J. not. Rubb. Res.

, 4(1), 1-21

Surface Treatment of Rubber to

Reduce Friction
A.D. ROBERTS* AND C.A. BRACKLEY*

Traditional methods for reducing the surface friction of rubber articles are reviewed and
explored experimentally. Results of simple friction tests are given for various treatments. These
include halogenation, oxidation and acid attack. The general conclusion is that the treatments
cause surface roughening which will contribute to reduced friction.

The aim of many surface treatments of rubber, chlorinated material is hard and probably has
traditionally performed by wet chemical attack, a cyclic structure 5 . Some details about the
is to reduce the surface tack and sliding friction reaction chemistry are available 4 - 5 - 6 .
of rubber articles, such as sheeting, liners and Chlorinated rubber can be obtained by directly
wipers. The problem in industry is to perform treating sheets of rubber, rubber in solution or
this simply and safely. latex with chlorine gas. Hydrogen chloride gas
adds readily to NR. The product remains flexible
This study presents the results of simple if the chlorine content is less than 30%; but
friction tests for various surface treatments. above this figure, it is hard and brittle.
The treated rubber samples have all been
compared in an identical way by rubbing Hypochlorous acid reacts with NR in solution
against a glass plate possessing small undula- but the chlorine content of the products is
tions in its surface ('wavy' glass). Not only are higher than expected from simple addition,
initial friction coefficients compared, but also suggesting substitutive chlorination of the
any change in value with continual rubbing rubber by free chlorine present in the reagent.
(durability test). The results show that a range The use of bleaching powder to reduce the
of friction levels can be obtained according to surface tack of rubber was described long ago7
the treatment employed, and that different and was once more widely used than nowadays.
treatments have different durabilities. When chlorine water or bleaching powder is
used, the strength of the reagent and time of
REVIEW OF TREATMENTS exposure require careful control to avoid
excessive hardening of the rubber surface.
Natural rubber (NR) is very reactive towards Typically, the reagent solution should contain
the halogen elements, acids and some organic about 0.3% free chlorine and exposure is about
halogen-containing compounds. Reactions with 2 min. It is reported8 that further treatment of
these are generally complex, involving both the hypochlorinated rubber surface with oxalic
addition and substitution. Cyclisation of the acid solution can decrease the friction coefficient
rubber molecule may ensue, but the detailed and prevent corrosion of contiguous metal
chemistry of many of the reactions is not parts.
known even though halogen reagents were
among the first with which NR was treated More complex ways have been reported for
many years ago1'2. In the case of chlorine, if its introducing chlorine into a rubber surface. A
content after reaction is less than 40%, the chemically resistant, less frictional surface is
product is still soft and flexible, but when claimed9 by treatment with a chlorine gas-
65% is reached rubberiness is lost. The fully sulphur dioxide mixture in the presence of

*MaIaysian Rubber Producers' Research Association, Brickendonbury, Hertford, SG13 8NL, United Kingdom

1
 Journal of Natural Rubber Research, Volume 4, Number 1, March 1989

ionising radiation. Alternatively, changes Interestingly to note, in 1961 the use of

occurring to NR by ihe action of chloro- peracclic acid to remove surface tack on rubber
sulphonyl isocyanate have been described 1 ". was recommended by a food company 19
The use of concentrated hypochlorous acid con- as a substitute for halogenation because it
taining an oxidising agent has been patented 11 was claimed to be as good, but safer. There
to make rubber surfaces resistant to abrasion. have been many attempts to substitute haloge-
On a wider point, quite how these and other nation for something else. For example, the
chlorinated rubber surfaces age is something yet following may be cited; grit blasting 20 , Teflon
to be fully explained. coating 21 , molybdenum disulphide 22 , graphite
powder or paint 23 , ultraviolet light to produce
Fluorine gas reacts violently with NR, but surface hardening 24 or polymer grafts 16 - 25 - 211 ,
products containing up to 30% fluorine have metal plating 2 '' 2 ", surfactant treatment 29 , glow
been obtained which are rubber-like. For discharge/plasmas in vacuum™- 31 .
example, articles such as surgeons' gloves can
be treated 12 by extending their surface by 10% The main object of surface chlorination of
by inflating with a fluorine/nitrogen gas rubber articles is to reduce their dry friction,
mixture. Alternatively, a rubber mix may be often against glass. A bonus is that it can
treated with a fluoro-organic compound and also slow down surface ageing. Frequently, a
after vulcanisation the article placed under surface treatment is simply described in terms
reduced pressure to subject it to glow dis- of a reduction in friction and little said about
charge 13 . Its surface will then have a low fric- the physico-chemical nature of the modified
tion coefficient and enhanced wear resistance. surface. Studies by scanning electron microscopy
Another way would be to treat a rubber surface (SEM) made sometime ago32 showed that
with gaseous antimony pentafluoride in nitrogen chlorination increases the fine scale roughness
gas as carrier 14 ; thus treated friction coeffi- of a rubber surface, with a corresponding
cients are claimed to decrease by a factor of decrease in dry friction. More recently, the
six times. Other ways are by fluoroplastic- chemistry of chlorine addition has been examined
containing lacquers applied to a rubber surface' 3 by XPS and the physical effects of chlorination
or by surface grafts 16 . on surface morphology noted 33 .
Anhydrous hydrogen fluoride has a powerful Two other classes of chemical treatment of
cyclising action on rubber. At low temperatures rubber surfaces must be mentioned. Sulphuric
(-150°C) cyclisation can be suppressed and acid has been used to reduce the tack of raw
addition of HF to the double bond is obtained, rubber. It renders the surface hard and micro-
the product containing 16% fluorine. It is rough 34 . Vulcanised rubber sheets have been
rubber-like although weak, but ages well. treated with dilute sulphuric acid to improve
their anti-blocking properties35. An extension
When NR is treated in solution at low of this comprises treating with an aqueous solu-
temperature with bromine, the element adds to tion containing an aromatic sulphonic acid36.
the double bond with very little substitution. The other treatment is based on permanganate.
Surface treatment with bromine water has been Some time ago, a study was made of the
used to impart surface hardness to certain products of surface reaction of aqueous acidified
rubber articles such as windscreen wiper blades. potassium permanganate with NR 3 ".
The effect of bromination on the surface
properties of rubber has been reported 17 ' 1 *.
The addition of hydrogen bromide to the rubber EXPERIMENTAL
molecule proceeds similarly to that of hydrogen
chloride. The frictional force between a rubber sample
(treated or not) and 'wavy' glass was measured
No well defined iodinated derivatives of using simple apparatus. This consisted of a
rubber have been reported. The addition of rubber hemisphere fastened to the traverse stage
hydrogen iodide has been but little studied. of a microscope, the traverse being driven by

COPYRIGHT © MALAYSIAN RUBBER BOARD

 A.D. Roberts and C.A. Brackley: Surface Treatment of Rubber to Reduce Friction

an electric motor. To the top of the rubber The NR sample vulcanisates used were
hemisphere was fixed the rubber sheet sample formulated as indicated (Table I) and moulded
under test. It was laid over the hemisphere and as 1 mm thick sheet. Rubber hardness was
held in place at its edges with sticky tape. The measured on a suitable meter. From the IRHD
'wavy' glass plate was part of a lever arm and value, Young's modulus, E, was directly found
rested under an applied load on the rubber lest from an established curve of the variation of
sample (Figure I). The applied load was 4 N, E with IRHD. Glass transition temperatures
which gave a contact circle of diameter 10 mm were measured by differential scanning calori-
with untreated rubber. When treated rubber metry. The moulded rubber samples were cut
was used there was no well defined contact to size (about 400 mm 2 ) for friction tests. Cut
periphery. The sliding distance was 27 mm and samples were cleaned with pure acetone. This
this was traversed at a fixed nominal speed of consisted of wiping a surface with cotton wool
1 mm per second, the motion being to and fro, soaked in acetone followed by brushing to
with the aid of a reversing switch. In this way, remove residual matter, the brush having been
durability of the surface treatment could be dipped in acetone. Thus cleaned, samples were
assessed. either directly friction tested or surface treated.
When treated, any residues were rinsed away
The friction force between rubber and glass with de-ionised water and the surface dried in
in the horizontal plane was detected by strain a stream of cool air. Thereafter, the surface was
gauged leaf springs on the lever arm. Strains friction tested at room temperature.
were measured on a suitable electric bridge, the
output amplified and displayed on a chart To treat the surface of rubber samples
recorder. The leaf springs were calibrated by chemically involved their immersion in appro-
applied forces in the horizontal plane, so that priate aqueous solutions. The solutions were
displacements on the chart recorder were stirred during treatment. The compositions of
known in terms of force. Friction coefficients solutions used are given for reference (Table!),
were then taken as that force divided by the though minor variants are likely to produce
applied normal load. similar results.

Load

Wavy glass 1

of rubber sample
Rubber
hemisphere
* r^™
1
1••-i
Sticky tape
Lever arm

Traverse

Figure I. Sliding contact arrangement for assessing various surface treatments of sheet rubber
samples. Sheet thickness ca. / mm, rubber hemisphere diameter 37 mm.
 TABLE 1. RUBBER SAMPLE FORMULATIONS

Formulation (parts by weight)

Compound
A B C

NR (SMR L) 100 100 100

Zinc oxide 5 5
Stearic acid 2 2
Fleet ol H 1
Santoflex 13 2
Dicumyl peroxide 1
Sulphur 0,5 2.5
Sulphasan R 2
MOR 1
Robac P2 0.2
N672 SRF black 60
Nonox ZA 1
Samocure NS 0.5

Cure time (min)/temp (°C) 24/150 60/150 40/140

Hardness (IRHD) 66 29 44

TABLE 2. COMPOSITION OF TREATMENT SOLUTIONS

Solution
Reagent
1 2 3 4 5 6 7 8 9

Sodium hypochlorite
aq. sol. (14% Cl) (ml) 5
Concentrated hydrochloric acid (ml) 2 90
Water (ml) 200 100 100 50 80
Bromine (ml) 5
Potassium iodide (g) 10
Iodine (g) 1
Potassium permanganate 10~ 4 M to
(variable concentrations in water) KT'M
Sulphuric acid (m!) IO~ 2 M 10~ 2 M 50
Potassium dichromate
(aq. sol.) 10~ 2 M
Concentrated nitric acid (ml) 30
40% Peracetic acid (ml) 20
Hydrogen peroxide (ml) 25
90% Formic acid (ml) 5
 A.D. Roberts and C.A. Brackley: Surface Treatment of Rubber to Reduce Friction

RESULTS distilled water and dried with cool air. Results

for a gum and a carbon-filled NR vulcanisate
Surface Chlorination (Figure 2) show that chlorination produces
The traditionally preferred method for about a three- to four-fold decrease in initial
reducing tack and friction may be regarded as sliding friction. Friction levels for the two
chlorination, usually by hypochlorite solution samples of carbon-filled vulcanisate are dif-
or chlorine water (gas bubbling). We have ferent because in one case the surface was very
friction tested after using the hypochlorite slightly talced beforehand. Traces of talc were
method (Solution 1, Table 2). Rubber samples present before chlorination, and on the un-
were suspended in the treatment solution for treated sample. Surprisingly, at long rubbing
3 min neutralised with 2% ammonia, rinsed in times, the differences between treated and un-

°——° Gum (Compound B) untreated, °——o chlorinated

•——• Carbon-filled (Compound A) untreated, •——• chlorinated
x
——x Carbon-filled (Compound A) residual talc, *——* plus chlorination

2.0

1.5

1.0

O
U
0.5

10 20 30 40
Rubbing time (min)
Figure 2. Effect of surface chlorination on the dry sliding friction of gum and carbon-filled NR
vulcanisates in reciprocal movement against 'wavy' glass.
 Journal of Natural Rubber Research, Volume 4, Number 1, March 1989

treated rubber surfaces diminish. Untreated To iodinate sample surfaces, the same
gum rubber friction increases with rubbing time method as above was employed with treatment
probably due to the enlarged contact area, Solution 3 (Table 2). Fluorination was carried
whereas untreated carbon-filled rubber decreases out by Bettix Ltd (UK) using their patent
due to surface roughening caused by abrasion. process.
Talc alone has a marked effect in reducing the A comparison of all the halogens for gum
friction of untreated rubber. and carbon-filled rubber is shown in Figures 3
and 4. Fluorination of gum rubber resulted in
Other Halogens very low friction and good wear resistance; but
for carbon-filled rubber, bromination gave the
A saturated solution of bromine (Solution 2, lowest friction.
Table 2), was used to treat rubber samples for
3 min. Thereafter, the sample surfaces were Oxidation
neutralised with 5% ammonia, rinsed with
distilled water and blow dried with cool air. The Previous work37 has shown that a polyglycol
brominated samples were then friction tested. film is formed on the surface of NR when

Untreated
—— — Brominated
x——* lodinated
——— Fluorinated
2.0 — -— Chlorinated

-X————X———X-

1.5

1.0

o
U
0.5

0
0 10 20 30 40
Rubbing time (min)
Figure 3. Comparison of halogen treatments for gum NR vulcanisate (Compound B).
 A.D. Roberts and C.A. Bracklev: Surface Treatment of Rubber to Reduce Friction

Untreated
—--— Brominated
x———x lodinated
——— Fluorinated
2.0 • — - — Chlorinated

=1 1.5
_c
o
'i—
'—
'o 1.0

Q 0.5

0 1 2 3 4 5 6 7
Rubbing time (h)
Figure 4. Comparison of halogen treatments for carbon-filled NR vulcanisate (Compound A).

treated with dilute acidic potassium per- the friction and surface roughness re-measured.
manganate. Unless the permanganate concen- The test was repeated for different treatment
tration is very low, other oxidative scission times and permanganate concentrations by re-
reactions occur at the same time. Tests were abrading the rubber to expose fresh surface.
undertaken to determine the treatment effect Results showed that permanganate treatment
of various permanganate concentrations in increased the surface roughness (Table 3)
0.01M sulphuric acid. according to solution concentration, and the
friction measurements showed a corresponding
A slightly modified experimental arrange- fall in level (Figure 5). This may be interpreted
ment was employed to that shown in Figure 1. as a fall in friction with increasing surface
A sulphur-cured rubber (Compound C, Table I) roughness.
moulded as a 25 x 25 mm square flat 10 mm
thick was first abraded with silicone carbide Rubber samples soaked in permanganate
paper (Tri-M-ite I80E) in a 'standard' manner. solutions acquired a patchy purple coating and
After acetone cleaning and surface roughness a black residue could be rubbed off. In view
measurements (Talysurf 10) had been taken, the of this, durability tests were carried out. The
flat was fixed to the friction lever arm and slid rise in friction with just a few to and fro
under a load of 10 N against 'wavy' glass. traversals against wavy glass was noticeable
Thereafter the flat was removed and placed in (Figures 6 and 7). Even so, in the longer term,
permanganate treatment solution (Solution 4, the rate of rise in friction diminished and the level
Table 2) for a given time, then rinsed, dried and remained below that of an untreated surface.
 Journal of Natural Rubber Research, Volume 4, Number 1, March 1989

TABLE 3. EFFECT OF PERMANGANATE TREATMENT ON SURFACE ROUGHNESS

Permanganate Immersion time Talvsurf Roughness

concentration (M) (min) (/im'CLA) (°/u increase)

None 3.58
io- d 30 3.70 3.3
3
io- 9 3.80 5.3
10"3 2 3.95 8.8
10-' 0.5 4.05 13.1

Treatment solutions were various mo!ar strengths of KMnO 4 in 10 M H,SO4

— — Untreated
x Permanganate dipped

0.50

o
s

£ 0.25
14-,
<L)
O
U

-4 -3 -2 -1
Permanganate concentration, Lg (M)
Figure 5. Surface treatment of gum NR vulcanisate (Compound C) with acidified solutions of
potassium permanganate.

For comparison, the effect of potassium Acid Attack

dichromate as oxidant was tried (Solution 5).
The rubber sample was treated in the same way The effectiveness of various concentrations
as for permanganate solution. The friction of sulpuric acid was investigated. Results for
results (Figure 6) suggest it is less effective than 3 min immersion of 1% dicup NR samples in
permanganate. acid (Figure 8) show a considerable fall in
 1.5

1.0

o
U

0.5 — — — 0.01 M permanganate in 0.01 M sulphuric acid

x—x—x 0.01 M dichromate in 0.01 M sulphuric acid
/ ———— Untreated

0
0 10 20 30 40
Rubbing time (min)
Figure 6. Comparison of oxidising agents for the surface treatment of gum NR vulcanisate.

Untreated
——— 0.1 M permanganate in 0.01 M sulphuric acid
- - --- 0.3 M permanganate, no acid

- 11
f \

o
U

10 20 30 40
Rubbing time (min)
Figure 7. Surface treatment of carbon-filled NR vulcanisate (Compound A) with potassium
permanganate solutions.
 Journal of Natural Rubber Research, Volume 4, Number 1, March 1989

-o 25% Sulphuric acid

- Untreated
-• Concentrated sulphuric acid

1.5
=4.
C
O

o 1.0

o
U
0.5

0
0 10 20 30 40
Rubbing time (min)
Figure 8. Sulphuric acid treatment of gum NR vulcanisate (Compound B).

friction for concentrated acid, but much less Surface Coating

when the acid is dilute.
Some rubber samples were coated with
The effect of some acids on very slightly 'Satreat' halogen primer18 which was scrubbed
talced carbon-filled rubber was examined. onto surfaces with a stiff bristled brush. Any
Immersion times varied, as indicated (Figure 9). residue was removed with 25% ethanol. Samples
All treated samples were rinsed with water, so treated were smooth to touch, had a 'milky'
neutralised in ammonia, rinsed again with appearance and gave low friction (Figure 10).
distilled water and dried at room temperature. Whether it is physical rough'ening by brushing,
The friction of the untreated rubber appears or halogen attack, that reduces the friction, is
surprisingly low. This is probably due to not clear.
residual talc particles. The high initial friction
with nitric acid may be due to the dissolution Samples were PTFE coated (Fothergill
of residual talc. Clean, untalced carbon- Engineered Surfaces Ltd) using an air cure resin
filled rubber showed higher levels of friction based system. Coated surfaces were smooth to
(Figure 10). the touch, but the coating cracked when rubber

10
 A.D. Roberts and C.A. Brackley: Surface Treatment of Rubber to Reduce Friction

Untreated
1.00
\. \
——— 10% aq. peracctic acid, 25 min (Solution 7, Table 2)
• 50% Sulphuric acid, 5 min (Solution 6, Table 2)
+•—— -(-Concentrated nitric acid, 2 min

0.75

0.50

0.25

10 20 30 40
Rubbing time (min)
Figure 9. Attack by various acids on slightly talced carbon-filled NR vulcanisate (Compound A).
Immersion times in acids are as indicated,

samples were flexed or stretched. Although the the effects of ageing upon friction of chlorinated
friction was low (Figure 10) and the wear rubber surfaces by either heating them in an
resistance good, the coating could be all too oven or exposing them to direct sunlight (in and
easily stripped off with sticky tape. outdoors).
Ageing Carbon-filled samples (Compound A,
Table I), some chlorinated several weeks
It is generally believed that for exposed before, were aged by heating at 120°C in an
rubber articles during their working life time oven for 15 min. It is noted that the chlorinated
chlorine in their surface skin eventually .samples had been stored in a polybag in the
becomes replaced by oxygen without loss in low dark for several weeks and during this period
friction. Exactly what happens on ageing is not displayed a low value of friction coefficient.
clear. Earlier work 39 has shown that photo- After heat ageing all samples were immediately
oxidised rubber (achieved by placing sample in friction tested. The results indicate (Figure II)
direct sunlight) acquires a hardened skin and reduced friction, even for chlorinated samples,
its friction coefficient can fall to the order of though brominated surfaces showed no change
0.2. In this study, we have attempted to examine after heating. The surface roughness of heat

11
 Journal of Natural Rubber Research, Volume 4, Number 1, March 1989

•——— Hydrogen peroxide/formic acid, 1 h (Solution 9 Table 2)

• Untreated
• • • - • • - - - Concentrated nitric acid, 3 min
— Satreat (Ref. 38)
Aqua regia, 3 min, (Solution 8 Table 2)
2.0 * Hypochlorous acid, 3 min (Solution 1 Table 2)
coating

c
o 1.5

o>
O
U

0.5

0 1 2 3 4
Rubbing time (h)
Figure 10. Various coatings and acid treatments of dean, freshly prepared, carbon-filled NR
vulcanisate samples (Compound A). Immersion times of samples in acids are indicated.

aged samples was measured (Talysurf 10) and complex was the setting-up of samples outdoors
found to be greater than unaged samples on an inclined (45°) board facing south and un-
(Table 4). This may, in part, explain the fall protected from wind and rain. In all cases, the
in friction caused by heat ageing. friction fell with increasing exposure time
(Figure 12). Conversely, the surface roughness
The effect of light exposure in air (photo- tended to increase with exposure time (Figure 13).
oxidation) was studied for carbon-filled samples It was interesting to find that for outdoor
(Compound A, Table I) in two ways. The exposure, chlorinated and untreated samples
simplest involved measuring the friction of gave the same low level of friction after a couple
chlorinated and untreated rubber samples of weeks.
before and after placing them on an indoor
window-sill in the sun. The change in friction When a plot is made of friction coefficient
was followed with the time of exposure. More against surface roughness all the above results

12
 A.D. Roberts and C.A. Bracklev: Surface Treatment of Rubber to Reduce Friction

-+ Untreated surface
~x Untreated, after heat ageing
o -° Chlorinated surface
•———• Chlorinated, heat aged
D———D Brominated surface
• ——• Brominated, heat aged

2.0

1.5

c
_o
;j
£

2 i.o
"c
"5

o
U

0.5

10 20 30 40
Rubbing time (min)
Figure 11. Heat ageing see treated and untreated carbon-filled NR vulcanisate
(Compound A).

for heat ageing and sunlight exposure fall on ting to see that surface chlorination of such
the same curve (Figure 14). This is strong samples delayed degradation by a factor of four
evidence for the role played by surface rough- in time (Table 5).
ness in reducing the level of sliding friction,
whatever the way of ageing. ANALYSIS

Additional outdoor sunlight exposure tests Surface analysis of chemically treated rubber
were carried out on unprotected rubber samples samples was made by using a profile machine
(Compound B, Table 1). Not unexpectedly, (Talysurf 10) and by examination with a scan-
these degraded quite quickly, but it was interes- ning electron microscope.

13
 Journal of Natural Rubber Research, Volume 4, Number 1, March 1989

TABLE 4. SURFACE ROUGHNESS OF HEAT separation (wavelength). The typical height of

AGED SAMPLES asperities, their number per unit length and the
Surface roughness (/im)
Ra values are tabulated with initial friction
values (Table 6). The Young's modulus, E, of
Sample Before After
samples was determined using a micro-hardness
heating heating
gauge. The values would appear to reflect bulk
Untreated 0.20 0.24 rather than surface modulus (Table 6). In all
Chlorinated 0.29 1.10 cases, it is apparent that treatment increased the
surface roughness, to varying extents, con-
Brominated 0.80 0.82
comitant with a fall in friction. In addition,
Compound A. Talysurf meler cut-off 0.8 mm surface hardness changes are likely to play an
important role, but this factor is difficult to
Surface Roughness assess owing to a general inability to measure
the surface hardness of the thin skin layer.
For surface profile measurements, rubber
samples were stuck with adhesive tape onto a The following attempts to account for the fall
glass plate and the treated surface lubricated in friction after surface treatment in terms of
with soapy water. Centre line average (Ra) the induced change in surface roughness and
roughness measurements were taken using hardness. If it is assumed that the frictional
0.08 mm and 0.8 mm cut-off lengths at 5000 X force, /, is proportional to the real area of
vertical magnification. Charts of treated sur- contact, A, then
faces were recorded at 20 x or 100 x horizontal / = As ...1
magnification. Typical results are shown in
Figure 15. The charts were assessed visually to where s is the interfacial shear strength. For the
find values for the radius of asperities and their point contact of a spherical rubber asperity on

•• Untreated, indoors
-x Untreated, outdoors
+———+ Chlorinated, indoors
D———n Chlorinated, outdoors

flj
o
U

8 12 16
Time (weeks)
Figure 12. Sunlight exposure of surface chlorinated and untreated samples of carbon-filled NR
vulcanisate (Compound A). Carried out in UK (52°N) during spring/summer.

14
 • Untreated, indoors
x Untreated, outdoors
f- Chlorinated, indoors
Chlorinated, outdoors

0.4

tu
c
oc 0.3 *£_/+---+---
3
o

0.2

0 4 8 12 16
Time (weeks)
Figure 13. Change in surface roughness on sunlight exposure in UK (Compound A).

o Heat aged, 5 min friction coefficients, Figure 11

• Sunlight exposed, data from Figures 12 and 13

=*.
cT
o

0.2 0.4 0.6 0.8 1.0

Surface roughness, R.,
Figure 14. Friction-roughness variation for aged samples of carbon-filled NR vulcanisate
(Compound A).
 Journal of Natural Rubber Research, Volume 4, Number 1, March 1989

TABLE 5. OUTDOOR WEATHERING OF the radius of the circle of contact. Thus

UNPROTECTED SAMPLES
9K
Exposure time Friction coefficients W~
~\6E
(weeks) Untreated Chlorinated
I f the rubber sphere represents a single asperity
0 1.50 0.25 on the surface of a rubber sample, then the total
1 0.97 0.25
friction is
2 0.90 0.23 , 9K . ™ (P .
jJ, = KITS ( —— ) ( — )
4 Degraded 0.21 \6E n
8 0.11 where n is the total number of asperities in
16 Degraded contact and P is the total applied load to the
rubber sample. This assumes all asperities are
Compound B of the same radius and height. Although a
considerable assumption, it brings out the
a rigid plane, Equation 1 may be re-written general point that the friction is likely to
to express the friction coefficient, f^asf,( = -V M ,), depend on total asperity numbers and their
related to the load, 14', the sphere radius, R, radius. The Hertz equation applies to a semi-
and its Young's modulus, E, through the use infinite body of uniform modulus, whereas in
of Hertz equation a- = 9l¥R/]6E, where a is the present problem each asperity of modulus

60 p
50 -
40
30
20
10
0
Before surface treatment

5000

After bromination
Figure 15. Surface profile charts of gum NR vulcanisate samples (Compound B) showing before
surface treatment and after bromination.

16
 A.D. Roberts and C.A. Bracklev: Surface Treatment of Rubber to Reduce Friction

TABLE 6. SURKACE ROUGHNESS OF TREATED RUBBER SAMPLES

Trcamicnt
E R I X R n n4/V-'
M {MIVl ( pm) (filll) (p.m) dim) (mm "") (iogio)
Noni; 1.55 0.9 0.06 0.! 100 12500 235 6.09
0.05 100 25 000 125 5.72
t- 0.15 1.0 0.38 1.2 100 1 040 110 4.73
1.6 100 780 85 4.49
Cl 0.50 1.0 0.29 0.4 100 3 125 380 5.76
0.6 100 2080 330 5.56
B,r 0.32 1.1 0.36 0.8 250 9 765 125 5.45
1.0 200 5 000 100 5.12
I 1.4? 0.9 0.2 0.2 .350 76 560 80 5.78
0.2 200 25 000 100 5.59
Aqua rt'gia 0.25 1.1 0.25 0-4 50 780 270 5.16
1.6 150 1 760 80 4.69
Sat real 0.40 0.9 1.55 0.8 100 1 560 180 5.13
KMnO 4 ?% aq. 0.60 1.0 0.35 1.6 300 7030 160 5.50
1.2 200 4 170 230 5.56
KMnO.i acidic 0.26 1.1 0.58 2.0 100 625 400 5.33
1.0 150 2 810 100 4.96
H2SO4 0.56 1.15 0.2 0.3 200 16660 225 5.94
0.45 1.25 0.27 0.3 100 4 160 200 5,47
H 2 S0 4 <25o/o) 1.40 1.0 0.16 0.4 125 7 880 200 5.52
HNO, 1.44 0.95 0.24 0.2 100 6 250 200 5.60

u = Friction coefficient; E = Young's modulus; Compound B

Ra = Centre line average roughness at 0.8 mm cut-off
t = Peak-to-valley height
X = Wavelength of asperities
R = Asperity radius = XV8i
n = Estimated number of asperities from Talysurf trace

E rests on a softer rubber bulk. However, with of asperity shape, as derived through measures
the assumption that all asperities are of the of asperity height and wavelength. Quite how
same radius and height, the error associated to handle parameter plots of high friction
with applying the simple Hertz equation should surfaces ( j u > 1.4) is not clear. Such surfaces
not be significant. appear (under the microscope) to form a
continuous contact over extensive areas when
The surface roughness results (Table 6) are pressed against wavy glass, whereas contact is
plotted in two ways. One is as a variation of visibly discontinuous (looks frosty) for low
friction coefficient with the parameter « 4/3 /? 2/3 friction surfaces (/i < 0.6). What does appear
as suggested by Equation 3, the other is simply to be more certain is that surface roughness
with Ra. The plots (Figure 16) indicate the influences the level of friction, outweighing any
importance of roughness in determining the physico-chemical interactions at the sliding
friction level. Better account needs to be taken interface. This is not to say that such interac-

17
 Journal of Natural Rubber Research, Volume 4, Number 1, March 1989

1.5

I 1.0

x+
0.5 + X

o
U
•H + x

+ X

0.2 0.4 0.6 4 5 6

Ra (/tm) 1 g n 4/3 R2 (jim2/3)
Surface parameters
Figure 16. Variation in friction with surface roughness (Compound B).

tions do not exist, but compared to roughness surface roughening (Figure 17). Mild attack
they are secondary in effect. (insufficient chlorine) shows only a small
change in appearance (Figure 17) but full
Scanning Electron Microscopy attack (3 min in 0.3% chlorine solution) gives
considerable surface modification (Figure 17).
Variously treated rubber samples were Difficulties in the manufacture of articles
examined by SEM. The result of immersion for involving surface chlorination may be associated
4 min in saturated bromine water was a matt with insufficient chlorine in solution (old
appearance to the sample, and SEM showed the reagents), so resulting in too high a level of
gross nature of the bromine attack {Figure 17). friction.
Electron probe micro-analysis indicated a high
level of bromine chemically bonded to the
rubber. There appear to be asperities upon CONCLUSION
asperities. How to relate these photographs to
the Talysurf trace (Figure 15) of the same By using Talysurf measurements, coupled with
surface is not clear. The Talysurf diamond SEM photographs, it becomes clear that
stylus has a square pyramid tip of side 2 /im. chemical surface treatments produce surface
Presumably features on a finer scale than this roughening of a rubber sample, and this
will not be resolved, so the Talysurf trace reduces the friction because the real area of
represents an average of the more gross features contact is diminished. It is observed that
seen in the SEM photographs. different treatments produce different degrees
of roughness. Thus a judicious choice of treat-
Chlorine attack by the standard method of ment could help to achieve the desired level of
acidified hypochlorite solution also shows friction.

18
 1

Brominated Brominated

Mild chlorination Full chlorination

Figure 17. Scanning electron micrographs of surface treated gum NR vulcanisate (Compound B).
 Journal of Natural Rubber Research, Volume 4, Number 1, March 1989

ACKNOWLEDGEMENT 16. DOEDE, C.M. (1961) Slippery Rubber. Rubb. Age,

89, 110.
Thanks are due to Dr K.N.G. Fuller for helpful 17. ORLOV, V.A. AND ROSTOVTSEVA, E.E. (1975)
comments. Surface Bromination of NR Vulcanizates for
Improvement in Performance Characteristic. Proc.
December 1988 Int. Rubb. Conf. 1975 Kuala Lumpur, 4, 263-
18. RADNOTI, A., SERA, J., SCHOBER, M. AND
FEDENYI, G. (1978) Surface Treatment of Rubber
REFERENCES with Bromine, ffuanyages Gummi, 15, 363.
1. PARKES, A. (1846) Treating Caoutchouc and Gutta 19. FOOD MACHINERY AND CHEMICAL CORP.
Percha, Br. Pat. No. 11147. (1961) Peracetic Acid Treatment of Vulcanized
Rubber. Br. Pat. No. 861 660.
2. GLADSTONE, J.H. AND HIBBERT, W. (1888) The
Optical and Chemical Properties of Caoutchouc. J. 20. POHR, H.C., RYCK, F.M., WALL1S, C.T. (I960)
Chem. Soc., 53, 679. Squeegee. General Motors U.S. Pat. No. 2952865.
3. LEE, D.F., SCANLON, J. AND WATSON, W.F. 21. RUNTON, L.A. AND MORTON, H.C. (1959)
(1963) The Cyclization of Natural Rubber. Proc. Russell Mfg Co. U.S. Pat. No. 2908028.
R. Soc., A273, 345. 22. SYMBOLIK, W.S. (1963) Squeegee Type Wiper
4. NAUNTON, W.J.S. (1961) The Applied Science of Blade. General Motors. U.S. Pat. No. 3080596.
Rubber. London: Edward Arnold. 23. SLADE, R.W. (1969) Method for Surface Modifica-
5. ROFF, W.J. AND SCOTT, J.R. (1971) Fibres, Films, tion of Elastomer and Windshield Wiper Produced
Plastics and Rubbers. London: Butterworths. by Same. Ford Motor Co. Ltd. Br. Pat. No. 1147577.
6. STERN, H.J. (1967) Rubber Natural and Synthetic. 24. RUSCH, K.C. AND FORRESTER, J.R. (1971)
London: Butterworths. Frictional Characteristics of EPDM Windshield
Wiper Blades. Rubb. Wld, 165, 54.
7. WEBER, L.E. (1926) Chemistry of Rubber Manufac-
ture. London: Griffin. 25. GORELIK, R.A. (1980) Modification of the Surfaces
of Rubber Articles. French Pat. No. 2465761.
8. USSR PATENT (1983) Volzhskii Plant of Rubber
Technical Articles. Surface Modification of Indus- 26. SUTURES INC. (1975) Improvements in or Relating
trial Rubber Products. SU 1002312. to Rubber Articles Having Improved Slip Coaling.
9. POLITECHNIKA LODZKA: INSTITUTE PRZE- Br. Pat. No. 1398552.
MYZLY GUMOWEGO (1980) Superficial Resistant 27. INCO EUROPE LTD (1976) Chromium Plating of
Film on Rubber Products. Pol. Pat. 105791 Polyolefin Compositions. Br. Pat. No. 1534638.
10. PAUTRAT, R., METIVIER, B. ANDMARTEAU, J. 28. MILLER, B. (1978) Sputtered Chrome New Route to
(1975) Changes Occurring in Natural Rubber Plated Elastomers. Plast. Wld, 36, 56.
Through Ihe Action of Chlorosulphonyl Isocyanate.
Proc. Int. Rubb. Conf. 1975 Kuala Lumpur, 4, 29. DAI-ICHI KOGYO SEIYAKU CO. LTD (1974)
267. Adhesion Prevention of Rubber to a Contiguous
Surface. Br. Pat. No. 1524721.
11. DUNLOP LTD (1941) Improvements in Treating
Surfaces of cr containing Rubber or the Like. Br. 30. DUKHOVSKII, E.A., KLEIMAN, A.M., PONO-
Pat. No. 535 983. MAREV, A.N., SILIN, A.A., STEPANOV, B.M.,
TAL'ROZE, V.L., T1KHOMIROV, L.A. AND
12. VITEK INC; MARCHEM INC. (1978) Method of
KHOMYAKOV, A.V. (1984) Surface Modification
Treating Extensible Hydrocarbon Articles. Br. Pat.
No. 1522395. of Rubber Articles. USSR Pat. SU 642950.

13. ANATOLIEVICH, E. (1981) Manufacturing Surface 31. YASUDA, H.K. (1984) Modification of Polymer
Modified Rubber Articles. U.K. Pat. Applic. Surfaces by Plasma Treatment and Plasma Polyme-
No. 2062649A. risation. 187th ACS Meet, St. Louis, Symp. on
Polym. Wear and Its Control, Ser. 287, 89.
14. NUDELMAN, Z.N. ANDPROKUDIN, I.P, (1974)
Process for Surface Fluorination of Rubber Articles. 32. ROBERTS, A.D. AND TABOR, D. (1971) Extrusion
Br. Pat. 1378485. o f Liquids between Highly Elastic Solids. Proc. R.
Soc., A325, 323.
15. CHESNOKOV, N.M., SEMENOV, I.V., YABLON-
SKI, N.S. AND PAL'MOVA, N.I. (1979) Seal 33. ROMBERG, V.S. (1986) Aqueous Chlorination of
Rubbers with Antifriction Coatings. Kauchuk i NR Surfaces. ACS (Rubb. Div.) Meet., Spring
Rezina, 7, 42. paper.

20
 A.D. Roberts and C.A. Brackley: Surface Treatment of Rubber to Reduce Friction

34. ONG, E.L. AND ROBERTS, A.D. (1986) Experi- 37. DEAN, W.R., PERERA, V. AND GLAZER, J.
ments on Lubrication of Raw NR. J. nat. Rubb. (1958) Momilayer Studies of Some Hydroxylated
Res., 1(1), 41. Polyolefin Rubbers. /. Polym. Set., 21, 489.
38
35. TIGER POLYMER, K.K. (1981) Prevention of ' SATREAT (Registered Trade Mark for a Process of
Adhesion of Vulcanized Rubber Sheets. Japanese Chlonnation Using an Organ.c ^lo^ne Cor,tain,ng
Pat 81120328 Substance). Br. Pat. No. 1278258, 1295676, 1295677.
39. BARQUINS, M. AND ROBERTS, A.D. (1986)
36. BRIDGESTONE TIRE AND RUBBER CO. (1975) Rubber Friction Variation with Rate and Tempera-
Preventing Adhesion between Substrates. U.S. Pat. ture: gome New Observations. /. Phys. D; Appl.
No. 3990990. p^ 19, 547.

21