7/30/2019 The Mechanisms of Erosion of Unfilled Elastomers http://slidepdf.com/reader/full/the-mechanisms-of-erosion-of-unfilled-elastomers 1/14 Wear, 138 (1990) 33 - 46 THE MECHANISMS OF EROSION OF UNFILLED ELASTOMERS BY SOLID PARTICLE IMPACT* J. C. ARNOLD and I. M. HUTCHINGS Department of Materials Science and Metallurgy, University of Cambridge, Pembroke Street, Cambridge (U.K.) The mechanisms of material removal were studied during the erosion of two unfilled elastomers (natural rubber and epoxidized natural rubber). The effects of impact velocity and of lubrication by silicone oil were in- vestigated. The development of surface features due to single impacts and during the early stages of erosion was followed by scanning electron mi- croscopy. The basic material removal mechanism at impact angles of both 30” and 90” involves the formation and growth of fine fatigue cracks under the tensile surface stresses caused by impact. No damage was observed after single impacts; it was found that many successive impacts are necessary for material removal. It was found that the erosion rate has a very strong dependence on impact velocity above about 70 m s-l. 1. Introduction Elastomers show very good resistance to erosive wear under certain conditions and are used in applications such as ore-handling plant and pipe linings [ 11. Despite this, understanding of the mechanisms of erosion and of the properties desirable for erosion resistance is still limited. There have been several studies attempting to correlate erosion resis- tance with various mechanical properties of elastomers [ 2 - 41. These have had some success, although it appears that the dependence of erosion rate upon mechanical properties is complicated, with no single property domi- nating. A better understanding of the mechanism of material removal is needed before any optimization of physical properties for erosion resistance can be achieved. The variation in erosion rate with angle of incidence is similar to that observed in ductile metals, with a high erosion rate at glancing angles of impact and a much lower erosion rate at normal incidence [3, 51. The mechanism of material removal in the case of metals, namely a cutting *Paper presented at the International Conference on Wear of Materials, Denver, CO, U.S.A., April 8 - 14, 1989. 0043-1648/90/$3.50 0 Elsevier Sequoia/Printed in The Netherlands
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
7/30/2019 The Mechanisms of Erosion of Unfilled Elastomers
and ploughing process, is, however, not generally thought to be responsible
for erosion in elastomers.
Elastomers eroded at glancing incidence show transverse features that
suggest the formation of tears and cracks perpendicular to the erosion direc-tion, possibly by a fatigue mechanism [2, 5, 61. The surface features pro-
duced during erosion bear a similarity to those produced during the abrasion
of elastomers on rough surfaces, to the extent that some workers have
drawn no distinction between the two processes [7, 81. The abrasion of
elastomers is thought to occur by a cyclic crack growth mechanism driven
by tensile stresses in the surface parallel to the sliding direction [9]. The
similarity of the surface features seen during both erosion and abrasion
suggests that a similar process may be occurring in erosion, The fact that
both processes are found to be accelerated by environmental degradation
[lo, 111 also points to some similarity in mechanism. There is, however,
a problem in that elastomers with good abrasion resistance tend to have
poor erosion resistance and vice versa. It has generally been found that
unfilled elastomers with a low modulus and high rebound resilience provide
the best erosion resistance [2, 3, 4, 81, whereas filled elastomers with a
high modulus tend to provide the best abrasion resistance [ 121.
The mode of deformation during wear has been found to be of impor-
tance by Muhr et al . [ 131, who found that lubrication during abrasion
caused a slight reduction in the frictional force, but a much larger reduction
in the wear rate due to an alteration in the mode of deformation. Thenature of the deformation caused by an impacting particle will be different
from that involved in abrasion and will be of great importance in deter-
mining the erosion rate. This could explain the discrepancy noted above
between resistance to the two types of wear.
Bartenev and Penkin [5] and Hutchings et al . [2] related the erosive
wear of elastomers to the amount of kinetic energy absorbed on impact.
Marei and Izvozchikov [ 41 suggested the necessity for a “stress build-up” by
successive impacts, by incomplete relaxation of the elastomer surface be-
tween impacts. It is clear that the impact stresses will play a large part in
determining the erosive wear of elastomers, although the nature of thedeformation induced in a roughened surface by an irregular particle is
likely to be complicated.
In this work, an attempt has been made to determine the material
removal mechanism during the erosion of elastomers by silica particles
under glancing and normal impact. The effects of lubrication were inves-
tigated, and the development of surface features was followed by scanning
electron microscopy (SEM) of the eroded specimens.
2. Experimental methods
2.1. Mater ia ls
The elastomers used in the present study were prepared at the Malay-
sian Rubber Producers Research Association (Brickendonbury, Hertford),
7/30/2019 The Mechanisms of Erosion of Unfilled Elastomers
Fig. 4. The variation in erosion rate produced by the incorporation of a silicone oillubricant into the air stream. (For ENR50 at 90” with lubricant, the erosion rate was
lower than the detectable limit of 10b6.)
at both 30” and 90” whereas, for mild steel, lubrication increases the erosion
rate slightly. The erosion rate for ENR50 at 90” with lubricant was lower
than the limits of detection in these experiments (about 10m6 owing to
mass fluctuations), as was the erosion rate for NR at 90” impact.
4. Discussion
4.1. The effects of velocity
As can be seen from Fig. 3, the rise in erosion rate with velocity is
very rapid above about 70 m s-l, particularly for the samples eroded at
an impact angle of 90”. The exponents given by plotting the results as a
power law are much larger than the value of 2 predicted by simple kinetic
energy considerations, and there are obviously other factors playing a part.
Figure 5 shows the surface features produced by erosion at an impact
angle of 30” at the two extremes of velocity (30 and 140 m s-l). In all the
micrographs of surfaces eroded at 30”, the erosion direction is from the top.
(a) (b)
Fig. 5. SEM micrographs of the surface of natural rubber specimens eroded at an impactangle of 30” (erosion direction from the top): (a) impact velocity of 30 m s-l; (b) impact
velocity of 140 m s-l.
7/30/2019 The Mechanisms of Erosion of Unfilled Elastomers
It can be seen that the features produced at the low velocity are very-
well-defined transverse ridges, as described previously [ 21. The similarity
between these features and the patterns produced by abrasion [9] should
be noted. At the higher impact velocity, however, the ridges although still
present are much more broken up and less well aligned. It seems to be a
general observation that a higher erosion rate leads to less-well-defined
ridges. Hutchings et al. [2] found that NR, with a low erosion rate, pro-
duced more-well-defined ridges than ENR50, with a higher erosion rate,
an observation that is confirmed by the present work.
Figure 6 shows the surface features produced by erosion at an impact
angle of 90”. At 140 m s-i, there is a rapid development of large and very
deep pits, many with silica particles embedded at the bottom. These pits
are absent at lower velocities (below 120 m s-l) and probably contributeto the very rapid rise in erosion rate with velocity at normal incidence.
The surface around the pits in the sample eroded at 140 m s-l is very rough
and almost granular in form. The surface features produced at a lower ve-
locity (90 m s-l) are about the same size, but the surface is much smoother
and can be seen to be merely a network of cracks.
(a) (b)
(cl (d)
Fig. 6. SEM micrographs of the surface of natural rubber specimens eroded at an im-pact angle of 90’: (a), (b) impact velocity of 90 m s-l; (c), (d) impact velocity of 140m s-l.
7/30/2019 The Mechanisms of Erosion of Unfilled Elastomers
4.2. The development of surface featur es i n nat ural rubber at an impact
angle of 30”
No observable damage was caused by single impacts. The impact sites
were visible when examined by SEM because silica debris in the form ofsmall particles (less than 10 pm) was evident over the impact site. These
particles originate from the surface of the impacting particle and adhere
to the elastomer surface on impact. From the size of the areas covered
with this debris, it seems that the size of the impact site is roughly the
same as the particle size, showing that the elastomer surface deforms appre-
ciably on impact.
The development of the surface features during the incubation period
is shown in Fig. 7. It should be noted in interpreting these micrographs
that, at an impact angle of 30”, 1 g of silica striking the surface leads to
about 200 impacts over each area the size of an impacting particle.
After erosion by 0.1 g of silica (about 20 successive impacts), only
isolated damaged areas are evident. These are in the form of raised ridges,
running roughly perpendicular to the erosion direction, with tears under-
neath (Fig. 7(a)). The facts that single impacts produce no observable
damage and that, after 20 impacts over each area, there are only isolated
areas of damage suggest that many successive impacts are required before
any surface damage becomes apparent.
(a)
(b) (d)
Fig. 7. SEM micrographs of the surface of NR samples at various stages during erosionat an impact angle of 30” and an impact velocity of 100 m s- 1 (erosion direction fromthe top): (a) after 0.1 g; (b) after 0.5 g; (c) after 5 g; (d) after 200 g (steady state).
7/30/2019 The Mechanisms of Erosion of Unfilled Elastomers
impact is evenly spread over the impact area, it corresponds to a depth
per impact of slightly less than 1 pm, about the same as the scale of the
irregularities in the subsurface cracks.
4.4. Su r face featu r es i n epoxidi zed nat ural rubber
Micrographs of the steady state surface features of ENR50 eroded at
impact angles of 30” and 90” are shown in Figs. 12(a) and 12(b) respec-
tively. There are no significant differences between the surface features
and their development in ENR50 and NR. .At 30”, the ENR50 has a surface
which is more broken up than that of NR, concomitant with its higher
erosion rate. At 90”, there is very little differences in surface appearance
between the two elastomers.
(a) (b)
Fig. 12. SEM micrographs of the steady state surfaces of samples of ENR50 eroded at an~~~o~t velocity of 100 m s-l: (a) impact angle of 30” (from the top); (b) impact angle
4.5. The mechani sm of mat eri al removal and the effect s of lubr i cat i on
The difference produced in the erosion rates of elastomers by lubrica-
tion (Fig. 4) reveals a great deal about the erosion mechanism. For a cutting
or ploughing process, lubrication would be expected to have little effect
or to increase the erosion rate due to reduced friction on the cutting edge
of a particle. This effect is probably responsible for the increase in erosion
rate with lubrication seen with the mild steel specimens. The dramatic
reduction in erosion rate with lubrication seen with the elastomer samples
at both 30” and 90” shows that it is the surface tensile stresses caused by
the impact that are important. At an impact angle of 30” a reduction in
friction between the elastomer surface and the impacting particle would
cause the surface tractions behind the impacting particle to be lower, leading
to a reduction in the erosion rate.
At normal incidence, as Poisson’s ratio for rubber is approximately
0.5, the surface tensile stresses due to an impacting particle will be pre-
dominantly frictional in nature. A reduction in friction due to lubrication
would cause the surface tensile stresses to be lowered, leading to the ob-
served reduction in erosion rate.
7/30/2019 The Mechanisms of Erosion of Unfilled Elastomers
1 V. K. Agarwal, D. Mills and J. S. Mason, A comparison of the erosive wear of steel
and rubber bends in pneumatic conveying system pipelines, in J. E. Field and N. S.Corney (eds.), Proc. 6th Int . Conf. on Erosion by Li qui d and Sol i d Impact, Cavendish
2 I. M. Hutchings, D. W. T. Deuchar and A. H. Muhr, J. M ateria ls Sci., 22 (1987)
4071 - 4076.
3 R. E. Morris and J. Oser, Rubber Age, 92 (1963) 96 - 103.
4 A. I. Marei and P. V. Izvozchikov, Determination of the wear of rubbers in a stream
of abrasive particles, in D. I. James (ed.), Abr asion of Rubber, McLaren, London,
1967, pp. 274 - 280.
5 G. M. Bartenev and N. S. Penkin, Sou. J. Frict. Wear, I (1980) 584 - 594.
6 S. Soderberg, S. Hogmark, U. Engman and H. Swahn, Tri bol. Znt ., 14 (1981) 333 -343.
7 I. A. Abu-Isa and C. B. Jaynes, Abrasion of elastomers by small particle impacts,
Res. Pub l . 4996 1985 (General Motors).
8 D. Bulgin and M. H. Waters, Abrasion of elastomers under laboratory and service
conditions, Proc. I nt . Rubber Conf., McLaren, London, 1967, pp. 445 - 469.
9 E. Southern and A. G. Thomas, Pl ast. Rubber: M at er. Appl ., 3 (1978) 133 - 138.10 A. Schallamach, J. Appl. Poiym. Sci., 12 (1968) 281 - 293.
11 J. C. Arnold and I. M. Hutchings, J. M ater. Sci., 24 (1989) 833 - 839.
12 Y. Uchiyama, In t . Pol ym . Sci. Technol ., 11 (1984) 93 - 102.13 A. H. Muhr, T. J. Pond and A. G. Thomas, J. Chim. Phys., 84 (1987) 331 - 334.14 A. W. Ruff and L. K. Ives, Wear, 35 (1975) 195 - 199.
15 I. M. Hutchings and R. E. Winter, J. Phy s. E, 8 (1975) 84 - 86.16 J. C. Arnold and I. M. Hutchings, Wear, 128 (1988) 339 - 342.