published by smithers rapra technology ltd, 2012 rubber...
TRANSCRIPT
Shawbury, Shrewsbury, Shropshire, SY4 4NR, UK Telephone: +44 (0)1939 250383 Fax: +44 (0)1939 251118 Web: www.polymer-books.com
Practical Guide to H
ydrogenated Nitrile B
utadiene Rubber TechnologyRobert Keller
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 author’s 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 author’s 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.
Published by Smithers Rapra Technology Ltd, 2012
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
Robert Keller
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
Typeset by Argil Services
ISBN: 978-1-84735-521-8 (softback)
978-1-84735-522-5 (ebook)
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.
A catalogue record for this book is available from the British Library.
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.
©2012, Smithers Rapra Technology Ltd
Smithers Rapra Technology LtdShawbury, Shrewsbury, Shropshire, SY4 4NR, UK
iii
Contents
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
iv
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
Contents
v
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 Ethylene–Propylene 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
vi
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
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
vii
Preface
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:
• ThemanufactureandchemistryofHNBRelastomers.
• Thecommercialgradescurrentlyavailable.
• HowHNBRfitsintothecurrentworldofspecialityelastomers.
• ThebasicsofcompoundingandformulatingHNBR.
• Examplesofcompoundingforspecificapplications.
• Itsrheologicalpropertiesandhowthesecanbeapplied.
• Problemsolvingtechniquesusingspecificexamples.
Much of the information will be found applicable not only to HNBR but also to other elastomers.
viii
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
1
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:
• Themonomersareusuallycontained inemulsifieddroplets in thecontinuousphase. 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.
• Theinitiator,inthiscaseafree-radicalspecies,isgenerallyquitesolubleinthecontinuous 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 oxidation–reduction reaction. Other NBR polymers are formed at elevated
1 Overview of the Chemistry and Manufacture of Hydrogenated Nitrile Butadiene Rubber Polymers
2
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
temperatures (typically around 95 °C) and are known as ‘hot’ polymers. Since the free-radical initiator is soluble in the continuous phase but only partially soluble in the emulsified phase, free radicals diffuse slowly into the emulsified phase. In contrast to bulk or solution free-radical polymerisation, in which a relatively high level of free radicals occur with the chain-growth polymer and cause early termination of polymerisation, emulsion polymerisation is diffusion-controlled and chain growth gives very high molecular weights. Chain transfer agents such as organo-mercaptans may be added to the emulsified phase to control the molecular weight of the polymer to give the desired bulk viscosity.
• Anemulsifier,whichmaybesimplyasodiumstearatesoap,assuresfinedispersionand emulsification of the monomer/polymer-rich phase.
• Thecontinuousphaseactsasanexcellentheattransfermediumintheexothermicpolymerisation 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.
• Theviscositydoesnotincreaseduringtheconversion,asisthecaseinbulkorsolution polymerisation. The viscosity remains essentially constant at the viscosity of the continuous phase.
R*
Initiator = free radicalOnly partially soluble in ‘oil’ phase
Monomers in emulsified ‘oil’ phasePolymer forms here
Typical size of emulsified phase = 100 nm
CH2=CH---CH=CH2 CH 2=CH >>>>> ---(CH2 ----CH=CH----CH2)x-(CH2----CH)y-(CH 2=CH )Z-| | |CΞ 2 CΞN
Typical composition of resulting NBR polymer = X = 84-50%, Y = 5% of butadiene incorporated, Z = 16-50%
+
N CH=CH
Figure 1.1 Emulsion free-radical polymerisation of NBR
3
Overview of the Chemistry and Manufacture of Hydrogenated Nitrile Butadiene Rubber Polymers
In general, NBR polymers contain between 18 and 50 mole% ACN content, the remainder being butadiene. A crude schematic of the emulsion polymerisation is shown in Figure 1.2.
Polymerisation Coagulation
Emulsifier
Acrylonitrile
Butadiene Initiator Water
Drying and dewateringFinal polymer bales
Figure 1.2 Schematic showing NBR polymerisation
NBRarecopolymersandfollowtheMayo–Lewiscopolymerequation[1].Sincetwomonomers, ACN and butadiene, are involved the active terminal grouping in the growing chain can undergo the following possible reactions:
M1* + M1 ----- M1–M1*, reaction rate = k11 (1.1)
M1* + M2 ----- M1–M2*, reaction rate = k12 (1.2)
M2* + M2 ----- M2–M2*, reaction rate = k22 (1.3)
M2* + M1 ----- M2–M1*, reaction rate = k21 (1.4)
4
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
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 givenbyEquation(1.7):
d[M1]
d[M2]
[M1]*(r1[M1]+[M2])
[M2]*([M1]+r2[M2])= (1.7)
where[M1]and[M2]arethemolarconcentrationsofmonomer1andmonomer2,respectively.
ThelimitingcasesinEquation(1.1)areasfollows:
• r1=r2>>1.Thetwomonomerreactivityratiosarehigh,indicatingthattheactivated 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.Whenmonomer1formsthereactivechainenditwillbemorelikelyto 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,whethermonomer1ormonomer 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=approaching0.Ifmonomer1formsthereactivechainendithaslittlelikelihood 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….).
5
Overview of the Chemistry and Manufacture of Hydrogenated Nitrile Butadiene Rubber Polymers
• r1>>1>>r2.Atthebeginningofthepolymerisationmonomer1hasamorereactive 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<1and/or0<r2<1.Thereactiontendstowardshomopolymerisation,and this tendency increases with decreasing rX.
WecanalsorearrangetheMayo–Lewisequationtoexpresstheratioofmonomer1 in the instantaneous polymer with respect to the monomer concentrations and reactivity ratios:
F1[r1f12 + f1f2]
[r1f12 + 2f1f2 + r2f22]= (1.8)
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.
Againstthisbackground,thequestionarisesastothevaluesofr1andr2fortheACNand 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 byEmbreeandco-workers[2].TheemulsionpolymerisationofNBRpolymershasa number of complicating factors:
• ACNhas reasonably high solubility in the aqueous continuous phase.ACNmonomer will therefore be present in both the emulsified and the continuous phase.
• TheresultingNBRpolymerisnotsolubleinbutadiene,andthismayresultinanon-homogeneous emulsified phase.
6
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
In their study of polymer composition and molecular weight, Embree and co-workers used previously published data in addition to a number of their own experiments. Their analyses indicated that r1 (butadiene) was 0.18 and r2 (ACN) was 0.03. In this case,r1>>r2,andbothr1andr2weremuchlowerthan1.0.Ifthereactivechainendwas 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.
Wecanalsouse thedataofEmbreeandco-workers [2] tounderstandtheeffectof 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 75/25molarchargeofbutadiene/ACNmonomersinthemixture,theirdatamaybereplotted as shown in Figure 1.3.
40
30
20
10
00 20 40
% Conversion
% ACN in polymer, 75/25Butadiene/ACN Charge mixture
% ACN incrementalpolymer
% ACN in polymer mix
% A
CN
60 80 100
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 30–35 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,
7
Overview of the Chemistry and Manufacture of Hydrogenated Nitrile Butadiene Rubber Polymers
since the remaining monomer will have to be stripped out of the polymer. From Figure 1.3,apracticalbalanceforthisblendofmonomersmightgive70–80%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 andresourcerequiredtostripresidualmonomerfromtheresultingemulsionmixture.WecanalsouseEquation(1.3)todeterminethemole%ACNintheinstantaneouspolymer formed as a function of the mole% composition of monomer (Figure 1.4):
0%10%20%30%40%50%60%70%80%90%
100%
0% 20% 40% 60% 80% 100%
poly
mer
NC
A%
%ACN in monomer feed
%ACN in instantaneous polymer
%ACN in polymer
Ideal
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:
• Thereactivityratiosofthetwomonomers,butadieneandACN,indicatethatthe 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.
8
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
• AthighconversionvaluesthecompositionoftheNBRpolymerchanges,andthe resulting bulk NBR polymer may well be a blend of a range of polymers of different ACN content.
• The reactivity ratios ofACNand butadiene tend to favour the presence ofalternating 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.
NBR Polymer
Solvent
Dissolution
Hydrogen
High pressurecatalytic hydrogenation
Separation fromsolvent and solventremoval
DryingBales of HNBR polymer
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
9
Overview of the Chemistry and Manufacture of Hydrogenated Nitrile Butadiene Rubber Polymers
catalyst[3],inwhichCydenotesacyclohexylresidue. The hydrogenation reaction is illustrated in Figure 1.6.
----{CH2—CH}x----{CH2—CH=CH—CH2}y---{CH2—CH}z----
CN CH=CH2
acrylonitrile 1,4-butadiene addition
1,2-butadiene addition, vinyl residue
NBR polymer
Metal catalysts, high pressure
+ (2x+2y) H2 (gas)
----{CH2—CH}x----{CH2—CH2 –CH2 —CH2}y ---{CH2—CH=CH—CH2}a ---{CH2—CH}z----
CN CH2 – CH3
acrylonitrile ethylene propylene
HNBR polymer
residual unsaturationfor vulcanisation
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:
• Thereactivityratiosofthetwomonomers,butadieneandACN,indicatethatthe 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 ethylene–propylene polymer of high ethylene content.
• Athighconversionvalues,thecompositionoftheNBRpolymerchanges,anditmay well comprise a blend of polymers of different ACN content.
These segments of ethylene–propylene polymer of high ethylene content create major challenges inthemanufactureofHNBRpolymersof lowACNcontent,required
10
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
when the end application demands improved low temperature flexibility. However, HNBR polymers of low ACN content produced from conventional NBR polymers actually have inferior low temperature flexibility to HNBR polymers of medium or high ACN content. The tendency for ethylene–propylene segments of high ethylene content to crystallise is an example of this. In the author’s laboratories the glass transition temperatures (Tg) of ethylene–propylene 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.
040 45 50 55 60 65 70 75 80
Tg (ºC)
linear Tg (ºC)
y = 1.2688 x – 102.79R2 = 0.9679
–10
–20
–30Tg
Mole% ethylene
Tg (ºC) of low vinyl EPDM
–40
–50
–60
Figure 1.7 Tg of ethylene–propylene 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 ethylene–propylene 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 ethylene–propylene 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.
11
Overview of the Chemistry and Manufacture of Hydrogenated Nitrile Butadiene Rubber Polymers
The challenge for manufacturers of HNBR polymers is thus to disrupt the regularity of the high ethylene content segments of the resulting HNBR polymer. This can be achieved by introducing a third diene monomer into the initial NBR polymerisation, since this inhibits the tendency of the ethylene segments to crystallise. An example is the emulsion copolymerisation of ACN, 1,3-butadiene and isoprene, with 1,3-butadiene/isopreneratioslyingwithintherange0.75/1.0to1.0/0.75[4].Thependantmethylgroup on the isoprene monomer then disrupts the regularity and crystallisation of the resultinghydrogenatedpolymer.Thereactionsequenceandfinalpolymerstructureare indicated in Figure 1.8.
----{CH2—CH}x----{CH2—CH=CH—CH2}y----{CH2—CH=CH—CH2}y--{CH2—CH}z----
CN CH3 CH=CH2
arylonitrile 1,4-butadiene addition
1,2-butadiene addition, vinyl residue
NBR Polymer
Metal catalysts, high pressure+ (2x+2y) H2 (gas)
----{CH2—CH}x----{CH2—CH2 –CH2 —CH2}y ---{CH2—CH=CH—CH2}a ---{CH2—CH}z----
CN CH2 –CH3
acrylonitrile ethylene propylene
HNBR Polymer
residual unsaturationfor vulcanisation
1,4-isopreneaddition
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 startingpolymerwouldresultinaroughly73/27molarratioofethylene/propylenein 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):
12
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
• 1moleof1,3-butadiene=0.95molesethylene+0.05molespropyleneinHNBRpolymer.
• 1moleofisoprene=0.5molesethylene+0.5molespropyleneinHNBRpolymer.
• Totalmolesofethylene=0.95+0.5=1.45.
• Totalmolesofpropylene=0.05+0.5=0.55.
• Ethylene/propyleneratiointhefinalHNBRpolymer=1.45/(1.45+0.55)=0.73.
• Ethylene/propylene ratio in thefinalHNBRpolymerwithno isoprene in thestarting polymer = 0.95.
The reduced ethylene content in the ethylene–propylene 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.MitchellandH.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.GuoandJ.Gamlin,inventors;BayerAG,assignee,USPatent7,091,284,2006.
13
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.
2 Types of Hydrogenated Nitrile Rubber Polymers Available
14
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
Table 2.1 HNBR grades commercially available, December 2010
Grade Manufacturer Mole% ACN
(nominal)
Mooney viscosity ML(1+4) at 100°C (nominal)
Residual % unsaturation
(approximate)
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
Zetpol 3310EP Zeon 25 30 5
Zetpol 3310 Zeon 25 80 5
Therban AP A 3404 Lanxess 34 39 0.9
Therban 3406 Lanxess 34 63 0.9
Therban 3407 Lanxess 34 70 0.9
Therban AT C 3443 VP Lanxess 34 39 4
Therban 3446 Lanxess 34 61 4
Therban 3467 Lanxess 34 68 5.5
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
15
Types of Hydrogenated Nitrile Rubber Polymers Available
Therban 3607 Lanxess 36 66 0.9
Therban 3627 Lanxess 36 87 2
Zetpol2010L Zeon 36 58 4
Zetpol 2010 Zeon 36 85 4
Zetpol 2010H Zeon 36 135 4
Therban AT A 3904 VP Lanxess 39 39 0.9
Therban 3907 Lanxess 39 70 0.9
Therban AT 4304 VP Lanxess 43 39 0.9
Therban 4307 Lanxess 43 63 0.9
Therban 4309 Lanxess 43 100 0.9
Therban AT 4364 VP Lanxess 43 39 5.5
Therban 4367 Lanxess 43 61 5.5
Therban 4369 Lanxess 43 97 5.5
Zetpol 1000L Zeon 44 65 2
Zetpol 1010EP Zeon 44 29 4
Zetpol 1010 Zeon 44 85 4
Zetpol 1020EP Zeon 44 30 9
Zetpol 1020L Zeon 44 57 9
Zetpol 1020 Zeon 44 78 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
Zetpol 0020EP Zeon 50 40 9
Zetpol 0020 Zeon 50 65 9
Therban® is a registered trademark of Lanxess Zetpol® is a registered trademark of Zeon Chemicals
16
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
In addition to these grades, speciality polymers are available for specific applications and end-products. These are summarised in Table 2.2.
Table 2.2 HNBR speciality grades available
Grade Manufacturer Mole% ACN
(nominal)
Mooney viscosity ML(1+4) at 100 °C (nominal)
Residual % unsaturation
(approximate)
Modification
Zeoforte ZSC 2295CX
Zeon 36 95 9 Zinc methacrylate
Zeoforte ZSC 2295L
Zeon 36 80 9 Zinc methacrylate
Zeoforte ZSC 2385
Zeon 36 70 15 Zinc methacrylate
Therban XT VP KA 8889
Lanxess 33 77 3.5 Carboxylated (XHNBR)
Therban VP KA 8796
Lanxess 34 22 5.5 Acrylate
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.
17
Types of Hydrogenated Nitrile Rubber Polymers Available
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
0
50
100
150
200
250
300
050100150200
% Swell in ASTM IRM–903 Oil
Tem
per
atur
e re
sist
ance
(ºC
)
NRCR
AEM
VMQ FVMQ
FKM
NBR
ACM
Class by Oil Swell
EPDMHNBR
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 ºC 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
18
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
end of their useful life before the end of the ageing test. The ACM material showed excellent long-term ageing performance in these fluids, but did not have the tear and abrasion resistance of HNBR materials.
–60–50–40–30–20–10
010203040
35% ACN HNBR
25% ACN HNBR
ACM AEM
% E
lon
gati
on
Lo
ss
Material
Comparison of Oil-ResistantElastomers, 10W30SG engine oil at 150 °C
% Elongation Loss, 504 h
% Elongation Loss, 1008 h
Figure 2.2 Comparison of engine oil resistance of HNBR, ACM and AEM elastomers
–30–25–20–15–10
–505
10152025
35% ACN HNBR
25% ACN HNBR
ACM AEM
% E
lon
gati
on
Lo
ss
Material
Comparison of Oil-resistant
Elastomers, Dexron III ATF at 150 °C
% Elongation Loss, 504 h
% Elongation Loss, 1008 h
Figure 2.3 Comparison of resistance to automatic transmission fluid (ATF) of HNBR, ACM and AEM elastomers
19
Types of Hydrogenated Nitrile Rubber Polymers Available
The relative abrasion resistance of HNBR materials compared to other oil-resistant elastomers, such as NBR, epichlorohydrin–ethylene 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
0
50
100
150
200
45% ACNHNBR
NBR ECO FKM ACM
Material
Rel
ativ
e re
sist
ance
to
ab
rasi
on
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 pound–volume 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 ºC, HNBR are much tougher and more abrasion-resistant. FKM and fluorosilicones (FVMQ) have excellent upper temperature limits, but their pound–volume cost is heavily influenced by their relatively high specific gravity.
20
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
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 ºC.
Table 2.3 Commercial comparison of oil-resistant elastomers
Polymer SG Cost of compound ($)
lb–vol cost ($) Operating temperature, upper limit (ºC)
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.
21
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 s–1.
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
3 The Flow and Processing Behaviour of Hydrogenated Nitrile Rubber
22
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
as this. However, polymer melts such as rubbers do not in general exhibit uniform viscosity across a range of shear rates, in fact they are described as pseudoplastic, which means that their viscosity is reduced as the shear rate increases.
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.
1000
10000
100000
1000000
1 10 100 1000
Shear rate (s–1)
Vis
cosi
ty (
poi
se)
Newtonian region
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
23
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 [1–11], and Equations (3.2) and (3.3) provide a starting point:
h = Kgn–1, (3.2)
where h is the viscosity, g is shear rate in s–1 and K and n are a curve-fitting parameters determined from the experimental data.
h = ho/(1 + Ag1–n), (3.3)
where ho 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.
Vis
cosi
ty
Shear rate
100000
10000
10001 10 100 1000
Figure 3.2 Comparison of theoretical models describing the pseudoplastic behaviour of polymers
24
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
The shear rate in compression moulding is normally within the range 1–10 s–1, for transfer moulding in the range 10–1,000 s–1, and for injection moulding in the range 500–10,000 s–1. 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 + ½(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
25
The Flow and Processing Behaviour of Hydrogenated Nitrile Rubber
rebound induced by increasing shear rate and shear stress. Using Equation (3.4) we can establish the proportionality between elastic recovery and shear stress or shear rate. Figure 3.3 shows the effect of a percentage increase in shear rate on the extrudate swell, B, or elastic recovery.
% D
ie s
wel
l
Shear rate (s–1)
454035302520151050
1 10 100 1000
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 manufacturer’s 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.
26
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
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 Williams–Landel–Ferry relationship for time–temperature 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(hT1/T1) – log(hTg/Tg) = log(aT1), (3.5)
where hT1 is the viscosity at absolute temperature T1, hTg 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(µT2/T2) – log(µT1/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 ºC. 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.
27
The Flow and Processing Behaviour of Hydrogenated Nitrile Rubber
1000000
100000
10000
1000
100100
100 °C
125 °C
150 °C
1000101
Vis
cosi
ty (
Pa. s
)
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 = 0∫t (µ/rc)g2 dt, (3.8)
where DT is the temperature change in ºC s–1, µ is the viscosity in Pa. s and g is shear rate in s–1.
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)
28
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
Using these proportionalities we can plot the effect of viscosity and shear rate on heat build-up, as illustrated in Figure 3.5.
350%300%250%200%150%100%50%0%
–50%–100%
–100% –50%
% change in parameter
0% 50% 100% 150%
% c
hang
e in
hea
t bu
ild-u
p
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.
29
The Flow and Processing Behaviour of Hydrogenated Nitrile Rubber
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 = (ΣNiMij)/(ΣNiMi
j–1), (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 styrene–butadiene 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 Ziegler–Natta-catalysed polyethylene or ethylene–propylene diene rubber (EPDM).
30
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
0
0.1
0.2
0.3
0.4
0.5
0 100000 200000 300000 400000 500000 600000 700000
Molecular weight
Fra
ctio
n
Mn Mw MzMv
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 [14–22]. 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 (carbon–carbon backbone) polymers it is dependent on the bulk of the pendant groups. The value of Mc for polyethylene (a carbon–carbon backbone with no bulky pendant groups) is approximately 4,000, while that for polystyrene (carbon–carbon 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):
31
The Flow and Processing Behaviour of Hydrogenated Nitrile Rubber
ho = kMw3.4, (3.12)
where ho is zero shear rate viscosity and k is a constant for a particular polymer system.
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 s–1, and this is very close to the zero shear rate viscosity, it should show proportionality to Mw
3.4. 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 s–1 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.
10
100
1000
10000
100000
1000000
1 10 100 1000
Shear rate (s–1)
Vis
cosi
ty (
pois
e)
Molecular weight = 80,000Mooney viscosity = 23
Molecular weight = 120,000Mooney viscosity = 93
Molecular weight = 100,000Mooney viscosity = 50
Figure 3.7 Effect of molecular weight on flow behaviour
32
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
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
33
The Flow and Processing Behaviour of Hydrogenated Nitrile Rubber
ASTM D2414, combines the effect of particle size and structure. The DBP number represents the quantity of dibutyl phthalate (DBP) that the carbon black can absorb, measured by titration to a mechanically sensed end-point; when the carbon black surface can absorb no more DBP the conversion from free-flowing powder to paste is sensed as a change in torque. The experimental design programme is summarised in Table 3.1 and the results for compression modulus and low-shear rate viscosity are shown in Figure 3.8.
Table 3.1 Experimental design formulations, NBR compound, triplicate data points for each formula basic recipeIngredient 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
34
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
Response surface for compression modulus in Pa
DBP number
Com
pres
sion
mod
ulus
phr Carbon black
15.5
13.5
11.5
9.5
7.5
5.5
30 40 50 60 700
3060
90120
150
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.
35
The Flow and Processing Behaviour of Hydrogenated Nitrile Rubber
Hysteresis as a function of filler and loading
phr Carbon Black
DBP Number
% H
yste
resi
s, C
ompr
essi
on
30 40 50 60 70 30507090
11013023
33
43
53
63
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 (gf2
1.4), (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(gf1
1.4)/ (f03.4F(gf0
1.4) = f14.8/f0
4.8 (3.15)
36
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
As more plasticiser is added, viscosity at a given shear rate falls. This is plotted in Figure 3.10, using our example from Figure 3.1. For a particular polymer and plasticiser system it should be straightforward to run the experiments and adapt the parameters of Equation (3.15) to the design of other compounds.
1000
10000
100000
1000000
1 10 100 1000
Vis
cosi
ty,
Pa. s
Shear rate (s–1)
Phi = 1
Phi = .9
Phi = .8
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.
37
The Flow and Processing Behaviour of Hydrogenated Nitrile Rubber
Techniques are available for studying the effect of process aids on mould flow, for example a capillary rheometer may be useful for this purpose. The pressure drop through the die of a capillary rheometer is made up of the total pressure drop at the ends plus the pressure drop within the die:
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
38
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
reasons the default diagnosis in elastomer production needs to be to fix this type of problem by changing or reformulating the rubber material. However, when we look at the underlying science this would seem to be the least effective solution to flow problems in production processing.
Viscosity is simply the ratio of shear stress and shear rate. For flow through a round capillary this can be expressed as follows:
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 / 4happ . 2L. (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.
39
The Flow and Processing Behaviour of Hydrogenated Nitrile Rubber
Radius of runner Length of runner Viscosity of material
% Change in factor
Net
% c
hang
e on
flo
w r
ate 120.0%
100.0%80.0%60.0%40.0%20.0%0.0%
–20.0%
–20.0% –10.0% 0.0% 10.0% 20.0% 30.0%
–40.0%–60.0%
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.
40
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
10. J.L. White and N. Tokita, Journal of Applied Polymer Science, 1974, 9, 1929.
11. W.E. Wolstenholme, Rubber Chemistry and Technology, 1965, 38, 769.
12. R.I. Tanner, Journal of Polymer Science, Part A2, 1970, 8, 2067.
13. M.L. Williams, R.F. Landel and J.D. Ferry, Journal of the American Chemical Society, 1955, 77, 3701.
14. P.J. Flory, Journal of the American Chemical Society, 1940, 62, 1057.
15. J.R. Shaefgen and P.J. Flory, Journal of the American Chemical Society, 1948, 70, 3709.
16. T.G. Fox and P.J. Flory, Journal of the American Chemical Society, 1949, 70, 2384.
17. T.G. Fox and V.R. Allen, Journal of Chemical Physics, 1964, 41, 344.
18. T.G. Fox and S. Loschaek, Journal of Applied Physics, 1955, 26, 1082.
19. W.F. Busse and R. Longworth, Transactions of the Society of Rheology, 1962, 6, 179.
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.
41
The Flow and Processing Behaviour of Hydrogenated Nitrile Rubber
28. C.D. Han, Journal of Applied Polymer Science, 1974, 18, 821.
29. N. Minagawa and J.L. White, Applied Polymer Science, 1975, 20, 501.
30. G. Kraus and J.T. Gruver, Transactions of the Society of Rheology, 1965, 9, 2, 17.
42
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
43
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 IRM–903 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 ºC test temperature, 40% or less volume swelling in IRM–903 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 IRM–903). 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.
4 Properties of Hydrogenated Nitrile Rubber and Comparison with Other Elastomers
44
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
Table 4.1 ASTM D2000 type and class system
Type Test temperature (°C)
Class %Volume swelling in IRM–903 oil 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
0
50
100
150
200
250
300
050100150200
Tem
pera
ture
res
ista
nce
(°C
)
% Swell in IRM–903 Oil
ASTM D2000 Classification
NBR
A B C D E F G H J K
Class by Oil Swell
ABCDEF
H
Typ
e by
tes
t te
mpe
ratu
re
Figure 4.1 Graphical representation of NBR in the ASTM D2000 system
45
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.
0
50
100
150
200
250
300
050100150200
Tem
pera
ture
res
ista
nce
(°C
)
% Swell in IRM–903 Oil
ASTM D2000 Classification
NBR
A B C D E F G H J K
Class by Oil Swell
ABCDEF
H
Typ
e by
tes
t te
mpe
ratu
re
HNBR
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.
46
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
0
50
100
150
200
250
300
050100150200
Tem
per
atur
e R
esis
tan
ce (
°C)
% Swell in ASTM IRM–903 Oil
ASTM D2000 Classification
NRCR
AEM
VMQ FVMQ
FKM
NBR
ACM
A B E F G H J K
Class by Oil Swell
EPDM
A
BCD
EF
H
HNBR
Note : NR = natural rubber, EPDM = ethylene–propylene diene rubber, CR = chloroprene rubber,
VMQ = vinyl–methyl silicone, ACM = acrylic, AEM = ethylene–acrylic, 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.
47
Properties of Hydrogenated Nitrile Rubber and Comparison with Other Elastomers
Table 4.2 Comparison of NBR and HNBR
Polymer A: HNBR polymer; ACN: 36%; nominal unsaturation: 4%; viscosity ML(1+4) at
120 ºC: 85 Mooney units
Polymer B: NBR polymer, cold polymerised; ACN: 33%; viscosity ML(1+4) at 100 ºC: 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: Recipe A (NBR)
Results: Recipe B (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 ºC 26.5 9.8
Compression set, ASTM D395, Method B: piled discs, 72 h at 150 ºC 32.6 10.5
Compression set, ASTM D395, Method B: piled discs, 1008 h at 150 ºC 79.8 32.1
% Volume change, IRM–901 oil, 70 h at 150 ºC, ASTM D471 –0.2 +0.3
% Volume change, IRM–902 oil, 70 h at 150 ºC, ASTM D471 +3.8 +10.1
% Volume change, IRM–903 oil, 70 h at 150 ºC, ASTM D471 +14.8 +16.8
% Volume change, ASTM Reference Fuel A, 70 h at 23 ºC, ASTM
D471
+1.5 +4.8
% Volume change, ASTM Reference Fuel B, 70 h at 23 ºC, ASTM
D471
+25.6 +45.2
% Volume change, ASTM Reference Fuel C, 70 h at 23 ºC, ASTM
D471
+32.1 +54.8
48
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
Since the HNBR recipe contained a slightly higher molar percentage of ACN this is not an exact comparison, but it is interesting that HNBR showed noticeably greater fluid swelling than NBR. From these data we can see that the hydrogenation of NBR to form HNBR results in:
• Betteroxidativethermalstability.
• Increasedswellinginhydrocarbons.
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 ºC 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 (phr)
Recipe B (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 antioxidant
1 1
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 ºC 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.
49
Properties of Hydrogenated Nitrile Rubber and Comparison with Other Elastomers
Mole % ACN
90
80
70
60
50
40
30
20
10
0
0 10 20 30 40 50 60–10
NBR
HNBR
Linear (HNBR)
Linear (NBR)
% V
olum
e ch
ange
NBRy=–1.9652x+97.386
R2=0.9995
HNBRy=–1.7361x+116.54
R2=0.9996
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 ºC 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.
50
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
-60
-40
-20
0
20
40
35% ACN HNBR
25% ACN HNBR
ACM AEM% E
lon
gati
on
loss
Material
Comparison of oil-resistantelastomers, 10W30SG engine oil at 150 °C
% Elongation Loss, 504 h
% Elongation Loss, 1008 h
Figure 4.5 Comparison of oil-resistant elastomers in loss of elongation in automotive engine oil at high temperature
-30-20-10
0102030
35% ACN HNBR
25% ACN HNBR
ACM AEM% E
long
atio
n lo
ss
Material
Comparison of oil-resistant
elastomers, Dexron III ATF at 150 °C
% Elongation loss, 504 h
% Elongation loss, 1008 h
Figure 4.6 Comparison of oil-resistant elastomers in loss of elongation in automatic transmission fluid (ATF)
51
Properties of Hydrogenated Nitrile Rubber and Comparison with Other Elastomers
An area in which HNBR excels is in its wear and abrasion resistance. This is shown graphically in Figure 4.7 by comparison with other oil-resistant speciality elastomers.
Akron-type abrasion resistance rating
0
50
100
150
200
45% ACNHNBR
NBR ECO FKM ACM
Material
Rel
ativ
e re
sist
ance
to
abra
sio
n
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.
52
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
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: IRM–902 oil containing 1% NACE Amine B corrosion inhibitorTest 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 pound–volume 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.
53
Properties of Hydrogenated Nitrile Rubber and Comparison with Other Elastomers
120
100
80
60
40
20
00 50 100 150 200
36%ACN HNBR, peroxidecured
NBR (41 mole%ACN),Semi-EVSulfur cured
Type 2 FKM (70 mole%Fluorine), peroxide cured
Type 1 FKM (66 mole%Fluorine), Bisphenol cured
FEPM, peroxide cured
45%ACN HNBR, peroxidecured
Ageing time (h)
% E
long
atio
n re
tain
ed
Figure 4.8 Comparison of various elastomers for resistance to sour gas at 23 ºC
Table 4.5 Cost comparison of oil-resistant elastomers
PolymerSpecific gravity of compound
Cost of compound
lb–volume cost
Upper operating temperature limit (ºC)
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.
54
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
References
1. American Society for Testing and Materials, D2000–08 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 SA–13, Applications for Oil Field Drilling, Recovery and Processing, Louisville, KY, USA, 2001.
55
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.
5 Typical Applications of Hydrogenated Nitrile Rubber
56
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
0
50
100
150
200
250
300
050100150200
Tem
per
atu
re r
esis
tan
ce (
°C)
% Swell in ASTM IRM–903 Oil
ASTM D2000 Classification
NRCR
AEM
VMQ FVMQ
FKM
NBR
ACM
A B E F G H J K
Class by Oil Swell
EPDM
A
BCD
EF
H
HNBR
Note : NR = natural rubber, EPDM = ethylene–propylene diene rubber, CR = chloroprene rubber,
VMQ = vinyl–methyl silicone, ACM = acrylic, AEM = ethylene–acrylic, 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.
57
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-reinforcedbeltsforautomotivepoweraccessories:
– Timing belts.
– Drive belts for air conditioning compressors, alternators and hydraulic steering pumps.
– Belt-driven power take-off pumps.
• Conveyorbelting,eitherhomogeneousrubberorfabric-reinforced:
– Acrylate-modified HNBR for conveying abrasive powders, including minerals and grain.
• Rubberrollers:
– 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
58
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
be too high for the thermoplastic elastomers traditionally used. HNBR materials offer a good balance of properties for high temperature protective boot and bellows applications, including:
• Drivetrainbootsinautomotiveandtruckapplications:
– Constant-velocity joints near exhaust headers.
– Axle joints close to exhaust pipes or catalytic converters.
• Steeringboots:
– Rack and pinion ends near exhaust headers.
– Input shafts close to the engine.
• Suspensioncomponents:
– 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.
59
Typical Applications of Hydrogenated Nitrile Rubber
Examples of static seals used in power transmission include:
• O-rings and square cross-section seals for long-life hydraulic power steeringgaskets, pumps and reservoirs:
– Pump seals.
– Fluid tube fitting seals.
– Fluid reservoir seals.
– Rack and pinion fluid seals.
• O-rings,gasketsandcross-sectionsealsforautomaticandmanualtransmissionfluids, 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:
• Inputandoutputradialshaftsinautomaticandmanualtransmissionsystems.
• Inputshaftsandreciprocatingracksinhydraulicpower-assistedrackandpinionsteering systems.
• Radialdifferentialshafts.
• Hydraulicpower-assistedsteeringpumps.
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.
60
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
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 CFC–12/mineral oil lubricants or the newer HFC–134a/polyalkylene glycol systems. HNBR also has good resistance in seals for high-volume air conditioning (HVAC) in buildings, such as HFC–22/mineral oil lubricant systems, making it an excellent choice for these applications:
• Staticseals:
– O-rings.
– Square cross-section seals.
– Gaskets (homogeneous rubber or composite-reinforced).
• Dynamicseals:
– 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:
• Packerelementsandblow-outpreventers.
• O-rings:
– Packer element static seals.
– Tubing seals.
– Pump static seals.
– Drill bit seals.
• Pumpradialshaftseals.
5.9 Other applications
These include:
61
Typical Applications of Hydrogenated Nitrile Rubber
• Static seals for hydrocarbon fluid cooler systems (e.g., engine oil, or powertransmission fluids):
– O-rings.
– Square cross-section seals.
– Gaskets.
• Flexibleprotectivebellowsonmechanicalfaceseals.
• Enginefuelstaticseals(awayfromtheheatoftheengine):
– O-rings for fuel injector return lines.
References
1. R.W. Keller, Rubber Chemistry and Technology, 1985, 58, 3, 637.
62
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
63
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.
6 Formulating Guidelines for Hydrogenated Nitrile Rubber
64
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
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
hexasulfide
0.7 0.7 0.7
Tetramethylthiuram disulfide 2.0 2.0 0.7 0.7 0.7 0.7
N-cyclohexyl-2-benzothiazole
sulfenamide
1.0
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
dithiocarbamate
0.5
Properties after vulcanisation, 20 min at 160 ºC
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% compression
Recipe from above
1 2 8 9 10 11
70 h at 100 ºC 35 34 37 41 42 41
70 h at 120 ºC 54 46 46 56 57 57
65
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 mercaptobenzimidazole
1.5 1.5 1.5 1.5 1.5 1.5
α,α-bis(t-butylperoxy)diisopropylbenzene
5.0 5.0 5.0 5.0 7.5 7.5
N,N′-m-phenylene dimaleimide 0.0 2.5 5.0 7.5 0.0 2.5
Ingredient phr in recipe
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- methylmercaptobenzimidazole
1.5 1.5 1.5 1.5 1.5 1.5
α,α-bis(t-butylperoxy)diisopropylbenzene
7.5 7.5 10.0 10.0 10.0 10.0
N,N′-m-phenylene dimaleimide 5.0 7.5 0.0 2.5 5.0 7.5
Properties after vulcanisation 15 min at 170 ºC
Property Recipe
1 2 3 4 5 6
Shore A hardness 70 73 76 78 71 73
Tensile strength, MPa 28.0 30.1 28.1 28.0 32.4 30.2
Ultimate elongation, % 560 340 270 240 400 270
50% Modulus, MPa 1.6 2.0 2.5 3.3 1.7 2.2
66
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
100% Modulus, MPa 2.2 4.1 6.4 9.0 2.9 5.0
Compression set, ASTM D395 Method B, 70 h at 150 ºC
61 32 26 22 42 29
Property Recipe
7 8 9 10 11 12
Shore A hardness 76 78 71 75 77 80
Tensile strength, MPa 29.3 27.1 32.0 28.2 28.6 27.7
Ultimate elongation, % 230 190 310 220 190 160
50% Modulus, MPa 2.8 3.8 1.8 2.5 3.2 4.4
100% Modulus, MPa 7.7 11.0 3.7 6.6 9.9 13.4
Compression set, ASTM D395 Method B, 70 h at 150 ºC
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:
• ShoreAhardness=74.70+0.75P+4.0C+0.5P2 – 0.375C2+0.15CP, (6.1)
where P=phrα,α-bis(t-butylperoxy)diisopropylbenzene peroxide, and
C=phrN,N′-m-phenylene dimaleimide coagent.
R-squared,adjustedfordegreesoffreedom=97.74%.
• 100%modulus=1.774–0.052P – 0.034C+0.006P2
+0.0187C2+0.0232PC (6.2)
R-squared,adjustedfordegreesoffreedom=99.54%.
• %elongation=711.6–49.6P – 76.9C+1.0P2+3.73C2+2.96PC (6.3)
R-squared,adjustedfordegreesoffreedom=97.70%.
• %compressionset=84.1–6.18P – 12.3C+0.12P2+0.56C2+0.688PC (6.4)
R-squared,adjustedfordegreesoffreedom=88.03%.
67
Formulating Guidelines for Hydrogenated Nitrile Rubber
78
76
74
72
Peroxide phr
Coa
gent
phr
1098765
7
6
5
4
3
2
1
0
Figure 6.1 Contour plot of Shore A hardness as a function of peroxide and coagent level
4.0
3.5
3.0
2.5
2.0
Peroxide phr
Coa
gent
phr
1098765
7
6
5
4
3
2
1
0
Figure 6.2 Contour plot of modulus at 100% elongation as a function of peroxide and coagent level
68
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
450400
350
300
250
200
Peroxide phr
Coa
gent
phr
1098765
7
6
5
4
3
2
1
0
Figure 6.3 Contour plot of ultimate elongation as a function of peroxide and coagent level
5045
40
35
30
25
1098765
7
6
5
4
3
2
1
0
Peroxide phr
Coa
gent
phr
Figure 6.4 Contour plot of compression set as a function of peroxide and coagent level
69
Formulating Guidelines for Hydrogenated Nitrile Rubber
Oven post-cure processes are also effective with HNBR compounds, as illustrated in a further study conducted by Zeon (Table 6.3 and Figure 6.5) [3].
Table 6.3 Zeon study on the effect of post-cure: Recipes and properties [3]
Ingredient phr in recipe
1 2 3 4 5 6 7 8
Zetpol 1020 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0
Zinc oxide 5.0 5.0 5.0 5.0 5.0 5.0 5.0 5.0
Stearic acid 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0
N770 carbon black 40.0 40.0 40.0 40.0 40.0 40.0 40.0 40.0
Dibutoxyethoxyethyl adipate 5.0 5.0 5.0 5.0 5.0 5.0 5.0 5.0
Sulfur (325 mesh) 0.5 0.5 0.5 0.5
Tetramethyl thiuram disulfide 2.5 2.5 2.5 2.5
2-Mercapto benzothiazole 0.5 0.5 0.5 0.5
40% Dicumyl peroxide on an inert carrier
8.0 8.0 8.0 8.0
Conditions Recipe
1 2 3 4 5 6 7 8
Press cure conditions 30 min at 160 ºC
30 min at 160 ºC
30 min at 160 ºC
30 min at 160 ºC
30 min at 160 ºC
30 min at 160 ºC
30 min at 160 ºC
30 min at 160 ºC
Oven post-cure conditions, h at 150 ºC
0 2 4 8 0 2 4 8
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 Method B, 70 h at 120 ºC
50 36 31 24 27 15 14 13
Compression set, ASTM D395 Method B, 70 h at 150 ºC
75 67 60 50 35 25 24 23
70
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
0
10
20
30
40
50
60
0 2 4 6 8 10
Com
pres
sion
Set
(70
h a
t 12
0 °C
)
Post-cure time (h at 150 °C)
Sulfur
Peroxide
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) 0.0–75.0
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
71
Formulating Guidelines for Hydrogenated Nitrile Rubber
The results for various carbon black levels and grades, along with nominal DBP absorption values, are illustrated in Figures 6.6 to 6.8.
1
125
0050
60
75
70
0
80
20 5040
60
Shor
e A
DBP
phr
Figure 6.6 Surface plot showing effect of carbon black on Shore A hardness
1
125
000
4
75
8
0
12
20 5040
60
10
0%
mod
ulus
DBP
phr
Figure 6.7 Surface plot showing effect of carbon black on modulus at 100% elongation
72
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
1
125
0050
100
750
150
20 5040
60
ML
(1+4
) at
10
0 °
C
DBP
phr
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=Mooneyviscosity,ML(1+4)at100°C
=60.16–0.929p+0.746D+0.00986p2 – 0.000540D2+0.0120pD, (6.5)
where p=phrcarbonblackand
D=DBPnumberofcarbonblack.
R-squared,adjustedfordegreesoffreedom=89.66%
• H=ShoreAdurometerhardness
=34.36+0.264p+0.372D – 0.00071p2 – 0.00202D2+0.00207pD (6.6)
R-squared,adjustedfordegreesoffreedom=91.19%
• M1=stressat100%elongation(100%modulus)inMPa
=–0.0804p +0.0593D+0.00149p2 – 0.000360D2+0.000808pD (6.7)
R-squared,adjustedfordegreesoffreedom=91.94%
73
Formulating Guidelines for Hydrogenated Nitrile Rubber
It is appreciated that these equations bear little resemblance to the theoretical equations for filler reinforcement, and they have no theoretical basis. In the real world, however, theoretical equations are rarely useful for predicting practical situations, and empirical relationships derived from actual studies such as these are often more valuable. From the standpoint of practical rubber compounding these equations are helpful for predicting the effect of carbon black reinforcement (high values of R-squared) and are therefore useful in formulating. In discussing the effects of fillers on viscosity and modulus (Chapter 3), the following theoretical equation was presented:
G1=G0(1+Af+Bf2) (6.8)
where G0 =shearmodulusofunfilledmaterial,
G1=shearmodulusoffilledmaterial,and
f=volumefractionoffillerinthemixture.
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.
74
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
Table 6.5 Zeon study of non-black fillers
Ingredient phr Relative particle
size
Shore A
100% modulus
Δ Shore A per phr
Δ Modulus at 100% per phr
None 0 50 1.1
Ultra-Pflex 90 small 65 1.6 0.167 0.0056
Roy-CalL 90 medium 65 1.8 0.167 0.0078
Zeeospheres 200 90 medium 66 2.4 0.178 0.0144
Imsil A–8 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
Cyprubond
90 medium 76 8.1 0.289 0.0778
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 TS–720 40 small 71 2.5 0.525 0.0350
HiSil 233 40 small 73 2.9 0.575 0.0450
Ultrasil VN–3 SP 40 small 74 3.0 0.600 0.0475
Aerosil 300 40 small 76 2.8 0.650 0.0425
Cab-O-Sil M–7D 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
75
Formulating Guidelines for Hydrogenated Nitrile Rubber
Zinc oxide 5.0
Stearic acid 0.5
Naugard 445 1.5
Vanox ZMTI 1.0
Carbowax 3350 2.0
A–172 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:
PhrNulok390=(70–50)/0.256=78phr
PhrHiSil233=(70–50)/0.450=44phr.
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.
76
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
Table 6.6 Plasticisers for HNBR
Plasticiser at 15 phr in base recipe
Property No plasticiser
Dioctyl phthalate
Dibutoxy ethoxyethyl
adipate
Ether ester
Trioctyl trimellitate
Triisononyl trimellitate
Compound
Mooney
viscosity, ML1+4(100
ºC)
133 86 78 81 91 87
Oscillating
disc rheometer
at 170 ºC,
Vmin
11.4 6.5 7.1 7.4 8.0 8.3
Oscillating
disc rheometer
at 170 ºC,
Vmax
51.6 40.4 46.2 41.4 53.1 56.3
Oscillating
disc rheometer
at 170 ºC,
T95, min.
19.5 18.3 15.1 16.5 17.2 17.0
Shore A
hardness
76 70 69 69 70 69
Ultimate
elongation, %
310 360 450 430 380 370
100%
Modulus,
MPa
6.2 4.1 3.4 3.3 4.2 4.3
Weight loss
after 168 h
at 150 ºC air
ageing, %
–1.4 –9.7 –8.6 –6.1 –3.4 –1.7
Temperature
retraction per
ASTM D1329
77
Formulating Guidelines for Hydrogenated Nitrile Rubber
TR-10, ºC –22.6 –28.7 –25.1 –25.0 –28.1 –27.9
TR-30, ºC –16.5 –20.6 –12.7 –12.5 –20.0 –19.6
TR-70, ºC –4.5 –8.5 –3.3 –3.8 –6.8 –6.8
Heat
resistance
weight loss,
168 h at 150
ºC
–1.4 –9.7 –8.6 –6.1 –3.4 –1.7
Softening Effects
Δ Shore A per
phr plasticiser
–0.4 –0.5 –0.5 –0.4 –0.5
Δ Mooney
viscosity per
phr plasticiser
–3.1 –3.7 –3.5 –2.8 –3.1
Δ 100%
Modulus per
phr plasticiser,
MPa
–0.140 –0.187 –0.193 –0.133 –0.127
Δ TR-10 per
phr plasticiser
(ºC)
–0.407 –0.167 –0.160 –0.367 –0.353
Base recipe ingredient
phr
Zetpol 2010 100
N550 carbon
black
50
α,α′-bis(t-
butylperoxy)
diisopropyl-
benzene,
40% on inert
carrier
6
78
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
Trioctyl trimellitate (TOTM) is one of the most common plasticisers used in HNBR formulations and the reasons for this are clear from Table 6.6, which shows that TOTM offers a good balance of raw compound viscosity reduction, reduction of low temperature properties as measured by TR–10, and excellent thermal stability in terms oflowweightlossat150°C.ThesofteningeffectlistedinTable 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(α,α′-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. Coumarone–indene resins, phenol–acetylene resins, aromatic hydrocarbon resins or styrene–vinyl 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.
79
Formulating Guidelines for Hydrogenated Nitrile Rubber
6.8 Blends with Ethylene–Propylene Diene Rubber
NBR can be blended with ethylene–propylene 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, 10–20 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 80–85 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 (17mole% acrylonitrile), to give the best lowtemperature flexibility, selecting the low-viscosity grade for injection moulding.
• Curesystem:Peroxidepluscoagentsystemforcompressionsetresistance:10.0phr α,α-bis(t-butylperoxy)diisopropylbenzene peroxide with 7.50 phr N,N′-m-phenylene dimaleimide.
• Carbonblack:N550(basedonpastexperiencewithNBRsealsinthisapplication).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
80
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
that our target for H is 82. We can also use a surface or contour plot such as that in Figure 6.6, giving 75.00 phr for N550.
• 1.5phr1,2-dihydro-2,2,4-trimethylquinolineantioxidant.
• PastexperiencewithHNBRcompoundshasshownthat5.00phrFrenchprocesszinc 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
antioxidant
1.5 Effective antioxidant
N550 Carbon black 75.0 Carbon black filler
α,α-Bis(t-butylperoxy)-
diisopropylbenzene peroxide
10.0 Efficient crosslinking, low
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
81
Formulating Guidelines for Hydrogenated Nitrile Rubber
requirement. We also need to keep the cost of the compound as low as possible to lessen the impact of the change from NBR to HNBR. We wish to select a mineral filler capable of high loadings, and will therefore use 10 phr of TOTM plasticiser to allow a higher filler content without seriously sacrificing high temperature compression set. The target Shore A durometer reading for this compound is 75.
• Polymer:Therban3446(34mole%ACN,4%residualunsaturationforhighcrosslink density and good compression set resistance.
• Heat-stablepigmentarysystem:1.00phrN990carbonblackplus4.00phrrediron oxide powder.
• Filler:Zeeospheres200toallowhighfillerloading:
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:
Totalfillerloading=(25+4)/0.178=163phr(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
antioxidant
1.5 Effective antioxidant
N990 carbon black 1 Pigment system
Red iron oxide powder 4 Pigment system
Zeeospheres 200 163 Filler
Trioctyl trimellitate plasticiser 10 Plasticiser
α,α-bis(t-butylperoxy)
diisopropylbenzene peroxide
10.00 Efficient crosslinking, low
compression set
N,N′-m-phenylene dimaleimide
coagent
7.50 Efficient crosslinking, low
compression set
82
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
These examples will give the reader some idea how to use this information to design formulations. In the chapter following we will discuss specimen formulations for specific applications of HNBR. We shall also discuss the value of statistical experimental design and desirability functions for studying formulation variables and selecting the optimal formula for the application concerned.
References
1. ZeonChemicalsLP,Semi-EV and EV Curing Systems for Zetpol 1020 (Z5.1.2),Louisville,KY,USA,1999.
2. ZeonChemicalsLP,Study of Level and Ratio of Peroxide and Co-Agent in Zetpol 2010 (Z5.1),Louisville,KY,USA,1999,p.1.
3. ZeonChemicalsLP,Effects of Post Curing on Zetpol 1020 (Z5.1), Louisville,KY,USA,1999,p.1.
4. ZeonChemicalsLP,Zetpol HNBR Elastomers: Carbon Black Study (SA–26), Louisville,KY,USA,April2001,p.1.
5. R.W. Keller, Hydrogenated Nitrile Rubber, in Handbook of Specialty Elastomers,Ed.,R.A.Klingender,CRCPress,BocaRaton,FL,USA,2008.
6. ZeonChemicalsLP, Evaluation of Larger Particle Non-Black Fillers in Peroxide Cured Zetpol 2010 (Z5.3.3),Louisville,KY,USA,1999,p.1.
7. ZeonChemicalsLP,Low Temperature Properties of Zetpol and Influence of Plasticizers and Tackifiers (Z5.4),Louisville,KY,USA,1999,p.1.
8. ZeonChemicalsLP,Protection Systems for Zetpol (Z5.5),Louisville,KY,USA, 1999.
83
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).
7 Examples of Hydrogenated Nitrile Rubber Formulae for Specific Applications
84
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
D = (d1, d2, d3, ... dn)
d=1∏nn
, (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 = (d1 * d2 * d3 * d4 * d5)5 (7.2)
85
Examples of Hydrogenated Nitrile Rubber Formulae for Specific Applications
For properties 1–4 (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.
0
50
100
150
200
250
300
050100150200
Tem
pera
ture
res
ista
nce
(°C
)
% Swell in Irm-903 OilASTM D2000 Classification
NBR
A B C D E F G H J K
Class by Oil Swell
A
BC
DEF
H
Typ
e by
tes
t te
mpe
ratu
re
NRCR
EPDMAEM
VMQFVMQ
FKM
ACM
HNBR
Figure 7.1 Graphical presentation of ASTM D2000
86
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
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 0–5%
d2 Definition: d2 = 1.00 if volume change in customer axle grease is > 0% after 1,008 hours at 125°C (ASTM D471); d2 = 0.00 if volume change is negative in the grease exposure test
d3 Definition: d3 = 0.00 if TR–10 temperature is above –40 °C; if TR–10 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 125°C; d6 = 0.00 if plied disc compression set is 100% or greater; linear between 0 and 100% compression set (B, below)
A
–52
d3
–47
TR-10, degrees C
0.8
0.6
0.4
0.2
0
1
–42
B
d6
2000
0.2
0.4
0.6
0.8
1
40 60 80 100
% Comp. set, 1,008 hrs. @ 125C in grease
87
Examples of Hydrogenated Nitrile Rubber Formulae for Specific Applications
A number of key desirabilities, d1, d2, d4 and d5, are pass/fail properties, for which our recipe must satisfy the minimum requirement. Two other properties, d3 and d6, are linear functions. Based on the information in Chapter 6, we have developed the basic recipe shown in Table 7.3, and the experimental design can be constructed around this recipe.
Table 7.3 Experimental design starting recipe for HNBR joint boot compound
Ingredient phr Requirements Experimental design optimisation
17 mol% HNBR polymer, 4% residual unsaturation
80–100
Good combination of freeze resistance and low–temperature flex properties
Vary level from 80 to 100 phr blending with EPDM polymer to achieve total 100 phr polymer
EPDM polymer – low ethylene, low viscosity, 4.5% third monomer unsaturation level
0–20 Low viscosity for injection mould flow; medium to high third monomer level for efficient vulcanisation
Vary level from 20 to 0 phr with blending of HNBR polymer to achieve total 100 phr polymer
Zinc oxide 2.0 Cure activator
Stearic acid 1.0 Cure activator
Polymerised 1,2-dihydro-2,2,4-trimethyl quinoline antioxidant
1.0 Optimal antioxidant and antiozonant balance
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 plasticiser
10.0 Reduction of compound viscosity for injection moulding; improvement of low temperature flexibility; good heat stability
Dicumyl peroxide, 40% on inert mineral carrier
3.0–5.0
Fast and efficient curing in injection moulding where cycle time is critical
Optimise level of peroxide for best balance of properties
88
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
The reader may be daunted at the prospect of making up and testing as many as nine formulae. On the other hand, guessing a formula and hoping it will perform well in all respects will almost certainly involve making up and evaluating many more formulations than this. The experiments are illustrated graphically in Figure 7.2.
3.003.203.403.603.804.004.204.404.604.805.00
00.0200.0100.0
phr EPDM in polymer blend
phr
pero
xide
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.
89
Examples of Hydrogenated Nitrile Rubber Formulae for Specific Applications
–2.4
–2.6
–2.8
–3.0
–3.2
–3.4
phr EPDM
phr
pero
xide
20151050
5.0
4.5
4.0
3.5
3.0
Figure 7.3 Experimental design of HNBR joint boot compound: Surface plot showing the effect of peroxide and EDPM on weight loss at 125 ºC
1.0
0.50.5
0.0
–0.5
–1.0
–1.5
–2.0
phr EPDM
phr
pero
xide
20151050
5.0
4.5
4.0
3.5
3.0
Figure 7.4 Experimental design of HNBR joint boot compound: Contour plot showing the effect of peroxide and EDPM on grease volume at 125 ºC
90
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
–42.0
–43.5
–45.0
–46.5 –48.0
phr EPDM
phr
pero
xide
20151050
5.0
4.5
4.0
3.5
3.0
Figure 7.5 Experimental design of HNBR joint boot compound: Results for low temperature retraction, TR–10
80
70
60
50
phr EPDM
phr
pero
xide
20151050
5.0
4.5
4.0
3.5
3.0
Figure 7.6 Experimental design of HNBR joint boot compound: Effect of peroxide and EDPM on compression set (Cset)
91
Examples of Hydrogenated Nitrile Rubber Formulae for Specific Applications
When we analyse the nine formulae in terms of our desirability function and calculate a composite desirability value for each formula, this gives the results summarised in Table 7.4.
Table 7.4 Experimental results for HNBR flex boot experimental designFactors 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
92
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
Table 7.4 shows that the formulae with no EPDM blended with the HNBR polymer has zero composite desirability, since these all show shrinkage on high temperature exposure to the customer’s 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.
0.00
0.25
0.50
010
0.75
4
5
320
com
posi
te D
phr peroxide
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 customer’s input of key properties.
93
Examples of Hydrogenated Nitrile Rubber Formulae for Specific Applications
The following sections illustrate further examples of formulae for HNBR in specific applications in terms of the variables studied and give suggestions for statistical experimental design.
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 100–120 °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 125–150 ºC include:
• Resistancetodieselengineoil.
• Resistancetowater/glycolcoolant.
• O-ringcompressionsetresistance.
94
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
Table 7.5 Oil cooler O-ring compound experimental design
Ingredient phr Logic/comments Experimental design
optimisation
HNBR polymer
with 34–36% ACN;
4–5.5% residual
unsaturation
100.0 Best overall balance of fluid resistance
and low temperature properties; the
higher levels of unsaturation give
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
antioxidant
0.7 Best overall heat resistance stabiliser
package
Vanox ZMTI
antioxidant
0.7 Best overall heat resistance stabiliser
package
Carbon black
(VARIABLE)
50–75 N762 black as low level in design at
75 phr, N550 black as high level in
design at 50 phr
Evaluate two
carbon black
types for best
overall balance of
properties
Varox VC40KE 10.0 Peroxide for compression moulding at
180 ºC
Vanax MBM
(VARIABLE)
4.5–7.5 Coagent for increased efficiency
of cure and best compression set
resistance
Optimise coagent
levels for best
balance of
properties
Optimisation:
2-factor 3-level full
factorial design; 8
compounds and
experiments
95
Examples of Hydrogenated Nitrile Rubber Formulae for Specific Applications
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:
• Heatresistance,125–150ºC.
• Crudeoilresistance.
• Hydrogensulfideresistance.
• Steamresistance.
• Aggressiveamineresistance.
• Pressureextrusionresistance.
96
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
Table 7.6 Oil well packer compound experimental design
Ingredient phr Logic/comments Experimental design optimisation
HNBR polymer with 34-36% ACN; less than 2% residual unsaturation
100.0 Best overall balance of heat and chemical resistance
Zinc oxide 2.0 Activator
Stearic acid 1.0 Activator
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 carbon black (VARIABLE)
55.0 To obtain > 90 Shore A durometer for pressure resistance
Evaluate two different carbon blacks for best overall balance of properties
Varox VC40KE 10.0 High temperature peroxide for compression moulding at 180 ºC
Coagent: Vanax MBM at 7.5 phr or TAIC at 5.0 phr (VARIABLE)
VARIABLE High level of coagent for efficient vulcanisation
Optimise compound based on evaluation of two chemical types of coagent
Optimisation: 2-factor 2-level full factorial design with centre point; 5 compounds and experiments
97
Examples of Hydrogenated Nitrile Rubber Formulae for Specific Applications
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 100–125 °C. Previously, the serpentine belt operated at 70–90 °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 100–125 °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:
• Heatresistance,100–125ºC.
• Staticanddynamicozoneresistance.
98
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
Table 7.7 Serpentine belt compound: Experimental design
Ingredient phr Logic/comments Experimental design optimisation
HNBR polymer with
34-36% ACN; 1-4%
residual unsaturation
(VARIABLE)
70-90 Best overall balance
between heat
resistance and efficient
vulcanisation
Vary levels of zinc
methacrylate modifier to
balance properties; total
polymer phr = 100
HNBR modified with
zinc methacrylate
30-10 Added tear and tensile
strength
Vary levels of zinc
methacrylate modifier to
balance properties; total
polymer phr = 100
Zinc oxide 2.0 Activator
Stearic acid 1.0 Activator
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
(VARIABLE)
10-20 Reinforcement without
loss of flex resistance
Vary level between 10 and
20 phr to obtain best balance
of properties
Varox VC40KE
(VARIABLE)
4-8 High temperature
vulcanisation system
Vary level between 4 and 8
phr to obtain best balance of
properties
Optimisation:
3-factor, 3-level
full factorial design
with centre point;
27 compounds and
experiments
99
Examples of Hydrogenated Nitrile Rubber Formulae for Specific Applications
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:
• Heatresistance,1008hat125ºC.
• Heatresistance,504hat150ºC.
• Ozoneresistance.
• Coolantresistance,1008hat125ºC.
• Lowtemperatureflexibility.
• Dynamicflexfatigueresistance.
• Mouldingperformance,flowandcurerates.
100
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
Table 7.8 Compound for water pump bellows experimental design
Ingredient phr Logic/comments Experimental design optimisation
21–25% Acrylonitrile
HNBR polymer, 1–4%
residual unsaturation
100.0 Best low temp. flexibility for
cold starts
Zinc oxide 1.0 Activator
Stearic acid 2.0 Activator
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) 0–20 Reinforcing filler Level 1 will be this
grade
HiSil 233 (VARIABLE) 0–20 Reinforcing filler Level 2 will be this
grade
Vinyl-tris(ß-
methoxyethoxy) silane
(Silane A–172)
1.0 Coupling agent for filler
dispersion
Red iron oxide 4.0 Stable pigment
Varox DCP–40C
(VARIABLE)
4–8 Rapid vulcanisation on
injection moulding at 170 ºC
Three levels to
optimise properties
Optimisation:
2-factor and 3-level full
factorial design;
6 compounds and
experiments
101
Examples of Hydrogenated Nitrile Rubber Formulae for Specific Applications
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 100–125 °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:
• Resistancetogearlubricantat125ºCfor1008h.
• Mouldingevaluations:flow,cleanlinessandbondingtothesteelcase.
• Shaftseallifetestingatextremesofoperatingenvironment.
102
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
Table 7.9 Differential shaft seal compound experimental design
Ingredient phr Logic/comments Experimental design optimisation
HNBR with 34-36% ACN;
1-4% residual unsaturation
100.0 Overall balance of oil
resistance and low
temperature properties
Zinc oxide 2.0 Activator
Stearic acid 1.0 Activator
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
Cyprubond (VARIABLE)
75.0 Mineral fillers show
best rotating shaft seal
performance
Evaluate two different
fillers for best overall
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
103
Examples of Hydrogenated Nitrile Rubber Formulae for Specific Applications
braided fabric reinforcement and the plies will be formed by extrusion. Due to the location of an exhaust header near the short section of hose the operating temperature of the hose will be normally at 125 °C, with an occasional excursions to 150 °C, for example when idling in heavy traffic. These requirements dictate the use of HNBR to provide temperature resistance, overall toughness, ozone resistance and retention of modulus at high temperatures. The critical properties of the compound and hose assembly include static and dynamic ozone resistance, flex fatigue resistance, smooth processing by extrusion, adhesion of the plies and the ability to withstand the customer’s 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
Ingredient phr Logic/comments Experimental design optimisation
HNBR polymer with
34–36% ACN; 1% and 5%
unsaturation
100.0 Best overall balance
of fluid resistance
Low level at 1%
unsaturation, high level at
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 DCP–40C 3.0–
5.0
Fast and efficient
curing
Variable level to optimise
properties
Optimisation:
2-factor, 2- and 3-level full
factorial design; 12 compounds
and experiments
104
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
Responses (properties) to be measured:
• Staticozoneresistance.
• Dynamicozoneresistance.
• Flex-fatigueresistance.
• Extrusionprocessing.
• High-temperaturepressurepulsetest,simulatingapplicationconditions.
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 ºC. 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 ºC, 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:
• Dynamichysteresis.
• Heatresistance,1008hat120ºC.
105
Examples of Hydrogenated Nitrile Rubber Formulae for Specific Applications
Table 7.11 Low hysteresis paper mill roller compound experimental design
Ingredient phr Logic/comments Experimental design optimisation
HNBR polymer with 34–36%
ACN; 5% unsaturation
50.0 Basic HNBR for
chemical resistance
Level 1 evaluation of
HNBR
HNBR polymer with 34–36%
ACN; 15% unsaturation
50.0 Higher unsaturation
gives less hysteresis
Level 2 evaluation of
HNBR
HNBR polymer modified with
zinc methacrylate
50.0 Adds toughness,
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 4.5–7.5 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.
106
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
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.
107
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
8 Solving Hydrogenated Nitrile Rubber Processing Issues
108
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
can see that this is unlikely to happen. From the discussion in Chapter 3, the typical effect of temperature on viscosity is shown in Figure 8.1.
Shear rate (s–1)
Temperature effect
1000000
100000
10000
1000
100100
100 ºC
125 ºC
150 ºC
1000101
Vis
cosi
ty (
pois
e)
Figure 8.1 Effect of temperature on pseudoplastic polymer viscosity
If the injection process is operating at a shear rate of 1,000 s–1 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
109
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.
–100%–100%–50%
0%50%
100%150%200%250%300%350%
–50% 0% 50% 100% 150%
% c
hang
e in
hea
t bu
ild-u
p
% 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 = 0∫t (h/rc)g2 dt, (8.1)
where DT = temperature change in ºC s–1,
h = viscosity in Pa. s and
g = shear rate in s–1.
110
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
Using Equation (8.1) we can establish the following proportionalities to estimate the effect of viscosity and shear rate on heat build-up, as shown in Figure 8.2:
DT ∝ h, and DT ∝ g2 (8.2)
Reducing the viscosity, h, 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)
111
Solving Hydrogenated Nitrile Rubber Processing Issues
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.
% Change in factor
120%100%80%60%40%20%0%
–20%
–20% –10% 0% 10% 20% 30%
–40%–60%
Radius of runner Length of runner Viscosity of material
Net
% c
hang
e in
flo
w r
ate
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.
112
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
8.2.4 Solving the Injection Flow Problem using the Above Factors
From Chapter 7 the reader may already have concluded that a statistical experimental design would be appropriate using the factors discussed above. Listed in Table 8.1 and illustrated in Figure 8.4 is an experimental programme designed to resolve the flow and non-fill issue.
Table 8.1 Experimental design for sporadic flow and non-fill injection moulding issues
Mould temperature
(°C)
Injection time (s)
Injection pressure holding time (s)
Responses (dimensions)
parts/100 with non-fills
Comments
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.
113
Solving Hydrogenated Nitrile Rubber Processing Issues
Injection feed temperature, °C
Run
ner
size
, mm
521001
100
1505.0
6.0
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/ºC. If room temperature is 25 ºC and the moulding temperature is 180 ºC, 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
114
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
temperature of 180 ºC on a 76 mm part and the moulding temperature drifted to 190 ºC, 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/ºC. If the mould was designed for a vulcanisation temperature of 180 ºC but the temperature was actually 190 ºC, 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.
115
Solving Hydrogenated Nitrile Rubber Processing Issues
4540353025201510
10 100 10001
50
% D
ie s
wel
l
Shear rate (s–1)
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 temperature (°C)
Injection time (s)
Injection pressure hold
time (s)
Responses (dimensions)
Comments
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
116
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
Mould temperature, °C
Pres
s ho
ld t
ime,
s
Inje
ctio
n ti
me,
s
091081
10
205
15
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:
117
Solving Hydrogenated Nitrile Rubber Processing Issues
• Slipagentusedinthecompound:
– Low level: paraffin wax at 1.00 phr.
– High level: fatty amide at 1.00 phr.
• Extruderbarreltemperature:
– Low level: 80 °C.
– High level: 100 °C.
• Extruderbreakerplateholesize:
– Low level: 7.00 mm hole size.
– High level: 8.00 mm hole size.
• Taperangleatentrancetothetubedie:
– Low level: 30º angle.
– High level: 45º angle.
• Extruderspeed:
– Low level: 30 rpm.
– High level: 40 rpm.
• Extruderdietemperature:
– 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 Plackett–Burman screening design would be appropriate, as illustrated in Table 8.3:
118
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
Table 8.3 Hose tube extrusion, 8-run Plackett–Burman 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 1.00B = Extruder barrel temperature: –1 = 80 ºC; +1 = 100 ºCC = Extruder breaker plate hole size: –1 = 7.00 mm; +1 = 8.00 mmD = Taper angle at entrance to die: –1 = 30º; +1 = 45ºE = Extruder speed: –1 = 30 rpm; +1 = 40 rpmF = Extruder die temperature: –1 = 100 ºC; +1 = 120 ºC
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
119
Solving Hydrogenated Nitrile Rubber Processing Issues
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 ofweighingtolerance.
• Compoundvariable:vulcanisingagent(peroxide)attheextremesofweighingtolerance.
• Processvariable:rawrubberweightsattheextremesoftolerance.
• Processvariable:variationsinmouldsurfacetemperature.Theplatentemperaturewas 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.
• Processvariable:mouldclosure.Duetotheplatenandmouldbeingmarginallyout of parallel and flatness, the mould could remain open by as much as 0.05 mm.
• Processvariable:ovenpost-curetemperature.Theproductunderwenta2hpost-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 Plackett–Burman 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 Plackett–Burman 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 Plackett–Burman experimental design, with results and analysis, is shown in Table 8.4 and Figure 8.7. The values given are ID measurements in mm.
120
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
Table 8.4 Screening design for dimensional resolution: 8-run Plackett–Burman 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 –96.63 –92.3 –92.3 92.3 92.3 92.3 –92.3 92.34 92.7 –92.7 –92.7 92.7 92.7 92.7 –92.75 –96.6 96.6 –96.6 –96.6 96.6 96.6 96.66 99.9 –99.9 99.9 –99.9 –99.9 99.9 99.97 96.3 96.3 –96.3 96.3 –96.3 –96.3 96.38 –100.7 –100.7 –100.7 –100.7 –100.7 –100.7 –100.7Sum + 385.2 385.9 385.2 377.9 377.9 385.9 385.2Sum – –386.3 –385.6 –386.3 –393.6 –393.6 –385.6 –386.3Average E+ 96.3 96.5 96.3 94.5 94.5 96.5 96.3Average E– –96.6 –96.4 –96.6 –98.4 –98.4 –96.4 –96.6Effect –0.3 0.1 –0.3 –3.9 –3.9 0.1 –0.3Significant? no no no YES YES no noRESULTSRun 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 52 96.8 96.5 96.2 96.8 96.9 96.6 0.1968 53 92.1 92.0 91.9 92.8 92.8 92.3 0.3272 54 92.6 93.0 93.1 92.1 92.5 92.7 0.6008 55 96.2 96.8 97.1 96.5 96.5 96.6 0.4392 56 100.2 100.1 99.5 99.6 100.3 99.9 0.2728 57 96.3 96.0 95.8 96.8 96.5 96.3 0.5312 58 101.0 100.8 100.2 101.1 100.5 100.7 0.4512 5
Si2 = sum of squares = Σ(yi–y(avg.))2 Sum Si2 2.882
Avg. Si2 0.36
Se = √Si 0.6
Seff 0.19 = Se × √(4/N)
DF = 32 (n–1) × number of runs
N = total number of experimentsA = carbon black filler level at maximum of weigh-up toleranceB = vulcanising agent levels at maximum of weigh-up toleranceC = preform weight levels at extremes of tolerancesD = mould surface temperature: – = 175 ºC, + = 190 ºCE = mould closure: – = 0.00 mm, + = 0.05 mmF = Post-cure temperature: – = 190 ºC, + = 215 ºCMSFE: 0.387 t × S(eff)t (0.950, df) 2.04 (n–1) × (number of runs) = degrees of freedom
121
Solving Hydrogenated Nitrile Rubber Processing Issues
Mou
ld te
mpe
ratu
re
Mou
ld cl
osur
e
Carbo
n bl
ack
level
Prefo
rm w
eight
Curat
ive le
vel
Post-
cure
tem
pera
ture
4.5
4.0
3.5
3.0
2.5
2.0
1.5
1.0
0.5
0.0
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.
122
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
8.5 Conclusions
Techniques have been demonstrated which will help in resolving production issues. In cases where multiple variables are involved, emphasis has been placed on statistical experimental design techniques in order to determine which are significant. This is intended to guard against the knee-jerk response that a change in formulation is the way to solve a production processing issue, and also attempting to examine the variables one at a time and missing the possibility of interaction between them. Statistical experimental design is a technique that can help to reduce variation in manufacture. Unfortunately many process and product improvement projects do not in practice employ these techniques properly, or fail to plan the experimental design effectively. The involvement of a wide range of personnel and disciplines is critical in planning stage and developing proper experimental design.
123
ACM Acrylic elastomer (per D1418)
ACN Acrylonitrile
AEM Ethylene–acrylic 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 Epichlorohydrin–ethylene oxide copolymer
EDPM Ethylene–propylene–diene terpolymer
FEF Fast extrusion furnace black
FEPM Polytetrafluoroethylene–propylene copolymer
FKM Fluorinated hydrocarbon rubber
FVMQ Fluorinated vinyl–methyl silicone elastomer
HFC Hydrofluorocarbon
Abbreviations
124
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
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 Styrene–butadiene rubber
Tg Glass transition temperature
TOTM Trioctyl trimellitate
VMQ Vinyl–methyl silicone elastomer
XHNBC Cross-carboxylated HNBR
γ Shear rate
125
Abbreviations
ΔP Pressure drop
Ø Volume fraction in a mixture
Ø2 Volume fraction of polymer
η Viscosity
τ Shear stress
126
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
127
2-Mercaptobenzothiazole 64100% Modulus 64, 66, 69, 72, 75200% Modulus 69α,α-bis(t-butylperoxy)diisopropylbenzene 65-66, 77, 79-81
A
ACM 17-20, 46, 49-51, 53, 56, 58, 85ACN 1, 3, 5-11, 13-14, 16, 18-19, 43, 45, 47-53, 63, 80-81, 94, 96, 98, 102-103,
105Acrylonitrile 1, 3, 9, 11, 33, 79, 100Activator 80-81, 87, 94, 96, 98, 100, 102AEM 17-20, 46, 49-50, 53, 56, 58, 85Aerosil 300 74Air conditioning 57, 60ASTM D2000 17, 43-46, 55-56, 85Automatic 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 74Cab-O-Sil TS-720 74Capillary 37-38
length 110radius 111reheometer 37
Carboxylated 16Celite 350 74, 102Coagent 63, 65-68, 78-81, 94, 96
Index
128
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
Compression modulus 33-34Compression moulding 24, 27, 93-96, 118, 121Compression set 34, 47, 64, 66, 68-70, 79-81, 86, 90, 93-94, 97Contour plot 67-68, 80, 89Conveyor 56-57
belt 16, 57component 56
Copolymerisation 11
D
DBP number 32-34, 72Desirability function 83, 85, 91-92Dibutoxyethoxyethyl adipate 69Dicumyl peroxide 47-48, 69, 87Die swell 24-25, 115Dimensional shift 113, 116Dioctyl phthalate 76Dynamic ozone resistance 78, 97, 103-104
E
Einstein equation 32Elasticity 24, 29, 32, 34, 52, 95Emulsion polymerisation 1-3, 5-9EPDM 10, 17, 29, 31, 46, 56, 79, 85, 87-93Ether ester 76Ethylene 9-12, 17, 19, 29, 46, 56, 79, 87, 93Extrusion 34, 95, 103-104, 116, 118
F
Fatty amide 117-118Filler 20, 32, 34-35, 63, 73-75, 78, 80-81, 100, 105, 119-120FKM 17, 19-20, 36, 46, 51-53, 56, 85, 93Flexible boots 57Full factorial design 83, 94, 96, 98, 100, 102-103, 105, 117FVMQ 17, 19-20, 46, 53, 56, 85
G
Glass transition temperature 10, 26
129
Index
H
Heat build-up 27-28, 104, 109-110, 114HiSil 233 74-75, 100HiSil 532EP 74, 100HNBR 1, 7-14, 16-21, 43, 45-60, 63, 69-70, 73, 75-76, 78-82, 85, 87-105, 107,
109, 113-114, 116Hydrogenation 1, 7-9, 11, 45-46, 48, 55, 57, 63, 79Hysteresis 34-35, 104-105
I
Imsil A-8 74Initiator 1-3Injection moulding 13, 24-25, 27, 34, 36, 79, 87, 99-100, 107-108, 112-113, 115Isoprene 11-12
K
Knit lines 36, 107
L
Lanxess 13-16
M
Manual transmission 59Mayo-Lewis copolymer equation 3Minimum significant factor effect (MSFE) 121Mistron Cyprubond 105Mn 30Modulus 26, 32-34, 64-67, 69, 71-77, 95, 103-104Mole % ACN 49Mole fraction of monomer 5Molecular weight 2, 6, 29-31Molecular weight distribution 29-31Mooney viscosity 13, 31, 72, 76Mould flow 36-37, 87Mz 30
N
N110 32N330 65, 96N550 32, 64, 77, 79-80, 94, 96, 102-103
130
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
N762 46-48, 87, 94, 98N990 32, 81NBR 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, 101Newtonian fluid 21Non-black fillers 73-74, 82Non-fills 107-108, 112NR 17, 46, 56, 85Nulok 321 74, 100Nulok 390 74-75
O
O-rings 59-61, 78, 93, 114, 118, 121Oil cooler O-ring 93-94Oil well packer 96Oxidation 17, 45, 49, 55
-reduction 1Ozone resistance 78-79, 84, 97-100, 103-104
P
Packer elements 60Paper mill roller 105Paraffin wax 87, 117-118Peroxide 9, 43, 47-48, 52-53, 63, 65-70, 78-82, 87-92, 94, 96, 99, 102, 119Petroleum exploration 51-52, 60phr 32-36, 46-48, 65-70, 72, 74-81, 87-92, 94, 96, 98, 100, 102-103, 105, 117Plackett-Burman screening design 117-120Plasticiser 35-36, 63, 75-79, 81, 87Polymerisation 1-9, 11, 31Post-cure 69-70, 119-121Pressure drop 37-38, 110Process aids 36-37Propylene 9-12, 29, 46, 56, 79Pseudoplastic 22-23, 108Pyrax B 74
R
Radial shaft seals 59-60Reactivity ratios 4-9Recipe 33, 46-48, 64-66, 69-70, 73, 75-77, 80-81, 87, 93, 97, 99, 101, 103
131
Index
Residual unsaturation 9-10, 13, 63, 81, 98, 100, 102Response surface 34, 66, 70, 83, 93, 95Rollers 57, 104Roy-Cal L 74Runners 37-38, 111
S
Serpentine belt 97-98Shaft seals 59-60, 101Shear 13, 21-25, 27-28, 31-33, 35-38, 73, 108-110, 114-115Shear rate 21-25, 27-28, 31-33, 35-36, 38, 108-110, 114-115Shear stress 21, 24-25, 37-38, 110Silene 732D 74Sprues 37-38, 111Static ozone resistance 104Static seals 58-61Statistical experimental design 19, 82-83, 85, 92-93, 99, 103, 105, 107, 112, 115,
122Stearic acid 33, 47-48, 64-65, 69-70, 75, 80-81, 87, 94, 96, 98, 100, 102-103Steering system hose 102-103Sulfur 9, 13, 17, 49, 52, 58, 63-64, 69-70, 101
T
Tensile strength 47, 64-66, 69Tetramethyl thiuram disulfide 69Tg 10, 24, 26Therban® 15-16Time-temperature superposition 26Transfer moulding 24, 101Translink 77 74Trioctyl trimellitate 78, 81, 87
U
Ultimate elongation 47, 64-66, 68-69Ultra-Pflex 74Ultrasil 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
132
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
VMQ 17, 46, 56, 85Volume fraction 32, 35, 73Volumetric flow 37-38, 110-111
W
Williams-Landel-Ferry relationship 26
Z
Zeeospheres 200 74, 81Zeolex 23 74Zeon Chemicals 13, 15-16, 20, 51, 54, 63, 78, 82Zetpol® 15-16Ziegler-Natta-catalyst 29Zinc methacrylate 16, 98, 104-105Zinc oxide 33, 47-48, 64-65, 69-70, 75, 80-81, 87, 94, 96, 98, 100, 102-103
a b c d
a b c d
a b c d
a b c d
a b c d
a b c d
a
a b c d
a
Shawbury, Shrewsbury, Shropshire, SY4 4NR, UK Telephone: +44 (0)1939 250383 Fax: +44 (0)1939 251118 Web: www.polymer-books.com
Practical Guide to H
ydrogenated Nitrile B
utadiene Rubber TechnologyRobert Keller
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 author’s 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 author’s 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.
Published by Smithers Rapra Technology Ltd, 2012
Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology
Robert Keller