Академический Документы
Профессиональный Документы
Культура Документы
Hydrogenated Nitrile
Butadiene Rubber
Technology
Robert Keller
A catalogue record for this book is available from the British Library.
Every effort has been made to contact copyright holders of any material reproduced
within the text and the authors and publishers apologise if
any have been overlooked.
iii
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
iv
Contents
v
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
Abbreviations..................................................................................... 123
vi
P
reface
The goal of this book has been to give the reader a short but comprehensive view of
the technology of hydrogenated nitrile rubber (HNBR). It is not intended to provide
an exhaustive literature search, but more to act as a guide for those new to HNBR
and seeking more information. My purpose has been to give an overview of:
Much of the information will be found applicable not only to HNBR but also to
other elastomers.
vii
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
viii
1
Overview of the Chemistry and
Manufacture of Hydrogenated Nitrile
Butadiene Rubber Polymers
1.1 Introduction
As the name suggests, hydrogenated nitrile polymers are manufactured by the catalytic
hydrogenation of conventional nitrile butadiene rubber (NBR) polymers. Firstly, the
manufacture of the basic NBR polymers will be discussed. Secondly, the process flow
and the chemistry of the catalytic hydrogenation process will be outlined. Finally,
the challenges presented by the manufacture of the HNBR polymers with lower
acrylonitrile (ACN) content will be briefly discussed. These HNBR polymers are
important for achieving low temperature flexibility in the resulting moulded goods.
The initiator, in this case a free-radical species, is generally quite soluble in the
continuous phase. The majority of NBR polymers are described as cold polymers,
since a free radical is formed at sub-ambient temperature (typically 5 C) in
an oxidationreduction reaction. Other NBR polymers are formed at elevated
1
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
temperatures (typically around 95 C) and are known as hot polymers. Since the
free-radical initiator is soluble in the continuous phase but only partially soluble
in the emulsified phase, free radicals diffuse slowly into the emulsified phase. In
contrast to bulk or solution free-radical polymerisation, in which a relatively
high level of free radicals occur with the chain-growth polymer and cause early
termination of polymerisation, emulsion polymerisation is diffusion-controlled
and chain growth gives very high molecular weights. Chain transfer agents such as
organo-mercaptans may be added to the emulsified phase to control the molecular
weight of the polymer to give the desired bulk viscosity.
An emulsifier, which may be simply a sodium stearate soap, assures fine dispersion
and emulsification of the monomer/polymer-rich phase.
The continuous phase acts as an excellent heat transfer medium in the exothermic
polymerisation reaction. In some cases bulk free-radical polymerisation generates
so much heat that the process becomes impractical. Due to heat transfer in the
continuous phase in emulsion polymerisation very tiny droplets of bulk polymer
are generated, eliminating the need for expensive solvents and solvent recovery.
The viscosity does not increase during the conversion, as is the case in bulk or
solution polymerisation. The viscosity remains essentially constant at the viscosity
of the continuous phase.
2
Overview of the Chemistry and Manufacture of Hydrogenated Nitrile
Butadiene Rubber Polymers
In general, NBR polymers contain between 18 and 50 mole% ACN content, the
remainder being butadiene. A crude schematic of the emulsion polymerisation is
shown in Figure 1.2.
Acrylonitrile
Emulsifier
Polymerisation Coagulation
Butadiene
Initiator
Water
NBR are copolymers and follow the MayoLewis copolymer equation [1]. Since two
monomers, ACN and butadiene, are involved the active terminal grouping in the
growing chain can undergo the following possible reactions:
3
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
The relative reactivity ratios of the monomers can then be defined as follows:
d[M1] [M1]*(r1[M1]+[M2])
= (1.7)
d[M2] [M2]*([M1]+r2[M2])
where [M1] and [M2] are the molar concentrations of monomer 1 and monomer 2,
respectively.
r1 = r2 >> 1. The two monomer reactivity ratios are high, indicating that the
activated monomers have little tendency to react with one another. Each activated
monomer would more probably react with the same monomer, leading to a mixture
of homopolymers in the resulting product (1-1-1-1-1-1-1-.. + 2-2-2-2-2-2-..).
r1 = r2 > 1. When monomer 1 forms the reactive chain end it will be more likely
to react with further monomer 1. In the instance where cross-polymerisation
occurs and monomer 2 becomes the reactive chain end, activated monomer 2
will probably react with further monomer 2. The resulting polymer will then
be a block copolymer containing long segments of monomer 1 attached to long
segments of monomer 2 (1-1-1-1-1-.2-2-2-2-2-2-2.).
4
Overview of the Chemistry and Manufacture of Hydrogenated Nitrile
Butadiene Rubber Polymers
r1 >> 1 >> r2. At the beginning of the polymerisation monomer 1 has a more
reactive chain end than monomer 2, and the initial copolymer is therefore rich
in monomer 1. Later, as the monomer 1 concentration is reduced, the copolymer
becomes rich in monomer 2. This is known as compositional drift (early in the
reaction: 1-1-1-1-1-2-1-1-2-1-1-1-.., followed later by: 1-2-2-2-2-2-2-1-2-2-2-
1-2-2-2.).
0 < r1 < 1 and/or 0 < r2 < 1. The reaction tends towards homopolymerisation,
and this tendency increases with decreasing rX.
We can also rearrange the MayoLewis equation to express the ratio of monomer
1 in the instantaneous polymer with respect to the monomer concentrations and
reactivity ratios:
[r1f1 2 + f1f2]
F1 = (1.8)
[r1f1 2 + 2f1f2 + r2f2 2]
where:
Against this background, the question arises as to the values of r1 and r2 for the ACN
and butadiene monomers used in the emulsion polymerisation of NBR. From this
we can gain a perspective on some of the challenges presented by the polymerisation
reaction.
A paper describing the reactivity ratios of ACN and butadiene has been published
by Embree and co-workers [2]. The emulsion polymerisation of NBR polymers has
a number of complicating factors:
ACN has reasonably high solubility in the aqueous continuous phase. ACN
monomer will therefore be present in both the emulsified and the continuous
phase.
The resulting NBR polymer is not soluble in butadiene, and this may result in a
non-homogeneous emulsified phase.
5
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
In their study of polymer composition and molecular weight, Embree and co-workers
used previously published data in addition to a number of their own experiments.
Their analyses indicated that r1 (butadiene) was 0.18 and r2 (ACN) was 0.03. In this
case, r1 >> r2, and both r1 and r2 were much lower than 1.0. If the reactive chain end
was a butadiene monomer, the reactivity ratios would indicate a greater probability
that the reactive chain end would react with a further butadiene monomer unit. On
the other hand, in the case where the reactive chain end was an ACN monomer the
reactivity ratios would indicate a greater likelihood that the reactive chain end would
also react with another butadiene monomer. From these reactivity ratios we would
expect the resulting NBR copolymer to contain small segments of repeating butadiene
and relatively isolated ACN units.
We can also use the data of Embree and co-workers [2] to understand the effect
of conversion on the composition of the polymer (i.e., how much of the initial
monomer charge is converted to polymer) for a given initial charge of monomers
into the emulsion polymerisation system. For example, if we take their example of a
75/25 molar charge of butadiene/ACN monomers in the mixture, their data may be
replotted as shown in Figure 1.3.
30
% ACN
% ACN incremental
20
polymer
10 % ACN in polymer mix
0
0 20 40 60 80 100
% Conversion
6
Overview of the Chemistry and Manufacture of Hydrogenated Nitrile
Butadiene Rubber Polymers
since the remaining monomer will have to be stripped out of the polymer. From Figure
1.3, a practical balance for this blend of monomers might give 7080% conversion,
when the overall polymer produced would contain roughly 33 mole% ACN, with
reasonably high conversion of monomer to polymer, and a reduction in the energy
and resource required to strip residual monomer from the resulting emulsion mixture.
We can also use Equation (1.3) to determine the mole% ACN in the instantaneous
polymer formed as a function of the mole% composition of monomer (Figure 1.4):
100%
90%
80%
70%
polymer NCA%
60%
50%
40% %ACN in polymer
30%
20% Ideal
10%
0%
0% 20% 40% 60% 80% 100%
%ACN in monomer feed
The ideal line in Figure 1.4 would be one at which the reactivity ratios are 1.00 and
the monomer feed composition results in the corresponding instantaneous polymer
composition.
The reactivity ratios of the two monomers, butadiene and ACN, indicate that
the NBR polymers formed will have at least short segments of repeat butadiene
monomer and relatively isolated ACN monomer units. This is particularly true
in the case of NBR polymers of very low ACN content.
7
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
At high conversion values the composition of the NBR polymer changes, and
the resulting bulk NBR polymer may well be a blend of a range of polymers of
different ACN content.
The reactivity ratios of ACN and butadiene tend to favour the presence of
alternating monomers in the resulting polymer.
The hydrogenation process follows the scheme shown in Figure 1.5. As discussed
previously, the first significant step is emulsion polymerisation to give the initial
NBR polymer. The NBR polymer is then coagulated, dried and stripped of residual
monomer. The hydrogenation reaction is completed by dissolving the NBR polymer
in a suitable solvent, followed by catalytic hydrogenation to give HNBR.
Hydrogen
NBR Polymer
Solvent
Separation from
Dissolution solvent and solvent
High pressure
catalytic hydrogenation removal
Figure 1.5 Schematic process for the catalytic hydrogenation of NBR to form
HNBR
8
Overview of the Chemistry and Manufacture of Hydrogenated Nitrile
Butadiene Rubber Polymers
NBR polymer
----{CH2CH}x----{CH2CH=CHCH2}y---{CH2CH}z ---- + (2x+2y) H 2 (gas)
CN CH=CH2
acrylonitrile 1,4-butadiene 1,2-butadiene
addition
addition, vinyl
residue
Metal catalysts, high pressure
HNBR polymer
----{CH2CH}x----{CH2CH2 CH2 CH2}y ---{CH2CH=CHCH2}a ---{CH2CH}z ----
CN CH 2 CH3
residual
unsaturation
acrylonitrile ethylene for vulcanisation propylene
The emulsion polymerisation process to give the NBR starting polymer may be
summarised as follows:
The reactivity ratios of the two monomers, butadiene and ACN, indicate that
the NBR polymer formed will comprise short segments of repeating butadiene
and relatively isolated ACN monomer units. This is particularly true for NBR
polymers of very low ACN content. The resulting HNBR polymer would thus
contain segments of an ethylenepropylene polymer of high ethylene content.
At high conversion values, the composition of the NBR polymer changes, and it
may well comprise a blend of polymers of different ACN content.
9
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
when the end application demands improved low temperature flexibility. However,
HNBR polymers of low ACN content produced from conventional NBR polymers
actually have inferior low temperature flexibility to HNBR polymers of medium or
high ACN content. The tendency for ethylenepropylene segments of high ethylene
content to crystallise is an example of this. In the authors laboratories the glass
transition temperatures (Tg) of ethylenepropylene rubber polymers over a range
of ethylene levels and low residual unsaturation have been studied using differential
scanning calorimetry (DSC). The results are shown in Figure 1.7.
20
30 Tg (C)
Tg
linear Tg (C)
40
y = 1.2688 x 102.79
50 R2 = 0.9679
60
Mole% ethylene
10
Overview of the Chemistry and Manufacture of Hydrogenated Nitrile
Butadiene Rubber Polymers
The challenge for manufacturers of HNBR polymers is thus to disrupt the regularity
of the high ethylene content segments of the resulting HNBR polymer. This can be
achieved by introducing a third diene monomer into the initial NBR polymerisation,
since this inhibits the tendency of the ethylene segments to crystallise. An example is the
emulsion copolymerisation of ACN, 1,3-butadiene and isoprene, with 1,3-butadiene/
isoprene ratios lying within the range 0.75/1.0 to 1.0/0.75 [4]. The pendant methyl
group on the isoprene monomer then disrupts the regularity and crystallisation of the
resulting hydrogenated polymer. The reaction sequence and final polymer structure
are indicated in Figure 1.8.
NBR Polymer
----{CH2CH}x----{CH2CH=CHCH2}y----{CH2CH=CHCH2}y--{CH2CH}z ----
CN CH3 CH=CH2
arylonitrile 1,4-butadiene 1,4-isoprene 1,2-butadiene
addition addition addition, vinyl
residue
HNBR Polymer
----{CH2CH}x----{CH2CH2 CH2 CH2}y ---{CH2CH=CHCH2}a ---{CH2CH}z----
CN CH2 CH3
residual
unsaturation
acrylonitrile ethylene for vulcanisation propylene
higher
propylene
content due to
Isoprene
11
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
1 mole of isoprene = 0.5 moles ethylene + 0.5 moles propylene in HNBR polymer.
1.4 Summary
The background to the technology of the HNBR polymer manufacturing process and
a brief review of some of the challenges involved in manufacturing these polymers
have been presented.
References
1. F.R. Mayo and F.M. Lewis, Journal of the American Chemical Society, 1944,
66, 9, 1594.
12
2
Types of Hydrogenated Nitrile Rubber
Polymers Available
2.1 Introduction
This chapter provides an overview of the types and grades of HNBR polymers
commercially available and their performance characteristics in the context of other
oil-resistant polymers.
Table 2.1 summarises the grades of HNBR polymers available from two manufacturers,
Zeon Chemicals and Lanxess. The primary sort was made by mole% ACN, a
secondary sort by approximate residual unsaturation and a tertiary sort by nominal
Mooney viscosity. It is seen that HNBR polymers are available with mole% ACN
within the range 17 to 50%, at a variety of residual unsaturation levels and bulk
viscosities. For a given mole% ACN content, a higher residual unsaturation figure
is useful for providing more rapid sulfur vulcanisation or when greater crosslink
density is required in the finished product. A lower bulk Mooney viscosity results in
improved high-shear flow in transfer or injection moulding.
13
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
14
Types of Hydrogenated Nitrile Rubber Polymers Available
15
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
In addition to these grades, speciality polymers are available for specific applications
and end-products. These are summarised in Table 2.2.
Modification with acrylates such as zinc methacrylate gives tough and mechanically
durable materials for use in power transmission belts and conveyor belts, and in
similar products requiring a high level of resistance to abrasion, cutting and wear.
16
Types of Hydrogenated Nitrile Rubber Polymers Available
One way to illustrate the position of HNBR within the elastomer market is to employ
the ASTM D2000 classification system [1]. This is illustrated in Figure 2.1, along with
a selection of other elastomers for comparison. Not only does the HNBR family fill a
gap in the properties identified, HNBR also offer tear resistance, abrasion resistance
and overall toughness that polymers such as acrylic elastomers (ACM) and ethylene
acrylic copolymer elastomers (AEM) cannot match.
0
200 150 100 50 0
Class by Oil Swell
NBR are among the oldest and most widely used oil-resistant polymers and provide
an excellent combination of properties and durability, but due to their high level of
unsaturation they are prone to oxidation and to attack by sulfur. HNBR materials are
much more resistant to oxidation and sulfur attack and combine the compounding
flexibility and toughness of NBR with improved temperature and chemical resistance.
This is illustrated in Figures 2.2 and 2.3 in the case of resistance to engine oil and
automatic transmission fluids, respectively. The loss of elongation after 1008 hours
at 150 C is shown for HNBR, NBR, ACM and AEM materials [2]. An estimate
of the life of an elastomer is the time required to reduce the elongation to 50% of
its original value. Figures 2.2 and 2.3 show that HNBR materials still retain some
useful life after long-term ageing, whereas NBR and CSM materials had reached the
17
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
end of their useful life before the end of the ageing test. The ACM material showed
excellent long-term ageing performance in these fluids, but did not have the tear and
abrasion resistance of HNBR materials.
Comparison of Oil-Resistant
Elastomers, 10W30SG engine oil at 150 C
40
30
% Elongation Loss
20
10
0 % Elongation Loss, 504 h
10
20 % Elongation Loss, 1008 h
30
40
50
60
35% ACN 25% ACN ACM AEM
HNBR HNBR
Material
Figure 2.2 Comparison of engine oil resistance of HNBR, ACM and AEM
elastomers
Comparison of Oil-resistant
Elastomers, Dexron III ATF at 150 C
25
20
% Elongation Loss
15
10
5 % Elongation Loss, 504 h
0
5 % Elongation Loss, 1008 h
10
15
20
25
30
35% ACN 25% ACN ACM AEM
HNBR HNBR
Material
18
Types of Hydrogenated Nitrile Rubber Polymers Available
200
Relative resistance to
150
abrasion
100
50
0
45% ACN NBR ECO FKM ACM
HNBR
Material
The effects of abrasion, wear and friction are not well understood, but this type
of testing is valuable in providing a relative rating for a range of materials. The
combination of increased high temperature performance and improved wear resistance
emphasises the niche occupied by HNBR among the available oil-resistant elastomers.
In the remainder of this chapter the effect of HNBR and compounding ingredients
on the final properties of the compound are discussed. This is not intended to be
an exhaustive account of every compounding possibility, but the use of statistical
experimental design will be emphasised and illustrated as a cost-effective way of
studying the compounding variables.
19
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
2.4 Summary
This chapter has given a brief overview of the polymers currently available and how
these fit into the overall range of oil-resistant elastomers. Formulation guidelines
and examples of formulations for specific applications will be given in detail later.
References
3. Zetpol Product Guide, Zeon Chemicals LP, Louisville, KY, USA, 1999.
20
3
The Flow and Processing Behaviour of
Hydrogenated Nitrile Rubber
3.1 Introduction
In this chapter the principles of polymer viscoelastic flow behaviour are discussed.
There are obviously a number of variables, such as compounding, and mould and
machinery design, which will influence the flow behaviour of fully compounded
HNBR materials. This chapter is not intended to discuss the finer details but will offer
some insight into the many factors affecting flow and processing behaviour. This will
provide guidelines for studying flow and processing behaviour, as well as indicating
some of the useful tools available for solving production processing problems.
3.2.1 Viscosity
Rubber and plastic melts can be regarded as extremely high viscosity fluids. This is
only an approximation, however, and polymers generally show viscoelastic properties,
a combination of viscous flow and elastic recovery. In turn viscosity is a measure of
resistance to flow under a given set of circumstances. Viscosity is designated by the
Greek letter h. For an ideal or Newtonian fluid, viscosity is simply the ratio between
the shear stress (t), the pressure applied to the fluid to create flow, and the shear rate
(g), the flow occurring over a given time:
h = t/g, (3.1)
where h = viscosity expressed as Pa. s, t = shear stress in Pa, and g = shear rate in s1.
For an ideal fluid the viscosity is thus constant at a given shear stress over the entire
range of shear rates. Life would certainly be much easier if viscosity were as simple
21
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
as this. However, polymer melts such as rubbers do not in general exhibit uniform
viscosity across a range of shear rates, in fact they are described as pseudoplastic,
which means that their viscosity is reduced as the shear rate increases.
A typical viscosity curve for a rubbery material as a function of shear rate is shown in
Figure 3.1. At low shear rates, the material shows Newtonian behaviour, which means
that its viscosity remains constant regardless of shear rate. As the shear rate increases,
the material is transformed from Newtonian to pseudoplastic, and its viscosity then
decreases with increasing shear rate. This is a curve that we will encounter many times
later in the chapter and we will see how it can help in understanding flow during
moulding and in overcoming flow-related problems. A mental picture of this curve
and its shape will help in practical situations.
Newtonian region
1000000
Viscos ity (poise )
100000
10000
1000
1 10 100 1000
Figure 3.1 Typical profile of viscosity versus shear rate for a rubber polymer
Once we appreciate that rubber materials are not ideal Newtonian fluids, and that
their viscosity decreases with increasing shear rate, we find that these variables have a
22
The Flow and Processing Behaviour of Hydrogenated Nitrile Rubber
mathematical relationship. A great deal of scientific analysis has been devoted to the
mathematics underlying the flow of rubber and other polymers [111], and Equations
(3.2) and (3.3) provide a starting point:
= Kgn1, (3.2)
where is the viscosity, g is shear rate in s1 and K and n are a curve-fitting parameters
determined from the experimental data.
where o is the viscosity at zero shear rate and A, n and K are curve-fitting parameters
determined from the experimental data.
Equation (3.2) is known as the Power Law and Equation (3.3) is the Cross Model.
Plots of these two models are shown in Figure 3.2. In general, the Cross Model is
the more successful in describing the actual viscosity versus shear rate data observed
for commercial polymers.
100000
Viscosity
10000
1000
1 10 100 1000
Shear rate
23
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
The shear rate in compression moulding is normally within the range 110 s1, for
transfer moulding in the range 101,000 s1, and for injection moulding in the range
50010,000 s1. Combinations and variations of these techniques give intermediate
shear rates and a number of fresh problems.
3.2.2 Elasticity
Unfortunately the elastic response of rubber to flow is probably the least well-
understood feature of its viscoelastic nature. A good illustration of elastic flow
is provided by chewing gum when pulled to breaking point. After chewing and
softening, if one pulls the chewing gum to fracture the broken ends of the gum are
seen to retract. The flow of the gum is the extension achieved on pulling, and its
elasticity is represented by the recovery of the broken ends. In compression, transfer
or injection moulding the flow of the rubber compound is similarly accompanied by
elastic rebound.
If the elastic component of rubber flow is not allowed for, the results can sometimes
be quite expensive. One of the best ways to study this effect has been to study the
extrudate swelling (die swell) of polymers forced through a die. As the polymer
exits the die it swells in diameter and decreases in length. Unfortunately, at the high
shear rates encountered in injection moulding and with lightly filled elastomers,
the extrudate swelling can be anything but smooth and predictable. The familiar
knotted and twisted appearance of unfilled or lightly loaded elastomers on exiting
a die at moderate to high shear rates is very difficult to quantify and study. This
may also account for the knotted, twisted or dimensionally non-compliant nature
of moulded rubber parts when the compound is lightly loaded or when shear rates
are particularly high.
In an attempt to quantify and describe extrudate swelling Tanner [12] has found that
the extrudate swell, B, can be estimated using Equation (3.4):
B = C + [1 + (tgw)2]1/6, (3.4)
where C is a constant, approximately 0.12, t is shear stress and gw is the shear rate
at the wall of the die.
Equation (3.4) applies to molten polymers but does not properly describe fully filled
or compounded elastomers. It is however useful for quantifying the change in elastic
24
The Flow and Processing Behaviour of Hydrogenated Nitrile Rubber
rebound induced by increasing shear rate and shear stress. Using Equation (3.4) we
can establish the proportionality between elastic recovery and shear stress or shear
rate. Figure 3.3 shows the effect of a percentage increase in shear rate on the extrudate
swell, B, or elastic recovery.
45
40
35
30
% Die swell
25
20
15
10
5
0
1 10 100 1000
Shear rate (s )
1
25
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
Like engine oil, polymer melts show reduced viscosity with increasing temperature. In
the absence of experimental data defining the curves, which is normally the case, this
can be determined using the WilliamsLandelFerry relationship for timetemperature
superposition [13]. This relationship can be applied to values of modulus and also
viscosity. Applying the relationship to viscosity, we obtain Equation (3.5):
where T1 is the viscosity at absolute temperature T1, Tg is the viscosity at the glass
transition temperature and aT is a shift factor derived from the data or by estimation.
This is shown in Figure 3.4 for the theoretical polymer shown in Figure 3.1, assuming
the Tg to be 20 C. As materials move further away from the Tg the reduction in
viscosity with temperature becomes less apparent, and for a material with a very
low Tg, such as a silicone, the reduction in viscosity with increasing temperature is
similarly less dramatic.
26
The Flow and Processing Behaviour of Hydrogenated Nitrile Rubber
1000000
Viscosity (Pa. s)
100000
100 C
10000
125 C
1000 150 C
100
1 10 100 1000
Shear rate (s )
1
One of the consequences of the viscoelastic behaviour of rubber is heat build-up during
high shear flow. During relatively low shear compression moulding this is probably
not a significant factor, but on the other hand during higher shear moulding, such
as in transfer and injection moulding, heat build-up can be the cause of premature
scorching and other quality problems. An exact relationship has not been established
to describe heat build-up during high shear flow, but Equation (3.8) can predict the
effect of an increase in viscosity and shear rate.
where DT is the temperature change in C s1, is the viscosity in Pa. s and g is shear
rate in s1.
DT h, and DT g2 (3.9)
27
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
Using these proportionalities we can plot the effect of viscosity and shear rate on heat
build-up, as illustrated in Figure 3.5.
350%
% change in heat build-up
300%
250%
200%
150%
100%
50%
0%
50%
100%
100% 50% 0% 50% 100% 150%
% change in parameter
Figure 3.5 Effect of viscosity and shear rate on heat build-up during flow
Using the example from the previous section we have studied the decrease in injection
cycle time resulting from increasing the injection pressure and by reducing the injection
time from 20 s to 10 s. Since the configuration of the runner system and the mould
cavities remained unchanged, a reduction in injection time by half had the effect of
doubling the shear rate. In turn, doubling the shear rate caused a four-fold increase in
heat build-up during flow. If this increase in heat build-up was sufficient for the critical
scorch temperature of the vulcanisation system to be reached, we would encounter
premature scorch problems, in addition to dimensional variances. Obviously, as we
increase the shear rate the viscosity decreases, causing a linear decrease in heat build-
up. However, the dependence of heat build-up on the square of the shear rate can
easily override the viscosity drop and give rise to scorching.
28
The Flow and Processing Behaviour of Hydrogenated Nitrile Rubber
This is the starting point for practical rubber compounds and will naturally be the
first item to consider. The molecular weights of commercial polymers, including
rubbers, tend to be skewed towards higher figures. A variety of averages are used to
describe the molecular weight distribution and its width. The general formula for
the j-average molecular weight (Mj) is given by Equation (3.10):
Mj = (NiMij)/(NiMij1), (3.10)
29
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
Mn Mw
Mv Mz
0.5
0.4
0.3
Fraction
0.2
0.1
0
0 100000 200000 300000 400000 500000 600000 700000
Molecular w eight
A great deal of experimental work has been carried out in the past to understand the
influence of molecular weight on viscosity [1422]. In linear polymers the viscosity
varies linearly with molecular weight until a critical value of molecular weight is
reached, at which entanglement of the polymer chains occurs, and the viscosity then
varies as the power to 3.4 of the molecular weight, as in Equation (3.11):
The value of Mc varies with the type of polymer, and in vinyl (carboncarbon
backbone) polymers it is dependent on the bulk of the pendant groups. The value of
Mc for polyethylene (a carboncarbon backbone with no bulky pendant groups) is
approximately 4,000, while that for polystyrene (carboncarbon backbone with a
bulky phenyl pendant group) is approximately 38,000. In general, rubber polymers
have molecular weights well above Mc and the viscosity shows a power dependency
of 3.4 on molecular weight.
Since rubber polymers contain a distribution of molecular weights, the question then
arises as to which average molecular weight should be used to establish the dependency
of viscosity. Fox and co-workers found that the weight-average molecular weight,
Mw, was best able to describe the dependence of viscosity on molecular weight,
Equation (3.12):
30
The Flow and Processing Behaviour of Hydrogenated Nitrile Rubber
ho = kMw3.4, (3.12)
where ho is zero shear rate viscosity and k is a constant for a particular polymer system.
A convenient way of estimating viscosity and molecular weight is by the use of the
traditional Mooney viscometer. Since this operates at a shear rate around 2 s1, and
this is very close to the zero shear rate viscosity, it should show proportionality to
Mw3.4. The data reported by polymer suppliers as Mooney viscosity can thus give an
estimate of the relative average molecular weight. As a first approximation we can
assume that two polymers varying in Mooney viscosity will have the same degree of
branching or linearity, and the viscosity shear rate curves will be parallel. Therefore,
if we know the viscosity difference between the two polymers at low shear rates and
we also have the viscosity shear rate curve for one of the polymers, we can estimate
the viscosity at higher shear rates. For example, if two polymers from the same
supplier and made by a similar polymerisation method have Mooney viscosities of
35 and 50, and we know the relative viscosities of the two polymers at high shear
rates, these should follow the ratio of Mooney viscosities. In other words, the polymer
of Mooney viscosity 50 s1 has roughly 43% higher viscosity at low shear rate, and
should also have roughly 43% higher viscosity at higher shear rates, since the curves
are parallel. This example is illustrated in Figure 3.7.
100000
Viscosity (poise)
31
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
Unfortunately this area is not well understood. The basic theory of particle
reinforcement of liquids was developed as a refinement to the Einstein equation [25,
26]:
h1 = h0 (1 + Af + Bf2), (3.13)
where h0 is the unfilled viscosity, h1 is the viscosity of filled material, f is the volume
fraction of filler in the mixture, A is a constant (2.5 for spherical particles) and B is
a different constant (within the range 10 to 14 for spherical particles).
G1 = G0 (1 + Af + Bf2), (3.14)
where G0 is the shear modulus of unfilled material, G1 the shear modulus of filled
material and f is the volume fraction of filler in the mixture.
As can be deduced from Equation (3.13), the type of filler should have neither an
impact on viscosity nor on modulus of elasticity. In the practical world that we live
in, we know that the loading of N110 SAF carbon black at the same volume or phr
(parts per hundred rubber polymer) gives much higher viscosities and much higher
modulus values compared to N990 MT carbon black. Thus, Equation (3.13) is not
very effective in predicting the effect of fillers. From this point onwards the situation
becomes unclear, since comprehensive studies of various fillers and their effect on
viscosity and modulus are relatively rare.
White and Crowder [27] studied the effect on viscosity of various low-structure carbon
blacks and found that those of smaller particle size caused an increase in viscosity
and shifted the viscosity relative to the shear rate curves. The use of low-structure
carbon blacks limited the structural effect caused by the agglomeration of particles.
Unfortunately, many useful commercial elastomer compounds employ high-structure
carbon blacks, such as furnace black N550 FEF.
In addition, the author has studied the combined effect of structure and particle size
of carbon black on the properties of an NBR compound. Obviously the particle size
on its own does not determine the results, but fortunately the DBP number, given in
32
The Flow and Processing Behaviour of Hydrogenated Nitrile Rubber
ASTM D2414, combines the effect of particle size and structure. The DBP number
represents the quantity of dibutyl phthalate (DBP) that the carbon black can absorb,
measured by titration to a mechanically sensed end-point; when the carbon black
surface can absorb no more DBP the conversion from free-flowing powder to paste
is sensed as a change in torque. The experimental design programme is summarised
in Table 3.1 and the results for compression modulus and low-shear rate viscosity
are shown in Figure 3.8.
33
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
15.5
Compression modulus
13.5
11.5
9.5
7.5 150
120
5.5 90
60
30 30 DBP number
40 50 0
60 70
phr Carbon black
The correlation between the experimental values and the plotted response surfaces was
greater than 0.900, confirming that the DBP number gives a good prediction of the
reinforcing nature of carbon blacks. The study was not extensive, however, and will
need to be developed further to include a comprehensive range of polymer families,
carbon blacks and other fillers. Prior to formulating compounds for compression,
transfer or injection moulding, the compounder should conduct a similar series of
experiments to optimise the fillers that will give a combination of properties such as
compression set, strength of vulcanisate and flow in the mould.
In the study, elasticity was seen to improve with increasing filler loading and DBP
value of carbon black, confirming earlier observations [28, 29]. Loss of elasticity was
measured in terms of the hysteresis observed in dynamic compression testing, and is
illustrated in Figure 3.9. It is of course well known in the rubber industry that high
structure carbon blacks of small particle size give smooth extrusion, which delivers
extrusions with smooth or glossy surfaces.
34
The Flow and Processing Behaviour of Hydrogenated Nitrile Rubber
% Hysteresis, Compression
63
53
43
33
130
110
23 90
70
30 40 50 DBP Number
5060 30
70
phr Carbon Black
where f2 is the volume fraction of polymer in the compound, F is a constant for the
polymer and plasticiser and g is shear rate.
From Equation (3.16) we can establish the proportionality between plasticised and
unplasticised material, as follows:
35
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
As more plasticiser is added, viscosity at a given shear rate falls. This is plotted
in Figure 3.10, using our example from Figure 3.1. For a particular polymer and
plasticiser system it should be straightforward to run the experiments and adapt the
parameters of Equation (3.15) to the design of other compounds.
1000000
100000 Phi = 1
Viscosity, Pa. s
Phi = .9
10000
Phi = .8
1000
1 10 100 1000
Shear rate (s )
1
Most process aids are designed to assist mould flow and mould release, and function
by having only limited solubility in the polymer matrix. For example, microcrystalline
polyethylene waxes make good process aids for highly polar polymers such as
fluorocarbon elastomers (FKM) since they are deposited at the surface of the shear flow
front and on the moulded parts, assisting lubrication and release. Similarly, specialised
FKM polymers are available as process aids and slip agents for use in polyethylene
manufacture. In general, effective process aids for mould flow and mould release
are used at very low and tightly controlled levels in the polymer. Using our example
of microcrystalline wax in FKM materials as a process aid in transfer or injection
moulding, while 0.500 phr is very effective, 1.00 phr is also effective but may lead
to mould deposits during extended operation involving several hundred cycles, and
levels of 1.50 phr or above will give knit lines and parts that are wet with excess
wax as they emerge from the mould.
36
The Flow and Processing Behaviour of Hydrogenated Nitrile Rubber
Techniques are available for studying the effect of process aids on mould flow, for
example a capillary rheometer may be useful for this purpose. The pressure drop
through the die of a capillary rheometer is made up of the total pressure drop at the
ends plus the pressure drop within the die:
where PT is the total pressure drop through the die, Dpe the pressure drop at the ends
of the die and Dp is the pressure drop within the die.
The shear stress at the wall of the capillary is governed by the total pressure drop
and the length and diameter of the die:
where (s12)w is the shear stress at the wall, L the die length and D the die diameter.
By running experiments at constant die diameter and total volumetric flow rate but
with different L/D ratios, the slope of the plot of PT versus L/D allows the shear stress
at the wall to be ascertained. From this, the effectiveness of process aids in improving
flow can be measured in terms of the reduction in shear stress at the wall caused by
their lubricating effect.
Assessing the effectiveness of a process aid in terms of mould cleanliness and release
can be challenging. In many cases the only accurate measure is a carefully monitored
extended production run, comparing long-term release and mould cleanliness with
and without the process aid, or by comparing different process aids.
This is an area that seems to lack a basic understanding at the plant and production
level in the speciality elastomer industry. A typical problem might involve a multi-deck
injection mould where there is insufficient flow to one of the decks. For a variety of
37
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
reasons the default diagnosis in elastomer production needs to be to fix this type of
problem by changing or reformulating the rubber material. However, when we look
at the underlying science this would seem to be the least effective solution to flow
problems in production processing.
Viscosity is simply the ratio of shear stress and shear rate. For flow through a round
capillary this can be expressed as follows:
where DP is pressure drop and R and L are the radius and length of the capillary,
respectively.
By only a small increase in the radius of the flow channel the volumetric flow rate will
thus increase by the fourth power of the radius. Decreasing the apparent viscosity of
the compound, happ, will therefore have not nearly as profound an effect on improving
flow rate as a small change in the radius or gap of the flow channel. The relative
impact of these various factors is illustrated in Figure 3.11.
From Equation (3.21) and Figure 3.11, it is clear that a minor change in a sprue or
runner system will have a disproportionate effect on rubber flow. In the case of a
double-deck injection mould in which one deck is not filling completely, it is obvious
that a small increase in the diameter of the sprues or runners leading to the problem
deck would provide the most effective solution to the problem. It is strange that
the immediate response often seems to be to reduce the viscosity by changing the
formulation.
38
The Flow and Processing Behaviour of Hydrogenated Nitrile Rubber
120.0%
Net % change on flow rate
100.0%
80.0%
60.0%
40.0%
20.0%
0.0%
20.0%
40.0%
60.0%
20.0% 10.0% 0.0% 10.0% 20.0% 30.0%
% Change in factor
Radius of runner Length of runner Viscosity of material
References
1. E.A. Collins and J.T. Oetzel, Rubber Age, 1970, 102, 64.
7. N. Nakajima and E.A. Collins, Polymer Engineering Science, 1974, 14, 137.
8. F.C. Weissert and B.L. Johnson, Rubber Chemistry and Technology, 1967,
40, 590.
9. J.L. White and J.W. Crowder, Journal of Applied Polymer Science, 1974, 18,
1013.
39
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
10. J.L. White and N. Tokita, Journal of Applied Polymer Science, 1974, 9, 1929.
11. W.E. Wolstenholme, Rubber Chemistry and Technology, 1965, 38, 769.
12. R.I. Tanner, Journal of Polymer Science, Part A2, 1970, 8, 2067.
13. M.L. Williams, R.F. Landel and J.D. Ferry, Journal of the American Chemical
Society, 1955, 77, 3701.
14. P.J. Flory, Journal of the American Chemical Society, 1940, 62, 1057.
15. J.R. Shaefgen and P.J. Flory, Journal of the American Chemical Society, 1948,
70, 3709.
16. T.G. Fox and P.J. Flory, Journal of the American Chemical Society, 1949, 70,
2384.
17. T.G. Fox and V.R. Allen, Journal of Chemical Physics, 1964, 41, 344.
18. T.G. Fox and S. Loschaek, Journal of Applied Physics, 1955, 26, 1082.
19. W.F. Busse and R. Longworth, Transactions of the Society of Rheology, 1962,
6, 179.
21. R.S. Porter and J.F. Johnson, Proceedings of the 4th International Rheology
Congress, Providence, RI, USA, 1965, 2, 467.
23. L.A. Belfiore, Physical Properties of Macromolecules, John Wiley and Sons,
Hoboken, New Jersey, 2010, p.403.
24. W.F. Busse and R. Longworth, Transactions of the Society of Rheology, 1962,
6, 179.
27. J.L. White and J.W. Crowder, Journal of Applied Polymer Science, 1974, 18,
1013.
40
The Flow and Processing Behaviour of Hydrogenated Nitrile Rubber
28. C.D. Han, Journal of Applied Polymer Science, 1974, 18, 821.
29. N. Minagawa and J.L. White, Applied Polymer Science, 1975, 20, 501.
30. G. Kraus and J.T. Gruver, Transactions of the Society of Rheology, 1965, 9,
2, 17.
41
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
42
4
Properties of Hydrogenated Nitrile Rubber
and Comparison with Other Elastomers
4.1 Introduction
In this chapter the properties of hydrogenated nitrile rubber (HNBR) are discussed
and compared with those of other speciality elastomers. This background will provide
the perspective for later chapters in which the applications of HNBR materials are
described.
A good starting point for this comparison is to show where HNBR materials fit into
the classification used by ASTM D2000 [1], which classifies elastomers by type and
class. The type classification gives an approximate upper temperature limit for the
material and the class characterises its degree of swelling in hydrocarbon-based
IRM903 oil, which has a high aromatic content. Table 4.1 gives a tabulation of
type and class in ASTM D2000 [1].
43
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
250 H
200 F
E
150 D
C
100 NBR B
A
50
0
200 150 100 50 0
Class by Oil Swell
A B C D E F G H J K
44
Properties of Hydrogenated Nitrile Rubber and Comparison with Other Elastomers
300
250 H
200 F
E
150 D
HNBR
C
100 NBR B
A
50
0
200 150 100 50 0
Figure 4.2 Graphical representation of NBR and HNBR in the ASTM D2000
system
Finally, Figure 4.3 places HNBR within the ASTM D2000 system in the perspective
of other speciality elastomers.
45
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
300
VMQ = vinylmethyl silicone, ACM = acrylic, AEM = ethyleneacrylic, FVMQ = fluorinated vinyl
46
Properties of Hydrogenated Nitrile Rubber and Comparison with Other Elastomers
47
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
Since the HNBR recipe contained a slightly higher molar percentage of ACN this is
not an exact comparison, but it is interesting that HNBR showed noticeably greater
fluid swelling than NBR. From these data we can see that the hydrogenation of NBR
to form HNBR results in:
Since both NBR and HNBR are now available with a wide range of molar percentages
of ACN, the author conducted an extended comparison using the recipes shown in
Table 4.3. The results in terms of swelling after immersion in ASTM Reference Fuel
C for 70 hours at 23 C are shown in Figure 4.4.
Table 4.3: Recipes used for comparison of swelling in ASTM Reference Fuel
C, HNBR versus NBR
Ingredient Recipe A Recipe B
(phr) (phr)
HNBR containing required mole% ACN 100
NBR containing required mole% ACN 100
N762 carbon black 50 50
Polymerised 1,2-dihydro-2,2,4-trimethyl quinoline 1 1
antioxidant
Stearic acid 1 1
Zinc oxide (French process) 5 5
Dicumyl peroxide, 40% on inert mineral carrier 7.5 3.5
Triallyl isocyanurate, 72% on inert mineral carrier 4.5
Test slabs vulcanised 5 min at 177 C in each case
Linear regressions are provided in Figure 4.4 since this may assist readers in designing
further recipes to give resistance to hydrocarbon fuels or other types of fluid. Detailed
information on the formulation of HNBR compounds is given in Chapter 6.
48
Properties of Hydrogenated Nitrile Rubber and Comparison with Other Elastomers
90
80 HNBR
70 y=1.7361x+116.54
R2=0.9996
% Volume change
60
50 NBR
40 HNBR
30 Linear (NBR)
49
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
Comparison of oil-resistant
elastomers, 10W30SG engine oil at 150 C
40
20
% Elongation loss
-40
-60
35% ACN 25% ACN ACM AEM
HNBR HNBR
Material
Comparison of oil-resistant
elastomers, Dexron III ATF at 150 C
30
20
% Elongation loss
50
Properties of Hydrogenated Nitrile Rubber and Comparison with Other Elastomers
An area in which HNBR excels is in its wear and abrasion resistance. This is shown
graphically in Figure 4.7 by comparison with other oil-resistant speciality elastomers.
200
150
100
50
0
45% ACN NBR ECO FKM ACM
HNBR
Material
One of the more demanding areas for elastomeric products is in the petroleum
exploration and drilling environment. As exploration and drilling become deeper
increased temperatures are encountered, making NBR no longer viable for these
applications. In addition, many new sources of crude petroleum contain significant
levels of hydrogen sulfide, which is reactive to a number of traditional elastomer
components. Moreover, exploration and drilling systems often require elastomeric
components having maximum ability to withstand high pressure and abrasion.
51
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
It is seen from Table 4.4 that HNBR materials showed excellent retention of flexibility
and elasticity and an increase in percentage elongation. FEPM materials had similar
properties, but in contrast FKM and conventional NBR showed a noticeable decrease
in elongation, indicating amine attack and hardening.
The results of sour gas testing (natural gas containing 5% hydrogen sulfide) are
shown in Figure 4.8. The inclusion of 36 mole% ACN in HNBR gave excellent
resistance to sour gas compared to other oil-resistant elastomers. Table 4.4 and
Figure 4.8 also confirm the relative resistance of HNBR in petroleum exploration and
drilling environments. These properties, combined with their toughness and abrasion
resistance, make HNBR materials particularly useful in petroleum exploration and
drilling.
52
Properties of Hydrogenated Nitrile Rubber and Comparison with Other Elastomers
120
100
36%ACN HNBR, peroxide
cured
% Elongation retained
80
45%ACN HNBR, peroxide
cured
0
0 50 100 150 200
Ageing time (h)
53
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
References
3. Zeon Chemicals LP, Zetpol Product Guide, Louisville, KY, USA, 1999.
54
5
Typical Applications of Hydrogenated
Nitrile Rubber
Introduction
Previous chapters have described the manufacture, chemistry, the grades available
and basic properties of HNBR compared to other speciality elastomers. The present
chapter provides examples of its applications, for which formulations are given in
Chapter 7.
Figure 5.1 illustrates the location of HNBR within the ASTM D2000 classification
system for rubbers. This provides the context for the applications for which HNBR
is important and shows that it has good resistance to hydrocarbon fluids, similar to
that of NBR but with improved high temperature performance.
Finally, the hydrogenation of NBR to form HNBR virtually eliminates the carbon
carbon unsaturation which is the primary site for oxidation [1]. Oxidative attack is
a feature of a number of high temperature applications, and resistance to oxidation
or ozonation is critical in extending the life of flexible dynamic rubber components.
55
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
300
VMQ = vinylmethyl silicone, ACM = acrylic, AEM = ethyleneacrylic, FVMQ = fluorinated vinyl
56
Typical Applications of Hydrogenated Nitrile Rubber
This applies for example in automotive engine compartments, where elastomers such
as NBR and CR do not have sufficient thermal and oxidative resistance to fulfil the
target life of the components.
Typical belt and conveyor applications for HNBR and acrylate-modified HNBR
vulcanisates include the following:
Timing belts.
Drive belts for air conditioning compressors, alternators and hydraulic steering
pumps.
Rubber rollers:
Rollers for conveyor systems handling abrasive minerals, grain or grain by-
products.
As mentioned earlier, HNBR materials offer excellent abrasion resistance and flex
life. In addition, the hydrogenation of NBR gives it enhanced thermal resistance.
As can be seen from Figure 4.3, HNBR is also reasonably resistant to hydrocarbon
fluids. This combination makes it a good choice for flexible boots and bellows for
sealing power transmission joints, keeping dust, dirt and water out of the joint and
retaining the hydrocarbon lubricant in place.
HNBR is relatively expensive, as shown in Chapter 4, but it may be the only choice
where boots or bellows are exposed to temperatures at which elastomers such as
thermoplastic esters, olefin/rubber blends and urethanes run the risk of partial melting.
One could, for example, envisage a protective power take-off boot on an all-wheel-
drive vehicle where the boot is close to the exhaust pipe or a catalytic converter. In
such cases the temperature near the boot might easily reach 120 C, which would
57
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
be too high for the thermoplastic elastomers traditionally used. HNBR materials
offer a good balance of properties for high temperature protective boot and bellows
applications, including:
Steering boots:
Suspension components:
Tie-rod end boots and strut end covers near exhaust systems.
From Figure 4.3 we can see that the properties of NBR are supplemented in HNBR
by good resistance to hydrocarbon fluids together with enhanced performance at high
temperatures. Many of the static seals used in power transmission, such as in power-
assisted steering, automatic transmission and differentials, have traditionally been
based on NBR due to its excellent hydrocarbon fluid resistance and relatively low cost.
NBR static seals are however no longer adequate now that component temperatures
are higher, and longer warranty periods have increased the life expectancy required.
Acrylic elastomers such as ACM or AEM may be more cost effective than HNBR,
but acrylic elastomers do not provide adequate fluid pressure resistance, especially
in hydraulically assisted steering and power transmission systems.
In addition, the fluids used in many power transmission applications have been
upgraded by the inclusion of antioxidants, sludge dispersants and metal protective
additives. The reactive organic sulfur or phosphorus compounds included as metal
protective additives may themselves act as vulcanisation chemicals in unsaturated
elastomers, and the saturation of HNBR makes it much more resistant to the hardening
effect of such additives.
58
Typical Applications of Hydrogenated Nitrile Rubber
O-rings and square cross-section seals for long-life hydraulic power steering
gaskets, pumps and reservoirs:
Pump seals.
O-rings, gaskets and cross-section seals for automatic and manual transmission
fluids, differential joints and pan seals.
Due to its resistance to hydrocarbon fluids and additives, combined with its enhanced
thermal resistance and excellent abrasion resistance, HNBR is an ideal alternative
to NBR for radial shaft seals in power transmission applications. In this application
dust exclusion and abrasion resistance are vital, and HNBR has found use in the
following power transmission applications:
Input and output radial shafts in automatic and manual transmission systems.
Input shafts and reciprocating racks in hydraulic power-assisted rack and pinion
steering systems.
Good resistance to hydrocarbon fluids, oxygen and ozone, combined with thermal
stability and abrasion resistance make HNBR valuable for hoses containing
hydrocarbon hydraulic fluids and lubricants, for which NBR is prone to fluid additive
attack at high temperatures. HNBR can be used in the liner, the reinforcing layer or
the cover for such hoses.
59
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
Static seals:
O-rings.
Dynamic seals:
O-rings:
Tubing seals.
These include:
60
Typical Applications of Hydrogenated Nitrile Rubber
Static seals for hydrocarbon fluid cooler systems (e.g., engine oil, or power
transmission fluids):
O-rings.
Gaskets.
Engine fuel static seals (away from the heat of the engine):
References
61
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
62
6
Formulating Guidelines for Hydrogenated
Nitrile Rubber
6.1 Introduction
In this chapter the general guidelines for HNBR elastomer formulations are discussed,
including those for vulcanisation, filler and plasticiser systems.
In Chapters 1 and 2 the manufacture, chemistry and the available grades of HNBR
polymers were discussed. Most commercial HNBR polymers contain a small degree
of residual unsaturation in the polymer backbone. This provides the opportunity for
vulcanisation, either using accelerated sulphur- or peroxide-based systems. The most
reactive areas of unsaturation in NBR polymers naturally provide preferential sites
for catalytic hydrogenation. It follows that any residual unsaturation in a HNBR
polymer is likely to be of a less reactive nature. This means that in the case of either
accelerated sulfur or peroxide vulcanisation a much higher level of crosslinking agent
and accelerator will normally be required for HNBR than for NBR.
A general study of sulfur-cured systems has been conducted by Zeon Chemicals, and
the results are summarised in Table 6.1 [1]. The Zeon study was carried out using
a base polymer with approximately 9% residual unsaturation. From the available
polymer grades listed in Chapter 2 it is seen that very low residual unsaturation may
not be adequate for accelerated sulfur vulcanisation.
Zeon has also conducted an interesting study on the peroxide plus coagent
vulcanisation of a polymer of medium ACN level and 4% residual unsaturation [2].
The results of this study are summarised in Table 6.2 and are plotted in Figures 6.1
to 6.4.
63
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
64
Formulating Guidelines for Hydrogenated Nitrile Rubber
Table 6.2 Peroxide + coagent: A study by Zeon [2], recipes and properties
Ingredient phr in recipe
1 2 3 4 5 6
Zetpol 2010 100.0 100.0 100.0 100.0 100.0 100.0
Zinc oxide 5.0 5.0 5.0 5.0 5.0 5.0
Stearic acid 0.5 0.5 0.5 0.5 0.5 0.5
N330 carbon black 40.0 40.0 40.0 40.0 40.0 40.0
Naugard 445 1.5 1.5 1.5 1.5 1.5 1.5
zinc salt of 4- and 5-methyl 1.5 1.5 1.5 1.5 1.5 1.5
mercaptobenzimidazole
,-bis(t-butylperoxy) 5.0 5.0 5.0 5.0 7.5 7.5
diisopropylbenzene
N,N-m-phenylene dimaleimide 0.0 2.5 5.0 7.5 0.0 2.5
65
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
Property Recipe
7 8 9 10 11 12
Shore A hardness 76 78 71 75 77 80
Tensile strength, MPa 29.3 27.1 32.0 28.2 28.6 27.7
Ultimate elongation, % 230 190 310 220 190 160
50% Modulus, MPa 2.8 3.8 1.8 2.5 3.2 4.4
100% Modulus, MPa 7.7 11.0 3.7 6.6 9.9 13.4
Compression set, ASTM D395 26 23 34 25 24 22
Method B, 70 h at 150 C
The response surface analysis of the data gives the following predictive equations for
the properties as a function of the peroxide and coagent levels:
66
Formulating Guidelines for Hydrogenated Nitrile Rubber
78
7
5
Coagent phr
4 76
74
3
1
72
0
5 6 7 8 9 10
Peroxide phr
Figure 6.1 Contour plot of Shore A hardness as a function of peroxide and coagent
level
3.5
7 4.0
5 2.5
3.0
Coagent phr
1 2.0
0
5 6 7 8 9 10
Peroxide phr
67
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
6
250
5
Coagent phr
4
200
3
350
2
1
450 300
400
0
5 6 7 8 9 10
Peroxide phr
5
Coagent phr
4
30
3
25
2 40
1
50
35
45
0
5 6 7 8 9 10
Peroxide phr
Figure 6.4 Contour plot of compression set as a function of peroxide and coagent
level
68
Formulating Guidelines for Hydrogenated Nitrile Rubber
Oven post-cure processes are also effective with HNBR compounds, as illustrated in
a further study conducted by Zeon (Table 6.3 and Figure 6.5) [3].
Table 6.3 Zeon study on the effect of post-cure: Recipes and properties [3]
Ingredient phr in recipe
1 2 3 4 5 6 7 8
Zetpol 1020 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0
Zinc oxide 5.0 5.0 5.0 5.0 5.0 5.0 5.0 5.0
Stearic acid 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0
N770 carbon black 40.0 40.0 40.0 40.0 40.0 40.0 40.0 40.0
Dibutoxyethoxyethyl adipate 5.0 5.0 5.0 5.0 5.0 5.0 5.0 5.0
Sulfur (325 mesh) 0.5 0.5 0.5 0.5
Tetramethyl thiuram disulfide 2.5 2.5 2.5 2.5
2-Mercapto benzothiazole 0.5 0.5 0.5 0.5
40% Dicumyl peroxide on an 8.0 8.0 8.0 8.0
inert carrier
Conditions Recipe
1 2 3 4 5 6 7 8
Press cure conditions 30 30 30 30 30 30 30 30
min min min min min min min min
at at at at at at at at
160 160 160 160 160 160 160 160
C C C C C C C C
Oven post-cure conditions, h 0 2 4 8 0 2 4 8
at 150 C
Property Recipe
1 2 3 4 5 6 7 8
Shore A hardness 65 66 66 67 65 67 67 67
Tensile strength, Mpa 27.5 23.5 25.1 25.6 24.3 26.8 26.4 25.6
Ultimate elongation, % 510 460 480 460 450 440 430 410
100% Modulus, Mpa 2.4 2.7 2.7 2.7 2.4 2.6 2.6 2.9
200% Modulus, Mpa 6.4 7.7 7.7 8.1 7.8 8.4 8.7 10
Compression set, ASTM D395 50 36 31 24 27 15 14 13
Method B, 70 h at 120 C
Compression set, ASTM D395 75 67 60 50 35 25 24 23
Method B, 70 h at 150 C
69
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
60
Compression Set (70 h at 120 C)
50
40
Sulfur
30
Peroxide
20
10
0
0 2 4 6 8 10
Post-cure time (h at 150 C)
Figure 6.5 Effect of post-cure on compression set
Carbon black is one of the most universally used fillers in elastomers, not only for
producing a black coloration but also for reinforcing and extending them. Zeon has
conducted an extensive study of the effect of various grades of carbon black on the
properties of HNBR vulcanisates [4]. In addition, work carried out by the author
also suggests that the reinforcing effect of carbon black can be predicted from its phr
and its dibutyl phthalate (DBP) absorption number [5]. By combining the results of
the Zeon study and those of the author, predictive equations can be derived from
the response surface analyses. The test recipe used by Zeon is given in Table 6.4 [4].
70
Formulating Guidelines for Hydrogenated Nitrile Rubber
The results for various carbon black levels and grades, along with nominal DBP
absorption values, are illustrated in Figures 6.6 to 6.8.
80
70
Shore A
60
125
50 100
75 DBP
0
20 50
40
60
phr
Figure 6.6 Surface plot showing effect of carbon black on Shore A hardness
12
100% modulus
4 125
100
0
75 DBP
0
20 50
40
phr 60
Figure 6.7 Surface plot showing effect of carbon black on modulus at 100%
elongation
71
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
150
ML(1+4) at 100 C
100
125
100
50
75 DBP
0
20 50
40
phr 60
Figure 6.8 Surface plot showing the effect of carbon black on Mooney viscosity
The predictive equations for these properties as a function of the phr and DBP number
of the carbon black are as follows:
72
Formulating Guidelines for Hydrogenated Nitrile Rubber
It is appreciated that these equations bear little resemblance to the theoretical equations
for filler reinforcement, and they have no theoretical basis. In the real world, however,
theoretical equations are rarely useful for predicting practical situations, and empirical
relationships derived from actual studies such as these are often more valuable. From
the standpoint of practical rubber compounding these equations are helpful for
predicting the effect of carbon black reinforcement (high values of R-squared) and
are therefore useful in formulating. In discussing the effects of fillers on viscosity and
modulus (Chapter 3), the following theoretical equation was presented:
G1 = G0 (1 + Af + Bf2) (6.8)
Equation (6.6) might suggest that the level of filler was the only important factor in
reinforcement, and not the nature of the filler. If that were the case there would really
be no requirement for the wide variety of carbon black grades available, each with
its own particle size and structure.
Here again Zeon have provided a comprehensive study of non-black fillers in a wide
range of particle sizes and reinforcing ability [6]. The test recipe and results are
summarised in Table 6.5.
73
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
74
Formulating Guidelines for Hydrogenated Nitrile Rubber
The author has calculated the reinforcing effect of various fillers in terms of the
change in hardness and 100% modulus per phr given by each. This information will
be of use in selecting the appropriate mineral filler. A typical example might be in the
production of a green article of 70 nominal Shore A hardness. A mineral filler with a
relatively low reinforcing factor is needed, and a higher level of filler might be used
to keep down the cost. We know from Table 6.5 that our base recipe without filler
would give 50 Shore A hardness. By selecting a couple of mineral fillers from Table
6.5 the reinforcing factor from the table can be used as follows:
If we needed to maximise the filler loading to keep the cost down, then Nulok 390
would be a better choice than HiSil 233 as a non-black filler, assuming the other
properties were satisfied.
6.5 Plasticisers
Since HNBR polymers are chemically related to NBR, the plasticisers commonly
used with NBR compounds are in general equally suitable for HNBR compounds.
In the case of HNBR, however, the polymer has a significantly higher temperature
limit and this means that the more volatile plasticisers suitable for NBR may not be
practical for HNBR. Zeon has conducted a study showing the effect of a variety of
plasticisers used in a simple HNBR recipe [7]. Some of the data are given in Table
6.6, together with the softening factors calculated by the author for each plasticiser.
75
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
Compound 133 86 78 81 91 87
Mooney
viscosity,
ML1+4 (100
C)
Oscillating 11.4 6.5 7.1 7.4 8.0 8.3
disc rheometer
at 170 C,
Vmin
Oscillating 51.6 40.4 46.2 41.4 53.1 56.3
disc rheometer
at 170 C,
Vmax
Oscillating 19.5 18.3 15.1 16.5 17.2 17.0
disc rheometer
at 170 C,
T95, min.
Shore A 76 70 69 69 70 69
hardness
Ultimate 310 360 450 430 380 370
elongation, %
100% 6.2 4.1 3.4 3.3 4.2 4.3
Modulus,
MPa
Temperature
retraction per
ASTM D1329
76
Formulating Guidelines for Hydrogenated Nitrile Rubber
N550 carbon 50
black
,-bis(t- 6
butylperoxy)
diisopropyl-
benzene,
40% on inert
carrier
77
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
Trioctyl trimellitate (TOTM) is one of the most common plasticisers used in HNBR
formulations and the reasons for this are clear from Table 6.6, which shows that
TOTM offers a good balance of raw compound viscosity reduction, reduction of low
temperature properties as measured by TR10, and excellent thermal stability in terms
of low weight loss at 150 C. The softening effect listed in Table 6.6 should help in
selecting the correct plasticiser and quantity, as well as determining the amount of
additional filler required to bring the hardness back to the required level.
While HNBR has much better stability than NBR to attack by oxygen or ozone, it
still needs to be stabilised against oxygen and, depending on the application, attack by
ozone. Studies by Zeon Chemicals [8] have shown that polymerised 1,2-dihydro-2,2,4-
trimethyl quinoline (available as Agerite Resin D, Flectol H or Naugard Q) at levels
between 1.5 and 3.0 phr are very effective for the protection of peroxide-vulcanised
HNBR against oxygen. The addition of 0.5 phr of a protective wax to the polymerised
1,2-dihydro-2,2,4-trimethyl quinoline further gives excellent dynamic ozone resistance.
These studies also demonstrate that an effective long-term antioxidant against air
ageing is a combination of 4,4-bis(,-dimethylbenzyl) diphenylamine, available as
Naugard 445, and a zinc salt blend of 4- or 5-methylmercaptobenzimidazole, under
the trade name Vanox ZMTI, each at 1.5 phr. It is important to note that some of
these systems may inhibit the peroxide curing mechanism and a higher proportion
of coagent may therefore be needed for effective vulcanisation [8].
Tackifiers may be required for combining plies of HNBR compounds with other HNBR
compounds or different polymer formulations for the construction of composites.
Coumaroneindene resins, phenolacetylene resins, aromatic hydrocarbon resins or
styrenevinyl toluene resins work well at levels up to15 phr to tackify the resulting
HNBR compound.
In some instances, for example in O-rings and seals for valves, internal lubrication
is desirable to prevent valve sticking or squeaking. Oleamides are by far the most
efficient internal lubricants in HNBR for this purpose. Commercial varieties of such
oleamides include Armoslip CP and Kenamide E.
78
Formulating Guidelines for Hydrogenated Nitrile Rubber
NBR can be blended with ethylenepropylene diene rubber (EPDM) to improve its
heat and ozone resistance, and this technique similarly works with HNBR. As might
be expected, EPDM improves low temperature flexibility and weather resistance, with
some sacrifice in resistance to swelling in hydrocarbon fluids. Generally, 1020 phr of
EPDM elastomer, with medium to high third monomer content to assist vulcanisation,
will give improvements in low temperature flexibility and weather resistance without
loss of hydrocarbon fluid resistance. The ethylene segments in HNBR resulting from
the hydrogenation process assist blending with EPDM.
6.9 Examples
In this example we know from our previous experience with NBR seals that for this
application we need a target of 8085 Shore A hardness. We also require good wear
resistance, and previous experience suggests that N550 carbon black will provide
a good balance of properties and processing performance. The seal has a rubber-
covered outer diameter (OD), which demands good high temperature compression
set resistance to prevent the seal from spinning in the bore after extended use. The
customer also requires the best low temperature performance available. The use of
plasticisers is not permissible since the finished seal cannot be allowed to shrink due
to plasticiser leaching into the power steering hydraulic fluid. With this knowledge
in hand we are able to design a starting compound as follows:
Polymer: Zetpol 4310 EP (17 mole% acrylonitrile), to give the best low
temperature flexibility, selecting the low-viscosity grade for injection moulding.
Cure system: Peroxide plus coagent system for compression set resistance: 10.0
phr ,-bis(t-butylperoxy)diisopropylbenzene peroxide with 7.50 phr N,N-m-
phenylene dimaleimide.
Carbon black: N550 (based on past experience with NBR seals in this application).
Shore A hardness H is provided by Equation (6.6) above:
The value of D for N550 carbon black is 121, so we can solve for phr knowing
79
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
that our target for H is 82. We can also use a surface or contour plot such as that
in Figure 6.6, giving 75.00 phr for N550.
Past experience with HNBR compounds has shown that 5.00 phr French process
zinc oxide and 1.0 phr stearic acid provide excellent cure activation and stability.
In this example the customer needs to replace an NBR O-ring, taking advantage of
the high temperature resistance of HNBR. The NBR O-ring is black, and the customer
needs a different colour for the HNBR replacement to be able to trace warranty
returns and avoid errors at the component service locations the instructions might
be no more than to replace the black O-ring with the light brown O-ring. Since we
are replacing a NBR compound of medium ACN a HNBR of medium ACN content
is suitable, and low temperature flexibility is not required in this application. On
the other hand, since the article is an O-ring compression set is a primary functional
80
Formulating Guidelines for Hydrogenated Nitrile Rubber
requirement. We also need to keep the cost of the compound as low as possible to
lessen the impact of the change from NBR to HNBR. We wish to select a mineral filler
capable of high loadings, and will therefore use 10 phr of TOTM plasticiser to allow
a higher filler content without seriously sacrificing high temperature compression set.
The target Shore A durometer reading for this compound is 75.
Polymer: Therban 3446 (34 mole% ACN, 4% residual unsaturation for high
crosslink density and good compression set resistance.
Heat-stable pigmentary system: 1.00 phr N990 carbon black plus 4.00 phr red
iron oxide powder.
We require 25 points Shore A from the polymer and a further 4 points Shore A from
the filler to offset the effect of the plasticiser:
Total filler loading = (25+4)/0.178 = 163 phr (Table 6.5 and Table 6.6).
Table 6.8 Possible compound for light brown O-ring for a hydraulic system
Ingredient phr Comments
Therban 3446 100
French process zinc oxide 5 Cure activator
Stearic acid 1 Cure activator, mould lubricant
1,2-dihydro-2,2,4-trimethyl quinoline 1.5 Effective antioxidant
antioxidant
N990 carbon black 1 Pigment system
Red iron oxide powder 4 Pigment system
Zeeospheres 200 163 Filler
Trioctyl trimellitate plasticiser 10 Plasticiser
,-bis(t-butylperoxy) 10.00 Efficient crosslinking, low
diisopropylbenzene peroxide compression set
N,N-m-phenylene dimaleimide 7.50 Efficient crosslinking, low
coagent compression set
81
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
These examples will give the reader some idea how to use this information to
design formulations. In the chapter following we will discuss specimen formulations
for specific applications of HNBR. We shall also discuss the value of statistical
experimental design and desirability functions for studying formulation variables
and selecting the optimal formula for the application concerned.
References
1. Zeon Chemicals LP, Semi-EV and EV Curing Systems for Zetpol 1020
(Z5.1.2), Louisville, KY, USA, 1999.
2. Zeon Chemicals LP, Study of Level and Ratio of Peroxide and Co-Agent in
Zetpol 2010 (Z5.1), Louisville, KY, USA, 1999, p.1.
3. Zeon Chemicals LP, Effects of Post Curing on Zetpol 1020 (Z5.1), Louisville,
KY, USA, 1999, p.1.
4. Zeon Chemicals LP, Zetpol HNBR Elastomers: Carbon Black Study (SA26),
Louisville, KY, USA, April 2001, p.1.
8. Zeon Chemicals LP, Protection Systems for Zetpol (Z5.5), Louisville, KY,
USA, 1999.
82
7
Examples of Hydrogenated Nitrile Rubber
Formulae for Specific Applications
7.1 Introduction
This discussion opens with an overview of statistical experimental design and the
use of the desirability function. If there was ever an area where the use of statistical
experimental design techniques was desperately needed, it is in the formulation of
speciality elastomers. The author has personally encountered a situation in which
variant 61 of an experimental formula was the one that finally went into production
after many months of trying one variable at a time. Rubber formulations are prone to
interaction between the ingredients, and a full factorial experimental design can detect
and quantify these interactions. A fictitious example will follow in which statistical
experimental design is used together with the desirability function to facilitate the
development of the process. Subsequent examples will involve recipes containing a
number of variables, and the only possible way to accommodate these effectively is
by statistical experimental design techniques together with desirability functions.
For background information the reader is encouraged to refer to two excellent books
on the subject of statistical experimental design [1, 2]. The first of these contains a
relatively brief but very adequate description of the techniques and calculations and
will give the reader a quick start in these techniques. The second book provides a
more comprehensive background to statistical experimental design.
As mentioned above, the variables in a rubber formulation may interact, and statistical
experimental design can detect and quantify these interactions, particularly when
full factorial design and response surface designs are employed. These techniques are
illustrated below using a theoretical example. Starting compounds and full factorial
design are discussed and employed in other examples in this chapter.
83
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
n n
D= (d1, d2, d3, ... dn)
d=1 , (7.1)
As an example, let us imagine that we are measuring five properties (d1 to d5) and we
have considered the desirability of each property. The properties and the individual
desirability of each have been carefully considered and agreed with the customer, with
full regard to the anticipated application of the product. Table 7.1 gives the individual
desirability of two formulae and the composite desirability values for each formula.
Note that a zero value for any individual desirability value (dn = 0) would make the
overall desirability for the formula zero, since it is a product function.
D= 5
(d1 * d2 * d3 * d4 * d5) (7.2)
84
Examples of Hydrogenated Nitrile Rubber Formulae for Specific Applications
For properties 14 (d1 to d4), Formula 2 (Equation (7.2) is suitable. However, the
desirability of property 5 (d5) is zero, rendering the composite desirability given by
Formula 2 also zero.
FKM
Temperature resistance (C)
250 H
VMQ
200 FVMQ F
AEM ACM E
150 EPDM
HNBR
D
C
100 CR NBR B
NR A
50
0
200 150 100 50 0
Class by Oil Swell
A B C D E F G H J K
85
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
Working with the customer, we have identified the individual desirabilities shown
in Table 7.2.
1
0.8
0.6
d3
0.4
0.2
0
52 47 42
A TR-10, degrees C
1
0.8
0.6
d6
0.4
0.2
0
0 20 40 60 80 100
B % Comp. set, 1,008 hrs. @ 125C in grease
86
Examples of Hydrogenated Nitrile Rubber Formulae for Specific Applications
A number of key desirabilities, d1, d2, d4 and d5, are pass/fail properties, for which
our recipe must satisfy the minimum requirement. Two other properties, d3 and d6,
are linear functions. Based on the information in Chapter 6, we have developed the
basic recipe shown in Table 7.3, and the experimental design can be constructed
around this recipe.
Table 7.3 Experimental design starting recipe for HNBR joint boot compound
87
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
The reader may be daunted at the prospect of making up and testing as many as nine
formulae. On the other hand, guessing a formula and hoping it will perform well
in all respects will almost certainly involve making up and evaluating many more
formulations than this. The experiments are illustrated graphically in Figure 7.2.
5.00
4.80
4.60
4.40
phr peroxide
4.20
4.00
3.80
3.60
3.40
3.20
3.00
0.00 10.00 20.00
It should be noted that one of the key principles is to randomise the order of the
experiments to eliminate bias, in this case in either the mixing or the testing. We
therefore submit the formulae to the mixing and testing laboratories in a random order,
using computer software to plot the results for each of the key properties involved,
as illustrated in Figures 7.3 to 7.6. All formulae are found to pass the dynamic flex
life requirement and they all give a 100% quality retention rating on accelerated
ozone exposure testing.
88
Examples of Hydrogenated Nitrile Rubber Formulae for Specific Applications
5.0
2.8 2.4
4.5
phr peroxide
4.0
3.2
3.5
2.6
3.4 3.0
3.0
0 5 10 15 20
phr EPDM
Figure 7.3 Experimental design of HNBR joint boot compound: Surface plot
showing the effect of peroxide and EDPM on weight loss at 125 C
5.0
0.5
0.5 0.5
1.5
4.5
phr peroxide
4.0
3.5
Figure 7.4 Experimental design of HNBR joint boot compound: Contour plot
showing the effect of peroxide and EDPM on grease volume at 125 C
89
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
5.0
46.5 48.0
43.5
4.5
phr peroxide
4.0
3.5
45.0
42.0
3.0
0 5 10 15 20
phr EPDM
Figure 7.5 Experimental design of HNBR joint boot compound: Results for low
temperature retraction, TR10
5.0
50
4.5
70
phr peroxide
4.0
3.5
80 60
3.0
0 5 10 15 20
phr EPDM
Figure 7.6 Experimental design of HNBR joint boot compound: Effect of peroxide
and EDPM on compression set (Cset)
90
Examples of Hydrogenated Nitrile Rubber Formulae for Specific Applications
When we analyse the nine formulae in terms of our desirability function and calculate
a composite desirability value for each formula, this gives the results summarised in
Table 7.4.
Table 7.4 Experimental results for HNBR flex boot experimental design
Factors Responses
phr EPDM phr peroxide d1 d2 d3 d4 d5 d6 D
0.00 3.00 1.00 0.00 0.05 1.00 1.00 0.15 0.00
0.00 3.00 1.00 0.00 0.11 1.00 1.00 0.14 0.00
0.00 3.00 1.00 0.00 0.08 1.00 1.00 0.13 0.00
10.00 3.00 1.00 1.00 0.35 1.00 1.00 0.32 0.69
10.00 3.00 1.00 1.00 0.30 1.00 1.00 0.32 0.68
10.00 3.00 1.00 1.00 0.32 1.00 1.00 0.31 0.68
20.00 3.00 1.00 1.00 0.55 1.00 1.00 0.45 0.79
20.00 3.00 1.00 1.00 0.56 1.00 1.00 0.44 0.79
20.00 3.00 1.00 1.00 0.61 1.00 1.00 0.45 0.81
0.00 4.00 1.00 0.00 0.12 1.00 1.00 0.25 0.00
0.00 4.00 1.00 0.00 0.15 1.00 1.00 0.24 0.00
0.00 4.00 1.00 0.00 0.16 1.00 1.00 0.24 0.00
10.00 4.00 1.00 1.00 0.46 1.00 1.00 0.40 0.75
10.00 4.00 1.00 1.00 0.50 1.00 1.00 0.40 0.76
10.00 4.00 1.00 1.00 0.50 1.00 1.00 0.40 0.77
20.00 4.00 1.00 1.00 0.72 1.00 1.00 0.52 0.85
20.00 4.00 1.00 1.00 0.72 1.00 1.00 0.52 0.85
20.00 4.00 1.00 1.00 0.75 1.00 1.00 0.52 0.86
0.00 5.00 1.00 0.00 0.20 1.00 1.00 0.40 0.00
0.00 5.00 1.00 0.00 0.25 1.00 1.00 0.40 0.00
0.00 5.00 1.00 0.00 0.21 1.00 1.00 0.39 0.00
10.00 5.00 1.00 1.00 0.58 1.00 1.00 0.50 0.81
10.00 5.00 1.00 1.00 0.58 1.00 1.00 0.49 0.81
10.00 5.00 1.00 1.00 0.60 1.00 1.00 0.50 0.82
20.00 5.00 1.00 1.00 0.81 1.00 1.00 0.61 0.89
20.00 5.00 1.00 1.00 0.81 1.00 1.00 0.60 0.89
20.00 5.00 1.00 1.00 0.85 1.00 1.00 0.61 0.90
91
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
Table 7.4 shows that the formulae with no EPDM blended with the HNBR polymer
has zero composite desirability, since these all show shrinkage on high temperature
exposure to the customers axle grease, and shrinkage in this test is defined as a zero
individual desirability. Finally, we can plot the composite desirability (D) values to
find the ideal formula from the nine under examination.
0.75
composite D
0.50
0.25 5
0.00
4
0 phr peroxide
10 3
20
phr EPDM
Figure 7.7 Surface plot showing the effect of peroxide and EDPM on composite
desirability (D) in the experimental design of a HNBR compound for a joint boot
We can see from Figure 7.7 that the most desirable formula contains at least 15 phr
EPDM polymer, but Table 7.4 indicates that it should have 20 phr EPDM polymer
and 5.0 phr of peroxide curative. Use of statistical experimental design and desirability
function calculations has now provided us with an optimal formulation over the range
studied and a numerical method of ranking the formulations based on the customers
input of key properties.
92
Examples of Hydrogenated Nitrile Rubber Formulae for Specific Applications
The following sections illustrate further examples of formulae for HNBR in specific
applications in terms of the variables studied and give suggestions for statistical
experimental design.
In this example the customer is a diesel engine manufacturer who employs circulating
water plus an ethylene glycol coolant in a small heat exchanger to lower the
temperature and extend the life of the engine lubricating oil. The O-ring seal will be
exposed to both engine oil and coolant, normally within the range 100120 C, but
with a maximum operating temperature of 125 C. In this application conventional
NBR would have insufficient durability at 125 C, EPDM have good coolant
resistance but poor resistance to engine oils, and FKM are expensive and also have
poor resistance to high temperature water-based systems. Based on these factors a
material based on HNBR is clearly the optimal choice.
Being an O-ring in a mainly static seal, its critical properties are resistance to the fluids
concerned and the application temperature, combined with excellent compression set
resistance. O-rings universally require a Shore A hardness of 75, and this will be our
target. Standard compression moulding will be used, and low unvulcanised viscosity is
therefore not a requirement. A starting recipe and recommendations for optimisation
are shown in Table 7.5. Based on the compounding parameters described earlier,
this compound will give the best overall compression set and fluid resistance. The
optimisation factors can then be combined in an experimental design with response
surface techniques for final development of the compound.
93
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
Carbon black 5075 N762 black as low level in design at Evaluate two
(VARIABLE) 75 phr, N550 black as high level in carbon black
design at 50 phr types for best
overall balance of
properties
Optimisation:
2-factor 3-level full
factorial design; 8
compounds and
experiments
94
Examples of Hydrogenated Nitrile Rubber Formulae for Specific Applications
In this case our customer is a manufacturer of oil well tools and requires a rubber
packer element with excellent resistance to crude oil, hydrogen sulfide, aggressive
amines and water. Fluorocarbon elastomers have outstanding high temperature
resistance but their inability to withstand aggressive amines rules them out.
Conventional NBR have the required oil, amine and water resistance, but cannot
operate at 125 C with excursions to 150 C. When the packer element seals the
well very high pressure differentials may be encountered, and we need to formulate
a tough material of high hardness.
The packer element has a relatively simple cylindrical shape, and compression
moulding is therefore suitable. Its critical properties are heat resistance and pressure,
as indicated above, plus resistance to crude oil, hydrogen sulfide, steam and
aggressive amines. HNBR show a smaller drop in modulus of elasticity and stiffness
at increasing temperatures, and a HNBR packer retains its pressure resistance better
than alternative materials at high temperature. A starting compound is shown in
Table 7.6, with variables listed for further study and optimisation using experimental
design, desirability functions and response surface methods.
Steam resistance.
95
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
Optimisation:
2-factor 2-level full
factorial design
with centre point;
5 compounds and
experiments
96
Examples of Hydrogenated Nitrile Rubber Formulae for Specific Applications
Critical properties are therefore heat resistance over the range 100125 C, static and
dynamic ozone resistance, and dynamic flex life. The better retention of stiffness at
high temperatures offered by HNBR compounds will ensure meshing of the belt at
high temperatures. A basic recipe for this application is shown in Table 7.7, including
compounding variables for experimental design and optimisation.
97
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
HNBR polymer with 70-90 Best overall balance Vary levels of zinc
34-36% ACN; 1-4% between heat methacrylate modifier to
residual unsaturation resistance and efficient balance properties; total
(VARIABLE) vulcanisation polymer phr = 100
HNBR modified with 30-10 Added tear and tensile Vary levels of zinc
zinc methacrylate strength methacrylate modifier to
balance properties; total
polymer phr = 100
N762 carbon black 10-20 Reinforcement without Vary level between 10 and
(VARIABLE) loss of flex resistance 20 phr to obtain best balance
of properties
Optimisation:
3-factor, 3-level
full factorial design
with centre point;
27 compounds and
experiments
98
Examples of Hydrogenated Nitrile Rubber Formulae for Specific Applications
Our customer is a manufacturer of mechanical face seals for automotive water pumps.
He has been using conventional NBR for the protective bellows around the water
pump seal but the engine compartment temperature has now risen to 125 C with
excursions up to 150 C, and NBR is unable to meet the 100,000-mile warranty.
Due to its combination of toughness, weather resistance, splash fluid resistance and
cost HNBR is a good alternative. The customer requires the bellows to be orange
in colour for the mechanics and assemblers to be able to distinguish them from the
previous black NBR version. There is some flexing of the bellows, demanding flex
fatigue resistance, but the flexing is relatively small.
The initial target hardness for the material is 65 Shore A, and other key properties
include colour, heat resistance continuously at 125 C with excursions to 150 C,
weather resistance, water and coolant resistance, low temperature flexibility for cold
starting and dynamic flex resistance. For high volume production of this fairly complex
shape injection moulding is the most suitable production process. A starting recipe is
shown in Table 7.8, with additional variables for statistical experimental design and
final compound optimisation. Organic pigments are not suitable for peroxide-cured
materials since the pigments can act as free-radical scavengers and may interfere
with the vulcanisation reaction. In addition, the oxidising effect of peroxides may
destroy the colour.
Ozone resistance.
99
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
Optimisation:
2-factor and 3-level full
factorial design;
6 compounds and
experiments
100
Examples of Hydrogenated Nitrile Rubber Formulae for Specific Applications
In this application the customer requires a high temperature seal for the output shaft
of the drive differential joint of an automobile. He has been using NBR-based shaft
seals for many years, but relocation of the exhaust and catalytic converter system
has now created an environment near the shaft seal in which the seal elastomer
lip encounters continuous temperatures in the range 100125 C; this has caused
premature hardening of the NBR seal, shortened life and warranty issues. The
extreme pressure lubricant oils used in the differential contain sulfur and phosphorus
compounds which tend to induce vulcanisation of the highly unsaturated NBR
material. The saturated nature of HNBR should offer resistance to differential gear
lubricants at the temperatures involved, and its toughness and abrasion resistance
will provide long-term life in the abrasive environment caused by dust and mud.
For the high volumes required transfer moulding is the preferred process, and in
addition the HNBR material must bond well to the metal casing of the shaft seal.
Optimal performance in this application dictates a Shore A hardness of 80. Other
critical properties include long-term resistance to gear lubricants at 125 C, moulding
performance in terms of flow and cleanliness, bonding to the steel case, and shaft seal
life in the simulated application extremes test programme. The starting recipe and
variables for experimental design and optimisation are shown in Table 7.9.
101
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
Celite 350 or Mistron 75.0 Mineral fillers show Evaluate two different
Cyprubond (VARIABLE) best rotating shaft seal fillers for best overall
performance properties
For this application the customer needs a short hose segment for power steering
hydraulic fluid. Due to restricted space, the fabric-reinforced hose needs to be based
on a compound that will resist heat, weathering and ozone on the outside of the hose
and mineral-based hydraulic fluid on the inside. The hose will be plied around the
102
Examples of Hydrogenated Nitrile Rubber Formulae for Specific Applications
braided fabric reinforcement and the plies will be formed by extrusion. Due to the
location of an exhaust header near the short section of hose the operating temperature
of the hose will be normally at 125 C, with an occasional excursions to 150 C, for
example when idling in heavy traffic. These requirements dictate the use of HNBR
to provide temperature resistance, overall toughness, ozone resistance and retention
of modulus at high temperatures. The critical properties of the compound and
hose assembly include static and dynamic ozone resistance, flex fatigue resistance,
smooth processing by extrusion, adhesion of the plies and the ability to withstand
the customers pressure pulse test on the finished hose section. A compound giving
75 Shore A hardness offers the best balance of pressure resistance and retention
of conformity. A starting recipe and variables for statistical experimental design
optimisation are given in Table 7.10.
103
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
Flex-fatigue resistance.
Extrusion processing.
The customer for this application is a paper mill with a requirement for long-running
gloss calender rolls operating at temperatures slightly above 120 C. The chemicals
employed dictate that the roll compound must have good resistance to water and
mineral-based fluids. The current polyurethane rolls are not sufficiently long-lasting
due to the combination of the chemicals and temperature. Low hysteresis is required to
prevent excessive heat build-up in the rollers which will upset the process. The desired
hardness of the roller compound is 60 Shore D. Because the rolls will be used with
brightly coloured papers a black compound is not considered acceptable. The abrasion
resistance, temperature resistance, chemical resistance and retention of modulus at
high temperatures make zinc methacrylate-modified HNBR a suitable choice for this
application. The rolls will be built from smooth calendered stock moulded on steel
cores and they will be ground to size after moulding. Critical properties for the roll
compound are low hysteresis, temperature resistance to 120 C, resistance to smooth
calendering and to the application chemicals. A starting compound for experimental
design optimisation is shown in Table 7.11.
Dynamic hysteresis.
104
Examples of Hydrogenated Nitrile Rubber Formulae for Specific Applications
Table 7.11 Low hysteresis paper mill roller compound experimental design
Ingredient phr Logic/comments Experimental design
optimisation
HNBR polymer with 3436% 50.0 Basic HNBR for Level 1 evaluation of
ACN; 5% unsaturation chemical resistance HNBR
Optimisation:
3-factor, 2,2- and 2-level full
factorial design; 8 compounds
and experiments
This chapter has reviewed the formulating principles for HNBR polymers covering a
range of specific examples. It has also introduced the reader to statistical experimental
design and desirability functions for optimising HNBR formulae to give the best overall
balance of properties. Using these techniques it is possible to formulate compounds
to give the best overall capability for meeting customer requirements.
105
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
References
2. G.E.P. Box, J.S. Hunter and W.G. Hunter, Statistics for experimenters:
Design, discovery and innovation, John Wiley and Sons, Hoboken, NJ, USA,
2005.
106
8
Solving Hydrogenated Nitrile Rubber
Processing Issues
8.1 Introduction
In this final chapter examples are discussed of typical production processing issues
that are likely to be encountered when processing speciality elastomers such as HNBR.
When a problem arises there is an unfortunate tendency to focus on a single solution,
and this is usually to change the compound. It must however be borne in mind that
even a minor change in the compound may require its complete requalification, and
particularly its effect on the nature of the parts produced. In any case simply changing
the compound without examining the other process variables is unlikely to lead to
a permanent solution, and the examples described below and the solutions offered
will provide some pointers to solving the issues that may arise.
At the beginning of the injection moulding process the first step is usually to preheat
the unvulcanised rubber and feed it into the injection mechanism, which can be either
a reciprocating screw or a screw and ram. Uniform preheating is achieved by use of
a circulating liquid. A tendency the author has observed is the general belief that a
lower injection feed temperature will result in less premature scorching and a reduced
tendency for non-fills. However, if we bear in mind the nature of elastomeric flow we
107
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
can see that this is unlikely to happen. From the discussion in Chapter 3, the typical
effect of temperature on viscosity is shown in Figure 8.1.
Temperature effect
1000000
100000
Viscosity (poise)
100 C
10000
125 C
1000
150 C
100
1 10 100 1000
Shear rate (s1)
If the injection process is operating at a shear rate of 1,000 s1 and the injection feed
mechanism is controlled at 100 C, the resulting viscosity of the material under these
conditions will be roughly 1,000 poise. If we raise the injection feed temperature
to 125 C the viscosity drops to 200 poise, or one-fifth of the viscosity at 100 C.
We obviously need to keep the injection feed mechanism temperature below the
temperature at which the material will begin to vulcanise prematurely and scorch
before it completely fills the mould cavity. A higher temperature in the feed zone
of the injection moulding machine may solve the issue, or it could be just one of a
number of factors to be considered in making the process more robust and avoiding
sporadic flow and non-fills.
After feeding and preheating the rubber prior to injection, the next step is its injection
into the mould. A possible factor at this point is the shear rate during injection. In
production processes time is always a pressing issue and the tendency is always to
inject the rubber as rapidly as possible. By elevating the injection pressure the rubber
108
Solving Hydrogenated Nitrile Rubber Processing Issues
enters the mould more quickly and in turn the shear rate is increased. Again, from the
discussions in Chapter 3, and noting that the shear will cause the HNBR compound to
become hot, too high a shear rate may be a major factor in the flow and non-fill issues
observed. A reduction in injection pressure will certainly reduce the flow rate of the
rubber and increase injection time, but this may be a small price to pay in preventing
the non-fill and flow issues caused by excessive heat build-up during injection.
350%
% change in heat build-up
300%
250%
200%
150%
100%
50%
0%
50%
100%
100% 50% 0% 50% 100% 150%
% change in parameter
109
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
Using Equation (8.1) we can establish the following proportionalities to estimate the
effect of viscosity and shear rate on heat build-up, as shown in Figure 8.2:
DT h, and DT g2 (8.2)
Reducing the viscosity, , of the unvulcanised rubber would certainly help, and we
may be able to accomplish this by increasing the temperature of the rubber feed, as
discussed in Section 8.2.1, but it can be seen from Equation (8.2) that increasing the
shear rate has a much greater effect on temperature build-up during injection. In the
particular case being considered, as a factor in solving the problem we may decide to
reduce the effective shear rate during injection by reducing the injection pressure from
150 to 100 bar. Again, this is just one factor involved in solving the overall problem,
and the reader may begin to suspect that an experimental problem-solving design is
on the horizon as we attempt to resolve the flow and non-fill issues:
L = length of capillary.
= (DP.R4.p) / (4Q.2L)
110
Solving Hydrogenated Nitrile Rubber Processing Issues
This issue may be a factor, particularly if the sporadic flow and non-fill issues are
isolated to the same cavities or decks of the injection mould. If we look at Equation
(8.5) and plot the effect of changing viscosity (happ) of the material against the radius
of the runner or sprue system, it is clear that the latter has a dramatic effect on the
volume flowing through, since the flow volume, Q, is proportional to the fourth
power of the capillary radius. These effects are illustrated in Figure 8.3.
120%
Net % change in flow rate
100%
80%
60%
40%
20%
0%
20%
40%
60%
20% 10% 0% 10% 20% 30%
% Change in factor
To improve the overall flow and to eradicate the sporadic flow and non-fill issues,
we could of course change the compound to reduce its viscosity, but it would seem
more sensible to make small increases in the size of the sprues and runners in the
injection mould. Again, the sprue and runner size may be just one of the factors to
be studied in an experimental design to solve the sporadic flow and non-fill problem.
111
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
8.2.4 Solving the Injection Flow Problem using the Above Factors
From Chapter 7 the reader may already have concluded that a statistical experimental
design would be appropriate using the factors discussed above. Listed in Table 8.1
and illustrated in Figure 8.4 is an experimental programme designed to resolve the
flow and non-fill issue.
Table 8.1 Experimental design for sporadic flow and non-fill injection
moulding issues
Mould Injection Injection pressure Responses Comments
temperature time (s) holding time (s) (dimensions)
(C) parts/100 with
non-fills
180 20 5 1, +1, 1
180 20 15 1, +1, +1
180 10 5 1, 1, 1
180 10 15 1, 1, +1
190 20 5 +1, +1, 1
190 20 15 +1, +1, +1
190 10 5 +1, 1, 1
190 10 15 +1, 1, +1
185 15 10 Centre point
Again, the order would need to be randomised and a number of runs are required to
determine the significance of the response. Multiple runs would for example help us
determine whether the difference between, say, 4 parts/100 and 6 parts/100 would
be statistically different in terms of non-fills. Obviously, we would require one of
these combinations to consistently produce zero non-fills in successive multiple runs.
112
Solving Hydrogenated Nitrile Rubber Processing Issues
6.0
Runner size, mm
150
5.0
100
100 125
Injection feed temperature, C
Figure 8.4 Injection moulding flow: Graphical depiction of experimental design
This problem arises in many production moulding operations, especially in the case of
parts with linear or diametric dimensions greater than about 7.5 cm. The author has
experienced this on many occasions, the initial reaction being that something must
be wrong with a particular batch or mix of rubber compound. If we look closely at
the physics involved, however, the problem could have little or nothing to do with
batch-to-batch variation in the rubber.
The first step is to examine the mould or platen temperature, since this is by far the
most critical factor in determining the dimensions of the finished part. Certainly,
production personnel and the tooling source will claim that it must be something
that has happened to the compound that is causing the change but temperature is
still the most likely factor. For example, a typical coefficient of linear expansion for
an HNBR compound is 0.00015 mm/mm/C. If room temperature is 25 C and the
moulding temperature is 180 C, the total shrinkage after moulding would be 0.023
mm/mm. If the mould were cut allowing for 0.023 mm/mm shrinkage at a moulding
113
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
This effect is amplified in the case of lightly filled elastomers with a high coefficient
of linear expansion, such as a HNBR elastomer of lower target hardness. A carbon
black-filled HNBR compound of this type might have a coefficient of linear expansion
of 0.00025 mm/mm/C. If the mould was designed for a vulcanisation temperature
of 180 C but the temperature was actually 190 C, this would reduce a nominally
76.20 mm dimension to 76.02 mm.
Step 2 should be to examine the flow conditions and any shifts in moulding. As
mentioned in Chapter 3 and confirmed by Figure 8.4, shear rate has a major impact
on the elastic recovery of the material. As shown in Figure 8.5, elastic recovery can
be pronounced after demoulding, when the part acts in the same way as stretched
chewing gum and retracts in the direction of flow. In addition, from Figure 8.2, the
effect of increasing shear rate on heat build-up in the stock can be pronounced. A
decrease in injection time from 20 s to 10 s would double the shear rate, and from
Figure 8.2 this would in turn cause a three-fold increase in the build-up of heat in
the compound during flow, ignoring any reduction in viscosity at higher shear rate.
In effect this increases the moulding temperature of the material without affecting
the mould or platen temperature and the moulding temperature is of course the
biggest factor in determining the dimensions of the finished part.
114
Solving Hydrogenated Nitrile Rubber Processing Issues
45
40
35
% Die swell
30
25
20
15
10
5
0
1 10 100 1000
Shear rate (s1)
Figure 8.5 Elastic recovery (% die swell) as a function of shear rate
Here again a statistical experimental design including the above factors will be helpful
in solving the problem. Table 8.2 and Figure 8.6 summarise an experimental design
appropriate for addressing the dimensional issue.
115
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
15
20
5
10
180 190
Mould temperature, C
Again, the order of the experiments would need to be randomised and multiple runs
required to determine the relative significance of the factors on the dimensions of
the moulded parts.
In this example, the product was a short non-reinforced HNBR hose. The hose was
manufactured by extruding the tube, followed by cutting to the proper length and
vulcanisation in an autoclave. A challenge in the manufacture of a short non-reinforced
HNBR hose is variation in wall thickness during extrusion. As part of a variability
reduction programme a team was formed to identify the variables affecting the
thickness of the extruded tube, comprising representatives of the technical disciplines
as well as production staff. Based on mutually exclusive and collectively exhaustive
(MECE) discussions, the following variables were identified for further investigation:
116
Solving Hydrogenated Nitrile Rubber Processing Issues
Low level: 80 C.
Extruder speed:
A full factorial design would have entailed 64 experimental runs, which was clearly
prohibitive. Initially we needed to determine which of the variables were the most
significant, and interactions between the variables would be ignored. If we discovered
two or three significant variables we could then go back and study these using a full
factorial experimental design to resolve interactions. In this case a PlackettBurman
screening design would be appropriate, as illustrated in Table 8.3:
117
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
As mentioned earlier, this is a screening design and interactions between the variables
are ignored, but the design will determine which variables are significant, and the next
step is a full factorial experimental design to identify and resolve any interactions.
118
Solving Hydrogenated Nitrile Rubber Processing Issues
this action, and focusing on variables that might possibly be contributing to the
dimensional variation, including the following:
Process variable: mould closure. Due to the platen and mould being marginally
out of parallel and flatness, the mould could remain open by as much as
0.05 mm.
With six possible factors, even a fractional factorial design would obviously be
prohibitive in terms of time and cost. To determine which of the factors might warrant
further study an 8-run PlackettBurman screening design was selected. As mentioned
earlier, fractional factorials and screening designs tend to confound and obscure
interactions between factors but can indicate which of these deserve further study. In
the present case it was believed that PlackettBurman screening design would identify
the key factors for more detailed study using full factorial experimental design.
The rubber compound was carefully weighed and mixed to the extremes of the
compound variable tolerances and the raw rubber preforms were segregated by weight
and sorted to provide the extremes of performance weight. To avoid interrupting
production, moulds were constructed in the prototype laboratory. Overall mould
closure was controlled by the use of shims, zero opening requiring no shims and 0.05
mm opening 0.05 mm shims. To provide better resolution of the significant factors
five replicate runs of each experimental design were carried out.
The 8-run PlackettBurman experimental design, with results and analysis, is shown
in Table 8.4 and Figure 8.7. The values given are ID measurements in mm.
119
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
120
Solving Hydrogenated Nitrile Rubber Processing Issues
4.5
4.0
3.5
3.0
2.5
2.0
1.5
1.0
0.5
0.0
e
re
re
ur
ve
ve
gh
tu
tu
os
le
le
ei
ra
ra
cl
w
k
e
pe
pe
iv
ac
ld
at
m
m
bl
ou
or
ur
te
te
n
ef
M
C
bo
ld
e
Pr
ur
ou
ar
-c
C
M
st
Po
Absolute factor effects MSFE
The bold line in Figure 8.7 is the minimum significant factor effect (MSFE) based
on a 95% confidence limit. The only two significant factors in this screening design
were the surface temperature of the mould and its actual degree of closure, none of
the other factors having a significant influence on the ID of the moulded O-rings.
Mould temperature and closure were therefore selected for more detailed study using
full factorial experimental design.
The outcome was the provision of multi-zone platen heating to narrow the temperature
variation across the mould, and the introduction of mould and platen machining and
maintenance schedules to reduce variation in mould closure. Had one variable at a
time been studied this would have taken considerably longer and other significant
factors might have been missed. Following capital investment to reduce variations
in mould surface temperature and mould closure, variation in ID of the moulded
O-rings was virtually eliminated.
121
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
8.5 Conclusions
Techniques have been demonstrated which will help in resolving production issues. In
cases where multiple variables are involved, emphasis has been placed on statistical
experimental design techniques in order to determine which are significant. This
is intended to guard against the knee-jerk response that a change in formulation
is the way to solve a production processing issue, and also attempting to examine
the variables one at a time and missing the possibility of interaction between them.
Statistical experimental design is a technique that can help to reduce variation in
manufacture. Unfortunately many process and product improvement projects do not
in practice employ these techniques properly, or fail to plan the experimental design
effectively. The involvement of a wide range of personnel and disciplines is critical
in planning stage and developing proper experimental design.
122
A
bbreviations
ACN Acrylonitrile
CFC Chlorofluorocarbon
CR Polychloroprene
D Composite desirability
HFC Hydrofluorocarbon
123
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
ID Internal diameter
NR Natural rubber
OD Outer diameter
Shear rate
124
Abbreviations
P Pressure drop
Viscosity
Shear stress
125
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
126
I
ndex
2-Mercaptobenzothiazole 64
100% Modulus 64, 66, 69, 72, 75
200% Modulus 69
,-bis(t-butylperoxy)diisopropylbenzene 65-66, 77, 79-81
A
ACM 17-20, 46, 49-51, 53, 56, 58, 85
ACN 1, 3, 5-11, 13-14, 16, 18-19, 43, 45, 47-53, 63, 80-81, 94, 96, 98, 102-103,
105
Acrylonitrile 1, 3, 9, 11, 33, 79, 100
Activator 80-81, 87, 94, 96, 98, 100, 102
AEM 17-20, 46, 49-50, 53, 56, 58, 85
Aerosil 300 74
Air conditioning 57, 60
ASTM D2000 17, 43-46, 55-56, 85
Automatic transmission 17-18, 49-50, 58
B
Butadiene 1-12, 14, 16, 18, 20, 22, 24, 26, 28-30, 32, 34, 36, 38, 40, 42, 44, 46,
48, 50, 52, 54, 56, 58, 60, 62, 64, 66, 68, 70, 72, 74, 76, 78, 80, 82, 84, 86, 88,
90, 92, 94, 96, 98, 100, 102, 104, 106, 108, 110, 112, 114, 116, 118, 120, 122
C
Cab-O-Sil M-7D 74
Cab-O-Sil TS-720 74
Capillary 37-38
length 110
radius 111
reheometer 37
Carboxylated 16
Celite 350 74, 102
Coagent 63, 65-68, 78-81, 94, 96
127
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
128
Index
H
Heat build-up 27-28, 104, 109-110, 114
HiSil 233 74-75, 100
HiSil 532EP 74, 100
HNBR 1, 7-14, 16-21, 43, 45-60, 63, 69-70, 73, 75-76, 78-82, 85, 87-105, 107,
109, 113-114, 116
Hydrogenation 1, 7-9, 11, 45-46, 48, 55, 57, 63, 79
Hysteresis 34-35, 104-105
I
Imsil A-8 74
Initiator 1-3
Injection moulding 13, 24-25, 27, 34, 36, 79, 87, 99-100, 107-108, 112-113, 115
Isoprene 11-12
K
Knit lines 36, 107
L
Lanxess 13-16
M
Manual transmission 59
Mayo-Lewis copolymer equation 3
Minimum significant factor effect (MSFE) 121
Mistron Cyprubond 105
Mn 30
Modulus 26, 32-34, 64-67, 69, 71-77, 95, 103-104
Mole % ACN 49
Mole fraction of monomer 5
Molecular weight 2, 6, 29-31
Molecular weight distribution 29-31
Mooney viscosity 13, 31, 72, 76
Mould flow 36-37, 87
Mz 30
N
N110 32
N330 65, 96
N550 32, 64, 77, 79-80, 94, 96, 102-103
129
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
130
Index
131
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
abcd
abcd
abcd
a b c d
abcd
abcd
132
Published by Smithers Rapra Technology Ltd, 2012
High performance engineering plastics are used in an increasingly wide range of applications
and environments. Their growth in importance is a response to the ever-increasing demand
for more reliable, high performance components.
This book is the product of the authors first-hand experience and understanding of high
performance engineering plastics; specifically hydrogenated nitrile rubbers, which are
progressively supplanting the simpler non-hydrogenated varieties thanks to their superior
properties. A practical overview of their key properties and formulation principles is
provided, based on the authors own background and practical experience. Each chapter
contains information on their product forms, properties, processing and applications, with
the emphasis on materials and concepts shown to work in practice.
Readers will learn why hydrogenated nitrile rubbers are now the first choice for a range of
demanding applications, how their characteristics arise and how their properties can be
adapted. Many readers will welcome the practical nature of the examples given and the
way in which problems can be resolved, for example by employing statistical experimental
design. Not only is this concept valuable in overcoming production issues in a logical and
cost-effective manner, it is also of help in communicating with raw material suppliers and
those equipment manufacturers who have become dependent on nitrile rubbers.