J. Mod. Phys., 2010, 1, 151-157
doi:10.4236/jmp.2010.13022 Published Online August 2010 (http://www.scirp.org/journal/jmp)
Effect of Kind Rolling on the Properties of Plasma-Formed
Nitride Layers on Fe93Ni4Ti3
Fayez Mahomoud El-Hossary1, Sayed Mohammed Khalil1,2, Magdy Abdel Wahab Kassem3,
Khaled Lotfy1
1Physics Department, Faculty of Science, Sohag University, Sohag, Egypt
2Physics Department, University of Colorado at Boulder, Boulder, USA
3Department of Materials and Metallurgical Engineering, Faculty of Petroleum and Mining Engineering,
Suez Canal University, Suez, Egypt
Received April 15, 2010; revised May 21, 2010; accepted June 10, 2010
Abstract
The nitriding behavior of cold rolled Fe93Ni4Ti3 specimens was compared with that of hot rolled specimens
of the same materials. Radio frequency (rf) nitriding was performed for 10 minutes in a 10-2 mbar nitrogen
atmosphere. The continuous plasma power was varied from 300-550 W in steps 50 W or less. Results of op-
tical microscopy (OM), x-ray diffraction (XRD) and microhardness measurements (Hv) are presented and
discussed with regard to the influence of kind rolling on the nitriding behavior, particularly nitride formation
and nitride layer growth on mechanical properties. The results show a remarkable increase of nitrogen diffu-
sivity and microhardness of cold rolled nitride samples. These best results may be attributed to enhancement
of the defect and/or a compressive stress.
Keywords: Cold Rolled, Hot Rolled-Radio Frequency (Rf), Microhardness, Optical Microscopy (OM),
X-Ray Diffraction (XRD)
1. Introduction
Surface plasma nitriding of steels is a well-established
technique to produce a modified (hard) surface layer on
steels or other iron based alloys with good anti-corrosive
and wear resistance properties. Both the nitrogen case
depth and the resulting nitride type commonly depend on
the nitriding conditions as well as on the properties of
material, such as composition, crystallographic structure
and the density of various lattice defects. In particular, the
thickness and the composition of the nitrided layer are
strongly affected by the type of chemical reactions oc-
curring at the specimen’s surface as well as by the diffu-
sivity of nitrogen in the treated material.
Nitrogen diffusivity in materials depends on many fa-
ctors including lattice structure, grain size, chemical com-
position and defect density. In ultrafine-grained materials,
which can be fabricated by consolidation of nanopow-
ders [1,2], sever plastic deformation [3], hot rolling [4]
and cold rolling [5], the increased grain boundary area
and the dislocation density will promote the diffusion of
alloying elements and nitrogen. Therefore, it can be ex-
pected that during a thermochemical treatment, e.g. pla-
sma nitriding, such materials will develop a large thick-
ness of nitriding layer. Ferkel et al. [6] reported that
sever deformation of X5CrNi1810 steel rf nitrided at
350°C can enhance the nitrogen diffusion. This enhance-
ment may be attributed to an increased grain boundary
area and dislocation density. In the present investigation,
an attempt has been made to evaluate the effect of kind
rolling prior to rf plasma nitriding mainly on the proper-
ties of Fe93Ni4Ti3 samples. After rf plasma nitriding, the
surface microhardness, nitrogen diffusion layer thickness,
surface morphology and formed phases have been ana-
lyzed using a Vickers microhardness tester, optical mi-
croscopy (OM) and x-ray diffraction (XRD).
2. Experimental Procedure
The material studied was Fe93Ni4Ti3 iron-based alloy.
Cupon-shaped specimens 1 cm × 2 cm in diameter and 2
mm thick. The cupons were prepared by arc-melting pure
metals with nominal purities of 99.99 wt.% in an induc-
tion furnace under an argon atmosphere (99.9999% pu-
rity). The produced castings were heated at 900ºC for
two hours. The heated specimens were hot rolled with
about 10% reduction each time at 800°C up to 2mm
thickness. After the initial hot rolling, these specimens

F. M. E. HOSSARY ET AL. 152
were divided into to two groups. One was rolled at room
temperature (cold rolled) and the other was rolled at
900°C (hot rolled). Nitriding treatment of specimens was
carried out using a radio frequency (rf) inductively cou-
pled glow discharge, with a continuous mode of opera-
tion at 13.56 MHz. The nitriding system consists of a
quartz reactor tube with 500 mm in length and 41.5 mm
in diameter evacuated by a two-stage rotary pump to a
base pressure of 10-2 mbarr. Iron-based alloy samples
were supported on copper sample holder fitted which
were equipped with water cooling pipes. Nitrogen (N2)
was introduced to establish a gas pressure of 8-8.4 × 10-2
mbar, measured with a pirani gauge. The distance be-
tween the sample holder surface and the rf coil was 2.9
cm and the water cooling rate for samples was 1500
cm3/min. The discharge is generated by an induction
copper coil energized by rf power generator (type HFS
2.5 KW, 13.65 MHz) via a tunable matching network.
The surface morphologies and cross-sectional micro-
graphs were examined using an optical microscope. The
phase compositions of the surface region of the nitrided
layer were studied by X-ray diffraction (XRD) using
CuKα radiation in the Ө-2Ө gemoetry. The microhard-
ness was measured by using a Wetzlar microhardness
tester with the load of 0.98 N (100 g).
3. Results and Discussion
3.1. Cross-Section Analysis by Optical
Microscope
An optical micrograph (OM) cross-section study was
employed to determine the thickness of the nitrided
cross-section layer (compound layer) of cold and hot (at
900°C) rolled Fe93Ni4Ti3 samples. The typical cross sec-
tion views of nitrided samples treated for 10 minutes
using different input plasma power levels of 300, 350,
400, 425, 450, 475, 500, 525 and 550 W are shown in
Figure 1 (for cold) and in Figure 2 (for hot at 900°C)
rolled treatments, respectively. From Figures 1(a) and (b)
and 2(a) and (b), it can be seen that in both types of roll-
ing, the two samples which were nitrided at 300 and 350
W do not contain any observed compound layer. This
behavior can be attributed to low plasma processing
power (from 300 W to 350 W) where the mobility of ni-
(a) (b) (c)
(d) (e) (f)
(g) (h) (i)
53μm
Figure 1. Optical micrographs of compound layer thickness of cold rolled samples Fe93Ni4Ti3 treated at various plasma proc-
essing power input: (a) 300 W; (b) 350 W; (c) 400 W; (d) 425 W; (e) 450 W; (f) 475 W; (g) 500 W; (h) 525 W; (i) 550 W.

F. M. E. HOSSARY ET AL. 153
(a) (b) (c)
(d) (e) (f)
(g) (h) (i)
49μm
Figure 2. Optical micrographs of compound layer thickness of hot rolled samples Fe93Ni4Ti3 at 900ºC treated at various
plasma processing power input. (a) 300 W; (b) 350 W; (c) 400 W; (d) 425 W; (e) 450 W; (f) 475 W; (g) 500 W; (h) 525 W and
(i) 550 W.
trogen atoms is low and penetration is done more easily-
through grain boundaries [7]. When the plasma power
input is increases with increasing the input plasma power
from 400 W to 550 W for cold and hot rolled samples,
respectively. The variation range of the layer thickness is
15 μm to 79 μm for cold rolled while from 11 μm to 56
μm for hot rolled at 900°C. This enhancement of thick-
ness for the cold rolled samples is probably due to the
promotion of nitrogen ionization, leading to high con-
centrations of high-energy ions supplied onto the speci-
men. However, this result agrees with Mahboubi et al.
[7]. Also, Ferkel et al. [8] reported that, the cold high
pressure torsion (HPT) processed material shows a
thicker and more homogeneous compound layer than the
material not subjected to HPT; the nitrogen uptake is
largest in the HPT-processed material.
3.2. Compound Layer Thickness
The variation of the compound layer thickness of cold
rolled Fe93Ni4Ti3 and hot rolled Fe93Ni4Ti3 at 900°C ni-
trided samples for different input plasma power are
shown in the Figures 3 and 4 respectively. However,
these values of thickness are measured from the cross
section morphology. From these figures, for both the
kinds of rolling, one can see that the thickness increases
continuously as the plasma power increases. The en-
hancement of thickness is probably due to the domina-
tion of lattice and the penetration of nitrogen atoms
through the grains, which all enhance the formation of a
more uniform compound layer [8].
3.3. Surface Morphology
Figure 5 shows typical OM micrograph of the surface
features of untreated and treated cold rolled Fe93Ni4Ti3
samples for different plasma processing power. The mi-
crograph of the untreated sample is shown in Figure 5(a).
This figure appears to be a relatively coarse-grained
structure composed of iron α–phase. The micrographs of
treated samples at 300 W and 350 W are shown in Fig-
ures 5(b) and (c), which reveal no observed change in

F. M. E. HOSSARY ET AL. 154
Figure 3. Thickness of compound layer as a function of
plasma power for cold rolled samples of Fe93Ni4Ti3.
Figure 4. Thickness of compound layer as a function of
plasma power for hot rolled samples of Fe93Ni4Ti3 at 900ºC.
(a) (b) (c)
(d) (e) (f)
(g) (h) (i)
53μm
(j)
Figure 5. Optical micrographs of the surface morphology of cold rolled samples Fe93Ni4Ti3 for plasma processing power input
(a) 0.0 W; (b) 300 W; (c) 350 W; (d) 400 W; (e) 425 W; (f) 450 W; (g) 475 W; (h) 500 W; (i) 525 W; (j) 550 W.
m)
m)

F. M. E. HOSSARY ET AL. 155
the microstructure with respect to the native sample. It
means that, this range of plasma processing power is not
adequate to form a clear compound layer. Figures 5(d-j)
shows a different microstructure with respect to the above
samples as plasma power increases from 400 W to 550
W. This microstructure is finer and finer in scale, as a
result of the increased nitrogen solubility and diffusivity
produced by the rf plasma process. This observation
agrees with Pantazopoulos et al. [9], in which they re-
ported that the liquid nitrocarburised of cold work steel
exhibits a very different microstructure with respect to
native sample.
3.4. Phase Analysis by XRD Diffraction
The X-ray diffraction patterns with CuKα radiation was
used to clarify the effect of kind rolling on the formed
phases in the original and nitrided samples at different
input plasma processing powers. Figures 6(a) and 7(a)
show the diffraction patterns of the original samples of
cold and hot (at 900°C) rolled Fe93Ni4Ti3, respectively.
These figures reveal that all intense peaks are assigned to
α-Fe-phase. The same diffractgrams can be seen in Fig-
ures 6(b) and 7(b) for the nitrided samples of cold and
hot rolled Fe93Ni4Ti3 respectively, when the plasma
power input is increased to 350 W. When plasma power
input was increased to 450 W and 550 W for cold ni-
trided samples, the most intense peaks are assigned to ε-
Fe2-3N and γ′-Fe4N phases while the peaks of α-phase are
disappeared as shown in Figures 6(c-d). It is worth men-
tioning that hcp ε-nitride exhibits a higher hardness than
fcc- γ′-nitride [10]. Figure 7(c) shows the diffraction
peaks of the hot nitrided sample when plasma power
input was adjusted to 450W. From this figure, only
ε-phase can be detected while the peaks of α and γ′ phas-
es have disappeared. When plasma power input reached
Figure 6. X-ray diffraction patterns for Fe93Ni4Ti3 samples
(a) untreated cold rolled and treated for plasma processing
powers; (b) 350 W; (c) 450 W; (d) 550 W.
Figure 7. X-ray diffraction patterns for Fe93Ni4Ti3 samples
(a) untreated hot rolled at 900ºC and treated for plasma
processing powers; (b) 350 W; (c) 450 W; (d) 550 W.
550W for the same sample in Figure 7(d), the intense
peaks of γ′–phases emerge, while the peaks of ε-phase
decrease significantly. From the above results, the fol-
lowing observations are made: 1) In both the types of
rolled samples, the peaks of NiN and TiN cannot be de-
tected. 2) The peak ratio of ε-phase to γ′-phase increases
by increasing plasma power input from 450W to 550W
for cold rolled samples, while it decreases at the same
condition for hot rolled samples. It can be concluded that
this is the reason for continuous increase in the micro-
hardness for the cold rolled treated samples after nitrided
for 450W, while the microhardness abruptly decreases
for the hot rolled treated samples after nitrided for the
same power (see the next paragraph). 3) The influence of
the Ni present in the samples has not been considered,
because the interaction of Ni with N that is even weaker
than the interaction between N and Fe. However, these
observations agree well with Chezan et al. [11], in which
they reported that the severe deformation caused by cold
rolling for Fe-Ni-Ti and Fe-Ni-Cr leads to formation of
large density nucleation sites for new phases, i.e. γ′ and ε.
3.5. Microhardness Measurements
In order to clarify the influence of kind rolling on the
mechanical properties of nitrided samples, microhard-
ness measurements of nitrided cold rolled and hot rolled
(at 900°C) Fe93Ni4Ti3 samples were performed at a load
of P = 100 g (0.89N).
Figure 8 displays the variation of microhardness val-
-ues of untreated and nitrided cold rolled samples Fe93
Ni4-Ti3 against the input plasma power. From this figure,
one can notice that the microhardness value of the ni-
trided samples at 300, 350W equals nearly the same value
of untreated sample 297 HV. This trend is probably due
to the fact that at 350 W and below, there is not enough rf
power to create precipitation, which would lead to sig-
nificant hard phases. This result is confirmed by XRD
analysis and OM observations. When plasma power in

F. M. E. HOSSARY ET AL. 156
Figure 8. Microhardness values of untreated and treated
cold rolled Fe93Ni4Ti3 samples for different plasma power
input.
put increases from 400 to 550W, the microhardness
value increases exponentially to reach the value of 2098
HV. This represents a 7-fold increase in the surface-
hardness comparing with the untreated sample (282 HV).
The anomalously high microhardness values are due to
the fact that high defect density in nitrided samples offers
enough nucleation sites for nitrides and enough diffusion
paths for nitrogen. In this case, a dense compound layer
with high hardness can easily form [6]. Also, the dra-
matic increase of hardness is a result of compressive
stresses, which is induced in the target surface by the
cold rolling and a compound layer produced on the target
surface by nitriding.
Figure 9 depicts the variation of microhardness values
of untreated and nitrided hot rolled (at 900ºC) samples
Fe93Ni4Ti3 against the input plasma power. From this
figure, one can notice that the microhardness values of
the nitrided samples at 300 W and 350 W equal nearly
the same value of untreated sample 274 HV. As plasma
power input increases from 400 W to 450 W, the micro-
hardness value increases continuously to reach the value
of 572 HV; this represents a 2-fold increase in the sur-
face hardness comparing with untreated sample. This
trend may be credited to the formation of a compound
layer thickness of 11 µm to 56 µm, which contains a
high concentration of nitrogen [12]. Similar behavior has
been observed by Devi and Mohanty [13], in which a
microhardness value of 1478 HV in hot rolled D2 steel
was achieved after plasma nitriding at 510ºC for 18 h.
With the high input plasma power of 475 W or more, the
microhardness value decreases. This reduction can be
attributed to the decrease the ratio of ε-phase to γ'-phase
[10]. However, this result agrees well with XRD analy-
sis.
In light of the microhardness measurements, it can be
concluded that a general hardness increase for cold rolled
treated samples can be associated with enhancement of
Figure 9. Microhardness values of untreated and treated
hot rolled Fe93Ni4Ti3 at 900°C samples for different plasma
power input.
defect density and/or a compressive stress. i.e., the hard-
ness of the plasma nitrided sample shows a kind of roll-
ing dependent behavior.
4. Conclusions
The thickness, hardness and phase composition of the
modified layer formed on cold and hot rolled nitrided
Fe93Ni4Ti3 are investigated in this work. All of the sam-
ples were hardened by forming nitrided layers. The ad-
vantage of the pre-cold rolling is that a thicker nitrided
layer with higher hardness is formed. The cold rolling
process is a well-known technique applied to improve
the mechanical properties of steel surfaces. Through this
process, a large number of grain boundaries, dislocations
and positive holes form through the surface of specimens,
which all enhance the nitrogen diffusion. The surface
hardness of nitrided cold-rolled Fe93Ni4Ti3 samples re-
presents a 3.7-fold increase with respect to the nitrided
hot-rolled samples. This trend may be credited to the
high defect density of nitrided cold rolled samples, which
offers enough nucleation sites for nitrides and enough
diffusion paths for nitrogen [12]. The modified surface of
nitrided cold rolled samples is characterized with a finer
microstructure. The phase analysis dictates that the peak
ratio of ε to γ′ phase increases by increasing plasma
power input from 450 to 550 W for nitrided cold rolled
samples, while it decreases at the same conditions for
nitrided hot-rolled samples. This is the reason behind the
continuous increase in the microhardness for the cold
rolled treated samples after application of 450 W plasma
power, with a decrease for the hot rolled treated samples
after being nitrided for the same power. The most strik-
ing results of this work are the increased thickness of the
nitrided cold rolled samples. Indeed, diffusion of nitro-

F. M. E. HOSSARY ET AL. 157
gen by the microcracks mechanism could be expected to
be significantly affected by an increase in grain boundary
area as a result of cold rolling. However, the experimen-
tal evidence demonstrates that the microstructure of cold
rolled material does have a significant influence on ni-
trogen diffusion.
5. Acknowledgments
The authors would like to thank Prof. T. Munsat at the
University of Colorado for his help in preparing this
manuscript.
6. References
[1] F. M. El-Hossary, “The Influence of Surface Microcracks
and Temperature Gradients on the Rf Plasma Nitriding
Rate,” Surface and Coatings Technology, Vol. 150, No. 2,
2002, pp. 277-281.
[2] F. M. EL-Hossary, N. Z. Negm, S. M. Khalil, A. M. Abd
Elrahman and D. N. Mcllroy, “RF Plasma Carbonitriding
of AISI Austenitic Stainless Steel,” Surface and Coatings
Technology, Vol. 141, No. 2-3, 2001, pp. 194-201.
[3] C. B. Craus, A. R. Chezan, N. G. Chechenin, D. O.
Boerma and L. Niesen, “Soft Magnetism in Nitrided
Fe93Ni4Cr3 and Fe94Ni4Ti2 Cold–Rolled Alloys,” Journal
of Magnetism and Magnetic Materials, Vol. 263, No. 1-2,
2003, pp. 47-56.
[4] G. Edrem and Y. Taptik, “Effect of Hot Rolling Condi-
tions to Produce Deep Drawing Quality Steels for Con-
tinuous Annealing Process,” Journal of Material Proc-
essing Technology, Vol. 170, No. 1-2, 2005, pp. 17- 23.
[5] H. Katoh, H. Takachi, N. Takahashi and M. Abe, “Cold
Rolled Steel Sheets Produced by Continuous Annealing,
Technology Continuously Annealed Cold-Rolled Steel,”
1984, pp. 37-38.
[6] H. Ferkel, M. Glatzer, Y. Estrin and R. Z. Valiev, “RF
Plasma Nitriding of Severely Deformed High Alloyed
Steel,” Scripta Material Science, Vol. 46, No. 9, 2002, pp.
623-628.
[7] F. Mahboubi and K. Abdolvahabi, “The Effect of Tem-
perature on Plasma Nitriding Behavior of DIN 1.6959
Low Alloy Steel,” Vacuum, Vol. 81, No. 3, 2006, pp.
239-243.
[8] H. Ferkel, M. Glatzer, Y. Estrin, R. Z. Valiev, C. Blawert
and B. L. Mordike, “RF Plasma Nitriding of Severely
Deformed Iron-Based Alloys,” Materials Science and
Engineering A, Vol. 348, No. 1-2, 2003, pp. 100-110.
[9] G. Pantazopoulos, P. Psyllaki, D. Kanakis, S. Antoniou,
K. Papadimitriou and J. Sideris, “Tribological Properties
of a Liquid Nitrocarburised Special Purpose Cold Work
Tool Steel,” Surface and Coatings Technology, Vol. 200,
No. 20-21, 2006, pp. 5889-5895.
[10] H. Ferkel, Y. Estrin, C. Blawert and R. Z. Valiev, “RF
Nitriding of Severely Deformed Armco Iron and St2
K50,” Surface and Coatings Technology, Vol. 173-174,
2003, pp. 1164-1170.
[11] A. R. Chezan, C. B. Craus, N. G. Chechenin, T. Vystavel,
L. Niesen, J. T. M. De Hosson and D. O. Boerma, “On
the Formation of Ultra-Fine Grained Fe-Base Alloys Via
Phase Transformations,” Materials Science and Engi-
neering A, Vol. 367, No. 1-2, 2004, pp. 176-184.
[12] F. M. El-Hossary, S. M. Khalil, K. Lotfy and M. A.
Kassem, “The Influence of Hot-Rolled Temperature on
Plasma Nitriding Behavior of Iron-Based Alloys,” Jour-
nal of Low Temperature Physics, Vol. 156, No. 1-2, 2009,
pp. 38-47.
[13] M. U. Devi and O. N. Mohanty, “Plasma-Nitriding of
Tool Steels for Combined Percussive Impact and Rolling
Fatigue Wear Applications,” Surface and Coatings Tech-
nology, Vol. 107, No. 1, 1998, pp. 55-64.