Group Presentation English Advancedy - Robert Keller (Smithers Rapra Technology, 2012)

Practical Guide to
Hydrogenated Nitrile
Butadiene Rubber
Technology
Robert Keller
A Smithers Group Company
Shawbury, Shrewsbury, Shropshire, SY4 4NR, United Kingdom

Telephone: +44 (0)1939 250383 Fax: +44 (0)1939 251118
http://www.polymer-books.com
First Published in 2012 by
Smithers Rapra Technology Ltd

Shawbury, Shrewsbury, Shropshire, SY4 4NR, UK
2012, Smithers Rapra Technology Ltd
All rights reserved. Except as permitted under current legislation no part

of this publication may be photocopied, reproduced or distributed in any
form or by any means or stored in a database or retrieval system, without
the prior permission from the copyright holder.
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.
ISBN: 978-1-84735-521-8 (softback)

978-1-84735-522-5 (ebook)
Typeset by Argil Services

C
ontents
1 Overview of the Chemistry and Manufacture of

Hydrogenated Nitrile Butadiene Rubber Polymers....................... 1
1.1 Introduction ...................................................................... 1
1.2 The Starting Polymer: Nitrile Butadiene Rubber................. 1
1.3 Hydrogenation of Nitrile Butadiene Rubber to Produce
the Hydrogenated Polymer................................................. 8
1.4 Summary.......................................................................... 12
2 Types of Hydrogenated Nitrile Rubber Polymers Available........ 13

2.1 Introduction..................................................................... 13
2.2 Summary of Grades Available........................................... 13
2.3 HNBR Grades and Technology......................................... 17
2.4 Summary.......................................................................... 20
3 The Flow and Processing Behaviour of Hydrogenated

Nitrile Rubber............................................................................ 21
3.1 Introduction..................................................................... 21
3.2 The Basics of Viscoelastic Flow in Rubber........................ 21
3.2.1 Viscosity................................................................ 21
3.2.2 Elasticity............................................................... 24
3.3 Effect of Process Variables................................................ 26
3.3.1 Effect of Temperature on Viscosity........................ 26
3.3.2 Heat Build-up During Flow................................... 27
iii
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
3.4 Effect of Compounding Variables on Viscosity,

Elasticity and Flow........................................................... 29
3.4.1 Effect of Polymer Molecular Weight...................... 29
3.4.2 Effect of Fillers...................................................... 32
3.4.3 Effect of Plasticisers............................................... 35
3.4.4 Effect of Process Aids............................................ 36
3.5 Flow through Sprues and Runners: Mould and
Machine Parameters Influencing Elastomer Flow.............. 37
4 Properties of Hydrogenated Nitrile Rubber and

Comparison with Other Elastomers............................................ 43
4.1 Introduction .................................................................... 43
4.2 Relationship between Hydrogenated Nitrile Rubbers
and Other Speciality Elastomers....................................... 43
4.2 Specific Comparison of HNBR and NBR.......................... 46
4.3 Comparison of HNBR with HNB and Other
Speciality Elastomers........................................................ 49
5 Typical Applications of Hydrogenated Nitrile Rubber................ 55

5.1 A Brief Recap................................................................... 55
5.2 Abrasion-resistant Belts and Conveyor Components........ 56
5.3 Flexible Boots for Power Transmission Joints................... 57
5.4 Static Seals for Power Transmission Fluids....................... 58
5.5 Dynamic Fluid Seals for Power Transmission................... 59
5.6 Hydraulic Fluid and Lubricant Hoses............................... 59
5.7 Seals for High-volume Air Conditioning . ........................ 60
5.8 Seals used in Petroleum Exploration and Drilling............. 60
5.9 Other applications............................................................ 60
6 Formulating Guidelines for Hydrogenated Nitrile Rubber.......... 63

6.1 Introduction..................................................................... 63
iv
Contents
6.2 Vulcanisation Systems....................................................... 63

6.3 Carbon Black Fillers......................................................... 70
6.4 Non-black Fillers.............................................................. 73
6.5 Plasticisers........................................................................ 75
6.6 Antioxidants and Antiozonants........................................ 78
6.7 Other Ingredients . ........................................................... 78
6.8 Blends with EthylenePropylene Diene Rubber................. 79
6.9 Examples.......................................................................... 79
6.9.1 Black HNBR Compound for a Reciprocating
Lip Seal on a Power Steering Rack End................. 79
6.9.2 Light Brown HNBR Compound for a
Hydraulic System O-ring....................................... 80
7 Examples of Hydrogenated Nitrile Rubber Formulae for

Specific Applications................................................................... 83
7.1 Introduction..................................................................... 83
7.2 Statistical Experimental Design and
Desirability Functions....................................................... 83
7.3 Theoretical Example: High-temperature HNBR
Joint Boot......................................................................... 85
7.4 Compound Example: High Temperature Oil Cooler
O-ring Seal....................................................................... 93
7.5 Compound Example: Oil Field High-pressure
Well Packer....................................................................... 95
7.6 Compound Example: High temperature Long-life
Automotive Serpentine Belt.............................................. 97
7.7 Compound Example: Orange Water Pump
Mechanical Seal Protective Bellows.................................. 99
7.8 Compound Example: High-temperature
Differential Shaft Seal..................................................... 101
7.9 Compound Example: Short Steering System Hose.......... 102
v
7.10 Compound Example: Chemically Resistant,

Low Hysteresis Roller for Paper Mills............................ 104
7.11 Concluding Comments................................................... 105
8 Solving Hydrogenated Nitrile Rubber Processing Issues........... 107

8.1 Introduction................................................................... 107
8.2 Example 1: Poor Flow, Knit Lines and Non-fills in
Injection Moulding......................................................... 107
8.2.1 Possible Factor: Temperature of the
Unvulcanised Rubber During Injection................ 107
8.2.2 Possible Factor: Excessive Shear During
Injection.............................................................. 108
8.2.3 Possible Factor: Sprue and Runner Design.......... 111
8.2.4 Solving the Injection Flow Problem using
the Above Factors............................................... 112
8.3 Example 2: Sporadic Dimensional Shift in
Injection Moulded Parts................................................. 113
8.3.1 Possible Factor: Mould Temperature................... 113
8.3.2 Possible Factor: Injection Pressures and
Holding Pressures and Times............................... 114
8.3.3 Solving the Injection Moulding Dimensional
Problem using the Above Factors........................ 115
8.4 Example 3: Variation in Hose Tube Thickness
during Extrusion............................................................. 116
8.5 Example 4: Variation in Dimensions of High-volume
Compression Moulded O-ring........................................ 118
8.5 Conclusions.................................................................... 122
Abbreviations..................................................................................... 123
Index ................................................................................................ 127
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:
The manufacture and chemistry of HNBR elastomers.
The commercial grades currently available.
How HNBR fits into the current world of speciality elastomers.
The basics of compounding and formulating HNBR.
Examples of compounding for specific applications.
Its rheological properties and how these can be applied.
Problem solving techniques using specific examples.
Much of the information will be found applicable not only to HNBR but also to
other elastomers.
vii
viii
1
Overview of the Chemistry and
Manufacture of Hydrogenated Nitrile
Butadiene Rubber Polymers
1.1 Introduction
This introductory chapter reviews the manufacture of hydrogenated nitrile butadiene

rubber (HNBR) polymers. It is not intended to cover every facet of the production
process, but the information provided will give the reader a background to the
manufacture of these industrially significant rubber polymers. It is also not the
intention to cover the proprietary specifics of polymer manufacturers.
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.
1.2 The Starting Polymer: Nitrile Butadiene Rubber
Nitrile rubber, ASTM D1418 designation NBR, is a copolymer of ACN and

1,3-butadiene. It is formed by free-radical polymerisation, shown in Figure 1.1.
The polymerisation is a typical emulsion polymerisation, which has the following

characteristics:
The monomers are usually contained in emulsified droplets in the continuous

phase. In the polymerisation of NBR the continuous phase is water, and the
butadiene and ACN monomers and resulting polymers are contained in the
emulsified droplets in the oil or emulsified phase.
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
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.
Initiator = free radical

Only partially soluble in oil phase
R*
Monomers in emulsified oil phase

Polymer forms here
Typical size of emulsified phase = 100 nm
CH 2=CH---CH=CH 2 + CH 2=CH >>>>> ---(CH2 ----CH=CH----CH2)x-(CH2----CH)y -(CH 2 =CH ) Z-

| | |
CN CH=CH 2 CN
Typical composition of resulting NBR polymer =

X = 84-50%, Y = 5% of butadiene incorporated, Z = 16-50%
Figure 1.1 Emulsion free-radical polymerisation of NBR
2
Overview of the Chemistry and Manufacture of Hydrogenated Nitrile
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
Final polymer bales

Drying and dewatering
Figure 1.2 Schematic showing NBR polymerisation
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:
M1* + M1 ----- M1M1*, reaction rate = k11 (1.1)
3
The relative reactivity ratios of the monomers can then be defined as follows:
r1 = k11/k12 = reactivity ratio for monomer 1 (1.5)
r2 = k22/k21 = reactivity ratio for monomer 2 (1.6)
The molar concentration of monomers at any instant during the polymerisation is

given by Equation (1.7):
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.
The limiting cases in Equation (1.1) are as follows:
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.).
r1 = r2 = approximately 1. The reactive chain ends, whether monomer 1 or

monomer 2, have no preference for either monomer 1 or monomer 2 and the
result is a random copolymer (e.g., 1-2-1-1-2-2-2-1-2-1-1-1-1-2-2-1-2-2-2-2-2-
1-2-2-2.).
r1 = r2 = approaching 0. If monomer 1 forms the reactive chain end it has little

likelihood of reacting with further monomer 1, and similarly, if monomer 2 is
the reactive chain end it has little likelihood of reacting with monomer 2. In this
case, a perfectly alternating copolymer will result (1-2-1-2-1-2-1-2.).
4
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:
F1 = instantaneous mole fraction of monomer 1 in the instantaneous polymer

formed,
f1 = mole fraction of monomer 1 in the instantaneous monomer feed, and
f2 = mole fraction of monomer 2 in the instantaneous monomer feed.
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
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.
% ACN in polymer, 75/25

Butadiene/ACN Charge mixture
40
30
% ACN
% ACN incremental
20
polymer
10 % ACN in polymer mix
0
0 20 40 60 80 100
% Conversion
Figure 1.3 Example of polymer ACN versus conversion
In this case, the incremental (instantaneous) polymer composition early in the

polymerisation is relatively constant at 3035 mole% ACN. At higher conversions
later in the reaction the incremental polymer becomes richer in butadiene, with a
lower mole% of ACN. It is obvious that very low conversion values are impractical,
6
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):
%ACN in instantaneous polymer
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
Figure 1.4 ACN content of polymer versus 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.
This understanding of the NBR polymer formed by emulsion polymerisation is

important in appreciating the properties of the HNBR polymer formed by the catalytic
hydrogenation of the NBR polymer, and the challenges presented by the process. The
key points in the emulsion polymerisation yielding NBR polymers are as follows:
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
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.
1.3 Hydrogenation of Nitrile Butadiene Rubber to Produce the

Hydrogenated 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
Bales of HNBR polymer

Drying
Figure 1.5 Schematic process for the catalytic hydrogenation of NBR to form
HNBR
A catalyst which facilitates hydrogenation is normally employed. A typical

combination is monochlorobenzene as solvent and OsHCl(CO)(O2)(PCy3)2 as the
8
catalyst [3], in which Cy denotes a cyclohexyl residue. The hydrogenation reaction

is illustrated in Figure 1.6.
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
Figure 1.6 Catalytic hydrogenation of NBR
In effect, the resulting HNBR is a tetrapolymer of ACN, ethylene and propylene,

containing a small amount of residual unsaturation which allows rapid vulcanisation
by means either of sulfur or a peroxide cure.
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.
These segments of ethylenepropylene polymer of high ethylene content create major

challenges in the manufacture of HNBR polymers of low ACN content, required
9
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.
Tg (C) of low vinyl EPDM

0
40 45 50 55 60 65 70 75 80
10
20
30 Tg (C)
Tg
linear Tg (C)
40
y = 1.2688 x 102.79
50 R2 = 0.9679
60
Mole% ethylene
Figure 1.7 Tg of ethylenepropylene copolymers versus ethylene content
As can be seen from Figure 1.7, low temperature flexibility as measured by Tg

deteriorates rapidly at higher ethylene content due to the tendency of the ethylene
segments to crystallise, in effect forming low-density polyethylene segments. In the
case of a typical cold-polymerised NBR base polymer, the ethylenepropylene segments
in the resulting HNBR polymer will comprise roughly 95 mole% ethylene, with very
poor low temperature flexibility, using Figure 1.7 to give an estimated value of Tg.
Put simply, if Figure 1.7 can be used to estimate the Tg of a HNBR of low molar
ACN content, and assuming that the ethylene content of the short ethylenepropylene
segments to be 95 mole% (typical for a cold-polymerised NBR starting polymer),
then an estimate of the Tg of the resulting HNBR polymer would be +18 C.
10
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
Metal catalysts, high pressure

+ (2x+2y) H2 (gas)
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
Figure 1.8 Hydrogenation reaction of ACN/1,3-butadiene/isoprene terpolymer
The incorporation, for example, of a 1/1 ratio of 1,3-butadiene/isoprene into the

starting polymer would result in a roughly 73/27 molar ratio of ethylene/propylene
in the final HNBR polymer (1 mole of 1,3-butadiene results in 0.95 moles of ethylene
and 0.05 moles of propylene in the final HNBR polymer, and 1 mole of isoprene
results in 0.5 moles of ethylene and 0.5 moles of propylene):
11
1 mole of 1,3-butadiene = 0.95 moles ethylene + 0.05 moles propylene in HNBR

polymer.
1 mole of isoprene = 0.5 moles ethylene + 0.5 moles propylene in HNBR polymer.
Total moles of ethylene = 0.95 + 0.5 = 1.45.
Total moles of propylene = 0.05 + 0.5 = 0.55.
Ethylene/propylene ratio in the final HNBR polymer = 1.45/(1.45 + 0.55) = 0.73.
Ethylene/propylene ratio in the final HNBR polymer with no isoprene in the

starting polymer = 0.95.
The reduced ethylene content in the ethylenepropylene segment gives significantly

improved low temperature flexibility.
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.
2. W.H. Embree, J.M. Mitchell and H.L. Williams, Canadian Journal of

Chemistry, 1951, 29, 3, 253.
3. C. Mouli, R. Madhuranthakam, Q. Pan and G.L. Rempel, AIChE Journal,

2009, 55, 11, 2934.
4. H. Bender, R. Casper, H.R. Winkelbach, H.C. Strauch, P. Nguyen, S.X. Guo

and J. Gamlin, inventors; Bayer AG, assignee, US Patent 7,091,284, 2006.
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.
2.2 Summary of Grades Available
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
Table 2.1 HNBR grades commercially available, December 2010

Grade Manufacturer Mole% Mooney Residual %
ACN viscosity unsaturation
(nominal) ML(1+4) (approximate)
at 100C
(nominal)
Zetpol 4300EP Zeon 17 30 0.5
Zetpol 4300 Zeon 17 75 0.5
Zetpol 4310EP Zeon 17 30 5
Zetpol 4310 Zeon 17 62 5
Therban AT LT
2004 VP Lanxess 21 39 0.9
Therban LT
2007 Lanxess 21 74 0.9
Therban LT
2057 Lanxess 21 67 5.5
Therban LT
2157 Lanxess 21 70 5.5
Therban AP A
3404 Lanxess 34 39 0.9
Therban 3406 Lanxess 34 63 0.9
Therban AT C
3443 VP Lanxess 34 39 4
Therban 3446 Lanxess 34 61 4

Therban VP KA
8837 Lanxess 34 55 18
Zetpol 2000EP Zeon 36 30 0.5
Zetpol 2000L Zeon 36 65 0.5
Zetpol 2000 Zeon 36 85 0.5
14
Types of Hydrogenated Nitrile Rubber Polymers Available

Therban 3627 Lanxess 36 87 2
Zetpol2010L Zeon 36 58 4
Zetpol 2010H Zeon 36 135 4
Therban AT A
3904 VP Lanxess 39 39 0.9
Therban AT
4304 VP Lanxess 43 39 0.9
Therban AT
4364 VP Lanxess 43 39 5.5
Zetpol 1000L Zeon 44 65 2
Zetpol 1020L Zeon 44 57 9
Therban AT
5005 VP Lanxess 49 55 0.9
Therban 5008
VP Lanxess 49 80 0.9
Therban AT
5065 VP Lanxess 49 55 6
Therban is a registered trademark of Lanxess
Zetpol is a registered trademark of Zeon Chemicals
15
In addition to these grades, speciality polymers are available for specific applications
and end-products. These are summarised in Table 2.2.
Table 2.2 HNBR speciality grades available

Grade Manufacturer Mole% Mooney Residual % Modification
ACN viscosity unsaturation
(nominal) ML(1+4) (approximate)
at 100 C
(nominal)
Zeoforte Zeon 36 95 9 Zinc
ZSC methacrylate
2295CX
ZSC methacrylate
2295L
ZSC methacrylate
2385
Therban Lanxess 33 77 3.5 Carboxylated
XT VP (XHNBR)
KA 8889
Therban Lanxess 34 22 5.5 Acrylate
VP KA
8796
Therban is a registered trademark of Lanxess
Zetpol and Zeoforte are registered trademarks of Zeon Chemicals
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
2.3 HNBR Grades and Technology
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.
ASTM D2000 Classification

% Swell in ASTM IRM903 Oil
300
Temperature resistance (C)

FKM 250
VMQ FVMQ 200

EPDM AEM ACM
150
HNBR
CR 100
NBR
NR
50
0
200 150 100 50 0
Class by Oil Swell
Figure 2.1 Graphical illustration of speciality elastomers, showing the position of

HNBR within the ASTM D2000 system
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
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
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
0
10
15
20
25
30
HNBR HNBR
Material
Figure 2.3 Comparison of resistance to automatic transmission fluid (ATF) of

HNBR, ACM and AEM elastomers
18
The relative abrasion resistance of HNBR materials compared to other oil-resistant

elastomers, such as NBR, epichlorohydrinethylene oxide copolymers (ECO),
fluorocarbons (FKM) and ACM, is shown in Figure 2.4 [3]. Traditional NBR materials
are given a relative rating of 100.
Akron-type abrasion resistance rating
200
Relative resistance to
150
abrasion
100
50
0
45% ACN NBR ECO FKM ACM
HNBR
Material
Figure 2.4 Comparison of the abrasion resistance of various elastomers
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.
Finally, it is important to compare the relative cost of the various oil-resistant

elastomers. The poundvolume cost is used, which is simply the cost of the compound
multiplied by its specific gravity. This gives the relative cost of the compound required
to fill a given volume such as a mould cavity. This has been shown earlier in Table
2.2, along with the upper temperature limits of the various materials. While acrylic
elastomers such as ACM and AEM are cost-effective up to 150 C, HNBR are much
tougher and more abrasion-resistant. FKM and fluorosilicones (FVMQ) have excellent
upper temperature limits, but their poundvolume cost is heavily influenced by their
relatively high specific gravity.
19
Table 2.3 gives a cost comparison of oil-resistant elastomers, and a combination

of Table 2.3 and Figures 2.2 to 2.4 puts the niche filled by HNBR materials in
perspective. HNBR combine toughness, abrasion resistance, fluid resistance and good
cost effectiveness at temperatures up to 150 C.
Table 2.3 Commercial comparison of oil-resistant elastomers

Polymer SG Cost of lbvol cost ($) Operating temperature,
compound ($) upper limit (C)
NBR 1.22 $1.00 $1.22 100
ACM 1.32 $2.30 $3.04 150
AEM 1.32 $2.40 $3.17 150
HNBR 1.22 $10.40 $12.69 150
FVMQ 1.53 $23.00 $35.19 200
FKM 1.86 $16.00 $29.76 250
Note: 75 Shore A, assumes carbon black filler, except for FVMQ
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
1. Rubber, American Society of Testing and Materials, Washington, DC, USA,

2006, 9, 2, D2000-06.
2. A Comparison of Oil Resistant Elastomers in Engine Oil and ATF, Z7.3.16,

Zeon Chemicals LP, Louisville, KY, USA, 1999, p.1.
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 The Basics of Viscoelastic Flow in Rubber
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
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.
An easy way to visualise pseudoplastic behaviour is to imagine a bottle of thick tomato

ketchup. If the bottle is allowed to stand and then upended the ketchup will remain
in the bottle, but if the bottle is shaken vigorously the ketchup will then flow freely.
At zero shear rate the viscosity (resistance to flow) is extremely high and flow does
not take place, but on agitation the shear rate increases dramatically, the viscosity
falls and the product can be poured without difficulty.
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
Shear rate (s1)
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.
= o/(1 + Ag1n), (3.3)
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
Figure 3.2 Comparison of theoretical models describing the pseudoplastic

behaviour of polymers
23
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
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
Figure 3.3 Die swelling (elastic recovery) versus shear rate
This type of relationship is important in describing dimensional problems at the high

shear rates encountered in transfer or injection moulding. For example, a rubber part
may have originally been formed by injection moulding, with a 20 s injection time
followed by 60 s vulcanisation. As a result of cost pressures, reductions in cycle time
are required and the manufacturer then increases the injection pressure to bring the
injection time down to 10 s. Unfortunately this increases the shear rate by a factor of
2, which increases the elastic rebound shown as B in Equation (3.4). The rebound is
manifested by a sudden shift in the diameter of the finished part, making it undersized.
Very soon the manufacturers injection press becomes an expensive scrap machine.
As mentioned above, this is the least understood facet of the viscoelastic flow of
rubber but it is important in understanding some of the unwelcome consequences
of process changes.
25
3.3 Effect of Process Variables
3.3.1 Effect of Temperature on Viscosity
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):
log(T1/T1) log(Tg/Tg) = log(aT1), (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.
Obviously, the determination of viscosity at Tg would be difficult or almost

impracticable. However, if we are only interested in comparing the viscosities at two
different process temperatures, T2 and T1, we can simplify matters and use a relative
comparison:
log(T2/T2) log(T1/T1) = log(aT2) log(aT1), (3.6)
in which the values of aTx can be estimated using Equation (3.7):
log(aTx) = [17.44(Tx Tg)] / [51.6 + (Tx Tg)]. (3.7)
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
1000000
Viscosity (Pa. s)
100000
100 C
10000
125 C
1000 150 C
100
1 10 100 1000
Shear rate (s )
1
Figure 3.4 Temperature effect on flow profile of a typical rubber material
3.3.2 Heat Build-up During Flow
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.
DT = 0t (/rc)g2 dt, (3.8)
where DT is the temperature change in C s1, is the viscosity in Pa. s and g is shear
rate in s1.
Using Equation 3.8 we can establish the following proportionalities in order to

estimate the effect of viscosity and shear rate on heat build-up:
DT h, and DT g2 (3.9)
27
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
Viscosity effect Shear rate effect
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
3.4 Effect of Compounding Variables on Viscosity, Elasticity and

Flow
3.4.1 Effect of Polymer Molecular Weight
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)
where Ni is the number of molecules of molecular weight Mi and is a weighting average.
If j = 1, then it is the number average or simple arithmetic mean, Mn.
If j = 2, then this is the weight-average molecular weight, Mw, which is sensitive to

higher molecular weight fractions.
If j = 3, this is known as the z-average molecular weight, Mz.
If j = 4, then this is described as the z + 1 average molecular weight, Mz+1.
A typical distribution of molecular weights, with various averages, is shown in Figure

3.6. The breadth of the distribution may be expressed as the weight-average molecular
weight, Mw, divided by the number-average molecular weight, Mn, and is described as
the polydispersity. For a typical emulsion polymer such as styrenebutadiene rubber
(SBR) or nitrile rubber (NBR), polydispersity values are typically around 2. The
polydispersity number can be larger than 2 for broad molecular weight distribution
polymers such as ZieglerNatta-catalysed polyethylene or ethylenepropylene diene
rubber (EPDM).
29
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
Figure 3.6 Example of molecular weight distribution of polymers
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):
h M when M < Mc, and h M3.4 when M > Mc, (3.11)
where h is viscosity, M is molecular weight and Mc is the critical molecular weight

for entanglement.
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
ho = kMw3.4, (3.12)
where ho is zero shear rate viscosity and k is a constant for a particular polymer system.
An interesting analysis carried out on polyethylene with a very broad molecular

weight distribution [24] has indicated that viscosity-average molecular weight gives
a better correlation with the experimental data. This may be relevant in the case of
EPDM polymers produced with similar polymerisation catalysts, giving a very broad
range of molecular weights.
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.
Molecular weight = 120,000

Mooney viscosity = 93
1000000
100000
Viscosity (poise)
10000 Molecular weight = 80,000

1000
100 Molecular weight = 100,000

10
1 10 100 1000
Shear rate (s1)

Figure 3.7 Effect of molecular weight on flow behaviour
31
3.4.2 Effect of Fillers
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
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.
Table 3.1 Experimental design formulations, NBR compound, triplicate data

points for each formula basic recipe
Ingredient Phr
33% Acrylonitrile, 35 ML(1+4)@100C polymer 100.00
Stearic acid 2.00
Agerite resin D 1.00
Vanox ZMTI 1.00
Carbon black (see below) 30.00 to 50.00
Zinc oxide 2.00
Varox DCP-40C 3.00
Compound Phr carbon black DBP number of carbon

black
1 30.00 43 (N-990)
2 50.00 43
3 70.00 43
4 30.00 65 (N-762)
5 50.00 65
6 70.00 65
7 30.00 90 (N-660)
8 50.00 90
9 70.00 90
10 30.00 121 (N-550)
11 50.00 121
12 70.00 121
33
Response surface for compression modulus in Pa
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
Figure 3.8 Effect of carbon black on compression modulus
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
Hysteresis as a function of filler and loading
% Hysteresis, Compression
63
53
43
33
130
110
23 90
70
30 40 50 DBP Number
5060 30
70
phr Carbon Black
R-Squared adjusted for degrees of freedom = 96%
Figure 3.9 Effect of carbon black on rubber material hysteresis
3.4.3 Effect of Plasticisers
This is a further area where comprehensive studies need to be undertaken, although

good work has been conducted to pave the way. Kraus and Gruver [30] have studied
the effects of various plasticisers on the viscosity of rubber compounds and have
proposed the following equation to describe their influence on viscosity:
h = f23.4 F (gf21.4), (3.14)
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:
h1/h0 = (f13.4F(gf11.4)/ (f03.4F(gf01.4) = f14.8/f04.8 (3.15)
35
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
Figure 3.10 Effect of plasticiser on viscosity
3.4.4 Effect of Process Aids
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
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:
PT = Dpe + Dp, (3.17)
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:
PT = Dpe + 4(s12)w L/D, (3.18)
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.
3.5 Flow through Sprues and Runners: Mould and Machine

Parameters Influencing Elastomer Flow
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
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:
Shear stress = DP.R / 2L, (3.19)
where DP is pressure drop and R and L are the radius and length of the capillary,
respectively.
Shear rate = 4Q / pR3, (3.20)
where Q and R is the radius of the capillary.
Apparent viscosity happ = shear stress / shear rate
= DP.R4.p / 4.Q.2L, and

the volumetric flow rate Q = DP.R4p / 4h .2L.
app
(3.21)
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
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
Figure 3.11 Effect of various parameters on the flow rate of an elastomer
References
1. E.A. Collins and J.T. Oetzel, Rubber Age, 1970, 102, 64.
2. E.A. Collins and J.T. Oetzel, Rubber Age, 1971, 103, 3.
3. S. Einhorn and S.B. Turetzky, Journal of Applied Polymer Science, 1964, 8,

1257.
4. J.R. Hopper, Rubber Chemistry and Technology, 1967, 40, 462.
5. M. Mooney, Physics, 1936, 7, 413.
6. M. Mooney in Rheology: Theory and Applications, Ed., F.R. Eirich,

Academic Press, New York, USA, 1958, p.196.
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
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.
20. T. Masuda, K. Kitagawa, T. Inoue and S Onogi, Macromolecules, 1970, 3,

116.
21. R.S. Porter and J.F. Johnson, Proceedings of the 4th International Rheology
Congress, Providence, RI, USA, 1965, 2, 467.
22. R.S. Porter and J. F. Johnson, Chemical Reviews, 1966, 66, 1.
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.
25. E. Guth and R. Simha, Kolloid-Zeitschrift, 1936, 74, 266.
26. D. Barthes-Biesel and A. Acrivos, International Journal of Muliphase Flow,

1973, 1, 1.
27. J.L. White and J.W. Crowder, Journal of Applied Polymer Science, 1974, 18,
1013.
40
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
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.
4.2 Relationship between Hydrogenated Nitrile Rubbers and Other

Speciality Elastomers
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].
Since HNBR is a hydrogenated derivative of NBR, a useful first step might be to

compare the two. NBR polymers are generally characterised as BG (100 C test
temperature, 40% or less volume swelling in IRM903 oil). NBR materials can
however be formulated with a high ACN content to give very low swelling in
hydrocarbons, and are then characterised as BK (100 C test temperature, 10% or less
volume swelling in IRM903). On the other hand, NBR with very low ACN content
gives greater swelling in hydrocarbons but much improved low temperature flexibility
(ASTM D2000 BF). Using peroxide curing and suitable antioxidants, NBR can also
be made to have a higher upper temperature limit (CH). NBR elastomers generally
lie within the range: BF, BG, BK or CH, as illustrated in Figure 4.1.
43
Table 4.1 ASTM D2000 type and class system

Type Test temperature Class %Volume swelling in IRM903 oil
(C) after 70 h immersion
A 70 A No requirement
B 100 B 140
C 125 C 120
D 150 D 100
E 175 E 80
F 200 F 60
G 225 G 40
H 250 H 30
J 275 J 20
K 300 K 10

% Swell in IRM903 Oil
300 Type by test temperature
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
Figure 4.1 Graphical representation of NBR in the ASTM D2000 system
44
Properties of Hydrogenated Nitrile Rubber and Comparison with Other Elastomers
Following hydrogenation the degree of unsaturation in NBR is significantly reduced

to give HNBR, as indicated in Chapter 1. This reduces the tendency for high
temperature oxidation and gives HNBR improved thermal stability in air and other
oxygen-containing environments. Due to the wide range of ACN contents used in
HNBR (from 17 to 50 mole% ACN) and its higher upper temperature limit, it is not
surprising that HNBR falls within the range CF, CH, DF, DH, CJ or DJ. In Figure
4.2 HNBR has been included with NBR in a graphical representation of the ASTM
D2000 system.

% Swell in IRM903 Oil
300
Type by test temperature

250 H
200 F
E
150 D
HNBR
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
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

300
Temperature Resistance (C)

FKM 250
VMQ FVMQ 200 H

AEM ACM
EPDM 150 F
HNBR E
CR 100 D
NBR C
NR
50 B
A
0
200 150 100 50 0
Class by Oil Swell
A B E F G H J K
Note : NR = natural rubber, EPDM = ethylenepropylene diene rubber, CR = chloroprene rubber,
VMQ = vinylmethyl silicone, ACM = acrylic, AEM = ethyleneacrylic, FVMQ = fluorinated vinyl
methyl silicone, FKM = fluorocarbon
Figure 4.3 Graphical representation of a variety of elastomers within the ASTM

D2000 system
4.2 Specific Comparison of HNBR and NBR
The effect of catalytic hydrogenation in converting NBR to HNBR is interesting, and

previously unpublished studies by the author may amplify this comparison. A simple,
basic recipe was used to compare NBR and HNBR filled with N762 carbon black
at 50 parts per hundred rubber (phr). The recipe, vulcanisation conditions and test
results are summarised in Table 4.2.
46
Table 4.2 Comparison of NBR and HNBR

Polymer A: HNBR polymer; ACN: 36%; nominal unsaturation: 4%; viscosity ML(1+4) at
120 C: 85 Mooney units
Polymer B: NBR polymer, cold polymerised; ACN: 33%; viscosity ML(1+4) at 100 C: 50
Mooney units
Ingredient Recipe A (phr) Recipe B (phr)
Polymer A (HNBR) 100
Polymer B (NBR) 100
N762 carbon black 50 50
Polymerised 1,2-dihydro-2,2,4-trimethyl quinoline 1 1
Stearic acid 1 1
Dicumyl peroxide, 40% on inert mineral carrier 7.5 3.5
Triallyl isocyanurate, 72% on inert mineral carrier 4.5
Zinc oxide 5 5
Note: phr = parts per hundred rubber polymer
Properties compared Results: Results:

Recipe A Recipe B
(NBR) (HNBR)
Shore A hardness, ASTM D2240 74 72
Tensile strength (mPa), ASTM D412, die C 15.8 16.5
Ultimate elongation (%), ASTM D412, die C 145 241
Compression set, ASTM D395, Method B: piled discs, 22 h at 150 C 26.5 9.8
% Volume change, IRM901 oil, 70 h at 150 C, ASTM D471 0.2 +0.3
% Volume change, IRM902 oil, 70 h at 150 C, ASTM D471 +3.8 +10.1
% Volume change, IRM903 oil, 70 h at 150 C, ASTM D471 +14.8 +16.8
% Volume change, ASTM Reference Fuel A, 70 h at 23 C, ASTM +1.5 +4.8
D471
% Volume change, ASTM Reference Fuel B, 70 h at 23 C, ASTM +25.6 +45.2
D471
% Volume change, ASTM Reference Fuel C, 70 h at 23 C, ASTM +32.1 +54.8
D471
47
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:
Better oxidative thermal stability.
Increased swelling in hydrocarbons.
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
90
80 HNBR
70 y=1.7361x+116.54
R2=0.9996
% Volume change
60
50 NBR
40 HNBR
30 Linear (NBR)
20 NBR Linear (HNBR)

y=1.9652x+97.386
10
R2=0.9995
0
10
0 10 20 30 40 50 60
Mole % ACN
Figure 4.4 Swelling of NHBR and NBR in Reference Fuel C
4.3 Comparison of HNBR with HNB and Other Speciality

Elastomers
As has been demonstrated earlier, HNBR combines the compounding flexibility

and toughness of NBR with improved temperature and chemical resistance. This
gives much better resistance to oxidation and sulfur attack at higher temperatures,
although at the cost of slightly greater swelling in hydrocarbons. A clear illustration
of this is given in Figures 4.5 and 4.6 for engine oil and automatic transmission
fluid, respectively. The loss of elongation after 1008 hours at 150 C is shown for
two HNBR, ACM and AEM materials [2]. An estimate of the life of an elastomer
is expressed as the time over which elongation is reduced to 50% of its original
value. Figures 4.5 and 4.6 confirm that HNBR materials retain their useful life after
long-term ageing. ACM and AEM materials show excellent long-term resistance to
hydrocarbon fluids, but in general do not offer the tear and abrasion resistance of
HNBR materials.
49
Comparison of oil-resistant
elastomers, 10W30SG engine oil at 150 C
40
20
% Elongation loss

-20 % Elongation Loss, 1008 h
-40
-60
HNBR HNBR
Material
Figure 4.5 Comparison of oil-resistant elastomers in loss of elongation in

automotive engine oil at high temperature
Comparison of oil-resistant
elastomers, Dexron III ATF at 150 C
30
20
% Elongation loss
10 % Elongation loss, 504 h

0
% Elongation loss, 1008 h
-10
-20
-30
HNBR HNBR
Material
Figure 4.6 Comparison of oil-resistant elastomers in loss of elongation in

automatic transmission fluid (ATF)
50
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.
Akron-type abrasion resistance rating

Relative resistance to abrasion
200
150
100
50
0
45% ACN NBR ECO FKM ACM
HNBR
Material
Figure 4.7 Comparison of oil-resistant elastomers in relative abrasion resistance
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.
Zeon Chemicals have tested a range of elastomers as inhibitors against aggressive

corrosion in petroleum applications and in mixtures of carbon dioxide, methane
and hydrogen sulfide [4]. The results for amine exposure are replotted in Table 4.4.
51
Table 4.4 Resistance of various elastomers for loss of elongation in reference

oil with amine corrosion inhibitor
Rubber material % Elongation change
36% ACN HNBR, peroxide cured 9
45% ACN HNBR, peroxide cured 32
FEPM, peroxide cured 15
Type 1 FKM (66 mole% fluorine), bisphenol cured 40
Type 2 FKM (70 mole% fluorine), peroxide cured 37
NBR (41 mole% ACN), Semi-EV sulfur cured 26
Note:
Test fluid: IRM902 oil containing 1% NACE Amine B corrosion inhibitor
Test immersion conditions: 168 h at 150 C
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.
Finally, it is useful to compare the relative costs of various oil-resistant elastomers.

These are summarised in Table 4.5. The poundvolume cost is the cost of the
compound multiplied by the specific gravity, and gives a relative idea of the cost of
the parts produced. These values are based on the published prices of raw materials
in late 2010.
52
120
100
36%ACN HNBR, peroxide
cured
% Elongation retained
80
45%ACN HNBR, peroxide
cured
60 FEPM, peroxide cured
Type 1 FKM (66 mole%

40 Fluorine), Bisphenol cured
Type 2 FKM (70 mole%
Fluorine), peroxide cured
20 NBR (41 mole%ACN),
Semi-EVSulfur cured
0
0 50 100 150 200
Ageing time (h)
Figure 4.8 Comparison of various elastomers for resistance to sour gas at 23 C
Table 4.5 Cost comparison of oil-resistant elastomers

Specific gravity Cost of lbvolume Upper operating
Polymer
of compound compound cost temperature limit (C)
NBR 1.22 $1.00 $1.22 100
ACM 1.32 $2.30 $3.04 150
AEM 1.32 $2.40 $3.17 150
HNBR 1.22 $11.10 $13.54 150
FVMQ 1.53 $23.00 $35.19 200
FKM 1.86 $15.64 $29.09 250
To summarise, while HNBR materials are similar in performance to ACM materials,

HNBR becomes the choice when toughness and abrasion resistance are required,
along with resistance to oils and high temperatures.
53
References
1. American Society for Testing and Materials, D200008 in Rubber, 9.02,

American Society of Testing and Materials, Washington, DC, 2009, p.1.
2. Zeon Chemicals LP, A Comparison of Oil Resistant Elastomers in Engine Oil

and ATF (Z7.3.16), Louisville, KY, USA, 1999, p.1.
3. Zeon Chemicals LP, Zetpol Product Guide, Louisville, KY, USA, 1999.
4. Zeon Chemicals LP, Zetpol HNBR Elastomers: Technical Bulletin SA13,

Applications for Oil Field Drilling, Recovery and Processing, Louisville, KY,
USA, 2001.
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.
5.1 A Brief Recap
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.
Information has been presented in Chapter 4 demonstrating the excellent abrasion

resistance and enhanced thermal stability of HNBR compared to other oil-resistant
elastomers. This feature is important in understanding that in many applications in
which NBR has historically been used, but for which higher application temperatures
and the demand for longer component life now require it to be replaced with HNBR.
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

300

FKM 250
VMQ FVMQ 200 H

AEM ACM
EPDM 150 F
HNBR E
CR 100 D
NBR C
NR
50 B
A
0
200 150 100 50 0
Class by Oil Swell
A B E F G H J K
Note : NR = natural rubber, EPDM = ethylenepropylene diene rubber, CR = chloroprene rubber,
VMQ = vinylmethyl silicone, ACM = acrylic, AEM = ethyleneacrylic, FVMQ = fluorinated vinyl
methyl silicone, FKM = fluorocarbon
Figure 5.1 Graphical representation of a variety of elastomers within the ASTM

D2000 system
5.2 Abrasion-resistant Belts and Conveyor Components
In Chapter 4 we discussed the relative abrasion resistance of HNBR compared to

other elastomers. HNBR materials also have excellent flex life in belts and other
conveying equipment, which results in the longer life of components. The removal
of most of the unsaturation in NBR is the key to the excellent flex life of HNBR.
Similarly, HNBR is much less susceptible to oxygen and ozone attack, and this also
greatly extends its flex life.
Acrylate modification of HNBR (Chapter 2) further improves its toughness, abrasion

resistance and flex life. The enhanced high temperature resistance of HNBR materials
also helps to extend the life of components in which long-term exposure to high
temperatures is important for improving the durability of elastomeric components.
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:
Fabric-reinforced belts for automotive power accessories:
Timing belts.
Drive belts for air conditioning compressors, alternators and hydraulic steering
pumps.
Belt-driven power take-off pumps.
Conveyor belting, either homogeneous rubber or fabric-reinforced:
Acrylate-modified HNBR for conveying abrasive powders, including minerals

and grain.
Rubber rollers:
In paper printing, especially acrylate-modified HNBR.
Rollers for conveyor systems handling abrasive minerals, grain or grain by-
products.
5.3 Flexible Boots for Power Transmission Joints
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
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:
Drive train boots in automotive and truck applications:
Constant-velocity joints near exhaust headers.
Axle joints close to exhaust pipes or catalytic converters.
Steering boots:
Rack and pinion ends near exhaust headers.
Input shafts close to the engine.
Suspension components:
Tie-rod end boots and strut end covers near exhaust systems.
5.4 Static Seals for Power Transmission Fluids
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
Examples of static seals used in power transmission include:
O-rings and square cross-section seals for long-life hydraulic power steering
gaskets, pumps and reservoirs:
Pump seals.
Fluid tube fitting seals.
Fluid reservoir seals.
Rack and pinion fluid seals.
O-rings, gaskets and cross-section seals for automatic and manual transmission
fluids, differential joints and pan seals.
5.5 Dynamic Fluid Seals for Power Transmission
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.
Radial differential shafts.
Hydraulic power-assisted steering pumps.
5.6 Hydraulic Fluid and Lubricant Hoses
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
5.7 Seals for High-volume Air Conditioning
HNBR provides an excellent combination of thermal resistance and resistance to

commercial refrigerants. Automotive systems include the older CFC12/mineral oil
lubricants or the newer HFC134a/polyalkylene glycol systems. HNBR also has
good resistance in seals for high-volume air conditioning (HVAC) in buildings, such
as HFC22/mineral oil lubricant systems, making it an excellent choice for these
applications:
Static seals:
O-rings.
Square cross-section seals.
Gaskets (homogeneous rubber or composite-reinforced).
Dynamic seals:
Radial shaft seals for rotation or reciprocation.
5.8 Seals used in Petroleum Exploration and Drilling
In common with other applications, a combination of thermal stability, resistance

to aggressive chemicals and abrasion resistance make HNBR an excellent choice in
petroleum handling, for example:
Packer elements and blow-out preventers.
O-rings:
Packer element static seals.
Tubing seals.
Pump static seals.
Drill bit seals.
Pump radial shaft seals.
5.9 Other applications
These include:
60
Static seals for hydrocarbon fluid cooler systems (e.g., engine oil, or power
transmission fluids):
O-rings.
Square cross-section seals.
Gaskets.
Flexible protective bellows on mechanical face seals.
Engine fuel static seals (away from the heat of the engine):
O-rings for fuel injector return lines.
References
1. R.W. Keller, Rubber Chemistry and Technology, 1985, 58, 3, 637.
61
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.
6.2 Vulcanisation 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
Table 6.1 Sulfur cure study by Zeon: Recipes and properties

Ingredient Recipe
1 2 8 9 10 11
Zetpol 1020 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 1.0 1.0 1.0 1.0 1.0 1.0
N550 carbon black 40.0 40.0 40.0 40.0 40.0 40.0
Permanox OD 2.0 2.0 2.0 2.0 2.0 2.0
Sulfur (325 mesh) 0.5 0.3
4,4-Dithiomorpholine 0.50 0.50 2.0 2.0 1.0 1.0
Dipentamethylene thiuram 0.7 0.7 0.7
hexasulfide
Tetramethylthiuram disulfide 2.0 2.0 0.7 0.7 0.7 0.7
N-cyclohexyl-2-benzothiazole 1.0
sulfenamide
2-Mercaptobenzothiazole 0.5
Zinc dimethyl dithiocarbamate 0.5 0.5 0.5
Zinc dibutyl dithiocarbamate 0.5 0.5 0.5 0.5
Tellurium diethyl 0.5
dithiocarbamate
Properties after vulcanisation, 20 min at 160 C

Property Recipe from above
1 2 8 9 10 11
Shore A hardness 74 74 75 75 75 75
Tensile strength, MPa 23.8 24.4 23 23.1 23.1 22.5
Ultimate elongation, % 460 450 410 400 420 390
100% Modulus, MPa 4.1 4.4 4.6 4.8 4.3 4.9
300% Modulus, MPa 16.6 16.9 18.3 18.6 17.8 18.6
Compression set, ASTM D395 Method B, 25% Recipe from above
compression
1 2 8 9 10 11
70 h at 100 C 35 34 37 41 42 41
70 h at 120 C 54 46 46 56 57 57
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

7 8 9 10 11 12
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- 1.5 1.5 1.5 1.5 1.5 1.5
methylmercaptobenzimidazole
,-bis(t-butylperoxy) 7.5 7.5 10.0 10.0 10.0 10.0
diisopropylbenzene
N,N-m-phenylene dimaleimide 5.0 7.5 0.0 2.5 5.0 7.5
Properties after vulcanisation 15 min at 170 C

Property Recipe
1 2 3 4 5 6
Tensile strength, MPa 28.0 30.1 28.1 28.0 32.4 30.2
50% Modulus, MPa 1.6 2.0 2.5 3.3 1.7 2.2
65
100% Modulus, MPa 2.2 4.1 6.4 9.0 2.9 5.0

Compression set, ASTM D395 61 32 26 22 42 29
Method B, 70 h at 150 C
Property Recipe
7 8 9 10 11 12
Tensile strength, MPa 29.3 27.1 32.0 28.2 28.6 27.7
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
The response surface analysis of the data gives the following predictive equations for
the properties as a function of the peroxide and coagent levels:
Shore A hardness = 74.70 + 0.75P + 4.0C + 0.5P2 0.375C2 + 0.15CP, (6.1)
where P = phr ,-bis(t-butylperoxy)diisopropylbenzene peroxide, and
C = phr N,N-m-phenylene dimaleimide coagent.
R-squared, adjusted for degrees of freedom = 97.74%.
100% modulus = 1.774 0.052P 0.034C + 0.006P2
+ 0.0187C2 + 0.0232PC (6.2)
% elongation = 711.6 49.6P 76.9C + 1.0P2 + 3.73C2 + 2.96PC (6.3)
% compression set = 84.1 6.18P 12.3C + 0.12P2 + 0.56C2 + 0.688PC (6.4)
66
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
Figure 6.2 Contour plot of modulus at 100% elongation as a function of peroxide

and coagent level
67
6
250
5
Coagent phr
4
200
3
350
2
1
450 300
400
0
5 6 7 8 9 10
Peroxide phr
Figure 6.3 Contour plot of ultimate elongation as a function of peroxide and

coagent level
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
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]
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
Compression set, ASTM D395 75 67 60 50 35 25 24 23
69
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
6.3 Carbon Black Fillers
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].
Table 6.4 Zeon study of carbon black in HNBR, recipes

Test recipe ingredient phr
Zetpol 2010 100.0
Carbon black (various grades) 0.075.0
Zinc oxide, Kadox 911C 5.0
Stearic acid 0.5
TOTM, Plasthall 5.0
Naugard 445 1.5
Vanox ZMTI 1.0
Vul-Cup 40KE 8.0
70
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
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:
MV = Mooney viscosity, ML(1+4) at 100 C
= 60.16 0.929p + 0.746D + 0.00986p2 0.000540D2 + 0.0120pD, (6.5)
where p = phr carbon black and
D = DBP number of carbon black.
R-squared, adjusted for degrees of freedom = 89.66%
H = Shore A durometer hardness
= 34.36 + 0.264p + 0.372D 0.00071p2 0.00202D2 + 0.00207pD (6.6)
M1 = stress at 100% elongation (100% modulus) in MPa
= 0.0804p + 0.0593D + 0.00149p2 0.000360D2 + 0.000808pD (6.7)
72
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)
where G0 = shear modulus of unfilled material,
G1 = shear modulus of filled material, and
f = volume fraction of filler in the mixture.
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.
6.4 Non-black Fillers
It is sometimes unacceptable for the finished rubber article to be black. An example

of this might be when a black NBR component was being replaced by a HNBR
component of higher temperature capability, giving longer life to the final assembly.
If the NBR and HNBR rubber articles were both black it would be difficult for the
customer to monitor the changeover from NBR to HNBR. It might therefore be
desirable to make the finished HNBR rubber article a different colour, say green, to
avoid mistakes in assembly and warranty situations. Mineral fillers are important for
producing colours other than black in the finished rubber articles.
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
Table 6.5 Zeon study of non-black fillers

Ingredient phr Relative Shore 100% Shore Modulus at
particle A modulus A per 100% per phr
size phr
None 0 50 1.1
Ultra-Pflex 90 small 65 1.6 0.167 0.0056
Roy-Cal L 90 medium 65 1.8 0.167 0.0078
Zeeospheres 200 90 medium 66 2.4 0.178 0.0144
Imsil A8 90 medium 68 3.4 0.200 0.0256
Nulok 321 90 medium 73 5.0 0.256 0.0433
Nulok 390 90 medium 73 7.0 0.256 0.0656
Pyrax B 90 large 74 7.5 0.267 0.0711
Mistron 90 medium 76 8.1 0.289 0.0778
Cyprubond
Translink 77 90 medium 77 9.2 0.300 0.0900
Celite 350 90 large 80 9.2 0.333 0.0900
Silene 732D 90 small 83 10.2 0.367 0.1011
Zeolex 23 40 medium 65 2.7 0.375 0.0400
HiSil 532EP 40 small 68 3.3 0.450 0.0550
Cab-O-Sil TS720 40 small 71 2.5 0.525 0.0350
HiSil 233 40 small 73 2.9 0.575 0.0450
Ultrasil VN3 SP 40 small 74 3.0 0.600 0.0475
Aerosil 300 40 small 76 2.8 0.650 0.0425
Cab-O-Sil M7D 40 small 78 3.1 0.700 0.0500
Base formula without filler

Ingredient phr
Zetpol 2010 100.0
TOTM, Plasthall 5.0
Titanium dioxide 5.0
74
Zinc oxide 5.0

Stearic acid 0.5
Naugard 445 1.5
Vanox ZMTI 1.0
Carbowax 3350 2.0
A172 Silane 1.0
Vul-Cup 40KE 8.0
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:
Phr Nulok 390 = (7050)/0.256 = 78 phr
Phr HiSil 233 = (7050)/0.450 = 44 phr.
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
Table 6.6 Plasticisers for HNBR
Plasticiser at 15 phr in base recipe
Property No Dioctyl Dibutoxy Ether Trioctyl Triisononyl

plasticiser phthalate ethoxyethyl ester trimellitate trimellitate
adipate
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
Weight loss 1.4 9.7 8.6 6.1 3.4 1.7

after 168 h
at 150 C air
ageing, %
Temperature
retraction per
ASTM D1329
76
TR-10, C 22.6 28.7 25.1 25.0 28.1 27.9

TR-30, C 16.5 20.6 12.7 12.5 20.0 19.6
TR-70, C 4.5 8.5 3.3 3.8 6.8 6.8
Heat 1.4 9.7 8.6 6.1 3.4 1.7
resistance
weight loss,
168 h at 150
C
Softening Effects
Shore A per 0.4 0.5 0.5 0.4 0.5

phr plasticiser
Mooney 3.1 3.7 3.5 2.8 3.1

viscosity per
phr plasticiser
100% 0.140 0.187 0.193 0.133 0.127
Modulus per
phr plasticiser,
MPa
TR-10 per 0.407 0.167 0.160 0.367 0.353
phr plasticiser
(C)
Base recipe phr

ingredient
Zetpol 2010 100
N550 carbon 50
black
,-bis(t- 6
butylperoxy)
diisopropyl-
benzene,
40% on inert
carrier
77
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.
6.6 Antioxidants and Antiozonants
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].
6.7 Other Ingredients
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
6.8 Blends with EthylenePropylene Diene 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
6.9.1 Black HNBR Compound for a Reciprocating Lip Seal on a Power

Steering Rack End
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:
H = 34.36 + 0.264p + 0.372D 0.00071p2 0.00202D2 + 0.00207pD.
The value of D for N550 carbon black is 121, so we can solve for phr knowing
79
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.
1.5 phr 1,2-dihydro-2,2,4-trimethyl quinoline antioxidant.
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.
This gives the starting recipe in Table 6.7
Table 6.7 Possible recipe for reciprocating lip seal
Ingredient phr Comments
Zetpol 4310 EP 100.0 Very good low temperature

flexibility polymer
French process zinc oxide 5.0 Cure activator
Stearic acid 1.0 Cure activator, mould lubricant
1,2-Dihydro-2,2,4-trimethyl quinoline 1.5 Effective antioxidant

antioxidant
N550 Carbon black 75.0 Carbon black filler
,-Bis(t-butylperoxy)- 10.0 Efficient crosslinking, low

diisopropylbenzene peroxide compression set
N,N-m-Phenylene dimaleimide coagent 7.5 Efficient crosslinking, low

compression set
6.9.2 Light Brown HNBR Compound for a Hydraulic System O-ring
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
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.
Filler: Zeeospheres 200 to allow high filler loading:
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).
The resulting starting recipe is given in Table 6.8.
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
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.
5. R.W. Keller, Hydrogenated Nitrile Rubber, in Handbook of Specialty

Elastomers, Ed., R.A. Klingender, CRC Press, Boca Raton, FL, USA, 2008.
6. Zeon Chemicals LP, Evaluation of Larger Particle Non-Black Fillers in

Peroxide Cured Zetpol 2010 (Z5.3.3), Louisville, KY, USA, 1999, p.1.
7. Zeon Chemicals LP, Low Temperature Properties of Zetpol and Influence of

Plasticizers and Tackifiers (Z5.4), Louisville, KY, USA, 1999, 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.
7.2 Statistical Experimental Design and 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.
In addition to statistical experimental design, the desirability function concept is also

useful in evaluating rubber formulations [3, 4]. The basics of the desirability function
are given by Equation (7.1).
83
n n
D= (d1, d2, d3, ... dn)
d=1 , (7.1)
where D is a composite desirability number between 0 and 1.00 (0 = completely

undesirable and 1.00 = completely desirable),
d1 is the desirability of property 1 in the range 0 to 1.00,
d2 is the desirability of property 2 in the range 0 to 1.00 and
dn is the desirability of property n in the range 0 to 1.00.
The individual desirabilities can have pass/fail, linear or curvilinear functions. An

example of a pass/fail property might be ozone resistance (ASTM D1171) above
168 h, any cracking being considered undesirable (0) and total absence of cracking
being completely desirable (1.00). The dn relationships are usually arrived at after
full involvement of the customer. The remainder is a straightforward calculation.
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.
Table 7.1 Example of composite desirability calculations

Formulation d1 d2 d3 d4 d5 D
Rubber Formula 1 0.1 0.8 0.9 0.3 0.2 0.336587
Rubber Formula 2 0.6 0.9 0.9 0.9 0 0
This results in Equation (7.2):
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.
Rubber formulations often need to balance a number of properties, making the

desirability function approach ideal for providing a composite numerical rating for
a number of formulations and pointing the way to the most appropriate overall
formula. In Section 7.3 both statistical experimental design and desirability function
are employed in a theoretical example of a HNBR formula for a specific customer
application.
7.3 Theoretical Example: High-temperature HNBR Joint Boot
In this example we will imagine that we have been approached by a customer, a

manufacturer of drive shaft components, who has a specific requirement. On an all-
wheel drive vehicle the joint boot must exclude dirt and moisture as well as retaining
the joint grease, but the joint happens to be close to the exhaust pipe and traditional
materials such as CR rubber or thermoplastic elastomers are unable to withstand
the temperature and satisfy long-term warranty requirements. The customer has
estimated a typical operating temperature at the joint boot of 125 C. A material is
required that gives good flex life, resistance to the joint grease, and wear and abrasion
resistance, all at this operating temperature. From the ASTM D2000 classification
chart in Chapter 4, shown in Figure 7.1, it is clear that HNBR is the correct choice.

% Swell in Irm-903 Oil
300
Type by test temperature
FKM
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
Figure 7.1 Graphical presentation of ASTM D2000
85
Working with the customer, we have identified the individual desirabilities shown
in Table 7.2.
Table 7.2 Definition of desirability of individual properties, theoretical joint boot

compound
d1 Definition: d1 = 1.00 if weight loss after 1,000 hours at 125 C in dry heat is 05%
d2 Definition: d2 = 1.00 if volume change in customer axle grease is > 0% after 1,008
hours at 125C (ASTM D471); d2 = 0.00 if volume change is negative in the grease
exposure test
d3 Definition: d3 = 0.00 if TR10 temperature is above 40 C; if TR10 temperature
is below 40 C, then linear scale applies from 0.00 at 40 C to 1.00 at 50 C (A,
below)
d4 Definition: d4 = 1.00 if dynamic flex life rating is < 3 after 1,000,000 cycles; d4 = 0.00
if dynamic flex life rating is 4 after 1,000,000 cycles
d5 Definition: d5 = 1.00 of static ozone quality retention rating is 100 after 168 h (ASTM
D1171); d5 = 0.00 if quality retention rating is < 100
d6 Definition: d6 = 1.00 if plied disc compression set in customer axle grease is 0% after
1,008 h at 125C; d6 = 0.00 if plied disc compression set is 100% or greater; linear
between 0 and 100% compression set (B, below)
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
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
Ingredient phr Requirements Experimental design

optimisation
17 mol% HNBR 80 Good combination of Vary level from 80 to

polymer, 4% residual 100 freeze resistance and low 100 phr blending with
unsaturation temperature flex properties EPDM polymer to
achieve total 100 phr
polymer
EPDM polymer low 020 Low viscosity for injection Vary level from 20 to
ethylene, low viscosity, mould flow; medium to high 0 phr with blending
4.5% third monomer third monomer level for of HNBR polymer to
unsaturation level efficient vulcanisation achieve total 100 phr
polymer
Zinc oxide 2.0 Cure activator
Stearic acid 1.0 Cure activator
Polymerised 1.0 Optimal antioxidant and
1,2-dihydro-2,2,4- antiozonant balance
trimethyl quinoline
antioxidant
Paraffin wax 2.0 Optimal antioxidant and
antiozonant balance
N762 Carbon black 55.0 Reasonable reinforcement

while maintaining high
elongation for flex resistance
Trioctyl trimellitate 10.0 Reduction of compound

plasticiser viscosity for injection
moulding; improvement of
low temperature flexibility;
good heat stability
Dicumyl peroxide, 40% 3.0 Fast and efficient curing in Optimise level of
on inert mineral carrier 5.0 injection moulding where peroxide for best balance
cycle time is critical of properties
87
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
phr EPDM in polymer blend
Figure 7.2 Graphical depiction of HNBR joint boot compound experimental

design
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
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
2.0 1.0 0.0 1.0

3.0
0 5 10 15 20
phr EPDM
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
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
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
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
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.
7.4 Compound Example: High Temperature Oil Cooler O-ring Seal
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.
Responses (properties) to be measured at 125150 C include:
Resistance to diesel engine oil.
Resistance to water/glycol coolant.
O-ring compression set resistance.
93
Table 7.5 Oil cooler O-ring compound experimental design
Ingredient phr Logic/comments Experimental

design
optimisation
HNBR polymer 100.0 Best overall balance of fluid resistance

with 3436% ACN; and low temperature properties; the
45.5% residual higher levels of unsaturation give
unsaturation the most efficient cure and balance
between temperature resistance and
compression set resistance
Zinc oxide 2.0 Activator
Stearic acid 1.0 Activator
Naugard 445 0.7 Best overall heat resistance stabiliser

antioxidant package
Vanox ZMTI 0.7 Best overall heat resistance stabiliser

antioxidant package
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
Varox VC40KE 10.0 Peroxide for compression moulding at

180 C
Vanax MBM 4.57.5 Coagent for increased efficiency Optimise coagent

(VARIABLE) of cure and best compression set levels for best
resistance balance of
properties
Optimisation:
2-factor 3-level full
factorial design; 8
compounds and
experiments
94
7.5 Compound Example: Oil Field High-pressure Well Packer
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.
Responses (properties) to be measured include:
Heat resistance, 125150 C.
Crude oil resistance.
Hydrogen sulfide resistance.
Steam resistance.
Aggressive amine resistance.
Pressure extrusion resistance.
95
Table 7.6 Oil well packer compound experimental design
Ingredient phr Logic/comments Experimental design

optimisation
HNBR polymer with 100.0 Best overall balance

34-36% ACN; less of heat and chemical
than 2% residual resistance
unsaturation
Naugard 445 1.5 Best antioxidant package

for long-term heat
resistance
Vanox ZMTI 1.5 Best antioxidant package

for long-term heat
resistance
N330 or N550 55.0 To obtain > 90 Shore A Evaluate two different

carbon black durometer for pressure carbon blacks for best
(VARIABLE) resistance overall balance of
properties
Varox VC40KE 10.0 High temperature

peroxide for compression
moulding at 180 C
Coagent: Vanax VARIABLE High level of coagent for Optimise compound

MBM at 7.5 phr efficient vulcanisation based on evaluation of
or TAIC at 5.0 phr two chemical types of
(VARIABLE) coagent
Optimisation:
2-factor 2-level full
factorial design
with centre point;
5 compounds and
experiments
96
7.6 Compound Example: High temperature Long-life Automotive

Serpentine Belt
In this example our customer is a manufacturer of automotive engines who has

encountered unacceptable warranty experience with chloroprene rubber serpentine
belts. While chloroprene rubber has excellent flex and weather resistance, the reduced
air flow through the engine compartment resulting from reduced drag on the car
has resulted in the serpentine belt operating continuously in the range 100125 C.
Previously, the serpentine belt operated at 7090 C and reduced life due to heat
ageing did not give rise to warranty claims. The belt is fabric reinforced to prevent
creep and elongation, so compression set is not critical. We require a compound of
relatively high elongation that gives excellent dynamic and static flex life.
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.
Responses (properties) to be measured include:
Heat resistance, 100125 C.
Static and dynamic ozone resistance.
97
Table 7.7 Serpentine belt compound: Experimental design

optimisation
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
Agerite Resin D 1.0 With wax, gives best

overall heat and ozone
resistance
Protective wax 2.0 With Agerite Resin D,

gives best overall heat
and ozone resistance
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
Varox VC40KE 4-8 High temperature Vary level between 4 and 8

(VARIABLE) vulcanisation system phr to obtain best balance of
properties
Optimisation:
3-factor, 3-level
full factorial design
with centre point;
27 compounds and
experiments
98
7.7 Compound Example: Orange Water Pump Mechanical Seal

Protective Bellows
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.
Responses (properties) to be measured:
Heat resistance, 1008 h at 125 C.
Ozone resistance.
Coolant resistance, 1008 h at 125 C.
Low temperature flexibility.
Dynamic flex fatigue resistance.
Moulding performance, flow and cure rates.
99
Table 7.8 Compound for water pump bellows experimental design

optimisation
2125% Acrylonitrile 100.0 Best low temp. flexibility for

HNBR polymer, 14% cold starts
residual unsaturation
Agerite Resin D 1.0 With wax, gives best overall
heat and ozone resistance
Protective wax 2.0 With Agerite Resin D, gives

best overall heat and ozone
resistance
Plasthall TOTM 10.0 Reduction of compound

viscosity for injection
moulding; improvement of low
temperature flex; good heat
stability
Nulok 321 50.0 Extending clay filler
HiSil 532EP (VARIABLE) 020 Reinforcing filler Level 1 will be this
grade
HiSil 233 (VARIABLE) 020 Reinforcing filler Level 2 will be this
grade
Vinyl-tris(- 1.0 Coupling agent for filler
methoxyethoxy) silane dispersion
(Silane A172)
Red iron oxide 4.0 Stable pigment
Varox DCP40C 48 Rapid vulcanisation on Three levels to
(VARIABLE) injection moulding at 170 C optimise properties
Optimisation:
2-factor and 3-level full
factorial design;
6 compounds and
experiments
100
7.8 Compound Example: High-temperature Differential Shaft Seal
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.
Responses (properties) to be measured are:
Resistance to gear lubricant at 125 C for 1008 h.
Moulding evaluations: flow, cleanliness and bonding to the steel case.
Shaft seal life testing at extremes of operating environment.
101
Table 7.9 Differential shaft seal compound experimental design

optimisation
HNBR with 34-36% ACN; 100.0 Overall balance of oil

1-4% residual unsaturation resistance and low
temperature properties

Naugard 445 0.7 With ZMTI, gives best
antioxidant performance
Vanox ZMTI 0.7 With Naugard 445,

gives best antioxidant
performance
Celite 350 or Mistron 75.0 Mineral fillers show Evaluate two different
Cyprubond (VARIABLE) best rotating shaft seal fillers for best overall
performance properties
N550 Carbon black 10.0 Black pigment

Varox VC40KE 5-10 Efficient vulcanisation Evaluate three levels of
peroxide for best overall
properties
Optimisation:
3-factor, 2,2- and 3- level
full factorial design;
12 compounds and
experiments
7.9 Compound Example: Short Steering System Hose
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
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.
Table 7.10 Steering system hose compound experimental design

optimisation
HNBR polymer with 100.0 Best overall balance Low level at 1%
3436% ACN; 1% and 5% of fluid resistance unsaturation, high level at
unsaturation 5% unsaturation
Zinc oxide 2.0
Stearic acid 1.0
N550 Carbon black 60.0
Plasthall TOTM 10.0 Viscosity reduction
without sacrifice of
heat ageing
Agerite Resin D 1.0 With protective wax,
gives best ozone
resistance
Protective Wax 2.0 With Agerite Resin
D gives best ozone
resistance
Varox DCP40C 3.0 Fast and efficient Variable level to optimise
5.0 curing properties
Optimisation:
2-factor, 2- and 3-level full
factorial design; 12 compounds
and experiments
103
Responses (properties) to be measured:
Static ozone resistance.
Dynamic ozone resistance.
Flex-fatigue resistance.
Extrusion processing.
High-temperature pressure pulse test, simulating application conditions.
7.10 Compound Example: Chemically Resistant, Low Hysteresis

Roller for Paper Mills
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.
Responses (properties) to be measured are:
Dynamic hysteresis.
104
Table 7.11 Low hysteresis paper mill roller compound experimental design
optimisation
HNBR polymer with 3436% 50.0 Basic HNBR for Level 1 evaluation of
ACN; 5% unsaturation chemical resistance HNBR
HNBR polymer with 3436% 50.0 Higher unsaturation Level 2 evaluation of

ACN; 15% unsaturation gives less hysteresis HNBR
HNBR polymer modified with 50.0 Adds toughness,

zinc methacrylate abrasion resistance and
good hardness
Silane-treated calcined clay 70.0 Mineral filler Level 1 filler evaluation
Mistron Cyprubond 70.0 Mineral filler Level 2 filler evaluation

Varox VC40KE 7.0
Vanax MBM 4.57.5 Efficient curing Variable levels to
optimise properties
Optimisation:
3-factor, 2,2- and 2-level full
factorial design; 8 compounds
and experiments
7.11 Concluding Comments
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
References
1. L.B. Barrentine, in An Introduction to Design of Experiments: A Simplified

Approach, American Society for Quality, Milwaukee, WI, USA, 1999.
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.
3. E.C. Harrington, Industrial Quality Control, 1965, 21, 494.
4. G. Derringer and R. Suich, Journal of Quality Technology, 1980, 12, 214.
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.
8.2 Example 1: Poor Flow, Knit Lines and Non-fills in Injection

Moulding
In this example a vulcanised HNBR article is being manufactured by multi-cavity or

multi-deck injection moulding. A sporadic problem has been encountered involving
poor flow and possible scorch (premature curing), causing unacceptable knit lines
and non-fills in the finished parts. A number of possible steps that may solve the
problem are described, and we will see how statistical experimental design can help
to determine the effect of each variable and possible interactions between them.
8.2.1 Possible Factor: Temperature of the Unvulcanised Rubber During

Injection
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
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)
Figure 8.1 Effect of temperature on pseudoplastic polymer viscosity
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.
8.2.2 Possible Factor: Excessive Shear During Injection
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
Viscosity effect Shear rate effect
Figure 8.2 Typical heat build-up during injection
As discussed in Chapter 3, Equation (8.1) can be used to establish correlation and

predict the effect of viscosity and shear rate increase:
DT = 0t (h/rc)g2 dt, (8.1)
where DT = temperature change in C s1,
h = viscosity in Pa. s and
g = shear rate in s1.
109
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:
Shear stress = (R.DP) / (2L), (8.3)
where DP = pressure drop,
R = radius of capillary and
L = length of capillary.
Shear rate = (4Q) / (pR3), (8.4)
where Q = volumetric flow rate,
R = radius of capillary and
happ = apparent viscosity
= (shear stress) / (shear rate)
= (DP.R4.p) / (4Q.2L)
110
Volumetric flow rate Q = (DP.R4.p) / ( 2L.4happ) (8.5)
8.2.3 Possible Factor: Sprue and Runner Design
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
Radius of runner Length of runner Viscosity of material
Figure 8.3 Effect of moulding parameters on flow rate
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
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
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
8.3 Example 2: Sporadic Dimensional Shift in Injection Moulded

Parts
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.
8.3.1 Possible Factor: Mould Temperature
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
temperature of 180 C on a 76 mm part and the moulding temperature drifted to

190 C, the shrinkage would increase to 0.025 mm/mm. While this might not seem a
major change in a 76 mm part, the change would reduce its finished dimension from
76.00 mm to 75.84 mm. Similarly, on a 127.00 mm target dimension the finished
part would be reduced in size from 127.00 mm to 126.62 mm.
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.
8.3.2 Possible Factor: Injection Pressures and Holding Pressures and

Times
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.
Perhaps the most surprising influence on the dimensions of an injection moulded

part is variation in the injection pressure holding time. Injection pressure is applied,
and then released after a specified time to allow the injector to recharge for the
next shot. Releasing the injection pressure early can have a profound effect on the
dimensions of the finished part. This has been frequently observed but has never been
thoroughly studied or quantified. However, in the case of large diameter parts such as
O-rings this factor can easily make the difference between dimensionally compliant
and rejected parts. In the initial set-up of the injection process, injection pressure
holding time must be considered for its effect on the finished parts. If the relationship
between injection holding time and final dimensions is established during the initial
production approval process, it is then more likely to be monitored and controlled
during routine production.
114
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
8.3.3 Solving the Injection Moulding Dimensional Problem using the

Above Factors
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.
Table 8.2 Experimental design for rectifying dimensional defects

Mould Injection Injection Responses Comments
temperature (C) time (s) pressure hold (dimensions)
time (s)
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
115
15
Press hold time, s

Injection time, s
20
5
10
180 190
Mould temperature, C
Figure 8.6 Injection dimensional shift: Graphical depiction of experimental design
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.
8.4 Example 3: Variation in Hose Tube Thickness during Extrusion
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
Slip agent used in the compound:
Low level: paraffin wax at 1.00 phr.
High level: fatty amide at 1.00 phr.
Extruder barrel temperature:
Low level: 80 C.
High level: 100 C.
Extruder breaker plate hole size:
Low level: 7.00 mm hole size.
High level: 8.00 mm hole size.
Taper angle at entrance to the tube die:
Low level: 30 angle.
High level: 45 angle.
Extruder speed:
Low level: 30 rpm.
High level: 40 rpm.
Extruder die temperature:
Low level: 100 C.
High level: 120 C.
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
Table 8.3 Hose tube extrusion, 8-run PlackettBurman screening design

Factors
Run A B C D E F G(dummy)
1 1.0 1.0 1.0 1.0 1.0 1.0 1.0
2 1.0 1.0 1.0 1.0 1.0 1.0 1.0
3 1.0 1.0 1.0 1.0 1.0 1.0 1.0
4 1.0 1.0 1.0 1.0 1.0 1.0 1.0
5 1.0 1.0 1.0 1.0 1.0 1.0 1.0
6 1.0 1.0 1.0 1.0 1.0 1.0 1.0
7 1.0 1.0 1.0 1.0 1.0 1.0 1.0
8 1.0 1.0 1.0 1.0 1.0 1.0 1.0
Significant variables (factors):
A = Slip agent in compound: 1 = paraffin wax at 1.00; +1 = fatty amide at 1.00
B = Extruder barrel temperature: 1 = 80 C; +1 = 100 C
C = Extruder breaker plate hole size: 1 = 7.00 mm; +1 = 8.00 mm
D = Taper angle at entrance to die: 1 = 30; +1 = 45
E = Extruder speed: 1 = 30 rpm; +1 = 40 rpm
F = Extruder die temperature: 1 = 100 C; +1 = 120 C
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.
8.5 Example 4: Variation in Dimensions of High-volume

Compression Moulded O-ring
This is an actual case in which a multi-disciplinary approach was used with

experimental design to solve a factory processing issue. A fairly large O-ring, nominal
ID 100 mm, was being produced by compression moulding in large moulds for a
high-volume application. The biggest issue was day-to-day and run-to-run variation
in the ID of the finished O-rings. Rather than attempting to study one variable at
a time, key personnel from technical and production areas were brought together
for a MECE discussion of possible variables. This discussion was conducted leaving
every possible factor open for discussion, eliminating any where the data supported
118
this action, and focusing on variables that might possibly be contributing to the
dimensional variation, including the following:
Compound variable: carbon black filler levels at the extremes of weighing

tolerance.
Compound variable: vulcanising agent (peroxide) at the extremes of weighing

tolerance.
Process variable: raw rubber weights at the extremes of tolerance.
Process variable: variations in mould surface temperature. The platen temperature

was set at 182 C but the actual temperature at the mould surface was found to
vary between 177 C at the corners and 188 C in the centre.
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.
Process variable: oven post-cure temperature. The product underwent a 2 h post-

cure with the oven set at 205 C, but measurements within the oven revealed that
the actual temperature could lie anywhere between 193 and 215 C.
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
Table 8.4 Screening design for dimensional resolution: 8-run PlackettBurman

screening design, compression moulded O-ring, 100 mm nominal ID
Run A B C D E F G (dummy)
1 96.3 96.3 96.3 96.3 96.3 96.3 96.3
2 96.6 96.6 96.6 96.6 96.6 96.6 96.6
3 92.3 92.3 92.3 92.3 92.3 92.3 92.3
4 92.7 92.7 92.7 92.7 92.7 92.7 92.7
5 96.6 96.6 96.6 96.6 96.6 96.6 96.6
6 99.9 99.9 99.9 99.9 99.9 99.9 99.9
7 96.3 96.3 96.3 96.3 96.3 96.3 96.3
8 100.7 100.7 100.7 100.7 100.7 100.7 100.7
Sum + 385.2 385.9 385.2 377.9 377.9 385.9 385.2
Sum 386.3 385.6 386.3 393.6 393.6 385.6 386.3
Average E+ 96.3 96.5 96.3 94.5 94.5 96.5 96.3
Average E 96.6 96.4 96.6 98.4 98.4 96.4 96.6
Effect 0.3 0.1 0.3 3.9 3.9 0.1 0.3
Significant? no no no YES YES no no
RESULTS
Run y1 y2 y3 y4 y5 Y (avg.) Si2 Replicates (n)
1 96.2 96.3 96.5 96.1 96.5 96.3 0.0632 5
2 96.8 96.5 96.2 96.8 96.9 96.6 0.1968 5
3 92.1 92.0 91.9 92.8 92.8 92.3 0.3272 5
4 92.6 93.0 93.1 92.1 92.5 92.7 0.6008 5
5 96.2 96.8 97.1 96.5 96.5 96.6 0.4392 5
6 100.2 100.1 99.5 99.6 100.3 99.9 0.2728 5
7 96.3 96.0 95.8 96.8 96.5 96.3 0.5312 5
8 101.0 100.8 100.2 101.1 100.5 100.7 0.4512 5
Si2 = sum of squares = (yiy(avg.))2 Sum Si2 2.882
Avg. Si2 0.36

Se = Si 0.6
Seff 0.19 = Se (4/N)
DF = 32 (n1) number of runs

N = total number of experiments
A = carbon black filler level at maximum of weigh-up tolerance
B = vulcanising agent levels at maximum of weigh-up tolerance
C = preform weight levels at extremes of tolerances
D = mould surface temperature: = 175 C, + = 190 C
E = mould closure: = 0.00 mm, + = 0.05 mm
F = Post-cure temperature: = 190 C, + = 215 C
MSFE: 0.387 t S(eff)
t (0.950, df) 2.04 (n1) (number of runs) = degrees of freedom
120
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
Figure 8.7 Bar chart rating magnitude of compression moulding dimensional

effects
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
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
ACM Acrylic elastomer (per D1418)
ACN Acrylonitrile
AEM Ethyleneacrylic elastomer
ASTM American Society for Testing and Materials
ATF Automatic transmission fluid
CFC Chlorofluorocarbon
CR Polychloroprene
D Composite desirability
DBP Dibutyl phthalate
d Desirability of an individual property
DSC Differential scanning calorimeter
ECO Epichlorohydrinethylene oxide copolymer
EDPM Ethylenepropylenediene terpolymer
FEF Fast extrusion furnace black
FEPM Polytetrafluoroethylenepropylene copolymer
FKM Fluorinated hydrocarbon rubber
FVMQ Fluorinated vinylmethyl silicone elastomer
HFC Hydrofluorocarbon
123
HNBR Hydrogenated nitrile rubber
HVAC High-volume air conditioning
ID Internal diameter
L Length of capillary flow
Mc Critical molecular weight for entanglement
MECE Mutually exclusive and collectively exhaustive
Mn Number average molecular weight
Mw Weight average molecular weight
MSFE Minimum significant factor effect
NBR Nitrile rubber
NR Natural rubber
OD Outer diameter
phr Parts per hundred rubber polymer
Pt Total pressure drop in capillary
Q Volumetric flow through capillary
R Radius of flow in capillary
SAF Super-abrasion furnace black
SBR Styrenebutadiene rubber
Tg Glass transition temperature
TOTM Trioctyl trimellitate
VMQ Vinylmethyl silicone elastomer
XHNBC Cross-carboxylated HNBR
Shear rate
124
Abbreviations
P Pressure drop
Volume fraction in a mixture
2 Volume fraction of polymer
Viscosity
Shear stress
125
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
Compression modulus 33-34

Compression moulding 24, 27, 93-96, 118, 121
Compression set 34, 47, 64, 66, 68-70, 79-81, 86, 90, 93-94, 97
Contour plot 67-68, 80, 89
Conveyor 56-57
belt 16, 57
component 56
Copolymerisation 11
D
DBP number 32-34, 72
Desirability function 83, 85, 91-92
Dibutoxyethoxyethyl adipate 69
Dicumyl peroxide 47-48, 69, 87
Die swell 24-25, 115
Dimensional shift 113, 116
Dioctyl phthalate 76
Dynamic ozone resistance 78, 97, 103-104
E
Einstein equation 32
Elasticity 24, 29, 32, 34, 52, 95
Emulsion polymerisation 1-3, 5-9
EPDM 10, 17, 29, 31, 46, 56, 79, 85, 87-93
Ether ester 76
Ethylene 9-12, 17, 19, 29, 46, 56, 79, 87, 93
Extrusion 34, 95, 103-104, 116, 118
F
Fatty amide 117-118
Filler 20, 32, 34-35, 63, 73-75, 78, 80-81, 100, 105, 119-120
FKM 17, 19-20, 36, 46, 51-53, 56, 85, 93
Flexible boots 57
Full factorial design 83, 94, 96, 98, 100, 102-103, 105, 117
FVMQ 17, 19-20, 46, 53, 56, 85
G
Glass transition temperature 10, 26
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
N762 46-48, 87, 94, 98

N990 32, 81
NBR 1-3, 5-11, 17, 19-20, 29, 32-33, 43-49, 51-53, 55-59, 63, 73, 75, 78-81, 85,
93, 95, 99, 101
Newtonian fluid 21
Non-black fillers 73-74, 82
Non-fills 107-108, 112
NR 17, 46, 56, 85
Nulok 321 74, 100
Nulok 390 74-75
O
O-rings 59-61, 78, 93, 114, 118, 121
Oil cooler O-ring 93-94
Oil well packer 96
Oxidation 17, 45, 49, 55
-reduction 1
Ozone resistance 78-79, 84, 97-100, 103-104
P
Packer elements 60
Paper mill roller 105
Paraffin wax 87, 117-118
Peroxide 9, 43, 47-48, 52-53, 63, 65-70, 78-82, 87-92, 94, 96, 99, 102, 119
Petroleum exploration 51-52, 60
phr 32-36, 46-48, 65-70, 72, 74-81, 87-92, 94, 96, 98, 100, 102-103, 105, 117
Plackett-Burman screening design 117-120
Plasticiser 35-36, 63, 75-79, 81, 87
Polymerisation 1-9, 11, 31
Post-cure 69-70, 119-121
Pressure drop 37-38, 110
Process aids 36-37
Propylene 9-12, 29, 46, 56, 79
Pseudoplastic 22-23, 108
Pyrax B 74
R
Radial shaft seals 59-60
Reactivity ratios 4-9
Recipe 33, 46-48, 64-66, 69-70, 73, 75-77, 80-81, 87, 93, 97, 99, 101, 103
130
Index
Residual unsaturation 9-10, 13, 63, 81, 98, 100, 102

Response surface 34, 66, 70, 83, 93, 95
Rollers 57, 104
Roy-Cal L 74
Runners 37-38, 111
S
Serpentine belt 97-98
Shaft seals 59-60, 101
Shear 13, 21-25, 27-28, 31-33, 35-38, 73, 108-110, 114-115
Shear rate 21-25, 27-28, 31-33, 35-36, 38, 108-110, 114-115
Shear stress 21, 24-25, 37-38, 110
Silene 732D 74
Sprues 37-38, 111
Static ozone resistance 104
Static seals 58-61
Statistical experimental design 19, 82-83, 85, 92-93, 99, 103, 105, 107, 112, 115,
122
Stearic acid 33, 47-48, 64-65, 69-70, 75, 80-81, 87, 94, 96, 98, 100, 102-103
Steering system hose 102-103
Sulfur 9, 13, 17, 49, 52, 58, 63-64, 69-70, 101
T
Tensile strength 47, 64-66, 69
Tetramethyl thiuram disulfide 69
Tg 10, 24, 26
Therban 15-16
Time-temperature superposition 26
Transfer moulding 24, 101
Translink 77 74
Trioctyl trimellitate 78, 81, 87
U
Ultimate elongation 47, 64-66, 68-69
Ultra-Pflex 74
Ultrasil VN-3 SP 74
V
Viscosity 2, 13-14, 16, 21-23, 26-33, 35-36, 38-39, 47, 72-73, 76-79, 87, 93, 100,
103, 108-111, 114
131
VMQ 17, 46, 56, 85

Volume fraction 32, 35, 73
Volumetric flow 37-38, 110-111
W
Williams-Landel-Ferry relationship 26
Z
Zeeospheres 200 74, 81
Zeolex 23 74
Zeon Chemicals 13, 15-16, 20, 51, 54, 63, 78, 82
Zetpol 15-16
Ziegler-Natta-catalyst 29
Zinc methacrylate 16, 98, 104-105
Zinc oxide 33, 47-48, 64-65, 69-70, 75, 80-81, 87, 94, 96, 98, 100, 102-103
abcd
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.
Shawbury, Shrewsbury, Shropshire, SY4 4NR, UK

Telephone: +44 (0)1939 250383
Fax: +44 (0)1939 251118
Web: www.polymer-books.com

Group Presentation English Advancedy - Robert Keller (Smithers Rapra Technology, 2012)

Загружено:

Сведения о документе

Авторское право

Доступные форматы

Поделиться этим документом

Поделиться или встроить документ

Параметры публикации

Этот документ был вам полезен?

Это неприемлемый материал?

Авторское право:

Доступные форматы

Group Presentation English Advancedy - Robert Keller (Smithers Rapra Technology, 2012)

Загружено:

Авторское право:

Доступные форматы

Practical Guide to

A Smithers Group Company

Shawbury, Shrewsbury, Shropshire, SY4 4NR, United Kingdom

Smithers Rapra Technology Ltd

2012, Smithers Rapra Technology Ltd

All rights reserved. Except as permitted under current legislation no part

ISBN: 978-1-84735-521-8 (softback)

Typeset by Argil Services

1 Overview of the Chemistry and Manufacture of

2 Types of Hydrogenated Nitrile Rubber Polymers Available........ 13

3 The Flow and Processing Behaviour of Hydrogenated

3.4 Effect of Compounding Variables on Viscosity,

4 Properties of Hydrogenated Nitrile Rubber and

5 Typical Applications of Hydrogenated Nitrile Rubber................ 55

6 Formulating Guidelines for Hydrogenated Nitrile Rubber.......... 63

6.2 Vulcanisation Systems....................................................... 63

7 Examples of Hydrogenated Nitrile Rubber Formulae for

7.10 Compound Example: Chemically Resistant,

8 Solving Hydrogenated Nitrile Rubber Processing Issues........... 107

Index ................................................................................................ 127

The manufacture and chemistry of HNBR elastomers.

The commercial grades currently available.

How HNBR fits into the current world of speciality elastomers.

The basics of compounding and formulating HNBR.

Examples of compounding for specific applications.

Its rheological properties and how these can be applied.

Problem solving techniques using specific examples.

This introductory chapter reviews the manufacture of hydrogenated nitrile butadiene

1.2 The Starting Polymer: Nitrile Butadiene Rubber

Nitrile rubber, ASTM D1418 designation NBR, is a copolymer of ACN and

The polymerisation is a typical emulsion polymerisation, which has the following

The monomers are usually contained in emulsified droplets in the continuous

Initiator = free radical

Monomers in emulsified oil phase

Typical size of emulsified phase = 100 nm

CH 2=CH---CH=CH 2 + CH 2=CH >>>>> ---(CH2 ----CH=CH----CH2)x-(CH2----CH)y -(CH 2 =CH ) Z-

Typical composition of resulting NBR polymer =

Figure 1.1 Emulsion free-radical polymerisation of NBR

Final polymer bales

Figure 1.2 Schematic showing NBR polymerisation

M1* + M1 ----- M1M1*, reaction rate = k11 (1.1)

M1* + M2 ----- M1M2*, reaction rate = k12 (1.2)

M2* + M2 ----- M2M2*, reaction rate = k22 (1.3)

M2* + M1 ----- M2M1*, reaction rate = k21 (1.4)

r1 = k11/k12 = reactivity ratio for monomer 1 (1.5)

r2 = k22/k21 = reactivity ratio for monomer 2 (1.6)

The molar concentration of monomers at any instant during the polymerisation is

The limiting cases in Equation (1.1) are as follows:

r1 = r2 = approximately 1. The reactive chain ends, whether monomer 1 or

r1 = r2 = approaching 0. If monomer 1 forms the reactive chain end it has little

F1 = instantaneous mole fraction of monomer 1 in the instantaneous polymer

f1 = mole fraction of monomer 1 in the instantaneous monomer feed, and

f2 = mole fraction of monomer 2 in the instantaneous monomer feed.

% ACN in polymer, 75/25

Figure 1.3 Example of polymer ACN versus conversion

In this case, the incremental (instantaneous) polymer composition early in the

%ACN in instantaneous polymer

Figure 1.4 ACN content of polymer versus monomer feed

This understanding of the NBR polymer formed by emulsion polymerisation is

1.3 Hydrogenation of Nitrile Butadiene Rubber to Produce the

Bales of HNBR polymer

A catalyst which facilitates hydrogenation is normally employed. A typical

catalyst [3], in which Cy denotes a cyclohexyl residue. The hydrogenation reaction

Figure 1.6 Catalytic hydrogenation of NBR