MPa;
temperature 140–180 C; sulfuric acid concentration 10–130 g/dm3; values Qa/Qs,
where Qa is the actual amount of sulfate ions, considering quantitative oxidation of
sulfur of matte to SO42 and Qs is the stoichiometrically required amount of sulfate
ions to extract copper matte, between 0.82 and 3.17, and duration of leaching
30–240 min.
Experimental conditions and results are shown in Table 7.2.
Acid consumption significantly affects the results of leaching. The best results of
the copper extraction (94.26%) were achieved at 180 C and an initial acid concen-
tration of 64 g/dm3 (Qa/Qs ¼ 1.35). Extraction of copper from SF matte with a
higher iron content was up to 56% at the same value of Qa/Qs ¼ 1.17. Under these
conditions the sulfide sulfur matte went into solution up to 71%, and the extraction of
iron reached 30%.
The highest arsenic extraction for the leaching of mattes at 160 C was as follows:
• For SF matte reached 57% ([H2SO4]0 ¼ 132 g/dm3, Qa/Qs ¼ 3.2).
• For OSF matte reached 53.8% ([H2SO4]0 ¼ 55 g/dm3, Qa/Qs ¼ 1.3).
At the same time, copper extraction decreased to 18–36% and sulfur to 15–20%.
Iron passing into solution under these conditions greatly increased and reached
80–90%. It is possible that the exchange reaction of copper deposition makes
7 High-Pressure Leaching Arsenic Containing Polymetallic Copper Mattes 53
Table 7.2 Experimental conditions and results of oxidative sulfuric acid pressure leaching of
mattes
H2SO4
Materials Test g/dm3 Qa/Qs t, C Duration, min Copper extraction, %
OSF 1 10 0.82 160 90 35.14
2 37 1.1 160 120 83.19
3 55 1.26 160 120 18.13
4 85 1.55 160 120 16.87
5 62 1.33 180 120 43.50
6 43 1.14 180 120 78.47
7 64 1.35 180 240 94.26
SF 8 10 1.02 140 120 9.90
9 10 1.02 160 120 54.83
10 132 3.17 160 60 36.32
11 22 1.17 180 120 30.40
12 21 1.17 180 240 55.78
cake completely. Total removal of arsenic in two stages of leaching was 44%. Cake
contained arsenic at 4.9%; this fact makes it difficult to arrange further cake
processing at lead plants.
Thus, oxidative sulfuric acid pressure leaching of matte shows the following
results:
• Maximum extraction of arsenic during oxidative pressure leaching was 14.7%.
• Extraction of copper was 94.2%.
• Final solution contains up to 4 g/dm3 of Fe and up to 1 g/dm3 of arsenic.
• Copper extraction is decreased from 94.2 to 55.8% with increasing iron content in
the matte from 7 to 23%, respectively.
The cake contains up to 7% of arsenic after leaching, and it became more difficult to
process it by traditional methods [6]; therefore pressure leaching with preliminary
treatment of mattes with copper sulfate was studied.
Pressure leaching with preliminary treatment of mattes with copper sulfate was
based on the exchange reaction mechanism (CuSO4 + MeS ¼ CuS + MeSO4, where
Me is Fe, Zn, and Pb), at temperatures above 150 C [7].
The experiments were carried out under the following conditions: pulp density L:
S ¼ 6:1; temperature of 140–180 C, sulfuric acid concentration of 5–30 g/dm3,
copper concentration of 10–130 g/dm3 (molar ratio of Cu/(Zn + Fe + As +
Pb) ¼ 0.5–1.1), and leaching time of 60–120 min.
The aim of the experiment was to extract iron, zinc, and arsenic and for lead and
copper to remain in the cake. The obtained results are shown in Table 7.3.
Treatment of mattes with copper sulfate at 180 C allows one to extract arsenic
to a solution for matte OSF up to 89% for SF up to 93%. At the same time, zinc and
Table 7.4 Results of the oxidative pressure leaching experiments for the cakes after treatment of
mattes with copper sulfate (copper extraction from the initial content of copper in mattes)
Copper
Copper content [H2SO4]0, Molar ratio extraction,
Mattes Tests in cake, % L:S g/dm3 of H2SO4/Cu t, C %
OSF 1 49.50 24.69 21.33 0.69 180 87.09
SF 2 60.90 30.00 30.00 0.96 180 99.00
OSF 3 48.50 30.00 26.67 1.07 160 97.50
OSF 4 46.20 12.00 63.67 1.07 160 98.10
OSF 5 47.00 6.00 127.33 1.05 160 98.20
OSF 6 56.30 6.67 115.00 0.88 160 98.60
Treatment of mattes
with copper sulphate
Solution (As,Fe)
Cake (Cu, Pb)
Deposition of
As-Fe cake Oxidative pressure leaching
of cake
Solution (As,Fe)
Regeneration
Lead cake Solution (Cu) (excess of
sulphate ion),
acid
Electrowinning neutralization
Lead production with lime
Cathode copper Electrolytic
solution
extracted into solution to 50%. The results obtained differ from those obtained by
oxidative pressure leaching of SF matte without preliminary treatment with copper
sulfate, probably due to a deficiency of iron in the cake which forms a ferric arsenate
with arsenic.
Lead remained in the cake after leaching. The cake contains 48–55% of lead at
copper extraction of 98–99% and is suitable for lead plants. The concentration of
iron, zinc, and arsenic in the solution was less than 1 g/dm3, whereas copper
concentration (at L:S ¼ 6:1) amounted to 83–86 g/dm3.
The obtained results for two-stage pressure leaching including preliminary treat-
ment of mattes with copper sulfate followed by oxidative pressure leaching of cake
allow one to compose a tentative flowchart to process mattes, which is shown in
Fig. 7.1.
7 High-Pressure Leaching Arsenic Containing Polymetallic Copper Mattes 57
7.4 Conclusions
References
1. Makhov I. E., Mikhaylov S. V., & Shishkina L. D. et al. (1991). Povedenie myshyaka i surmy pri
pirometallurgicheskom proizvodstve medi (behavior of arsenic and antimony during the pyro-
metallurgical manufacturing of copper). Moscow: Central Scientific and Research Institute of
Economics and Information of Non-ferrous Metallurgy. Iss. 2. 56 p.
2. Skopov, G. V., & Matveev, A. V. (2011). Combined processing of polymetallic semifinished
products of metallurgical production. Metallurgist, 7–8, 596–600.
3. Neustroev, V. I., Karimov, K. A., & Naboychenko, S. S. (2013). Autoclave leaching of mattes
from smelting of intermediate metallurgical products. Tsvetnye Metally Nonferrous Metals, 8,
75–78.
4. Gomez, M. A., Becze, L., & Celikin, M. (2011). The effect of copper on the precipitation of
scorodite (FeAsO4*2H2O) under hydrothermal conditions: Evidence for a hydrated copper
containing ferric arsenate sulfate–short lived intermediate. Journal of Colloid and Interface
Science, 360, 508–518.
5. Monhemius, A. J., & Swash, P. M. (1999). Removal and stabilizing as from copper refining
circuits by hydrothermal processing. JOM, 51, 30–33.
6. Pozdnyakov, V. Y. A., & Chetvertkov, M. S. (1975). Povedenie mishyaka pri proizvodstve
cvetnih metallov. Tsvetnye Metally Nonferrous Metals, 11, 17–19.
7. Semenov, M. Y. U., Sirkis, A. L., & Khudyakov, I. F. (1984). Study of the hydrothermal reaction
of copper, nickel and iron sulfides with a copper sulfate solution. Tsvetnye Metally Nonferrous
Metals, 6, 15–18.
Chapter 8
Investigation of TSL Furnace
Hydrodynamics Using Cold Modeling
Method
8.1 Introduction
Cold modeling allows one to observe flows in media and determine the main
dynamic zones in a vessel avoiding problems related to high temperatures and
hard access to sealed internal space of a real metallurgical vessel. Our main goal
here was to observe the torch formation zone at the lance tip and measure various
parameters depending on blowing intensity and lance geometry.
A cold model was made using Plexiglas at a scale of 1:12 to real vessel. Compressed
air was used for blowing (up to 300 l/min). The lance was made using plastic tubes of
various diameters (most experiments were made with a 5 mm lance). The model was
mounted on a wooden stand to minimize vibrations. The stand was equipped with
LED lights to ensure caption quality.
The lance of a real furnace has an inner swirl to make possible tangential blowing
input, so the model got a steel swirl glued into the lance tube too.
Pressure at the lance tip was measured via manometer attached to the lance. Two
rotameters were used to measure the amount of air blown with four vents for blowing
intensity regulation.
Images were captured using two 100 fps cameras with a resolution of
1920 1080 mounted at a fixed distance from the model. Video capture was used
along with a photo series to ensure maximum quality of images received.
Figure 8.1 shows the scheme of cold model setup main elements.
The modeling liquid used was water and water-glycerol emulsion. The physical
properties of gas phase were not so important in the approach chosen according to
[4]. The main factors in terms of blowing were volume and velocity. Taking this into
consideration, the criteria of modeling liquid selection come to
ηl 4 ∙ g
M¼ ð8:1Þ
σ 3 ∙ ρl
where g is standard acceleration of free fall, m/s2; ηl is the dynamic viscosity, Pa•s; σ
is surface tension, N/m; and ρl is liquid density, kg/m3.
According to [5, 6], real slag has the following properties at 1300 C: ηl ¼ 0.15 Pa•s,
σ ¼ 0.4 N/m, and ρl ¼ 3700 kg/m3. Entering this data into Eq. (8.1), we obtain
M ¼ 2.1105 which corresponds to water-glycerol emulsion. Using water as the
modeling liquid corresponds to slag systems with lower viscosity and density.
The amount of air blown varied from 0.001 to 0.005 m3/s. For each value, four
test runs were made to ensure minimal error in results received. Some experiments
included usage of an organic indicator to determine the solid particles flow trajec-
tories and get a more consistent image of bath dynamics.
To make sure that optical distraction of media and/or vessel walls is taken into
account properly, a number of scale marks were placed on the lance tip and on the metal
plate that was submerged into liquid along with the lance in one run of each series.
8 Investigation of TSL Furnace Hydrodynamics Using Cold Modeling Method 61
Fig. 8.1 Scheme of modeling setup. 1 – Vessel, 2 – mounting points, 3 – movable plate, 4 – lance,
5 – rotary elements, 6 – liquid level in vessel, 7 – redactor, 8 – vents, 9 – manometer, 10 – safety
pressure vent, 11 – rotameter
At each blowing intensity and each geometrical parameters set (lance diameter,
lance submergence, number of lance tubes, swirl form), four series each consisting
of ten photos were taken.
Figure 8.2 shows an example of the image series taken during one of the test runs.
It should be mentioned that blowing intensity of 0.002–0.0025 m3/s is the closest
experimental condition to real operation modes of the TSL furnace in Karabash.
Blowing modes of higher intensity shall be considered as extreme and are not
suitable to a real process because of lining wear increase and lance lifetime decrease
in such modes.
Figure 8.3 shows the profile of the torch obtained at 0.002 m3/s blowing. It will
serve as initial data to verify results of CFD modeling which is still in an early stage
of development.
62 B. V. Kolmachikhin et al.
Fig. 8.2 Images taken during a test run at 0.002 m3/s blowing intensity
Each image was processed using photo editing software to increase quality and to
add the coordinate grid on it. Figure 8.4 shows an example of the image after this
stage’s processing.
The real size of each grid segment was calculated for each series individually
according to the mentioned physical marks on the lance and on the submerged steel
plate. To make possible quantity comparison between series, calculations of blowing
torch volume were simplified to those of a flattened cone.
Figure 8.5 shows the dependency of blowing torch volume on the amount of
blowing with 5 mm lance submergence and 5 mm lance diameter in single-phase
(water) media.
8 Investigation of TSL Furnace Hydrodynamics Using Cold Modeling Method 63
700
600
Torch volume, sm3
500
400
300
200
100
0
0.001 0.0015 0.002 0.0025 0.003
Blowing intensity, m3/s
A plateau in Fig. 8.5 ranging from 0.0022 to 0.0028 m3/s blowing intensity was
repeatedly appearing both in water and water-glycerol media tests. It shows that for a
further increase of torch penetration, a depth excess of air blowing is required.
Also we measured pressure at the lance tip under different conditions, but no
significant change was observed: pressure raised along with blowing intensity with
no difference of increase curve for submergence level or media density.
The Reynolds criterion was calculated for different blown air velocities using
Eq. (8.2):
υ∙D∙ρ
Re ¼ ð8:2Þ
η
where υ is blown air velocity, m/s; D is lance diameter, mm; ρ is air density, kg/m3;
and η is media dynamic viscosity, Pa•s.
The lance diameter was 0.005 m; media density was 1.84•105 Pa•s.
Calculation results are presented in Table 8.1.
64 B. V. Kolmachikhin et al.
The results obtained allowed us to compare data with real process conditions and
determine if chosen blowing intensities were correct. A comparison showed that
blowing intensity of 0.002–0.0025 m3/s corresponds to real operation modes.
8.4 Conclusion
The data obtained allowed us to determine the basic dynamic characteristics of TSL
furnace liquid bath: torch penetration depth and size increase with blowing intensity,
blowing conditions, and main flows in reaction zone. It is shown that to increase
torch penetration depth from real operation mode excessive blowing is required,
which is not an option for real furnace because of lining damage and lance lifetime
decrease.
Future work will include CFD modeling based on this data and further hydrody-
namic research including colored indicators used and detailed comparison of differ-
ent swirl modes to find optimal lance geometry parameters.
References
1. Matusewicz, R. W., & Lin S. L. (2010). Large scale copper smelting using Ausmelt TSL
technology at the Tongling Jinchang Smelter, Proc. “Copper 2010”: Vol. 3. Hamburg, June
6–10, 2010 (pp. 961–970).
2. Floyd, J. M. (2004). Converting an idea into a worldwide business commercializing smelting
technology. Metallurgical and Material Transactions B, 36(B), 557–575.
3. Floyd, J. M. (1992). The emerging role of New Bath Smelting, Mervyn Willis Symp. and
Smelting and Refining Course. Melbourne: University of Melbourne (pp. 13:1–13:51).
4. Markov, B. L. (1975). Methody produvki martenovskoy vanni (p. 279). Moscow: Metallurgia.
5. Rossinsky, E. E. (1974). Metallurgicheskie shlaki (p. 156). Moscow: Metallurgia.
6. Shmonin, U. B. (1981). Pyrometallurgicheskoe obednenie shlakov tsvetnoy metallurgii (p. 131).
Moscow: Metallurgia.
Chapter 9
Physical Simulation of Melt Lancing
in a TROF Converter
9.1 Introduction
The tilting rotating oxy-fuel (TROF) converter is a rotary bevel drum-type furnace.
Furnaces of this type are becoming more and more popular. These furnaces have a
number of advantages:
• They are suitable for a wide range of tasks.
• They may process both molten and solid materials.
• The furnaces are compact and do not require a lot of space.
• They are widely automated.
• The gas removal system prevents the release of gases in a working area.
The most important advantage of these furnaces is the ability of a smelting
chamber to rotate during smelting.
While recycling copper-containing materials, slags with high viscosity might be
obtained, which entails high copper losses with slag. Most of all, because of high
slag viscosity, they are very difficult to process, for example, in a refractory furnace.
The composition of copper-containing materials for recycling is unstable. Especially
the content of zinc and tin has the greatest impact on slag viscosity. In the work of
Naboychenko [1], a ratio of tin content in the slag to its viscosity is given. The TROF
converter is perfect for processing such slags and materials. The ability to mix the
produced slag allows one to receive waste slag with low copper containment. The
construction of the TROF converter allows the application of unsubmerged blowing.
In this case, the furnace might be used as a converter unit. For this purpose, the steel
lance embeds through converters nose.
Due to the unsubmerged blowing and the rotating smelting chamber, the material
flow trajectory is very difficult to predict. The better understanding of TROF
converter hydrodynamics is necessary for:
• More efficient use of blowing air
• Lining lifetime increase
• Achievement of best conditions for copper recovery from slag
Cold simulation has been carried out for a better understanding of the TROF
converter operation and to find possible ways for hydrodynamic improvement. For
this purpose, a model of the TROF converter smelting chamber was made of
plexiglass in scale 1:10 to a real unit. As with a real furnace, the model has the
ability to rotate about the longitudinal axis to simulate rotation of smelting chamber
and about the transverse axis to change the model angle and tape of the liquid.
Special frame and bearing allowed the observation of the movement of liquid in any
part of the model. The scheme of the experimental setup is shown in Fig. 9.1.
Fig. 9.1 The scheme of the experimental setup. 1 – The smelting chamber (the model’s body), 2 –
bearing support, 3 – guide rail for the lance, 4 – lance, 5 – driving actuator, 6 – level of liquid in the
smelting chamber; 7 – compressed air inlet and pressure regulator, 8 – valves, 9 – manometer, 10 –
safety valve
9 Physical Simulation of Melt Lancing in a TROF Converter 67
The model is equipped with a steel lance with an external diameter of 10 mm and
internal of 5 mm. In order to minimize the hydraulic resistance of the lance, its
channel was polished. Blowing was performed by compressed air at a temperature of
20 C. A special tank was used to maintain air temperature. The airflow rate was
measured with rate-of-flow meter PM-16 with a working range up to 16 m3/h.
According to airflow rate used in real furnace, it was calculated that the airflow
rate for modeling ought to be about 8.4 m3/h.
During modeling, the following parameters were varied:
• The airflow rate
• The model’s body angle
• Smelting chamber rotational speed
• The position of lance and its angle
In order to determine the speed and trajectory of liquid movement, a dispersed
organic indicator was used. The indicator consisted of a great number of particles
with different sizes (from 1 mm to 8 mm) and different flotation abilities. It allowed
the observation of the trajectory and speed of liquid flows in the bulk of
model’s bath.
In the capacity of liquid, water and mixes of water with glycerin and different oils
were used. The best results were achieved in experiments with pure water.
In order to achieve conditions of similarity, the Reynolds number was calculated.
The physical properties of the gas phase were not so important in the approach
chosen according to Yavoisky et al. [2]. The main factors are the blowing volume
and its velocity. Taking this into consideration, the criterion of modeling a liquid
selection come to
ηl 4 ∙ g
М¼ : ð9:1Þ
σ 3 ρl
Analysis of all obtained data allowed us to reveal the next picture of flows inside the
smelting chamber. They are presented in Fig. 9.3.
All flows arise under the impact of blowing torch, and their form, trajectory, and
speed depend much on the blowing airflow rate. It was revealed that the rotation of
the smelting chamber does not affect much the bath hydrodynamics if solid materials
are not used.
As can be seen from Fig. 9.3, when the airflow rate is 8.0 m3/h., four main flows
form. Flows numbered 1 and 2 spread closer to the liquid surface and move in the
direction of a spherical bottom. There they change direction and move backward
along the chamber’s body to the zone before the cavern.
The third flow unlike the first one develops below the liquid level. It moves along
the side of the chamber and rises in front of the cavern, but it does not crop out.
9 Physical Simulation of Melt Lancing in a TROF Converter 69
The flow numbered 4 is also initiated by blowing torch, but it moves down along
the spherical bottom and moves there along the lower element of a cylinder toward
the zone before the cavern. There it divides into two flows, one of them continues its
movement and imparts the material which is far away from the blowing torch, and
the other one moves up to the cavern. In a place where three flows divide into two,
the dead zone appears.
As already mentioned, the airflow rate greatly influences the flow’s trajectory. At
the flow rate of about 8.4 m3/h directions of same flows change; thus a third flow
joins with the second and forms one, which goes alongside elements of the cylinder.
According to the results of video filming, the average speed of liquid flows was 101
m/s. Taking into consideration the higher viscosity of real slags, it may match the
speed of real melt flow in the TROF converter by about 0.01–0.05 m/s.
One of the main parameters measured during experiments was the size of the
blowing torch (cavern); in order to determine it, the axial and cross airflow penetra-
tion depths were measured. Results are presented in Fig. 9.4.
By means of these measurements, the volume of displaced liquid was calculated.
The volumes of displaced liquid depending on airflow rate are provided in Fig. 9.5.
For different flow rate speeds, the values Re were estimated using Eq. (9.2).
Results are presented in Table 9.1.
Re ¼ υDρ=μ, ð9:2Þ
where υ is flow rate speed, m/c; D is nozzle diameter, m; ρ is air density, kg/m3; and
η is dynamic viscosity, Pa•c
The nozzle diameter was 0.005 m; the dynamic viscosity was 1.84105 Pa•c.
20
18
Penetration depths, d/L
16
14
12
10
8
6
axial
4
2 cross
0
50 70 90 110 130 150 170 190
Air flow rate, dm3/min
300
200
cm3 150
100
50
Volume of liquid
0
50 100 150 200
Air flow rate dm3 /min
Fig. 9.5 The relationship between the volume of displaced liquid and airflow rate
For every airflow rate, the kinetic energy Ek and potential energy Ep were calcu-
lated. According to the volume of displaced liquid, the energy required for its lift
(El) was estimated. Thereby only one percent of introduced energy is consumed for
lifting and splashing of melt. The results are presented in Table 9.2.
It means that practically all energy of blowing air passes to the liquid.
9.4 Conclusions
Cold simulation was carried out with an airflow speed from 52 to 230 m/s. That
matches the levels of kinetic energy from 1.8 to 223 W/s. Cold modeling permitted
the estimation of speed of liquid flows. It was about 101 m/s. Taking into consid-
eration the higher viscosity of real slags, it may match the speed of real melt in a
9 Physical Simulation of Melt Lancing in a TROF Converter 71
TROF converter by about 0.01–0.05 m/s, and locally the flow speed may exceed the
average flow speed by 5–10 times. The obtained results might be used to improve the
operation of the TROF converter unit.
References
10.1 Introduction
10.2 Experimental
The investigated raw material was an intractable division of the selective flotation
multicomponent sulfurous materials of “Maykain B” deposit, Kazakhstan. The
chemical composition of middling is presented in Table 10.1.
Figure 10.1 shows the phase composition of the investigated raw material
obtained via X-ray diffraction analysis.
The X-ray diffraction analysis showed that sphalerite, pyrite, chalcopyrite, and
galena are the most common minerals represented in the raw material.
Electron microscopic studies were performed to determine the fine structure of
the material, phase composition, chemical composition of the individual phases, and
their crystallographic orientation of raw materials using a transmission electron
microscope JEM 2100 prefix for microanalysis Oxford Inca.
Figures 10.2 and 10.3 show the results of the solid sulfide grains middlings
composition study.
Table 10.1 The chemical composition of raw material “Maykain B,” % mass
Cu Zn Fe S Pb Ca Na Mg As Al
6.37 9.21 33.57 46.3 1.18 1.72 1.66 0.24 0.21 0.1
Au, g/t Ag, g/t Sn Ni P Cd Mn Sb Si Ba
11.5 81.3 0.005 0.009 0.005 0.03 0.01 0.0063 <0.05 0.06
Sr Se Te Ti V Cr K Co Tl
0.013 0.005 <0.005 0.001 <0.001 0.006 0.09 0.016 <0.001
30000
Intensity, pulse/min
1
1
2
2
25000
1 1
20000 2 2
3
15000
2 2
10000 3 2
4 3
3 2 4 1 1
4
5000
0
20 30 40 50 60 70 80 90 100
2 teta, degree
Fig. 10.1 X-ray phase composition of the investigated raw material “Maykain B” : 1 – ZnS;
2 – FeS2; 3 – CuFeS2; and 4 – PbS
10 Thermodynamic Features Research of Polymetallic Sulfide Raw Material Leaching 75
Fig. 10.2 The results of the solid sulfide grains middlings composition study
Fig. 10.3 The results of the study of the solid sulfide grains middlings composition
Table 10.2 shows the results of the Gibbs energy change (ΔG, kJ) calculations for
reactions (10.1)–(10.11) for 373 K and 353 K temperatures. The temperature range
was determined by technological considerations.
Fig. 10.4 Existence diagrams of copper (a) and zinc (b) compounds
Fig. 10.5 Existence diagrams of iron compounds (a) and joint diagram of copper, zinc, and iron in
the nitric medium (b)
The diagram of Fig. 10.4b shows that in the cathodic potential range, formation of
gaseous NH3 and H2S is possible. Sphalerite begins to decompose with the forma-
tion of Zn2+ and S0 at E >0.3 and pH 1.5. With further increase in the acidity,
formation of sulfate ions is observed. This may be accompanied by release of NO
(g). At E >1.2 V, the formation of Zn(NO3)2 is occurred.
Figure 10.5a shows the existence diagrams of iron in nitric medium. Dissolution
of iron raw material components may lead to slower opening speed of minerals
containing copper and zinc. This is due to the fact that films (oxides of iron,
elemental sulfur, etc.) are formed on the surface of the mineral zinc and copper,
preventing oxidation of the middlings components.
At sufficiently low oxidation potential values (up to 0.45 V), in the whole range,
the formation of S0 and Fe2O3 occurs.
To convert iron in the trivalent form, a high potential value of E >1.1 V should
be used.
The joint diagram of copper, zinc, and iron existence during the nitric acid
oxidation of sulfides (Fig. 10.5b) allows prediction of the behavior of all the studied
complex polymetallic middlings main components and also allows one to choose
modes of the sulfide oxidation process with the aim to maximize the transfer of
components from raw materials in a convenient form for further processing.
Diagram analysis showed that the full transfer of sulfide copper-zinc components
of raw materials to the sulfate form required initial high oxidation potentials of
E >0.9 V. It is important that the copper and zinc are completely converted into
cationic form.
The interaction of metal sulfides with nitric acid leads to its degradation and the
formation of nitrous gases. The resulting nitrogen oxides are oxidized to higher
oxides in the absorption column to form a mixture of nitric and nitrous acids. The
resulting mixture was used in the next leaching stages [10].
10 Thermodynamic Features Research of Polymetallic Sulfide Raw Material Leaching 79
10.4 Conclusions
References
1. Kozyrev, V. (1991). Some trends in the development of raw material base of ferrous metallurgy
capitalist and developing countries. Tsvetnye Metally, 12, 16–19.
2. Bolatbayev, K. (2001). State, problems and reserves enrichment technology polymetallic raw
materials. Industry of Kazakhstan, 10, 91–93.
3. Bolatbayev, K. N. (2002). Complex use of mineral resources – State, reserves, priorities (p. 33).
Almaty: KazGos INTI.
4. Rogozhnikov, D., Karelov, S., Mamyachenkov, S., & Anisimova, O. (2013). Technology for
the hydrometallurgical processing of a complex multicomponent sulfide-based raw material.
Metallurgist, 57(3–4), 247–250.
5. Rogozhnikov, D., Mamyachenkov, S., Karelov, S., & Anisimova, O. (2013). Nitric acid
leaching of polymetallic middlings of concentration. Russian Journal of Non-Ferrous Metals,
54(6), 440–442.
6. Semun, N., Mamyachenkov, S., & Rogozhnikov, D. (2013). Combined processing of Erdenet
ore-dressing plant pyrite concentrates. Metallurgist, 57(1–2), 77–79.
7. Rogozhnikov, D., Mamyachenkov, S., & Anisimova, O. (2015). Kinetics of nitric acid leaching
of multicomponent sulfide middlings products. Russian Journal of Non-Ferrous Metals, 8,
26–29.
8. Bogacheva, L. M., & Ismatov, K. R. (1989). Hydrometallurgical processing of copper-
containing materials. Tashkent: Publications.
9. Sokolenko, L. M. (2009). Collection and recycling of nitrous gases. Cherkassy: Publications.
10. Locman A. A., Karavaev M. M., & Ivanov Y. A. (1999). A method of nitric acid producing, Pat.
RF № 2127224.
Chapter 11
Investigation of Lignosulfonate Adsorption
on Zinc Materials in Acidic Media
11.1 Introduction
11.2.1 Materials
Zinc concentrate was obtained from Chelyabinsk zinc plant and Uchalinsky mining
and processing complex. The concentrate composition is shown in Table 11.1.
Natural sphalerite was provided by Berezovsky deposit. The specific area was
determined by means of the Kozeny-Carman method. Specific areas of sphalerite
concentrate and natural sphalerite were 0.031 and 0.030 m2/g, respectively.
Studied sodium lignosulfonates (Solikamsky, Kamsky) were preliminary ana-
lyzed by means of FTIR (Fourier-transform infrared) spectroscopy (FTIR spectrom-
eter, Alpha, Bruker Optics, Germany).
All adsorption experiments were conducted with 1% solid suspension; solid and
liquid phases were separated by centrifugation at 3,000 RPM. The adsorption
experiments were performed with the following varying parameters: Ls concentra-
tion 25–300 mg/l, temperature 298–393 K, and sulfuric acid concentration 5–100 g/l.
The range of sulfuric acid concentration was chosen based on factual concentration at
pressure leaching of sphalerite concentrate. Initial and final concentrations were
determined by means of UV-Vis spectrophotometry (Specord 250, Analytik Jena).
FTIR spectroscopy analysis (Fig. 11.1) allows one to semiquantitatively appraise the
lignosulfonate composition. Broad absorption band at 3300 cm1 is typical for
2
3500 3000 2500 2000 1500 1000 500
Wavenumber, cm-1
84 E. Kolmachikhina and S. Naboychenko
presence of low molecular weight electrolyte. For example, in Ref. [7], CMC of
lignosulfonate was estimated equal to 0.3 g/l. During leaching, increase of surfactant
concentration above CMC is inefficient.
Adsorption isotherm models are often used for describing the adsorption mech-
anism. Both Langmuir (Eq. 11.2) and Freundlich (Eq. 11.3) isotherms were applied
to analyze the experimental data. Langmuir isotherm assumes that adsorption occurs
on homogeneous solid surface, whereas Freundlich isotherm is associated with
nonideal adsorption of heterogeneous surface by multilayer sorption [8]. The valid-
ity of adsorption isotherm models was estimated by regression coefficient, R2 (4).
kL ∙ Ce
qe ¼ q1 ð11:2Þ
1 þ kL Ce
qe ¼ K F ∙ C e 1=n ð11:3Þ
Pn 2
i¼1 qei q c
ei
R ¼ 1 Pn
2
2 ð11:4Þ
i¼1 qei q
where:
q1 – maximum adsorbed amount, mg/m2
kL – Langmuir constant, l/mg
KF – Fruendlich constant, mg/m2
n – dimensionless constant
qei , qc
ei – virtual and calculated adsorption capacity, mg/m
2
and
1 Xn
q ¼ q :
i¼1 ei
n
The values of regression coefficient indicate that the Langmuir model is well
fitted with adsorption isotherm of lignosulfonate on sphalerite concentrate
(Table 11.2). Kamsky lignosulfonate has the greatest adsorption capability. It con-
tains more –COOH, –OH, and –SO3H groups than Solikamsky Ls. As is known,
these groups are capable of hydrogen bonding, so they act as a bridge between the
solid and liquid phases [9]. Study of additional FTIR spectra of sphalerite concen-
trate after adsorption shows the presence of these groups. On the samples with
Kamsky Ls, intensive absorbance was found in characteristic area of –OH and –
COOH, along with weak absorbance for –SO3H.
Sulfuric acid addition (up to 20 g/l) results in noticeable decline of lignosulfonate
adsorption. Adsorption capability reduced by 58% and 34% for Solikamsky and
Kamsky lignosulfonates, respectively. Further growth of sulfuric acid concentration
(from 20 g/l to 100 g/l) increases Ls adsorption.
Temperature increase up to 393 K leads to adsorption isotherm change: the
plateau disappears; the isotherm is consistent with empirical equation of Freundlich
(Fig. 11.3).
86 E. Kolmachikhina and S. Naboychenko
120
100
Kamsky Experiment
Adsorption, mg/m2
80
60 Solikamsky Experiment
40 Kamsky Freundlich
20
Solikamsky Freundlich
0
0 100 200 300
Equillibrium Lignosulfonate Concentration, mg/l
Fig. 11.3 Isotherm of adsorption of lignosulfonate on zinc concentrate at 393 K, C H 2 SO4 ¼ 100 g/l
45
40
Adsorption, mg/m2 35
30
25 Solikamsky Experiment
20 Kamsky Experiment
15 Solikamsky Henry
10
Kamsky Henry
5
0
0 100 200 300
Equillibrium Lignosulfonate Concentration, mg/l
Fig. 11.4 Isotherms of lignosulfonate adsorption on sphalerite at 393 K, C H 2 SO4 ¼ 100 g/l
11.4 Conclusions
References
12.1 Introduction
wide range from 0.7% to 3% when the content of zinc is changed from 0.7% to 3.5%
or more.
Successful selective flotation of ores also depends on the correlation between the
activated and nonactivated types of sphalerite, on the content of impurities in martite,
wolframite (hübnerite) and cleiophane.
The presence of tennantite, secondary copper sulphides and oxidized minerals in
the ores, has a significant impact on the mineral separation.
Pyrite copper-zinc ores are most involved in the processing; they contain 20–30%
of the total copper reserves and production in Russia.
Ores are complex, since gold, silver and PGM and rare earth metals have
commercial value as well as the basic materials such as copper and zinc minerals.
The technology of ore processing in metallurgy provides simultaneous recovery
of basic noble and rare metals. Copper-zinc-pyrite-pyrrhotite gold ores, most wide-
spread in the Ural region, contain a wide range of ores and gangue. Producing high-
quality copper and zinc concentrates presents a challenge, as copper sulphides are
correlated with iron sulphides and with sphalerite and the high flotation response of
the minerals variety.
The loss of metals with pyrite concentrate and pyrite products reaches 20% Cu,
40% Zn, 90% Au and 50–70% Ag; the loss of zinc with copper concentrate is up to
30–40%.
The analysis of the causes for metal loss has shown that considerable losses occur
due to the thin complex mineral splice joints of ferrous metal sulphides with pyrite
and pyrrhotite (with the size of 100 + 10 microns or less), as well as due to the
substantial loose subtly dispersed grains ( 20 + 5 and less microns), which indicates
the imperfection in schemes and regimes of grinding and flotation.
The opening [6, 11] of the pyrite splice joints, splice joints of copper, zinc and
native gold, is not equal.
The problem of pyrite, other sulphides and their modifications and kinds of
separation is an essential factor and reserve in raising the selective flotation indica-
tors [8]. Super thin grinding pyrite ore of non-ferrous metals in Isomill and Vertimill
does not provide high results of selective flotation of copper and zinc sulphides due
to the formation of secondary sludge and a sharp rise in the flotation activity of pyrite
and sphalerite grains caused by the oxidation of sulphides and release of copper
cations.
Flotation activity of pyrite and other sulphide minerals is comparable while using
butyl xanthogenate as a collector; in relation to iron sulphides, xanthogenates with
highest oxidation state are not selective collectors. The flotation separation of
collective copper and zinc concentrates is a complex technological task. The usage
of hydrometallurgical processes may be an alternative solution, but despite being
highly efficient, these processes are much more expensive than the most expensive
enrichment processes and are not safe for the environment.
12 Complex Processing of Refractory Pyrite Copper, Copper-Zinc. . . 91
Various technological schemes [9, 10] of copper ore and copper-zinc ore processing,
technological schemes of direct selective flotation and variants of collectively
selective schemes have been developed and introduced for refractory pyrite-pyrrho-
tite-copper-zinc ores of Tarnersky, New Shemursky, Podolsky, Uzelginsky,
Safyanovskoye, etc. and other deposits of the Ural region.
Varieties of sphalerite, zinc magnitude module (the ratio of the original content of
zinc and original content of copper) and the presence of ferrous sphalerite require
special conditioning of rough zinc concentrate.
The success of sphalerite depression depends on the concentration of sulphide
ions in the desorption operations of the collector, in the selection of the collective
concentrate and in pH values. Best sphalerite depression is achieved when using
combinations of bisulphite, sulphide and sulphite of sodium thiosulphate in different
ratios often accompanied by zinc sulphate [6, 11, 12]. The practice of enrichment has
shown that naturally active and easily floating sphalerite with particle size from 0.2
to 0.005 mm including emulsive extrusions of copper minerals contains 3–5 times
more isomorphic copper and iron than nonactivated varieties. At the first stage of
selection of the collective copper-zinc concentrate in the “soft” depression mode, the
flotation of nonactivated sphalerite gets suppressed during the flotation of copper
minerals and flotation of the part of naturally activated sphalerite. After that the
process of sphalerite activation, the copper-zinc flotation out of the chamber product
of copper flotation with following cleaning of the rough zinc concentrate is
performed.
The second stage of “hard” depression mode selection is the copper-pyrite
flotation with pre-desorption of the collector and higher depressor consumption.
Optimum pH values during the selection in the copper flotation are being
regulated by the value of lime, zinc sulphate and sodium sulphide consumption
and comprise about 7.5–9.5.
The improvement of quality of the zinc concentrate obtained in a chamber
product can be attained under the condition of the zinc content in rougher concen-
trate being no less than 10–12% and pyrite content being no more than 25%.
The selective flotation of pyrite-pyrrhotite-copper-zinc ores of Tarnierskoe
deposit is problematic due to the high content of pyrrhotite (up to 70%) compared
to pyrite content. If the content of pyrrhotite equals 20% or less, the simple
technological method of aeration in lime medium is performed before the
collective-selective flotation with following flotation of the butyl xanthate. This
allows obtaining satisfactory flotation performance. The aeration mode and weak
(compared to pyrite and pyrrhotite and inactive sphalerite) sulphhydryl collectors of
М-TF type along with butyl xanthate allow improving the selectivity of flotation.
At the beginning of the ore flotation, the copper “head” liberates. With the return
of cleaning tailings in the ore flotation, the sphalerite is flotated out of the chamber
product. This reduces the circulation of the pyrite-pyrrhotite intermediate products
circulation significantly. An important problem is the presence of tennantite in some
92 V. Bocharov et al.
ores. The floatability of tennantite in lime medium is lower than floatability of other
copper sulphides, which reduces the overall copper abstraction. The high content of
tennantite helps to concentrate an arsenic up to 1% in copper concentrate, which
complicates the technology of its pyrometallurgical processing; it requires the
disposal of arsenic from waste gases. A significant part of tennantite can be isolated
in a separated copper-arsenic concentrate by picking up selective collectors, depres-
sors and regulators [13].
Technological schemes with fractional flotation release of monometallic concen-
trates in different cycles of grinding and flotation in accordance with the technolog-
ical properties of minerals are designed and implemented to improve the abstraction
of metals (Fig. 12.1).
A method for complex ore processing has been developed to produce rich
flotation concentrates and poor polymineral products for hydrometallurgy
(Fig. 12.2).
Chlorine processes, intensive tank-leaching or bio-leaching processes can be
applied to poor products. As for rich products, autoclave methods and acid leaching
with preliminary firing are preferable.
Gold ores in the Ural region are mainly associated with pyrite; the content of the
free gold is 20% or less.
Gold abstraction into copper concentrate increases along with the output of the
concentrate and abstraction of sulphur (pyrite), which leads to a decrease in copper
content of the concentrate to 15%. Therefore, improving the quality of the copper
concentrate is accompanied by the loss of gold from pyrite-bearing tailings. Addi-
tional gold abstraction can be achieved by interstadial gravity separation by means of
centrifugal concentrators and concentration tables or combined gravity-flotation-
chemical and metallurgical technologies on gold pyrite tailings.
Sulphide polymetallic lead-zinc and copper-lead-zinc ores are also considered
refractory and have distinctive technological features because of complex mineral,
chemical and phase compositions and the proximity of technological properties of
sulphide minerals, easily sludging galena. For ingrained pyritic and oxidized sul-
phide ores and roughing concentrates, combined techniques are proposed, with the
release at the stage of processing rich selective concentrates and intermediate
products followed by hydrometallurgical processing schemes, as well as for
copper-zinc ores [17, 18].
Stepnoe polymetallic ore and other deposits have complicated textural and
structural characteristics: from crystalline up to corrosion and skeletal. It has been
proposed to use the flotation scheme with allocation of the copper-lead at the
beginning, and with allocation of the zinc and pyrite concentrates afterwards. The
resulting concentrates contain copper in copper 25.0% with abstraction 72.0%; lead
in lead 46.0% with abstraction lead 81%; and zinc in zinc 52.8% with abstraction
73.0%.
12 Complex Processing of Refractory Pyrite Copper, Copper-Zinc. . . 93
Ore
1 Grinding,
classification
Flotation
Tailings 1 Cu
concentrate
2 Classification
2 Cu
2 Grinding flotation
Tailings
Cu
3 Classification concentrate
3 Grinding Conditioning
Cu-Zn flotation
Tailings
Cu-Zn Tailings 2 Cu
concentrate concentrate
Conditioning
1 Zn flotation
2 FeS Tailings
concentrate Cn-Zn Conditioning 1 FeS2
concentrate concentrate
Tailings
1 Zn
concentrate
3 Cu flotation
Cu 2 Zn
concentrate flotation
Conditioning Zn 3 FeS2
concentrate concentrate
Conditioning
3 Cu
concentrate 2 Zn
concentrate
Ore
Grinding, classification
Fractional flotation
Selection Conditioning
2 Cu 2 Zn Cu-Zn FeS2
concentrate concentrate product concentrate
Conditioning Conditioning
Cu Cu-zn-FeS2 Zn-Cu-FeS2 Zn
Concentrate product product Concentrate
Fig. 12.2 Schematic diagram of selective flotation of pyrite copper-zinc ore with fractional release
concentrates
During the processing of complex ores of non-ferrous metals in the tailings concen-
trating factories, hundreds of millions tons of pyrite tailings have accumulated. In
these tailings, thousands of tons of non-ferrous and precious metals are contained.
The necessity to use technogenic pyrite-bearing tailings and other products allocated
during the processing of complex copper-zinc ores has arisen due to the reduction in
production volumes and ease of processing rich ores and also in relation with
reduction of the production of marketable concentrates for the metallurgical produc-
tion of copper, zinc, gold and silver. Cost-effective development of technogenic
deposits and their complex enrichment is possible if a number of technological and
organizational issues are solved. Those tasks are related with the inventory of
technogenic tails, geological mapping technology between rich and poor areas and
with the usage of rational processing technologies along with modern ways of
abstraction of the main components into commercial products of high quality.
Pyrite technogenic tailings are more complex and refractory compared to other
technogenic mineral objects. Pyrite technogenic products are highly subjected to
technological properties changing than the raw ore materials in warehouses or raw
12 Complex Processing of Refractory Pyrite Copper, Copper-Zinc. . . 95
ore during preparation and flotation. The oxidation of pyrite and other sulphides is
intensified due to the tight interaction between the micro-galvanic pairs, mostly
made of pyrite, and other components of the solid phase (copper sulphide, gold
particles form iron compounds and others). Compared to ore material, the portion of
thin-sized fractions exceeds 20%. The concentration of thiosulphate, sulphite, sul-
phate and polythionate ions is much higher than the concentration of the ionic
composition of the ore material due to the more active oxidation of sulphides.
Ion-molecular composition of the liquid phase influences the formation of com-
pounds on the separated minerals and on the electrochemical condition of the pyrite
pulp. Optimum composition of the pulp stimulates the rise of the contrast properties
of separatable minerals and differences in the oxidation of sulphide minerals in the
changing pH, temperature and oxygen concentration; other oxidants allow us to
determine the effective regimes of mineral separation.
A number of studies [1, 6, 7] show the basic technological possibility of fractional
allocation of minerals of non-ferrous metals and gold into concentrates and products
in accordance with the technological properties. The research on material composi-
tion of technogenic materials makes it possible to determine the optimal conditions
for gravitational separation of native free gold particles and its gold open splices into
the gold-containing product; also choose the regime of fractionation and concentra-
tion of copper, zinc and pyrite minerals in individual products and their subsequent
conditioning using individual technological schemes based on mineral processing
and hydrometallurgy.
The most used method for better decomposition of pyrite with the release of gold
and non-ferrous metal sulphide scanned in it is a process of chlorine metallurgy,
mastered in the factories of Germany, Spain and Japan; the process was created by
the Gintsvetmet Institute and was applied at the Ryazan pilot plant. Acid leaching
and bio-oxidation are among the most effective types of impact [14, 15, 16].
12.4 Conclusions
1. Distinctive features of ores: complex and variable chemical and mineral compo-
sition, complex and variable phase constitution, different structures, size, differ-
ent oxidations and slime formation tendencies.
2. Combined processing with separation of the mineral varieties into different
flowsheet cycles has been developed.
3. The combined technologies providing the acquisition of rich concentrates of
non-ferrous metals and pyrite products for the hydrometallurgical processing
are described.
96 V. Bocharov et al.
References
13.1 Introduction
In the market of building materials to increase durability and improve the water
resistance of concrete and reinforced concrete structures, “Vandex”, “Xypex” and
“Penetron” waterproofing penetrative capillary mixture (WPCM) brands are
presented. They usually consist of рortland cement, silica sand and a certain disper-
sion of the complex chemical active substances that dissolve in water and migrate
through the pores and capillaries in the concrete body, interacting with the products
of hydration of portland cement to form sparingly soluble compounds, filling the
pore space and sealing cement stone (CS).
The permeability reduction of the concrete treated takes place at the expense of
the chemical interaction of soluble mineral salts contained in waterproof mixture
with the hydration products of рortland cement to form heavily soluble compounds
which are produced in the pores and capillaries of concrete thus decreasing their
radius [1, 2]. The permeability increase of the cement stone is also achieved due to
the interaction character change of the porous mortar with the pores surface,
i.e. transformation of the capillary surface in such a way as to increase the end
angle value of their wetting.
However, X-ray diffraction (XRD) and differential thermal analyses (DTA)
methods fail to give a full picture of the processes taking place in the cement stone
under the influence of WPCM. At the same time, the insignificant phase constitution
of cement stone structure which is practically inconspicuous for the researchers can
considerably influence the concrete structure permeability. It is our opinion that
electron microscopy with X-ray spectrum microanalyser of the elements, chemical
composition of the hydration and hardening products is one of the most effective
modern complex methods to study the cement stone structure features including the
changes occurring in it on a capillary level by the action of WPCM.
To study the WPCM action on the cement stone microstructure, samples of рortland
cement (PC) СEM I 42.5 EN 197–1 of ZAO Nevyansky Tsementnik production
have been made, and as WPCM – type the mixture Penetron was used. The cube
samples of cement stone with the rib size of 40 mm were prepared from cement paste
with water-cement ratio 0.4.
In 28 days of air-moist hardening, the cement samples in question were sub-
merged in water for 2 days, and then on one of the edges of the sample cubes, half of
them have been treated with WPCM, the layer thickness being 1.2–1.5 mm (com-
position of cement stone No. 2). The second half of the samples under consideration
which had not been treated with WPCM was the control one (composition of cement
stone No. 1). All the samples in use were periodically moistened (twice a day) within
the first 3 days to allow more effective permeability of the WPCM soluble ion
penetration into cement stone. All the samples under discussion were stored in an
exicator before testing. Then the WPCM mortar was removed mechanically from the
surface of some samples, and all the samples removed were fractured. The cement
stone probe was taken from the central part of the samples to investigate their
structure using electron microscope.
The microstructure of the cement stone was evaluated with the help of a Tescan
MIRAS LMU electron microscope (magnification up to 100,000 times) and JSM–
5900 series microscope with energy-dispersive spectrometer with which not only the
thin structure images can be produced, but element analyses of the samples can be
carried out, too.
To determine the active part of the composition, WPCM was mixed with distilled
water, blended for 15 min and filtrated, and the content of different ions in the mortar
investigated was determined. It was established that carbonates, hydrocarbonate
sodium ions in general and also sulphates, chlorides and calcium ions in less
concentration were available in the solution investigated. It was also established
from the cement stone surface square treated with WPCM that the concentration of
carbon ions was 50 g/m2 and hydrocarbon ions 25 g/m2. It can be assumed that the
present ions have been formed on dissolving the complex ion addition present in
WPCM. Comparatively small concentrations of Ca2+ and SO42 could be formed as
13 Impact of Waterproofing Permeable Capillary Mixtures on Cement Stone 101
Fig. 13.1 Cement stone structure (600 times magnification): (а) control composition (composition
of cement stone No. 1); (b) covered with WPCM (composition of cement stone No. 2)
13 Impact of Waterproofing Permeable Capillary Mixtures on Cement Stone 103
Fig. 13.2 Cement stone of the composition of cement stone No. 2 (Region A, Fig. 13.1b, 2,000
times magnification)
Fig. 13.3 Cement stone of composition of cement stone No. 1 (1,500 times magnification)
content than the main mass of calcium hydrosilicates removed from the crack edges
and cavities.
Also the composition of cement stone No. 2 under study consists of relatively
large needle-like crystals of ettringite, the length of which is approximately 3.5 μm
and the diameter 0.3 μm (Fig. 13.4, position 4). The quantity of the ettringite crystals
is comparatively small, and, in general, they are located in big cavities, which are
practically fully filled with needle-like calcium hydrosilicate and relatively large
ettringite. The distance between crystals is from 0.1 to 0.5 μm (Fig. 13.4a). It should
104 F. Kapustin et al.
Fig. 13.4 Cement stone of the composition of cement stone No. 2 (4,000 times magnification)
be pointed out that aggregates of cube-like crystals of calcite type are available
which will also decrease cement stone permeability (Fig. 13.4, positions 5, 6).
Thus, when mixing WPCM with water and putting it on the prepared concrete
surface, the dissolution of the chemically active components, which are the part of
ionic additions, and рortland cement components takes place. In this case, on the one
hand, the high ion concentration of Ca32, HCO3 and Na+ and, on the other hand,
comparatively small Ca2+, SO42 and Cl are formed on the concrete surface.
Because ion concentration of the concrete surface is considerably higher than that
of the porous mortar, these ions will move into the depth of the concrete structure
along the developed pore net and capillaries towards the concentration decrease or
reduction on account of their high wetness.
The original ion concentration on the concrete surface will shrink due to the ion
migration into the protective composition hydration on the other one. In the course of
time, the soluble ions will bind in hardly soluble compounds such as calcium
hydrosulphoaluminate and hydrocarbon aluminate of calcium and also calcite.
Therefore, in the initial period after WPCM hardening on the concrete surface, it is
necessary to provide the deep moistening of its surface to give the possibility to
many ions to be able to migrate from the concrete surface into its depth.
The ions penetrating into the concrete will react chemically with the products of
PC hydration and hardening. The most soluble of them is portlandite, and therefore,
the probable reaction with its participation will be
13 Impact of Waterproofing Permeable Capillary Mixtures on Cement Stone 105
C3 AH 6 þ 2CaðOH Þ2 þ 2СО3 2 þ 6Н 2 О
! C 3 A СaСO3 12H 2 O þ СаCO3 þ OH ð13:2Þ
To decrease the alkali ion concentration in the concrete pores after the WPCM
action is over (according to the point of view of the mixture’s producers, this period
lasts 28 days), one must wash out the treated surface with water which results in the
probability of occureence of alkali corrosion process development on the concrete
construction surface.
It is our opinion that the WPCM action will be most effective for concretes
consisting of sufficient quantity of Ca(OH)2 and having a definite porous structure in
which the predominant capillary pores are 1–2 μm in size. For concretes with coarser
pores, the effect from needle-like compound formation on the pore walls having a
length of 0.5–0.1 μm can be less significant.
13.5 Conclusions
The alteration of the cement stone structure after its surface treatment with WPCM
takes place at the expense of CO32, HCO3, Ca2+, SO42 and Cl ions to form the
proper AFt and AFm phases and calcite, too. The particular role of forming a
waterproof concrete structure belongs to the calcium needle-like hydrosilicates of
low basicity of (1.03–1.11)CaO∙SiO2∙(3.9–4.3)H2O content. Needle-like
hydrosilicates are formed on the pores, cracks and capillary walls of the cement
stone in question in considerable quantity, narrowing them and increasing the angle
edge wetting magnitude. But the effectiveness of the waterproof action of WPCM is
determined primarily by the peculiarities of both the pore structure and the mineral
composition of the cement stone in the concrete under review which depends on
concrete and the type of the cement used.
References
1. Brovkina, V. G., Ovcharenko, Y. I., Bykov, V. G., & Izosimov, M. P. (2010). Influence of
permeable hydro insulation on the phase constitution of the porous cement. Vestnik UrGU:
Series: Building and architecture, 10, 19–21.
2. Kapustin, F. L., Spiridonova, A. M., & Pomazkin, F. P. (2013). Changing of the cement stone
structure on the treatment with hydro insulating permeable capillary mixture. Cement and its
application, 6, 52–56.
3. Taylor, C. (1996). Chemistry of cement. Moscow: Mir.
4. Shark, I., & Bernd, V. (2004). Durability of concrete. Oranta: Kiev.
5. Kubal, M. T. (1993). Waterproofing the building envelope. New York: McCraw. Inc.
6. Kapustin, F. L., & Semerikov, I. S. (2013). Chemistry of mineral binders. UrFU: Ekaterinburg.
7. Royar, Y. S., & Yranoxsraya, I. Y. (1986). Prevention of alkali concrete corrosion with active
mineral additions. Concrete and reinforced concrete, 7, 16–17.
8. Salnikov, N. S., & Ivanov, F. M. (1971). Corrosion destruction of concrete, containing large
additions of potash. Concrete and reinforced concrete, 10, 20–21.
9. Vicktorov, A. M. (1986). Prevention of alkali corrosion of moistened concrete. Concrete and
reinforced concrete, 8, 38–39.
Chapter 14
Composite Cement with a Polymer Addition
E. Gerasimova
14.1 Introduction
The complex use of raw materials and industrial waste is of great practical impor-
tance nowadays [1]. It will contribute substantially to the development of ecologi-
cally pure, resource-saving and nonwaste commercial technologies, whereas
utilization of industrial waste provides the solution to the environmental issues
which are aggravating both in Russia and elsewhere annually.
Various kinds of industrial waste can serve as valuable raw materials due to their
composition and properties. They can be used most effectively and extensively in the
production of building materials, specifically, to obtain composite binding
substances.
Composite binding materials are widely used in modern construction for inside
and outside finishing. Up to three different types of mineral additions of both natural
and technogenic origin can be introduced into their composition, e.g. fly ash, blast
furnace slag, pozzolanas, limestone ground, marble powder, etc. [2, 3].
Besides, the production of high-technology finishing materials requires polymer
modifier application, which allows improving and changing the final properties of
the product in terms of its practical usage [4].
Another objective regarding practical implementation is to investigate the
strength characteristics of composite cements consisting of a certain type of polymer
modifier as well as the quantity of mineral fillers.
E. Gerasimova (*)
Institute of Materials Science and Metallurgy, Ural Federal University, Yekaterinburg, Russia
In the course of the experiments, the samples from the mixtures were prepared in
accordance with a specific formulation after storage under the given conditions and
time. Then the necessary quantity of the samples in question was tested.
Two types of polymer modifiers of non-Russian production have been used in the
investigation proper, namely, vinyl-acetic ester redispersible polymer powder
(PAV-22, France) and styrene-acrylate liquid dispersion (Acronal 290D, Germany).
The polymer characteristics are given in Table 14.1.
Fly ash from Berezovskaya district power station (FY), limestone meal (LM) and
marble powder (MP) have been used as mineral additions to ordinary Portland
cement CEM I 42.5 N (PC). The main properties and chemical composition of
mineral components are presented in Tables 14.2 and 14.3.
A small-scale method has been used for this investigation. A standard mixing
procedure was used.
The polymer additions were combined with water and then added to the cement
while mixing. The time after the addition of polymers was kept constant throughout
the study. The water present in the polymer dispersion was included in the total water
content of a mix. From each mix small prisms were cast in steel moulds. The
specimens were demoulded after 24 h and cured under combined conditions.
The tests for the compressive and flexural strengths were carried out after 3, 7,
14, 21 and 28 days’ time. For flexural strength determination, a Michaelis device for
small samples was used.
All the results presented in the paper are the averages over a maximum of six
specimens and a minimum of three.
FA = 27.3929-0.0098*x-0.0048*x^2
LM = 49.208-1.2061*x+0.0104*x^2
MP = 34.4642-0.1967*x-0.0031*x^2
Rcom,
MPa 40
35
30
25
20
15
10
0
10 15 20 25 30 35 40 45 50
Quantity of addition, %
FA; LM; MP
Fig. 14.1 Compressive strength of Portland cement stone with mineral additions (28 days)
FA = 1.908-0.0116*x-6.4286E-5*x^2
LM = 2.044-0.0086*x-8.7921E-5*x^2
MP = 1.9584-0.0109*x-1.1554E-5*x^2
Rfl, 2.0
MPa
1.8
1.6
1.4
1.2
1.0
0.8
0.6
0.4
0.2
0.0
10 15 20 25 30 35 40 45 50
Quantity of addition, %
FA; LM; МP
Fig. 14.2 Flexural strength of Portland cement stone with mineral additions (28 days)
decrease. It can be concluded that it is possible to introduce the chosen fillers in the
quantity of up to 20% of the cement mass.
The last stage of the present work focused on the properties of the complex three-
component system containing mineral additives and modified polymers. This stage
of our work was carried out by using the simplex lattice experiment planning
method. The composition of three systems “cement-fly ash-marble powder”,
“cement-fly ash-limestone meal” and “cement-marble powder-limestone meal” has
been formed to fit the planning matrix (Table 14.5).
The test results concerning the strength of the hardened binders obtained on the
28th day were processed with the help of “Statistica 10.0” software package
assuming that the total amount of materials was taken as 100% (Figs. 14.3 and 14.4).
112 E. Gerasimova
> 38
> 2.5 < 38
< 2.5 < 34
< 2.2 < 30
0.75 0.25 < 1.9 0.75 0.25 < 26
< 1.6 < 22
< 1.3 < 18
<1 < 14
< 0.7 < 10
< 0.4 <6
1.00 0.00 < 0.1 1.00 0.00 <2
0.00 0.25 0.50 0.75 1.00 0.00 0.25 0.50 0.75 1.00
CEMENT (%) FLY ASH (%) CEMENT (%) FLY ASH (%)
> 38
> 2.6 < 38
< 2.6 < 34
< 2.3 < 30
0.75 0.25 <2 0.75 0.25 < 26
< 1.7 < 22
< 1.4 < 18
< 1.1 < 14
< 0.8 < 10
< 0.5 <6
1.00 0.00 < 0.2 1.00 0.00 <2
0.00 0.25 0.50 0.75 1.00 0.00 0.25 0.50 0.75 1.00
CEMENT (%) FLY ASH (%) CEMENT (%) FLY ASH (%)
Fig. 14.3 The projection of curvilinear strength dependence of three-component systems with
PAV-22 polymer on the 28th day. (On the left side, the flexural; on the right side, the compressive.)
14 Composite Cement with a Polymer Addition 113
> 31
>2 < 31
0.75 0.25 <2 0.75 0.25 < 27
< 1.7 < 23
< 1.4 < 19
< 1.1 < 15
< 0.8 < 11
< 0.5 <7
1.00 0.00 < 0.2 1.00 0.00 <3
0.00 0.25 0.50 0.75 1.00 0.00 0.25 0.50 0.75 1.00
FLY ASH (%) CEMENT (%) FLY ASH (%)
CEMENT (%)
> 32
>2 < 32
0.75 0.25 <2 0.75 0.25 < 28
< 1.7 < 24
< 1.4 < 20
< 1.1 < 16
< 0.8 < 12
< 0.5 <8
1.00 0.00 < 0.2 1.00 0.00 <4
0.00 0.25 0.50 0.75 1.00 0.00 0.25 0.50 0.75 1.00
CEMENT (%) FLY ASH (%) CEMENT (%) FLY ASH (%)
> 29
< 29
> 1.9 < 26
0.75 0.25 < 1.9 0.75 0.25 < 23
< 1.6 < 20
< 1.3 < 17
<1 < 14
< 0.7 < 11
< 0.4 <8
1.00 0.00 < 0.1 1.00 0.00 <5
0.00 0.25 0.50 0.75 1.00 0.00 0.25 0.50 0.75 1.00
CEMENT (%) MARBLE CEMENT (%) MARBLE
POWDER (%) POWDER (%)
Fig. 14.4 The projection of curvilinear strength dependence of three-component systems with
Acronal290D polymer on the 28th day. (On the left, flexural; on the right, compressive)
114 E. Gerasimova
14.4 Conclusions
In the course of the work, the potentialities of using mineral fillers of various origins
to produce materials based on Portland cement CEM I modified with polymer
addition have been investigated.
The experiments on the impact of mineral fillers upon the strength of water
cement stone have shown that it is preferable to introduce carbonate fillers in the
amount of up to 20%. When replacing cement by using only one type of fillers, it is
better to use limestone meal to make the polymer-modified cement binder.
The three-component polymer-modified systems have been studied with the help
of the simplex lattice method of mathematical planning experiment. The study has
revealed that in order to produce composite materials based on the polymer cement
binders selected, it is recommended to use both fly ash and marble powder with
PAV-22 or Acronal290D with regard to the strength loss. The combined additive can
be introduced in the amount of 25% of cement mass. Under these conditions the ratio
of fly ash to marble powder is to be 1:2.
References
1. Bozhenov, P. I. (1986). Complex use of mineral raw materials for production of building
materials (pp. 10–11). Moscow: Stroyizdat.
2. Kopanitsa, N. O., & Anikavova, L. A. (2002). Fine dispersed additives for filled binding
materials on the basis of cement. Constuction Materials, 9, 2–3.
3. Entin, Z. B. (1993). Chemistry and technology of finely ground multi-component cements, Thesis
for DPh: Moscow.
4. Frederik, A., & Nikolya, R. (2000). Using redispersible powders “Rhoximat” in dry mixes
production. Construction Materials, 5, 8–9.
5. Domanskaya, I. K., Kapustin, F. L., Mokhort, E. S., & Oleynik, V. N. (2003). The main
approaches of component selection for fine polymercement products and coatings. In F. A.
Finger, J. Stark, & H.-B. Fischer (Eds.), Proc. of the 15th Int. Conf. IBAUSIL (pp. 523–530).
Weimar: Bauhaus University.
14 Composite Cement with a Polymer Addition 115
6. Rogante, M., Domanskaya, I. K., Gerasimova, E. S., & Vladimirova, E. V. (2013). Feasibility
study for a neutron-based investigation of polymer cement concretes. Universal Journal of
Applied Science, 1(1), 11–17.
7. Bhikshma, V., Rao, K., & Balaji, B. (2011). An experimental study on behavior of polymer
cement concrete. Asian Journal of Civil Engineering (Building and Housing), 5, 563–573.
Chapter 15
The Manufacture of Concretes Based
on Unburnt Ash Gravel
15.1 Introduction
15.2 Experiments
The present paper aims to investigate the composition and structure of unburnt ash
gravel as well as the possibilities of using it for the production of high-strength light
concrete.
In the study, the following materials have been used: PC of two types Cement I
42.5 H (Cement I), Cement II/SA 32.5 (Cement II), fly ash from bituminous coal
burning (FA), screening of hornblende and a solidification accelerator, namely
sodium sulphate (Na2SO4).
Fly ash (FA) includes up to 93% of SiO2 + Al2O3 + Fe2O3. It is also acidic and
fine. In terms of its principal properties, it meets the GOST 25818 requirements and
belongs to type 3, i.e. it can be used for the production of both cellular and ordinary
concrete.
There has been developed a new composition and method for the production of
unburnt ash gravel of high strength, consisting of fly ash obtained from heat power
industry (65%), Portland cement (20%), screening from rock crushing (15%) and the
accelerator of hardening. The method for the production of unburnt ash gravel
involves proportioning and mixing the components, moistening the mixture,
forming granules of the required coarseness, steaming and fractioning.
The granules are subjected to steam treatment in a steam chamber for 6 h at 85 C
and are then tested according to the standard (GOST 9757-90 Gravel, artificial
porous crushed stone and sand). During the selection of components for burnt ash
gravel mixture, the influence of the cement type on its properties was investigated.
Table 15.1 presents the compositions of burnt ash gravel and the properties of
unburnt granules before steam treatment. The compositions of granules after steam
treatment are shown in Table 15.2.
It was established that the grain composition of all types of burnt ash granules
meets all the GOST 9757-90 requirements.
The aggregate strength in Cement I is 1.5 times higher than that of burnt ash
gravel in Cement II. As far as physical-mechanical properties are concerned, the
unburnt ash gravel obtained meets the GOST 9757-90 requirements and has both the
strength of P350 and the bulk density of 900.
Fig. 15.1 Macrostructure of unburnt ash gravel: (a) in Cement I; (b) in Cement II
In order to justify the differing density of unburnt ash gravel, its microstructure
has been investigated (Fig. 15.1).
The study has revealed that burnt ash gravel in Cement I has small pores, their
size being 0.17–0.32 mm, whereas in Cement II, pores are coarser, and their size
ranges from 0.32 to 0.62 mm. It should be noted that there are fewer open pores in
burnt ash gravel in Cement I. The strength of granules depends on the degree of
porosity to a large extent. It is higher in the case of closed porosity as compared with
open porosity. Open porosity of burnt ash gravel in Cement II is 1.62 times higher
than that in Cement I. Open porosity is caused by excessive water during the
granulation. The amount of water used to obtain burnt ash gravel based on Cement
II is 1.17 times greater than that of based on Cement I, which suggests that the
porosity of the former is higher and the strength is lower.
Burnt ash gravel has a combined structure, i.e. it includes both loose-grained and
conglomerate structure.
Differential thermal analysis of unburnt gravel (Fig. 15.2) confirmed that steam
treatment of ash cement stone with hornblendite addition accelerates cement hydra-
tion, which is supported by large mass loss (19.55%). In the temperature range of
120 F. Kapustin et al.
Fig. 15.2 Derivate diagram of burnt ash gravel containing 20% PC + 15% hornblendite+65% fly
ash + 2% of Na2SO4
130–150 C, the crystallization of tobermorite gel is observed. In this case, the mass
loss increases up to 14.40%. In the temperature range of 301–380 C, the dehydra-
tion of calcium hydrosulphoaluminate occurs. It should be added that during deep
endothermic effect in the temperature range of 460–475 C, Na2SO4 dissolves Ca
(OH)2 and thereby decreases its contents. Moreover, the endothermic peak in the
temperature range of 640–669 C is observed on the derivative diagram. The peak is
caused by the burnout of semi-coke and coke residues in ash. In the temperature
range of 950–980 C, the crystallization of the glassy phase of ash occurs [3].
Mix design of the construction blocks of V25 and V40 strength class was carried out
using burnt ash gravel produced from Cement I (Tables 15.3 and 15.4). Since bulk
density and hollowness of crushed granite stone and burnt ash gravel differ, during
the concrete proportioning, it was taken into account that the lighter burnt ash gravel
demands a greater amount of crushed stone, the volume being equal. It was
established that the strength characteristics of the optimum concrete contents made
15 The Manufacture of Concretes Based on Unburnt Ash Gravel 121
Table 15.3 The composition and the properties of concrete mixtures and concrete
Materials consumption, kg/m3 Concrete mix properties
No C S UAG G W С-3 Mobility, sm Density kg/m3
1 334 932 837 – 191 1.8 5.0 2131
2 511 550 819 – 210 3.1 10.5 2055
3 293 814 – 1202 173 1.8 5.0 2458
of burnt ash gravel correspond to V25 and V40 classes allowing using it instead of
granite crushed stone.
The durability of concrete is determined not so much by the durability of
aggregates but by the firmness of coupling of cement stone with the grain surface
of the aggregates. Grain porosity of aggregates has a positive impact on coupling.
Due to water suction by porous aggregates in concrete mix, cement paste penetrates
open pores, thereby joining the cement stone and the aggregate.
Concrete deteriorates due to the cracks in both the cement stone and the aggregate
(Fig. 15.3) [4].
122 F. Kapustin et al.
15.4 Conclusions
References
1. Volzhensky, A. (1970). Unburnt artificial aggregates for lightweight concrete. Building Mate-
rials, 80–81.
2. Cherry, B. L., Ufimtsev, V. M., & Kapustin, F. L. (2006). Promising technology for removal,
storage and use of TPP ash-and-slag (pp. 11–110). Ekaterinburg.
3. Gorchkov, V. C., & Timashev, V. V. (1963). Methods physico-chemical analysis of binders
(pp. 80–130). Moscow: Higher School.
4. Chumakov, L. D. (2011). Technology of fillers of concrete (pp. 55–138). Moscow: Publishing
House of the Association of Construction Higher Education Institutions.
Chapter 16
Physical–Chemical Bases of a Glass Batch
Preparation Technique
16.1 Introduction
The increasingly demanding requirements for quality of modern glass call for a
serious analysis of the influence of various factors on glass batch uniformity. A glass
batch is a homogenous mixture of raw materials prepared beforehand and weighed in
accordance with a specific recipe. The main parameters of the glass batch quality are
its chemical homogeneity and the conformance to its chemical composition. Uni-
form distribution of batch components allows one to obtain glass homogeneity in
composition and structure. Batch homogeneity depends on the size distribution of
raw materials, the constancy of its chemical composition, the dosing accuracy, the
mixing conditions of batch components and the absence of segregation, dusting and
volatilization at the stages of batch transportation.
The preparation of raw materials is often referred to as the preparation of the
composition. This attitude to the preparation of initial components for the glass
melting was not formed accidentally and is due to the nature of the processes
occurring in the glass melting furnace. Due to relatively high viscosity of the glass
melt, glass furnaces are characterized by laminar motion of glass melt. Under these
conditions the only possibility of glass melt composition averaging is the diffusion
of individual ions and molecules. Since the diffusion rate is low, the homogenization
is possible only in small areas of melt during the time in the furnace. The scale of
glass melt inhomogeneity is predetermined.
The preparation of glass involves several operations, such as grinding, screening,
drying materials, dosing, feeding into the mixer of the individual components and
mixing them. At the last stage, glass batch is also wetted as this is prerequisite to
obtaining homogeneous glass and can significantly reduce dust particle carryover as
1.7 1.7
1.6 1.6
90/10 90/10
1.5 1.5
80/20 80/20
70/30
3
1.4
Bulk density, g/cm
70/30
3
1.4
1.2 1.2
1.1 1.1
1.0 1.0
0.9 0.9
0.8 0.8
1 2 3 4 5 1 2 3 4 5
Layer no. Layer no.
1.7 1.7
90/10 90/10
1.6 1.6
80/20 80/20
1.5 70/30 1.5 70/30
3
3
1.4 1.4
Bulk density, g/cm
1.3 1.3
1.2 1.2
1.1 1.1
1.0 1.0
0.9 0.9
0.8 0.8
1 2 3 4 5 1 2 3 4 5
Layer no. Layer no.
(а) 30 times (b) 300 times
Fig. 16.2 The bulk density of two-component batch “quartz sand–soda” at different layers with 4%
wetting. (а) 30 times (b) 300 times
126 V. Deryabin et al.
1.7
1.7
1.6 90/10
1.6 90/10
80/20 80/20
1.5
70/30 1.5 70/30
3
1.4
Bulk density, g/cm
3
1.4
1.1 1.1
1.0 1.0
0.9 0.9
0.8 0.8
1 2 3 4 5 1 2 3 4 5
Layer no. Layer no.
(а) 30 times (b) 300 times
Fig. 16.3 The bulk density of two-component batch “quartz sand–soda” at different layers with 4%
wetting and 40 min conditioning time. (а) 30 times (b) 300 times
As a result, the bulk density of the first layer to the fourth remains constant. When
mechanical impact is prolonged, conglomerates of sand particles fixed by layers of
water are destroyed, and there is a gradual increase in bulk density from the upper
layer to the bottom.
Conditioning moistened “silica sand–soda” mixture (Fig. 16.3) results in a more
uniform distribution of components given the same mechanical effects. In the upper
layers, bulk density is higher, and it is lower at the bottom. The reason for such
distribution is capillary-crystallization fixing of individual conglomerates, which
occurs during the conditioning period. The characteristics of this fixing have been
studied in Refs. [2–4]. The bigger the mechanical impact, the more proportional is
the change in bulk density of the mixture.
Dry dolomite was crushed and ground to study its fraction separation. As a result,
we obtained the following ratio of fractions of dolomite wt.%: 1–2 mm – 38.89;
0.5–1 mm – 22.22; lower than 0.5 mm – 37.
The initial batch with uniform distribution of fractions was shaken 30, 100 and
300 times. Each composition was divided into three layers, and the content of each
dolomite fraction in portion was evaluated. Firstly the segregation of various frac-
tions of dried powder dolomite was studied, and then the dolomite powder was
moistened with water. The moisture of the powder under consideration was 1%
and 4%.
Well-mixed powder containing particles of different sizes in each case was
shaken 100 times. For further experiments the dolomite powder was moistened
with 1 mol/l soda solution. The powder was shaken 100 times in all cases.
16 Physical–Chemical Bases of a Glass Batch Preparation Technique 127
100%
Fracture content, wt.%
100%
90% 90%
36.60
30.47
37.00
43.58
37.00
38.29
80%
80%
51.71
70%
70%
60%
17.87
60%
22.22
17.75
22.22
19.28
50%
47.20
50%
15.88
40.01
40%
40%
30%
45.53
30%
38.89
37.19
43.96
38.89
20%
32.32
20%
16.41
22.33
10%
10%
0%
0%
initial 1 2 3
initial 1 2 3
Layer no. Layer no.
< 0.5 mm 1-0.5 mm 2-1 mm < 0.5 mm 1-0.5 mm 2-1 mm
(а) 30 times (b) 100 times
Fig. 16.4 Distribution of fractions of a dry dolomite powder in layers of the batch after mechanical
impact. (а) 30 times (b) 100 times
100% 100%
23.10 13.39
21.22
90% 90%
32.93
37.00
37.00
40.77
45.43
80% 80%
Fracture content, wt.%
65.50
70% 70%
17.09
60%
23.14
60%
22.22
22.22
20.23
50% 50%
40%
39.57
40%
63.51
44.92
30% 30%
38.89
43.94
38.98
27.34
38.89
20% 20%
14.99
10% 10%
0% 7.16
0%
initial 1 2 3 1 2 3
initial
Layer no. Layer no.
< 0.5 mm 1-0.5 mm 2-1 mm < 0.5 mm 1-0.5 mm 2-1 mm
(a) 1% wetting (b) 4% wetting
Fig. 16.5 Distribution of fractions of moistened dolomite in layers of the batch. (a) 1% wetting (b)
4% wetting
The moisture of the batch was 4%. Experiments involving the batch were performed
immediately after wetting and after a 40-min conditioning period. The graphs
(Figs. 16.4, 16.5, 16.6, and 16.7) illustrate the dependences of the distribution of
various fractions of material throughout batch layers (upper layer of the mixture
components is marked 1, the lower layer – 3).
128 V. Deryabin et al.
100%
100%
90%
36.85
37.00
38.39
29.32
90%
41.60
37.00
39.41
80%
70%
70% 4.84
60% 9.65
22.22
60% 8.89
6.07
22.22
50%
50%
46.65
40%
40%
65.84
53.50
49.41
30%
49.51
54.52
30%
38.89
38.89
20%
20%
14.97
10%
10%
4.74 0%
0%
initial 1 2 3 initial 1 2 3
Layer no. Layer no.
<0.5 mm 1-0.5 mm 2-1 mm 2-1 mm 1-0.5 mm < 0.5 mm
Fig. 16.6 Distribution of fractions of powder dolomite 1% moistened by 1 mol/l soda solution in
layers of batch. (а) without conditioning time (b) 40 min of conditioning time
100%
100%
90% 90%
35.17
35.62
37.00
36.58
37.00
39.79
43.99
41.20
80%
Fracture content, wt.%
Fracture content, wt.%
80%
70% 70%
60% 7.77 10.21 60%
22.22
22.22
10.78 8.10
50% 50%
40.94
40.29
40% 40%
56.61
54.62
52.11
30% 30%
48.02
38.89
38.89
20% 20%
23.12
15.72
10% 10%
0% 0%
initial 1 2 3 initial 1 2 3
Layer no. Layer no.
< 0.5 mm 1-0.5 mm 2-1 mm < 0.5 mm 1-0.5 mm 2-1 mm
(а) without conditioning time (b) 40 minutes of conditioning time
Fig. 16.7 Distribution of fractions of powder dolomite 4% moistened by 1 mol/l soda solution in
layers of batch. (а) without conditioning time (b) 40 min of conditioning time
As shown in Fig. 16.4, the proportion of the coarse fraction is reduced when
passing to the bottom layer.
This tendency can be observed in all experiments. However, the middle fraction
demonstrates the opposite behaviour: its content increases when passing to the
bottom layers. These two fractions change their contents consistently. The content
of the fine fraction increases slightly only in response to minor external impact when
16 Physical–Chemical Bases of a Glass Batch Preparation Technique 129
passing to the bottom layer. With significant external impact, its share increases
when passing to the middle level and decreases in the bottom layer. It should also be
noted that the most significant change is observed on passing from 30 to 100 shak-
ings; the character of changes further remains the same.
When the powder dolomite is moistened by 1 wt.% water (Fig. 16.5), the amount
of coarse fraction is reduced from the upper layer to the bottom. This is especially
evident when passing to the third layer. The content of the fine fraction increases
uniformly. The amount of middle fraction becomes smaller only in the bottom layer;
the content of middle fraction virtually does not change on the other layers. At 4%
wetting the amount of coarse fraction is reduced uniformly when passing to the
bottom layer. The content of the fine fraction is increasing. It should be noted that on
the first and second levels, there is an averaging of batch composition: fine and
coarse fractions can be found in roughly equal proportions.
In general the dependences show that wetting with water changes the distribution
of fractions throughout batch levels. The amount of coarse fraction is decreased
when passing to the lower level, whereas the amount of small and middle fraction is
increased. Moreover, with a small amount of water, the changes between them are
insignificant, and there is redistribution when water content is of 4%.
It is known that the capillary force is directly proportional to the size of particles
[2]. A large fraction under water effect becomes independent of other fractions; the
capillary interaction of large particles leads to their consolidation and contributes to
the transition to upper level at the batch column height. Small fraction goes down,
and the capillary force is insufficient to hold the particles. For further experiments,
the dolomite powder was moistened by 1 mol/l soda solution. Humidity was 1 and
4%. The experiments were carried out with the prepared batch immediately after
wetting and with a 40-min conditioning time (Figs. 16.6 and 16.7).
Both experiments indicated almost identical dependences: the components fixing
on the first level and its change when passing to the third layer. The amount of
middle fraction on the first two layers does not change, and on the last layer, it
increases sharply. The characteristic feature of this series of experiments is that the
amount of fine fraction does not change significantly when passing from the upper
level to the lower one. This trend can be most clearly observed in the experiments
with 40-min conditioning time. The fine fraction was fixed in the pores between the
middle and the coarse fraction by means of crystalline aggregates obtained after
batch drying.
16.4 Conclusions
experiment is concerned, it has been revealed that the segregation of its components
occurs because of different particle sizes; the segregation is about 150–200%.
Wetting the mixture of “quartz sand–soda” fixes the initial distribution depending
on the conditions created for the formation of primary particle conglomerates fixed
by capillary forces as well as by the establishment of capillary-crystallization
contacts: the fuller the conditioning is, the higher the degree of the fixation. Wetting
powder dolomite with water does not significantly change the redistribution of
particles after shaking. When dolomite powder is moistened with salt solutions,
the crystals resulting from water evaporation reduce the mobility of the powder
particles thereby preventing segregation. Soda solution serves as a fixing component
due to the appearance of crystal (cementing) phase whereby the powder particles are
immobilized.
References
1. Deryabin, V. A., Farafontova, E. P., Paramonova, O. L., & Panov, I. V. (2012). Study of glass
batch components segregation. Glass and Ceramics, 68(9–10), 319–322.
2. Deryabin, V. A., Malygina, O. L., & Farafontova, E. P. (2006). Capillary counteraction to
segregation of particles in glass batch preparation. Glass and Ceramics, 63(1–2), 3–6.
3. Deryabin, V. A., Malygina, O. L., & Farafontova, E. P. (2006). Interaction between glass batch
particles via potassium-bearing interlayers. Glass and Ceramics, 63(1–2), 40–42.
4. Paramonova, O. L., Deryabin, V. A., & Farafontova, E. P. (2014). Investigation of component
segregation in wetted glass batch. Glass and Ceramics, 70(9–10), 315–318.
Chapter 17
Influence of Ion-Electron Interaction
at the Metal-Oxide Melt Boundary
on the Protective Properties of Enamel
Coatings
17.1 Introduction
Silicate-enamel coatings are widely used for the protection of metal pipeline trans-
portation equipment which is considered to be the most reliable and cheap variety at
the moment. In single-layered thin enamel coatings for industrial use, the enamel
layer is less than 0.2 mm. In that case the coating quality depends on the extent to
which a thin film of melt is enriched by iron oxides formed during the burning
process in enamelling. At high concentrations of these oxides during the cooling
process, the protective coating crystallizes and crumbles off the metal surface.
Modern research methods make it possible to study the boundary of interaction
metal-oxide melt, to determine the thickness of the adhesion layer and the distribu-
tion of elements throughout this layer. The study aims to investigate the mechanism
of metal oxidation during enamel coating burning and to identify the factors affect-
ing the formation of protective coating.
17.2 Experiment
The metal oxidation rate under the enamel melt layer in various atmospheres and
diffusive mobility of iron ions with given amount of Fe2O3 (10–20 wt. %) in melt
was measured by the gravimetric method using derivatograph Q-1500-D. The study
of the adhesion layer at the metal-enamel coating boundary was conducted using
SEM with energy dispersive microanalyzer Oxford Inca Energy 350.
In previous studies [1], we have determined that enamel enrichment by iron oxides
occurs not gradually due to the oxidation of iron surface but through the diffusion
into the areas of oxidation in spots where the coating has worn thin (sharp corners,
edges, holes, ribs).
The oxidation process under the melt layer enriched by iron oxides is realized due
to air oxygen in air atmosphere or due to oxygen in argon and oxygen atmosphere
[2]. In that case, oxygen at the film melt boundary is involved in the reaction
1 1
Fe2þ þ O2 ¼ Fe3þ þ O2 : ð17:1Þ
4 2
The concentration of Fe3+ ions at the boundary with the atmosphere is determined
by the law of mass action:
pffiffiffiffiffiffiffiffiffi
CFe3þ aO2
K¼γ , ð17:2Þ
C Fe2þ PO1=4
2
where K, equilibrium constant for reaction (17.2), CFe3þ and C Fe2þ are ion concen-
trations at the boundary with atmosphere, γ, coefficient taking into account the
activity of iron ions:
aFe2þ C 3þ
¼ γ Fe : ð17:3Þ
aFe3þ C Fe2þ
Hence the concentration of the oxidizer (Fe3+ ions) at the boundary with atmo-
sphere is
k CFe2þ PO1=4
0
CFe3þ ¼ pffiffiffiffiffiffiffiffiffi 2 ¼ k PO1=4 : ð17:4Þ
γ aO2 2
At the boundary with metal, the concentration of Fe3+ ions is determined by the
equilibrium of iron oxidation reaction (17.5). The equilibrium of this reaction is
significantly shifted to the right [1], and we can, therefore, assume that CFe3þ 0. Let
us suppose that the process (17.3) rate is limited by the diffusion of Fe3+ ions
through the melt film:
In that case
DFe3þ D 3þ
V gr ¼ C Fe3þ or taking into account Eq:ð17:4Þ, V gr ¼ Fe k 0 PO1=4 :
δ δ 2
ð17:6Þ
17 Influence of Ion-Electron Interaction at the Metal-Oxide Melt. . . 133
lgPO2
0.8
0.6
0.4
0.2
2.00 2.02 2.04 2.06 2.08 2.10 2.12 2.14 2.16 2.18
lgVgr
Fig. 17.1 Correlation of advancement rate of the oxidation boundary along 08KP steel under a
layer of Na2O∙aB2O3 melt and partial pressure at 1073 K
160
Vgr, mm/h
140
120
100
2 4 6 8 10 12 14 16 18
PO2, vol. %
Fig. 17.2 Advancement rate of oxidation boundary for Na2O∙aB2O3 melt along 08KP steel versus
air partial pressure
This implies that reducing oxygen partial pressure in the gas phase leads to the
decrease in the oxidation rate Vgr. Therefore, one can expect that for sufficiently
small values PO2 , the oxidation zone distribution rate will be determined by the iron
oxidation rate according to reaction (17.1). In this case the magnitude of Vgr should
be proportional to PO1=4 .
2
Figure 17.1 illustrates the dependence of lgV gr ðlgPO2 Þ at PO2 ¼ 2.18.26102 hPa
for melt Na2O∙2B2O3 at 1073 К. It can be seen that dependence is linear at PO2 ¼
6.22 hPa.
The value of n found from slope ratio of this curve equals 0.276, which is
congruent to the theoretic value calculated according to the equation
E
V gr ¼ Aexp : ð17:7Þ
RT
As expected, in this case the oxidation boundary moves progressively rather than
cyclically. At PO2 ¼ 1025102 hPa, the rate Vgr decreases from 153 to 130 mm/h
(see Fig. 17.2). This is because at PO2 102 hPa the diffusion of the oxidation zone
becomes cyclic.
The dependence of the oxidation rate on the temperature in air atmosphere can be
described by an exponential Eq. (17.7). For Na2O∙2B2O3 and B2O3 melt, this rate
equals 185 and 189 kJ/mol, respectively. These values are close to the theoretical
ones [1]. These data once more confirm that viscous flow of melt significantly
influences the speed of the process described above.
134 O. Lazutkina et al.
As shown above, if melt contains iron ions, there is an appreciable oxidation rate
under the enamel melt layer in oxidation atmosphere. The air oxygen is involved in
the oxidation process, and the total oxidation process can be described as follows:
1
2Fe2þ þ O2 ¼ 2Fe3þ þ O2 , ð17:8Þ
2
Fe þ 2Fe3þ ¼ 3Fe2þ , ð17:9Þ
1
Fe þ O2 ¼ Fe2þ þ O2 ¼ FeO: ð17:10Þ
2
Process (17.9) can occur without air oxygen but in this case as the concentration
of Fe3+ ions in the enamel melt film decreases, so does the metal oxidation rate.
Let us compare Fe, Co, Ni oxidation curves obtained by thermogravimetric
method in Pt crucible. The melt layer under samples was about 5 mm. The mass
ratio of sample and melt was as 1:5. The experiments were carried out in air
atmosphere. One can see a definite effect of ions of different valence on metal
oxidation under enamel melt layer. At high temperatures 11001400 К, only the
NiO is stable [3], so in the melt Ni, it is presented only as Ni2+. It was established that
at 1373 K, the nickel sample under enamel melt layer did not vary its mass
significantly.
When comparing the oxidation curves for Fe and Co samples covered by enamel
melt layer, it can be observed that the Fe oxidation is more intensive than that of Co.
This is due to the fact that the Co3+ ions equilibrium concentration is higher than that
of Fe3+ at PO2 ¼ const.
Let us assume that the ion portion of Мe2+ (Fe2+, Co2+) is N Me2þ in two initial
melts, and in equilibrium state, the concentration of Fe3+, Fe2+, Co3+, Co2+ ions is
P P P P
equal to N Fe 3þ , N Fe 2þ , N Co3þ , and N Co 2þ , respectively. It is evident that
N Me2þ ¼ N Me3þ þ N Me2þ . In air atmosphere, the equilibrium of Me2+ and Me3+ ions
P P
1
2FeO þ O2 ¼ Fe2 O3 , ð17:11Þ
2
1
3CoO þ O2 ¼ Co3 O4 : ð17:12Þ
2
Reaction (17.12) considers the fact that at high temperatures only Co3O4 [3] is
stable. The ionic form of these reactions can be written as follows:
1
2FeO þ O2 ¼ 2Fe3þ þ O2 , ð17:13Þ
2
1
2CoO þ O2 ¼ 2Co3þ þ O2 : ð17:14Þ
2
The equilibrium constants of these reactions are:
17 Influence of Ion-Electron Interaction at the Metal-Oxide Melt. . . 135
a2Fe3þ þ aO2
K 31 ¼ 1=2
, ð17:15Þ
a2Fe2þ þ PO2
a2Co3þ þ aO2
K 32 ¼ 1=2
: ð17:16Þ
a2Co2þ þ PO2
1=2 N P 3þ 1:48aO2
K 31 ¼ Fe 2 , ð17:17Þ
N Me2þ N FeP
3þ
1=2 N P 3þ 1:48aO2
K 32 ¼ Co 2 : ð17:18Þ
N Me2þ N CoP
3þ
Consequently
1=2
K 31 N Me2þ
3þ ¼ ð17:19Þ
P
N Fe 1=2 1=2
,
K 31 þ 1:48aO2
1=2
K 32 N Me2þ
3þ ¼ : ð17:20Þ
P
N Co 1=2 1=2
K 32 þ 1:48aO2
At 1373 K lgК31 ¼ 4.4 and lgК32 ¼ 0.75. The value aO2 < 1, which is why the
3þ N Me2þ . Substituting
P
second term in Eq. (17.19) can be neglected; in that case N Fe
the result into Eq. (17.20), we obtain
1=2
3þ K 32
P
N Fe
P
N Co3þ ¼ 1=2 1=2
¼ bN Fe
P
3þ : ð17:21Þ
K 32 þ 1:48aO2
3þ > N
P P
As b < 1 then it can be concluded that N Fe Co3þ .
In the real conditions of oxidation at the air boundary, the equilibrium of reactions
(17.13) and (17.14) is established. The concentration of Me3+ ions at the boundary
with metal is determined as follows:
The equilibrium constants of these reactions equal lgК40 ¼ 2.1, lgК41 ¼ 5.9 at
1373 K, respectively. As these values are too high, considering that Me3+ ions
concentration at the boundary with metal is very small, it can be assumed that
3þ 0.
P
N Me
136 O. Lazutkina et al.
1.0
Dm, mg/sm2
Fe
0.5 Co
Ni
0.0
5 10 15 20 25
t, min
Fig. 17.3 Oxidation of Fe, Co, Ni under Na2O∙aSiO2 melt layer in air at 1273 K
Assuming that the metal oxidation rate is determined by Me3+ ions diffusion to
metal surface, we find
P
CMe 3þ
V k ¼ DMe3þ , ð17:24Þ
δ
3+
P
where C Me 3þ is the concentration of Co and Fe3+ ions near the melt boundary with
3þ > C
P P
gas. As C Fe Co3þ , hence Fe, according to reaction (17.22), should oxidize more
quickly than Co, according to reaction (17.23), and Ni should not oxidize in the least
3þ ¼ 0. Experimental data (Fig. 17.3)
P
with higher valence ions participation as C Ni
support this conclusion. Figure 17.3 represents the oxidation results at 1273 K of Fe,
Co, Ni samples covered by melt film with insert of FeO, CoO, NiO in amount of
4 mol. %, respectively.
The average oxidation rates of iron and nickel found using straight lines slope at
τ > 5 min are 2.4 and 0.36 mg/sm2, respectively. The iron oxidation rate is
significantly higher than that of cobalt, whereas the nickel oxidation rate is about
zero. In air atmosphere, the decisive role in metal oxidation process under enamel
melt layer is played by metal ions of different valence in the melt. The contribution
of air oxygen is restricted to ions oxidation Me2+ to Me3+ on the film-air boundary.
If the oxidation process of metal with oxygen depolarization, which can occur
during the molecular oxygen diffusion through the film, has evolved considerably,
the nickel would oxidize at 1273 K, according to the reaction
Experience has shown that this is not observed. This confirms the conclusion that
the corrosion process with oxygen depolarization in the system oxygen-oxide film
enamel melt-metal is not essential.
Using energy dispersive microanalyzer, we have performed the qualitative and
quantitative X-ray microanalysis of the adhesion layer at metal-enamel boundary.
The metal-enamel penetration zone is a rough surface. The inclusions of irregular
round shape consisting of metal-oxide film fragments break off from the metal
surface to enamel. Moreover, in metal-enamel interface, heterogenic structures are
formed [4, 5]. These structures consist of dendritic crystals Fe2SiO4, NiO-Fe2O3,
CoO-Fe2O3 of lighter colouration (see Fig. 17.4).
17 Influence of Ion-Electron Interaction at the Metal-Oxide Melt. . . 137
Fig. 17.4 The results of X-ray analysis of metal-enamel coating boundary layer
17.4 Conclusion
The experiments have shown that the iron oxidation rate is considerably higher than
that of cobalt, while nickel oxidation rate virtually equals zero. We can, therefore,
conclude that in air metal ions of different valence play the key role in the metal
oxidation process under enamel melt layer.
The role of atmosphere oxygen is restricted to the oxidation of Me2+ to Me3+ on
the boundary between the film and the air. The corrosion process with oxygen
depolarization in the system oxygen-oxide film of enamel melt-metal is of no
particular significance. During Fe and Co oxidation under the enamel melt layer
containing ions Fe3+, Fe2+, Co3+, Co2+ of variable valence, the metal ions in melt
play the central role in the oxidation process.
During the firing process on certain spots of the surface of steel covered with
enamel melt layer in air, there appear foci of intense metal oxidation with pro-
nounced dark colour, where Fe2O3 concentration is significantly higher than 20%.
At these sites the enamel melt is enriched with oxides of variable valence to the
extent that it obtains electronic conductivity, and this makes the oxidation process
spread in molten enamel throughout the metal surface with constant speed.
There are no depositions of oxide film or metal films of adhesion oxides (Co, Ni)
on the Fe surface. These metals are presented in inclusions of NiO-Fe2O3, CoO-
Fe2O3 compositions drifting in enamel near the metal surface.
References
1. Lazutkina, O. R. (2013). Substrate metal oxidation mechanisms for the appearance of defects in
enamel coatings. Glass and Ceramics, 69(11–12), 416–419.
2. Gutiérrez-Castañeda, E., Salinas-Rodríguez, A., Deaquino-Lara, R., & Márquez-Torres,
F. (2015). High temperature oxidation and its effects on microstructural changes of hot-rolled
138 O. Lazutkina et al.
low carbon non-oriented electrical steels during air annealing. Oxidation of Metals, 83(3–4),
237–252.
3. Esin, O. A., & Gel’d, P. V. (1962). Physical chemistry of pirometallurgical processes [in -
Russian] (p. 671). Moscow: Metallurgiya.
4. Lazutkina, O. R., Kazak, A. K., & Ponomareva, M. V. (2011). Effect of enamel layer thickness
on defectiveness of enamel coatings. Glass and Ceramics, 68(1–2), 33–35.
5. Deryabin, V. A., & Uglinskikh, M. Y. (2014). Change in the structure of fired glass enamels with
different iron oxide content. Glass and Ceramics, 71(7–8), 258–259.
Chapter 18
Study of Enamel Slip Particles Interaction
Using Model Systems
18.1 Introduction
During batch preparation for enamel frit synthesis, obtaining a homogeneous mix-
ture of raw materials is of considerable importance. Water droplets falling in
between the solid particles of batch result in the formation of capillary water layers
causing particle attraction [1]. Water menisci arising at low moisture content
between solid particles cause capillary attraction and particle convergence
[2, 3]. As capillary forces have considerable significance in the process of enamel
batch preparation, it is necessary to study them.
18.2 Experiments
I II III IV
clay particles get closer together resulting in shrinkage and the formation of air
voids.
Due to the reduction of liquid volume, isolated capillary layers between particles
appear [6]. Concave capillary isthmuses arising at the contacts of solid particles
trigger the convergence of solid surfaces. While drying, solid particles are pulled
together by capillary bridges and are drawn to the product surface, and hereby biscuit
is formed. At the stage of free capillary adhesion of separate particles, the attractive
forces may contribute to the reorganization of the slurry structure. Until the forma-
tion of crystalline aggregates the reorientation and convergence of solid constituents
of low water, slurry is possible. The greater the capillary adhesion force is, the more
likely the local compaction and reorientation of solids is.
The mechanisms of capillary-dispersed system behavior mentioned above are the
subject of the research. During the experiments, the glass particle was in contact with
the liquid phase since a droplet was applied on it. Throughout the experiment,
persistent evaporation of water from the slurry was observed. There are four typical
stages of the experimental test of capillary-crystallization strength. These stages are
schematically presented in Fig. 18.1.
Stage I – the beginning of the experiment. A liquid drop is deposited onto the top
of the glass spherical particle. The model particle is withdrawn from the flat bottom
substrate.
Stage II – a flat substrate is placed under the glass particle using a microscrew. A
liquid drop forms a meniscus between the spherical particle and the substrate. This
position is observable both during the attachment and the detachment of the glass
particle and the substrate.
Stage III – the flat glass substrate comes into contact with the spherical particle.
Step IV – the glass particle is withdrawn from the flat substrate, whereby the
liquid droplets prevail both on the glass particle and on the flat substrate.
As part of the present research, experiments using model glass particles have been
conducted to study the capillary and crystallization strength of solid particle adhe-
sion during the process of capillary layer evaporation. Capillary layer was obtained
through the immersion of a glass ball into distilled water, slip of coating enamel,
ground enamel, kombi-ground enamel, and their water constituent. The experiments
were performed with changing cup volume due to the evaporation of liquid. In the
18 Study of Enamel Slip Particles Interaction Using Model Systems 141
case of mobile liquid, the evaporation progresses more rapidly, which can cause
more rapid crystallization of liquid solution and adhesion of particles.
The droplets of solutions under consideration form concave menisci attracting
particles and bringing them closer up to the immediate contact. When the external
force causing the deformation of torsion balance coil exceeds the maximum capillary
strength, the capillary isthmus ruptures. The force was fixed at the time of formation
of the cup as well as at specified intervals.
The measurements were performed in dynamics since the crystal structure was
constantly forming and further collapsing.
The experiment was repeated several times until the liquid layer dried out or the
attraction forces of model particles disappeared.
Based on the experimental data, the dependence of capillary-crystallization
strength of coupled model particles on contact duration has been plotted on the
graphs.
The average adhesion force at the initial time is 0.6 mN (Fig. 18.2). The measure-
ment ends at time lags of 6–8 min. Increasing water contents makes it possible to
take a large number of measurements. As a result, the experiment is prolonged.
This is due to the increased volume of the liquid layer. Extending the exposure
time during the successive measurements leads to the rise of the adhesion force. This
is due to the occurrence of crystallization component as the filtrate is a salt solution.
In the withdrawn state, crystals grow from the interface. The solid content in slip
slurry is from 50% to 60% whereas the content of particles smaller than 0.14 mm in
4 0.036 g
Adhesion force, mN
0.037 g
3 0.034 g
0
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 5.5 6 6.5 7 7.5 8
Duration, min
4 0.050 g
Adhesion force, mN 0.051 g
3 0.045 g
0
0 0.5 1 1.5 2
Duration, min
the filtrate is not more than 0.2% in accordance with the requirements of the
laboratory filter paper (GOST 12026–76).
The interaction forces between different particles between molecules or ions in
solution are not identical [7]. Water has higher surface tension σ ¼ 72.8 MJ/m2
which is due to the relatively strong hydrogen bonds. In molecule solutions, the
particles forming weaker bonds are pushed up into the surface layer. Thereby the
surface tension decreases. The reduction of the surface tension in the solutions under
study is caused by such compounds as sodium nitrite and others.
It should be noted that in this case, the number of experiments is far lower than in
previous one. This is due to the high viscosity (103.5–104 Pa∙s) and low water content
(approximately 40%) in the slurry, which resulted in rapid drying out. Figure 18.3
shows that not all measurements could be taken. The drying process occurred too
fast, and, as a result, the adhesion force exceeds the upper limit of the measuring
instrument. The exposure was conducted in the withdrawn state (without solids
contact). Crystals grew from the interface. At high viscosity evaporation is in
progress too. In this case there appear crystallized areas in the layer bordering with
air. Since evaporation occurs from the liquid surface, the high viscosity fluid does
not have time to flow to the surface layer, and solely the cup shell hardens.
Therefore, when viscosity is high, only the outer layer of the cup hardens.
The maximum adhesion force of enamel coating is 2.36 mN. The maximum
number of measurements is 2.
18 Study of Enamel Slip Particles Interaction Using Model Systems 143
4 0.035 g
Adhesion force, mN 0.034 g
3 0.037 g
0
0 0.5 1 1.5 2 2.5 3 3.5 4
Duration, min
During the experiment with water component of ground enamel slip, it had been
established that virtually all measurements finish within 3 min. The sample weight is
between 0.034 and 0.037 g. It should be mentioned that the viscosity of the aqueous
component is much lower than that of the enamel slip. A greater mass leads to the
increase in the volume of the solid phase, adhesion force and the number of
experiments. The pattern persists. However, the number of points of the ground
enamel filtrate fixed was lower than that of the coating enamel filtrate. Measuring at
time lags of 1 min made it possible to register the maximum adhesion force, which
equals 0.83 mN as the crystallization component appears (Fig. 18.4).
In contrast to the water component, the number of measurements for the slip ground
is much lower since the slip ground enamel is characterized by high viscosity and
low amount of liquid layer. The greater the sample mass is, the longer the experiment
is. The average value of adhesion strength at the initial time ranges from 0.8 to 1.0
mN. Thus, the general pattern in case of the slip ground enamel compared to the
coating enamel slip persists. The adhesion force of the slip is considerably greater
than in previous measurements. The viscosity of the ground enamel slip (102–103
Pa∙s) is lower than that of the coating enamel slip (103.5–104 Pa∙s); water content is
greater, and so is the number of measurements.
In the case of a cup at a low viscosity, the solidification occurs throughout the cup
volume. At low viscosity liquid flows more freely to the air interface, and it can be,
therefore, assumed that the entire capillary layer hardens as a result. If the device had
144 V. Deryabin and I. Tochilova
0
0 0.5 1 1.5 2 2.5 3 3.5 4
Duration, min
Fig. 18.5 Dependence of capillary-crystallization strength on contact duration (cup material – the
ground enamel slip, diameter of spherical particle 2.96 mm)
4 0.016 g
Adhesion force, mN
0.015 g
3 0.014 g
0
0 0.5 1 1.5 2
Duration, min
had a broader measuring range, it would have shown that low viscosity causes higher
strength, as the whole cup is crystallized (Fig. 18.5).
It is vital to note that the average initial weight equaled 0.015 g, which is more than
two times higher than the average weight of the water component of the coating
enamel slip at the initial point of the experiment. This is due to the increased
viscosity of the slip. The average adhesion strength at the initial time is 0.75
mN. As the graph demonstrates, few values could be measured due to fast drying
(Fig. 18.6).
18 Study of Enamel Slip Particles Interaction Using Model Systems 145
4 0.013 g
Adhesion force, mN 0.016 g
3 0.015 g
0
0 0.5 1 1.5 2
Duration, min
In the study of the adhesion force of the kombi-ground slip, only two graphs could be
drawn since at 1.5 min intervals the slip dries out very fast, and the time is
insufficient to make any measurements. Both lines are directed straight upward
due to the emergence of the crystallization component. When measured at a time
lag of 30 s, the slurry starts to dry out at the point of 0.5 min (Fig. 18.7).
To study the crystallization process of the water component of enamel slip, similar
experiments have been carried out, but not taking into account the adhesion force.
The drops of water component of enamel slip were analyzed under the microscope
after liquid evaporation to determine the patterns of particle distribution (Fig. 18.8).
The macrostructure of crystalline precipitate obtained after drop drying of water
component of kombi-ground slip (Fig. 18.8a) is inhomogeneous. A crystalline roller
has been formed at the drop edge. A secondary ring of small particles is emerging
along the drop edge.
The structure of the water component of coating enamel slip (Fig. 18.8b) is also
inhomogeneous. There are dense rings of particles resulted from approximating and
withdrawing the glass particle. The area of ring formation is wider than in case of
kombi-ground slip, which is probably due to the composition of water component.
As far as the water component of ground enamel slip (Fig. 18.8c) is concerned, it
shows heterogeneous structure and clear zoning. There is a crystalline roller at the
drop edge. There is a dark area: dark crystals fall from outside whereas smaller
crystals are inside.
146 V. Deryabin and I. Tochilova
Fig. 18.8 Macrostructure of crystalline precipitate (35x) obtained after drop drying of water
component of kombi-ground slip (a), coating enamel slip (b), and ground enamel slip (c)
The analysis of droplets of all filtrates has made it clear that there is consistent
formation of precipitate with crystals deposition mainly at the drop edge. As the
wetted perimeter is extended, salt deposition due to evaporation is observed, and this
is the cause for the ring-shaped distribution of precipitate.
18.4 Conclusion
When the capillary liquid volume is increased, the retention of liquid phase at the
particles contact is prolonged. If distilled water is used as capillary liquid, it is
possible to register the maximum amount of experimental points. When the filtrate
is used as capillary liquid, crystallization occurs due to water evaporation, and, as a
result, the retention time of liquid capillary phase is reduced. If enamel slip is used as
capillary liquid, in some cases the contact strength exceeds the upper limit of the
measuring instrument.
For various types of slips the capillary-crystallization effect reveals itself differ-
ently. For example, ground enamel slip shows longer retention time of liquid phase
at the particles contact in contrast to coating enamel slip as the viscosity of ground
enamel (102–103 Pa∙s) is lower than that of coating enamel (103.5–104 Pa∙s), and
hence water content is greater. The maximum number of measurements was taken
for kombi-ground slip.
References
1. Saavedra, J. H., Rozas, R. E., & Toledo, P. G. (2014). A molecular dynamics study of the force
between planar substrates due to capillary bridges. Journal of Colloid and Interface Science, 426,
145–151.
2. Amarasinghe, P. M., Anandarajah, A., & Ghosh, P. (2014). Molecular dynamic study of capillary
forces on clay particles. Applied Clay Science, 88–89, 170–177.
18 Study of Enamel Slip Particles Interaction Using Model Systems 147
3. Shinto, H. (2012). Computer simulation of wetting, capillary forces, and particle stabilized
emulsions: From molecular-scale to mesoscale modeling. Advanced Powder Technology, 23
(5), 538–547.
4. Deryabin, V. A., Farafontova, E. P., & Malygina, O. L. (2001). Specifics of interaction of glass
batch particles via interlayers of two-component solutions. Glass and Ceramics, 62(1–2), 39–41.
5. Zarate, N. V., Harrison, A. J., & Litster, J. D. (2013). Effect of relative humidity on onset of
capillary forces for rough surfaces. Journal of Colloid and Interface Science, 411, 265–272.
6. Deryabin, V. A. (1997). Capillary forces in disperse systems [in Russian]: Izd-vo UGTU,
Ekaterinburg, 64 p.
7. Breinlinger, T., Hashibon, A., & Kraft, T. (2015). Simulation of the influence of surface tension
on granule morphology during spray drying using a simple capillary force model. Powder
Technology, 283, 1–8.
Chapter 19
Peculiarities of Formation of Phase
Structure and the Hydraulic Activity
of High-Calcium Fly Ash
19.1 Introduction
The mineral part of the original solid fuel is a complex, heterogeneous substance
consisting of formations related to multiple genetic classes. Among its basic types
are biogenic, sorption, concretion, infiltration, and terrigenous ashes. The latter is
composed of minerals consisting of coal seams: quartz, clay substance, feldspars,
micas, volcanic ash, and others. As a rule, this is the predominant class dominating in
all coals possessing moderate- and high-ash content. Low-ash lignites are charac-
terized mainly by biogenic and sorption ash, the main element being calcium. In
sorption ash calcium and, to a lesser extent, aluminum, iron, and magnesium form
salt-like compounds – humates – or complex ones of chelates type. Biogenic ash is
formed from the basic components of chlorophyll (calcium, silicon, magnesium,
potassium) and also traces elements entering the enzymes composition. As the
thermal metamorphism of coal from lignite to anthracite, their ash is depleted of
biogenic and sorption components [1, 2].
Concretionary ash is formed together with sorption ash on peat coal formation
stage. It is not associated with organic matter but represented by such mineral forms
as siderite, ankerite, dolomite, kaolinite, and others. Infiltration ash is formed in the
cracks of coal seams from hot or cold groundwaters and is represented by sulfides,
carbonates, sulfates, calcium, iron, quartz, and other minerals. The particle size of
сoncretion and infiltration ash can be very different, from molecular size to tens of
centimeters, and the share in the overall balance is extremely changeable.
In Russian scientific literature devoted to the issues of burning of Kansk-Achinsk
coal, there is a conventional distinction between “external” and “internal” ash. It is
assumed that “internal” ash is a combination of sorption and biogenic ash, while
“external” ash is represented by terrigenous ash and separate large particles belong-
ing to other genetic classes. Young Kansk-Achinsk coals are characterized by high
content of “internal” ash, up to 50–60%. With a decrease in ash content to less than
3%, the “internal” mineral part is revealed in coals. The content of Cao may increase
to 70% and above [3].
Three groups of fly ashes were investigated in the present work. Samples Nos. 1–6
were obtained at the Siberia TPS when burning brown coals in a flame burner in
question according to the molten slag removal (the temperature in the core of the
torch reaches 1600–1700 C). Samples Nos. 7–12 were obtained at the TPS when
burning brown coals in a flame burner in question according to the hard slag removal
(the temperature in the core of the flame is 1300–1400 C). Samples Nos. 13–16
were obtained as a result of experimental burning of brown coals at the commercial
pilot plant with circular fluidized bed (CFB) at a temperature of about 900 C.
19 Peculiarities of Formation of Phase Structure and the Hydraulic. . . 151
All the samples of ashes studied contain free lime (free СаО) and possess binding
properties. The hydraulic activity of ashes was assessed by the size limit of the
compressive strength. The results of the determination of the binding properties of
the fly ashes are presented in Table 19.1.
It was determined that the lightest fly ashes with bulk density 470–800 kg/m3
were formed in the furnaces CFB. The particles of such ashes are the aggregates of
tiny grains vitrified and mineralized remnants of organic origin in the form of
“chains” and “branches.” Fly ashes obtained by burning brown coals by flame
burners having bulk density of 990–1200 kg/m3 consist of coarse ball-type particles
of different size and color with the refractive index 1.49–1.79 [4].
As to the chemical composition of fly ashes, it is characterized by instability
(Table 19.2).
The reduction of the temperature of burning brown coals leads to a decrease in the
content of vitrified phase from 67.32% to 16.98% (according to the rational analysis
of chemical). The quantity of CaSO4 and СаСО3 increases (according to X-ray
phase analysis). It should be noted that among crystalline phases regardless of the
method of producing ash, β-quartz can be identified, free СаО, MgO, and also some
clinker minerals identified in the range of angles θ ¼ 15–18 [4].
152 I. Domanskaya and V. Oleinik
From the analysis of fly ashes, we can conclude that considerable differences of the
composition and fly ash properties obtained from burning brown coals in different
furnaces are first and foremost conditioned by the availability of “external ash”
(quartz, clay, feldspars, limestones, gypsum, and other enclosing rocks) and “inter-
nal ash” of biogenic origin. The flame burning way is characterized by a high degree
of coal grinding (up to 0.2 mm) providing more uniform distribution of mineral part
on fractions and better interaction of components of “external” and “internal” ashes.
For coal grains to be burnt in CFB, their diameter is adjusted – to 10 mm – and is
decreased as the even burning out takes place by multiple circulation through the
burning zone. Redistribution of mineral components depending on their grinding
and enrichment of fly ash CFB by components of “internal ash” is likely to take place
in this case.
In order to determine the most probable low temperature reactions of the products
influencing the hydraulic activity of fly ash afterward, thermodynamic calculations
of Gibbs energy were performed. It was established that the supposed CaO isolation
from the organic substance (400–600 C) and its interaction with sulfate component
of gas phase are more advantageous thermodynamically. Therefore, CFB fly ashes
compared to flame burned ones are enriched by anhydrite. Free CaO is likely to be a
product of CaCO3 decomposition coming from the “external” mineral part, also the
secondary one, the formation of which cannot be excluded taking into account the
calculated data (Table 19.3). As far as flame burned fly ashes are concerned, the main
19 Peculiarities of Formation of Phase Structure and the Hydraulic. . . 153
Table 19.3 Data of Gibbs –ΔGТ, kJ/mol/ C
energy calculation
Compounds 600 800 1000 1200
CaSO4 280.69 243.94 208.11 173.62
СаСОз 82.80 52.47 22.97 5.79
CaОSiО2 90.17 90.63 90.97 –
2 CaОSiО2 130.48 132.29 134.29 136.44
CaOA2O3 27.19 31.40 36.01 41.11
12 CaO7A2O3 191.42 234.17 276.75 319.99
2 CaOFe2O3 61.13 68.47 75.67 82.62
Fig. 19.1 Scheme of the transformations of the mineral part of brown coal
processes of the phase formation include desulfurization, solid phase interaction and
caking, melting, and crystallization from the melt (Fig. 19.1).
The analysis of the CaO-Al2O3–SiO2 state diagram and the fusibility curve
construction of Nos. 1–12 compositions allowed us to establish that the most likely
phase crystallizing on melt cooling consisting of more than 33–35% CaO is belite,
and on that consisting of less CaO content is gelenite. However, the low viscosity
indicators of the compositions under investigation in the latter case contribute to the
conservation of vitrified state at the moment of fly ash particles cooling down, which
is confirmed by the X-ray diagram character and is also indicative mainly of
pozzolana mechanism of their hardening when mixed with water. CaO free present
in flame burned ashes is likely to be the product of the secondary sulfates
decomposition.
Thus, it can be concluded that the phase composition and the hydraulic activity of
high-calcium fly ashes depend on their mineral part and the “external”: “internal”
ratio of ashes, the degree of grinding, temperature, the way of coal burning proper,
oxidation-reduction character of gas medium, the duration of the particles stay in the
furnace chamber, and the place of sampling in the system of fly ash trapping.
154 I. Domanskaya and V. Oleinik
References
1. Karlowska, M. S., & Rivkin, A. S. (1973). In A. Ots (Ed.), The influence of the mineral part of the
Siberian coal contamination of the heating surfaces of steam generators. Novosibirsk: The
Institute of physical and chemical principles of mineral processing.
2. Yudovich Ya, E., & Ketris, М. P. (2005). Toxic trace elements in coal. Ekaterinburg: UrB RAS.
3. Gamow, M. I., & Granovsky, N. (2012). In S. V. Levchenko (Ed.), Metals in coals. Rostov-on-
Don: SFU.
4. Domanskaya, I. K., & Oleinik, V. N. (2006). High-calcium fly ash: Forming of composition and
hydraulic activity. Ibausil 16. Internationale Baustofftagung – Weimar, рр. 1-1071–1-1079.
Part IV
High-Pressure Treatment of Metals and
Alloys
Chapter 20
Modernisation of the Method of Rolling
of Hot-Rolled Tubes
20.1 Introduction
The advantages of pipe-rolling plants (PRP) with an automatic plug rolling mill are
high maneuverability enabling one to change over from one pipe dimension type to
another within a short time; relatively high output (150,000–250,000 ton of pipes per
year) [1–3]; and the ability of making large tonnage batches of a wide range of rolled
products from carbon, alloy and high-alloy steels. The disadvantages of PRP with an
automatic plug rolling mill are low elongation ratio, long time of auxiliary operations
and non-uniform rough tube wall thickness. The installation of a continuous mill
instead of an automatic one would eliminate the above-mentioned shortcomings;
however, the reconstruction of a PRP always involves considerable material costs.
Besides, new equipment may fail to be installed in place of the old within the existent
shop layout. In these terms, it seems expedient to improve the current process of
plugging.
The PRP-140 pipe-rolling plant made by the “Sintz” public joint-stock company,
which comprises a tandem-type automatic plug rolling mill consisting of two length-
wise rolling mills (LRM-1 and LRM-2) has enabled the time of auxiliary operations
to be reduced due to a sequential arrangement of the stands. However, the elongation
ratios remain rather low. With LRM-1, the elongation ratio is λ ¼ 1.161.5, and with
LRM-2 it is λ ¼ 1.071.15. Rough tubes have transversely non-uniform wall
thickness, which is partially eliminated during rolling by a three-roll sizing mill.
The study aims at improving the plug rolling process, when rolling is performed
with high elongation ratios, and wall thickness variation of rough tubes becomes
less. A higher elongation ratio in plug rolling will enable PRP-140 to produce
thin-walled pipes with the ratio D/S >33 and to reduce the elongation ratio in
piercing. A lower elongation ratio in piercing will increase the precision of shells
and decrease rolling skin.
It was found during industrial research that a higher elongation ratio on an automatic
tandem mill causes increased spoilage due to various defects, including “guide
marks.” A “guide mark” defect appears as follows (Fig. 20.1): (1) when rolling is
performed on LRM-1, there is intensive metal flowing into the tapers, and this results
in thicker pipe walls in the tapers than in the upper part of the groove. This causes
stripes on the pipe surface (Fig. 20.1a); (2) after turning about the pipe axis through
90 , the pipe is rolled in LRM-2; and the stripes get into the tapers (Fig. 20.1b). After
the groove is filled (Fig. 20.1c), the stripes are deformed on the mandrel, and this
causes jamming (Fig. 20.1d); (3) When rolling is performed in a three-roll blooming
mill and then in a stretch-reducing or sizing mill, jamming on the inner tube surface
turns into longitudinal “guide marks.”
Staper
Wall thickness in a taper is characterised by the nondimensional parameter Supper
(Staper and Supper are rough tube wall thickness, respectively, in the taper and the
upper part). Figure 20.1 presents the cross sections of a deformation zone for
plugging on LRM-1 (a) and LRM-2 (d, b, c).
Fig. 20.1 Scheme “guide mark” defect formation on the inner tube surface. (a) The formation of
stripes on the pipe surface. (b) Deformation of the stripes into the tapers. (c) The groove filling.
(d) The formation of the jamming
20 Modernisation of the Method of Rolling of Hot-Rolled Tubes 159
Fig. 20.3 A schematic of bite in the rolling of round (a) and oval (b) shells
In the course of computer simulation in the DEFORM software, it was found that
the stripe size can be decreased by mounting edger rolls before the stand of LRM-1,
which make the shell oval before plugging in LRM-1 (Fig. 20.2). When tasks were
set in DEFORM, the boundary conditions were specified according to the practical
data and recommendations given in Ref. [4].
When a round shell is fed into LRM-1, the bite begins at the tapers (Fig. 20.3a).
Flattening and some reduction result in the filling of the whole groove, the metal
flows intensively into the groove tapers, and the pipe wall becomes thicker in the
groove taper. When an oval shell is fed into a hexahedral groove, it is bitten by the
working rolls of LRM-1 in the upper part of the groove (Fig. 20.3b), not in the taper
zone. This reduces shell flattening and offers a better bite. Besides, edger rolls
restrain metal flow into the groove tapers and reduce the size of the stripes. Thus,
the stripes appearing during oval shell rolling are smaller than those appearing
during round shell rolling, and, consequently, the appearance of a “guide mark”
becomes less probable.
160 A. Bogatov et al.
1.9
1.85
1.8
1.75
Staper 1.7
Sup
1.65
1.6
1.55
1.5
1.5 1.6 1.7 1.8 1.9 2.0 2.1 2.2
λ
Staper
Fig. 20.4 The parameter Sup as dependent on the elongation ratio λ in rolling on LRM-1 with edger
rolls
The computer simulation has shown that, when the shell ovalisation coefficient
S
ξ ¼ 1.25 [5, 6], the nondimensional parameter Staper
up
decreases by 31%, i.e., the use of
edger rolls reduces the size of stripes.
According to experimental data on shell plugging
on LRM-1 without edger rolls,
Staper S
the ratio Sup must not exceed the critical value Staper
up
¼ 1.80 [7, 8]. When it does, a
cr
S
“guide mark” defect occurs. Figure 20.4 presents a plot of Staper
up
as a function of the
elongation ratio for rolling in LRM-1 with edger rolls.
It is obvious from Fig. 20.4 that, as the elongation ratio λ increases, the wall
S
thickness in the groove taper increases. When λ ¼ 2.00 and Staper
up
¼ 1.77, it does not
exceed the critical value. Thus, when edger rolls are placed before the stand of
LRM-1, the elongation factor can be increased from 1.50 to 2.00.
Fig. 20.5 A hard-body model of LRM-1 with two pairs of edger rolls
Staper
Table 20.1 Measured values of Sup as dependent on the kinematic tension coefficient ω
It is obvious from Table 20.1 that, as the kinematic tension coefficient grows, the
S S
parameter Staper
up
decreases. The lowest value of Staper
up
¼1.42 is observed when ω ¼ 1.1. It
has been found by computer simulation that this value of kinematic tension on
one to increase the elongation ratio to λ ¼ 2.40, with the critical
LRM-1 enables
S S
value of Staper
up
not being exceeded (when λ ¼ 2.40, Staper
up
¼ 1.73).
cr
20.4 Conclusion
It has been found within the study that shell ovalisation with the aid of edger rolls
before rolling in LRM-1 reduces the non-uniformity of rough tube walls. Besides,
with the maximum possible kinematic tension on LRM-1, that is, ω ¼ 1.1, one
manages to reduce wall thickening in the groove tapers considerably, thus decreas-
ing the probability of “guide mark” defects at high elongation ratios. Computer
simulation of pipe rolling in LRM-1 suggests that the elongation ratio can be raised
from 1.5 to 2.4. A higher elongation ratio in plugging enables one to produce thin-
wall pipes with D/S >33 on PRP-140 and to increase the precision of shell dimen-
sions due to lower elongation ratio in piercing.
162 A. Bogatov et al.
References
1. Kharitonov, V. V., Bogatov, A. A., & Vakhrushev, V. Y. (2008). Equipment for pipe rolling,
extrusion and drawing. Ekaterinburg: USTU-UPI.
2. Potapov, I. N., Kolikov, A. P., & Druyan, V. M. (1991). Theory of pipe making. Moskva:
Metallurgiya.
3. Danchenko, V. N., & Kolikov, A. P. (2002). Romantsev, B.A., pipe-making technology. Moskva:
Intermetinzhiniring.
4. Karamyshev, A. P., Nekrasov, I. I., & Pugin, A. I. (2010). A manual for the program deform-3D
software package. Ekaterinburg: UrFU Publ.
5. Bogatov, A. A., Pavlov, D. A., Lipnyagov, S. V., & Suvorov, V. N. (2012). Simulating the
formation of a “guide mark” defect on the inner tube surface in lengthwise rolling. Proizvodstvo
prokata, 8, 37–39.
6. Bogatov, A. A., Pavlov, D. A., & Dresvyankina, E. A. (2014). Development of a mathematical
model for a new process of lengthwise pipe rolling: In Metal forming 2014, fundamental
problems (Advanced materials and technologies, pp. 102–106). Ekaterinburg: UrFu Publication.
7. Bogatov, A. A., Pavlova, E. A., & Pavlov, D. A. (2014). Finite-element simulation of the process
of lengthwise pipe rolling on a stub mandrel. In: Proceedings of the 3rd International Conference
COMAT.
8. Bogatov, A. A., Pavlov, D. A., Lipnyagov, S. V., Suvorov, V. N., & Pavlova, E. A. (2014).
Rolling hot deformed pipes on a stub mandrel. Steel in Translation, 44(12), 912–915.
Chapter 21
New Engineering Solutions
in the Production of Laminated Composite
Pipes for the Oil Industry
21.1 Introduction
Russia is one of the world leaders in oil production and the largest participant in the
world energy market. Since the 2000s, oil production in Russia had been growing
steadily before this growth slowed down some time ago. Note that the conditions for
oil production in Russia have changed since the early 2000s. Thus, the development
of new oil fields shifts to more and more unfavorable regions in terms of geological
and climatic conditions, whereas oil extraction in regions discovered and developed
30–50 years ago is characterized by high water cut and intensive corrosive attack of
downhole equipment. Due to the simultaneous effect of such factors as the presence
of dissolved H2S and СO2 gases in the fluid to be extracted, abrasive wear and tensile
stresses, pump-compressor pipes (PCP) suffer the hardest conditions. This is verified
by the analysis of the reasons for the failure of downhole pumping equipment, which
demonstrates that the contribution from the failure of PCP into malfunction is
predominant, and it amounts to about 60%. The experience of operating corrosive
wells shows that the service life of PCP in them sometimes does not exceed 4 months
[1–7]. The operation of such a well is extremely difficult and involves considerable
costs for the restoration of the efficiency of the pump-compressor column. The
restoration costs for each well amount to about 350,000 RUB, annual emergency
failures of pump-compressor columns, including those in oil industry, are numbered
in thousands, and expenditures for failure elimination amount to hundreds of
millions RUB.
N. A. Bogatov (*)
“TEMP” Limited Scientific Production Association, Moscow, Russia
e-mail: nb0612@yandex.ru
A. Bogatov · D. R. Salikhyanov
Institute of Material Science and Metallurgy, Ural Federal University, Yekaterinburg, Russia
To replace PCP that have exhausted their service life, the pipe industry of Russia
annually provides oil companies with about 250,000 ton of new pipes. Medium-
carbon manganese steels (30G2, 35G2S, etc.) and chromium-molybdenum steels
(30KhMA) are more often used as pipe material; however, they are not corrosion
resistant enough.
The analysis of the scientific and technical documentation for PCP [8, 9] shows
that the Russian standards do not provide any sufficient requirements in terms of
resistance to corrosive attack, and this, to some extent, restrains the improvement of
the quality of pipe products. The world practice of oil production pays much
attention to the resistance of PCP to attack by aggressive media. Thus, PCP made
of various groups of stainless steels and corrosion-resistant alloys, namely, austenitic
stainless steels, ferritic stainless steels, and martensitic stainless steels, has found a
wide application under conditions of complicated oil extraction. A number of
Russian and foreign studies deal with an efficient choice of steel or alloy grades
and formulation of the matrix of choice for the pipe material depending on specific
conditions of oil production, namely, on temperature and the content of the CO2 and
H2S components in the well [1–5, 10–12].
The main trends in solving the problem of increasing the life of downhole pumping
equipment [13] used by oil-producing companies are (1) corrosion inhibition by
feeding a chemical reagent to the pump intake and to the hole clearance,
(2) downhole treatment by periodic pumping of a chemical reagent, (3) use of
corrosion protectors against electrochemical corrosion (sacrificial anodes), (4) intro-
duction of PCP made of high-alloy steels in the anticorrosion version, and (5) appli-
cation of special anticorrosion coatings onto the inner surface of PCP. These
measures require extra costs. Another line of increasing the life of PCP is the
application of repair technologies, service companies having been created for this
purpose. The current repair technology consists of cleaning the inner surface of pipes
from deposition and cutting out defective areas; however, the lifetime of repaired
pipes is reduced due to the retention of defects and corrosion pits intensifying the
processes of corrosion in the second stage of pipe operation. The performance,
reliability, and lifetime of pipes can be radically increased through making them
from laminated composite materials. In the world practice, these materials have
found application in many industries. Laminated composite pipes can be produced in
several principally different ways, namely, longitudinal welding of bimetal sheets,
producing pipes by simultaneous expansion, liquid diffusion welding along inter-
faces, explosion welding, centrifugal casting, and hot isostatic extrusion [14–18]. In
oil and gas industry, there is a positive experience of testing experimental-industrial
batches of laminated composite pipes in oil wells in the Netherlands, the USA,
Indonesia, New Zealand, Norway, the UK, the Philippines, Malaysia, Japan, and
21 New Engineering Solutions in the Production of Laminated Composite. . . 165
Russia. Among the known methods of manufacturing pipes from laminated com-
posite materials, lining should be distinguished. Lining consists of simultaneous
expansion of a pipe made of conventional carbon or low-alloy steel (external shell)
and a thin-wall pipe made of corrosion-resistant high-alloy steel (internal shell), the
pipes being joined due to compressive residual stresses on the interlayer boundary.
Fig. 21.2 A general view of the computer simulation of simultaneous pipe expansion on a mandrel
21 New Engineering Solutions in the Production of Laminated Composite. . . 167
Pilot models of a composite pipe were made from six pieces of worn PCP, 1700 to
1900 mm long. Before lining, on three of the six worn PCP with more than 1.9 mm
deep defects, 3 mm diameter through holes were additionally made in the midsection
(four on each). The worn PCP was lined by expansion with 47 1.5 mm electrically
welded pipes made of steel 10. The interlayer clearances and thread connections of
the fabricated pipe were sealed with a sealant. The pieces were joined on the thread
by sockets. The fabricated bimetal pipes were tested on a “Brocker” hydraulic press
in pipe-making department No 4 of the Pervouralskiy Novotrubniy Zavod JSC
(Pervouralsk New Pipe Plant). The hydraulic pressure was changed in six steps
from 14.7 to 56.9 MPa with a 60 s hold. The test results have demonstrated that the
fabricated 9500 mm long bimetal pipe withstands a test under pressure exceeding by
13% the standard value according to the Russian standard GOST R 52203–2004 for
conventionally made PCP of the D strength group. No depressurization of the
interlayer clearances and thread connections with complete preservation of the
carrying capacity along the entire bimetal pipe, including the portions with 3 mm
diameter through holes drilled in body of the PCP, is observed.
Pilot batches of bimetal PCP made by lining from worn 73 5.5 mm pipes with
1.9–3.4 mm deep defects have been tested under field conditions on four facilities
the “Tatneft” JSC. The PCP service conditions on all the facilities are estimated as
severe. Thin-wall electrically welded liner pipes were made of steel 22GYu
according to the Russian standard specifications TU 1373-021-05757850-07. The
lined PCP from the pilot batch have been operated since March 4, 2009 (for more
than 2000 days), and there is no criticism about the quality and efficiency of repaired
pipes. Regular inspection of the quality of PCP by visual examination and
hydrotesting, both involving pipe lifting from the wells, has not revealed any quality
faults. The inspected pipes continue being operated in the normal mode.
21.5 Conclusion
The creation of the process of producing lined PCP and bringing it to commercial
level is aimed at increasing the engineering and economic performance of oil
production in Russia and multiply decreasing power inputs for the maintenance
and operation of oil wells, and this corresponds to the national and global scientific
168 N. A. Bogatov et al.
and technological priorities, the state program “Energy Efficiency and Energy Sector
Development” approved by Russian Federation Government Regulation No
321 dated April 15, 2014, the priority lines of research, technology and engineering,
and the list of critical technologies approved by RF Presidential Edict No 899 dated
July 7, 2011.
The cost-effectiveness of the production and consumption of lined pump-
compressor pipes results from lower maintenance costs of pump-compressor col-
umns due to longer service life of lined PCP than that of new pipes. The restoration
of PCP by lining offers a more than four times increase in durability and a multiple
decrease in the expenditure of material and financial resources for buying new pipes
(instead of lined), repairing, and maintenance of PCP and oil wells.
The study was made within the basic part of the state job in the field of scientific
activity No. 11.9538.2017/8.9; it was supported by Act 211 of the Government of
the Russian Federation (agreement No. 02.A03.21.0006).
References
1. Ioffe, A. V., Vyboishchik, M. A., Trifonova, E. A., & Suvorov, P. V. (2010). Effect of chemical
composition and structure on the resistance of oil pipelines to carbon dioxide corrosion. Metal
Science and Heat Treatment, 52(1–2), 46–51.
2. Ioffe, A. V., Tetyueva, T. V., Vyboyshchik, M. A., Trifonova, E. A., & Lutsenko, E. S. (2010).
Tubing with high corrosion resistance. Metal Science and Heat Treatment, 52(1–2), 13–19.
3. Ioffe, A. V., Tetyueva, T. V., Vyboyshchik, M. A., Knyaz’kin, S. A., & Zyryanov, A. O.
(2010). Corrosion-mechanical fracture of tubing from carbon and alloy steels operating in
environments containing hydrogen sulfide. Metal Science and Heat Treatment, 54(9–10),
492–497.
4. Knyaz’kin, S. A., Ioffe, A. V., Vyboyshchik, M. A., & Zyryanov, A. O. (2010). Special features
of corrosion fracture of tubing operating in environments with elevated content of carbon
dioxide. Metal Science and Heat Treatment, 54(9–10), 498–503.
5. Ioffe, A. V., Tetyueva, T. V., Revyakin, V. A., Borisenkova, E. A., Knyaz’kin, S. A., &
Denisova, T. V. (2010). Corrosion-mechanical fracture of tube steel in operation. Metal Science
and Heat Treatment, 54(9–10), 512–518.
6. Smith, L. (1999). Control of corrosion in oil and gas production tubing. British Corrosion
Journal, 34(4), 247–253.
7. Brondel, D., Edwards, R., Hayman, A., Hill, D., Mehta, S., & Semerad, T. (1994). Corrosion in
the oil industry. Oilfield review, 6(2), 4–18.
8. GOST R 52203–2004. Pump-compressor tubes and couplings for them. Specifications.
9. API Specification 5CT. Specification for casing and tubing. 8th ed, July 1, 2005. ISO
11960:2004, Petroleum and natural gas industries – Steel pipes for use as casing or tubing for
wells.
10. Cerruti, S. (1998). An overview of corrosion resistant alloy steel selection and requirements for
oil and gas industry. Conference: Convegno IGF XIV Trento. 1–9.
11. Craig, B. D., & Smith, L. (2011). Corrosion Resistant Alloys (CRAs) in the oil and gas industry
– selection guidelines update. 3rd ed. US: The Nickel Institute.
12. Craig, B. D. (2011). Selection guidelines for corrosion resistant alloys in the oil and gas
industry. Materials selection for the oil and gas industry, Nickel Institute Technical, 10(73),
1–11.
21 New Engineering Solutions in the Production of Laminated Composite. . . 169
13. Safonov, V. N., & Kim, S. K. (2012). Performance of the wells of the OOO Lukoil-Komi under
corrosive conditions. Inzh. Praktika, 1, 50–59.
14. Smith, L. M. (1992). Engineering with clad steel. NiDI Series, 10(064), 20.
15. Wang, X., Li, P., & Wang, R. (2005). Study on hydro-forming technology of manufacturing
bimetallic CRA-lined pipe. International Journal of Machine Tools & Manufacture, 45,
373–378.
16. Koning, A. C., Nakasugi, H., & Ping, L. (2004). TFP and TFT back in town (Tight Fit CRA
lined Pipe and Tubing). Stainless steel world, 53–61.
17. Bogatov, N. A., Bogatov, A. A., & Salikhyanov, D. R. (2014). Corrosion-resistant lined pump
and compressor pipe. Steel in Translation, 44(11), 867–869.
18. Bogatov, N. A., Bogatov, A. A., & Salikhyanov, D. R. (2014). Use of the lining method to
restore the service characteristic of pump-compressor tubing that has exhausted its original
service life. Metallurgist, 58(11–12), 1006–1010.
19. Bogatov, N. A. Patent № 2344266 RF, IPC E21V17/01, “Method of making pump-compressor
tubing”, subm. 04.17.2007, publ. 09.20.2007.
Chapter 22
Precision of the Tubes Produced
by Drawing and Cold Rolling
G. Orlov
22.1 Introduction
Classical methods of cold deformation оf tubes are drawing through stationary dies
and cold rolling by rolls and rollers. The combination of these methods provides a
wide assortment of cold-deformed tubes and their high quality. However, classical
methods have a number of deficiencies: in the case of drawing, the probability of
formation marks is great, and cold rolling is characterized by high power consump-
tion, small productivity, complexity of adjustment of camps, and tool manufactur-
ing. Therefore, the problem of working out and research of new or combined
methods of cold deformation is topical.
There are “technological niches” that can be discovered in the course of devel-
oping new and combining the existing methods of drawing and cold rolling. One
such opportunity is the design of dies with specific technological characteristics. For
example, for the drawing of tubes – in which case the main tools are a die and a
mandrel – it is possible to construct a matrix of drawing methods Bij based on the
type of movement of the tools (see Table 22.1).
The use of a die in accordance with scheme B11 corresponds to the traditional
method of drawing metal through stationary mandrel; method B31 corresponds to
drawing on a long moving mandrel. Methods B12 and B32 entail drawing through
rotating dies. Methods of type B41 make use of a reciprocating mandrel. One of these
methods, called periodic drawing, was developed with the author’s participation and
is examined below. Methods B13 and B14, which also employ a rotating mandrel and
have the die move counter to the tube, are not known to have yet been put into
practical use and are probably regarded as technological niches. А countermoving
die can also be employed for the cold rolling of tubes. In this case, the types of
G. Orlov (*)
Institute of Materials Science and Metallurgy, Ural Federal University, Yekaterinburg, Russia
movement of the working stand and the mandrel and the number of rolls in the stand
are among the most important technical factors.
This chapter describes the use of mathematical modeling to examine several
approaches to improving the operations of drawing and cold rolling.
The finite-element method was used to study the correction of initial eccentric
thickness variation in the course of roller drawing of tubes through one and two
consistently established die cases [1] (Fig. 22.1). As cross-section thickness varies
by 70–80%, it is defined by an eccentric component; for theoretical analysis, the
initial thickness variation is presented in the following way:
Sθ ¼ Sm E cos θ, ð22:1Þ
where Sm ¼ (R0 – r0) is the medium thickness of a wall; R0 and r0, respectively, the
external and internal radiuses of cross section of stock; Е, excentricity; and θ ¼ 0 . . . 2π,
angular coordinate of the cross section (zero value corresponds to the minimum
thickness of a wall).
22 Precision of the Tubes Produced by Drawing and Cold Rolling 173
where μ is the elongation ratio factor; Kp, factor of roller pass; δ0, initial excentric
thickness variation; and ψ, friction factor by Siebel.
The optimum factor of roller pass providing minimum thickness variation is
found from Eq. (22.2) on the condition ∂δK/∂Кр ¼ 0:
νℓk
v∗ ¼ ,
μ 1 2πarccos 1 2 ℓℓ0k
1
where ν is the rate of back and forth motion mandrel; ℓk, a way of mandrel at wall
calibration (length of a calibrating corbel of die case); ℓ0, a full way of mandrel; and
μ, elongation ratio factor.
For the correction of thickness variation, Eq. (22.1), in traditional methods of cold
deformation, it is necessary to provide the nonuniform scheme оf drafts in cross
section, which can be presented in the following way:
where constants a1, a2, and θ0, accordingly, characterize the minimum and maxi-
mum speed оf draft and an orientation angle of scheme оf drafts concerning the main
axis of thickness variation, connecting the maximum and minimum thickness of
a wall.
Unknown parameters of function, Eq. (22.4), are found from the decision of
variation problems of theory of metal forming with the use of a principle of virtual
speeds. It was shown that the nonuniform scheme оf drafts can be realized by
application of the tool of not round cross section with the subsequent deformation
in round caliber. The parameters of the tool are found from dependence
The proposed method of periodic drawing has been tested at production of trial parties
of tubes in the sizes 16–20 0.25–1.0 mm from copper alloys and also molybdenic
tubes in the sizes 12–18 0.5–1.5 mm from hot-pressed stock 26 2.5 mm.
Under the dependence, Eq. (22.5), optimum parameters of mandrel profiles for
maintenance of demanded level of thickness variation δk were defined. Intermediate
profiling is realized by applying profile mandrels at the rolling by rolls and rollers [3]
and pressing and drawing of copper and brass tubes. For profiling at a pressing stage,
a trihedral pin was used and at a drawing stage, profile self-established mandrel.
At a cold rolling, advanced calibration for the reduction of longitudinal thickness
variation formed in the end of a reducing site is proposed. Mathematically the
problem of calculation of calibration is reduced to plotting of the forming caliber
D (x) and mandrel d (x) on a reducing site in length l, satisfying boundary conditions
22 Precision of the Tubes Produced by Drawing and Cold Rolling 175
Fig. 22.3 Distribution of true relative drafts for various design procedures: 1, MISiS; 2, PU tube
plant; 3, Ural tube institute; 4, according to Eq. (22.6)
and some criteria. Thus the tube section is considered ideally round, not dependent
on angular coordinate. Boundary conditions provide the necessary sizes for rolling
of the tubes: D (0) ¼ D0, D (0) ¼ d (0) + 2S0 + Δ, D (l) ¼ DТ, and d (l) ¼ DТ-2ST,
where D0, S0, DТ, and SТ are the initial and final sizes of a tube and Δ is a backlash
for input mandrel. Curves of the forming caliber and mandrel are coordinated as D
(x) ¼ d (x) + 2S (x) and D / (l) ¼ d / (l) where the stroke designates the derivative of a
function. The last condition provides a reduction in longitudinal thickness variation
of tubes.
Initial parcel for definition of function S (x) and calculation of a profile of calibers
is the assignment of a mode true relative drafts f (x) ¼ ΔSi/Si. In order to prevent the
formation of longitudinal thickness variation, this function should satisfy the condi-
tion lim f ðxÞ ¼ 0:
x!l
The analysis of known design procedures of calibrations has shown that the last
condition often is not satisfied (Fig. 22.3); therefore, it is proposed to set a mode in
the following way [4]:
x x
f ð xÞ ¼ B 1 eA l , ð22:6Þ
l
where factor A ¼1.2 and В is defined from a boundary condition.
Factor A in Eq. (22.6) varied and was found from the condition of the minimum
loading on a precalibrating site in maintaining the minimum longitudinal thickness
variation of tubes.
On the tandem cold rolling, the tube is deformed simultaneously by two succes-
sive pairs of rolls of the same size. This scheme makes it possible to increase the
draft during the rolling operation by a factor of 1.5–2. The first pair of rolls, which
176 G. Orlov
Dg2
Dp2 ¼ Dp1 ð22:7Þ
Dg1
where Dg2 and Dg1 are the pitch-circle diameters of the gear wheels on the break-
down and sizing rolls, respectively.
To a large extent, Eq. (22.7) determines the deformation conditions and the axial
forces transmitted by the second pair of rolls. Thus, it is possible to choose an
efficient ratio for rolling with substantial drafts while avoiding having the tube
become unstable due to large axial forces.
The deformation conditions in the breakdown pair of rolls differ from a conven-
tional mill, in which the metal is subjected to an additional load (tension or thrust)
from the sizing roll pair, depending on the direction of the axial forces. Thus, the
axial force in the i-th instantaneous deformation zone (IDZ) is determined with
allowance for the axial force in the corresponding IDZ of the second roll pair. The
additional axial load has a substantial effect on the stress state of the metal, its
cumulative damage content, and the degree to which any variation in the thickness of
the wall of the tube can be corrected.
The method discussed above was used on a computer to design new rolls and
analyze the rolling of steel 45 tubes on the tandem rolling mill at the pipe plant in
accordance with the new schedule 45 3 ! 25 1 mm. The draft was 5.25.
It is investigated that the compressive axial forces increase more rapidly in the
first pair of rolls. The axial forces decrease in the zone where the deformation cones
overlap due to the low rate of compression. The forces exerted by the metal on the
rolls in this zone decrease for the same reason. Because the amount of metal rolled in
the sizing roll pair during the reverse stroke of the stand is greater than in a normal
mill, the forces on the rolls are greater during this stroke than during the forward
stroke.
We subsequently studied the effect of the gear ratio of the diagonal transmission
in Eq. (22.7), which determines the nominal diameter of the pinion of the second roll
pair and the axial forces that the pair transmits. The gear ratio on the stand being
discussed is 1.057, which led to high axial forces and destabilization of the tube.
22 Precision of the Tubes Produced by Drawing and Cold Rolling 177
To lower the axial forces, it was proposed that the gear ratio be reduced to 1.032 by
decreasing the diameter of the gear wheel of the second roll pair. A larger decrease in
the gear ratio would be inadvisable, since it would increase the area of the lead zone
and increase the axial compressive force created during the forward stroke of the
stand. This situation is illustrated by the data below:
Gear ratio 1.057 1.032 1.028
Maximum axial force, kN 79.8 52.1 56.2
Maximum relative stress 0.92 0.70 0.76
The change in the gear ratio lowered the axial forces by 32–37%, which makes it
possible to avoid butt joining of the ends of tubes. The change in gear ratio had
almost no effect on the distribution of the forces on the rolls.
22.4 Summary
References
1. Orlov, G. A., & Vagapov, E. N. (2012). The use of modeling to develop a method for the roller
drawing of tubes on a long mandrel. Metallurgist, 56(1–2), 109–114.
2. Pupyshev, A. A., Akchurin, R. Z., & Orlov, G. A. (2000). Geometric and kinematic constraints of
the periodic tube drawing. Nonferrous Metallurgy (Russia), 3, 49–55.
3. Orlov, G. A., Igoshin, V. F., & Mizhiritskij, O. I. (1993). Reduction of thickness variation during
rolling on shaped mandrels of rocking and roller-type rolling mills. Steel in Translation, 10,
59–61.
4. Orlov, G. A. (2005). Estimating the quality of cold-deformed pipes. Steel in Translation, 35(3),
49–54.
Chapter 23
Features of Multihole Stamping of Platinum
Glass Melting Apparatus Elements
23.1 Introduction
The use of platinum and platinum alloys in industry is caused by its unique
properties: resistance to high-temperature corrosion and oxidation in contact with
oxide materials at high temperatures [1, 2] as well as the ability to deep drawing
[3]. One of the important fields of precious metal industry is the application of
platinum alloys in manufacturing of the equipment for the production of glass and
basalt fiber [4, 5]. The glass melting apparatus (bushing) is a container for placement
of melted glass or basalt, made of heat-resistant platinum-rhodium alloy [6]. The
essential component element of platinum bushing is bottom tip-plate with tips.
There are two basic technologies for bushing bottom tip-plate manufacturing: the
first one is multistage stamping of tips from a solid plate, and the second one is
welding of tips into a plate with prepunched holes. Stamped tip-plate has a number
of advantages, including the possibility of a more close packing of tips on tip-plate.
However, the close packing of tips creates some technological challenges, one of
which is maintaining the symmetrical execution of each tip to ensure the required
dimensional accuracy.
Y. Loginov (*)
Institute of Materials Science and Metallurgy, Ural Federal University, Yekaterinburg, Russia
e-mail: j.n.loginov@urfu.ru
A. Fomin
EZOCM-ENGINEERING, Verhnjaja Pyshma city, Sverdlovsk Region, Russia
Fig. 23.1 Side view (a) and top view (b) of platinum strip with stamped tips after the first stage of
stamping
(Fig. 23.2) formed by the recess and flows through the path of least resistance
because of the minimum energy law, i.e., in the hole of the stamp second semi-row.
As a result, the second semi-row has a larger volume. It happens that the volume of the
first semi-row is not even enough for the formation of a high-grade tip at the second
stamping stage.
All the above-described defects occur due to the step-by-step processing tech-
nology, resulting in intensive longitudinal metal flow. One approach to improve this
technology could be an increase in the contact area of the upper die to the maximum
value, based on the capabilities of stamping equipment, and the second approach is
stamping step selection, which must be a multiple equal to the distance between two
adjacent rows of projections.
the model of the workpiece. Since the workpiece has a simple geometric shape, the
automatic determination of the finite element mesh was used. As a result, the mesh,
consisting of 85,160 tetrahedra uniformly distributed throughout the volume of the
workpiece, was obtained. The next step was to set the properties of the workpiece
material. In glass fiber industry platinum-rhodium alloys are the most commonly
used; their chemical composition is constantly improving [11, 12]. In this work, as
the material alloy PtRh-10DS (dispersion-strengthened) was selected, whose
Young’s modulus is equal to 193,100 MPa and the Poisson’s ratio is 0.34. Plastic
properties were set by an equation, which establishes a relationship between the
deformation resistance and the plastic strain. This equation is
σ S ¼ a þ b ∙ εC , ð23:1Þ
where σ S is the deformation resistance, ε is the plastic strain, a is the yield strength,
b is the hardening coefficient, and c is the hardening exponent. The applicability of
such law for platinum hardening was shown in Refs. [13, 14]. For alloy PtRh-10DS,
these values were adopted as follows: a ¼ 241 MPa, b ¼ 444 MPa, and c ¼ 0.419. At
the boundary of contact of the workpiece with the stamp the implemented friction is
described by the friction law of Siebel with the friction coefficient of 0.12 [15]. The
displacement of the upper die is defined by the direction and a constant speed of
50 mm/s. The calculation lasted until the ultimate distance between the dies is equal
to a specified tip-plate thickness.
The results of the first simulation are shown in Fig. 23.3a. In the figure, the blue
triangles denote the finite element mesh of the workpiece; the green triangles depict
the body of the stamp. It is seen that the results correspond to reality: the first semi-
row of projections has a small volume and a sloping site, and the second semi-row
has a large amount of metal and a conical form.
Fig. 23.3 Modeling of the workpiece deformation with a different processing step: (a) 3 mm, (b)
12 mm
23 Features of Multihole Stamping of Platinum Glass Melting Apparatus Elements 183
The results of the second simulation are shown in Fig. 23.3b. You may notice a
significant change in the pattern of filling the stamp: the first semi-row of projections
began to have a more correct conical shape; its volume has increased about two
times. In addition, the sloping site appeared on the second semi-row, but at this
height of the projections, this site is not a defect, because the workpiece is subjected
to further machining.
Therefore, this approach of increasing the stamping step resulted in a more
symmetrical and more complete filling of the holes of the lower die. However, it
also implies the increase of the stamping force by several times.
Another approach of improving the uniformity of filling of the stamp is the
selection of the geometric parameters of the upper die. The first stamping stage is
carried out by the upper die, which is shown in Fig. 23.4 and has a flat base and a
rounded edge.
The flat base is the deforming element; the rounded edge creates a longitudinal
flow of metal. The influence of a rounding radius value on the filling of the stamp
was investigated using finite element modeling. Three radiuses – 2 mm, 10 mm, and
20 mm – were selected; the simulation results are shown in Fig. 23.5a, b, c,
respectively.
The projections on the workpiece bottom at a rounding radius of 2 mm and
10 mm have a more regular shape than at a radius of 20 mm. The largest volume of
projections is observed at a radius of 2 mm and 20 mm.
Therefore, the radius of 2 mm is the best of the three selected options. The
comparison of the simulation results shows that the smaller the rounding radius of
the upper die, the more correct form and a greater volume of the projections have the
first and the second projections semi-row. This pattern of the stamp filling shows the
positive effect of reducing longitudinal metal flow due to the decrease of the upper
die rounding radius. However, the complete absence of rounding leads to the
appearance of stripes of metal on the top of the workpiece, so this solution is not
rational.
Thus, the computer simulation shows that a change in stamping process param-
eters can lead to a more symmetrical filling of the stamp.
184 Y. Loginov and A. Fomin
Fig. 23.5 Modeling of the workpiece deformation with different rounding radius: (a) 2 mm, (b)
10 mm, (c) 20 mm
23.5 Conclusion
References
1. McGrath, R. B., & Badcock, G. C. (1987). New dispersion strengthened platinum alloy.
Platinum Metals Review, 31(I), 8–11.
2. ElGuindy, M. (2001). Platinum group metals: Alloying, properties, and applications. Encyclo-
pedia of Materials: Science and Technology (2nd ed, pp. 7117–7121).
3. Kameneckij, B. I., Grohovskaja, L. G., Loginov, Ju. N., & Studenok, G. I. (2007). Issledovanie
deformacionnyh harakteristik i struktury splava PlN-4,5 pri izgotovlenii polyh zagotovok
metodom glubokoj vytjazhki. Cvetnye metally. № 8. С. 51–53.
4. Rytvin, Y. (1999). I. Platinum metals and silicates: from the 20th century to the 21st century.
Proceedings of International Symposium on Precious Metals. Kunming, Yunnan Science and
Technology Press (pp. 89–98).
23 Features of Multihole Stamping of Platinum Glass Melting Apparatus Elements 185
5. Koch, M., & Lupton, D. (2006). Design and manufacture of bushings for glass fibre production.
HVG-Colloquium Production, Properties and Applications of Refractory and Platinum Group
Metals for the Contact with Glass Melts. Düsseldorf (pp. 1–12).
6. Burcev, V. M., Tykochinskij, D. S., & Rudenko, A. E. (1977). Perspektivy sozdanija
cel’noshtampovannyh fil’ernyh uzlov stekloplavil’nyh apparatov, Platinovye splavy dlja
stekloplavil’nyh apparatov. M.: VNIISPV. S. 96–101.
7. Loginov, Ju. N., & Fomin, A. A. (2013). Kinematicheskie uslovija vydavlivanija
plasticheskogo sloja v mnogorjadnom shhelevom shtampe. Kuznechno-shtampovochnoe
proizvodstvo. Obrabotka materialov davleniem. № 4. S. 14–17.
8. Loginov Ju. N., Kameneckij, B. I., & Studenok, G. I. (2006). Modelirovanie deformirovannogo
sostojanija krugloj plastiny pri vytjazhke. Izvestija vysshih uchebnyh zavedenij. Chernaja
metallurgija. № 3. S. 26–28.
9. Völkl R., Fischer, B. Teschner R., & Lupton, D. F. (2001). Finite element modelling of strains
and stresses in platinum alloy bushings for textile glass fibre production, Glass Science and
Technology, Glastechnische Berichte. Vol. 74 (pp. 142–151).
10. Ershov, A. A., & Loginov, Y. N. (2013). Using the INVERSE solver “INVERSE” in
PAM-STAMP 2G. Metallurgist, 57(5–6), 372–377. https://doi.org/10.1007/s11015-013-
9741-x.
11. Darling, A. S., Selman, G. L., & Bourne, A. A. (1968). Dispersion strengthened platinum.
Improved high temperature creep properties. Platinum Metals Review, 12(l), 7–13.
12. Rdzawski, Z. M., & Stobrawa, J. P. (2004). Microstructure and properties of the new Pt–Rh
based alloys for high-temperature applications. Journal of Materials Processing Technology,
153–154, 681–687. http://dx.doi.org/10.1016/j.jmatprotec.2004.04.130.
13. Loginov, Y. N., Yermakov, A. V., Grohovskaya, L. G., & Studenok, G. I. (2007). Annealing
characteristics and strain resistance of 99.93 wt.% platinum. Platinum Metals Review, 51(4),
178–184. https://doi.org/10.1595/147106707X237708.
14. Loginov, Ju. N., Ermakov, A. V., Grohovskaja, L. G., & Studenok, G. I. (2006). Uslovija
razuprochnenija i soprotivlenie deformacii platiny. Cvetnye metally. №6. S. 85–88.
15. Loginov, Ju. N., & Studenok, G. I. (2010). Izuchenie trenija pri listovoj prokatke platiny i ee
splavov. Proizvodstvo prokata. № 7. S. 14–16.
Chapter 24
Research on Metal Forming in the Pipe
Ends Upsetting Process
24.1 Introduction
The pipe ends upsetting process is a local increase of the wall thickness of a pipe at a
predetermined distance from the end. Wall thickening is performed outside, inside,
or on both sides using dies and a punch [1, 2]. Greater wall thickness increases the
strength of threaded tubing connection with couplings or welded drill pipe connec-
tion with tool joints. Couplings and tool joints are used to build the pipes in the
columns in the extraction of oil and gas.
Now, the conditions of oil and gas production are becoming more complex. This
fact sets high demands on the quality of pipes produced. The practice of the
production of pipes with upset ends shows that most defects are located on the
inner surface. These imperfections are surface cracks that reduce the effectiveness of
pipe wall thickness [3]. They are detected close to the ends of pipes (Fig. 24.1).
Mueller and Arem [4] show that the cause of such defects is a lower temperature at
the end of the pipe relative to the rest of the heated part of the pipe. However, the
temperature increase directly at the pipe end does not improve the product quality.
Therefore, the purpose of this paper is to research metal forming in pipe ends
upsetting process and to identify the defects associated with the process. The study
involved finite element simulation and a full-scale experiment. The influence of the
diameter and wall thickness of the pipe, friction conditions, the temperature of the
heated pipe ends, and the temperature of the tool on the defect creation has been
studied.
To achieve the goals of this research, finite element simulation in DEFORM-3D was
performed. The simulation was carried out for the tubing with nominal dimensions
73.02 5.51 mm in accordance with API specification 5CT [5] and for drill pipe
73 9.2 mm in accordance with State Standard R 50278-92 [6]. The actual
dimensions 73.23 4.82 mm and 73.81 6.2 mm of the tubing were also used
taking into account the tolerances on the diameter and the wall thickness of the pipe.
The steel grade AISI-1045 was selected in DEFORM-3D’s material library. These
blanks were used for computer simulation with the following upsetting process
variables: Siebel friction factor is ψ ¼ 0.3; the length and the temperature of heated
pipe ends are Lh ¼ 240 mm and T ¼ 1200 С, respectively; and the dies and the
punch temperature are Td ¼ 200 С and Tp ¼ 130 С, respectively. The tool calibra-
tion and abovenamed variables of the process are used taking into account the
practice of production of pipes with upset ends at JSC Pervouralsk new pipe plant.
In addition, the problems with Siebel friction factor ψ ¼ 0.7, the temperature of
heated pipe ends T ¼ 1100 С, the temperature of dies Td ¼ 150 С and Td ¼ 250 С,
and the temperature of punch Tp ¼ 100 С and Tp ¼ 180 С, respectively, had been
solved. Moreover, the temperature in a narrow region near the face plane of the pipe
was taken equal to Tf.p. ¼ 1100 С and Tf.p. ¼ 1290 С. All additional problems were
solved for the tubing 73.02 5.51 mm upsetting. In total, ten problems had been at
the stage of computer simulation.
According to the simulation results, there are several stages of the upsetting process
which are characterized by different velocity fields [7]. The first stage is the expansion
of the pipe by the punch. It begins from the time when the punch first contacts the
blank and continues until the punch contacts the face plane of the pipe. The velocity
values at this stage are close to zero. In this regard, there is no attention paid to this
stage in the literature. However, it is characterized by intensive heat removal from the
heated end of the pipe through the contact surface into a cold tool (Fig. 24.2).
The value of cooling defines the change of metal thermomechanical behavior of
blank in the region close to the pipe end. The maximum temperature change near the
contact surface of the blank is as follows:
24 Research on Metal Forming in the Pipe Ends Upsetting Process 189
1300
T0
1200
Temperature, °С
1100 T1max
1000
900 T1min
800
0 50 100 150 200 250
The distance from the conical section of the dies to the face
plane of the pipe, mm
Temperature field in the initial time Temperature field at the end of pipe expansion
Fig. 24.2 The change of the temperature field of the blank during the first stage of the process
L2 15
50 L3
40 5
1 2 3
30
r1 r2 r3 -5
20
-15
10
0 -25
-80 -60 -40 -20 0 20
Axial coordinate of pipe profile, mm
The pipe profile in longitudinal section Radial velocity field
T 1max T 1min
K temp ¼ : ð24:1Þ
T 1max
In the process of the pipe profile buckling, there is a gradual increase in the outer
and inner diameter of the end of the blank, whereby the contact of a pipe with a
punch disappears. Then the filling of the die space starts. The die space is formed
with the working surfaces of the die and the punch. The filling of the dies occurs in
the direction from the periphery to the axis of the blank [8–10]. Due to the
nonuniform deformation resistance of the metal in the axial direction specified by
cooling in the initial stage of upsetting, the field of radial velocity is characterized by
two local maxima (Fig. 24.3).
A closed air cavity appears between the surfaces of the punch and the inner
surface of the pipe. It appears in the case when the metal particles that are on the
boundary of domains 1–2 have a larger radial velocity than the particles on the
boundary 2–3 (Fig. 24.3). This cavity does not disappear completely to the end of the
upsetting process. To quantify the tendency to the creation of defects on the inner
surface of the upset pipe, the following parameter is introduced:
190 M. Erpalov and A. Bogatov
0.05 Tf.p.=1100°C
0
0.12 0.17 0.22 0.27
-0.05 KTemp
ψ=0.7
-0.1
Kdef
-0.15
-0.2
-0.25
-0.3
r2 r1
K def ¼ : ð24:2Þ
L2
Figure 24.4 shows the dependence of a parameter Kdef on the parameter Ktemp that
characterizes nonuniform deformation resistance along the axis of the blank at the
end of the stage of expansion of the pipe. It is seen from the diagram that the inner
defect is formed when the temperature of metal in face plane of the pipe is
Т ¼ 1100 С. It is 100 C below the temperature of the rest heated part of the pipe.
The decrease of tool temperature leads to a slightly increased probability of defect
formation. The increase of friction factor leads to the increased nonuniformity of
deformation in the pipe end [10], which in turn increases the tendency to the creation
of defects. This can be seen from Fig. 24.4 where the value Кdef is very different from
the trend line.
The increase of the tool temperature, as well as the increase of temperature in the
face plane of the pipe, provides more uniform temperature on the end of the blank. It
should be said that the values of parameter Ktemp are equal when two pipe ends are
heated, one to the temperature 1200 С and the other one to 1100 С. This leads to the
fact that the tendency to the creation of defects is equal in both cases.
An additional parameter is added in the process:
2S
K tool ¼ , ð24:3Þ
Dd d p
0.1
73.81x6.2
0.05
0
0.5 0.55 0.6 0.65 0.7 0.75
-0.05 Ktool
Kdef
-0.1
73.23x4.82 73.02x5.51
-0.15
-0.2
-0.25
-0.3 73x9.2
73.81 6.2 mm. Statistical analysis of experimental data suggests that there is a
linear relation between the parameters Kdef and Ktool regardless of the pipe dimen-
sions and the tool sizes. Thus, correlation has been found between the tool calibra-
tion parameters with the maximum wall thickness at which the defect creation is
observed:
Smax ¼ 0:3486 Dd dp : ð24:4Þ
In order to confirm the results obtained at the stage of computer simulation, a full-
scale experiment was carried out at JSC Pervouralsk new pipe plant on a hydraulic
press made by SMS Meer. Blanks with nominal dimensions 73.02 5.51 mm were
chosen for this experiment. According to the existing tool calibration at JSC
Pervouralsk new pipe plant and Eq. (24.4), the maximum wall thickness is
5.96 mm. Blanks with average wall thickness of 5.28 mm, 5.36 mm, 5.49 mm,
5.79 mm, 5.87 mm, and 6.09 mm were chosen for the full-scale experiment. The
upsetting was made with the following process variables: the length of the heating
pipe ends is Lh ¼ 310 mm, the heating temperature is T ¼ 1290 C, the pressure in
the main hydro cylinder is 10 MPa, and the time delay before moving the punch in
the opposite direction is 1.8 s.
The full-scale experiment has shown that the pipe in which inner defects were
found had wall thickness equal to 6.09 mm. This exceeded the maximum wall
thickness equal to 5.96 mm. Inner defects were not found for the rest pipes with
wall thickness of less than 5.96 mm. Thus, the full-scale experiment has confirmed
the validity of Eq. (24.4) that allows one to establish requirements either to blanks
coming on the upsetting operation or to the tool calibration based on the actual
dimensions of the pipes.
192 M. Erpalov and A. Bogatov
Figure 24.5 shows that the parameter Kdef is minimal in the case of upsetting the
drill pipe with dimensions 73 9.2 mm in accordance with State Standard R 50278-
92 in one pass. This is caused by the fact that the step of expansion of pipe is omitted.
So, inner defects near the pipe end are excluded. The experimental work was carried
out at JSC Pervouralsk new pipe plant to test the possibility of upsetting drill pipes
73 9.2 mm in an industrial environment in one step. As a result, 30 tons of the
pipes were successfully treated, and the coefficient of the pipe wall thickening was
about 1.85. Defects on the inner surface of the pipes were not detected (Fig. 24.6).
This confirmed the results of theoretical research.
24.4 Conclusion
This paper presents theoretical and industrial research on the pipe ends upsetting
process in one step. Special attention is paid to the quality of the inner surface of final
products. The results of computer simulation show the influence of technological
parameters of upsetting on defect formation on the inner surface of the pipe. Full-
scale experiment confirms the correlation between maximum wall thickness and the
calibration parameters of the process tool (Eq. 24.4).
This research was performed as part of the project of the state task in the field of
scientific activity №11.1369.2014/K from 07/18/2014 (state registration number:
114122470051).
References
1. Development of OCTG ends upsetting technology at JSC “PNTZ”, Erpalov M. V. et al. OMD
2014. Fundamental problems. Innovative Materials and Technologies: Proceedings of the
International Scientific and Engineering Congress. Moscow: Publishing House “Bely veter”.
2014 (Vol. 2, pp. 64–69).
2. Erpalov, M. V., Kondratieva, G. N., & Bogatov, A. A. (2014). Upgrading pipe ends upsetting
technology in order to increase the efficiency of hydraulic presses SMS Meer, Innovative
24 Research on Metal Forming in the Pipe Ends Upsetting Process 193
25.1 Introduction
Nuclear power development and the necessity to increase the durability of power
units up to 80–100 years [1] have established higher expectations for reliability and
safe operation of A-plant equipment.
The steam generator (SG) is one of the most important as well as the most
damageable elements of an A-plant. The main factor affecting the durability of a
steam generator (SG) is the condition of heat exchanger pipes. The process of
deterioration of heat exchanger pipes in steam generators has been thoroughly studied
by the “Prometey” construction materials scientific institution and “Hydropress”
construction bureau (Podolsk). One of the most common types of damage related
to heat exchanger tubes in SG is corrosion cracking – the deterioration of metal
caused by tensile and residual stresses along with the impact of corrosive environ-
ment [2–5].
The techniques for manufacturing heat exchanger tubes are represented by the
following operations: boring and grinding hot-rolled pipe billet, cold rolling middle-
sized pipes, heating, rolling final-sized pipes, straightening, electropolishing the
inner surface and burnishing the outer surface, and checking the final quality.
Correcting and burnishing operational modes predetermine residual stresses in
pipes during the delivery process.
This chapter presents the results of the research on residual stresses and surface
peening formed during finishing operations (correcting and burnishing).
Fig. 25.2 Ring samples for defining residual stresses: a – ring with increased diameter (tension
residual stresses); b – ring with reduced diameter (compression residual stresses)
Cutting the ring in generatrix and measuring the changes of its diameter make it
possible to calculate residual stresses (Eq. (25.1)):
σ res ¼ E Δd S
ð25:1Þ
D,
σ res – residual stresses value, mPa
E – Young modulus of elasticity that equals to 205,000 mPa
Δd ¼ (d1 d0) – the difference of diameters measured before and after cutting, mm
S – the pipe wall thickness, mm
D ¼ ðDout SÞ – the average diameter, mm
Dout – the outer diameter before cutting, mm
This method is based on the assumption that the distribution of circumferential
residual stresses throughout the pipe wall section is linear. It gives an opportunity to
estimate the influence of technological operations on residual stresses.
Surface peening. Solidity measurement was done using FISCHERSCOPE
HM2000 XYm series measurement device that is designed for microhardness
estimation. This method conforms to ISO 14577, DIN 50359, DIN EN ISO
14577–1, DIN55676, and VDI/VDE Guidance N 2616 standards that define the
requirements for tools operating by an indentation principle to determine solidity
value and other parameters. Measuring and analyzing procedures are automated in
the system.
The standardized interpretation of the term hardness is defined in ISO 14577
standard approved in 2002. This standard was elaborated by ISO/TC 164 technical
committee (metal mechanic testings) and SC3 subcommittee (solidity testing).
According to this document, solidity is defined as resistance of a material-to-gradual
penetration of harder matter of another kind.
Scratch hardness value HM is calculated through indenter submergence depth
and load value in accordance with the following international standards: ISO 14577
(ISO 14577 Metallic Materials. Instrumented indentation test for hardness and
materials parameters. Part 1: Test method) and DIN 50239 (DIN 50359–1 Testing
metallic materials – Universal Hardness test Part 1: Test method).
The recording of the indentation diagram and preliminary results processing were
done with the use of WIN-HCU device software. The following indenters were used
for the measurements: Vickers four-sided diamond pyramid nanoindenter with
restricted requirements to geometry (VV064 DKD-K-06301 № 430,009,544 certif-
icate dated 11/23/2005) and Berkovich three-sided diamond pyramid indenter
(VB034 DKD-K-06301 # 430009940 dated 3/16/2006). Hardness calibration of
the devices was done according to BK7 material.
ISO 14577 standard regulates using two main hardness indicators: scratch hard-
ness HM and indention hardness Hit. With the continuous increase of indention force
(F), the corresponding depth value (h) of the indenter penetration is determined.
Taking into consideration the geometrical correlation between penetration depth
and the shape of the indenter, the measuring system generates the universal hardness
198 M. Rozenbaum et al.
index value according to HM ¼ F/A physics formula in which А ¼ f(h) stands for
imprint surface area. Scratch hardness denotes metal hardness value.
Table 25.1 demonstrates the average values of samples diameters before and after
cutting as well as the calculations of tangential residual stresses. Calculation results
with + mark apply to stretching residual stresses, and results with – mark apply to
shrinking residual stresses. The data from Table 25.1 are illustrated in Figs. 25.3 and
25.4.
25 The Study of Residual Stresses and Surface Hardening in Tubes After. . . 199
140
130
120
110
100
90
Residual stresses, σres mPa
80
70
60
50
40
30
20
10
0
-10
-20
-30
-40
1 2 3 4
Technological operations
Figures 25.3 and 25.4 show the changes of the average values of residual stresses
depending on finishing operations. As can be seen in Figs. 25.3 and 25.4, residual
stresses shrink after heat treatment and equal to 29.53 MPa. Further pipe correction
creates stretching residual stresses equal to 40.72 MPa. Sand jetting the surface of
inner pipes additionally increases stretching residual stresses up to 129.28 MPa;
subsequent grinding reduces the tensile residual stresses down to 78.29 MPa.
Excluding sandblasting (Fig. 25.4) causes residual stresses to retain the same level
of tension as after straightening.
The data from Table 25.1 are illustrated in the boxplot (Fig. 25.5), consisting of
the following elements:
– Central quartiles (shown as rectangles that point out that 50% of data are located
within required interval, 25% are above those limits, and the remaining 25% are
below those limits)
– Distribution range (intervals limited by lines coming out of smaller rectangle
sides show distribution interval for the majority of values or all of them)
– Median/average value (line inside a rectangle)
This diagram enables an analysis considering the wide range of experimental
data. The analysis of the diagram supports the results presented in Figs. 25.3 and
25.4.
200 M. Rozenbaum et al.
55
45
35
Residual stresses, σres mPa
25
15
-5
-15
-25
-35
1 2 4
Technological operations
Pipe samples after heating, correcting, and polishing were taken for this test. Each
sample underwent ten measurements. The measurements were made with 200 g
loading and indenter penetration depth of 7 μm. The results of the measurements
were obtained as files containing electronic tables. A file formed by the measuring
system contains 6400 records. These files contain information about the experimen-
tal results: indentation pressure, the depth of indenter penetration into a sample,
pressure time, and scratch hardness values.
Then the test data are adjusted for automatic analysis in open package statistics
programming R designed for statistics data processing and graphic operations [7]. In
order to analyze large amount of data, a special script program was installed in R
programming package. The program provided the dependencies of hardness on the
depth of indenter penetration for the pipe samples under study as well as generated
the averaged dependencies of hardness on the indentation depth based on the
experiments. The results of processing the experimental data are graphically
shown in Fig. 25.6.
Figure 25.6 shows the averaged dependencies of hardness on indentation depth of
ten measurements after heating, correcting, and burnishing.
25 The Study of Residual Stresses and Surface Hardening in Tubes After. . . 201
200
180
160
140
120
Residual stresses, σres mPa
100
80
60
40
20
-20
-40
-60
1 2 3 4 5
Technological operations
25.4 Conclusion
1. Shrinking residual stresses occur in pipes after heat treatment. Correcting oper-
ations form stretching residual stresses. Further sand jetting further increases the
level of stretching residual stresses. Burnishing decreases stretching residual
stresses but not significantly.
2. According to the research results, finishing operations of correcting and burnish-
ing in pipe production cause surface peening. The lowest hardness level is found
after heating the pipe samples. Further correction and burnishing additionally
increase hardness.
3. Applied finishing operations during production create residual stresses and sur-
face peening that decrease pipe resistance to corrosion during exploitation. The
objective of improving the quality of heat exchanger pipes can be reached by
202 M. Rozenbaum et al.
Fig. 25.6 Hardness dependencies on indentation depth after finishing operations: 1 – pipe heat
treatment; 2 – correcting; 3 – correcting and grinding
creating new production techniques that will prevent the formation of stretching
residual stresses and increase pipe surface peening during the delivery process.
4. The ring tangential method can determine the integrated tangential residual
stresses. An interesting development of the work may be the determination of
the residual stresses on the surface of the tubes [8, 9].
Acknowledgements The study was made within the grant of Russian Science Foundation
(№ 17-79-100072), supported by Act 211 of the Government of the Russian Federation (agreement
No. 02.A03.21.0006).
References
1. Karzov, G. P. (2013). VVER reactors lifetime increase based on construction materials improve-
ment. Proceedings of the 8th International Conference on A-plants with VVER safety securing
Podolsk.
2. Karzov, G. P. Dynamics evaluation of VVER-1000 steam generator heat exchanger pipes
damages appearance and development in operating conditions. www.gidropress.podolsk.ru/
files/proceedings/seminar7/documents/f63.pdf
25 The Study of Residual Stresses and Surface Hardening in Tubes After. . . 203
3. Karzov, G. P. Main heat exchanger pipes damages mechanisms in various exploitation stages of
VVER-1000 type steam generators. Proceedings of the 7th International Conference on Hori-
zontal Steam Generators. www.gidropress.podolsk.ru/files/proceedings/seminar7/seminar7/
authors.htm
4. Neklyudov, I. M. (2003). Researching reasons of corrosion defects appearance in VVER-1000
steam generators heat exchanger pipes. Proceedings of the International Conference on South-
Ukraine A-plant steam generators heat exchanger pipes corrosion damages: Sevastopol
National University of Atomic Energy and Industry collected articles (pp. 5–63).
5. Fedoseev, A. E., Bulanova, T. M., & Goncharenko, U.D. (1998). Researching Balakovo
A-plant’s second part steam generator heat exchanger pipes damages origin, New technologies
for energetics, industry, and construction: collected articles (pp. 38–42). Dimitrovgrad: Atomic
reactors scientific research Institute.
6. Birger, I. A. (1963). Residual stresses, M.: State scientific and technical publishing office
engineering literature.
7. R Development Core Team. (2011). R: A language and environment for statistical computing. R
Foundation for Statistical Computing, Vienna, Austria. ISBN 3–900051–07-0. www.R-project.org
8. Shimov, G. V. (2018). Rapid Method for Determining the Distribution of Residual Stresses in
Pipes. Materials Performance and Characterization 7(4):20170050
9. Shimov, G., Rozenbaum, M., Serebryakov, A., & Serebryakov, A. (2016). The method for
measuring residual stress in stainless steel pipes. Journal of Chemical Technology and Metal-
lurgy, 51(4), 458–464.
Author Index
K
D Kapustin, F., 99–106
Deryabin, V., 123–130, 139–146 Karimov, K.A., 51–57
Domanskaya, I., 149–153 Kayumov, A., 89–95
Khotinov, V.A., 27–31
Kolmachikhin, B.V., 59–71
E Kolmachikhina, E., 43–48, 81–88
Erpalov, M., 187–192
L
F Lazutkina, O., 131–137
Farafontova, E., 123–130 Levina, A., 19–25
Farber, V.M., 27–31 Loginov, Yu., 179–184
Finkelstein, A., 33–39
Fomin, A., 179–184
Fomina, I., 117–122 M
Maltsev, V., 89–95
Mal’tseva, L., 19–25
G Mal’tseva, T., 19–25
Gerasimov, S., 11–16 Mamyachenkov, S., 73–79
Gerasimova, E., 107–114 Mamyachenkov, S.V., 43–48
N S
Naboychenko, S., 81–88 Salikhyanov, D.R., 163–168
Naboychenko, S.S., 51–57 Serebryakov, An., 195–202
Neustroev, V.I., 51–57 Sharapova, V., 19–25
Novgorodov, D., 3–9 Shimov, G., 195–202
Spiridonova, A., 99–106
O
Oleinik, V., 99–106, 117–122, 149–153 T
Orlov, G., 171–177 Tochilova, I., 139–146
P V
Paramonova, O., 123–130 Viduetsky, M., 89–95
Pavlov, D.A., 157–162 Voroshilova, I., 131–137
Pavlova, E.A., 157–162
Pomazkin, E., 99–106
Z
Zhukov, V.P., 59–64
R
Riehemann, W., 33–39
Rogozhnikov, D., 73–79
Subject Index
A phase ratio, 19
Activation energy, pearlitic and bainitic quenching temperature, on physical
transformations properties, 21
Avrami equation, 4, 5 structure formation, 22
calculation, 5 tensile strength and yield strength, 24
existing TTT diagrams, 5 03Kh14N10K5M2Yu2T steel’s structure, 19
on supercooling, 9 Avrami equation, 4, 5, 7, 9
temperature dependence, 4
time dependence, transformation
development, 5 B
Adhesion layer, 131, 136 Bimetal pipes, 167
Alloy behavior, Cu-Ni-Zn, 11 Biogenic ash, 150
Alloying addition, Sn and Al, 11, 12, 16 Blowing intensity, 59–64
Aluminum porous filters, 34 Brown coal, 150, 153
API X80 steel
Charpy fracture surface zones, 31
Charpy impact tests, 28 C
fracture zone length, 29 Capillary-crystallization strength
impact load vs. displacement curves, 28, 29 contact duration, 141–144
low-carbon (LC), 28 slip types, 139
Arsenic solid particle adhesion, 140
cakes, OSF matte, 55 typical stages, 140
extraction, 52, 53, 55 Capillary-dispersed system, 140
removal, 54, 55 Capillary force, 129
Austenite, undercooled, see Undercooled Casting bar structure, nickel silver, 15
austenite Casting mold fillability, Cu-15Ni-20Zn alloy,
Austenitic-ferritic steel 11–14, 16
chemical composition, 19 Cement-fly ash-limestone meal, 111
DSC heating curves, 23, 24 Cement-fly ash-marble powder, 111
hardness, 19 Cement-marble powder-limestone meal, 111
heat treatment, 21 Cement stone
hot metal forming, 24 Cа aluminate and alum ferrite hydration, 101
laser surface treatment, 21 composition, 103
microstructure, 20, 22 crystalline hydrates, 101
W Z
Water-glycerol emulsion, 60 Zinc, 44
Waterproofing penetrative capillary mixture extraction of, 44
(WPCM), 99–101, 104, 106 response surfaces, 45
temperature and pulp density, 46
X
X80 grade steels, 27
X-ray diffraction (XRD), 99