Performance of Low Temperature Curing EPDM

Extruded Profiles for Weatherstrip Applications

*
J A Harris , M V Lewan and R P Campion, MERL Ltd, Hertford, UK

*Presenting author
Dr John Harris
Technical Director
Materials Engineering Research Laboratory (MERL) Ltd,
89-91 Tamworth Rd,
Hertford
SG13 7DG,
UK
phone +44 (0) 1992 500120
fax +44 (0) 1992 586439
email [email protected]
web www.merl-ltd.co.uk

Biography
Dr John Harris is Technical Director of MERL with over 25 years experience as a specialist in R & D with non-
metallic materials including elastomers, adhesives, thermoplastics and composites. His degree is in
mechanical engineering and PhD in structural adhesives. Main interests with elastomers are: dynamic
properties for vibration isolation, design analysis including FEA, creep, fatigue, fracture and performance
testing for life assessment.

1
SUMMARY

This paper aims to demonstrate the feasibility and benefits of using rubber compounds that can be cured at
lower temperatures than are currently used for EPDM extruded profiles for automotive weatherstrip.

The basic cure properties of an LTC compound are compared with a conventionally cured control weatherstrip
EPDM compound. The relative cure rate of the LTC compound is about 2 times faster than the control at
190ºC and this difference increases for lower cure temperatures to more than 300 times faster at 100ºC.

The extrusion performance of an LTC compound was confirmed on an NFM Iddon High Intensity Mixing scroll
extruder with a Garvey die where perfect profile shapes were obtained. Energy consumption comparisons
were made with a conventional compound extruded at normal temperatures at different scroll speeds. When
processed at low temperature the LTC compound gave similar energy consumption to the control with a 15%
lower production rate. When processed at normal temperature the LTC compound required 20 – 26% less
energy than the control with a 10% lower production rate.

During the cure stage, because of the greater reactivity of the LTC compound, an energy saving on heating of
30% or more is shown to be feasible.

Controlled cure of test sheets was achieved using the cure-control package CURIGHT, which uses a cure
units approach to account for variations in temperature profile.

Similar physical properties (modulus, tensile strength, elongation at break) were obtained for the LTC
compound which matched the hardness of a commercial control. In addition the LTC compound gave more
than a factor of two greater tear strength. Accelerated ageing resulted in somewhat greater changes in
physical properties and stress relaxation for an LTC compound compared to a control due to additional cure.
However in the stress relaxation tests there was little further force reduction after about 1d at elevated
temperature.

It is concluded that by employing the LTC approach, energy (and hence cost and environmental) savings are
possible, without substantial loss of long-term functional performance.

INTRODUCTION

The LTC Concept

The concept of this work is based on the proven success of a particular set of pliant low temperature curing
elastomers developed for the repair of elastomeric components. These materials were originally used for on-
site repair of large elastomeric components used in offshore engineering and storage tank applications.
MERL developed low viscosity materials, which, could be moulded into intricate shapes at 70°C, be cured in a
short time, be bonded to existing elastomers and metals, and possess acceptable mechanical properties. This
led to the more general concept of curing elastomeric products at low temperatures i.e. in the range 70ºC to
120ºC compared with ‘normal’ elastomers cured at temperatures of 130°C to 180°C. As illustrated in Figure 1,
the use of a lower cure temperature means that less heat input is required to achieve cure so that significant
cost savings can be achieved. Benefits may also accrue from higher rates of production, the ability to
incorporate lower melting temperature reinforcement materials and the reduction in cure-related and thermal
shrinkage effects.

The two key features of LTC compound recipes are:

(i) the use of ultra-active accelerators to achieve a sufficient rate of cure at the reduced cure temperature
and
(ii) the incorporation of a compatible liquid polymer (clp) to reduce viscosity and enable processing at
reduced temperatures.

The concept is covered by US provisional patent application no. 60/365,466 and European patent application
no. 03252588.3.

2
Work Programme
To demonstrate the feasibility of the LTC concept for automotive weatherstrip, three aspects of compound
performance have been considered:

(i) processing properties for mixing and extrusion,

(ii) cure properties,
(iii) physical properties after cure.

All compounds were based on EPDM, being the preferred elastomer for weatherstrip applications. Control
compounds, representative of current practice were used for comparison with LTC compounds. The control
compounds were formulated based on published information or were current commercial materials.

The work was carried out in five parts.

• Formulation of LTC compounds and a Control compound.

• Evaluation of cure behaviour to compare LTC with a Control compound.
• Factory-scale mixing of an LTC compound on a K2A Intermixer to demonstrate the processability
• Extrusion trials on an NFM Iddon High Intensity Mixing (HIM) scroll extruder to compare processing
energy consumption and heating between an LTC compound and a high grade automotive profile
commercial control compound.
• Mechanical property tests on an LTC compound and a control including stress relaxation and accelerated
ageing.

COMPOUND FORMULATIONS

The target for the first stage of the work was the development of an LTC recipe with adequate flow properties
for extrusion and the ability to be cured at 90°C - 100°C. This involved the formulation of some LTC non-
curable masterbatches that include the low viscosity compatible liquid polymer (clp) miscible with the EPDM.
Full LTC formulations were obtained by addition of the ultra-active accelerators to the masterbatches. These
accelerators increase vulcanisation rate significantly at normal cure temperature by an amount which depends
on the types used and their quantity. Consequently, they can also bring about cure in reasonable times at
much lower temperatures than normal.

LTC Masterbatch Formulations

Three EPDM-based masterbatches (M/B) were defined, as shown in Table 1. M/B #B was a factory-scale
K2A Intermixer mix used as basis for the extrusion studies. M/B #C was a laboratory mix masterbatch variant
used for producing a compound with increased hardness for mechanical property assessment. M/B #D was
similar to #C but without clp and was used for a control formulation (MERL control).

Table 1. EPDM / clp* masterbatch formulations

Masterbatch #B #C #D
® 75 75 100
Vistalon 7500
clp* 25 25 -
® 45 35 35
Indopol H7
FEF N550 80 100 100
Zinc oxide 5 5 5
Stearic acid 0.5 0.5 0.5
Clay 20 - -
Magnesium oxide 2 2 2
TOTAL 252.5 242.5 242.5
*clp = compatible liquid polymer

3
The clp is employed for its ability to make the EPDM compounds more pliable and therefore easier to process
at lower temperatures. Indopol H7, a very low molecular weight polyisobutene polymer, was included to aid
extrudability.

LTC Full Recipes

After an initial investigation of six LTC compounds, two LTC compounds and one control were formulated as
shown in Table 2. For LTC #7, M/B #B was used with the optimum combination of ultra-active accelerators
and sulphur for extrusion testing. Details of the factory scale mixing of this compound are given in the next
section. LTC #7 possessed a lower hardness than the control compound so LTC #8 with a higher hardness
was formulated. Control #12 has a “normal” cure system so that comparisons can be made with the LTC
compounds.

Table 2. Compound formulations LTC #7, #8 & Control #12

LTC #7 LTC #8 Control #12
M/B #B 252.5 - -
M/B #C - 242.5 -
M/B #D - - 242.5
Sulphur 1.0 1.0 1.5
MBTS 1.0 1.0 1.5
Ultra-active accelerator I 3.0 3.0 3.0
Ultra-active accelerator II 3.0 3.0 3.0
Ultra-active accelerator III 0.52 0.52 0.52
Ultra-active accelerator IV 1.0 1.0 1.0
ZDBC - - 0.8
ZDEC - - 0.8
TMTD - - 0.8

CURE CHARACTERISITICS

To measure cure characteristics of compounds, curometry was performed at a range of different temperatures
between 100°C and 190°C using an Alpha Technologies MDR 2000. One problem encountered with some
synthetic rubbers such as EPDM is the “marching modulus” of the cure profile i.e. a horizontal equilibrium or
maximum torque value is not achieved; thus the conventional definitions of cure time, which are based upon
the maximum torque value, cannot be applied. Investigation of the cure profiles of the compounds has shown
that the rate of change in the torque in the cure profile can be used as a convenient basis to define the cure
time. For these compounds cure time has been defined as:

Cure time = 6 x (time to maximum torque rate)

As examples, Figure 2 shows the cure profiles for Control #12 compound at 150°C and LTC #8 at 120°C, with
the points of maximum torque rate and their subsequent cure times shown.

From the cure time measurements, an Arrhenius-type plot of the natural log of reciprocal cure time versus
reciprocal of absolute temperature was obtained for both compounds over the temperature range 100°C to
190°C as shown in Figure 3. This comparison indicates the faster rate of cure and hence shorter cure times
of the LTC #8 compared to Control #12 at each specified temperature. The difference is greater for lower
temperatures i.e. LTC #8 is more than 300 times faster curing at 100ºC and 2 times faster at 190ºC. Thus with
LTC #8 either a lower cure temperature or faster throughput at the “normal” curing temperature (scorch safety
has been shown not to be an issue) than with control #12.

4
PROCESSING AND ENERGY SAVINGS

Factory scale mix

A 30 litre K2A Intermixer was employed for the factory scale mixing. A total of 80kg of compound was
required for set-up trials and actual extrusion tests (40kg masterbatch and 40kg finalised compound
respectively), requiring three mixes of the masterbatch to be carried out in the K2A. Its standard rotor revs of
50 rpm were reduced to 25 rpm to accommodate the relatively low-viscosity compound, and a “normal” mixing
cycle used (see Appendix 1). After the mixing was finished, the rubber was dumped (temperature of 100°C),
and then consolidated on a two-roll mill.

For the 40kg of compound being finalised, the next stage was the addition of the curatives. Two equal
batches were mixed, each consolidated on a two-roll mill and then combined to ensure a uniform rubber mix.

Extrusion trials
Laboratory trials were conducted at NFM Iddon on their 50mm High Intensity Mixing (HIM) scroll extruder.
This contains a patented screw design that, besides the regular feeding and metering zone, contains a middle
section whereby the flights are interrupted so that a material countercurrent is produced which leads to a very
effective mixing process. This also ensures that an even temperature is maintained throughout the rubber, as
well as a lower heat input being required because the mechanical action will generate enough heat. The die
used for this work was a Garvey die, a standard die used in the industry for rating the extrusion characteristics
of a rubber compound due to its prescribed contour and dimensions. If the extrudate emerges with no loss in
shape then it indicates that there are no fluctuations in the extruding process.

A series of tests were performed to the conditions shown in Table 4 on the LTC #7 and on the profile control
compound for comparison.

Table 4. Test temperatures (°C)

Test EPDM Screw Feed Barrel Barrel Die head
Material cylinder zone 1 zone 2
1 LTC #7 40 50 40 40 70
2 Profile control 40 50 40 40 70
3 Profile control 80 65 65 55 85
4 LTC #7 60 50 40 40 60
5 LTC #7 80 65 65 55 85

Test details were:

• Test 1 used low process temperature settings with LTC #7 as a start basis.

• Test 2 was a repeat of the conditions used in Test 1 but this time the EPDM profile compound was used.

• Test 3 used higher temperature conditions that are generally employed for a “normal” rubber compound,
again with the EPDM profile compound.

• Test 4 was similar to Test 1 except that the screw temperature was increased by 20°C and the die head
temperature reduced by 10°C.

• Test 5 was a repeat of the higher temperature conditions used in Test 3 but this time using the LTC #7
compound.

The resulting extrudates all exhibited a perfect profile to the shape of the Garvey die, thus the ability to extrude
LTC type compounds acceptably was demonstrated.

The variation in process power and extrudate output with scroll speed is shown in Figures 4 and 5
respectively. From the results of these trials several conclusions can be drawn.

5
Power consumption (Figure 4)
The process power is given in terms of the measured current in Figure 4 since the voltage is constant. The
product “voltage times process amps” reflects the actual power to process the polymer and not the total power
to drive the extruder and process the polymer. All plots against scroll speed show an increase until a plateau
is effectively reached. Comparison at the lower temperatures of process amps between Test 1 (LTC #7) and
Test 2 (profile control) shows a significant reduction in the power consumed processing the LTC material,
typically around 30%. Test 1 also displayed 15% reduction compared with Test 3 – the control at standard
temperatures. The higher screw temperatures in LTC compound Tests 4 and 5 gave even lower process
power consumption.

Extrudate output (Figure 5)

The actual output weight for LTC #7 was up to 16% below that of the control profile EPDM. However, higher
screw temperatures for the LTC compound in Test 4 and Test 5 gave marginally increased outputs, as would
be expected.

Energy savings analysis

A comparison between LTC #7 and the profile control compound can be made for a typical production
situation where, for example, a 25kg batch of each rubber compound is required to be extruded. Two scroll
speeds at the extremes of the power plateaux (see Figure 4) are chosen for making these comparisons.
Table 5 shows the steps involved and the resulting estimates.

Table 5. Estimations of energy required to produce 25kg of extrudate

* 1 3 5
Test no.

Estimation stages Factor LTC at low Profile Control LTC at

temperature at normal normal
temperature temperature
Step Comment Scroll speed (rpm) 25 65 25 65 25 65
1 From Figure 4 Current (amps) 16.5 15.5 18.1 18 11.8 13.1
2 From Figure 5 Output (kg/hour) 17.1 42.9 20.4 49.2 18 45
3 25kg ÷”Step 2 value” Time (hrs) to extrude 1.46 0.58 1.23 0.51 1.39 0.56
a 25kg batch
4 “Step 1 value” x voltage x Energy used+ (kWh) 10.0 3.75 9.21 3.80 6.80 3.02
“Step 3 value”
*as described in Table 4
+
assuming 415 volt supply

From this illustration the following is concluded.

• Extrusion time, arising inversely from scroll speed, has a major effect on energy used
• LTC #7 extruded at normal higher temperatures (Test 5) needs the least energy of the systems analysed,
at 20 – 26% lower levels than the corresponding control ones from Test 3. These temperatures should not
cause scorch problems.
• LTC #7 at low extrusion temperatures employs a similar amount of energy to the EPDM profile control
compound extruded at normal higher temperatures.
• In practice, these energy (and hence cost) savings per batch associated with LTC EPDM at normal
extrusion temperatures might have to be weighed against daily output issues.

6
Cure Energy
The energy required to cure a rubber compound has two stages; the initial heating and then maintaining the
required temperature.

1. Initial heating
At 100°C, the cure time for LTC #8 is 9.1 minutes. Control #12 has this cure time when its cure temperature is
132°C (see Figure 3). Thus, the initial heating is proportional to thermal capacity times temperature rise.
Hence, the energy used will proportionate to the temperature difference, i.e.

100°C − 23°C
= 0.70
132°C − 23°C

Thus, for this step there will be a 30% energy saving.

2. Maintaining temperature
Heat losses from the press and mould will depend upon Newton’s law of cooling where “the rate of cooling is
proportional to the temperature difference above ambient”. Hence more energy is required to keep the press
at higher cure temperature – the feedback system controlling the press or equivalent will switch the power on
for much of the cure time. This could well offset the effect of the shorter cure times involved; this point cannot
be quantified precisely in general terms as the rate of heat loss will depend on how well a specific press and
mould are insulated, and on how short the LTC elastomer cure times can become by further compound
development. However, the potential benefits of a lower temperature cure are clear.

Material Costs
Appendix 2 illustrates the approximate prices appropriate to modest usage of the clp and the ultra-active
accelerators. This only gives a rough guide since costs depend on volumes of material. Based upon these
costs and previous experience, a worst case comparison could give a 20% cost increase in ingredients.
However, this would be easily offset by the savings that occur in both the processing and curing of the LTC
rubber.

PHYSICAL PROPERTIES

Samples for physical property tests were cured under controlled conditions using the CURIGHT program.
This program uses the principle of cure unit analysis to enable precise state of cure conditions to be obtained
for a particular temperature profile. Details are given in Appendix 3. Using this procedure a number of
compounds were cured and a set of mechanical and stress relaxation tests carried out that included
consideration of the effects of ageing.

Tensile, Tear, Hardness and Ageing

As an indication of the viability of these materials, mechanical property values were obtained for LTC #7 and
LTC #8 cured at 90°C. For comparison, a proprietary automotive profile compound (different from that used in
the extrusion trials) was cured to its optimum state of cure at 170°C as the control. The control and LTC
compounds were cured into nominally 2mm sheets in a compression mould; testpieces were cut out, basic
mechanical property testing conducted and their mechanical properties obtained. Table 6 shows the values
measured.

Table 6. Mechanical properties of LTC #7 & #8, and Control

Control* LTC #7 LTC #8
M100, MPa 2.6 1.3 2.2
Tensile strength, MPa 8.8 5.9 6.9
Elongation at break, % 407 526 412
Trouser tear, N/mm 6.2 11.6 15.6
Hardness, IRHD 60 45 58
*Data obtained at MERL from a production profile compound (Cure: 6 mins. at 170°C)

7
Extrudable LTC #7 is a relatively soft elastomer, i.e. relatively low hardness and modulus. However, the other
mechanical property levels such as trouser tear and elongation at break are superior to those of the control.

LTC #8 displayed hardness and modulus values similar to those of the control, and trouser tear strength which
was even greater. Elongation at break was the same as the control, while tensile strength was slightly lower.
Clearly, a range of acceptable property magnitudes are obtainable for LTC elastomers as required by suitable
compounding.

Generally, the LTC elastomers exhibit trouser tear values consistently much higher than that of the control. It
is likely that the presence of the clp improves the tear strength which would otherwise be attained with these
ultra-active curing systems.

In a limited exercise, the effect of aerobic ageing on both LTC #8 and the control was investigated.
Vulcanised 2mm sheets cured as before were aged in an oven for 7 days at 100°C. The resultant mechanical
properties and their respective magnitude changes from the unaged values are shown in Table 7.

Table 7. Effect of aerobic ageing on the mechanical properties of LTC #8 and Control
Control LTC #8
M100, MPa 4.0 (+54%) 11.6 (+427%)
Tensile strength, MPa 10.0 (+13.6%) 14.8 (+114%)
Elongation at break, % 298 (-26.8%)* 155 (-62.4%)*
Trouser tear, N/mm 2.3 (-62.9%) 3.0 (-80.8%)
Hardness, IRHD 69 (+9) 83 (+25)
* proportional percentage

Both the control and the LTC compound have large changes in their mechanical property values following
ageing. They both are stiffer materials when aged, as seen by the higher hardness and modulus values, and
also stronger materials, as seen by the higher tensile strength values. However, both have reduced
elongation at break and trouser tear values.

Stress relaxation and ageing

Stress relaxation is a key performance property for weatherstrip since a sealing force must be maintained for
the service life. Stress relaxation tests were performed on strip specimens of the compounds cured to
optimum and twice optimum at 100°C. A permanent strain of 20% was then applied and the force on the
rubber measured. Short-term continuous measurements were carried out on a Zwick Z250 test machine at
ambient and 100ºC. Longer-term tests with intermittent force measurement were carried out on sets of
samples, which were placed in ovens at 70°C and 100°C for a period of one to two weeks. During this time
they were periodically removed and the force measured at ambient temperature.

The results from the short-term continuous tests are summarised in Figure 6 which shows the variation in the
magnitude of the force reduction as a percentage of the initial force after 0.1 hr as a function of log time. The
linear variation on this basis indicates that the changes are purely due to physical effects. Similar changes
were obtained for LTC #7 at ambient and 100ºC and the control #12 at ambient. For LTC #7 the initial force at
100ºC was about half of that obtained at ambient.

For the longer-term tests, the results for optimally cured compounds aged at 70ºC are given in Figure 7 and for
compounds cured to twice optimum aged at 100ºC are given in Figure 8. In both tests the initial force
recorded at 20% strain is highest for the control compound, indicating that it is a stiffer material than the LTC
compound. Samples cured to twice the optimum value have higher initial force values than their counterparts
cured to optimum; indicating that stiffer materials are produced with longer cure times.

The majority of the relaxation occurs within the first day of ageing the materials with very little further relaxation
occurring after that period. In both cases, the relaxation of LTC #8 is somewhat higher than the control
compound, however both appear to be thermally stable i.e. there is no evidence of chemical ageing over the
period of the measurements.

8
CONCLUSIONS

From this study, the following conclusions are drawn concerning the benefits of using LTC type compounds for
automotive weatherstrip.

• At 100ºC, LTC cures 300 times faster than a convention cure EPDM and at a similar rate to the
conventional cure at 132ºC. Thus significant energy savings during cure are possible (circa 30% or more).

• Acceptable extrusion profiles can be obtained using LTC and if extruded at normal extrusion temperature
less energy is required (circa 20% or more).

• Physical properties can be obtained with LTC that are comparable with conventional cure. In addition LTC
gives significantly greater tear strength (factor of 2 or more).

• Accelerated ageing gives somewhat greater changes in physical properties and stress relaxation with LTC
than conventional cure. However stress relaxation in LTC is stable after a short time period, so that
acceptable functional performance can still be obtained.

• Further optimisation is possible to improve cure rates, scorch safety, shelf life and functional property
levels.

9
APPENDIX 1. “NORMAL” MIXING CYCLE EMPLOYED IN THE K2A MIXER

Time (mins) Procedure

0 Add rubber
1.0 Add powders (zinc oxide, stearic acid, magnesium oxide)
2.0 Add ½ black filler + ½ clay + ½ oils (clp, Indopol H7)
4.0 Add rest of black filler + clay + oils
6.0 Sweep chute
6.5 Dump rubber compound

10
APPENDIX 2. MATERIAL COSTS

Cost estimates
The cost for polymers and their ingredients will generally depend on volume. However, for comparison
purposes MERL have been provided with some approximate prices appropriate to modest usage.

Polymer/liquid polymer costs

Vistalon 7500 £2.0 – 2.5 per kg
clp £14.87 per kg (< 250kg) £14.07 per kg (> 250kg)

Because costs may be calculated on a volume basis, values for polymer/liquid polymer densities are:
Vistalon 7500 0.86 g/cm3
clp 0.84 g/cm3

Although a small advantage on a volume basis arises for the clp from its lower density, clearly material cost for
the LTC recipe will be increased compared with the full-EPDM based formulation.

Ultra-fast accelerator costs

As quoted by the supplier (£/kg) for prices regarding different quantity levels:

Quantities bought (tons) 0.2 – 0.5 1.0 and above

Ultra-fast accelerator 1 9.76 9.26
Ultra-fast accelerator 2 7.16 6.66
Ultra-fast accelerator 3 7.02 6.52
Ultra-fast accelerator 4 7.45 6.95

(On the same basis CBS would be ca £2.20/kg, and TMTD ca £4.12/kg).

11
APPENDIX 3. CURIGHT

A technique is required whereby each compound can be vulcanised to the same point of cure each time. This
produces duplicates that are virtually identical and should therefore eliminate any variation in test results.

As rheometer curves show, when heat is applied to uncured rubber compound during moulding, it initially
softens for physical reasons (viscosity decreases with increasing temperature) after which the formation of
vulcanisation crosslinks stiffens the rubber until it reaches its maximum modulus (torque): whether reversion
(further softening) subsequently occurs is dependent on compound recipe details and curing temperature.

The network theory of crosslinked rubber, developed many years ago, shows that to a first approximation,
applied stress (and hence related parameters modulus and stiffness) is proportional to the crosslink density,
i.e. the concentration of crosslinks present. This means that to follow the course of vulcanisation we can use
chemical kinetics, which consider changes in concentration as chemistry proceeds, by measuring changes in
rubber modulus (i.e. rheometer torque). Hence we can obtain appropriate information from curves from a
rheometer possessing good heat transfer properties.

The Arrhenius equation compares the rate of chemical reaction with temperature. From a rheometer curve,
the reciprocal time to reach a particular feature (normally t95)1 is taken as the rate. Hence

1/t95 = A exp (-Ea/RT) ---------------------------------------------------------------------------------------1

where T is absolute temperature, R is the Gas Constant (1.9872 cal/deg/mole [8.31441 JK-1mol-1]), A is
another constant and Ea the activation energy2 . For two temperatures, by expressing logarithmically and then
subtracting we get:

(t )
95 2 − Ea  1 1
ln =  −  -------------------------------------------------------------------2
(t )
95 1
R  T1 T2 

If we select 150oC as being our reference temperature, making this T2, we can define one ‘cure unit’3 as being
a cure of one minute at 150oC. Then, if n is the number of cure units per minute at another temperature T1, n
= 1/(t95)1, so that

− Ea  1 
ln(n) =  − 0.00236317  -------------------------------------------------------------3
R  T1 

(Clearly, this equation satisfies the original definition for one cure unit.)

T1 and R are, of course, known. To obtain Ea, a logarithm plot of 1/t95 values obtained by rheometer testing at
a series of different temperatures versus 1/T gives (from equation 2) a line of gradient -Ea/R. Hence n can
now be obtained from equation 3 for any selected value of T1.

1
t95 is chosen for convenience. The subscript refers to the percentage of maximum torque attained at that time.
However this approach can be applied to times for reaching other features, e.g. t90, t100 (more difficult to identify), or even
scorch t2 (the subscript here referring to 2 minutes of cure).
2
Equation 1 has a statistical derivation, so that the right hand term represents the probability of two reactant molecules
which have come together receiving enough kinetic energy - Ea, from the motions of, and collisions with, surrounding
molecules - to react.
3
Although developed by Dunlop Tyre Division in the 1960’s, the ‘cure unit’ approach is based on standard physical
chemistry and so is open to all. In any case, this company (now SP Tyres) has kindly given RPC permission to use the
term freely.
12
Of course, in reality, cure is not performed at a single temperature - the article heats up and then cools down
during curing. However, as shown below, the consequence of the cure unit approach is that the total amount
of cure received by a small rubber component can be measured, whatever the temperature/time profiles
applying. (Ideally, this total amount should be the same for a series of mouldings of replicate articles - see last
sentence in this section). With large components, because of the significant heating and cooling times
applying across them, this approach can be applied for local regions, and then a balance struck across the
whole component (or a non-reverting recipe employed so that overcure at the outside is less important).

The approach is:

a) view the complete temperature/time curves ‘digitally’, i.e. as a series of many incremental steps
(conveniently of constant time).
b) calculate the average temperature for each step.
c) determine n (cure units per minute) for that temperature as outlined above.
d) multiply n by the time increment (conveniently 0.05 minutes for the present work) to give the number of
cure units for that step.
e) summate for the total number of steps to give the overall cure received, in cure units.

MERL has developed a PC program to make all of these calculations, providing t95 input values for 3 or 4
temperatures are known (so that it can calculate Ea), and the temperature/time profiles are available
(preferably in digital form). Once t95 values are known, the MERL program can be used concurrently with
curing, causing it to be stopped at a pre-set number of cure units, providing the program is being fed
temperature/time information as it is being measured for that cure.

13
 Figure 1. Energy saving curve

12

10 Control #12 - 150°C

8
Torque (dNm)

LTC #8 - 120°C

Dotted lines represent the time to maximum

2 torque rate and 6 x that value i.e. cure time

0
0 5 10 15 20 25 30 35 40 45 50 55 60
Time (min)

Figure 2. Cure profiles of Control #12 and LTC #8 showing the points of maximum torque rate and
the subsequent cure times

14
 1/T
0.0021 0.0022 0.0023 0.0024 0.0025 0.0026 0.0027
0

-1

-2
LTC #8

-3
In Rate

-4 Control #12

-5

-6

-7 190° 150° 120°C 100°C

Figure 3. Arrhenius-type plot of In rate versus 1/absolute temperature

25

Test 2
20

Test 3
Process power (Amps)

Test 1
15 Test 4

Test 5

10

0
0 5 10 15 20 25 30 35 40 45 50 55 60 65
Scroll speed (rpm)
Figure 4. Process current (proportional to power) versus scroll speed

15
 50 Test 3
Test 2
45 Test 5
Test 4
Test 1
40

35

30
Output (kg/hr)

25

20

15

10

0
0 5 10 15 20 25 30 35 40 45 50 55 60 65
Scroll speed (rpm)

Figure 5 - Output versus scroll speed

25

LTC #8 100C Control #12 RT

20
(F0-F)/F0 (%)

15

LTC #8 RT
10

0
-1 0 1
Log Time (hr)

Figure 6. Short-term continuous stress relaxation for LTC #8 at room temperature (RT) and 100ºC
and control #12 at room temperature.

16
 60

LTC #8

50

40
(Fo - F)/Fo (%)

30
Control #12

20

10

0
0.0 0.5 1.0 1.5 2.0 2.5
Log Time (hours)

Figure 7. Relaxation over 7 days at 70°C for samples cured to optimum

60

50

LTC #8
40
(Fo - F)/Fo (%)

30

Control #12
20

10

0
0.0 0.5 1.0 1.5 2.0 2.5
Log Time (hours)

Figure 8. Relaxation over 14 days at 100°C for samples cured to twice optimum

17