Belkin 2016

SCT-21269; No of Pages 25
Surface & Coatings Technology xxx (2016) xxxxxx
Contents lists available at ScienceDirect
Surface & Coatings Technology
journal homepage: www.elsevier.com/locate/surfcoat
Plasma electrolytic saturation of steels with nitrogen and carbon

P.N. Belkin a, A. Yerokhin b,, S.A. Kusmanov a
a
Nekrasov Kostroma State University, Kostroma 156961, Russia
b
University of Manchester, UK
a r t i c l e i n f o a b s t r a c t
Article history: Studies in plasma electrolysis resulted in development of various surface treatments for metal components.
Received 5 April 2016 These treatments include formation of protective ceramic layers on some metals (e.g. oxide coatings), saturation
Revised 7 June 2016 of metal surfaces with interstitial elements (e.g. nitrogen, carbon and boron), and plasma electrolytic deposition
Accepted in revised form 10 June 2016
of extrinsic compounds, heat-treatments (e.g. hardening and annealing), surface cleaning and polishing. The
Available online xxxx
main advantages of plasma electrolytic treatments are high processing speeds and low costs. The treatments en-
Keywords:
able production of surface nanostructures and local area processing. This review examines recent results in plas-
Plasma electrolysis ma electrolytic carburising, nitriding, and nitrocarburising (as the most common diffusion-based treatments),
Carburising including treatment modes, electrolyte compositions, structures, and properties of hardened materials. Analysis
Nitriding of the results obtained up to date indicates that pulse plasma electrolytic saturation treatments leading formation
Nitrocarburising of surface nano-structures appear to be the most promising to advance further this type of electrolytic plasma
Phase composition technology. Moreover, electrolytic plasma treatments provide considerable research interest in terms of funda-
Protective properties mental science, in particular for development of models of heat transfer on at vertically or horizontally oriented
surfaces and electrochemical processes occurring in the studied systems. These processes include stages of liber-
ation of saturating components, adsorption of active atoms, and their diffusion into the metal surface; therefore
understanding associated kinetics and limiting factors is important for gaining proper control over these surface
treatments.
2016 Elsevier B.V. All rights reserved.
Contents
1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0
2. Physical features of plasma electrolysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0
2.1. Formation of the vapour-gaseous envelope . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0
2.2. Heat exchange in the system electrolyteVGEmetal electrode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0
3. Electrochemical features of plasma electrolysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0
4. Carburising . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0
4.1. Electrolyte compositions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0
4.2. Structure and phase composition of carburised layers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0
4.2.1. Low-carbon steels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0
4.2.2. Low-alloy steel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0
4.3. Carbon diffusion kinetics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0
4.3.2. High-carbon steel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0
4.4. Tribological behaviour . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0
4.4.1. Low-carbon steel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0
4.4.2. Medium carbon steel. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0
4.4.3. High-carbon steel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0
4.4.4. Low-alloy steel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0
Corresponding author at: School of Materials, University of Manchester, Oxford Road, Manchester M13 9PL, UK.
E-mail address: Aleksey.Yerokhin@manchester.ac.uk (A. Yerokhin).
http://dx.doi.org/10.1016/j.surfcoat.2016.06.027
0257-8972/ 2016 Elsevier B.V. All rights reserved.
Please cite this article as: P.N. Belkin, et al., Surf. Coat. Technol. (2016), http://dx.doi.org/10.1016/j.surfcoat.2016.06.027
2 P.N. Belkin et al. / Surface & Coatings Technology xxx (2016) xxxxxx
4.5. Corrosion resistance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0

4.5.2. Medium carbon steels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0
4.5.3. Stainless steels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0
5. Nitriding and nitrohardening . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0
5.2. Structure and phase composition of nitrided layers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0
5.2.1. Medium carbon steel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0
5.2.2. Stainless steel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0
5.3. Nitrogen diffusion kinetics. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0
5.4. Hardness, strength, and toughness of PEN modied steels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0
5.4.2. Stainless steel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0
5.5. Fatigue strength and residual stresses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0
5.5.1. Endurance limit of nitrided medium carbon steels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0
5.5.2. Residual stress distribution in the surface layer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0
5.6.1. Medium carbon steels and cast iron . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0
5.6.2. High-speed steels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0
5.7.2. Stainless steel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0
6. Nitrocarburising. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0
6.2. Structure and phase composition of nitrocarburised layers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0
6.2.1. Low carbon steel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0
6.2.3. Stainless steel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0
6.2.4. Cast iron . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0
6.3. Mechanical properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0
6.3.1. Low-carbon steel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0
6.3.3. Stainless steel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0
6.4.2. Cast iron . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0
6.4.3. Stainless steels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0
6.5.1. Low carbon steel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0
6.5.3. Stainless steel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0
7. Future challenges . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0
8. Conclusions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0
Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0
References. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0
1. Introduction In this article, we review examples of PES of steels. In Section 2, we

consider key concepts and physical fundamentals of electrolytic plasma
Various techniques of electrolytic-plasma surface treatment initiation and its existence while no dielectric lm is formed on the elec-
technology were classied in the review by Yerokhin et al. [1] who trode surface and no substances are deposited from the electrolyte on
also discussed physical and chemical features associated with phe- the substrate. In Section 3, we analyse oxidation and anode dissolution
nomena and obtained results. By now, a number of research papers of steel workpieces. We cover PEC conditions, characteristics and
on plasma electrolytic treatment (PET) has increased signicantly. properties of carburised layers in Section 4 which is the main section
For this reason, the analysis of relatively new publications seems of the review. Sections 5 and 6 are devoted to the nitriding and
topical. nitrocarburising of steels. Finally, Section 7 discusses challenges and de-
Cathode and anode plasma electrolytic carburising (PEC) processes velopment prospects of PEC.
have been studied since the 1960s [2], and later applied for industrial
use. Anode PEC units were produced by the Academy of Sciences of 2. Physical features of plasma electrolysis
Moldova and used industrially in Russia, Ukraine, Belarus, Kazakhstan,
Moldova and Romania [2,3]. A new step in development of plasma elec- 2.1. Formation of the vapour-gaseous envelope
trolytic saturation (PES) was provided by pulse technology which en-
abled formation of nanostructured layers on the surface of metallic Conventional electrolysis is known to be interrupted at a certain crit-
substrates. Fundamentals of plasma-assisted cathode electrolysis, in- ical voltage U1 by formation of a vapourgas envelope (VGE) around
cluding pulse current modes, electrolyte compositions and mechanical one of the electrodes (Fig. 1) [1]. The main cause of VGE initiation is
and electrochemical properties of resulting coatings were reviewed by electrolyte boiling [7] while the gaseous products of electrolysis are sec-
Aliofkhazraei et al. [4]. Selected physical, chemical and technological ondary consideration. The VGE is formed at the electrode where opti-
features of PES were discussed elsewhere [3,5,6]. mum conditions for electrolyte boiling exist [8], e.g. at the electrode
P.N. Belkin et al. / Surface & Coatings Technology xxx (2016) xxxxxx 3
Fig. 1. Typical currentvoltage (a) and temperaturevoltage (b) characteristics for the process of anode plasma electrolysis without dielectric lm on the anode surface [2] and current
voltage characteristics of cathode plasma electrolysis (c) [12].
with smaller surface area. Therefore, the rst critical voltage U1 is asso- (mill-scales) removal is short-term (0.001 s) electrical discharges which
ciated with the thermal energy required for local electrolyte boiling provide signicant pressure and high temperature gradient in the elec-
around the electrode [2]. Conventional electrolysis shifts to the current trode surface layer.
oscillation mode when the cell operates as a pulse generator because of Regularities of cleaning including formation of micro-craters and
instabilities in the VGE which is periodically destroyed by pulsed dis- spheroids on the workpiece surface are explained by Gupta et al. by ef-
charges [9]. Critical eld strength causing VGE breakdown determines fects attributed to the collapse of plasma bubbles [5]. These bubbles can
discharge inception. At this stage, temperature of the electrode with melt micro-volumes of processed metal within fractions of a second but
the small surface area does not exceed 100 C due to the periodic con- these volumes are cooled and quenched after the bubbles collapse. The
tact renewal between the electrode and the electrolyte. Thus continu- surface morphology is characterised by two unique features, namely,
ous but unstable VGE is a characteristic feature of the oscillation mode. micro-craters and spheroids. These features are the result of plasma
The second critical voltage U2 determines transition from the current bubble implosion and localised quenching of melted surface layer.
oscillation mode to the stable VGE state, in which current oscillations This hypothesis can probably explain the discrete nature of dis-
are relatively small. The steady-state mode is described by decreasing charges, but needs to be claried. Gas bubbles can be formed in liquid
currentvoltage characteristics (CVC) typical of stationary electric dis- or in micro-craters on the metal surface between the workpieces and
charges. The VGE stabilisation occurs in the region U2U3 while U2 de- the electrolyte. Electrical eld in these bubbles can cause their break-
pends sufciently on the active electrode polarity. The minimal down and heating of metal surface up to its melting, whereupon the
voltage to maintain the true glow discharge is considerably lower on bubble will collapse and the metal micro-volume would be cooled.
the cathode (160 V) than on the anode (420 V) [8]. It is likely that the This process is not possible in a continuous VGE. A stable and continuous
marked difference of the voltages can be associated with possible elec- VGE can exist only due to its own electrical conductivity. In this case, the
tron emission from the metallic cathode. The CVC falling section is ex- VGE will receive the energy to maintain stable interface between the
plained as follows. The voltage increase results in evaporation electrolyte and the vapour-gaseous phase. Concentration of positive
intensication, an increase in VGE thickness and decrease in its electri- charges in the bubble in a continuous VGE presumed by the authors of
cal resistance [10,11]. This mode can be realised under conditions of [5] appears to be impossible, the charges will move and discharge on
both natural convection and controlled hydrodynamic ow of the elec- the electrode surface. Formation of bubbles and emergence of electrical
trolyte. As a rule, this stationary mode promoting the electrode heating discharges in the bubbles are possible only outside the continuous VGE.
up to high temperatures is used for PES. For example, the VGE continuity may be destroyed by electrolyte mov-
In the case of cathode treatment, a critical voltage U3 corresponds to ing with sufcient speed. It should be noted that the combination of
transformation of the normal glow discharge to intensive arcing with short-term local overheating of the metal surface with its cooling by
specic low-frequency acoustic emission [1]. In the case of anode treat- the electrolyte could be achieved by switching voltage. This method is
ment, arcing is impossible owing to the absence of a metal cathode pro- proposed for steel hardening with the use of plasma electrolytic
viding intense emission of electrons. The mode of stationary anode quenching [17].
heating up to maximum temperature of 1000 C exists in the range of Traditional methods are hardly applicable in experimental studies
100320 V. With increasing voltage, anode temperature drops sharply on physical characteristics of plasma electrolysis because of small VGE
to 200250 C probably due to increased possibility of local contact re- thickness. Electrical discharge formation in the cathode VGE is con-
covery between the anode and the electrolyte [2]. rmed by the data of optical emission spectrum which contains almost
The situation is different when electrolyte ows through the perfo- all elements of the system [18]. However, the optical emission spectrum
rated anode onto the workpiece (cathode) [12]. An intermediate of glow discharge in the anode VGE contains only lines of alkali or alka-
mode appears in the region U2U2, where the current decreases but line earth metals with low ionization potentials. These elements are not
plasma intensity increases when the voltage raises (Fig. 1c). The bulk necessary in provision of PES which requires the use of other
sample temperature remains relatively low (b 200300 C) due to the electrolytes.
simultaneous cooling by the electrolyte solution. However, the local Electrical conductivity of the anode VGE in the voltage range of
surface temperature around the plasma bubbles is very high. A specic 100300 V can be provided by anion emission from boiling electrolyte
feature of this intermediate mode is a continuous and unstable VGE in the VGE followed by transfer of these ions to the anode under the in-
with local disturbances of its continuity. The oscillation mode can be uence of electric eld. This conclusion is based on modelling of electro-
promoted by application of pulsed current [13]. lytic plasmas through lm boiling [2,19]. An electrode (nickel
The intermediate mode is used for cleaning steel surfaces (oxides chromium alloy wire) is immersed in an aqueous solution of ammoni-
and contamination) using the perforated anode [12]. Surface cleaning um nitrate and heated by the electric current passing it through until
of a steel band moving through electrolyte can be carried out in the cur- a vapour lm is formed completely surrounding the electrode. Thereaf-
rent oscillation mode when the bulk metal temperature does not exceed ter, the voltage is applied between the wire and the other electrode that
80100 C [1416]. The authors assume that the cause of an oxide layer is placed in the electrolyte. In the case of positive wire polarity current
ow through the vapour lm starts at 4050 V (ammonium nitrate con- Heat transfer in the studied system can be considered with a number
centration (2530 wt.%). The current ow does not change the observed of assumptions [11]. Electric energy is dissipated in the bulk of the VGE
glow colour; its emission spectrum is appropriate to that of a hot body since its thickness exceeds the mean free path of the particles in the
with a maximum at the wavelength of 700 nm. Conversely, when the vapourgaseous medium by two or three orders of magnitude at atmo-
wire is connected as the cathode, an unstable current develops at spheric pressure. In the simplest case, it is assumed that the current is
30 V, with bright blue ashes at the orange-red background. At described by Ohm's law, that is specic VGE conductivity is constant.
5758 V, a stationary blue glow is observed, the current increases sharp- A more sophisticated case takes into account the space charge in the
ly and the vapour lm is broken [20]. anode VGE, which appears to be caused by electrolyte anions. Evidently,
Evaporation does not provide electric current. The current through the space charge in the VGE affects the quantitative values of calculated
the anode VGE may be provided by means of the eld and thermal evap- current and temperature but does not change the CVC type. A physically
oration of the electrolyte ions [21]. Critical voltage U2 promoting the justied general description of the current ow through the VGE has not
anion emission is calculated on the base of GouyChapman model and yet been identied. It is shown that the heating conditions of a cylindri-
TonksFrenkel aperiodic instability [22]. Theoretical dependence of U2 cal workpiece may be described by two complexes: the dimensionless
on the electrolyte concentration is conrmed experimentally. energy released in the envelope, and the ratio of the densities of heat
The anion mechanism of the anode VGE conduction accounts for uxes from the envelope into the electrolyte and into the sample [24].
practically all the experimental data. These data include the inuence For the stationary anode PES, the heat ux per unit area from the
of electrolyte composition on the anode mass change (anodic dissolu- VGE to the liquid is described by a linear dependency of ql on the voltage
tion and oxidation) [3], heating temperature limit (~ 1000 C) due to and the electrolyte ow rate along the cylindrical sample [25]. Accord-
limited emissive capacity of the electrolyte [19], and the observed depo- ing to the experimental data for anode temperatures 600910 C, the
sition of solution components on the surface of the anode and on the heat ux per unit area from the VGE to the electrolyte ranges from 1.3
walls of the electrolyser [2]. Discrete current character in the case of to 2.4 MW/m2. The share of heat transferred to the electrolyte can
small anode surface area was found in [19], which may be related to reach 90% of the total power consumption [2].
both low- [23] and high-frequency [19,21] current pulses. The description of heat exchange between the VGE and the metallic
Some authors believe that the transition from the current oscillation electrode is difcult because of the absence of material surface that
mode to the stable electrode heating can be explained by transition transfers heat from the VGE to the anode. Heat transfers from the VGE
from bubble to lm boiling. Then the second critical voltage U2 is asso- to the electrolyte and the workpiece; therefore, the maximum VGE tem-
ciated with the power supplied to the system [1,5,16]. However, it perature is reached not on its boundaries but somewhere in the VGE
should be claried that the true bubble boiling in this system is impos- (Fig. 2) [3]. The mathematical surface of the maximum VGE tempera-
sible. The bubble boiling occurs on the surface heated by an external en- ture is analogous to the heating surface which may coincide with the
ergy source that is capable of providing sufcient heat ux into the outer workpiece surface at its top. For this reason, the formal use of
liquid. In the system under study, such a source does not exist. Electrical the heat transfer coefcient from the VGE to the workpiece results in a
energy is liberated in the VGE where the electrical resistance is maxi- number of contradictions owing to inhomogeneous distributions of
mum. It is the envelope that appears to be the primary source of heat temperature and heat ux on the workpiece surface [26]. Description
transferred to the electrode and the electrolyte. of the heat transfer using the heat ux density from the VGE to the
workpiece qA is more adequate. The method for determining qA is devel-
2.2. Heat exchange in the system electrolyteVGEmetal electrode oped by solving the inverse heat conduction problem with the use of
stepping regulation [27]. There is also a less accurate method for the de-
Nevertheless, the stationary heating mode is often considered anal- termination of qA by numerical differentiation of the experimental tem-
ogous to the lm boiling in liquid. Common features of conditions that perature distribution on the workpiece surface.
exist during PES and lm boiling are the liquidvapour interface and The heat ux density from the VGE to the metal electrode is linearly
the heat ux into the liquid, which ensures the interface stability. Essen- dependent on voltage and is equal to 0.10.21 MW/m2 in the range of
tial difference between the lm boiling and the PES is a presence of in- 140 to 270 V. The share of heat transferred to the anode is b10% of the
ternal heat sources in the VGE caused by the passage of electric current. total power consumption but it can reach 30% in the case of treating cy-
Fig. 2 shows the scheme of energy transfer in the electrode region for lindrical workpieces.
the stationary PES condition. The share of heat consumed by electrolyte evaporation does not ex-
ceed 8% of the total power [28]. In the range of 120 to 280 V, the heat
ux density from the VGE to atmosphere qs associated with evaporation
depends linearly on the voltage and increases with it from 0.02 to
0.15 MW/m2. An increase in the sample size results in this heat ux den-
sity to decrease.
When the end of a cylindrical sample is heated, the heat ux distri-
bution becomes signicantly different. The fraction of heat consumed
by evaporation can reach 60%. In this case, vapour pathway into the at-
mosphere is signicantly reduced. Conversely, the vapour motion along
the sample surface submerged into electrolyte increases, the fraction of
molecules returning to the electrolyte, which increases ql and reduces
qs.
The knowledge of heat uxes allows us to calculate the temperature
of the treated cylindrical sample as a function of voltage by solving the
equation of heat conduction in the VGE [11]. Obtained increase in
temperaturevoltage characteristics (TVC) is qualitatively consistent
with the experimental data up to the voltage range of 240250 V. Spe-
Fig. 2. Heat ow distribution in the vapour-gaseous envelope. Nomenclature: x cic conductivity of the VGE was established to increase from
Cartesian coordinate, thickness of the envelope T temperature, TA anode
temperature, TS water saturation temperature, W heat generation rate per unit
1.66 103 to 2.20 103 S/m with the increment in NH4NO3 concen-
volume, qA heat ux density from the envelope to the anode, ql heat ux density tration from 1 to 3 mol/l. Based on these results, an increasing depen-
from the envelope to the liquid, qs heat ux density from the envelope to air [11]. dency of the average VGE thickness on the voltage was obtained. The
voltage rise results in the increase in energy supplied to the VGE and its dissolution rate of titanium [3], chromium [34], and copper [35]. Iron
volume expansion due to intensied evaporation. Moreover, the calcu- dissolution rate evaluated based on the specimen weight loss due to
lation conrms the increase in the VGE thickness with the electrolyte iron transfer to the electrolyte always increases with voltage and pro-
temperature rise that leads to a reduction in the sample-anode temper- cessing temperature.
ature. Conrmed experimentally [7,29], this dependence accounts for Pores in the oxide layer aid not only anodic dissolution, but also
the need of the electrolyte temperature stabilisation and consequently, adsorption of carbon, nitrogen and boron on the surface followed
the saturation conditions. by their diffusion into the material. Fig. 3 shows the surface morphol-
Calculations of the VGE prole formed during the treatment of the ogy and elemental composition of the oxide layer formed during
vertical surfaces were carried out in [30]. The VGE expansion in the ver- anode carburising. Weight of the steel sample always decreases
tical direction results in the non-uniform temperature distribution on after anodic PES since iron dissolution prevails over its oxidation
the workpiece surface, thus leading to non-uniform thickness of the [36]. It should be noted that oxide formation is also observed during
modied layer, differential hardness, and other characteristic features. cathodic nitrocarburising for AISI 316 stainless steel in an aqueous
This disadvantage can be offset by promoting special hydrodynamic electrolyte containing urea [37].
conditions that would equalise the temperature distribution on the Fig. 4 shows dependences of the sample weight and the weight of
workpiece surface [3133] and even raise the temperature of individual iron found in the electrolyte on the treatment time [38]. The sample
sections where an increased hardness is required. weight loss is less than the amount of iron in solution due to the oxygen
Physical and chemical mechanisms underlying development of elec- transfer from the VGE to the sample. Iron oxidation may occur by the
trolytic plasmas are of signicant interest to researchers. Qualitatively a following reaction:
different state of the electrochemical system compared to conventional
electrolysis allows new information about laws of nature to be obtained. 3Fe 4H2 OFe3 O4 8H 8e 0:025 V: 1
In practical terms, insights into mechanisms of plasma electrolytic
modication are important when justifying a selection of treatment Hereinafter, the potentials are referred to the standard hydrogen
modes and electrolyte compositions, as well as developing technologi- electrode. The iron dissolution may occur with the formation of both
cal equipment. Depending on the hydrodynamic conditions in the Fe2+ and Fe3+. However, Fe2+ ions were not found in the solution. It
electrolyser, heat transfer regularities in the electrode region affect sig- is possible that several electrochemical reactions proceed simulta-
nicantly the liberation of saturating components and their transfer to neously. The anodic reaction of steel may be expressed in the following
the workpiece surface. Associated regularities depend on the electrolyte form [38]:
composition which, together with treatment modes, determines the
material properties after PES. Fe 3H2 OFeOH3 3H 3e : 2
The PES temperature is affected by applied voltage, electrolyte com-
position, temperature and ow rate as well as workpiece shape and size.
The composition of saturating medium (VGE) is determined by the elec-
trolyte composition and the PES temperature. Therefore, the main pro-
cessing parameters inuencing anodic treatments comprise PES
temperature controlled by voltage, electrolyte composition and ow
rate, treatment time and post-saturation cooling method (solution or
air).
Physical mechanisms of electrolytic plasmas and their theoretical
description are found only for the PES version of steady-state anode
heating. However, there are unexplored aspects in this simplest case.
For example, effects of electrolyte concentration on the sample temper-
ature are not explained and the reasons behind impossibility of anode
heating for some metals (aluminium, tantalum, and zirconium) are
not clear.
More complex conditions of transient heating are even less studied.
Regimes of thermal cycling found empirically enable to reduce signi-
cantly the heat transfer from VGE to the electrolyte and to control the
surface temperature of at components. However, the lack of under-
standing of the regularities of transient heating makes it impossible to
use these developments for treatments of complex shape components.
Moreover, the most complex processes are observed at the pulse cath-
ode saturation where short-term discharges migrating along the treated
surface provide its local heating followed by cooling. It is assumed that
such regimes lead to the formation of nanoscale structures. In this
case, the control of the results is carried out only by empirical selection
of the pulse current parameters without understanding the mecha-
nisms underlying this phenomenon.
3. Electrochemical features of plasma electrolysis
Anode dissolution and high-temperature oxidation of the sample

surface occur in parallel with oxide layer formation [2,3]. Many regular-
ities of anodic dissolution established for conventional electrolysis re-
main unchanged under the PES conditions. According to experimental Fig. 3. SEM image of the low-carbon steel surface after carburising in aqueous solution
data, iron dissolves more actively in chloride, than in nitrate or ammo- containing 10% NH4Cl and 10% glycerol (900 , 210 V, 5 min) followed by cooling in air
nium sulfate solutions. Similarly, the nature of anions determines the [36].
insufciently. Meanwhile, these processes affect very important proper-

ties of the surface layers, including wear and corrosion resistance.
4. Carburising
4.1. Electrolyte compositions
In most cases, binary aqueous electrolytes are used for PET. The rst
component provides to the solution a sufciently high electrical con-
ductivity for possible electrolyte boiling with VGE formation. The sec-
ond component is the source of diffusing species (in this case carbon).
Known electrolyte compositions for carburising and associated results
are shown in Tables 1 and 2.
It can be assumed that carbon-containing compounds in the form of
short, easily cleavable molecular chains are more preferable for PEC. For
example, dissociation of ethanol occurs in the case of cathode PEC as fol-
lows:
Fig. 4. Kinetics of iron transfer from anode to an aqueous solution (10% NH4Cl) during
stationary anode heating (200 V, 850 ) [38].
C2 H5 OH H C2 H5 H2 O: 8
This reaction cannot proceed in one-step, it is likely to be a combina- The C2H+

5 cation is then discharged at the cathode:
tion of one or two anodic and one chemical reactions:
C2 H5 e C2 H5 : 9
FeFe2 2e 0:44 V 3
Under sparking conditions, ethyl radical dissociates liberating atom-
Fe2 Fe3 e 0:77 V 4 ic carbon which can be adsorbed on the cathode surface and diffuse into
its crystal lattice:
Fe3 3OH FeOH3 : 5
2C2 H5 4C 5H2 : 10
Formation of Fe(OH)3 may also occur following reaction (1):
A hypothesis regarding glycerol decomposition in plasma discharge
Fe3 O4 FeO Fe2 O3 6 zone with formation of the H2, CO and some fragments of hydrocarbon
radicals has been put forward in [44]. Hydrocarbon groups such as CH
Fe2 O3 3H2 O2FeOH3 : 7 and CH3 have priority to be absorbed on cathode surface through
particle collisions [55]. In the end, the intermediate products are
Fe(OH)3 is an idealised description of multiple forms Fe2O3(H2O)x decomposed into C and H reactive species on the sample surface. The ac-
and FeOOH(H2O)x. Proton liberation in reactions (1) and (2) suggests a tive radicals and ionic components accelerated by electric eld, bom-
possibility of electrolyte acidication which is observed experimentally. bard and heat the sample, providing enhanced diffusion of atoms in
Information on dissolution and oxidation of PES treated steels is the iron lattice [44].
available only for selected electrolyte compositions. Inuence of electri- Chromatographic analysis of the organic volatile components
cal discharge on erosion of workpieces at cathode PES is also studied formed in the VGE during anode carburising at 1000 C in the electrolyte
Table 1
Examples of the cathode PEC. Nomenclature: U voltage, T sample temperature, layer thickness, t treatment time, Ra roughness (in brackets initial roughness), dc duty
cycle.
Electrolyte composition U (V) T Material and sizes sample (mm) t HV Ra (m) Research group
(C) (mm) (min) (load)
10% ammonium chloride + 10% glycerol in 210 850 Fe 0.078 60 850 Tarakci et al. [39]
water 10 12 2 (0.1)
25% potassium acetate in glycerol 300 Steel (0.15% ) 0.4 10 Inoue et al. [40]
10 10
10% ammonium chloride + 20% glycerol in 180 750 Fe 0.030 30 935 0.118 avulu et al. [41]
water 240 950 10 12 2 0.160 30 900 0.819
(0.1) (0.18)
50% urea + sodium chloride in water 240 SAE 1015 0.6 10 1200 Rezaei et al. [42]
10 5 HK
(0.03)
50% glycerol + 2% NaOH 150, 10 kHz 944 AISI H13 0.13 25 864 Yaghmazadeh et al.
26 3 (0.05) [43]
80% glycerol in water 380, 150 Hz, dc 794 T8 (0.75% C) 0.03 1 574 0.3 Wu et al. [44]
20% (0.01) (0.1)
10% sodium carbonate + 10% glycerol 850 30CrMnSiA 0.04 6 750 Skakov et al. [45]
(0.1)
10% sodium carbonate + 10% glycerol 121810 0.030.065 Kurbanbekov et al.
[46]
Glycerol based electrolyte in water 350 AISI 304 85 3 762 Xue et al. [47]
(0.02)
80% glycerol in water 350 304 0.085 3 762 Xue et al. [48]
40 5 1 (0.02)
Table 2
Examples of the anode PEC. Nomenclature: U voltage, T sample temperature, layer thickness, t treatment time, Ra roughness (in brackets initial). Hydrodynamic conditions
longitudinal ow the sample by electrolyte.
Electrolyte composition U T Material and sample sizes (mm) t HV Ra Research group

(V) (C) (mm) (min) (load) (m)
10% ammonium chloride and 10% acetone in water 950 Steel 0.2% 0.14 7 750 Belkin et al. [49]
12 14 (0.05)
5% ammonium chloride and 5% glycerol in water 250 900 Steel 0.2% 5 630 0.22 Zhirov et al. [50]
12 14 (0.05) (0.6)
10% ammonium chloride and 10% glycerol in water 275 900 Steel 0.1% 0.1 5 700 Zhirov et al. [50]
10 60 (0.05)
10% ammonium chloride and 4% sucrose in water 210 900 Steel 0.2% 0.14a 10 Belkin et al. [51]
Hexagonal samples; facet 4 mm and length 15
mm
10% ammonium chloride and 4% ethylene glycol in 210 900 Steel 0.2% 0.1a 10 Belkin et al. [51]
water Hexagonal samples; facet 4 mm and length 15
mm
15% ammonium chloride and 10% acetone in water 860 Fe 0.22 26 Duradji et al. [52]
10% ammonium chloride and 10% glycerol in water 900 Steel 0.1% 0.1 10 700 0.08 Kusmanov [53]
12 14 (0.05) (0.42)
10% ammonium chloride and 12% glycerol in water 235 950 Powder irongraphite 1% C, 2% Cu 0.4 20 Belikhov et al.
16 15 [54]
a
Thickness of martensite layer.
containing glycerol (10%) and ammonium chloride (10%) showed pre- carburising was 0.9 wt.% in aqueous solution of acetone, 0.8, 0.7 and
vailing formation of formaldehyde and methanol with minor amounts 0.6 wt.% in solutions of glycerol, sucrose and ethylene glycol, respec-
of acetaldehyde, acetone and ethanol [53]. Decomposition mechanism tively [3,49]. When the sample was cooled in electrolyte after
of glycerol during anode carburising can be described as follows. The carburising, a martensite layer with high hardness is formed (Fig. 5)
glycerol is oxidised to glyceraldehyde and dihydroxyacetone that are in- [39]. In the case of air-cooling, a slight quenching or its absence are pos-
volved into the VGE and decomposed into low molecular weight com- sible with perlite layer formation which may contain cementite net-
pounds detected by chromatography. These compounds are absorbed work [3,49]. The phases usually obtained in carburising (cementite,
at the anode surface and subjected to electro-oxidation and dissociation martensite or pearlite, ferrite and austenite) are conrmed by X-ray dif-
to form active atomic carbon. fraction (XRD) analysis [39].
Apart from the components providing electrical conductivity and According to XRD results, the oxide layer formed after the anode
supplying carbon, additives are recommended to improve some electro- carburising contains FeO, Fe2O3 and Fe3O4 (Fig. 6). Oxygen distribution
lyte properties. Surfactants (isoamyl alcohol, oxyethylated higher fatty in the surface layer determined by proton nuclear backscattering
alcohols, etc.) result in signicant reduction of current density together shows that the thickness of the oxide layer is several micrometres;
with a small decrease in temperature, which provides electric energy deeper within tenths of micrometres an iron zone enriched with
savings [56]. To increase viscosity of the electrolyte solution, additives its oxides is formed [3,51,58].
of soluble inert substances, such as gelatine or starch are also proposed Anode PEC of powder iron-graphite results in perlite layer formation
[57]. without carbon diffusion from graphite pores [54]. This layer with thick-
ness of 0.5 mm was obtained after a four-stage anode carburising at
4.2. Structure and phase composition of carburised layers 950 C, each lasting for 5 min, followed by air-cooling.
4.2.1. Low-carbon steels

The maximum surface carbon concentration (1.3 wt.%) was obtained
through cathodic carburising using potassium acetate solution in glyc-
erol [40]. The surface carbon concentration obtained by anode
Fig. 6. Cross-section of the sample (diameter 12 mm, length 14 mm) treated at 850 C for
Fig. 5. SEM image of pure iron carburised at 850 C for 30 min, showing the carbon rich 10 min in an aqueous solution of ammonium chloride (10%) and glycerol (10%). 1 oxide
layer (1), the transition layer (2), and the base metal (3) [39]. layer, 2 carburised layer [49].
4.2.2. Low-alloy steel The measured carbon distribution in the carburised layer can be ap-
Noteworthy are the results of AISI H13 steel carburising using high- proximated by a well-known equation [36]:
frequency voltage. The current at frequency of 10 kHz is shown to re-
duce the crystal size to the nanometre scale and to improve the wear re- C xC 0 x
1erf p ; 11
sistance of the material [43]. C S C 0 2 Dt
where (x) is the nal carbon concentration at depth x, 0 is the initial

4.3. Carbon diffusion kinetics carbon concentration in steel, S is the nal carbon concentration on
the sample surface, D is the carbon diffusion coefcient in steel, t is
4.3.1. Low-carbon steels time, and erf(x) is the Gaussian error function. This equation represents
Dependence of layer thickness on treatment time is adequately de- the solution to the problem of nonstationary diffusion for a half-innite
scribed by a parabolic function (Fig. 7). Cathodic PEC of pure iron in body with the following boundary conditions: (0, 0) = 0, C(, t) = C0,
glycerol electrolyte also yields a parabolic dependency of the carburised and C(0, t) = CS. Thickness dependence of the carburised layer on pro-
layer thickness on processing time (line 5, Fig. 7), as derived from data cessing time as well as S enables rough estimation of the carbon diffu-
of [39]. These dependencies conrm the diffusive nature of the process sion coefcient using Eq. (11). The data obtained for different
and indicate the constant carbon concentration on the sample surface. electrolytes containing 10% ammonium chloride and 10% of one of the
Furthermore, these data indicate a rather rapid stabilisation of condi- above mentioned carbon components are presented in Table 3.
tions in the VGE, which is an important advantage of PEC. The increase in diffusion coefcient of carbon in steel with increasing
The layer thickness increases expectedly with the carbon saturation temperature is common since it follows exponential law. The inuence
temperature. It is important to note that different electrolyte composi- of carbon component on carbon diffusion coefcient can be explained
tions result in different maximum achievable temperatures of treated by the change in the thickness of the layer enriched in iron oxides.
sample [59]. Therefore, the thickness of the layer obtained in a Oxidising ability of electrolytes determined by XRD analyses according
glycerol-based electrolyte at the temperature of 1000 C is not signi- to the intensity of the iron oxide peaks increases in the sequence of ac-
cantly different from the thickness observed after carburising in the etoneglycerolsucrose [54]. Reduction in carbon diffusion coefcient
electrolyte with acetone at 900 C, in spite of a higher saturating capac- with increasing oxygen content proves inhibitory effects of the oxide
ity of the latter. layer.
A layer thickness formed by the anode PEC depends not only on the The inuence of electrolyte composition on the carbon diffusion co-
concentration of the carbon-containing components in the electrolyte efcient highlights different structure and phase composition of obtain-
but also on dissolution and oxidising rates of the sample material [3, ed layers. For this reason, it is preferable to refer to an effective diffusion
51]. Increase in concentration of ammonium chloride containing no car- coefcient in the multiphase system. Glycerol concentration fall or am-
bon results in the thickening of carburised layer owing to intensied an- monium chloride concentration rise lead to an increase in the estimated
odic dissolution and reduction in the thickness of the oxide layer which value of carbon diffusion coefcient. These facts conrm both the pre-
retards carbon diffusion. vention of carbon diffusion in the bulk of the sample by the surface
Dependence of the carburised layer thickness on the concentration oxide layer and the inuence of electrolyte composition on the oxygen
of carbon components has a maximum which is associated with the content in it. Increases in both temperature and treatment time also
competition between the following two processes. When the concen- lead to an increase in the carbon diffusion coefcient, but to a different
tration increases from 0 to 2%, the saturating capacity of the VGE rises extent. Increasing processing time from 5 to 10 min is more effective
because of the ow increase of carbon compounds into the VGE from than raising sample temperature to 1000 C to increase both the layer
the electrolyte. Further concentration increase leads to reduction in thickness and carbon content in it. The results presented in Table 4 indi-
the carburised layer thickness due to the oxide layer growth owing to cate that this fact is due to an increase in the carbon surface concentra-
the decrease in the rate of anodic dissolution. Accordingly, the current tion, rather than in diffusion coefcient. Substitution of acetone for
density decreases as the concentration of carbon compounds is glycerol results in an increase in the effective diffusion coefcient and
increased. higher carbon surface concentrations, which is consistent with the
data in Table 3 demonstrating a higher saturation capacity of the ace-
tone electrolyte.
4.3.2. High-carbon steel

Enhanced carbon diffusion is discovered under the conditions of
pulse cathode PEC of high-carbon steel forming a carburised layer con-
taining diamond-like carbon [44]. In this case, the carbon concentration
on the sample surface reached 13.25 wt.%. Carbon diffusion coefcient is
(1.39 0.14) 107 cm2/s. Plasma discharge provides the activation
energy reduction of carbon diffusion from 151 kJ/mol for conventional
carburising process to 132.8 kJ/mol. More accurate values of carbon
Table 3
Average carbon diffusion coefcients (1012, m2/s) in low-carbon steel (0.2% C) at differ-
ent temperatures T (t = 5 min).
Carbon containing component
T () Acetone Glycerol Sucrose Ethylene glycol
800 103 21 52
Fig. 7. Dependence of the square of layer thickness on carburising time in aqueous
850 346 36 102 18 60 12 83
electrolytes with ammonium chloride concentration, 10%. It presents the martensite
900 710 52 172 23 107 16 25 6
layers on steel with 0.2% C obtained at 900 C (14) and carburised layer on pure iron
950 261 28 190 21
obtained at 850 C (5). Carbon containing components (10%): 1 acetone, 2 and 5
1000 494 38 269 24
glycerol, 3 sucrose, 4 ethylene glycol.
Table 4 In other circumstances, roughness rise after cathode carburising

Calculated data of carbon diffusion coefcients and carbon surface concentrations. does not lead to an increase in the coefcient of friction. For example,
C1 (wt.%) C2 (wt.%) T (C) t (min) D (108 cm2/s) CC (wt.%) pulse carburising of T8 steel in a glycerol electrolyte results in the sur-
Glycerol 10 10 900 5 0.78 0.12 0.91
face roughness increase from 0.1 to 0.3 m, but the coefcient of dry
Glycerol 5 10 900 5 3.5 0.4 0.93 friction against ZrO2 decreases from 0.4 to 0.30.35 [44]. The wear rate
Glycerol 10 15 900 5 12.9 1.6 0.87 of PEC-treated steel is about 1.13 105 mm3/Nm which is less than
Glycerol 10 10 1000 5 7.8 0.8 0.86 half of that for T8 steel. High wear resistance of PEC-treated steel is
Glycerol 10 10 900 10 6.8 1.7 0.95
caused by high hardness of a diamond-like carbon containing
Acetone 10 10 900 5 12.5 1.7 1.27
carburised layer and possibly graphitisation of this surface during fric-
Nomenclature: C1 is the concentration of carbon containing components, C2 is the concen- tion process. Moreover, high oxygen content (18.98 wt.%) conrmed
tration of ammonium chloride, T is the sample temperature, t is treatment time, D and Cc
are carbon diffusion coefcient and surface concentration respectively.
by energy dispersive X-ray spectroscopy (EDS) analysis indicates that
the wear mechanism of PEC-treated steel transforms to oxidation
wear. The friction heat causes steel oxidation in the contact region,
diffusion coefcients were obtained using spectral emission analysis of and iron oxides act essentially as lubricants.
carburised surfaces after PEC (Table 4) [53].
4.4.4. Low-alloy steel
4.4. Tribological behaviour An increase in wear resistance is achieved using pulse processing.
This result is obtained at the cathode carburising of steel AISI H13 in
4.4.1. Low-carbon steel electrolyte containing 50% glycerol and 2% NaOH at 150 V and the
Cathode carburising of pure iron is carried out in a solution contain- pulse duty cycle, 90% [43]. Weight loss was measured at the sliding dis-
ing about 10% ammonium chloride and 20% glycerin [41]. The highest tance of 250 m. The minimal weight loss equal to 0.022 g was observed
hardness is obtained at 750 C when martensite and cementite are de- for the carburised layer obtained at a frequency of 10,000 Hz. Worse
tected in the surface layer by XRD analysis. Cementite is not identied data were obtained at 100 Hz (0.033 g) and at DC (0.028 g). The weight
at higher processing temperatures (900 and 950 C). An increase in loss of the untreated sample was 0.043 g. It should be noted that the
the saturation time results in the hardness rise probably owing to car- highest temperature of the sample 944 C was obtained at the frequency
bon content increment due to diffusion. As a rule, their surface rough- of 10 kHz, whereas the frequency of 100 Hz corresponded to 918 C. Yet
ness Ra increases from 0.18 m (untreated sample) to 0.250.98 m as lower temperature was achieved at DC for the same voltage of 150 V.
a result of cathode carburising; no effect of PEC duration and tempera- These results are consistent with the hardness measurements, where
ture on the surface roughness is identied. samples processed at high and low frequencies had the maximum and
Anode carburising of low-carbon steel (0.08% C) in a solution con- the minimum surface hardness, respectively.
taining 10% ammonium chloride and 10% glycerol (treatment mode: Information available indicates that friction mechanisms of steels
voltage 200 V, current 3 A, 950 C) demonstrates similar results. Surface subjected to PEC treatments have been studied insufciently. Typically,
hardness of cooled in electrolyte carburised layers exceeds 850 HV50 cathodic PEC leads to increases in the surface roughness and friction co-
owing to martensite phase formation. Microhardness is not signicantly efcient, but the surface hardness also increases owing to the formation
affected by treatment temperature and time. In this case, the modied of martensitic layer, providing improved wear resistance. Anode PEC
layer thickness increases with treatment time. usually results in surface smoothing and causes friction coefcient to
Cathode carburising may result in the increment of friction coef- decrease, but it is always accompanied by the formation of an oxide
cient of the samples. The untreated and carburised samples were tested layer. In this case, an improvement in the wear resistance is not always
against 6-mm Al2O3 ball counterparts under 10 N load in dry sliding associated with increased hardness, being determined by a combination
conditions. The mean average friction coefcients increased from 0.6 of the hard martensitic sublayer and a porous oxide surface layer which
to 1.0 (maximum value). The wear rate of carburised samples was can retain lubricants. In addition, a certain amount of austenite retained
lower than that of the untreated one. after quenching can improve the running-in ability of carburised steels
Anode carburising usually results in decreased surface roughness and provide localised plastic deformation in the thin surface layer.
due to anodic dissolution. For example, surface roughness of mild steel
specimens treated at 900 C decreases from 1.2 m (untreated sample) 4.5. Corrosion resistance
to 0.2 m during the rst 2 min of carburising and remains constant af-
terwards [50]. 4.5.1. Low-carbon steels
Corrosion properties of samples after the anodic PEC are improved.
One of the qualitative characteristics of the oxide layer is the corrosion
4.4.2. Medium carbon steel behaviour of samples after processing. The open circuit potential de-
Anodic carburising improves the mechanical properties of steel [50]. creases during the rst 10 min [36] and then oscillates, which can be as-
Nominal tensile strength increases from 420 20 MPa (untreated sam- sociated with the saline passivation and development of diffusion
ple) to 930 50 MPa (carburised with quenching). The toughness (im- limitations in the porous oxide layer. Polarisation curves indicate a
pact test) of the material is the same in comparison with the traditional stronger tendency for passivation for the samples cooled in air. The
method of heat treatment. cathode branches contain nearly linear regions, indicating the only the
reaction of hydrogen reduction taking place. The absence of such re-
4.4.3. High-carbon steel gions in the anode branches can be associated with the formation of a
Enhanced carbon diffusion is discovered under the conditions of relatively dense surface oxide layer which limits the kinetics of anodic
pulse cathode PEC of high-carbon steel forming a carburised layer con- process.
taining diamond-like carbon [44]. In this case, the carbon concentration
on the sample surface reached 13.25 wt.%. Carbon diffusion coefcient is 4.5.2. Medium carbon steels
(1.39 0.14) 107 cm2/s. Plasma discharge provides the activation Treatment of medium carbon steel (0.45% C) in a solution of ammo-
energy reduction of carbon diffusion from 151 kJ/mol for conventional nium acetate provides a positive potential shift for the specimens in
carburising process to 132.8 kJ/mol. More accurate values of carbon dif- 0.5 M sulfuric acid solution [60]. The best results are obtained after ox-
fusion coefcient were obtained using spectral emission analysis of idation of samples at voltages of 200240 V, which corresponds to the
carburised surfaces after PEC (Table 4) [53]. processing temperature of 850950 C. In this case, the sample potential
of OCP vs a saturated silver-chloride electrode is in the range of shown [64]. The hydrolysis of ammonium carbonate is suggested to
0.200.35 V and the layer conductivity is 1050 S/cm. Extension of occur as follows:
treatment time over 3 min leads to expanding of surface which nega-
tively affects the workpiece corrosion characteristics. NH4 2 CO3 2NH 2
4 CO3 12

H2 OH OH :
4.5.3. Stainless steels
Similar results were obtained for anodically carburised stainless At low potentials, these ions are neutralised at the cathode and am-
steels (austenitic and martensitic) studied by the electrochemical im- monia and hydrogen are produced. However, at high potentials, plasma
pedance method [61]. Corrosion rate of samples treated in glycerol elec- generated at the cathode may also contain other ions, like N+, NH+, and
trolyte decreased by a factor of 7 (AISI 403) to 30 (AISI 304) with the use NH+2 , similar to plasma nitriding.
of boric acid or borax solutions as the corrosion medium. The following reaction paths could lead to formation of iron nitrides
The effect of carburising on the corrosion behaviour of three dif- via NHx species [65]
ferent stainless steels (AISI 304, 316L and 430) was studied in [62]
by scanning potentiodynamic and electrochemical impedance spec- Fesolid NHx;gas FeNH23 Fe24 N N2 H2 NH3 : 13
troscopies. Less compact structures and/or smaller amounts of
alloying elements (as occurs with the body centred cubic structure A mixture of ammonium chloride and nitrate can be used as a source
in ferritic AISI 430 steel) are affected stronger by this modication. of nitrogen for the anode PEN which is followed by hardening [6]. In this
Conversely, stainless steels with a larger amount of alloying ele- electrolyte, the ammonium salts are hydrolysed to ammonium hydrox-
ments and/or more compact structures show only slightly changed ide that liberates ammonia to the VGE at the electrolyte boiling temper-
corrosion behaviour. Corrosion rates of the carburised samples are ature [66]:
slower than of the untreated one, and most of the treated samples
NH4 Cl H2 ONH4 OH HCl 14
maintains the best behaviour for longer periods of time. Carburising
of AISI 304 increases the localised corrosion resistance. Increasing
NH4 NO3 H2 ONH4 OH HNO3 15
applied voltages above 260 V strongly affects the corrosion mecha-
nism by producing a surface layer, which could favour a more passive
NH4 OHNH3 H2 O: 16
behaviour. In contrast, carburising produces a harmful effect on AISI
430 by decreasing its pitting corrosion resistance.
It should be noted that gaseous nitrogen added into PES process can
The most common component for carburising of steels is glycerol.
serve as a source of active nitrogen too [67]. Known electrolyte compo-
However, comprehensive studies of the properties of the base-glycerol
sitions for nitriding processes and associated results are shown in
electrolytes include their saturation ability, resources, and energy con-
Table 5.
sumption does not carry out. There are some positive examples of PEC
steels for improving their protective properties, such as resistance to
5.2. Structure and phase composition of nitrided layers
wear and corrosion. Relationship between the test results and structural
features of the modied layers must be found for the scientic justica-
5.2.1. Medium carbon steel
tion and practical application. Similar, still unresolved, problems arise
The cathode PEN of cast steel S0050A allowed nitride and diffusion
when studying corrosion behaviour of carburised steels where the in-
layers to be obtained, with thicknesses of 45 and 100 m, respectively
crease in their resistance may be due to the formation of the martensitic
[69]. This treatment was carried out in an aqueous solution of carbam-
phase or the presence of iron oxides.
ide during 8 min at 280 V (sample surface temperature ~ 600 C). The
The protective properties of iron oxides are most clearly manifested
optical micrograph of cross-section of a cast iron G3500 nitrided for
after the anode PEC of steels. In this case, the oxide layer is easily detect-
5 min at 250 V (sample surface temperature 570 C) (Fig. 8) shows
able. A positive role is also presumed for spinel oxides formed during
that the thickness of the nitride layer is 10 m and that of the diffusion
cathode PEC [62]. Another phase leading to increased corrosion resis-
layer 40 m, with both layers being not uniform because of the effect
tance is martensite. Currently, PEC regimes and electrolyte composi-
of graphite blocking nitrogen diffusion.
tions contributing to the increase in corrosion resistance have been
No nitride layer was formed at all on the surface areas where the
established for various steels. However, the study of the structural fea-
graphite akes were located. The diffusion of nitrogen was also blocked
tures of surface oxide layers is in the initial stage. The pore size in the
by graphite akes underneath the surface layer (i.e., subsurface). It was
oxide layer has been determined and the possibility of its peeling off
found that both of the XRD patterns, obtained from the treated cast iron
due to sharp changes in temperature noticed. Nevertheless, the effects
and cast steel, matched the reference of Fe2.4N.
of anode PEC conditions on oxide layer structure and its protective
Surface saturation of steel with nitrogen results in both formation of
properties have not been evaluated comprehensively, which compli-
nitrides (or solid solution of nitrogen in iron) and decrease in
cates justication of processing modes.
austenitisation temperature. According to the FeN phase diagram,
the eutectoid transformation temperature decreases from 727 C (Fe
5. Nitriding and nitrohardening
C phase diagram) to 700 C in presence of 0.07 wt.% N and becomes
590 C when steel contains 0.1 wt.% N or more [63]. Therefore, the sat-
uration with nitrogen at 650750 C during several minutes results in
austenitisation of the surface layer and formation of nitrogenised mar-
Ammonia is widely used in the processes of gas and plasma nitriding
tensite in the case of rapid cooling. PEN combines easy with quenching
as a source of atomic nitrogen. At the specimen surface, ammonia mol-
in the same electrolyte; therefore, this process can be termed
ecules are adsorbed and dissociate by stepwise removal of hydrogen
nitrohardening.
atoms. The adsorbed nitrogen atoms can diffuse into the bulk of the
The SEM analysis shows a laminated structure of the steel sur-
metal [63]. Under the condition of plasma electrolytic nitriding (PEN),
face after nitrohardening. It is composed of the following layers
volatile ammonia is evaporated from the electrolyte to the VGE and
(Fig. 9):
adsorbed on the workpiece surface.
Besides ammonia, some ammonium salts can serve as a source of ni- surface oxide layer;
trogen. The feasibility of nitriding of austenitic stainless steel in the nitridemartensite layer (including martensite, retained austenite
aqueous electrolysis of 5 wt.% solution of ammonium carbonate is and Fe4N and Fe23N nitrides);
Table 5
Examples of PEN treatments.
Cathode PEN
Nomenclature: U voltage, T sample temperature, layer thickness, t treatment time
Electrolyte composition U (V) T (C) Material and sizes sample (mm) (mm) t (min) HV Research group
(load)
25% potassium acetate in formamide 200 Steel (0.15% ) 0.55 20 850 Inoue et al. [40]
10 10
Potassium chloride in formamide Q235 Li et al. [68]
Carbamide in water 280 600 Steel S0050 A 0.145 8 700 Nie et al. [69]
(0.05)
20% carbamide and 2% potassium hydroxide 600a SCM420 0.0027 20 832 Kong et al. [70]
5 30 (0.05)
5% ammonium carbonate in water 440 Austenitic stainless steel Disc 8 0.08 40 1040 Roy et al. [65]
(0.02)
7.5% ammonium carbonate in water 550 0.05 20 875 Roy et al. [65]
(0.02)
20% carbamide and 10% sodium carbonate 320 550 R65 7 Skakov et al. [71]
Carbamide in water 250 570 Cast iron G3500 0.05 5 780 Nie et al. [69]
24.5 24.5 8 (0.05)
Anode PEN
Nomenclature: T sample temperature, t treatment time, 1 compound layer thickness, 2 diffusion layer thickness
Electrolyte composition T (C) Material and sizes sample (mm) 1 (m) 2 (mm) t (min) HV Research group
(load)
10% ammonium chloride and 5% ammonia 750 Steel (0.4% C, 1% Cr) 15 20 20 5 1000 Belkin. et al. [49]
(0.05)
8% ammonium chloride and 10% ammonium nitrate 650 Fe 20 0.12 5 650 Belkin et al. [72]
(0.05)
9% ammonium chloride and 15% ammonium nitrate 570 Fe 5 0.4 5 650 Belkin et al. [72]
(0.05)
15% ammonium chloride and 10% ammonium nitrate 750 Steel (0.45% C) 30 0.14 5 1060 Kusmanov et al. [66]
12 15 (0.05)
a
Pulse power supply with 2 s width and 15 kHz frequency.
Fig. 8. Microscopic images of (a, b) the untreated sample and (c, d) cross sections of the treated ones [69].
patterns of plasma treated austenitic stainless steel show diffraction

peaks corresponding to nitrides of iron and chromium, i.e. Fe4N, Fe2N
and Cr2N. The peak intensity of Fe2N increases with nitriding tempera-
ture, which is associated with the gain of the VGE nitriding potential.
It should be noted that the processing conditions in this research were
not stable, as current variation during the process was random making
temperature control difcult. In order to maintain the temperature
within the desired range, the voltage was controlled between 135 and
200 V and the electrolyte was not cooled during the treatment.
After the cathode PEN, a 23 m thick iron nitride (FeN0.076) layer
was formed in the outmost surface, resulting in hardness values of
about 800 HV20 [64]. However, the cathode treatment led to erosion
of the sample surface. The surface porosity generated by discharge
coarsened when the amount of KOH in the electrolyte has increased
from 10 g/l to 40 g/l.
5.3. Nitrogen diffusion kinetics
Under the stationary PEN conditions, the thickness of the compound

layer on low-carbon steels increases parabolically with the treatment
time (Fig. 10) [6], which testies a constant nitrogen concentration on
the surface. All layers enlarge with the increase in temperature owing
to the intensication of the nitrogen diffusion. Fig. 11 shows the diffu-
sion layer thickness versus treatment time. More precisely, this depen-
dency represents only a part of the diffusion layer containing excessive
Fe4N nitrides that were identied through their coagulation after aging
of the nitrided samples. The PEN treatment in a chloridenitrate electro-
lyte leads to a more developed diffusion zone. It can be assumed that the
rate of ammonia evaporation from nitrate solution into the VGE in-
creases with temperature rising to 700 C and becomes equal to its for-
mation rate. Therefore, the thickness of the nitride layer formed in the
chloridenitrate solution at 650750 C remains the same. On the con-
trary, ammonia evaporation rate from the solution containing ammonia
into the VGE increases with temperature monotonically. This explains
both increasing in the nitride layer thickness and a relatively low re-
source of this solution due to the rapid depletion with ammonia in the
operation.
Fig. 9. SEM image of cross-section of the medium carbon steel surface after anode PENH in
5.4. Hardness, strength, and toughness of PEN modied steels
a solution of ammonium nitrate (10%) and ammonium chloride (10%) at 750 C for 5 min.
1 surface oxide layer; 2 nitridemartensite layer; 3 martensiteferrite layer; and 4
initial pearliteferrite structure [66]. 5.4.1. Medium carbon steels
Similar values of hardness were obtained by PEN of medium carbon
steel (0.45% C) followed by quenching in chloridenitrate electrolyte
martensiteferrite layer (including martensite, retained austenite
and solid solution of nitrogen);
original pearliteferrite structure of the substrate material.
Formation of nitrides in the surface layer after the PEN treatment was
conrmed by XRD analysis, with Fe4N and Fe23N nitrides, FeO and
Fe3O4 oxides, martensite, retained austenite and ferrite detected.
The Fe23N phase was detected only after PEN at 750 C. The anode
PEN of the steel samples resulted in thicker nitride layers using the
electrolyte containing ammonia and ammonium chloride [6].
Measurements of the surface nitrogen concentration carried out by
different methods showed similar results. According to EDS analysis
the maximum concentration of nitrogen (5.6 wt.%) was observed be-
neath the oxide layer and further down into the bulk of the sample
[66]. The nitrogen concentration in the nitride zone after PEN in the
electrolyte containing ammonia and ammonium chloride was
4.6 wt.% according to the X-ray probe analysis [73] or 7.3 wt.% accord-
ing to the Auger electron spectroscopy [6].
5.2.2. Stainless steel Fig. 10. Dependence of the square of nitride layer thickness on processing time in a
Generally, the microstructure of a nitrided layer formed at low tem- solution of ammonium nitrate (11%) and ammonium chloride (11%) for different
peratures is similar to that obtained by conventional methods [64]. XRD treatment temperatures [6].
electrolyte, although the maximum concentration of nitrogen on the

surface did not change. Adjustment of quenching temperature after ni-
trogen saturation may provide additional means of controlling thick-
ness and hardness of the modied surface layer. Decreasing quenching
temperature compared to that of nitriding leads to reduction of layer
thickness and its maximum microhardness.
Maximum microhardness of the nitrided layer obtained by anode
PEN followed by quenching is observed at the some depth where the
martensite concentration is maximum [49]. The outer part of the nitride
layer contains Fe23N nitride and retained austenite with a lower hard-
ness. A similar decrease in hardness is observed for gas-nitrided steels
[74]. The PEN temperature rise results in increased surface hardness of
medium carbon steels due to the growth of nitrogen concentration in
martensite. The surface hardness of the nitrided steel increases with
processing time prolonged from 1 to 5 min owing to increases of nitro-
gen concentration in both nitride and martensite phases. However fur-
ther saturation with nitrogen does not increase the hardness.
The modied layer thickness can be controlled using the three-step
processing that includes heating for the nitrogen saturation, cooling,
Fig. 11. Dependence of the square of diffusion layer thickness on processing time in a and second heating for hardening in the same electrolyte. For example,
solution of ammonium nitrate (11%) and ammonium chloride (11%) for different
a medium carbon steel sample (0.45% C) was treated in the electrolyte
treatment temperatures [6].
containing 11 wt.% ammonia chloride and 11 wt.% ammonia nitrate at
750 C (160 V) during 2 min. Further, the voltage dropped to 80 V and
[66]. The surface microhardness prole is shown in Fig. 12. It is associat- the sample temperature decreased to 520 C during 30 s. Then, before
ed with the phase composition of the modied layers. The maximum quenching the voltage increased for a short period of time to 160 V
value of microhardness and the greatest width of the hardened zone and switched off. When the second step decreased from 6.5 s to 2.7 s,
were observed after nitriding at 750 C (Fig. 12a). The microhardness the hardened layer thickness reduces from 2.2 mm (900950 HV50) to
of the modied layer increased with the increase in treatment time 1.05 mm (800850 HV50).
from 2 to 10 min (Fig. 12b). At various electrolyte concentrations, the It was established that the anode PEN of a medium carbon steel sam-
maximum values of microhardness and depth of the hardened zone cor- ple (0.45% C) leads to an increase in its tensile strength similarly to
related with the thickness of the modied layer (Fig. 12c and d). The nitrocarburising followed by quenching [58] (Table 6). Fracture tough-
best results were obtained after the PEN treatment in a solution of am- ness of round samples with U-type notches was measured after three-
monium nitrate (10%) and ammonium chloride (15%) at 750 for step PEN in the aqueous solution of ammonium chloride (11 wt.%)
5 min. This is due to an increase in nitrogen content in the surface and ammonium nitrate (11 wt.%) at 770 (130 V) during 2 min. Fur-
layer with increased ammonium chloride concentration in the ther, the voltage was dropped to 97 V at the sample temperature
Fig. 12. Microhardness distribution in the modied layer after anode nitriding at different treatment temperatures (a) and time (b), concentration of ammonium chloride (c) and
ammonium nitrate (d) at: (a) 10% NH4NO3, 10% NH4Cl, 5 min; (b) 10% NH4NO3, 10% NH4Cl, 750 C; (c) 10% NH4NO3, 5 min, 750 C; and (d) 10% NH4Cl, 5 min, 750 C [66].
Table 6 Table 8
Characteristics of hardened steel after the PEN during 5 min. Effects of nitriding regimes on the fatigue limit (MPa) of the samples with different stress
concentration.
Saturation Layer Tensile Maximum Layer
temperature thickness strength elongation ductility Anode PEN with quenching
() (mm) HV0.05 (MPa) (%) (%)
Nitriding time
Stress Average
Aqueous solution of ammonium chloride (10%) and ammonia (5%) (min)
Concentration concentration coefcient of
650 0.4 500600 730 14.0 85
number factor Untreated 3 6 9 hardening
700 0.8 800900 820 11.0 83
750 1.0 800900 900 10.0 75 PEN temperature is 650 C
1 4.6 112 236 248 226 2.11
Aqueous solution of ammonium chloride (11%) and ammonium nitrate (11%)
2 2.5 137 239 232 228 1.70
650 0.4 550650 740 12.0 80
4 2.2 200 259 248 244 1.25
700 0.8 850950 880 6.1 76
3 2.0 189 222 240 242 1.24
750 1.0 850950 890 5.4 75
5 2.0 230 244 256 258 1.10
PEN temperature is 750 C

decreased to 550 C during 15 s. Then, before quenching, the voltage 1 4.6 112 251 247 247 2.21
2 2.5 137 276 245 259 1.90
was increased to 180 V and after a short exposure, switched off
4 2.2 200 317 323 419 1.76
(Table 7). Fracture toughness decreases from 1.11 MJ/m2 (untreated 3 2.0 189 335 300 330 1.70
sample) to 0.19 MJ/m2 (one-step PEN) but only to 0.52 MJ/m2 (three- 5 2.0 230 296 327 326 1.38
step PEN) with the second heating exposure during 5.2 s.
5.5.2. Residual stress distribution in the surface layer

5.4.2. Stainless steel The improvement of the fatigue properties of steels after the anode
Cathode PEN of austenitic stainless steel in the solution containing PEN is associated with the increase in their strength and formation of re-
7.5 wt.% ammonium carbonate permits high microhardness of the sur- sidual compressive stresses in the surface layer [6]. The observed stress
face layer to be obtained [65]. Higher concentration of ammonium car- state is caused by a combination of thermally induced plastic ow and
bonate did not lead to an increase in hardness due to a decrease in the structural (phase transformation) changes. These include high-speed
voltage required to maintain a certain temperature. The voltage de- heating and rapid cooling, formation of nitride phases with differential
crease results in the reduced velocity of the ions responsible for supply- volume and martensitic transformation. The residual compressive stress
ing nitrogen to the sample surface. The highest hardness was obtained is shown to always form in the surface layer when nitrided samples are
at a temperature of 550 C. The nitride layer thickness increases from cooled down in the electrolyte.
20 m (478 C) to 40 m (379 C) when the processing time is The shift in the maximum compressive stress to the depth of the
prolonged from 20 to 60 min. hardened layer is also observed after carburising followed by quenching
[75]. Such stress prole has a positive effect on the fatigue strength since
the fatigue cracks observed at the tests initiated in the subsurface layer.
5.5. Fatigue strength and residual stresses
A similar regularity is found in fatigue tests of the samples that were
treated by gas or plasma nitriding [74].
5.5.1. Endurance limit of nitrided medium carbon steels
The practical importance of nitriding follows from the pronounced
increase of fatigue strength, wear and corrosion resistance of medium
5.6. Tribological behaviour
carbon steels [63], which can be achieved by application of PEN. The
anode PEN followed by quenching promotes sufcient increases in the
5.6.1. Medium carbon steels and cast iron
endurance limit of samples with stress concentrators, which is in agree-
It was found that the PEN treatment could signicantly increase
ment with the experimental data obtained by other nitriding tech-
hardness and wear resistance of the cast iron and cast steel [69]. Cath-
niques [63]. Table 8 shows the examples of the anode PEN of medium
ode nitriding was carried out in the aqueous solution of carbamide at
carbon steel (0.35% C) samples, 10 mm in diameter with various stress
voltages of 200300 V, resulting in sample surface temperature of
concentrators:
about 400600 C. The PEN-treated cast iron (G3500) and steel
(S0050A) had about three times higher hardness than the untreated
1. V-shaped circumferential groove (d/D = 0.8; = 60; R = 0.1 mm);
samples of G3500 and S0050A. The treated cast iron and steel had a sim-
2. U-shaped circumferential groove (d/D = 0.8; R/d = 0.0625); ilar coefcient of frictions that was higher than those of original untreat-
3. Transverse hole (d/D = 0.3); ed samples due to increasing of surface roughness (Fig. 13). Lubrication
4. Keyway (b t l = 4 2.5 20 mm); signicantly reduced the friction coefcient of cast steel S0050A ( =
5. Groove for woodruff (b t = 3.5 3.5 mm), cutter diameter is 0.109) to the level of that of the cast iron G3500 ( = 0.105). Under the
13 mm. lubricated testing condition, the PEN treated cast iron and steel samples
The samples were saturated with nitrogen and quenched in the elec- had only slightly higher friction coefcient = 0.12, than the untreated
trolyte containing 11 wt.% ammonium chloride and 11 wt.% ammo- samples.
nium nitrate. The cast iron G3500 (Fig. 13a) showed a much higher wear resis-
tance than cast steel (Fig. 13c) since graphite in G3500 acted as a solid
Table 7 lubricant. Both of the treated samples G3500 and S0050A (Fig. 13b
Effects of surface layer thickness and microhardness on the impact toughness of medium and d) had an improved wear resistance compared to corresponding
carbon steel (0.45% C).
untreated substrates (Fig. 13a and c). The wear rates of nitrided samples
Second heating exposure Layer thickness Toughness decreased from 120 to 40 m3/(Nm) for S0050 steel and from 23 to
(s) (mm) HV0.05 (MJ/m2) 8 m3/(Nm) for G3500 cast iron. It should be mentioned that the sur-
Untreated 0 250300 1.11 face of cast iron G3500 has become rougher after the EPN treatment,
5.2 2.5 900950 0.52 the pin-on-disc test however seemed to make the surface of the wear
10.0 4.0 900950 0.29 track smoother rather than deeper. Moreover, the samples G3500 and
One-step PEN 5.0 900950 0.19
TG3500 exhibited very low wear rates at the dry testing condition.
Fig. 13. Appearance of wear tracks after pin-on-disc tribological tests. Untreated samples: (a) G3500 and (c) S0050A. PEN-treated samples: (b) G3500 and (d) S0050A [69].
Particularly at a lubricated testing condition, all the treated and untreat- out during 5 min. Fig. 14 shows that the friction coefcient of the nitrid-
ed samples appeared to have a negligible wear rate. ed samples decreases when the normal load increased. Hence, elastic
SEM micrographs of wear scars of the untreated and treated samples contact takes place during friction of nitrided samples. On the contrary,
after pin-on-disc dry tests against WC pins show a large fatigue cracking implantation of the counter-body in the untreated sample is accompa-
occurred on the wear track of untreated G3500. The cracks were initiat- nied by plastic deformation, which increases the area of real contact
ed at the edges of graphite akes by stress concentrations when the WC and facilitates development of adhesive wear. Consequently, the friction
pin counterface was rubbed against the G3500 sample. After 200 m of coefcient increases with increasing load. In this case, wear tracks and
sliding, only a small amount of localised peeling was observed. Although secondary structures are formed in the initial period of the test, which
cracks developed in the wear track, the cracked material was still at-
tached to it, which appeared to have a low wear rate when the wear
track was measured by a surface prolometer. However, this may
change after testing at longer sliding distances and the cracked material
would detach, causing severe wear. Interestingly, no fatigue cracking
but only slight wear tracks and asperity smoothing was observed on
the EPN-treated G3500 sample under the same dry testing condition.
Severe wear and plastic deformation were observed on the untreated
S0050A cast steel (Fig. 13c). The EPN treatment improved wear resis-
tance of the cast steel; however, the wear mechanism did not change.
In tests under lubricated conditions, cracking and localised peeling
still occurred in a relatively narrow wear track on the untreated cast
iron. Contrarily, no cracking was observed on the EPN-treated cast
iron. Surface polishing appeared on both the treated cast iron and cast
steel. Those observations implied that all treated and untreated samples
had similar low wear rates under the applied loads, lubrication and slid-
ing distances. However, a longer sliding distance could cause cast iron to
suffer from severe localised wear, also due to the fatigue cracking and
peeling.
The friction characteristics of these layers in dry pin-on-disc sliding
Fig. 14. Dependence of friction coefcient on the normal load. Nitriding time is 5 min,
against sintered TiC (hardness 2900 HV50) were studied [49]. The nor- temperature (C): 650 (3) and 750 (2, 4). 1 untreated sample. Solutions: 10%
mal load on the sample ranged from 0.5 to 4 MPa at an average sliding ammonium chloride and 5% ammonia (2, 3); 11% ammonium chloride and 11%
speed of 1.25 m/min. The PEN of steel (0.4% C and 1.0% Cr) was carried ammonium nitrate (4).
Table 9 (10 wt.%) and ammonia (5 wt.%) at 750 during 5 min resulted in
Regimes of anode PEN and tribological characteristics of medium carbon steel (0.45%C) the decrease in corrosion current density by 45 orders of magnitude
samples.
in neutral media due to passivating effects provided by a combination
Voltage Temperature Processing time Weight loss Friction of the oxide layer and the nitrided zone [81]. The nitrided samples
(V) () (s) (mg) coefcient were tested by full immersion in an aqueous solution of sodium sulfate
Untreated 24.80 0.26 (0.1 N) or their exposure above the surface of 3% sodium chloride solu-
80 900 300 21.60 0.20 tion [6]. The maximum corrosion potentials and minimum corrosion
150 1000 300 15.73 0.22
current densities were observed for steels nitrided at 750 C in the aque-
250 1200 300 8.70 0.24
225 1150 200 22.95 0.22 ous solution of 11 wt.% ammonium chloride and 11 wt.% ammonium ni-
225 1150 400 18.75 0.21 trate during 5 min, followed by air cooling [82].
225 1150 600 13.85 0.20 The maximum corrosion rate was observed during the rst few
225 1150 800 17.80 0.22 hours of the test; then it reduced and stabilised for all studied samples.
The steady-state corrosion rate of the nitrided steel was 3 times less
than the untreated one. After the oxide layer has been removed, the ni-
results in reduced linear wear rate. The wear is localised in the thin sur- trided steel showed an intermediate corrosion rate. After the tests, the
face layer owing to a combination of hard martensite with oxides and untreated samples were covered totally in a continuous brown layer
nitrides. The high hardness of martensite prevents the subsurface of corrosion products. The nitrided samples exhibited localised corro-
layer from plastic deformation. sion which can be explained by the presence of pores in the nitrided
Friction characteristics of the medium carbon steel samples (0.4% C, layer. The area affected by corrosion damage after a 96-hour test was
1% Cr) after their anode PEN treatments were compared with those of about 21% on the nitrided samples, 70% on those without oxide layer,
tool steel samples (1% C, 1.5% W, 1% Mn, 1% Cr) treated by thermal dif- and 100% on the untreated steel samples.
fusion boronising or thermal hardening. The anode PEN of other steels in the electrolyte comprising NH4Cl
The anode PEN followed by quenching in the solution of sodium ni- and NH4OH usually leads to increased corrosion resistance measured
trite (45 wt.%) shows similar results [76]. Wear tests were carried out by electrochemical impedance spectroscopy in solutions containing
under dry sliding conditions at the normal load of 1 MPa, sliding boric acid and borax [61].
speed of 15 m/min and sliding distance of 500 m, against hardened me- The corrosion resistance of nitrided steels can be increased by addi-
dium carbon steel (HRC 50) counterface. The friction coefcient was tional oxidation in an aqueous solution of sodium nitrite (1030 wt.%)
found to decrease when the normal load increased. Table 9 shows fur- [83], wherein the nal quenching could also be carried out. The resul-
ther details of the test results. tant 2- to 3-fold increase in the thickness of the oxide layer and its den-
sication, which improve signicantly the corrosion resistance of the
5.6.2. High-speed steels workpiece, can be explained by high oxidative capacity of NO 2 ions:
It is established that cathode PEN of high-speed steel R6M5 in the
carbamide solution results in the increase of abrasive wear resistance 2Fe NaNO2 2H2 OFe2 O3 NaOH: 17
[71]. The anode PEN followed by quenching also increases the wear re-
sistance of medium carbon steels [2]. This treatment creates the condi- It was established that the additional oxidation of the nitrided medi-
tions for localisation of contact stress in thin surface layers. A thin layer um carbon steel (0.45% C) in the solution of sodium nitrite reduces an-
of nitrides Fe23N and Fe4N with retained austenite is formed on the sur- odic dissolution current. When the anode PEN was carried out in an
face, a solid or carrier phase (martensite) is placed lower. The surface aqueous solution of NH4Cl (11 wt.%) and NH4NO3 (11 wt.%), the corro-
nitride zone is characterised by a sufcient plasticity and good wear-in. sion rate of samples in 0.05 M Na2SO4 solution reduced from 6.3 to
High hardness of martensite allows plastic deformation in subsurface 0.85 g/(m2day). The additional oxidation of steel parts after the
layers to be prevented. anode PEN is patented and recommended in situations, when enhance-
At the moment, the information on wear mechanisms of steels sub- ment in corrosion resistance of machine parts and tools is required [84].
jected to PEN treatments is rather limited. SEM studies of wear tracks
have been carried out only for cathode PEN treated samples of cast 5.7.2. Stainless steel
steel and cast iron tested at a xed normal load and sliding velocity. Ef- Corrosion behaviour of stainless steels (AISI 304, 316L and 430) in
fects of the loading on the friction coefcient under dry sliding condi- solutions of sodium chloride was improved signicantly by the cathode
tions have been established only for medium carbon steel treated by PEN, in particular the 304 grade with lower contents of alloying ele-
anode PEN followed by quenching. Tribological properties of nitrided ments [62]. Changes in the electrochemical response of nitrided AISI
high-speed steel have been studied only under conditions of abrasive 304 and 430 steels could be due to the spinel oxide sublayer formed
wear. Thus, different wear tests employed by various authors prevent in the outermost region of the passive layer, which improves the corro-
comparative analysis of data obtained be performed. sion resistance. The potentiodynamic polarisation curves for all steel
samples show that PEN shifts Ecorr towards higher potentials and re-
5.7. Corrosion resistance duces corresponding values of icorr. Accordingly, the samples' resistance
to pitting corrosion increases signicantly. Bode diagrams of nitrided
5.7.1. Medium carbon steels AISI 430 samples shows that higher PEN voltages cause corrosion resis-
It is known that corrosion properties of steels are inuenced by the tance to increase. In this case, the effect of PEN treatment lasts longer
possibility of formation and stability of passive surface layers [69]. Addi- than for other stainless steels, so that the corrosion mechanism remains
tional protection can be provided by the oxide layer thermally grown in the same for all treated samples, although with time most of the surface
special atmospheres containing oxygen or air [77,78]. This process is layer becomes damaged due to a partial dissolution in the electrolyte. By
recommended for manufacture of automobile engine valves made contrast, the untreated AISI 430 sample shows response characteristic
from high-temperature grades of steel [79]. It was found that dispersed of pitting corrosion.
inclusions of Fe2O3, along with excessive nitrides in the diffusion zone, Efcient sources of nitrogen are much scarcer than the carbon-
increase the microhardness of diffusion layer [80]. containing organic compounds suitable for PES. The most common
The positive role of protective oxide layers is also conrmed for the component is ammonia which provides a sufcient concentration of ni-
samples subjected to anode PEN treatments [2,3]. The treatment of me- trogen in the surface layer. However, the nitrogen generation capacity
dium carbon steels in the aqueous solution of ammonium chloride of base ammonia electrolytes is quite low. The search for new
nitrogen-containing compounds is highly challenging. It is also of inter- 97,98,105,107]. For example, it is suggested that monoethanolamine
est to supply the saturating components, in particular gaseous nitrogen decomposition can yield carbon monoxide, methane, hydrogen and hy-
[67] or powdered materials [85], directly in the processing zone. drogen cyanide according to the following reaction [97]:
Although high anticorrosion properties of iron nitride Fe23N are
well known, iron oxides may play a positive role in provision of corro- C2 H5 ONH2 CO HCN H2 CH4 18
sion resistance by PEN treatments too. This is conrmed by evaluation
of corrosion rates of nitrided samples with the outer oxide layer re- CH4 2H2 C 19
moved or further oxidation in air or in a solution of sodium nitrite per-
formed. In addition, alloying elements in stainless steels may also have 2COCO2 C 20
effects on corrosion behaviour of nitrided samples.
H2 CO2 CO H2 O 21
6. Nitrocarburising
2HCNH2 2C 2N: 22
Known electrolyte compositions for the nitrocarburising process and Thermal decomposition of triethanolamine and formamide is also
its results are shown in Tables 10 and 11. As a rule, appropriate electro- possible as follows [107]:
lytes contain components similar to those used for carburising or nitrid-
HCONH2 NH3 CO 23
ing individually.
Hypothetical reactions of decomposition of electrolyte components
as sources of nitrogen and carbon are presented in some articles [37, HCONH2 HCN H2 O 24
Table 10
Examples of the cathode PEN/C. Nomenclature: U voltage, T sample temperature, layer thickness, t treatment time, Ra surface roughness (in brackets initial roughness), dc
duty cycle.
Electrolyte composition U (V)/T Material and sizes sample t HV Ra (m) Research group
(C) (mm) (mm) (min) (load)
15% potassium hexacyanoferrate (II) in glycerol /1100 Steel (0.15% C) 0.4 5 920 Vanin et al. [86]
Tube 10 20
Wall is 1 mm
510% hydrochloric acid (or sodium carbonate), 1015% ammonia, /950 Malleable and ductile cast 0.06 1.5 850
1520% acetone iron
6 12
65% carbamide, 27% water 8% sodium carbonate 200/700 AISI 316L 0.02 Taheri et al. [87]
25 3
Carbamide in water /900 AISI H13 0.35% 0.16 15 Taheri et al. [88]
25 3
Borax and carbamide /900 H13 0.35% 0.16 15 1890 Taheri et al. [88]
25 3 (0.05)
Ethanolamine in water 150, Q235 0.05 0.75 750 Shen et al. [89]
50 Hz 8 6 1,5 0.07 1.25
0.09 2.00 730
Carbamide in water and other ingredients 230 AISI 316 0.5 HK0.01 0.003 Nie et al. [37]
40 20 2 400
230 1.0 450 0.013
250 0.5 550 0.004
250 0.063 1.0 1200 0.15
(0.002)
Formamide-ethanolamine aqueous solution 150 Q235 0.05 0.5 720 Li et al. [90]
(0.01)
6192% carbamide, 0.79% sodium carbonate, 730% water +300; DIN1.8509 (0.5%) 0.5 10 1250 Tavakoli et al. [91]
100; 20 5 (0.05)
5 kHz; dc
80%
Carbamide and sodium carbonate 240/715 AISI 1020 25 5 0.23 5 930 Zarchi et al. [92]
(0.025)
Carbamide and sodium carbonate 240260 316L 0.1 36 980 Aliofkhazraei et al.
20 5 (0.025) [93]
Acetamide, glycerol and sodium chloride 350; Cast iron 0.02 1.0 550 0.97 Pang et al. [94]
dc 50% 15 10 1 (0.02) (0.06)
100 Hz
Saturated solution of carbamide 220 AISI 1045 0.1 9 1280 0.49 Rastkar et al. [95]
20 2 (0.1) (0.03)
10% ammonium chloride, 3% sodium carbonate, carbamide 150/400 316L 0.065 10 783 5.6* Kazerooniy et al.
20 13 1 (0.05) (0.4) [96]
Monoethanolamine and potassium chloride 220 Q235 0.7 3 779 0.130.29 Jiang et al. [97]
50 10 2 (0.030.05)
Carbamide and ammonium salt in water 250 Q235 0.1 20 600 2.17 Jiang et al. [98]
15 5 3 (0.1) (0.11)
Carbamide in water 250 Steel 304 0.06 10 1380 0,14 Yerokhin et al. [99]
50 25 3 HK10 (0.025)
Carbamide 1 kg/l, potassium hydroxide 10 g/l 300 316L 30 550 Kumruoglu et al.
10 10 1.5 (0.01) [100]
Table 11
Examples of the anode PEN/C. Nomenclature: T sample temperature, layer thickness, t treatment time.
Electrolyte composition T Material and sizes sample (mm) t HV0.05 Research group
(C) (mm) (min)
510% nitric acid, 515% ammonium chloride, 1015% glycerol 850 Steel (0.1% C) 0.19 4 700 Lazarenko et al. [101]
20% ammonium chloride, 17% ammonium borftoride, 15% 860 Steel (0.4% C) 0.46 5 950 Duradji et al. [102]
glycerol 10 18
20% ammonium nitrate, 4.5% ammonia, 7% acetone 860 Steel (0.1% C) 0.24 5 750 Duradji et al. [103]
8 18
10% ammonium chloride, 20% carbamide 950 12Cr18Ni10Ti 0.1 5 450 Belkin et al. [104]
8 10
10% ammonium chloride, 20% carbamide 850 Steel (0.45% C) 0.5 10 600
10 10
10% carbamide, 10% ammonium chloride 850 Steel (0.2% C) hexagonal samples (height 15, facet 4) 0.1 5 700 Belkin et al. [58]
10% carbamide 10% ammonium chloride 850 Steel (0.2% C) 0.18 10 740 Kusmanov et al.
10 15 [105]
10% acetonitrile, 10% ammonium chloride 850 Steel (0.2% C) 0.18 10 740 Kusmanov et al.
10 15 [106]
C2 H4 OH3 N2CH4 3CO HCN 3H2 : 25 The anode PEN/C of low carbon steel leads to increase in ammonia
concentration in the electrolyte and decrease in that of carbamide
The most common component in electrolytes for plasma electrolytic (Fig. 15) [109]. Consequently, carbamide is decomposed in the adjacent
nitrocarburising (PEN/C) is carbamide. This component is shown to de- to VGE electrolyte region. The sample weight decreases at a rate of
compose to N2, CO2 and nitrate or nitrite anions by anodic oxidation 0.024 0.001 g/(cm2min). Accordingly, the concentration of iron
under condition of conventional electrolysis [108]. Therefore, carbam- ions in the solution is observed to increase.
ide is believed to decompose forming ammonia and carbon dioxide Apparently, ammonia as a product of carbamide decomposition is ac-
[37]. However, the chemical inertness of carbon dioxide gives grounds cumulated in the electrolyte faster than evaporated in the VGE wherein
for a doubt that this is a nal product in plasma-assisted electrolysis. the limiting pressure of ammonia is reached. Accumulation of ammonia
Carbon dioxide is suggested to decompose in strong electric elds but in the solution results in its buffering, which is conrmed by an increase
it should be noted that carbamide is an effective source of carbon in in pH from 5.5 to 7.7 during 250 min of PEN/C. During 300 min of PEN/
the case of anode PEN/C even with no visible electric discharges. Ac- C processing at 220 V, changes in the electrolyte composition result in sur-
cording to data of ultra violet spectroscopy, the groups of atoms present face temperature increasing from 830 to 880 C. In parallel, the current in-
in the electrolyte following processing point out to the formation of car- creases from 8 to 10 A, which indicates higher energy liberated in the
bon monoxide, atomic carbon and cyanides [98]. Carbamide may be VGE, probably due to its thinning. The reason for the VGE thinning is likely
decomposed in an aqueous solution at temperatures below 50 via to be due to the increase in surface tension at the electrolyte/VGE inter-
the formation of isocyanic acid [105]: face induced by the changes in electrolyte chemical composition.
After the PEN/C in the fresh electrolyte, the compound layer thick-
NH2 2 CONH3 HNCO 26 ness is 68 m, which is diminished to 60 m when the electrolyte served
for 30 min was used. On the contrary, the thickness of diffusion layer in-
HNCO H2 ONH3 CO2 : 27 creases from 93 m to 125 m due to the temperature rise at the same
rate of anode dissolution. It can be expected the nitrogen diffusion to de-
These reactions are conrmed by the detection of isocyanate-ions in crease and carbon to increase. The microhardness proles in the surface
the electrolyte. Nitrogen diffusion is associated with decomposition of layers produced during 300 min of electrolyte operation remain
the adsorbed ammonia and oxidation of isocyanic acid: unchanged.
Acetonitrile can be used for anode PEN/C as a source of nitrogen and
2NCO O2 2 O CO2 CO 2N Fe N : 28 carbon [106]. Acetonitrile and ethanol were found in the VGE during
PEN/C treatments at 850 by chromatography techniques. The data
obtained permit us to propose the following description of the transport
mechanism of saturating components. During PEN/C, acetonitrile is
evaporated in the VGE where it is adsorbed and subjected to thermal
decomposition to form atomic nitrogen and carbon on the anode sur-
face. Ethanol found in the VGE can be obtained as follows:
H H2 O
CH3 CN C2 H5 NH2 C2 H5 OH NH3 : 29
Ammonia is adsorbed on the anode and decomposed to atomic ni-

trogen. Ethanol may be subjected to both thermal decomposition on
the anode to atomic carbon and multistep oxidation:
C2 H5 OHCH3 CHO 2H 2e 30
Fig. 15. Change of electrolyte composition during PEN/C. Initial composition is 10%
ammonium chloride and 15% carbamide [109]. CH3 CHOCH3 CO H e 31
Fig. 16. Cross-sectional SEM micrographs of Q235 steel PES treated for different times: (a) 30 s, (b) 75 s and (c) 120 s [89].
CH3 COCO CH2 H e 32 or nitrides and perlite, after cooling in air:

inner diffusion layer containing solid solution of nitrogen and/or car-
bon, depending on the treatment temperature and cooling condi-
tions;
original pearliteferrite substrate structure.
CH2 C 2H 2e : 33
The iron oxide FeO and FeN0.05 phases were detected between 650
In this case, carbon monoxide and atomic carbon provide the sources and 850 C, the latter suggesting development of nitrogen diffusion
for carbon diffusion in the surface. into steel surface. Iron oxides Fe2O3 and Fe3O4 were detected be-
tween 750 and 950 C. The concentration of iron oxides increases
with treatment temperature, which is explained by the intensica-
6.2. Structure and phase composition of nitrocarburised layers
tion of high-temperature oxidation.
According to the results of EDX analysis, nitrogen concentration
6.2.1. Low carbon steel
reaches the maximum value underneath the oxide layer, decreasing
Surface layer composition and elemental distributions are similar to
further into the bulk of the sample (Fig. 18a). Kinetics of steel satu-
those observed following conventional thermochemical nitrocarburising
ration with nitrogen during PEN/C are virtually identical at all treat-
of steels. The cathode PEN/C of Q235 steel in all studied electrolytes re-
ment temperatures, except for 650 C, where an increase in nitrogen
sults in formation of iron carbides and nitrides.
concentration is only observed around the interface with the oxide
The modied layer contains Fe3C, Fe5C2, Fe4N, Fe23N, and Fe(C,N). The
layer.
compound layer formed during nitrocarburising treatment includes vari-
The carbon concentration in the diffusion layer increases with the
ous iron nitrides and/or carbonitrides such as Fe4N or Fe4(N,C) and Fe23N
saturation temperature (Fig. 18b). Pearlite structure is clearly
or Fe23(N,C). In some cases, martensite [92,97] and retained austenite
observed in the internal diffusion layer after PEN/C at 850 and
[89,90] can be found when processing temperature is adequate for hard-
950 C [105]. At 950 C the fragments of pearlite structure appear
ening. Fig. 16 shows the carbonitrided layer obtained in an aqueous solu-
tion of ethanolamine using pulsed voltage with magnitude and frequency
of 150 V and 50 Hz respectively [89]. It was found that the layer growth
rate decreased and the quality worsened after 75 s of treatment. For ex-
ample, when the treatment time increased to 120 s, the layer formed
had a thickness of only about 90 m, and a continuous network of car-
bides appeared in it. According to the results of EDS analysis, carbon con-
centration on the surface of the sample treated in an aqueous solution of
monoethanolamine and potassium chloride was 1.06 wt.% [97].
The surface modied by the anode PEN/C is characterised by the
oxide layer and reduced roughness. According to the XRD analysis, the
nitrocarburised surface layer on low carbon steel (0.2% ) contains
iron oxides FeO, -Fe2O3 [105], -Fe2O3 [106] and Fe3O4; carbonitride
-Fex(NC); nitrides FeN0.05, and Fe23N; carbide Fe4C; martensite and
austenite.
According to data of the nuclear backscattering of protons the oxygen
concentration in the surface layer of nitrocarburised low carbon steel
(0.1% C) increases signicantly when processing temperature rises up
to 950 C and reaches to 50 at.% at the depth of 5 m [104]. As usual, ni-
trogen diffusion dominates over that of carbon, at low temperatures.
According to SEM analysis, the surface microstructure of the treated
steel is composed of several layers (Fig. 17) [105]:
Fig. 17. Cross-sectional SEM image of steel surface after anode PEN/C treatment for 5 min
oxide layer containing iron oxides; in the electrolyte containing 15% (NH2)2CO, 10% NH4Cl, at 850 . 1 oxide layer, 2
outer nitrocarburised layer containing nitrides, martensite and external nitrocarburised layer, 3 internal diffusion layer, 4 initial pearlite-ferrite
retained austenite, when the sample is quenched in the electrolyte, structure [105].
dense appearance (while those treated at 220 V for 6 min and for the
longer time of 9 min contained some ne porosity dispersed in the
layers). The voltages around 220 V in plasma electrolytic process are
equal to temperatures in the region of 900 C and air quenching from
that temperature results in bainite or martensite structures.
The cathode DC PEN/C of nitriding steel DIN1.8509 in the electrolyte
containing carbamide and sodium carbonate leads to the formation of
iron oxides and nitrides in equal proportions, in the surface layer
[110]. When the carbamide concentration was above 90%, the Fe4N
and Fe23N nitrides are detected. The latter nitride phase dominates in
the surface layer when the PEN/C treatment was carried out in the elec-
trolyte containing N 84% carbamide, whereas the former prevails when
the electrolyte containing 6080% carbamide was used.
Nitrocarburising of nitriding steel DIN1.8509 using pulsed bipolar
plasma electrolytic techniques results in a submicron-size surface
layer [91]. The treatments were performed for 600 s in an electrolyte
composed of 85.8 wt.% urea, 12 wt.% water and 2.2 wt.% sodium carbon-
ate at various frequencies with duty cycle of 80%. It was found that the
treatment at 5000 Hz reduces the grain size in the compound layer to
110 nm. XRD analysis conrmed the presence of Fe3O4 phase on the sur-
face of the samples treated at lower frequencies and Fe4N rather than
Fe23 N nitride was found to be the dominant phase at higher frequen-
cies. The results also suggest that the size of the cavities formed on the
surface is reduced to 300 nm.
The inuence of the pulsed PEN/C on the phase composition can be
explained as following. In the case of DC mode, the gaseous envelope
surrounding the workpiece is stable. The stability of this envelope
causes a sudden increase in the workpiece temperature. However, if
the voltage regime is changed to pulsed bipolar, short-term voltage in-
terruptions may partially destroy the envelope, enabling brief contacts
with electrolyte that would temporarily reduce the workpiece temper-
ature. At lower frequencies, the resulting mean average temperature
still keeps increasing and the phase Fe23N becomes dominant. At
higher frequencies, there is less tendency for the temperature to rise
and the Fe4N phase remains prevailing.
6.2.3. Stainless steel

The cathode PEN/C of AISI 316L steel results in formation of oxides
and nitrides of iron and alloying elements. The sample microstructure
consists of an outer thin (23 m) magnetitechromite layer and an ex-
panded austenite diffusion layer which possesses ne recrystallised
grains [37]. Because of the relatively high temperature on the sample
surface and rapid quenching by the electrolyte after the power supply
is switched off, nitrogen and carbon diffused inward cannot precipitate
as grain-boundary nitrides or carbides and remain within the recrystal-
lised grains as a solid solution (expanded austenite). A lower, compared
to the sample surface, bulk temperature prevents recrystallisation (or
chromium carbide precipitation) in the metal substrate.
Fig. 18. EDX analysis of distribution of nitrogen (a), carbon (b) and oxygen (c) in the 6.2.4. Cast iron
surface layers of samples treated by anode PEN/C in a solution of carbamide (10%) and The PEC/N treatment was applied to cast iron using an aqueous solu-
ammonium chloride (10%) during 5 min for different treatment temperature [105]. tion of acetamide and glycerine as the electrolyte [94]. After the treat-
ment, the carbonitrided layer consisting of martensite, austenite, Fe2C,
throughout the modied surface layer. The temperature of 650 C Fe3C, Fe5C2, FeN and Fe23N was produced on the sample surface.
can be considered as a threshold below which no appreciable carbon
diffusion is observed under the studied conditions.
6.3. Mechanical properties
6.3.1. Low-carbon steel

6.2.2. Medium carbon steel The PEN/C of low carbon steels causes surface microhardness to in-
Nitrocarburising of medium carbon steel AISI 1045 in a saturated crease from 600 to 1000 HV (Tables 10 and 11). The higher values of mi-
aqueous solution of urea results in the formation of iron nitrides in the crohardness are observed for the medium carbon or stainless steel. The
surface layer [95]. The compound layer on the surface of the samples nitrocarburising of AISI 304 stainless steel with a diamond-like carbon
treated for a shorter time of 6 min and at voltages of up to 180 V had coating should also be noted [99].
6.3.2. Medium carbon steel 6.4.3. Stainless steels

It is shown that anodic PEN/ results in increase in tensile strength Considering the effects of PEN/C treatments in the carbamide con-
of the medium carbon steel from 360 20 to 1200 50 MPa [104]. taining electrolyte on the tribological performance AISI 316L steel [37],
Therefore, KCU impact tests were performed after the anodic PEN/C at it was noticed that despite of the sharp increase in surface roughness
the same conditions, yielding similar values of toughness as for the sam- from 0.002 to 0.15 m the average friction coefcient of the treated sam-
ples subjected to conventional hardening. ples was slightly lower than that of the untreated one. The sample treat-
ed at 250 V during 1 min demonstrated a much higher wear resistance
6.3.3. Stainless steel than both the untreated sample and those treated under different con-
Comparative fatigue-wear behaviour of untreated and PEN/C treated ditions, for which severe ploughing through the surface was observed.
AISI 316L austenitic stainless steel was investigated in [111] under dif- Additional increase in wear resistance of AISI 304 stainless steel can be
ferent combinations of cyclic loading and contact pressure. As a result, obtained using a diamond-like carbon coating on the PEN/C pre-
the PEN/C treated specimens exhibited a higher resistance (about 40% treated substrate [99]. This treatment results in a simultaneous reduc-
for 15.6 N contact load and about 60% for 25 N contact load) to simulta- tion of the friction coefcient and wear rate due to changes in the
neous cyclic stress and contact pressure. Also it was shown that under a wear mechanism from adhesion/abrasion to asperity deformation and
range of combined fatigue and wear loads, the specimens exhibit a bet- polishing. The primary wear mechanism observed is one of mild asper-
ter endurance than under wear or fatigue loads applied separately, with ity deformation and polishing and, in extreme cases, localised delamina-
this effect being signicantly stronger for the PEN/C-treated samples. In tion of the DLC coating. The volumetric wear rate was almost constant
these types of fatigue-wear tests, the surface plastic deformation in- for loads up to 15 N, particularly when sliding against WCCo, where
duced by the superposition of compressive and bending stresses material loss from the ball counterface is less severe than from the
work-hardens the surface region of the specimen. Also the favourable SAE 52100 steel ball, despite the higher contact pressure at an equiva-
residual compressive stresses in the surface layer may postpone or lent load. No failure of the DLC coating was observed for 10 and 15 N
even prevent the propagation of primary fatigue cracks. loads against WCCo and at up to 25 N against SAE 52100 steel [112].
Fatigue-wear behaviour of the PEN/C 316L steel was studied under a
6.4. Tribological behaviour range of combined fatigue and wear stresses [111]. This test was carried
out at the contact stress of 6.25 MPa and the bending stress of 87 MPa, in
6.4.1. Low-carbon steels Ringer's solution used as both a corrosive medium and a lubricant. It is
The PEN/C of steels increases in their wear resistance almost in all established that the induced surface plastic deformations due to the su-
cases. Fig. 19 shows wear behaviour of untreated low carbon steel perposition of compressive and bending stresses work-harden the sam-
(0.2% C) samples and those treated using the anode PEN/C in the elec- ple surface. Favourable residual compressive stresses in the surface
trolyte containing 1015% ammonium chloride, 15% carbamide [105]. layer may postpone or even prevent the propagation of primary fatigue
It can be seen that the nitrocarburised samples exhibited lower wear cracks. The weight loss at the rst stage (up to 16,000 cycles) for PEN/C-
rates and friction coefcient than the untreated one ( = 0.16 vs treated samples was signicantly lower than that of the untreated
0.19). Similarly, the cathode PEN/C of AISI 1020 steel in a urea electro- sample.
lyte shows that the wear properties of the nitrocarburised samples There are attempts to clarify wear mechanisms of PEN/C treated
were increased signicantly in relation to the untreated sample [92, steels. The cathode PEN/C of low-carbon steel was established to pro-
98]. It was concluded that the adhesive wear has developed on the orig- mote transition from adhesive to abrasive wear. Moderate abrasive/ad-
inal sample; on the contrary, the abrasive wear mainly appeared on the hesive wear has been observed in the case of cathode PEN/C of stainless
PEN/C treated one. steel. In general however, effects of loading and sliding speed on tribo-
logical behaviour of PEN/C treated steels have also been studied
6.4.2. Cast iron insufciently.
The cathode PEN/C of cast iron in an aqueous solution of acetamide
and glycerin results in the increase in the wear resistance [94]. It was 6.5. Corrosion resistance
shown that the tracks created on the original sample surface are wider
and deeper compared to those on the PEN/C treated sample. The wear 6.5.1. Low carbon steel
rate of the treated sample drops down 6.5 times compared to the un- The corrosion resistance of nitrocarburised samples is usually in-
treated sample. creased compared to the untreated substrate. The values of corrosion
current densities and corrosion potentials are often determined by
Tafel analysis (extrapolation of linear segments of anodic and cathodic
branches of polarisation curves towards their intersection in the semi-
logarithmic, E vs log I, plane). Disc specimens of AISI 1020 steel were
treated in the aqueous solution of urea and sodium carbonate [92].
The corrosion potential of the sample nitrocarburised for 5 min at
240 V (715 C) increased from 795 mV (for the untreated sample)
to 545 mV vs AgAgCl saturated electrode; corrosion current densi-
ties of the treated samples were also reduced.
Although Lee et al. [113] reported that the corrosion resistance of
AISI 1020 steel samples nitrocarburised at 220 and 240 V has been im-
proved in comparison with the untreated sample, a further increase in
this property is only possible using oxidation post-treatments. An ad-
herent Fe3O4 lm with 0.2 to 1 m thickness formed on the top of the
compound layer during the oxidation post-treatment, enhances the cor-
rosion resistance of other nitrocarburised low carbon steels [113,114].
Fig. 19. Wear behaviour of untreated substrate of low carbon steel (0.2% C) and samples
6.5.2. Medium carbon steels
treated in a solution of carbamide (15%) and ammonium chloride (10%) at 850 for The corrosion resistance of nitriding steel (DIN1.8509) can be en-
5 min [105]. hanced by the PEN/C in the solution containing urea (6192 wt.%),
sodium carbonates (0.79 wt.%) and water (730 wt.%) [91]. The EIS expanded austenite and a thin yet dense magnetite-based iron-
technique was used for corrosion evaluation. The samples were exposed chromium oxide layer [37]. Potentiodynamic polarisation curves for
to a solution of 3 wt.% NaCl for 1 h, 4 h, 2 days and 7 days. It was found the treated and untreated samples are shown in Fig. 20. The corrosion
that the corrosion resistance of samples processed in electrolytes with potential of 270 mV exhibited by the untreated substrate has in-
higher water contents is reduced. Developing a submicron-size surface creased 500 mV for the sample treated at 230 V during 30 s. Corrosion
layer on a nitriding steel using pulsed bipolar plasma electrolytic potentials of samples treated at 230 V (60 s) or 250 V (30 or 60 s) also
nitrocarburising results in improvement of the corrosion resistance increased, while both icorr and the limiting current of passive state sig-
[110]. All the treatments were performed for 600 s in the electrolyte nicantly decreased, providing signicant improvements in corrosion
containing 85.8 wt.% urea, 12 wt.% water and 2.2 wt.% sodium carbon- resistance. For the samples treated at 230 V (60 s) and 250 V (30 s),
ate. According to the results of EIS analysis, the sample corrosion resis- the layers of expanded austenite were signicantly thicker than that
tance increases with increasing pulse frequency in the PEN/C treatment. in the sample treated at 230 V during 30 s, resulting in increased corro-
Similar results were obtained for the anode PEN/C of medium carbon sion resistance. The sample treated for a longer time at the relatively
steel [51]. Corrosion tests were performed by immersion in decinormal high process temperature, showed the best performance in all corrosion
solution of sodium sulfate. The weight loss of specimens treated in the tests, which may be due to a combination of oxide layer and a thick (and
carbamide containing electrolyte for 3 and 5 min decreased more than ner-grained) diffusion layer of expanded austenite.
vefold and eightfold respectively; however, the decrease was only It is shown that addition of gaseous nitrogen during the PEN/C treat-
fourfold following the 10-min treatment. It can be assumed that the in- ment of AISI 316L steel in a saturated aqueous solution of carbamide and
crease in treatment time over 5 min impairs the integrity of the surface sodium carbonate can signicantly increase nitrogen diffusion rate and
oxide layer which plays an important role in metal passivation. The in- result in a higher corrosion resistance of the treated material [67].
crease in the saturation temperature to 950 signicantly increases Among the samples nitrocarburised in gas free conditions, only one
the thickness of the oxide layer, but this does not necessarily result in in- processed at 500 V has shown slight increases in both corrosion poten-
creased corrosive resistance. This can be attributed to the pores and tial and polarisation resistance compared to those for the untreated sub-
microcracks developed in the oxide layer after hardening. Thicker strate (Ecorr = 200 mV vs SCE; Rp = 126.5 k/cm2). In contrast, for
oxide layers obtained by cooling specimens after saturation in air can the samples treated with nitrogen purged through plasma, these char-
stratify. As a rule, the corrosion rate decreases which can be attributed acteristics increased signicantly following treatments in the range of
to the gradual lling of the pores with corrosion products. The mini- 600 to 700 V. The highest values of Ecorr = + 3 mV and Rp =
mum corrosion rate is observed after saturation with nitrogen and car- 296.5 k/cm2 were observed for the samples treated at 650 and 600 V
bon at 850 for 5 min. The decrease in corrosion rate was attained by respectively, which is probably due to the formation of dense iron-
an increase in saturation time from 3 to 5 min regardless of tempera- oxi-nitrocarburised layers on the surface under these conditions.
ture, which can be attributed to an increase in the thickness of the mod- Finally, results of rapid anodic hardening of austenitic stainless steel
ied layer. As suggested previously, the increase in corrosion rate after 12Cr18Ni10Ti, consisting in saturation with nitrogen and carbon in a
saturation for 10 min is associated with the increase in porosity of the chloridecarbamide electrolyte followed by quenching, are discussed
oxide layer. in [51]. The treatment at 850 for 10 min yields a hardened surface
layer of up to 80 m thick with microhardness of 450 HV50. Corrosion
6.5.3. Stainless steel tests revealed no evidence of intercrystalline corrosion along the grain
The corrosion resistance of the AISI 316L austenitic stainless steel boundaries, although a localised corrosion was observed, with depths
can be increased by N 50% using the PEN/C treatment in the electrolyte of corrosion spots approximately equal to their widths. These did not
containing urea and sodium carbonate [93]. It is established that the exceed 30 and 60 m for all specimens treated for 5 and 10 min respec-
percentage contributions of the applied voltage, treatment time, elec- tively. The average corrosion rates probably did not exceed the corre-
trolyte conductivity and urea concentration to the corrosion resistance sponding value for the untreated substrate.
are 78.24%, 19.05%, 2.41% and 0.3%, respectively. The optimised treat- Simultaneous saturation of steels with nitrogen and carbon has a
ment parameters for the maximum corrosion resistance include number of well-known advantages; consequently, carbonitriding and
1150 g/l for urea concentration, 360 mS/cm for electrolyte conductivity, nitrocarburising are being developed successfully. New possibilities
260 V for applied voltage, 6 min for treatment time. emerge for other multicomponent saturation processes, such as
Dramatic improvements in corrosion resistance by the PEN/C treat- borocarburising and borocarbonitriding [115,116]. It should be noted
ment may be due to the combination of thick, ne-grained layer of that electrolyte components added to adjust solution conductivity affect
sufciently intensity of electrical discharges and characteristics of oxide
layer formed on the surface, which determines surface roughness, wear
and corrosion resistance. Hence, careful selection of these compo-
nents can signicantly improve electrolyte properties and quality
of treatment.
The corrosion resistance of PEN/C treated steels has been paid more
attention to compared to nitriding or carburising, including pulse treat-
ment. Protective properties of PEN/C layers on steels may be enhanced
by grain renement in the porous layer. Increased pulse frequency
and duty cycle facilitate formation of nanocrystalline carbonitrides,
which improves corrosion resistance. In the case of anodic nitriding, a
positive role of iron oxides has also been noticed.
7. Future challenges
Further advancement of PES technology requires a better under-

standing of its physical features. Thermal potential of PES was studied
Fig. 20. Potentiodynamic polarisation curves in a 3% NaCl solution of untreated AISI 316
only for simplied steady-state conditions of a continuous vapour
steel sample and those subjected to PEN/C treatments under different conditions: (1) layer on vertical cylindrical surfaces. More complex, practically impor-
230 V, 30 s; (2) 230 V, 60 s; (3) 250 V, 30 s; (4) 250 V, 60 s. [37]. tant but theoretically challenging, systems with unsteady conditions
of heating and cooling are expected to be addressed. Particularly chal- also be sought to improve electrolyte properties, e.g. by adding compo-
lenging issues include modelling of heat transfer during PES treatments nents that reduce energy consumption, or using non-aqueous solutions.
of immersed at vertically or horizontally oriented surfaces as well as Finally, new counter-electrode designs, e.g. enabling purging plasma
those treated by electrolyte jets. supporting gases through the electrolyte ow directed to the sample
At present, PES treatments are not standardised, which makes it dif- [67], may provide new technological opportunities, including extension
cult to compare results obtained by different authors. Various hydro- of the range of saturating species. Moreover, the diffusion saturation can
dynamic conditions are used in the apparatuses for PES. Natural be facilitated using a supply of particulate precursor materials (boron
convection in a cylindrical vessel without electrolyte cooling [70] or carbide, aluminium or nickel) directly in the VGE [85].
cooling electrolyte in a jacketed vessel do not provide adequate control
over electrolyte temperature and therefore processing stability. Condi- 8. Conclusions
tions of forced convection with electrolyte stirring [4] or longitudinal
ow are more preferable. The latter provides both stable [39,41,53] Modern studies have led to the development of a number of electro-
and more uniform [31] temperature distribution over the treated sur- lyte compositions and processing regimes enabling hardness, wear and
face. Moreover, PETs can be implemented in stagnant electrolytes corrosion resistances of a wide range of steels to be enhanced using PES.
cooled by airow. The possibility of formation of surface nanostructures with promising
It is necessary to nd the optimum hydrodynamic conditions in the properties by pulsed cathodic PES was shown. One of the main advan-
electrolyser, which can provide a stable heating up to the temperatures tages of electrolytic plasma treatments, such as carburising, nitriding
practically suitable for saturation with interstitial elements. A unied or carbonitriding, is the possibility to combine PES with hardening by
electrolyser design would facilitate comparison between results of dif- means of interrupting polarisation and cooling in the same electrolyte
ferent studies, development and scaling up of technology and equip- without reheating. Continuous motion of the medium in the VGE pro-
ment. Studies into hydrodynamic conditions and regularities of heat vides rapid supply of saturating components to workpieces and removal
transfer for jet processing using perforated counter-electrodes are also of reaction products. Owing to small VGE thickness, high workpiece
important. heating rates (up to 100/s) can be achieved, which allows grain growth
Up to date, most of the results have been obtained on samples con- and associated deterioration of material's properties to be avoided. In
venient for analysis and testing. At the same time, processing features addition, small VGE thickness contributes to rapid stabilisation of pro-
of real components could be different because the shape of the working cessing conditions as the equilibrium concentration of adsorbed
electrode would inuence local hydrodynamic, thermal and chemical diffusant is achieved in several seconds. As a result, processing time is
conditions on the surface [16]. Systematic studies into these practically reduced to several minutes. Pulse PES treatments appear to be of partic-
important aspects of PES technology would therefore be of particular ular signicance as a means of obtaining nanoscale surface structures
interest. with promising properties.
Erosion of the workpiece, its oxidation and dissolution should be in- Surface modication can be carried out by jetting electrolyte onto
vestigated more deeply because they determine important characteris- the workpiece surface as well as by conventional immersion technique.
tics of state surface, crucial for its tribological behaviour. The thickness The jetting scheme opens up new technological opportunities, including
and microstructure of the surface oxide layer inuence diffusion localised processing, treatments of inner surfaces and tailoring in ser-
characteristics of carbon and nitrogen, thus affecting structure and com- vice performance by processing separate areas of the component
position corresponding diffusion layers. Furthermore, iron oxides con- under different conditions.
tribute to the corrosion resistance of steels, under certain conditions. It has been established that the stationary heating mode is analogous
In the case of the cathode PES, electrical discharges are more important to the lm boiling in liquids distinguished by the presence of internal
than the oxidation process, but they may cause surface erosion, which heat sources in the VGE. These sources are caused by energy libera-
characteristics depend also on electrolyte composition. This inuence tion associated with electrical discharges or ion emission from the
should be studied too. electrolyte.
Pulse PES has already demonstrated certain advantages, but further The mechanism of nitrogen and carbon liberation from the electro-
research of this method is required. Effects of pulse parameters (ampli- lyte has been proposed. The saturating components are evaporated
tude, frequency, and duty cycle) on the wear and corrosion resistance from the electrolyte to VGE where they are oxidised, decomposed, sub-
were studied. However, the formation of nanostructures is explained jected to electro-oxidation to low-molecular products, and dissociated
by different hypotheses which require conrmation. to form active atomic carbon and nitrogen species. It was shown that
There is also theoretical and practical interest in studies of electro- carbon and nitrogen diffusion is affected by the surface oxide layer
chemical processes that always occur in the studied systems and deter- formed in the case of anode PES. The layer thickness is determined by
mine many aspects of PET technology. These include conditions of anode oxidation and dissolution processes that are inuenced by elec-
current ow across the VGE, liberation of saturating components, ad- trolyte composition.
sorption of active species, and their diffusion into the bulk of the work- These results convincingly demonstrate signicant potential of PES
ing electrode. A limited number of publications addresses transport technology for scientic research and various industrial applications.
mechanisms of saturating components from the electrolyte to the VGE
and further into the treated material. There are only assumptions Acknowledgements
about possible reactions in the system without sufcient experimental
conrmation. It is desirable to investigate chemical and electrochemical The research leading to these results has received funding from the
reactions in the system electrolyteVGEmetal electrode, nature of Russian Science Foundation (Contract #15-13-10018) (Sections 1, 3, 5,
charge carrier in the VGE, and transport mechanisms of nitrogen, car- 6, 7) and the European Research Council under the ERC Advanced
bon, boron and oxygen from the electrolyte to the VGE and further to Grant (#320879 IMPUNEP) (Sections 2, 4, 8).
the modied material.
The obtained results make it possible to choose the electrolyte com-
References
position with regard to its saturating capacity, service life and energy
consumption. Attention should be paid to the electrolyte components [1] A.L. Yerokhin, X. Nie, A. Leyland, A. Matthews, S.J. Dowey, Plasma electrolysis for
that determine both solution conductivity and properties of the treated surface engineering, Surf. Coat. Technol. 122 (1999) 7393.
[2] P.N. Belkin, V.I. Ganchar, A.D. Davydov, A.I. Dikusar, E.A. Pasinkovskii, Anodic
surfaces. A range of such components should be expanded to enable ef- heating in aqueous solutions of electrolytes and its use for treating metal surfaces,
cient optimisation of electrolyte composition. Opportunities should Surf. Eng. Appl. Electrochem. (2) (1997) 115.
[3] P.N. Belkin, Anode electrochemical thermal modication of metals and alloys, Surf. [38] I.G. Dyakov, A.R. Naumov, Electrochemical reactions on anodic heating in aqueous
Eng. Appl. Electrochem. 46 (6) (2010) 558569. ammonium-chloride electrolytes, Surf. Eng. Appl. Electrochem. (6) (2006) 14.
[4] M. Aliofkhazraei, A. Sabour Rouhaghdam, P. Gupta, Nano-fabrication by cathodic [39] M. Tarakci, K. Korkmaz, Y. Gencer, M. Usta, Plasma electrolytic surface carburized
plasma electrolysis, Crit. Rev. Solid State Mater. Sci. 36 (2011) 174190, http:// and hardening of pure iron, Surf. Coat. Technol. 199 (23) (2005) 205212.
dx.doi.org/10.1080/10408436.2011.593269. [40] Pat. 44-1049 Japan, (1234, 1231, 1232). Solution for Electrolytic Thermal
[5] P. Gupta, G. Tenhundfeld, E.O. Daigle, D. Ryabkov, Electrolytic plasma technology: Treatment/Inoue Kiyoshi, Kaneko Hideo; 18.01.1979.
science and engineeringan overview, Surf. Coat. Technol. 201 (21) (2007) [41] F. avulu, M. Usta, Kinetics and mechanical study of plasma electrolytic carburiz-
87468760. ing for pure iron, Appl. Surf. Sci. 257 (9) (2011) 40144020.
[6] P.N. Belkin, E.A. Pasinkovskij, Heat treatment and case hardening of steels subject- [42] A. Rezaei, A. Shokuhtar, M. Asadi, M.M. Hosseinzadeh, S. Ahmadi, Carburizing of
ed to heat in electrolytic solutions, Met. Sci. Heat Treat. 31 (56) (1989) 331337. low carbon steel with plasma electrolysis in aqueous solution, Defect Diffus.
[7] H.H. Kellogg, Anode effect in the aqueous electrolyses, J. Electrochem. Soc. 97 (4) Forum 258260 (2006) 2631, http://dx.doi.org/10.4028/www.scienticnet/DDF.
(1950) 133142. 258-260.26.
[8] S.K. Sengupta, O.P. Singh, Contact glow discharge electrolysis: a study of its onset [43] M. Yaghmazadeh, C. Dehghanian, Hardening of AISI H13 steel using pulsed plasma
and location, J. Electroanal. Chem. 301 (1991) 189197. electrolytic carburizing (PPEC), Plasma Process. Polym. 6 (2009) S168S172.
[9] T. Mizuno, T. Ohmori, T. Akimoto, A. Takahashi, Production of heat during plasma [44] J. Wu, W. Xue, B. Wang, X. Jin, J. Du, Y. Li, Characterization of carburized layer on T8
electrolysis in liquid, Jpn. J. Appl. Phys. 39 (2000) 60556061. steel fabricated by cathodic plasma electrolysis, Surf. Coat. Technol. 245 (2014) 915.
[10] T. Paulmier, J.M. Bell, P.M. Fredericks, Development of a novel cathodic plasma/ [45] M. Skakov, E. Batyrbekov, L. Zhurerova, M. Schefer, Microstructure and micro-
electrolytic deposition technique. Part 2: physico-chemical analysis of the plasma hardness changes of 30CrMnSiA steel modied surface layers by electrolyte-plas-
discharge, J. Mater. Process. Technol. 208 (2008) 117123. ma processing, Appl. Mech. Mater. 423-426 (2013) 824827.
[11] S.Y. Shadrin, P.N. Belkin, Analysis of models for calculation of temperature of anode [46] S. Kurbanbekov, M. Skakov, M. Schefer, A. Naltaev, Changes of mechanical prop-
plasma electrolytic heating, Int. J. Heat Mass Transf. 55 (2012) 179186. erties of steel 12Cr18Ni10Ti after electrolytic-plasma cementation, Adv. Mater.
[12] E.I. Meletis, X. Nie, F.L. Wang, J.C. Jiang, Electrolytic plasma processing for cleaning Res. 601 (2013) 5963.
and metal-coating of steel surface, Surf. Coat. Technol. 150 (2002) 246256. [47] W. Xue, Q. Jin, Q. Zhu, X. Wu, Characterization of plasma electrolytic carburized
[13] A. Yerokhin, A. Pilkington, A. Matthews, Pulse current plasma assisted electrolytic stainless steel in glycerin aqueous solution, J. Aeronaut. Mater. 30 (4) (2010)
cleaning of AISI 4340 steel, J. Mater. Process. Technol. 210 (2010) 5463. 3842.
[14] Y.A. Koshelev, G.A. Kitaev, E.I. Bazhin, Use of thyristor converters in the electrolytic [48] W. Xue, J. Qian, R. Liu, X. Jin, B. Wang, J. Du, Inuence of glycerin concentration on
cleaning of continuous wide strip, Surf. Eng. Appl. Electrochem. (6) (1984) 9395. plasma electrolytic saturation process of stainless steel surface, Chin. J. Nonferrous
[15] V.M. Rudim, Electrolytic descaling technology for strip steel, Surf. Eng. Appl. Metals 23 (3) (2013) 882887.
Electrochem. (5) (1986) 100104. [49] P. Belkin, A. Naumov, S. Shadrin, I. Dyakov, A. Zhirov, S. Kusmanov, T. Mukhacheva,
[16] A.Y. Zanin, Y.I. Nazarenko, M.I. Serdyuk, Heating a continuously moving circular Anodic plasma electrolytic saturation of steels by carbon and nitrogen, Adv. Mater.
plate in an electrolyte plasmotron, Surf. Eng. Appl. Electrochem. (3) (1984) 3638. Res. 704 (2013) 3742.
[17] Y.N. Tyurin, A.D. Pogrebnjak, Electric heating using a liquid electrode, Surf. Coat. [50] A.V. Zhirov, A.O. Komarov, V.V. Danilov, S.A. Shorokhov, Effect of glycerine concen-
Technol. 142144 (2001) 293299. tration on dissolution and oxidation of mild steel during anodic cementation, Surf.
[18] J. Garbarz-Olivier, C. Guilpin, Etude des discharges electriques produites entre Eng. Appl. Electrochem. 48 (3) (2012) 289291.
l'electrode et la solution lors des effects d'anode et de cathode dans les electrolytes [51] P.N. Belkin, I.G. Dyakov, A.V. Zhirov, S.A. Kusmanov, T.L. Mukhacheva, Effect of
aqueux, J. Chim. Phys. 72 (2) (1975) 207214. compositions of active electrolytes on properties of anodic carburization, Prot.
[19] P.N. Belkin, V.I. Ganchar, Y.N. Petrov, A study of vapor lm conduction at anodic Met. Phys. Chem. 46 (6) (2010) 715720.
electrolytic heating, Sov. Phys. Dokl. 31 (1986) 10011004. [52] V.N. Duradji, A.M. Mokrova, T.S. Lavrova, Carbon distribution in steel that
[20] P.N. Belkin, V.I. Ganchar, Passage of a current through a vaporgas sheath during underwent case hardening in electrolytic plasma, Surf. Eng. Appl. Electrochem.
anodic electrolytic heating, Surf. Eng. Appl. Electrochem. (5) (1988) 97102. (5) (1984) 4346.
[21] S.O. Shiryaeva, A.I. Grigor'ev, V.V. Morozov, On the appearance of ions near the [53] S.A. Kusmanov, S.Y. Shadrin, P.N. Belkin, Carbon transfer from aqueous electrolytes
charged surface of an intensely evaporating electrolyte, Tech. Phys. 48 (7) to steel by anode plasma electrolytic carburizing, Surf. Coat. Technol. 258 (2014)
(2003) 822828. 727733.
[22] S.Y. Shadrin, A.V. Zhirov, P.N. Belkin, Formation regularities of gaseous vapour plas- [54] A.B. Belikhov, P.N. Belkin, Features of iron-graphite anode carburising, Surf. Eng.
ma envelope in electrolyser, Surf. Eng. Appl. Electrochem. (1) (2016) 110116. Appl. Electrochem. (6) (1998) 917.
[23] T. Mizuno, T. Akimoto, K. Azumi, T. Ohmori, Y. Aoki, A. Takahashi, Hydrogen evolu- [55] J. Wu, W. Xue, X. Jin, B. Wang, J. Du, Z. Wu, Preparation and characterization of
tion by plasma electrolysis in aqueous solution, Jpn. J. Appl. Phys. 44 (1A) (2005) diamond-like carbon/oxides composite lm on carbon steel by cathodic plasma
396401. electrolysis, Appl. Phys. Lett. 103 (2013) 031905, http://dx.doi.org/10.1063/1.
[24] I.G. D'yakov, V.S. Belkin, S.Y. Shadrin, P.N. Belkin, Peculiarities of heat transfer at an- 4813830.
odic plasma electrolytic treatment of cylindrical pieces, Surf. Eng. Appl. [56] A.O. Komarov, P.N. Belkin, The effect of surfactants on characteristics of anodic ce-
Electrochem. 50 (4) (2014) 346355. mentation of structural steels, Powder Metall. Funct. Coat. (2) (2008) 4649.
[25] P.N. Belkin, A.B. Belikhov, Stationary temperature of an anode heated in aqueous [57] Pat. 3,098,151 USA (Cl. 21971). Electrical Discharge of Metals in Electrolytes/
electrolytes, J. Eng. Phys. Thermophys. 75 (6) (2002) 12711277. Inoue Kiyoshi, 16.07.1963.
[26] P.N. Belkin, T.L. Mukhacheva, I.G. Dyakov, Features of the distribution of heat uxes [58] P. Belkin, S. Kusmanov, A. Naumov, Y. Parkaeva, Anodic plasma electrolytic
in the anodevaporgas sheath system in anodic electrolytic heating, J. Eng. Phys. nitrocarburizing of low-carbon steel, Adv. Mater. Res. 704 (2013) 3136.
Thermophys. 81 (6) (2008) 10691075. [59] P. Belkin, I. Dyakov, S. Kusmanov, A. Naumov, Composition and properties of struc-
[27] A.V. Zhirov, S.Y. Shadrin, Determination of heat ux from vaporgas envelope into tural steel surface after plasma electrolytic carburising, Proceedings of the Interna-
short anode heated by plasma electrolysis, 6th International Conference on Mate- tional Conference BALTRIB, Kaunas 2011, pp. 184188.
rial Science and Condensed Matter Physics, Chisinau 2012, p. 279. [60] E.P. Grishina, A.V. Zhirov, P.N. Belkin, A.I. Dikusar, Inuence of anodic
[28] I.G. Dyakov, S.Y. Shadrin, P.N. Belkin, Features of anode heating in movement of electrothermochemical oxidation on the corrosion stability of steel 45, Surf. Eng.
electrolyte in a free convection regime, Surf. Eng. Appl. Electrochem. (4) (2004) Appl. Electrochem. 44 (5) (2008) 390395.
712. [61] V. Andrei, G. Vlaicu, M. Fulger, C. Ducu, C. Diaconu, G. Oncioiu, E. Andrei, M. Bahrim,
[29] V.I. Ganchar, I.M. Zgardan, A.I. Dikusar, Anodic dissolution of iron in the process of A. Gheboianu, Chemical and structural modications induced in structural mate-
electrolytic heating, Surf. Eng. Appl. Electrochem. (4) (1994) 6977. rials by electrochemical processes, Rom. Rep. Phys. 61 (1) (2009) 95104.
[30] I.G. Dyakov, P.N. Belkin, Thickness of vapour/gas envelope during anode heating of [62] M.K. Aliev, A. Sabour, P. Taheri, Study of corrosion protection of different stainless
vertically immersed cylinder, Surf. Eng. Appl. Electrochem. (4) (2002) 4451. steels by nanocrystalline plasma electrolysis, Prot. Met. Phys. Chem. 44 (4) (2008)
[31] Pat. 124 682 U1 RU, (C23D 1/00). Device for Electrolyte Treatment of Metal Prod- 402407.
ucts, A.O. Komarov, T.L. Mukhacheva, P.N. Belkin. 07.03.2012. [63] E.J. Mittemeijer, Fundamentals of nitriding and nitrocarburising. ASM Handbook,
[32] A.O. Komarov, T.L. Mukhacheva, I.G. Dyakov, P.N. Belkin, Inuence of hydrodynam- Volume 4A, Steel Heat Treating Fundamentals and Processes, 2013. (http://www.
ical peculiarities of electrolyte ows on temperature of cylindrical workpiece by is.mpg.de/14343245/Mittemeijer_ASM_Handbook_4A_2013.pdf)
plasma electrolysis, Surf. Eng. Appl. Electrochem. 48 (2) (2012) 141186. [64] A. Roy, R.K. Tewari, R.C. Sharma, R. Sherhar, Feasibility study of aqueous electrolyte
[33] I.G. Dyakov, S.A. Kusmanov, A.R. Naumov, Dynamics of 12X18H10T steel and VT1-0 plasma nitriding, Surf. Eng. 23 (4) (2007) 243246.
alloy dissolving in the anodic electrolytic heating, Book of the Abstracts of the In- [65] A. Roy, Aqueous Electrolyte Plasma Nitriding of Austenitic Stainless Steel(MT the-
ternational Conference Dedicated to the 50th Anniversary from the Foundation sis) Indian Institute of Technology, Kanpur, DC, 2006.
of the Institute of the Chemistry of the Academy of Sciences of Moldova, Chisinau [66] S.A. Kusmanov, A.A. Smirnov, Y.V. Kusmanova, P.N. Belkin, Anode plasma electro-
2009, p. 232. lytic nitrohardening of medium carbon steel, Surf. Coat. Technol. 269 (2015)
[34] V.I. Ganchar, I.M. Zgardan, A.I. Dicusar, Anodic dissolution of chromium during 308313.
electrolytic heating, Surf. Eng. Appl. Electrochem. (5) (1996) 1319. [67] L. Cenk Kumruoglu, A. Yerokhin, A. Ozel, A. Matthews, Effect of nitrogen gas addi-
[35] I.M. Zgardan, V.I. Ganchar, A.I. Dikusar, Abnormal anodic dissolution of copper tion onto the process of plasma electrolytic nitrocarburising of AISI 316L stainless
under electrolysis heating conditions, Russ. J. Electrochem. 35 (4) (1999) 494495. steel, Proceed. 1st ISTS International Surface Treatment Symposium, Istanbul
[36] S.A. Kusmanov, P.N. Belkin, I.G. D'yakov, A.V. Zhirov, T.L. Mukhacheva, A.R. 2011, pp. 295310.
Naumov, Inuence of oxide layer on carbon diffusion during anode plasma electro- [68] J. Li, D. Shen, Z. Tian, Y. Wang, K. Liu, Hot working technology (2006)-01, http://en.
lytic carburizing, Prot. Met. Phys. Chem. Surf. 50 (2) (2014) 223229. cnki.com.cn/Article_en/CJFDTOTAL-SJGY200601012.htm.
[37] X. Nie, C. Tsotsos, A. Wilson, A.L. Yerokhin, A. Leyland, A. Matthews, Characteristics [69] X. Nie, L. Wang, Z.C. Yao, L. Zhang, F. Cheng, Sliding wear behaviour of electrolytic
of a plasma electrolytic nitrocarburising treatment for stainless steels, Surf. Coat. plasma nitrided cast iron and steel, Surf. Coat. Technol. 200 (56) (2005)
Technol. 139 (23) (2001) 135142. 17451750.
[70] J.H. Kong, M. Okumiya, Y. Tsunekawa, T. Takeda, K.Y. Yun, M. Yoshida, S.G. Kim, [95] A.R. Rastkar, B. Shokri, Surface modication and wear test of carbon steel by plas-
Surface modication of SCM420 steel by plasma electrolytic treatment, Surf. ma electrolytic nitrocarburizing, Surf. Interface Anal. 44 (2012) 342351, http://dx.
Coat. Technol. 232 (2013) 275282. doi.org/10.1002/sia.3808.
[71] M. Skakov, B. Rakhadilov, M. Scheffner, G. Karipbaeva, M. Rakhadilov, Electrolyte [96] N. Afsar Kazerooniy, M.E. Bahrololoom, M.H. Shariat, F. Mahzoon, T. Jozaghi, Effect
plasma nitriding of high-speed steel, Appl. Mech. Mater. 379 (2013) 161166. of Ringer's solution on wear and friction of stainless steel 316L after plasma elec-
[72] Pat. 621799 USSR, (C23C 9/12). Electrolyte for Anode Nitriding of Steel Products. trolytic nitrocarburising at low voltages, J. Mater. Sci. Technol. 27 (10) (2011)
P.N. Belkin, A.A. Factorovich, E.A. Pasinkovski. 09.03.1977. 906912.
[73] P.N. Belkin, A.A. Burbelko, E.A. Pasinkovski, A.V. Rabinovich, A.A. Faktorovich, Ni- [97] Y.F. Jiang, Y.F. Bao, K. Yang, Effect of C/N concentration uctuation on formation of
triding kinetics of technical iron and steel 40Kh under electrolytic heating condi- plasma electrolytic carbonitriding coating on Q235, J. Iron Steel Res. Int. 19 (11)
tions, Surf. Eng. Appl. Electrochem. (2) (1984) 6870. (2012) 3945.
[74] J.M. Cowling, J.W. Martin, Effect of internal residual stresses on the fatique behavior [98] Y.F. Jiang, T. Geng, Y.F. Bao, Y. Zhu, Electrolyteelectrode interface and surface char-
of nitrided En41b steel, Heat. Treat. 79 Proc. Int. Conf. Birmingham 1980, acterization of plasma electrolytic nitrocarburizing, Surf. Coat. Technol. 216 (2013)
pp. 178181. 232236.
[75] L.J. Ebert, The role of residual stresses in the mechanical performance of case car- [99] A.L. Yerokhin, A. Leyland, C. Tsotsos, A.D. Wilson, X. Nie, A. Matthews, Duplex sur-
burized steels, Metall. Mater. Trans. A 98 (11) (1978) 15371551. face treatments combining plasma electrolytic nitrocarburising and plasma-
[76] S.E. Kuzenkov, B.P. Saushkin, Borating of steel 45 in electrolytic plasma, Surf. Eng. immersion ion-assisted deposition, Surf. Coat. Technol. 142144 (2001)
Appl. Electrochem. (6) (1996) 1015. 11291136.
[77] Pat. 4,881,983 USA, C23C 8/34; C23C 11/08. Manufacture of Corrosion Resistant [100] L.C. Kumruoglu, A. Ozel, Plasma electrolytic saturation of 316L stainless steel in an
Components, John D. Smith, Stephen E. Vanes. aqueous electrolyte containing urea and ammonium nitrate, Mater. Technol. 47 (3)
[78] Pat. 4,131,492 USA, C23C 11/16. Steel Article Having a Nitrided and Partly Oxidizes (2013) 307310.
Surface and Method for Producing Same, Shingi Fushimi, Takeshi Miyata. [101] Pat. 461 161 USSR, (C23C 9/10). Method of Thermochemical Treatment of Metals,
[79] S. Fuji, S. Fushimi, M. Hashimoto, Analysis of the surface of the heat-resistant steel B.R. Lazarenko, V.N. Duradji, A.A. Factorovich, I.V. Bryantsev. 25.02.1975.
subjected oxy nitriding, J. Jpn. Soc. Heat Treat. 22 (2) (1982) 105110. [102] Pat. 922 177 USSR, (C23C 9/16). Electrolyte for Treatment of Steel Products, V.N.
[80] Z. Li, D. Yu, H. Pau, D. Wang, Mechanism study of increasing the depth and hard- Duradji, A.K. Tovarkov. 23.04.1982.
ness of diffusion layer on steel after its liquid nitriding in urea, J. Zhejiang Univ. [103] Pat. 618 447 USSR, (C23C 9/16). Electrolyte for Nitrocarburising of Steel Products,
(2) (1982) 91104. V.N. Duradji, I.V. Bryantsev, A.K. Tovarkov. 05.08.1978.
[81] V.G. Revenko, V.V. Parshutin, G.P. Chernova, N.L. Bogdashkina, N.D. Tomashov, P.N. [104] P. Belkin, B. Krit, I. Dyakov, V. Vostrikov, T. Mukhacheva, Anode saturation with ni-
Belkin, E.A. Pasinkovskii, A.A. Faktorovich, Inuence of nitriding during electrolytic trogen and carbon in aqueous solution of carbamide-bearing electrolytes, Met. Sci.
heating on electrochemical and corrosion behaviour of steel St 45, Elektron. Obrab. Heat Treat. 52 (12) (2010) 2024.
Mater. (5) (1985) 5659. [105] S.A. Kusmanov, Y.V. Kusmanova, A.R. Naumov, P.N. Belkin, Features of anode plas-
[82] P.N. Belkin, S.A. Kusmanov, A.A. Smirnov, Plasma electrolytic hardening and ma electrolytic nitrocarburising of low carbon steel, Surf. Coat. Technol. 272 (2015)
nitrohardening of medium carbon steels, Mater. Sci. Forum 844 (2016) 146152. 149157.
[83] V.V. Parshutin, E.A. Pasinkovslii, Promotion of steel corrosion resistance by chemi- [106] S.A. Kusmanov, Y.V. Kusmanova, A.R. Naumov, P.N. Belkin, Formation of diffusion
calthermal treatment in electrolytes, Surf. Eng. Appl. Electrochem. 43 (6) (2007) layers by anode plasma electrolytic nitrocarburising of low carbon steel, J. Mater.
431433. Eng. Perform. 24 (8) (2015) 31873193.
[84] Patent 2959 Moldova, 23C 20/00, 23C 20/06, 23C 20/08. Method of Steel Parts [107] X.-M. Li, Y. Han, Porous nanocrystalline Ti(CxN1 x) thick lms by plasma electro-
Treatment for Formation of Surface Layer With High Corrosion Resistance. V. lytic carbonitriding, Electrochem. Commun. 8 (2006) 267272.
Parshutin, E. Pasincovschi. 2006.01.31. [108] N.V. Osetrova, A.M. Skundin, Products of anodic oxidation of carbamide: effect of
[85] Pat. 870 486 USSR, (C23C 9/00). Method of Thermochemical Treatment of Metals anionic composition of solution, Russ. J. Electrochem. 38 (3) (2002) 266269.
and Alloys. A.K. Tovarkov, V.N. Duradji. 07.10.1981. [109] S.A. Kusmanov, Y.V. Parkaeva, I.S. Frolov, A.R. Naumov, P.N. Belkin, Effect of electro-
[86] V.S. Vanin, G.A. Semenova, Cyaniding of steel by heating in an electrolyte, Met. Sci. lyte depletion on characteristics of anodic plasma electrolytic nitrocarburizing,
Heat Treat. 7 (10) (1965) 670672. Surf. Eng. Appl. Electrochem. 51 (3) (2015) 213219.
[87] P. Taheri, C. Dehghanian, A phenomenological model of nanocrystalline coating [110] H. Tavakoli, S.M. Mousavi Khoie, S.P.H. Marashi, O. Bolhasani, Effect of electrolyte
production using plasma electrolytic saturation (PES) technique, Trans. B Mech. composition on characteristics of plasma electrolysis nitrocarburizing, J. Mater.
Eng. 16 (1) (2009) 8791. Eng. Perform. 22 (8) (2013) 2351.
[88] P. Taheri, C. Dehghanian, M. Aliofkhazraei, A.S. Rouhaghdam, Nanocrystalline [111] F. Mahzoon, S.A. Behgozin, M.E. Bahrololoom, S. Javadpour, Study the fatigue-wear
structure produced by complex surface treatments: plasma electrolytic behavior of a plasma electrolytic nitrocarburized (PEN/C) 316L stainless steel, J.
nitrocarburizing, boronitriding, borocarburizing, and borocarbonitriding, Plasma Mater. Eng. Perform. 21 (8) (2012) 17511756.
Process. Polym. 4 (2007) S721S727, http://dx.doi.org/10.1002/ppap.200731805. [112] C. Tsotsos, A.L. Yerokhin, A.D. Wilson, A. Leyland, A. Matthews, Tribological evalu-
[89] D.J. Shen, Y.L. Wang, P. Nash, G.Z. Xing, A novel method of surface modication for ation of AISI 304 stainless steel duplex treated by plasma electrolytic
steel by plasma electrolysis carbonitriding, Mater. Sci. Eng. A 458 (2007) 240243. nitrocarburising and diamond-like carbon coating, Wear 253 (910) (2002)
[90] J. Li, D. Shen, Y. Wang, K. Liu, Research on the formation conditions of plasma elec- 986993.
trolytic carbonitriding in liquid, Chin. Surf. Eng. 18 (4) (2005) 3133. [113] K.H. Lee, K.S. Nam, P.W. Shin, D.Y. Lee, Y.S. Song, Effect of post-oxidizing time on
[91] H. Tavakoli, S.M. Mousavi Khoie, S.P.H. Marashi, S.A. Hosseini Mogadam, Character- corrosion properties of plasma nitrocarburized AISI 1020 steel, Mater. Lett. 57
ization of submicron-size layer produced by pulsed bipolar plasma electrolytic (2003) 20602065.
carbonitriding, J. Alloys Compd. 583 (2014) 382389. [114] T. Bell, Y. Sun, A. Suhadi, Environmental and technical aspects of plasma
[92] M.K. Zarchi, M.H. Shariat, S.A. Dehghan, S. Solhjoo, Characterization of nitrocarburising, Vacuum 59 (2000) 1423.
nitrocarburized surface layer on AISI 1020 steel by electrolytic plasma processing [115] B. Wang, W. Xue, J. Wu, X. Jin, M. Hu, Z. Wu, Characterization of surface hardened
in an urea electrolyte, J. Mater. Res. Technol. 2 (3) (2013) 213220 (http://www. layers on Q235 low-carbon steel treated by plasma electrolytic borocarburizing, J.
sciencedirect.com/science/article/pii/S2238785413000574). Alloys Compd. 578 (2013) 162169.
[93] M. Aliofkhazraei, P. Taheri, A. Sabour Rouhaghdam, C. Dehghanian, Systematic [116] P. Taheri, C. Dehghanian, M. Aliofkhazraei, A.S. Rouhaghdam, Nanocrystalline
study of nanocrystalline plasma electrolytic nitrocarburising of 316L austenitic structure produced by complex surface treatments: plasma electrolytic
stainless steel for corrosion protection, J. Mater. Sci. Technol. 23 (5) (2007) nitrocarburizing, boronitriding, borocarburizing, and borocarbonitriding, Plasma
665671. Process. Polym. (4) (2007) S721S727.
[94] H. Pang, G.-L. Zhang, X.Q. Wang, G.-H. Lv, H. Chen, S.-Z. Yang, Mechanical perfor-
mances of carbonitriding lms on cast iron by plasma electrolytic carbonitriding,
Chin. Phys. Lett. 28 (11) (2011) 118103.

Belkin 2016

Enviado por

Dados do documento

Direitos autorais

Formatos disponíveis

Compartilhar este documento

Compartilhar ou incorporar documento

Opções de compartilhamento

Você considera este documento útil?

Este conteúdo é inapropriado?

Direitos autorais:

Formatos disponíveis

Belkin 2016

Enviado por

Direitos autorais:

Formatos disponíveis

SCT-21269; No of Pages 25

Surface & Coatings Technology xxx (2016) xxxxxx

Contents lists available at ScienceDirect

Surface & Coatings Technology

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

Plasma electrolytic saturation of steels with nitrogen and carbon

4.5. Corrosion resistance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0

1. Introduction In this article, we review examples of PES of steels. In Section 2, we

3. Electrochemical features of plasma electrolysis

Anode dissolution and high-temperature oxidation of the sample

insufciently. Meanwhile, these processes affect very important proper-

4.1. Electrolyte compositions

This reaction cannot proceed in one-step, it is likely to be a combina- The C2H+

Electrolyte composition U T Material and sample sizes (mm) t HV Ra Research group

4.2.1. Low-carbon steels

where (x) is the nal carbon concentration at depth x, 0 is the initial

4.3.2. High-carbon steel

Carbon containing component

T () Acetone Glycerol Sucrose Ethylene glycol

Table 4 In other circumstances, roughness rise after cathode carburising

Nomenclature: U voltage, T sample temperature, layer thickness, t treatment time

patterns of plasma treated austenitic stainless steel show diffraction

5.3. Nitrogen diffusion kinetics

Under the stationary PEN conditions, the thickness of the compound

electrolyte, although the maximum concentration of nitrogen on the

PEN temperature is 750 C

5.5.2. Residual stress distribution in the surface layer

Ammonia is adsorbed on the anode and decomposed to atomic ni-

CH3 COCO CH2 H e 32 or nitrides and perlite, after cooling in air:

6.2.3. Stainless steel

6.3.1. Low-carbon steel

6.3.2. Medium carbon steel 6.4.3. Stainless steels

Further advancement of PES technology requires a better under-

Você também pode gostar