Nippon Gomu Kyokaishi, No. 10, 2010, pp.

331–336

Influence of moisture in uncrosslinked rubber on

mechanical properties and crosslinking reaction

K. Nakayama, T. Saito, and Y. Ohtake

Chemicals Evaluation and Research Institute, Japan
1600, Shimo-Takano, Sugito-Machi,Kitakatsushika-Gun, Saitame, 345-0043, Japan

Selected from International Polymer Science and Technology, 38, No. 1, 2011, reference NG 10/10/331; transl. serial no. 16263

Translated by K. Halpin

1. Introduction of the water content of the uncrosslinked rubber on cured

rubber properties has hitherto eluded quantification.
Rubber compounding and crosslinking conditions
are often subtly adjusted according to season and The study reported here investigated the effect of
manufacturers’ know-how. This is because of an empirical moisture in uncrosslinked rubber on crosslinking and
acceptance that the moisture in uncrosslinked rubber cured rubber properties for two general purpose rubbers,
affects crosslinking reactions. Previous studies have in NR and ethylene-propylene diene terpolymer (EPDM),
fact shown that crosslinking is faster in the presence of using the most commonly utilised sulphur and peroxide
water [1-3]: for example, it has been verified from the crosslinking agents.
curing curves of pure gum compounds of sulphur-cured
natural rubber (NR), styrene butadiene rubber (SBR) and
acrylonitrile butadiene rubber (NBR) with sulphenamide 2. Experimental
vulcanisation accelerator that the crosslinking reactions in
NR and SBR are accelerated the higher the water content
of the uncrosslinked rubber, while NBR is unaffected. 2.1 Production of samples
Crosslinking in silica compounds of NR and SBR is likewise
significantly accelerated by water in the uncrosslinked The recipes of the rubbers compounded are shown in
rubber, while the effect in silica compounded NBR is Table 1. The rubber stock materials used were JSR’s EP33
small [1]. In compounding with carbon, the water in the (ethylene content 52%, diene content 8.1%) for EPDM
uncross­linked rubber is reported to promote crosslinking compounds, and RSS#1 for natural rubber compounds.
in NR with sulphenamide or thiuram vulcanisation The crosslinking agents used to prepare peroxide
accelerator [2]. It is clear from this past research alone compounded EPDM, sulphur compounded EPDM, and
that water has an effect on crosslinking reactions, and sulphur compounded NR were dicumyl peroxide (DCP,
as a factor responsible for such problems as scorching Percumyl D40, NOF Corp.) or sulphur powder (200 mesh,
in the crosslinking process and substandard products, purity not less than 99.9%, Tsurumi Chemical Industry
the water content of uncrosslinked rubber therefore Co.). Nippon Kasei Chemical Co.’s triallyl isocyanurate
assumes great importance in the crosslinking process. (TAIC) was used as crosslinking promoter for peroxide
Again, unforeseen difficulties can arise from the fact that compounded EPDM, 2‑mercaptobenzothiazole (MBT,
the water content of uncrosslinked rubber fluctuates with Noceller M, Ouchi Shinko Chemical Industrial Co.)
ambient humidity, a factor heavily dependent on weather and tetramethylthiuram disulphide (TMTD, Noceller
and season. However, since the degree to which water TT, Ouchi Shinko Chemical Industrial Co.) were used
affects crosslinking varies with polymer species and as vulcanisation accelerators for sulphur compounded
rubber compounding and storage conditions, the effect EPDM; N-tert-butyl-2-benzothiazolylsulphenamide

© 2011 Smithers Rapra Technology T/9

Table 1. Compound recipe
Amount, phr
Ingredient EPDM/DCP EPDM/DCP/W EPDM/S EPDM/S/W NR/S NR/S/W
EPDM (EP33) 100 100 100 100 - -
NR (RSS#1) - - - - 100 100
ZnO 5 5 5 5 5 5
Stearic acid 1 1 1 1 2 2
Carbon black ISAF 40 40 40 40 35 35
DCP (D40)*1 6.75 6.75 - - - -
TAIC*2 3 3 - - - -
Sulfur - - 1.5 1.5 2.25 2.25
Accelerator MBT*3 - - 0.5 0.5 - -
Accelerator TMTD *4
- - 1 1 - -
Accelerator BBS*5 - - - - 0.7 0.7
Water - 5 - 5 - 5
Cure temp., °C 170 160 140
Cure time, min 21 35 13 and 18
*1
Dicumyl peroxide (purity 40%)
*2
Triallyl isocyanurate
*3
2-Mercaptobenzothiazole
*4
Tetramethylthiuramdisulfide
*5
N-tert-butyl-2-benzothiazolylsulfenamide

Table 2. Water content in uncrosslinked rubbers and mechanical properties of crosslinked rubbers. ne: Crosslink density;
TB: Tensile strength; EB: Elongation at break
Samples Water content, % Cure time, min ne,mol·m-3 Hardness TB, MPa EB, %
EPDM/DCP 0.02 21 411 67.0 16.9 170
EPDM/DCP/W 0.35 21 424 67.4 15.2 160
EPDM/S 0.08 35 427 71.4 22.8 340
EPDM/S/W 0.44 35 437 72.7 22.4 320
NR/S 0.21 13 131 52.5 27.9 600
NR/S/W 0.94 13 161 55.9 28.4 570
NR/S 0.21 18 175 56.9 29.6 550
NR/S/W 0.94 18 179 57.4 28.9 560

(BBS, Noceller NS, Ouchi Shinko Chemical Industrial NR/S/W). The water content of uncrosslinked rubber
Co.) was used as vulcanisation accelerator for sulphur was determined by heated coulometric-Karl Fischer
compounded NR. The compounds were mixed on a analyzer (CA-200, VA-124S, Mitsubishi Chemical Co.).
6 inch open roll in the sequence rubber stock, stearic Crosslinked rubber sheet of thickness 2 mm was obtained
acid, zinc oxide, DCP or sulphur powder, TAIC or by press curing under the conditions given in Table 2.
vulcanisation accelerator, and carbon black dried by
heating at 125°C; the following uncrosslinked rubbers
were produced without addition of water, viz. DCP 2.2 Methods of evaluation and analysis
compounded EPDM (here denoted EPDM/DCP), sulphur To examine the correspondence between storage
compounded EPDM (denoted EPDM/S), and sulphur conditions and the water content of the uncrosslinked
compounded NR (denoted NR/S). Compounds were rubber, samples of the uncrosslinked EPDM/DCP,
similarly mixed using carbon black mixed with water EPDM/S and NR/S were left to stand in a constant
(water purified by ultrafiltration and ion-exchange) to temperature and humidity cabinet at 23°C × 70% RH
produce the following uncrosslinked rubbers, each or 23°C × 20% RH for a predetermined time after which
containing water: DCP compounded EPDM (denoted the water contents of the respective uncrosslinked rubbers
EPDM/DCP/W), sulphur compounded EPDM (denoted were determined with a heated coulometric-Karl Fischer
EPDM/S/W), and sulphur compounded NR (denoted analyzer.

T/10 International Polymer Science and Technology, Vol. 38, No. 2, 2011
 To provide an index of crosslinking induction time, 3. Results and Discussion
the 10% crosslinking time tC(10) was found from the
curing curve obtained by measurements with a rotorless
vulcanisation tester as in JIS K 6300-2; the value of
3.1 Water content of uncrosslinked rubber
|90%ME-10%ME|/|tC(90)- tC(10)|, the slope of the
straight line joining the 10% and 90% crosslinked points, About 3% of water was added in each case. However,
was found as an index of rate of crosslinking. since some water evaporates owing to the heat generated
Hardness was measured with a Micro rubber hardness in mixing, the uncrosslinked rubber will contain less than
tester; tensile strength and elongation at break were the amount added. The water content of the uncrosslinked
tested as in JIS K 6251 using No.3 dumbbell test pieces rubber samples was therefore determined by heated
at a tensile rate of 500 mm/min‑1; and the ageing tests coulometric-Karl Fischer analyzer. Table 2 shows the
followed JIS K 6257 with treatment in a Geer oven results of water determination, which confirmed that
for 72 hours and 168 hours at 70°C, followed by uncrosslinked rubber with water added contained much
measurement of the changes in hardness, tensile strength more water than uncrosslinked rubber without water
and elongation at break. To determine crosslink density, added.
swelling tests were run at 30°C with cyclohexane as Figure 1 plots the change in water content with time
solvent for EPDM and toluene as solvent for NR, and the when uncrosslinked rubber without water added was left
apparent network chain concentration was found from to stand in a thermostatic-hygrostatic environment. The
the equilibrium percentage swelling after 72 hours using water content in EPDM/DCP, EPDM/S and NR/S alike
the corrected Flory-Rehner equation [4]. increased at 23°C/70% RH, conditions simulating a
Changes in chemical structure in the crosslinked humid environment as in the Japanese rainy season, and
rubber due to the presence of water in the uncrosslinked
rubber were evaluated by microscope Fourier transform
infrared spectrophotometry (FT-IR) and high resolution
solid-state 13C nuclear magnetic resonance (NMR)
spectroscopy. The FT-IR microscope instruments were a
Varian FT2-6000 and UMA-500, and measurements
were made by microscope ATR (with a Ge prism) at
a resolution of 8 cm-1 and 256 scans. The solid state
13
C high resolution NMR used a JEOL JNM-ECX400;
measurements were made by the heteronuclear dipolar
interaction-magic angle rotation (DD/MAS) technique at
an observation frequency of 100.53 MHz and rotation
frequency of 18 kHz.
To analyse the mechanism of action of water in
the uncrosslinked rubber on the crosslinking reactions
in EPDM/DCP, we determined the contents of the
a-methylstyrene ionic decomposition product of
DCP and the stearic acid that can promote this ionic
decomposition by solvent extraction-gas chromatography
mass spectrometry (GC-MS). The solvent used to extract
a-methylstyrene was diethyl ether; stearic acid was
extracted with acetone. To find the zinc stearate content
affecting the progress of vulcanisation reactions in
EPDM/S and NR/S, the stearic acid in the uncrosslinked
rubber was estimated by acetone extraction followed
by GC/MS. The zinc stearate content was the found by
subtracting the observed content of stearic acid from the
recipe content of stearic acid. GC/MS was carried out
with an Agilent Technologies GC/MS 6890 and 5973N
using a J&W Co. DB-5 column (length 30 m, internal
diameter 0.25 mm, film thickness 0.25 µm).

Figure 1. Change in water content in uncrosslinked rubbers

under the constant relative humidity

© 2011 Smithers Rapra Technology T/11

reached roughly the water content of the uncrosslinked gave a slightly earlier rise in torque in both the DCP
EPDM/DCP/W, EPDM/S/W or NR/S/W with water compound and sulphur compound: tC(10) was 1.13 min
added after about 10 days. On the other hand, almost no in EPDM/DCP/W, 1.22 min in EPDM/DCP, 1.45 min
change in water content was seen in any of the rubbers in EPDM/S/W and 1.60 min in EPDM/S. The curing
at 23°C/20% RH, which simulated dry conditions as curves also showed good reproducibility. Hence, while
in winter. It was hence established that uncrosslinked the presence of water in the uncured rubber in DCP and
rubber with water added had the water content the sulphur compounds of EPDM would have no significant
rubber would have after storage for around 10 days at effect on the rubber products, it slightly shortens the
high humidity as in the rainy season, while uncrosslinked crosslinking induction time. Calculation of the slope of
rubber without water added had the water content it the straight line connecting the 10% and 90% crosslinked
would have in storage at low humidity as in winter. points as an index of crosslinking reaction rate after
initiation gave a result of 0.09 N.m.min-1 for both
the NR sulphur compounds irrespective of addition of
3.2 Relation between the water content water, thus showing no significant difference due to the
of uncrosslinked rubber and crosslinking presence of water. In DCP compounded EPDM the results
characteristics were 0.29 N.m.min-1 for the uncrosslinked rubber with
water added and 0.30 N.m.min-1 for the same without
Figure 2 shows the curing curves and Table 3 lists the water added, while in the sulphur compounded EPDM
10% crosslinking time tC(10) and the rate of crosslinking the results were 0.27 N.m.min-1 for the uncrosslinked
reaction |90%ME-10%ME|/|tC(90)- tC(10)|. When rubber with water added and 0.25 N.m.min-1 for the
water is present in the uncured NR sulphur compound, same without water added; thus in neither case did water
the torque begins to rise early: tC(10) was 3.51 min have any clear-cut effect on the rate of crosslinking.
in NR/S/W and 7.11 min in NR/S, the presence of For DCP compounded EPDM, sulphur compounded
water thus greatly shortening the induction time for EPDM and sulphur compounded NR alike, therefore,
crosslinking. In the case of EPDM, the addition of water the presence of water in the uncrosslinked rubber
was shown to have no significant effect on the rate of
crosslinking after initiation of the crosslinking reaction. The
following vulcanisation mechanism generally operates
when a vulcanisation accelerator is used [5,6]. (1) The
vulcanisation accelerator is activated, and the zinc salt of
the vulcanisation accelerator forms; (2) the sulphur reacts
with the vulcanisation accelerator to form an intermediate
that releases active sulphur; (3) polysulphide pendant
linkages containing terminal accelerator residues are
introduced into the rubber molecular chains, forming
a crosslinking precursor; and (4) the polysulphide
linkages shorten as crosslinking progresses, ultimately
shortening to monosulphide linkages. The point at which
crosslinked structure is first formed in this process is
step (4), and the preceding sequence up to formation
of a crosslinking precursor that has accelerator residue
introduced into the molecular chain as polysulphide
Figure 2. Cure curves of the compounds with different water
pendant linkages would correspond to the induction time
content. The values in parentheses are water content in for crosslinking before torque rises on the curing curve.
uncrosslinked rubbers In the case of sulphur compounded EPDM, on the other

Table 3. Water content in uncrosslinked rubbers, tc (10) and crosslinking rate ({90%ME-10%ME} / {tc (90) - tc (10})
Samples Water content, % tc (10), min Crosslinking rate, N·m·min-1
EPDM/DCP 0.02 1.22 0.30
EPDM/DCP/W 0.35 1.13 0.29
EPDM/S 0.08 1.60 0.27
EPDM/S/W 0.44 1.45 0.25
NR/S 0.21 7.11 0.09
NR/S/W 0.94 3.51 0.09

T/12 International Polymer Science and Technology, Vol. 38, No. 2, 2011
hand, there is almost no induction time in crosslinking
since a thiuram vulcanisation accelerator is used, and
the process should therefore be resistant to the effect of
water in the uncrosslinked rubber. DCP compounded
EPDM would likewise be resistant to the effect of water
since the crosslinker again allows almost no induction
time in crosslinking.

3.3 Effect of water in the uncrosslinked rubber on

crosslinked steady-state properties and ageing
characteristics
Table 2 sets out the crosslink density, hardness and
tensile properties of the crosslinked rubbers in the steady
state. DCP crosslinked EPDM and sulphur crosslinked
EPDM showed almost no difference in crosslink density,
Figure 3. Change in hardness by ageing test at 70°C
hardness, tensile strength and elongation at break
regardless of addition of water to the uncrosslinked
rubber. Sulphur crosslinked NR, on the other hand, shows
a rise in crosslink density and hardness in the samples
of higher water content in the uncrosslinked rubber as
compared with their lower water content counter­parts
at cure times of 13 min (corresponding to the tC(70) of
NR/S) and 18 min (corresponding to the tC(90) of NR/S),
confirming that the water in the uncrosslinked rubber had
a considerable effect on the properties of the crosslinked
rubber. Some difference in steady state properties would
be expected since crosslinking in sulphur crosslinked NR
at the same cure time is more advanced when water
is present, as shown by the curing curves in Figure 2.
Figures 3, 4 and 5, respectively, show the effect of
ageing treatment at 70°C on hardness, tensile strength
and elongation at break. The presence of water in the
uncrosslinked rubber in DCP crosslinked EPDM and
sulphur crosslinked EPDM has no effect on the changes Figure 4. Change in tensile strength (TB) by ageing test at
in hardness, tensile strength and elongation at break 70°C
due to ageing treatment. This is consistent with the
observation in 3.2 above that water had almost no effect
on crosslinking reaction in DCP crosslinked EPDM and
sulphur crosslinked EPDM. In sulphur crosslinked NR the
change in hardness due to ageing treatment is smaller
and the decrease in tensile strength and elongation at
rupture occurs earlier when more water is present in the
uncrosslinked rubber. This may be attributed to the fact
that water in the uncrosslinked rubber greatly promotes
crosslinking, driving the reaction further than when no
water is added, as noted in 3.2.

3.4 Effect of water in the uncrosslinked rubber on

the chemical structure of the crosslinked rubber
The water in the uncrosslinked rubber has only a slight
effect on the crosslinking reaction in DCP crosslinked
EPDM and sulphur cured EPDM but has a marked Figure 5. Change in elongation at break (EB) by ageing test
effect in sulphur cured NR. FT-IR and solid state 13C at 70°C

© 2011 Smithers Rapra Technology T/13

high resolution NMR measurements were made to
ascertain whether the water reacts directly with rubber
molecules. The FT-IR spectra showed no difference due
to the presence of water in the uncrosslinked rubbers in
DCP crosslinked EPDM, sulphur crosslinked EPDM and
sulphur crosslinked NR; similarly, no specific signals
generated by the presence of water in the uncrosslinked
rubber were observed in the 13C high resolution NMR
spectra. In the case of sulphur crosslinked NR, the very
small signals observed at 35-60 ppm were somewhat
stronger with water added than without water added to
the uncrosslinked rubber, but these originate from carbon
next to a sulphur bridge [7], indicating that crosslinking
was more advanced. It was hence shown that water in
the uncrosslinked rubber does not react directly with
rubber molecules in DCP crosslinked EPDM, sulphur
crosslinked EPDM or sulphur crosslinked NR.

3.5 Investigation of the mechanism whereby water

promotes crosslinking reaction

3.5.1 Promotion of crosslinking reaction in DCP

compounded EPDM
The crosslinking reaction in EPDM crosslinked with
DCP proceeds as the radicals formed by free radical
decomposition generate polymer radicals, but any acidic
substance present will promote ionic decomposition, in
which none of the radicals responsible for crosslinking are
formed [8]. Among the products of ionic decomposition
are a-methylstyrene, phenol and acetone [8]. Taking
the amount of a-methylstyrene evolved as a measure Figure 6. 13C-NMR spectra of the crosslinked rubbers made
of the progress of ionic decomposition, therefore, we from the compounds with different water content
estimated the yield of a-methylstyrene as a function
of time when DCP compounded EPDM was heated at
170°C. The results are presented in Figure 7. The yield
of a-methylstyrene is lower at the higher water content,
ionic decomposition thus being inhibited. Of the various
compounding agents, stearic acid may be singled out
as an acidic substance; estimation of the stearic acid
content in uncrosslinked DCP-compounded EPDM gave
results of 0.43% in EPDM/DCP and 0.38% in EPDM/
DCP/W as against 0.64% in the recipe, more stearic
acid thus being converted to zinc stearate the greater
the water content. This implies that crosslinking in DCP
crosslinked EPDM is somewhat accelerated by water
in the uncrosslinked rubber as a result of inhibition of
ionic decomposition of DCP because the amount of
stearic acid present has been reduced by water. As to
the mechanism whereby water promotes consumption
of stearic acid to form zinc stearate, we may suppose
that water creates basic zinc hydroxide from the zinc
oxide component and the zinc hydroxide then reacts Figure 7. Change of a-methylstyrene content in EPDMs at
with stearic acid, promoting formation of zinc stearate: 170°C

T/14 International Polymer Science and Technology, Vol. 38, No. 2, 2011
ZnO + H2O → Zn(OH)2 (1) the formation of crosslinked structure, i.e. shortens the
crosslinking induction time.
Zn(OH)2 + 2C17H35COOH →
(C17H35COO)2Zn + 2H2O (2)
4. Conclusions
The effect that the water present during the crosslinking of
EPDM has on crosslinked rubber properties and ageing
3.5.2 Mechanism whereby water promotes
characteristics is minimal for both DCP crosslinking and
crosslinking reaction in sulphur compounded
sulphur crosslinking. In sulphur crosslinked NR, on the
EPDM and sulphur compounded NR
other hand, the presence of water during crosslinking
Vulcanisation with a vulcanisation accelerator proceeds drives the process too far and therefore alters the properties
via the steps (1) to (5) outlined in 3.2. It is likely, of the crosslinked rubber while also shortening service
however, that a vulcanisation accelerator comprising life. Again, although the water present at crosslinking
a zinc complex formed from zinc stearate and the does not react with rubber molecules, it encourages
nominal acceler­ator functions as the actual accelerator, transformation of the stearic acid compounded in the
the zinc stearate thereby promoting the formation rubber to zinc stearate. It may be inferred that this serves
of crosslinking reaction intermediates in the various to promote the crosslinking reactions.
processes represented here as (1), (4) and (5) [5,6]. The
zinc stearate content in the uncrosslinked rubber was
therefore determined. The results of 0.28% for EPDM/S, References
0.51% for EPDM/S/W, 0.06% for NR/S, and 0.66%
1. Kodama S., Nippon Gomu Kyokaishi, 78
for NR/S/W indicated that transformation of stearic
(2005) 187.
acid to zinc stearate occurs more easily at the higher
water content in both sulphur compounded EPDM and 2. Butler J., Freakley P.K., Rubber Chem. Technol.,
sulphur compounded NR. It would also appear that 65 (1992) 374.
water promotes the formation of zinc stearate by the 3. Kim K-J., Vanderkooi J., Rubber Chem. Technol.,
mechanism shown above in Equations (1) and (2). It 78 (2005) 84.
may hence be inferred that, since the amount of zinc 4. Ono K., ‘Gomu Kogyo Binran (4th Ed.)’,
stearate encouraging formation of crosslinking reaction Nippon Gomu Kyokai Ed. Nippon Gomu
intermediates in sulphur compounded EPDM is increased Kyokai, Tokyo, p.1231 (1994).
by water, the water in the uncrosslinked rubber will
5. Morrison N.J., Porter M., Plast. Rubber Proc.
accelerate crosslinking somewhat while also producing
Appl., 3 (1983) 295.
some increase in crosslink density. Likewise for sulphur
compounded NR, it may be inferred in particular that, 6. Morrison N.J., Porter M., Rubber Chem
since water in the uncrosslinked rubber promotes the Technol., 57 (1984) 63.
formation of zinc stearate, encouraging formation of 7. Mori M., Koenig J.L., Nippon Gomu Kyokaishi,
crosslinking reaction intermediates, the presence of water 71 (1998) 68.
greatly shortens the process as far as the formation of 8. Dluzneski P.R., Rubber Chem. Technol., 74
crosslink precursor (when polysulphide pendant linkages (2001) 451.
have been introduced into the polymer) directly preceding

© 2011 Smithers Rapra Technology T/15