Wear 390–391 (2017) 113–124

Contents lists available at ScienceDirect

Wear
journal homepage: www.elsevier.com/locate/wear

Effect of manganese, silicon and chromium additions on microstructure and MARK

wear characteristics of grey cast iron for sugar industries applications
⁎
E.E.T. ELSawyc, , M.R. EL-Hebearya, I.S.E. El Mahallawib
a
Mechanical Design and Production Engineering Department, Faculty of Engineering, Cairo University, Egypt
b
Department of Metallurgical Engineering, Faculty of Engineering, Cairo University, Egypt
c
Faculty of Engineering, Cairo University, Egypt

A R T I C L E I N F O A B S T R A C T

Keywords: Mill roller shells are amongst the most important manufacturing-line elements in the sugar production process.
Wear resistance They are carefully designed according to many surface criteria amongst which wear resistance is the most
Mill roller shells important. To increase the life of the mill roller shells, the resistance of the surface to failure by abrasion should
Microstructure be increased. In this investigation, a series of sixty-six cast iron samples with varied contents of chromium,
Grey cast iron
manganese and silicon were prepared. The microstructure and the mechanical properties were studied. Hence,
the wear of all the specimens was evaluated using the pin-on-disc test, with a high-speed steel disc. The accu-
mulated weight loss was measured and the wear rates were determined at constant normal test load of 20 N and
constant pin velocity of 1.28 m/s during the pin-on-disc test. The accumulated weight loss was measured as a
function of the number of revolutions. From test results, it was found that the wear rate, for all conditions under
investigation, decreases with the increase of manganese and chromium additions at constant silicon additions.
The wear rate obtained for the samples containing 3% manganese and 2.02% silicon was equal to 1.2267 mg/s,
and the minimum wear rate was obtained for the samples containing 3% manganese, 2.11% silicon, and 2.11%
chromium (equal to 0.3400 mg/s). The enhancement in the wear resistance of the chromium alloyed samples
was attributed to the increase in the chromium carbides volume fraction.

1. Introduction
Mill roller shells are usually constructed of a shaft of steel [1] onto
which a cast iron shell is shrunk [2]. Sugar Mill roller shells are im-
portant surface engineered insert elements, which are manufactured in
the Foundry of Hawamdia Machinery & Equipment Factories, from grey
cast iron. Mill roller breakages are fairly common and invariably costly,
not only because of the need to replace them but also due to the
downtime that results from their breakage. Moreover, their breakage is
usually associated with secondary damage to the bearings and the
headstock. The purpose of the mill headstocks is to maintain the
working elements (particularly the rolls) in their desired orientation.
This orientation needs to be flexible to allow for different roll sizes and
setting. The basic mill comprises three mills: the top roll, which needs
to be able to "float" upwards during operation, and the feed and dis-
charge rolls that need to be adjustable sideways. The company con-
sumes a large quantity of mill roller shells, approximately (1200)
tonne/year and the cost of this production rate is nearly (12) million
Egyptian pounds. The Egyptian Sugar Integrated Industries Company
adopts a program for annual maintenance for the mill roller shells to
prevent failure during the season as failure costs a big pity.
The average lifetime of mill roller shells in Egyptian
Sugar & Integrated Industries Company (ESIIC) before the first re-
storation is approximately one year the main damage occurring to the
surface is abrasive wear. For example, the Kawasaki mill roller shell
diameter changes by 20 mm each season; from 1120 mm to 1100 mm,
1080 mm and 1060 mm, consecutively, as shown in Fig. 1. Usually,
there are three restoration maintenance repairs conducted when the
worn surface reaches about (1.8%) of the diameter. The restoration
includes machining of the worn surface and regrooving of the new
surface. This is associated with a decrease in the diameter resulting in a
loss in productivity of the mills. For economic purposes, the restoration
is done three times till the diameter is drastically reduced and the mill
roller is put out of the service.
There are many reasons for corrosion and wear in mill roller shells.
such as rust of metal mill rollers (as a result of the presence of acidic
juice); friction between the top scraper-trash plate-cane- sucker and
mill rollers; and on top of all the high speed of the moving rolls [3],
where the wear in the middle of the mill rollers is greater than their
ends. Usually, the wear is 1.8% (or in other words 20 mm) annually,

⁎
Corresponding author.
E-mail address: eng_elsawytolba@yahoo.com (E.E.T. ELSawy).

http://dx.doi.org/10.1016/j.wear.2017.07.007
Received 24 January 2017; Received in revised form 12 June 2017; Accepted 6 July 2017
Available online 11 July 2017
0043-1648/ © 2017 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/BY-NC-ND/4.0/).
E.E.T. ELSawy et al. Wear 390–391 (2017) 113–124

and < 0.07% P), with the overriding constraint that the Carbon
Equivalent (CE) level = 3.9–4.2%.
Some studies focused on the hard facing of the grey cast iron rollers
using similar welding electrodes [8]. The engineered surfaces were
characterised for their abrasion and corrosion wear resistance. The re-
sults of the previous studies [8] suggested that the predominant wear
mechanism is corrosion associated with porosity and Cr depletion in the
weld zone.
The main objective of this study is to bring into focus the sig-
nificance of the chemical composition of the cast iron used for manu-
facturing the shells in order to increase juice extraction with minimum
wear at the lowest cost. According to Krauss [9], the addition of each
1.00%, silicon reduces the amount of carbon in the eutectic by 0.33%.
Fig. 1. Annual wear of mill roller shell.
The most common range for manganese in grey iron is from 0.55% to
0.75%. Increasing the manganese content tends to promote the for-
and the tool life for mill rollers is about (3–5) years.
mation of pearlite while cooling through the critical range. It is ne-
The favoured specification for mill roller shells used to be "open-
cessary to recognise that only that portion of the manganese not com-
grained cast iron", since this produced a rough finish to aid feeding.
bined with sulphur is effective. Virtually, all of the sulphur in grey iron
However, with higher loads and stresses and the development of roll
is present as manganese sulphide, and the manganese necessary for this
surface roughening techniques such as carbon-arcing and subsequently
purpose is 1.7 times the sulphur content. Manganese is often raised
hard welding a close-grained, the higher tensile material is now pre-
beyond 1.00%, but in some types of green sand castings, pinholes may
ferred. The material should be free from blowholes and cracks. Wear
be encountered. Chromium is generally present in amounts below
rate increases with the decrease of angle grooves and the optimum
0.10% as a residual element carried over from the charge materials.
angle for mill roller is 45°: 55° [3]. Three type of grooving is generally
Chromium is often added to improve hardness and strength of grey
used on mill rollers [3,4].
iron, and for this purpose, the chromium level is raised to 0.20–0.35%.

• Fine grooves 5–20 mm pitch.

Beyond this range, it is necessary to add a graphitised to avoid the

• Medium grooves 20–50 mm pitch.

formation of carbides and hard edges. Chromium improves the elevated

• Coarse grooves above 50 mm pitch.

temperature properties of grey iron. The aim of this experimental study
is to study the effect of adding (2.11% chromium, 3% manganese and
2.02% silicon) to grey cast iron cast shells in order to improve the wear
The fine grooves are best for extraction and the coarse ones are best
resistance of the surface. The alloy design is based on adding Cr as
for crushing. Some engineers adopt identical grooves on all mill rollers,
carbide former with increased Si content to approach the upper limit in
from the extraction point of view, while others favour differential
grey cast iron to act as a graphitised. Mn is also raised up to 2% to
grooving as by these juice handling capacity increases. The wear and
provide plentiful amounts to stabilise the pearlite.
tear, however, in the case of differential grooving is greater than in the
case of identical grooving. The angle of grooving is also important. In
2. Experimental work
recent practices, wider angles of (55–60°) are generally employed on
the top rollers and (52–58°) on the bottom rollers.
2.1. System description
Wear resistance is considered as the most important parameter for
proper selection of the material to be used for the manufacturing of mill
The main objective of milling is to separate the sucrose containing
roller shells. Hardness, microstructure and chemical composition of mill
juice from the cane. The prepared cane is pushed through a three-roller
roller shells play a marked role in wear resistance [5]. Although
mill and squeezed between the top roller and the feed roller. The juice is
hardness and relative wear are linearly proportional for most of the
extracted and collected in a trough and a trash plate guides the bagasse
commercial pure metals, the same simple relation does not hold for a
resulting from squeezing to the opening between the top roller and the
range of mill roller shells where it becomes necessary to consider che-
discharge roller, where it is squeezed once again in the set of the three-
mical composition as well.
roller mill. In a three- roller mill, three rollers are arranged in a trian-
Shells have historically been made of low strength cast iron. The
gular pattern for removing sucrose up to 96–97% max. The arrange-
desired properties of the final shell material include resistance to wear
ment of the rollers is shown in Fig. 2.
and polishing, suitable strength and good machineability and weld-
The three roller mills used for extraction of juice consist of top feed
ability. Being of a large section, the shell casting requires careful at-
and discharge rollers. Sugar cane is fed into the top and feed rollers,
tention to prevent undue porosity and shrinkage cavities. Once the
then further pass through the top and discharge rollers along with trash
casting has cooled at the correct rate, it is machined for the retelling of
plate. This trash plate has a downside such that 25% of the total hy-
a shaft. Precise machining of the internal through hole is critical to
draulic load is shared by this trash plate in overcoming friction and the
ensure an adequate shrink fit onto the shaft to avoid shell slippage in
remaining 75% only is useful, i.e. 25% of the hydraulic load is shared
service [6]. Despite the significance of this issue, there is not much
by feed roller and 50% is shared by the discharge roller, as shown in
research on this subject. Previous work only studied the surface of grey
Fig. 3. The crushing rolls are designed to have a high coefficient of
cast iron rollers arced in traditional ways.
friction and are operated at very low rotational speeds up to 4–7 rpm.
A recent study [7] on eight factories in Jamaica using grey cast iron
The power needed for crushing the sugarcane is applied on to the
for their mill roller shells revealed that much more direction and testing
top roller, which rotates the feed and discharge rollers, with pinion
of the material specification are required from factories to manu-
arrangement attached on one side of the roller. The direction of rotation
facturers when purchasing mill roller shells. The findings showed that
of top and feed–discharge rollers is opposite. The top roller is the most
specifications are seldom requested. Later on, researches started to re-
critical component amongst all, as the drive torque, hydraulic load and
cognise the importance of chemical analysis [8] but research on this
crushing load all act on it. The forces acting on the mill roller gives rise
subject is still lacking. Good practice, based upon the work carried out
to shearing, bending, torsion and compressive stress. The top roller is
by Sare, Constantine, Mason and Thwaite [8] suggested an alloy com-
most highly stressed since it consumes about half of the mill torque. Out
position of (3.3–3.5% C, 1.6–2.0% Si, 2.0–3.0% Mn, 0.07–0.10%S,
of total power, 50% is taken by top roller, 35% is taken by discharge

114
E.E.T. ELSawy et al. Wear 390–391 (2017) 113–124

Table 1
Material charge composition.

Materials Composition

C Si Mn P S Cr

Pig iron 3.28 1.8 0.56 0.12 0.04

Steel scrap 0.1 0.1 0.5 0.02 0.01
Carburizer 99.7 0.27
Ferro-silicon 65
Ferro-manganese 7 74
Ferro-Chromium 0.3 73.5

Table 2
The chemical composition (wt%) of tested materials.

Elements C% Si % Mn % P% S% Cr %

Condition 1 3.39 2.06 0.58 0.12 0.040 0.062

Condition 2 3.38 2.08 1.05 0.12 0.041 0.062
Condition 3 3.39 2.09 1.53 0.12 0.041 0.062
Condition 4 3.39 2.03 2.03 0.12 0.039 0.062
Condition 5 3.39 2.05 2.51 0.13 0.043 0.062
Condition 6 3.39 2.02 3.00 0.13 0.044 0.062
Condition 7 3.39 2.59 2.89 0.12 0.044 0.062
Condition 8 3.38 3.00 2.91 0.12 0.044 0.063
Fig. 2. Assembly of conventional sugar three- roller mills. Condition 9 3.30 2.00 3.00 0.12 0.039 0.520
Condition 10 3.31 2.04 2.93 0.12 0.040 1.070
Condition 11 3.34 2.11 3.00 0.12 0.039 2.110

Fig. 3. Conventional Roller mill Hydraulic load details. Fig. 5. Schematic configuration of the Pin-on- Disc machine employed for the wear test.

Table 3
Summary of experimental results.

Conditions Hardness (HB)

Sample (1) Sample (2) Sample (3) Mean Hardness

Condition 1 185 HB 175 HB 190 HB 183 HB

Condition 2 182 HB 187 HB 200 HB 190 HB
Condition 3 190 HB 180 HB 211 HB 194 HB
Condition 4 200 HB 200 HB 214 HB 205 HB
Condition 5 214 HB 204 HB 207 HB 208 HB
Condition 6 211 HB 215 HB 235 HB 220 HB
Condition 7 223 HB 210 HB 207 HB 213 HB
Condition 8 208 HB 203 HB 206 HB 205 HB
Condition 9 248HB 268HB 252HB 256 HB
Condition 10 269 HB 296 HB 276 HB 280 HB
Condition 11 328 HB 332 HB 326 HB 329 HB
Fig. 4. Image of worn roller surfaces.

roller, 15% is taken by the feed roller. Fig. 4 shows the worn roller
surfaces, from which it can be deduced that the main wear mechanism
is abrasive wear by forming large grooves across the surface of the rolls.
The chemical composition of the current working mill roller shells is
(3.2–3.5% C, 1.5–2.5%Si, 1.5–2.5% Mn, 0.5% max. P and 0.15% max.
S).
2.2. Melting, casting and materials
The melts in the present study were produced in a 1000 kg capacity
medium frequency induction melting furnace, with a power rating of
330 kW/ 950 Hz. Low manganese pig iron, steel scrap and ferroalloys
were used as the raw materials. The chemical analysis of the raw ma-
terials used in this work is shown in Table 1. The melts were cast in

115
E.E.T. ELSawy et al. Wear 390–391 (2017) 113–124

Fig. 6. The hardness distribution for different tested materials. Fig. 7. The variation of ultimate tensile strength for different tested materials.

resin bounded sand moulds to produce round bars of 50 mm diameter

and 350 mm length, from a pouring temperature of 1300 °C. After
pouring, the castings were left to cool in the mould to room tempera-
ture, and allowed to solidify. The melt was inoculated by adding
commercial inoculants (70–75% Si, 1.5–2% Ba, 0.75–1.25% Ca and
0.75–1.25 Al) to the molten metal in the ladle. The weight of inoculant
FeSi 0.7 was equal to 4 kg /tonne of melt and the inoculation tem-
perature was 1380 °C.
Table 2 lists the chemical composition of the cast alloys produced in
this study. The material used in this work is cast iron with carbon in the
range of 3.30% [10]. The materials were designed to include chromium
in the range of (0.52–2.11%), manganese in the range of (0.58–3.00%)
and silicon in the range of (2–3%). Eleven material groups of the che-
mical analysis shown in Table 2 were prepared for this work.
The chemical composition of the currently used mill roller shells
(which is 3.2–3.5% C, 1.5–2.5%Si, 1.5–2.5% Mn, 0.5% max. P and
0.15% max. S) is represented by the condition number (4) in Table 2. Fig. 8. The microstructure of condition (1).

2.3. Hardness test

The hardness was measured using a Portable Hardness Tester, in-

strument type–Hartip 3000, Serial Number: E 12030043. The hardness
test was carried out by using 11 samples of dimensions 40 mm diameter
and 20 mm length, three readings were made for each sample. The tests
were conducted at Central Metallurgical Research and development
Institute (CMRDI), and some validation tests were made at Hawamdia
Laboratories.

2.4. Tension test

The tensile properties of the materials were determined at

Hawamdia Machinery & Equipment Factories, According to ASTM
2004 vol 03.01 E8M using tensile test machine type VEB 283 / 22,
capacity 100 t. The tension test was carried out for the eleven material
groups; each group consisted of three samples according to ASTM
2004 vol 03.01 E8M, turned from casting round bars 50 mm diameter
and 350 mm length.
2.5. Microstructure examination
The aim of this experimental study was to evaluate the effect of
adding some carbide forming alloying elements to the cast iron used for
shelling the mill rollers in order to enhance their wear resistance.
Accordingly, the microstructure examination was an important element
of the study and it was carried out using an Optical Microscope type
PME OLYMPUS X50 - X400. Specimens were prepared ground, polished
and etched using (3%), Nital solution as an etching solution. The
Fig. 9. The microstructure of condition (6).
microstructure examination was carried out for eleven samples (re-
presenting the 11 groups of material prepared in this work) of dimen-
sions 20 mm diameter and 10 mm length. Microstructure examinations
at higher magnifications were performed.
2.6. X-Ray diffraction examination
This test was used to determine the type of carbides, which are
present in the material, by using instrument type Panalytical Xport Pro,
located in the Central Metallurgical R & D Institute (CMRDI). The X-Ray
diffraction examination was carried out on eleven samples of

Table 4
Tension test for conditions.

Conditions 1 2 3 4 5 6 7 8 9 10 11

U.T.S (MPa) 142.88 148.86 153.96 179.66 182.80 187.90 179.46 145.33 145.73 156.12 166.42

116
E.E.T. ELSawy et al. Wear 390–391 (2017) 113–124

Table 5
Phases present in conditions from 1 to 8.

Name and formula A B

Reference code: 00 − 023 – 1113 03 – 065 – 4899

Mineral name: Cohenite, syn
Common name: Cementite Α – Fe
PDF index name: Iron Carbide Iron
Empirical formula: CFe3 Fe
Chemical formula: Fe3C Fe
Crystal system: Orthorhombic Cubic

Peak list:
A = 00 – 023 – 1113; Cementite; Fe3C; Orthorhombic.
B = 03 – 065 – 4899; α – Fe; Fe; Cubic.

Table 6
Fig. 10. The microstructure of condition (8). Phases present in conditions from 9 to 11.

Name and formula C B

Reference code: 01 − 023–1113 03 – 065 – 4899

Mineral name: Cohenite, syn
Common name: Α – Fe
PDF index name: Iron Carbide Iron
Empirical formula: CFe3 Fe
Chemical formula: Fe3C Fe
Crystal system: Orthorhombic Cubic

Peak list.
C = 01 – 085 – 1317; Cohenite, syn; Fe3C; Orthorhombic.
B = 03 – 065 – 4899; α – Fe; Fe; Cubic.

dimensions 20 mm diameter and 10 mm length.

2.7. Wear test

Fig. 11. The microstructure of condition (11).
2.7.1. Analysing operational conditions
The pressure acting on the mill roller shells is very large, and it
ranges from 550 to 600 t and the mill driving power ranges from 650 to
850 HP. These conditions give stresses in the range of approximately
2.2 Kgf/mm2, at a speed range of 4–7 rpm. The diameters of mill roller
shells range from 1070 to 1165 mm all depending on the production
rate. Based on the working conditions, the lab wear test was selected to
simulate those conditions during the operation. During the rotation of
the mill rollers, friction is generated between the top and feed rollers; as
well as between the top and discharge rollers, leading to abrasive wear
in the mill roller shells. The testing conditions in this work were se-
lected to represent the working conditions.

2.7.2. Wear testing machine (pin on disc)

Fig. 12. X – Ray diffraction pattern of condition (1). The wear test was conducted on a pin- on- disc, wear testing ma-
chine type Plint multi-axis Tribology Machine, and serial number TE
79/8532. In the pin-on-disc, the contact between the pin and disc may
be either end-on or side-on. If the disc takes the form of a grinding
wheel, then with the pin in the end-on position we have a simple
grinding set-up, which can be used for fundamental studies of abrasive
wear while the side-on position simulates simple sliding, as shown in
Fig. 5. In this test, the contact was end-on corresponding to the pin-on-
disc type. This configuration was selected to represent the conditions
during juice extraction from the sugar cane.
The technical specifications for the pin-on-disc wear testing ma-
chine, used for this study are as follows:

• Normal load: 1–50 N.

• Disc diameter: 100 mm.
• Track radius: 0–40 mm.
Fig. 13. X – Ray diffraction pattern of condition (9). • Rotation speed: 0–350 rpm.
117
E.E.T. ELSawy et al. Wear 390–391 (2017) 113–124

Table 7
Wear rate for conditions from 1 to 11 by a pin on disc machine.

Conditions 1 2 3 4 5 6 7 8 9 10 11

Wear rate 2.49 2.48 2.27 2.24 1.77 1.23 1.38 1.45 0.903 0.660 0.340
(mg/s)

• Contact configuration: pin on disc.

• Grinding disc: high-speed steel.
• Disc Hardness: 60 HRC.
2.7.3. Wear test procedure
The procedure for wear test experiments is summarised in the fol-
lowing steps.

• Round cast bars, 50 mm diameter and 350 mm length were pre-

pared for the wear test.
• After casting the specimens, they were turned to pins of 8 mm dia-
Fig. 14. The variation of accumulated loss of weight for wear test machine pin-on-disc.
meter and 15 mm length, which correspond to the size of the pin in
the pin-on-disc machine.
Table 8 • The surface roughness values Ra (μm) were equal to 0.8 and the
Wear data compared to life time expectancy. roughness grade number was equal to N6, after turning process.
• Before each test, the pins were carefully cleaned.
Conditions Wear rate (mg/s) Life time
• The tests were carried out at a constant load of 20 N.
• The test was conducted at a constant speed of 1.28 m/s.
•
4 (reference material) 2.2367 4 years
6 1.2267 7 years The machine was stopped at time intervals for weighing the pin. Any
11 0.3400 26 years loose debris at the surface of the pin specimens was removed by
cleaning before weighing the specimens.

Fig. 15. SEM micrographs worn out surfaces (a) condition 1, (b) condition 6, (c) condition 8 and (d) condition 11.

118
E.E.T. ELSawy et al. Wear 390–391 (2017) 113–124

Fig. 16. SEM images of (a) condition 1 (b) condition 6 (c) condition 8 (d) condition 11in etched condition.

• On restarting the wear tests, the pins were located in their supports 3. Results and discussion
in exactly the same orientation as before stopping the tests, in order
to ensure that the wear in the pin has occurred in the same direction 3.1. Hardness test
through out the test.
• At the end of the test, the accumulated wear of the pins was cal- The hardness of the studied material is shown in Table 3 and Fig. 6.
culated (expressed in mg). The time and the wear rate were calcu- The hardness ranges were from (183 HB to 329 HB) as shown in
lated for each test material. Table 3. The hardness values were found to increase, as the manganese
percent increased. It is clear from the Fig. 6 that the hardness increases
The test conditions were as follows: with increasing the manganese percent for the tested materials from
Wear test machine pin-on-disc. condition 1 to condition 6 with the highest manganese content. In-
Grinding disc: high-speed steel. creasing the manganese content increases the hardness in conditions (1,
Wear test parameters. 2, 3, 4, 5, and 6), corresponding to (0.58% Mn, 1.05% Mn, 1.53% Mn,
Time: 300 s. 2.03% Mn, 2.51% Mn, and 3.00% Mn), respectively. It is also shown
Load: 20 N. from Fig. 6 that the hardness decreases with increasing the silicon
Temperature: 23 °C. percent for the tested materials from conditions (7 and 8). This may be
Samples mean area: 25 mm2 (diameter of pin=8 mm). attributed to the increase in the volume of the soft graphite flakes as the
Velocity: 1.28 m/s. silicon content increases. The hardness test results, also show that in-
creasing the chromium content increases the hardness in conditions 9,
• Track radius: 70 mm. 10, and 11, corresponding to (0.520% Cr, 1.070% Cr, and 2.110% Cr),
• Rotation speed: 350 rpm. respectively. Fig. 6 shows, on the other hand, that increasing the
chromium content to (2.11%) resulted in the highest hardness (condi-
tion number 11) and that increasing the manganese to (3%) resulted in
2.8. Scanning electron microscope examination
the highest hardness for the alloys without chromium (condition
number 6 compared to 1–5). Previous work [11] has attributed similar
Scanning Electron Microscopy was used in this work in order to
observations to the formation of chromium carbides, which results in
study the worn surfaces, as well as the microstructure at higher mag-
enhancement in its hardenability and resistance to abrasion and wear.
nifications. A SEM Machine Type XTm 810 Inspect S, built in date
The increasing carbide formation with increasing Cr addition was ob-
01–05–2007 was used. The examination aimed at identifying the mi-
served in this work, associated with a decrease in free graphite content.
crostructural features. These tests were carried out using SEM with
EDX.

119
E.E.T. ELSawy et al. Wear 390–391 (2017) 113–124

Fig. 17. SEM images of (a) condition (1), (b) condition (6), (c) condition (8) and (d) condition (11) unetched condition.

3.2. Tension test

The values of the ultimate tensile strength are shown in Table 4 and
Fig. 7. It is shown from Fig. 7 that the condition number 6 (3% man-
ganese and 2.02%silicon) has the highest ultimate tensile strength.
Fig. 7 also shows that when the silicon increased for conditions 7 & 8,
the ultimate tensile strength dropped significantly. It is also observed
from Fig. 7 that the addition of chromium raised the ultimate tensile
strength relative to conditions (7 & 8), though it did not result in a high
ultimate tensile strength as that obtained for condition 6.

3.3. Microstructure

The microstructure of the studied grey cast iron materials is shown

in Figs. 8–11, respectively. The addition of silicon is known to allow the
formation of graphite more easily, particularly its formation from the
liquid. Perhaps it might be more correct to say that the addition of Si
makes it more difficult to form Fe3C.
Fig. 18. Black region and white region condition (1).
The size and the number of graphite nodules formed during solidi-
fication are influenced by the amount of carbon, the number of graphite
nuclei, and the choice of inoculation practice. Almost all samples in this

120
E.E.T. ELSawy et al. Wear 390–391 (2017) 113–124

Fig. 19. SEM/EDXA spectrum at the black region of condition (1).

Fig. 20. SEM/EDXA spectrum at the white region of condition

(1).

responsible for the increase in hardness with a corresponding increase

in toughness. The reduction in the volume of soft graphite flakes and
the strengthening effect of tough carbide combines to fortify the matrix
and help improve the wear resistance of grey cast iron. The volume and
distribution of soft flake graphite decrease with increase in manganese
addition in the samples [12].
From Figs. 9 and 10 show the conditions (3.00% Mn and 2.02% Si)
and (2.91% Mn and 3.00% Si), respectively. The effect of increased Si is
evident on the increase of the free graphite with increasing silicon
content. The size of the flakes was observed to increase as well with
increasing the silicon content. The effect of the Si percentage on the size
of the flakes was discussed [13] and it was shown that increasing the
silicon content of the cast iron increases the grain size and decreases the
hardness. Grain size and hardness are both dependent on silicon content
of the melt.
Fig. 11 show the micrographs of the as–cast conditions (9, 10 and
11) which contain Cr, where increased carbide formation is observed as
the chromium content increased from 0.520–2.110%). The micrographs
show a general trend of increasing the number of carbides with chro-
mium addition. It is well known [14] that the chromium addition
Fig. 21. Black region and white region condition (11).
strongly favours the formation of carbides during solidification. Pre-
vious work [15] has shown that the presence of the elements (Cr, V) in
work (apart from samples number 7 and 8) contain the same Si content.
grey iron raises the metastable eutectic temperature TEM, and depresses
It can be seen from Fig. 9 that those two conditions show longer gra-
the stable eutectic temperature TEG, giving rise to a reduction of the
phite flakes.
TEM-TEG interval. When the content of those elements is high, they
The effect of varying manganese contents (0.58%, 1.05%, 1.53%,
segregate along the grain boundaries and form eutectic carbides be-
2.03%, 2.51%, and 3.00%) is also revealed from Figs. 8–11. It can be
tween the eutectic cells. The carbides observed in this work (Fig. 11)
seen that the volume of the soft graphite flakes is substituted by hard
resemble the Cr carbides previously identified by Zhou Jiyang [16].
carbides (increasingly) as the manganese addition increased. Those
Correlating the microstructure observations with the hardness
hard carbides are responsible for the slight increase in the hardness for
measurements reported in Section 3.1, suggests that the increase in the
heats 4, 5, 6, 7 and 8 compared to heats 1, 2 and 3. The volume of the
length of the carbon flakes leads to a decrease in the hardness of the
soft graphite flakes has been substituted by hard carbides, which were
castings for all the values of silicon content used in this work. The
reduced as manganese addition increases. This may have been
decrease in hardness with increasing the silicon content of the cast iron

Fig. 22. SEM/EDXA spectrum at the black region of condition

(11).

121
E.E.T. ELSawy et al. Wear 390–391 (2017) 113–124

Fig. 23. SEM/EDXA spectrum at the white region of condition

(11).

Table 9
Chemical composition at the black region for condition 1.

KV: 30.0 Tilt: 0.0 Take-off: 35.0 Det Type: SUTW+ Res: 132 Amp. T: 102.4
Condition (1) /Black Region. FS: 4256 Lsec:60

Element wt% at% K-Ratio Z A F

CK 80.52 94.99 0.4010 1.0245 0.4860 1.0001

Si K 0.26 0.13 0.0016 0.9751 0.6401 1.0009
Mn K 0.26 0.07 0.0023 0.8562 1.0527 1.0000
Fe K 18.95 4.81 0.1738 0.8736 1.0495 1.0000
Total 100.00 100.00

Table 10
Chemical composition at the white region for condition 1.

KV: 30.0 Tilt: 0.0 Take-off: 35.0 Det Type: SUTW+ Res: 132 Amp. T: 102.4
Condition (1)/White Region. FS: 4256 Lsec:60

Element wt% at% K-Ratio Z A F

CK 13.06 40.50 0.0265 1.1323 0.1792 1.0006

Si K 2.27 3.01 0.0074 1.0758 0.3015 1.0022
Mn K 0.58 0.04 0.0056 0.9528 0.9988 1.0000
Fe K 84.09 56.09 0.8235 0.9727 1.0068 1.0000
Total 100.00 100.00

Table 11
Chemical composition at the black region for condition 11.

KV: 30.0 Tilt: 0.0 Take-off: 35.0 Det Type: SUTW+ Res: 132 Amp. T: 102.4
Condition (11)/Black Region. FS: 3070 Lsec:60

Element wt% at% K-Ratio Z A F

CK 73.41 92.61 0.3108 1.0341 0.4094 1.0001

Si K 0.59 0.32 0.0034 0.9840 0.5802 1.0011
Cr K 0.57 0.17 0.0064 0.8789 1.0477 1.2157
Mn K 1.02 0.28 0.0092 0.8648 1.0468 1.0000
Fe K 24.40 6.62 0.2245 0.8825 1.0425 1.0000
Total 100.00 100.00

Table 12
Chemical composition at the white region for condition 11.

KV: 30.0 Tilt: 0.0 Take-off: 35.0 Det Type: SUTW+ Res: 132 Amp. T: 102.4
Condition (11)/White Region. FS: 3832 Lsec:60

Element wt% at% K-Ratio Z A F

CK 16.47 47.50 0.0356 1.1281 0.1913 1.0005

Si K 0.79 0.98 0.0026 1.0720 0.3094 1.0022
Cr K 3.20 2.13 0.0399 0.9647 0.9925 1.3038
Mn K 3.07 1.94 0.0292 0.9490 1.0032 1.0000
Fe K 76.47 47.45 0.7393 0.9689 0.9978 1.0000
Total 100.00 100.00

122
E.E.T. ELSawy et al.
is also reported elsewhere [17]. The microstructure observations also
confirm the effect of the Cr- carbides on increasing the hardness of the
grey cast iron from 184 HV to 329 HV.
3.4. X-Ray diffraction examination
Fig. 12 shows the x-ray chart for condition (1). This is typical for all
conditions 1–8). Fig. 13 shows the x-ray chart for condition (9). This is
typical for all conditions 9–11). Table 5 lists the types and names of the
phases identified for conditions 1–8), and Table 6 lists the phases
identified for conditions (9, 10 and 11).
It is seen from the X-ray analysis shown in Figs. 12 and 13, and
Tables 5, 6 that the main phases formed by the alloying elements in this
work are mainly α-Fe and Fe3C. The x-ray failed to detect the carbides
formed after adding the alloying elements, probably due to their small
volume fraction.
The x – ray proved the presence of the phases shown in Tables 5, 6
from which the presence of Fe3C is capital. This phase is responsible for
increasing the wear resistance.
3.5. Wear test
As mentioned before, the wear testing conditions selected for this
work compromised loading under pin-on-disc configuration. The con-
tact configurations were such that the pin was loaded vertically
downwards onto a horizontally rotating disc. The pin was made from
the cast material prepared in this work, and the disc was made from
high-speed steel [18]. The coefficients of friction between the pins (test
specimens) and the disc for the pin-on-disc wear testing machine
ranged between 0.512 and 0.721. The wear testing conditions were
selected to be similar to the operating conditions of the mill roller
shells. Usually, they operate at constant speeds and loads, and ac-
cordingly, those two parameters were kept constant during the tests.
Table 7 and Fig. 14 show the results obtained from the wear test. It
is shown from Fig. 14 that the wear rate decreases with increasing the
hardness for all tested materials (conditions). By comparing the wear
rates for conditions (1, 6 and 11), where condition (11) shows the least
wear and condition (6) shows less wear than condition (1), the effect of
Mn addition is shown to improve the wear resistance. This may be at-
tributed to the increased hardness of the Mn-containing cast iron. The
reduction of the wear rate for condition (11) compared to conditions (1
and 6), again confirms the relationship between wear resistance and
hardness for this group of cast iron alloys. The least wear rate in this
work is associated with the maximum hardness for condition (11).
Considering conditions (7 and 8), it is shown that the wear rate is
less for condition (7) than that for condition (8). This suggests that the
wear increases as the silicon content increases. The wear rate for con-
dition (11) is less than that for conditions (9 and 10), suggesting that
the wear decreases as the chromium content increases. The wear rates
for conditions (9, 10, and 11) are equal to (0.9033, 0.6600 and 0.3400)
mg/s, which represent the least wear rates obtained in this work. These
results confirm the beneficial role of adding Cr to grey cast iron for
improving the wear resistance. The formation of hard Cr carbides with
finer sizes and more uniform distribution explains the increase in the
hardness of the grey cast iron.
Condition (4) represents the currently used material for manu-
facturing the mill roller shells, and the obtained wear rate is equal to
(2.2367) mg/s. The service life for this material is about (3–5) years at
the reported wear rate of (2.2367) mg/s. This suggests that replacing
the current material (condition 4) with the suggested material (condi-
tion 11) is expected to increase the service life of the sugar mill roller
shell six times. Since the minimum wear rate was obtained in this work
for condition (11) and was equal to (0.3400) mg/s. Table 8 shows the
wear data compared to life time expectancy, based on comparison with
the reference material.
It can be demonstrated from Table 8 that the life time expectancy of
123
Wear 390–391 (2017) 113–124
the sugar mill roller shells is expected to increase significantly with the
new material proposed in this work. However, actual field measure-
ments are needed.
3.6. Study of worn surfaces by scanning electron microscope
Figs. 15a–d represent the morphologies of the worn out surfaces of
conditions (1, 6, 8 and 11) at (2.06% Si and 0.58% Mn; 2.02% Si and
3.00% Mn; 3.00% Si and 2.91% Mn; and 2.11% Si and 3.00% Mn,
2.11% Cr), respectively. It can be detected from Fig. 15 that the hypo
eutectic material, developed in this work, attains its wear resistance
through the dispersion of the carbides within the matrix, as well as its
capacity to undergo strain hardening.
It can be seen from Fig. 15 that the abrasive surface damage in-
creased with increasing the silicon content (Fig. 15c), the surface da-
mage decreased with increasing the manganese and chromium content
(Figs. 15b and d). It appears from Fig. 15, which represents conditions
(1, 6, 8 and 11); respectively that a change in the wear surface is
pronounced from conditions (1–11), as the manganese, silicon and
chromium content varies.
The analysis of the worn surface features shows that the wear me-
chanism that would result in such surface features resemble typical
abrasive conditions, where sharp particles or grits remove material by
micro cutting, microfracture, pull-out of individual grains or ac-
celerated fatigue by repeated deformations as illustrated in Fig. 15.
Sugar cane, which is an organic material, is associated with abrasive
wear of cane cutters and shredders because of the small fraction of silica
present in the plant fibres. A major difficulty in the prevention and
control of abrasive wear is that the term ‘abrasive wear’ does not
precisely describe the wear mechanisms involved. There are, in fact,
almost always several different mechanisms of wear acting in concert,
all of which have different characteristics.
3.7. Scanning electron microscope examination
Fig. 16 shows SEM images obtained for conditions (1, 6, 8 and 11),
after etching to reveal the matrix and the carbides. Higher magnifica-
tions offered by SEM were sought to aim at identifying the carbides that
may have formed after the alloying additions in this work. The SEM
images revealed in Fig. 16 have confirmed previous observations by
Optical Microscopy, in a clearer manner. The replacement of the soft
graphite flakes by the hard carbides, with the chromium addition, is
obvious in Fig. 16d. The replacement of the graphite flakes by the
carbide phase for condition 11 (Fig. 16d) explains the enhancement in
wear resistance for this condition.
On the other hand, Figs. 17a–d represent the SEM images obtained
for conditions (1, 6, 8 and 11), without etching.
It can be seen from Figs. 17a and b that the volume of the soft
graphite flakes decreased, as the manganese addition increases, which
supports the slight increase in the hardness. Another factor, which may
have contributed to the high hardness is the increase in the pearlite
volume fraction for conditions 6 and 8, as shown from Fig. 16(b) and
(d). The micrographs show a general trend of increasing amount of
carbides with increasing chromium addition (Fig. 17(d)) on the expense
of a decrease in free graphite content.
3.8. Scanning electron microscope with EDS examination
From Figs. 18–23 and Tables 9–12 show the SEM images with EDS
obtained for conditions (1, 6, 8 and 11). The increase in the carbon
content and a low proportion of manganese and silicon in the graphite
zones (Black region) is shown from the figure. On the other hand, the
decrease in the carbon content and high proportion of manganese and
silicon in the pearlite zones (White region) is also shown from the figure
(Figs. 18–20).
The SEM study combined with the EDS analysis provided some
E.E.T. ELSawy et al.
explanation for the previously reported findings on the role of the al-
loying elements on the hardness and tensile strength. These observa-
tions explain the role of both elements, as silicon stabilises the ferrite,
and manganese stabilises the cementite phases. The presence of chro-
mium carbides in chromium containing cast irons (heats number 11) is
also shown from the EDS analysis (Figs. 21–23).
4. Conclusions
• In this work, a modified cast iron is developed with the composition
3.34% C, 2.11% Si, 3.00% Mn, 0.12% P, 0.039% S and 2.11% Cr, for
the manufacturing of sugar mill roller shells. Based on the me-
chanical properties and the wear test, this cast iron is expected to
demonstrate a six-time longer life in service, compared to the cur-
rently used cast iron.
• The role of silicon, manganese and chromium on the mechanical
properties of the investigated cast iron has shown that increasing
silicon and carbon equivalent results in a decrease in the tensile
strength and a decrease in the hardness. The increase in the man-
ganese and chromium additions results in an increase in the tensile
strength and hardness.
• The wear resistance improved with increasing manganese and
chromium additions. The minimum wear rate was obtained at
2.11% chromium, 3% manganese and 2.11% silicon additions.
Appendix A. Supporting information
Supplementary data associated with this article can be found in the
online version at http://dx.doi.org/10.1016/j.wear.2017.07.007.
References
[1] Chetan T. Rathod, Walmiki S. Rathod, Design and analysis of two roller sugar mill
using FEA techniques, Int. J. Sci. Eng. Technol. 1 (3) (2012) 148–152 (19).
[2] Santosh Y. Salunkhe, Static structural analysis of conventional sugar mill roller
124
Wear 390–391 (2017) 113–124
shaft for Φ40" x Φ80" milling plant, Int. J. Latest Trends Eng. Technol. 5 (4)
(2015) (20).
[3] Peter Rein, Cane Sugar Engineering, Verlag Dr. Albert Bartens KG, Berlin,
2007 (21).
[4] E. Hujot, Handbook of Cane Sugar Engineering, Elsevier publishing company,
Amsterdam/London/New York, 1972 (22).
[5] G.D. Oliver, V.E. Buchanan, K.O. Cooke, A critiq ue on the wear mechanism of
surface engineered su gar mill roller shells Oliver GD et al. Proc. Aust. Soc. Sugar
Cane Technol. vol. 30, 2008. 2.
[6] S.I. Anderson, J.G. Loughran, Finite Element and Durability Modelling Roller Shells
and Shafts, Department of Mechanical Engineering, James Cook University of North
Queensland, 1998.
[7] Johnson O. Agunsoye, Talabi S. Isaac, Olumuyiwa I. Awe, Afemefuna
T. Onwuegbuzie, Effect of silicon additions on the wear properties of grey cast iron,
J. Miner. Mater. Charact. Eng. 1 (2013) 61–67 (4).
[8] G.D. Oliver, B. Wilson, A Study of Roughening the Surface Cane Mill Roll Shells as
Practiced in Jamaica Jamaica Association of Sugar Technologists Conference
RenaissanceJamaica Grande Resort – Ocho Rios, Jamaica.3.
[9] D.E. Krause, “Grey Iron-A Unique Engineering Material” Grey, Ductile, and
Malleable Iron Castings-Current Capabilities, ASTM STP 455, American Society for
[10]
Testing and Materials, Philadelphia, 1969, pp. 3–28.
I.A. Mohamed, A.A. Ibraheem, M.I. Khashaba, W. Ali, Influence of heat treatment
on friction and wear of ductile Iron: i. Role of copper and molybdenum, Int. J.
Control Autom. Syst. 2 (3) (2013) (10).
[11] P. Atanda, A. Okeowo, O. Oluwole, Microstructural study of heat treated chromium
alloyed grey cast iron, J. Mater. Mater. Charact. Eng. 9 (3) (2010) 263–274 (11).
[12] J.O. Agunsoye, E.F. Ochulor, S.I. Talabi, S. Olatunji, Effect of manganese additions
and wear parameter of NF Grey [8] Cast Iron, Tribol. Ind. 34 (4) (2012)
[13]
239–246 (1).
P. Atanda, G. Oluwadare, O. Oluwole, Effect of silicon content and shake-out time
on hardness and grain size properties of GL 250 cast iron, J. Mater. Mater. Charact.
Eng. 10 (3) (2011) 257–266 (13).
[14] E. Zumelze, O. Opitz, C. Cabezas, High-chromium (22-34 per cent) cast iron alloys
and their simulated behavior at the sugar industry, J. Sci. Ind. Res. 62 (2003)
583–588 (12).
[15] S. Bockus. A Study of the Microstructure and Mechanical Properties of Continuously
Cast Iron Products)) Original Scientific Paper-Izvorni znanstveni rad, Accepted –
Prihvacen.
[16] Zhou Jiyang, China colour metallography of cast iron, China Foundry 6 (4) (2009).
[17] S.O. Seidu, Effect of compositional changes on the mechanical behaviour of grey
cast iron, J. Metall. Eng. 3 (2) (2014) 14.
[18] Giovanni Straffelini, Piyush Chandra Verma, Ibrahim Metinoz, Rodica Ciudin,
Guido Perricone, Stefano Gialanella, Wear behavior of a low metallic friction ma-
terial dry sliding against a cast iron disc: role of heat-treatment of the disc, Wear
348–349 (2016) 10–16.