Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology 1847355218, 1847355226, 9781847355218, 9781847355225

Citation preview

Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology Robert Keller

A Smithers Group Company Shawbury, Shrewsbury, Shropshire, SY4 4NR, United Kingdom Telephone: +44 (0)1939 250383 Fax: +44 (0)1939 251118 http://www.polymer-books.com

First Published in 2012 by

Smithers Rapra Technology Ltd Shawbury, Shrewsbury, Shropshire, SY4 4NR, UK

©2012, Smithers Rapra Technology Ltd

All rights reserved. Except as permitted under current legislation no part of this publication may be photocopied, reproduced or distributed in any form or by any means or stored in a database or retrieval system, without the prior permission from the copyright holder.

A catalogue record for this book is available from the British Library.

Every effort has been made to contact copyright holders of any material reproduced within the text and the authors and publishers apologise if any have been overlooked.

ISBN: 978-1-84735-521-8 (softback) 978-1-84735-522-5 (ebook)

Typeset by Argil Services

C

ontents

1

2

3

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

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

The Flow and Processing Behaviour of Hydrogenated Nitrile Rubber ........................................................................... 21 3.1

Introduction .................................................................... 21

3.2

The Basics of Viscoelastic Flow in Rubber ....................... 21

3.3

3.2.1

Viscosity ............................................................... 21

3.2.2

Elasticity .............................................................. 24

Effect of Process Variables ............................................... 26 3.3.1

Effect of Temperature on Viscosity ....................... 26

3.3.2

Heat Build-up During Flow .................................. 27

iii

Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology

3.4

3.5

4

5

6

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

Flow through Sprues and Runners: Mould and Machine Parameters Influencing Elastomer Flow ............. 37

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

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

Formulating Guidelines for Hydrogenated Nitrile Rubber ......... 63 6.1

iv

Effect of Compounding Variables on Viscosity, Elasticity and Flow .......................................................... 29

Introduction .................................................................... 63

Contents

7

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

Examples of Hydrogenated Nitrile Rubber Formulae for Specific Applications .................................................................. 83 7.1

Introduction .................................................................... 83

7.2

Statistical Experimental Design and Desirability Functions ...................................................... 83

7.3

Theoretical Example: High-temperature HNBR Joint Boot ........................................................................ 85

7.4

Compound Example: High Temperature Oil Cooler O-ring Seal ...................................................................... 93

7.5

Compound Example: Oil Field High-pressure Well Packer ...................................................................... 95

7.6

Compound Example: High temperature Long-life Automotive Serpentine Belt ............................................. 97

7.7

Compound Example: Orange Water Pump Mechanical Seal Protective Bellows ................................. 99

7.8

Compound Example: High-temperature Differential Shaft Seal .................................................... 101

7.9

Compound Example: Short Steering System Hose ......... 102

v

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

8.3

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

Example 2: Sporadic Dimensional Shift in Injection Moulded Parts ................................................ 113 8.3.1

Possible Factor: Mould Temperature .................. 113

8.3.2

Possible Factor: Injection Pressures and Holding Pressures and Times.............................. 114

8.3.3

Solving the Injection Moulding Dimensional Problem using the Above Factors ....................... 115

8.4

Example 3: Variation in Hose Tube Thickness during Extrusion ............................................................ 116

8.5

Example 4: Variation in Dimensions of High-volume Compression Moulded O-ring ....................................... 118

8.5

Conclusions ................................................................... 122

Abbreviations .................................................................................... 123 Index ............................................................................................... 127

vi

P

reface

The goal of this book has been to give the reader a short but comprehensive view of the technology of hydrogenated nitrile rubber (HNBR). It is not intended to provide an exhaustive literature search, but more to act as a guide for those new to HNBR and seeking more information. My purpose has been to give an overview of: • The manufacture and chemistry of HNBR elastomers. • The commercial grades currently available. • How HNBR its into the current world of speciality elastomers. • The basics of compounding and formulating HNBR. • Examples of compounding for speciic applications. • Its rheological properties and how these can be applied. • Problem solving techniques using speciic examples. Much of the information will be found applicable not only to HNBR but also to other elastomers.

vii

Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology

viii

1

Overview of the Chemistry and Manufacture of Hydrogenated Nitrile Butadiene Rubber Polymers

1.1 Introduction 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 signiicant rubber polymers. It is also not the intention to cover the proprietary speciics 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 low 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 briely discussed. These HNBR polymers are important for achieving low temperature lexibility in the resulting moulded goods.

1.2 The Starting Polymer: Nitrile Butadiene Rubber Nitrile rubber, ASTM D1418 designation NBR, is a copolymer of ACN and 1,3-butadiene. It is formed by free-radical polymerisation, shown in Figure 1.1. The polymerisation is a typical emulsion polymerisation, which has the following characteristics: • The monomers are usually contained in emulsiied droplets in the continuous phase. In the polymerisation of NBR the continuous phase is water, and the butadiene and ACN monomers and resulting polymers are contained in the emulsiied droplets in the ‘oil’ or emulsiied phase. • The initiator, in this case a free-radical species, is generally quite soluble in the continuous phase. The majority of NBR polymers are described as ‘cold’ polymers, since a free radical is formed at sub-ambient temperature (typically 5 °C) in an oxidation–reduction reaction. Other NBR polymers are formed at elevated

1

Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology temperatures (typically around 95 °C) and are known as ‘hot’ polymers. Since the free-radical initiator is soluble in the continuous phase but only partially soluble in the emulsiied phase, free radicals diffuse slowly into the emulsiied 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 emulsiied phase to control the molecular weight of the polymer to give the desired bulk viscosity. • An emulsiier, which may be simply a sodium stearate soap, assures ine dispersion and emulsiication of the monomer/polymer-rich phase. • The continuous phase acts as an excellent heat transfer medium in the exothermic polymerisation reaction. In some cases bulk free-radical polymerisation generates so much heat that the process becomes impractical. Due to heat transfer in the continuous phase in emulsion polymerisation very tiny droplets of bulk polymer are generated, eliminating the need for expensive solvents and solvent recovery. • The viscosity does not increase during the conversion, as is the case in bulk or solution polymerisation. The viscosity remains essentially constant at the viscosity of the continuous phase. Initiator = free radical Only partially soluble in ‘oil’ phase R*

Monomers in emulsified ‘oil’ phase Polymer forms here Typical size of emulsified phase = 100 nm CH 2=CH---CH=CH 2 +

CH 2=CH >>>>> ---(CH2 ----CH=CH----CH2)x-(CH2----CH)y -(CH 2 =CH ) Z| | | CH=CH 2 CΞN CΞN

Typical composition of resulting NBR polymer = X = 84-50%, Y = 5% of butadiene incorporated, Z = 16-50%

Figure 1.1 Emulsion free-radical polymerisation of NBR

2

Overview of the Chemistry and Manufacture of Hydrogenated Nitrile Butadiene Rubber Polymers In general, NBR polymers contain between 18 and 50 mole% ACN content, the remainder being butadiene. A crude schematic of the emulsion polymerisation is shown in Figure 1.2.

Acrylonitrile Emulsifier

Polymerisation

Butadiene

Initiator

Coagulation

Water

Drying and dewatering

Final polymer bales

Figure 1.2 Schematic showing NBR polymerisation

NBR are copolymers and follow the Mayo–Lewis copolymer equation [1]. Since two monomers, ACN and butadiene, are involved the active terminal grouping in the growing chain can undergo the following possible reactions:

M1* + M1 ----- 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)

3

Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology The relative reactivity ratios of the monomers can then be deined as follows:

r1 = k11/k12 = reactivity ratio for monomer 1

(1.5)

r2 = k22/k21 = reactivity ratio for monomer 2

(1.6)

The molar concentration of monomers at any instant during the polymerisation is given by Equation (1.7):

d[M1] d[M2]

=

[M1]*(r1[M1]+[M2]) [M2]*([M1]+r2[M2])

(1.7)

where [M1] and [M2] are the molar concentrations of monomer 1 and monomer 2, respectively. The limiting cases in Equation (1.1) are as follows: • r1 = r2 >> 1. The two monomer reactivity ratios are high, indicating that the activated monomers have little tendency to react with one another. Each activated monomer would more probably react with the same monomer, leading to a mixture of homopolymers in the resulting product (1-1-1-1-1-1-1-….. + 2-2-2-2-2-2-…..). • r1 = r2 > 1. When monomer 1 forms the reactive chain end it will be more likely to react with further monomer 1. In the instance where cross-polymerisation occurs and monomer 2 becomes the reactive chain end, activated monomer 2 will probably react with further monomer 2. The resulting polymer will then be a block copolymer containing long segments of monomer 1 attached to long segments of monomer 2 (1-1-1-1-1-…….2-2-2-2-2-2-2….). • r1 = r2 = approximately 1. The reactive chain ends, whether monomer 1 or monomer 2, have no preference for either monomer 1 or monomer 2 and the result is a random copolymer (e.g., 1-2-1-1-2-2-2-1-2-1-1-1-1-2-2-1-2-2-2-2-21-2-2-2….). • r1 = r2 = approaching 0. If monomer 1 forms the reactive chain end it has little likelihood of reacting with further monomer 1, and similarly, if monomer 2 is the reactive chain end it has little likelihood of reacting with monomer 2. In this case, a perfectly alternating copolymer will result (1-2-1-2-1-2-1-2….).

4

Overview of the Chemistry and Manufacture of Hydrogenated Nitrile Butadiene Rubber Polymers • r1 >> 1 >> r2. At the beginning of the polymerisation monomer 1 has a more reactive chain end than monomer 2, and the initial copolymer is therefore rich in monomer 1. Later, as the monomer 1 concentration is reduced, the copolymer becomes rich in monomer 2. This is known as compositional drift (early in the reaction: 1-1-1-1-1-2-1-1-2-1-1-1-….., followed later by: 1-2-2-2-2-2-2-1-2-2-21-2-2-2….). • 0 < r1 < 1 and/or 0 < r2 < 1. The reaction tends towards homopolymerisation, and this tendency increases with decreasing rX. We can also rearrange the Mayo–Lewis equation to express the ratio of monomer 1 in the instantaneous polymer with respect to the monomer concentrations and reactivity ratios:

F1 =

[r1f1 2 + f1f2] [r1f1 2 + 2f1f2 + r2f2 2]

(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. Against this background, the question arises as to the values of r1 and r2 for the ACN and butadiene monomers used in the emulsion polymerisation of NBR. From this we can gain a perspective on some of the challenges presented by the polymerisation reaction. A paper describing the reactivity ratios of ACN and butadiene has been published by Embree and co-workers [2]. The emulsion polymerisation of NBR polymers has a number of complicating factors: • ACN has reasonably high solubility in the aqueous continuous phase. ACN monomer will therefore be present in both the emulsiied and the continuous phase. • The resulting NBR polymer is not soluble in butadiene, and this may result in a non-homogeneous emulsiied phase.

5

Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology In their study of polymer composition and molecular weight, Embree and co-workers used previously published data in addition to a number of their own experiments. Their analyses indicated that r1 (butadiene) was 0.18 and r2 (ACN) was 0.03. In this case, r1 >> r2, and both r1 and r2 were much lower than 1.0. If the reactive chain end was a butadiene monomer, the reactivity ratios would indicate a greater probability that the reactive chain end would react with a further butadiene monomer unit. On the other hand, in the case where the reactive chain end was an ACN monomer the reactivity ratios would indicate a greater likelihood that the reactive chain end would also react with another butadiene monomer. From these reactivity ratios we would expect the resulting NBR copolymer to contain small segments of repeating butadiene and relatively isolated ACN units. We can also use the data of Embree and co-workers [2] to understand the effect of conversion on the composition of the polymer (i.e., how much of the initial monomer charge is converted to polymer) for a given initial charge of monomers into the emulsion polymerisation system. For example, if we take their example of a 75/25 molar charge of butadiene/ACN monomers in the mixture, their data may be replotted as shown in Figure 1.3.

% ACN in polymer, 75/25 Butadiene/ACN Charge mixture

% ACN

40 30 20

% ACN incremental polymer

10

% ACN in polymer mix

0 0

20

60 80 40 % Conversion

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,

6

Overview of the Chemistry and Manufacture of Hydrogenated Nitrile Butadiene Rubber Polymers since the remaining monomer will have to be stripped out of the polymer. From Figure 1.3, a practical balance for this blend of monomers might give 70–80% 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 and resource required to strip residual monomer from the resulting emulsion mixture. We can also use Equation (1.3) to determine the mole% ACN in the instantaneous polymer formed as a function of the mole% composition of monomer (Figure 1.4):

polymer NCA%

%ACN in instantaneous polymer 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0%

%ACN in polymer Ideal

0% 20% 40% 60% 80% 100% %ACN in monomer feed

Figure 1.4 ACN content of polymer versus monomer feed

The ideal line in Figure 1.4 would be one at which the reactivity ratios are 1.00 and the monomer feed composition results in the corresponding instantaneous polymer composition. This understanding of the NBR polymer formed by emulsion polymerisation is important in appreciating the properties of the HNBR polymer formed by the catalytic hydrogenation of the NBR polymer, and the challenges presented by the process. The key points in the emulsion polymerisation yielding NBR polymers are as follows: • The reactivity ratios of the two monomers, butadiene and ACN, indicate that the NBR polymers formed will have at least short segments of repeat butadiene monomer and relatively isolated ACN monomer units. This is particularly true in the case of NBR polymers of very low ACN content.

7

Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology • At high conversion values the composition of the NBR polymer changes, and the resulting bulk NBR polymer may well be a blend of a range of polymers of different ACN content. • The reactivity ratios of ACN and butadiene tend to favour the presence of alternating monomers in the resulting polymer.

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 irst signiicant step is emulsion polymerisation to give the initial NBR polymer. The NBR polymer is then coagulated, dried and stripped of residual monomer. The hydrogenation reaction is completed by dissolving the NBR polymer in a suitable solvent, followed by catalytic hydrogenation to give HNBR.

Hydrogen

NBR Polymer

Solvent

Dissolution

High pressure catalytic hydrogenation

Separation from solvent and solvent removal

Bales of HNBR polymer Drying

Figure 1.5 Schematic process for the catalytic hydrogenation of NBR to form HNBR

A catalyst which facilitates hydrogenation is normally employed. A typical combination is monochlorobenzene as solvent and OsHCl(CO)(O2)(PCy3)2 as the

8

Overview of the Chemistry and Manufacture of Hydrogenated Nitrile Butadiene Rubber Polymers catalyst [3], in which Cy denotes a cyclohexyl residue. The hydrogenation reaction is illustrated in Figure 1.6.

NBR polymer ----{CH2—CH}x----{CH2—CH=CH—CH2}y---{CH2—CH}z ---- + (2x+2y) H 2 (gas) CN

acrylonitrile

CH=CH2

1,4-butadiene addition

1,2-butadiene addition, vinyl residue

Metal catalysts, high pressure

HNBR polymer ----{CH2—CH}x----{CH2—CH2 –CH2 —CH2}y ---{CH2—CH=CH—CH2}a ---{CH2—CH}z ---CN

acrylonitrile

ethylene

residual unsaturation for vulcanisation

CH 2 – CH3

propylene

Figure 1.6 Catalytic hydrogenation of NBR

In effect, the resulting HNBR is a tetrapolymer of ACN, ethylene and propylene, containing a small amount of residual unsaturation which allows rapid vulcanisation by means either of sulfur or a peroxide cure. The emulsion polymerisation process to give the NBR starting polymer may be summarised as follows: • The reactivity ratios of the two monomers, butadiene and ACN, indicate that the NBR polymer formed will comprise short segments of repeating butadiene and relatively isolated ACN monomer units. This is particularly true for NBR polymers of very low ACN content. The resulting HNBR polymer would thus contain segments of an ethylene–propylene polymer of high ethylene content. • At high conversion values, the composition of the NBR polymer changes, and it may well comprise a blend of polymers of different ACN content. These segments of ethylene–propylene polymer of high ethylene content create major challenges in the manufacture of HNBR polymers of low ACN content, required

9

Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology when the end application demands improved low temperature lexibility. However, HNBR polymers of low ACN content produced from conventional NBR polymers actually have inferior low temperature lexibility 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.

Tg (ºC) of low vinyl EPDM 0 40

45

50

55

60

65

70

75

80

–10

Tg

–20 Tg (ºC)

–30

linear Tg (ºC)

–40

y = 1.2688 x – 102.79 R2 = 0.9679

–50 –60 Mole% ethylene

Figure 1.7 Tg of ethylene–propylene copolymers versus ethylene content

As can be seen from Figure 1.7, low temperature lexibility 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 lexibility, 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.

10

Overview of the Chemistry and Manufacture of Hydrogenated Nitrile Butadiene Rubber Polymers The challenge for manufacturers of HNBR polymers is thus to disrupt the regularity of the high ethylene content segments of the resulting HNBR polymer. This can be achieved by introducing a third diene monomer into the initial NBR polymerisation, since this inhibits the tendency of the ethylene segments to crystallise. An example is the emulsion copolymerisation of ACN, 1,3-butadiene and isoprene, with 1,3-butadiene/ isoprene ratios lying within the range 0.75/1.0 to 1.0/0.75 [4]. The pendant methyl group on the isoprene monomer then disrupts the regularity and crystallisation of the resulting hydrogenated polymer. The reaction sequence and inal polymer structure are indicated in Figure 1.8.

NBR Polymer ----{CH2—CH}x----{CH2—CH=CH—CH2}y----{CH2—CH=CH—CH2}y--{CH2—CH}z ---CN

arylonitrile

CH3

1,4-butadiene addition

CH=CH2

1,2-butadiene addition, vinyl residue

1,4-isoprene addition

Metal catalysts, high pressure +

(2x+2y) H2 (gas)

HNBR Polymer ----{CH2—CH}x----{CH2—CH2 –CH2 —CH2}y ---{CH2—CH=CH—CH2}a ---{CH2—CH}z---CN

acrylonitrile

residual unsaturation for vulcanisation

ethylene higher propylene content due to Isoprene

CH2 –CH3

propylene

Figure 1.8 Hydrogenation reaction of ACN/1,3-butadiene/isoprene terpolymer

The incorporation, for example, of a 1/1 ratio of 1,3-butadiene/isoprene into the starting polymer would result in a roughly 73/27 molar ratio of ethylene/propylene in the inal HNBR polymer (1 mole of 1,3-butadiene results in 0.95 moles of ethylene and 0.05 moles of propylene in the inal HNBR polymer, and 1 mole of isoprene results in 0.5 moles of ethylene and 0.5 moles of propylene):

11

Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology • 1 mole of 1,3-butadiene = 0.95 moles ethylene + 0.05 moles propylene in HNBR polymer. • 1 mole of isoprene = 0.5 moles ethylene + 0.5 moles propylene in HNBR polymer. • Total moles of ethylene = 0.95 + 0.5 = 1.45. • Total moles of propylene = 0.05 + 0.5 = 0.55. • Ethylene/propylene ratio in the inal HNBR polymer = 1.45/(1.45 + 0.55) = 0.73. • Ethylene/propylene ratio in the inal HNBR polymer with no isoprene in the starting polymer = 0.95. The reduced ethylene content in the ethylene–propylene segment gives signiicantly improved low temperature lexibility.

1.4 Summary The background to the technology of the HNBR polymer manufacturing process and a brief review of some of the challenges involved in manufacturing these polymers have been presented.

References 1.

F.R. Mayo and F.M. Lewis, Journal of the American Chemical Society, 1944, 66, 9, 1594.

2.

W.H. Embree, J.M. Mitchell and H.L. Williams, Canadian Journal of Chemistry, 1951, 29, 3, 253.

3.

C. Mouli, R. Madhuranthakam, Q. Pan and G.L. Rempel, AIChE Journal, 2009, 55, 11, 2934.

4.

H. Bender, R. Casper, H.R. Winkelbach, H.C. Strauch, P. Nguyen, S.X. Guo and J. Gamlin, inventors; Bayer AG, assignee, US Patent 7,091,284, 2006.

12

2

Types of Hydrogenated Nitrile Rubber Polymers Available

2.1 Introduction This chapter provides an overview of the types and grades of HNBR polymers commercially available and their performance characteristics in the context of other oil-resistant polymers.

2.2 Summary of Grades Available Table 2.1 summarises the grades of HNBR polymers available from two manufacturers, Zeon Chemicals and Lanxess. The primary sort was made by mole% ACN, a secondary sort by approximate residual unsaturation and a tertiary sort by nominal Mooney viscosity. It is seen that HNBR polymers are available with mole% ACN within the range 17 to 50%, at a variety of residual unsaturation levels and bulk viscosities. For a given mole% ACN content, a higher residual unsaturation igure is useful for providing more rapid sulfur vulcanisation or when greater crosslink density is required in the inished product. A lower bulk Mooney viscosity results in improved high-shear low in transfer or injection moulding.

13

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

14

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

15

Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology In addition to these grades, speciality polymers are available for speciic 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

Modiication with acrylates such as zinc methacrylate gives tough and mechanically durable materials for use in power transmission belts and conveyor belts, and in similar products requiring a high level of resistance to abrasion, cutting and wear.

16

Types of Hydrogenated Nitrile Rubber Polymers Available

2.3 HNBR Grades and Technology One way to illustrate the position of HNBR within the elastomer market is to employ the ASTM D2000 classiication system [1]. This is illustrated in Figure 2.1, along with a selection of other elastomers for comparison. Not only does the HNBR family ill a gap in the properties identiied, HNBR also offer tear resistance, abrasion resistance and overall toughness that polymers such as acrylic elastomers (ACM) and ethylene acrylic copolymer elastomers (AEM) cannot match.

ASTM D2000 Classification % Swell in ASTM IRM–903 Oil

VMQ AEM

EPDM CR

FKM

250

FVMQ

200

ACM HNBR

150 100

NBR

NR

50

Temperature resistance (ºC)

300

0 200

150

100

50

0

Class by Oil Swell

Figure 2.1 Graphical illustration of speciality elastomers, showing the position of HNBR within the ASTM D2000 system

NBR are among the oldest and most widely used oil-resistant polymers and provide an excellent combination of properties and durability, but due to their high level of unsaturation they are prone to oxidation and to attack by sulfur. HNBR materials are much more resistant to oxidation and sulfur attack and combine the compounding lexibility 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 luids, 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

17

Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology end of their useful life before the end of the ageing test. The ACM material showed excellent long-term ageing performance in these luids, but did not have the tear and abrasion resistance of HNBR materials.

% Elongation Loss

Comparison of Oil-Resistant Elastomers, 10W30SG engine oil at 150 °C 40 30 20 10 0 –10 –20 –30 –40 –50 –60

% Elongation Loss, 504 h % Elongation Loss, 1008 h

35% ACN 25% ACN HNBR HNBR

ACM

AEM

Material

Figure 2.2 Comparison of engine oil resistance of HNBR, ACM and AEM elastomers

Comparison of Oil-resistant

% Elongation Loss

Elastomers, Dexron III ATF at 150 °C 25 20 15 10 5 0 –5 –10 –15 –20 –25 –30

% Elongation Loss, 504 h % Elongation Loss, 1008 h

35% ACN 25% ACN HNBR HNBR

ACM

AEM

Material

Figure 2.3 Comparison of resistance to automatic transmission luid (ATF) of HNBR, ACM and AEM elastomers

18

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), luorocarbons (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 Relative resistance to abrasion

200 150 100 50 0 45% ACN HNBR

NBR

ECO

FKM

ACM

Material

Figure 2.4 Comparison of the abrasion resistance of various elastomers

The effects of abrasion, wear and friction are not well understood, but this type of testing is valuable in providing a relative rating for a range of materials. The combination of increased high temperature performance and improved wear resistance emphasises the niche occupied by HNBR among the available oil-resistant elastomers. In the remainder of this chapter the effect of HNBR and compounding ingredients on the inal 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 speciic gravity. This gives the relative cost of the compound required to ill 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 luorosilicones (FVMQ) have excellent upper temperature limits, but their pound–volume cost is heavily inluenced by their relatively high speciic gravity. 19

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 illed by HNBR materials in perspective. HNBR combine toughness, abrasion resistance, luid 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 iller, except for FVMQ

2.4 Summary This chapter has given a brief overview of the polymers currently available and how these it into the overall range of oil-resistant elastomers. Formulation guidelines and examples of formulations for speciic applications will be given in detail later.

References 1.

Rubber, American Society of Testing and Materials, Washington, DC, USA, 2006, 9, 2, D2000-06.

2.

A Comparison of Oil Resistant Elastomers in Engine Oil and ATF, Z7.3.16, Zeon Chemicals LP, Louisville, KY, USA, 1999, p.1.

3.

Zetpol Product Guide, Zeon Chemicals LP, Louisville, KY, USA, 1999.

20

3

The Flow and Processing Behaviour of Hydrogenated Nitrile Rubber

3.1 Introduction In this chapter the principles of polymer viscoelastic flow behaviour are discussed. There are obviously a number of variables, such as compounding, and mould and machinery design, which will influence the flow behaviour of fully compounded HNBR materials. This chapter is not intended to discuss the finer details but will offer some insight into the many factors affecting flow and processing behaviour. This will provide guidelines for studying flow and processing behaviour, as well as indicating some of the useful tools available for solving production processing problems.

3.2 The Basics of Viscoelastic Flow in Rubber

3.2.1 Viscosity Rubber and plastic melts can be regarded as extremely high viscosity fluids. This is only an approximation, however, and polymers generally show viscoelastic properties, a combination of viscous flow and elastic recovery. In turn viscosity is a measure of resistance to flow under a given set of circumstances. Viscosity is designated by the Greek letter η. For an ideal or Newtonian fluid, viscosity is simply the ratio between the shear stress (τ), the pressure applied to the fluid to create flow, and the shear rate (γ), the flow occurring over a given time:

η = τ/γ,

(3.1)

where η = viscosity expressed as Pa. s, τ = shear stress in Pa, and γ = 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

21

Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology as this. However, polymer melts such as rubbers do not in general exhibit uniform viscosity across a range of shear rates, in fact they are described as pseudoplastic, which means that their viscosity is reduced as the shear rate increases. An easy way to visualise pseudoplastic behaviour is to imagine a bottle of thick tomato ketchup. If the bottle is allowed to stand and then upended the ketchup will remain in the bottle, but if the bottle is shaken vigorously the ketchup will then flow freely. At zero shear rate the viscosity (resistance to flow) is extremely high and flow does not take place, but on agitation the shear rate increases dramatically, the viscosity falls and the product can be poured without difficulty. A typical viscosity curve for a rubbery material as a function of shear rate is shown in Figure 3.1. At low shear rates, the material shows Newtonian behaviour, which means that its viscosity remains constant regardless of shear rate. As the shear rate increases, the material is transformed from Newtonian to pseudoplastic, and its viscosity then decreases with increasing shear rate. This is a curve that we will encounter many times later in the chapter and we will see how it can help in understanding flow during moulding and in overcoming flow-related problems. A mental picture of this curve and its shape will help in practical situations.

Newtonian region

Viscos ity (poise )

1000000

100000

10000

1000 1

10

100

1000

Shear rate (s–1)

Figure 3.1 Typical profile of viscosity versus shear rate for a rubber polymer

Once we appreciate that rubber materials are not ideal Newtonian fluids, and that their viscosity decreases with increasing shear rate, we find that these variables have a

22

The Flow and Processing Behaviour of Hydrogenated Nitrile Rubber mathematical relationship. A great deal of scientific analysis has been devoted to the mathematics underlying the flow of rubber and other polymers [1–11], and Equations (3.2) and (3.3) provide a starting point:

η = Kγn–1,

(3.2)

where η is the viscosity, γ is shear rate in s–1 and K and n are a curve-fitting parameters determined from the experimental data.

η = ηo/(1 + Aγ1–n),

(3.3)

where ηo 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.

Viscosity

100000

10000

1000 1

10

100

1000

Shear rate

Figure 3.2 Comparison of theoretical models describing the pseudoplastic behaviour of polymers

23

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 wellunderstood 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(τγw)2]1/6,

(3.4)

where C is a constant, approximately 0.12, τ is shear stress and γw is the shear rate at the wall of the die. Equation (3.4) applies to molten polymers but does not properly describe fully filled or compounded elastomers. It is however useful for quantifying the change in elastic

24

The Flow and Processing Behaviour of Hydrogenated Nitrile Rubber

% Die swell

rebound induced by increasing shear rate and shear stress. Using Equation (3.4) we can establish the proportionality between elastic recovery and shear stress or shear rate. Figure 3.3 shows the effect of a percentage increase in shear rate on the extrudate swell, B, or elastic recovery.

45 40 35 30 25 20 15 10 5 0 1

10

100

1000

–1

Shear rate (s )

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.

25

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(ηT1/T1) – log(ηTg/Tg) = log(aT1),

(3.5)

where ηT1 is the viscosity at absolute temperature T1, ηTg 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.

26

The Flow and Processing Behaviour of Hydrogenated Nitrile Rubber

Viscosity (Pa. s)

1000000 100000 100 °C 10000

125 °C

1000

150 °C

100 1

100

10

1000

–1

Shear rate (s )

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.

∆T = 0∫t (µ/ρc)γ2 dt,

(3.8)

where ∆T is the temperature change in ºC s–1, µ is the viscosity in Pa. s and γ 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:

∆T ∝ η, and ∆T ∝ γ2

(3.9)

27

Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology

% change in heat build-up

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%

0%

50%

100%

150%

% change in parameter Viscosity effect

Shear rate effect

Figure 3.5 Effect of viscosity and shear rate on heat build-up during flow

Using the example from the previous section we have studied the decrease in injection cycle time resulting from increasing the injection pressure and by reducing the injection time from 20 s to 10 s. Since the configuration of the runner system and the mould cavities remained unchanged, a reduction in injection time by half had the effect of doubling the shear rate. In turn, doubling the shear rate caused a four-fold increase in heat build-up during flow. If this increase in heat build-up was sufficient for the critical scorch temperature of the vulcanisation system to be reached, we would encounter premature scorch problems, in addition to dimensional variances. Obviously, as we increase the shear rate the viscosity decreases, causing a linear decrease in heat buildup. However, the dependence of heat build-up on the square of the shear rate can easily override the viscosity drop and give rise to scorching.

28

The Flow and Processing Behaviour of Hydrogenated Nitrile Rubber

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)/(ΣNiMij–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).

29

Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology

Mn 0.5

Mv

Mw

Mz

Fraction

0.4 0.3 0.2 0.1 0 0

100000

200000

300000

400000

500000

600000

700000

Molecular w eight

Figure 3.6 Example of molecular weight distribution of polymers

A great deal of experimental work has been carried out in the past to understand the influence of molecular weight on viscosity [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):

η ∝ M when M < Mc, and η ∝ M3.4 when M > Mc,

(3.11)

where η 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):

30

The Flow and Processing Behaviour of Hydrogenated Nitrile Rubber

ηo = kMw3.4,

(3.12)

where ηo 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 Mw3.4. The data reported by polymer suppliers as Mooney viscosity can thus give an estimate of the relative average molecular weight. As a first approximation we can assume that two polymers varying in Mooney viscosity will have the same degree of branching or linearity, and the viscosity shear rate curves will be parallel. Therefore, if we know the viscosity difference between the two polymers at low shear rates and we also have the viscosity shear rate curve for one of the polymers, we can estimate the viscosity at higher shear rates. For example, if two polymers from the same supplier and made by a similar polymerisation method have Mooney viscosities of 35 and 50, and we know the relative viscosities of the two polymers at high shear rates, these should follow the ratio of Mooney viscosities. In other words, the polymer of Mooney viscosity 50 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.

Molecular weight = 120,000 Mooney viscosity = 93

1000000

Viscosity (poise)

100000 10000

Molecular weight = 80,000 Mooney viscosity = 23

1000 100

Molecular weight = 100,000 Mooney viscosity = 50

10 1

10

100

1000

Shear rate (s–1)

Figure 3.7 Effect of molecular weight on flow behaviour

31

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]:

η1 = η0 (1 + Aφ + Bφ2),

(3.13)

where η0 is the unfilled viscosity, η1 is the viscosity of filled material, φ 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 + Aφ + Bφ2),

(3.14)

where G0 is the shear modulus of unfilled material, G1 the shear modulus of filled material and φ is the volume fraction of filler in the mixture. As can be deduced from Equation (3.13), the type of filler should have neither an impact on viscosity nor on modulus of elasticity. In the practical world that we live in, we know that the loading of N110 SAF carbon black at the same volume or phr (parts per hundred rubber polymer) gives much higher viscosities and much higher modulus values compared to N990 MT carbon black. Thus, Equation (3.13) is not very effective in predicting the effect of fillers. From this point onwards the situation becomes unclear, since comprehensive studies of various fillers and their effect on viscosity and modulus are relatively rare. White and Crowder [27] studied the effect on viscosity of various low-structure carbon blacks and found that those of smaller particle size caused an increase in viscosity and shifted the viscosity relative to the shear rate curves. The use of low-structure carbon blacks limited the structural effect caused by the agglomeration of particles. Unfortunately, many useful commercial elastomer compounds employ high-structure carbon blacks, such as furnace black N550 FEF. In addition, the author has studied the combined effect of structure and particle size of carbon black on the properties of an NBR compound. Obviously the particle size on its own does not determine the results, but fortunately the DBP number, given in

32

The Flow and Processing Behaviour of Hydrogenated Nitrile Rubber ASTM D2414, combines the effect of particle size and structure. The DBP number represents the quantity of dibutyl phthalate (DBP) that the carbon black can absorb, measured by titration to a mechanically sensed end-point; when the carbon black surface can absorb no more DBP the conversion from free-flowing powder to paste is sensed as a change in torque. The experimental design programme is summarised in Table 3.1 and the results for compression modulus and low-shear rate viscosity are shown in Figure 3.8.

Table 3.1 Experimental design formulations, NBR compound, triplicate data points for each formula basic recipe Ingredient Phr 33% Acrylonitrile, 35 ML(1+4)@100C polymer 100.00 Stearic acid 2.00 Agerite resin D 1.00 Vanox ZMTI 1.00 Carbon black (see below) 30.00 to 50.00 Zinc oxide 2.00 Varox DCP-40C 3.00 Compound

Phr carbon black

DBP number of carbon black

1

30.00

43 (N-990)

2

50.00

43

3

70.00

43

4

30.00

65 (N-762)

5

50.00

65

6

70.00

65

7

30.00

90 (N-660)

8

50.00

90

9

70.00

90

10

30.00

121 (N-550)

11

50.00

121

12

70.00

121

33

Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology Response surface for compression modulus in Pa

Compression modulus

15.5 13.5 11.5 9.5 7.5

150 120

5.5 30

40

50

60

70

0

30

60

90 DBP number

phr Carbon black

Figure 3.8 Effect of carbon black on compression modulus

The correlation between the experimental values and the plotted response surfaces was greater than 0.900, confirming that the DBP number gives a good prediction of the reinforcing nature of carbon blacks. The study was not extensive, however, and will need to be developed further to include a comprehensive range of polymer families, carbon blacks and other fillers. Prior to formulating compounds for compression, transfer or injection moulding, the compounder should conduct a similar series of experiments to optimise the fillers that will give a combination of properties such as compression set, strength of vulcanisate and flow in the mould. In the study, elasticity was seen to improve with increasing filler loading and DBP value of carbon black, confirming earlier observations [28, 29]. Loss of elasticity was measured in terms of the hysteresis observed in dynamic compression testing, and is illustrated in Figure 3.9. It is of course well known in the rubber industry that high structure carbon blacks of small particle size give smooth extrusion, which delivers extrusions with smooth or glossy surfaces.

34

The Flow and Processing Behaviour of Hydrogenated Nitrile Rubber

% Hysteresis, Compression

Hysteresis as a function of filler and loading

63 53 43 33 23 30

40

50

60 phr Carbon Black

70

130 110 90 70 50 DBP Number 30

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:

η = φ23.4 F (γφ21.4),

(3.14)

where φ2 is the volume fraction of polymer in the compound, F is a constant for the polymer and plasticiser and γ is shear rate. From Equation (3.16) we can establish the proportionality between plasticised and unplasticised material, as follows:

η1/η0 = (φ13.4F(γφ11.4)/ (φ03.4F(γφ01.4) = φ14.8/φ04.8

(3.15)

35

Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology As more plasticiser is added, viscosity at a given shear rate falls. This is plotted in Figure 3.10, using our example from Figure 3.1. For a particular polymer and plasticiser system it should be straightforward to run the experiments and adapt the parameters of Equation (3.15) to the design of other compounds.

1000000

Viscosity, Pa. s

100000

Phi = 1 Phi = .9

10000

Phi = .8

1000 1

10

100

1000

–1

Shear rate (s )

Figure 3.10 Effect of plasticiser on viscosity

3.4.4 Effect of Process Aids Most process aids are designed to assist mould flow and mould release, and function by having only limited solubility in the polymer matrix. For example, microcrystalline polyethylene waxes make good process aids for highly polar polymers such as fluorocarbon elastomers (FKM) since they are deposited at the surface of the shear flow front and on the moulded parts, assisting lubrication and release. Similarly, specialised FKM polymers are available as process aids and slip agents for use in polyethylene manufacture. In general, effective process aids for mould flow and mould release are used at very low and tightly controlled levels in the polymer. Using our example of microcrystalline wax in FKM materials as a process aid in transfer or injection moulding, while 0.500 phr is very effective, 1.00 phr is also effective but may lead to mould deposits during extended operation involving several hundred cycles, and levels of 1.50 phr or above will give knit lines and parts that are ‘wet’ with excess wax as they emerge from the mould.

36

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 = ∆pe + ∆p,

(3.17)

where PT is the total pressure drop through the die, ∆pe the pressure drop at the ends of the die and ∆p 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 = ∆pe + 4(σ12)w L/D,

(3.18)

where (σ12)w is the shear stress at the wall, L the die length and D the die diameter. By running experiments at constant die diameter and total volumetric flow rate but with different L/D ratios, the slope of the plot of PT versus L/D allows the shear stress at the wall to be ascertained. From this, the effectiveness of process aids in improving flow can be measured in terms of the reduction in shear stress at the wall caused by their lubricating effect. Assessing the effectiveness of a process aid in terms of mould cleanliness and release can be challenging. In many cases the only accurate measure is a carefully monitored extended production run, comparing long-term release and mould cleanliness with and without the process aid, or by comparing different process aids.

3.5 Flow through Sprues and Runners: Mould and Machine Parameters Influencing Elastomer Flow This is an area that seems to lack a basic understanding at the plant and production level in the speciality elastomer industry. A typical problem might involve a multi-deck injection mould where there is insufficient flow to one of the decks. For a variety of

37

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 = ∆P . R / 2L,

(3.19)

where ∆P is pressure drop and R and L are the radius and length of the capillary, respectively.

Shear rate = 4Q / πR3,

(3.20)

where Q and R is the radius of the capillary. Apparent viscosity ηapp = shear stress / shear rate = ∆P . R4 . π / 4 . Q . 2L, and the volumetric flow rate Q = ∆P . R4π / 4η . 2L. app

(3.21)

By only a small increase in the radius of the flow channel the volumetric flow rate will thus increase by the fourth power of the radius. Decreasing the apparent viscosity of the compound, ηapp, will therefore have not nearly as profound an effect on improving flow rate as a small change in the radius or gap of the flow channel. The relative impact of these various factors is illustrated in Figure 3.11. From Equation (3.21) and Figure 3.11, it is clear that a minor change in a sprue or runner system will have a disproportionate effect on rubber flow. In the case of a double-deck injection mould in which one deck is not filling completely, it is obvious that a small increase in the diameter of the sprues or runners leading to the problem deck would provide the most effective solution to the problem. It is strange that the immediate response often seems to be to reduce the viscosity by changing the formulation.

38

Net % change on flow rate

The Flow and Processing Behaviour of Hydrogenated Nitrile Rubber 120.0% 100.0% 80.0% 60.0% 40.0% 20.0% 0.0% –20.0% –40.0% –60.0% –20.0%

–10.0%

0.0%

10.0%

20.0%

30.0%

% Change in factor Radius of runner

Length of runner

Viscosity of material

Figure 3.11 Effect of various parameters on the flow rate of an elastomer

References 1.

E.A. Collins and J.T. Oetzel, Rubber Age, 1970, 102, 64.

2.

E.A. Collins and J.T. Oetzel, Rubber Age, 1971, 103, 3.

3.

S. Einhorn and S.B. Turetzky, Journal of Applied Polymer Science, 1964, 8, 1257.

4.

J.R. Hopper, Rubber Chemistry and Technology, 1967, 40, 462.

5.

M. Mooney, Physics, 1936, 7, 413.

6.

M. Mooney in Rheology: Theory and Applications, Ed., F.R. Eirich, Academic Press, New York, USA, 1958, p.196.

7.

N. Nakajima and E.A. Collins, Polymer Engineering Science, 1974, 14, 137.

8.

F.C. Weissert and B.L. Johnson, Rubber Chemistry and Technology, 1967, 40, 590.

9.

J.L. White and J.W. Crowder, Journal of Applied Polymer Science, 1974, 18, 1013.

39

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.

40

The Flow and Processing Behaviour of Hydrogenated Nitrile Rubber 28. C.D. Han, Journal of Applied Polymer Science, 1974, 18, 821. 29. N. Minagawa and J.L. White, Applied Polymer Science, 1975, 20, 501. 30. G. Kraus and J.T. Gruver, Transactions of the Society of Rheology, 1965, 9, 2, 17.

41

Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology

42

4

Properties of Hydrogenated Nitrile Rubber and Comparison with Other Elastomers

4.1 Introduction In this chapter the properties of hydrogenated nitrile rubber (HNBR) are discussed and compared with those of other speciality elastomers. This background will provide the perspective for later chapters in which the applications of HNBR materials are described.

4.2 Relationship between Hydrogenated Nitrile Rubbers and Other Speciality Elastomers A good starting point for this comparison is to show where HNBR materials it into the classiication used by ASTM D2000 [1], which classiies elastomers by type and class. The ‘type’ classiication 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 irst 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 lexibility (ASTM D2000 BF). Using peroxide curing and suitable antioxidants, NBR can also be made to have a higher upper temperature limit (CH). NBR elastomers generally lie within the range: BF, BG, BK or CH, as illustrated in Figure 4.1.

43

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

ASTM D2000 Classification % Swell in IRM–903 Oil

200 150 100 50

H F E D C B A

NBR

0 200

150

100

50

0

Class by Oil Swell A

B

C

D

E

F

G

H J

K

Figure 4.1 Graphical representation of NBR in the ASTM D2000 system

44

Type by test temperature

250

Temperature resistance (°C)

300

Properties of Hydrogenated Nitrile Rubber and Comparison with Other Elastomers Following hydrogenation the degree of unsaturation in NBR is signiicantly 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.

ASTM D2000 Classification % Swell in IRM–903 Oil

200 150 100 50

H F E D C B A

HNBR NBR

Type by test temperature

250

Temperature resistance (°C)

300

0 200

150

100

50

0

Class by Oil Swell A

B

C

D

E

F

G

H J

K

Figure 4.2 Graphical representation of NBR and HNBR in the ASTM D2000 system

Finally, Figure 4.3 places HNBR within the ASTM D2000 system in the perspective of other speciality elastomers.

45

Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology ASTM D2000 Classification % Swell in ASTM IRM–903 Oil

250

FKM

AEM

EPDM

200 H

FVMQ

VMQ

ACM

150 F

E

HNBR CR

100 D

NBR

NR

C 50 B A

Temperature Resistance (°C)

300

0 200

150

100

50

0

Class by Oil Swell A

B

E

F

G

H J

K

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 illed with N762 carbon black at 50 parts per hundred rubber (phr). The recipe, vulcanisation conditions and test results are summarised in Table 4.2.

46

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)

100

–

–

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

–

5

5

Polymer A (HNBR) Polymer B (NBR)

Zinc oxide Note: phr = parts per hundred rubber polymer

Properties compared

Shore A hardness, ASTM D2240

Results: Results: Recipe A Recipe B (NBR) (HNBR) 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

47

Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology Since the HNBR recipe contained a slightly higher molar percentage of ACN this is not an exact comparison, but it is interesting that HNBR showed noticeably greater luid swelling than NBR. From these data we can see that the hydrogenation of NBR to form HNBR results in: • Better oxidative thermal stability. • Increased swelling in hydrocarbons. Since both NBR and HNBR are now available with a wide range of molar percentages of ACN, the author conducted an extended comparison using the recipes shown in Table 4.3. The results in terms of swelling after immersion in ASTM Reference Fuel C for 70 hours at 23 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)

100

–

–

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

–

HNBR containing required mole% ACN NBR containing required mole% ACN

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 luid. Detailed information on the formulation of HNBR compounds is given in Chapter 6.

48

Properties of Hydrogenated Nitrile Rubber and Comparison with Other Elastomers 90 80

HNBR y=–1.7361x+116.54 R2=0.9996

% Volume change

70 60 50

NBR

40

HNBR

30

Linear (NBR) Linear (HNBR)

NBR y=–1.9652x+97.386 R2=0.9995

20 10 0 –10 0

10

20

30

40

50

60

Mole % ACN

Figure 4.4 Swelling of NHBR and NBR in Reference Fuel C

4.3 Comparison of HNBR with HNB and Other Speciality Elastomers As has been demonstrated earlier, HNBR combines the compounding lexibility 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 luid, 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 conirm that HNBR materials retain their useful life after long-term ageing. ACM and AEM materials show excellent long-term resistance to hydrocarbon luids, but in general do not offer the tear and abrasion resistance of HNBR materials.

49

Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology

Comparison of oil-resistant elastomers, 10W30SG engine oil at 150 °C % Elongation loss

40 20 0

% Elongation Loss, 504 h % Elongation Loss, 1008 h

-20 -40 -60 35% ACN 25% ACN ACM HNBR HNBR Material

AEM

Figure 4.5 Comparison of oil-resistant elastomers in loss of elongation in automotive engine oil at high temperature

Comparison of oil-resistant

% Elongation loss

elastomers, Dexron III ATF at 150 °C 30 20 10 0 -10 -20 -30

% Elongation loss, 504 h % Elongation loss, 1008 h

35% ACN 25% ACN ACM HNBR HNBR Material

AEM

Figure 4.6 Comparison of oil-resistant elastomers in loss of elongation in automatic transmission luid (ATF)

50

Properties of Hydrogenated Nitrile Rubber and Comparison with Other Elastomers An area in which HNBR excels is in its wear and abrasion resistance. This is shown graphically in Figure 4.7 by comparison with other oil-resistant speciality elastomers.

Relative resistance to abrasion

Akron-type abrasion resistance rating 200 150 100 50 0 45% ACN HNBR

NBR

ECO

FKM

ACM

Material Figure 4.7 Comparison of oil-resistant elastomers in relative abrasion resistance

One of the more demanding areas for elastomeric products is in the petroleum exploration and drilling environment. As exploration and drilling become deeper increased temperatures are encountered, making NBR no longer viable for these applications. In addition, many new sources of crude petroleum contain signiicant levels of hydrogen sulide, 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 sulide [4]. The results for amine exposure are replotted in Table 4.4.

51

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% luorine), bisphenol cured

–40

Type 2 FKM (70 mole% luorine), peroxide cured

–37

NBR (41 mole% ACN), Semi-EV sulfur cured

–26

Note: Test luid: IRM–902 oil containing 1% NACE Amine B corrosion inhibitor Test immersion conditions: 168 h at 150 °C

It is seen from Table 4.4 that HNBR materials showed excellent retention of lexibility 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 sulide) 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 conirm 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 speciic gravity, and gives a relative idea of the cost of the parts produced. These values are based on the published prices of raw materials in late 2010.

52

Properties of Hydrogenated Nitrile Rubber and Comparison with Other Elastomers 120

% Elongation retained

100 36%ACN HNBR, peroxide cured

80

45%ACN HNBR, peroxide cured

60

FEPM, peroxide cured

40

Type 1 FKM (66 mole% Fluorine), Bisphenol cured Type 2 FKM (70 mole% Fluorine), peroxide cured

20

NBR (41 mole%ACN), Semi-EVSulfur cured

0 0

50

100 150 Ageing time (h)

200

Figure 4.8 Comparison of various elastomers for resistance to sour gas at 23 ºC

Table 4.5 Cost comparison of oil-resistant elastomers Specific gravity Cost of lb–volume Upper operating Polymer of compound compound cost 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.

53

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.

54

5

Typical Applications of Hydrogenated Nitrile Rubber

Introduction Previous chapters have described the manufacture, chemistry, the grades available and basic properties of HNBR compared to other speciality elastomers. The present chapter provides examples of its applications, for which formulations are given in Chapter 7.

5.1 A Brief Recap Figure 5.1 illustrates the location of HNBR within the ASTM D2000 classiication system for rubbers. This provides the context for the applications for which HNBR is important and shows that it has good resistance to hydrocarbon luids, 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 lexible dynamic rubber components.

55

Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology ASTM D2000 Classification 300 250

FKM

AEM

EPDM

200 H

FVMQ

VMQ

ACM

150 F

E

HNBR CR

100 D

NBR

NR

C

50 B A

Temperature resistance (°C)

% Swell in ASTM IRM–903 Oil

0 200

150

100

50

0

Class by Oil Swell A

B

E

F

G

H J

K

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 lex 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 lex life of HNBR. Similarly, HNBR is much less susceptible to oxygen and ozone attack, and this also greatly extends its lex life. Acrylate modiication of HNBR (Chapter 2) further improves its toughness, abrasion resistance and lex life. The enhanced high temperature resistance of HNBR materials also helps to extend the life of components in which long-term exposure to high temperatures is important for improving the durability of elastomeric components.

56

Typical Applications of Hydrogenated Nitrile Rubber This applies for example in automotive engine compartments, where elastomers such as NBR and CR do not have suficient thermal and oxidative resistance to fulil the target life of the components. Typical belt and conveyor applications for HNBR and acrylate-modiied HNBR vulcanisates include the following: • Fabric-reinforced belts for automotive power accessories: –

Timing belts.

–

Drive belts for air conditioning compressors, alternators and hydraulic steering pumps.

–

Belt-driven power take-off pumps.

• Conveyor belting, either homogeneous rubber or fabric-reinforced: –

Acrylate-modiied HNBR for conveying abrasive powders, including minerals and grain.

• Rubber rollers: –

In paper printing, especially acrylate-modiied HNBR.

–

Rollers for conveyor systems handling abrasive minerals, grain or grain byproducts.

5.3 Flexible Boots for Power Transmission Joints As mentioned earlier, HNBR materials offer excellent abrasion resistance and lex 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 luids. This combination makes it a good choice for lexible 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, olein/rubber blends and urethanes run the risk of partial melting. One could, for example, envisage a protective power take-off boot on an all-wheeldrive vehicle where the boot is close to the exhaust pipe or a catalytic converter. In such cases the temperature near the boot might easily reach 120 °C, which would

57

Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology be too high for the thermoplastic elastomers traditionally used. HNBR materials offer a good balance of properties for high temperature protective boot and bellows applications, including: • Drive train boots in automotive and truck applications: –

Constant-velocity joints near exhaust headers.

–

Axle joints close to exhaust pipes or catalytic converters.

• Steering boots: –

Rack and pinion ends near exhaust headers.

–

Input shafts close to the engine.

• Suspension components: –

Tie-rod end boots and strut end covers near exhaust systems.

5.4 Static Seals for Power Transmission Fluids From Figure 4.3 we can see that the properties of NBR are supplemented in HNBR by good resistance to hydrocarbon luids together with enhanced performance at high temperatures. Many of the static seals used in power transmission, such as in powerassisted steering, automatic transmission and differentials, have traditionally been based on NBR due to its excellent hydrocarbon luid 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 luid pressure resistance, especially in hydraulically assisted steering and power transmission systems. In addition, the luids used in many power transmission applications have been upgraded by the inclusion of antioxidants, sludge dispersants and metal protective additives. The reactive organic sulfur or phosphorus compounds included as metal protective additives may themselves act as vulcanisation chemicals in unsaturated elastomers, and the saturation of HNBR makes it much more resistant to the hardening effect of such additives.

58

Typical Applications of Hydrogenated Nitrile Rubber Examples of static seals used in power transmission include: • O-rings and square cross-section seals for long-life hydraulic power steering gaskets, pumps and reservoirs: –

Pump seals.

–

Fluid tube itting seals.

–

Fluid reservoir seals.

–

Rack and pinion luid seals.

• O-rings, gaskets and cross-section seals for automatic and manual transmission luids, differential joints and pan seals.

5.5 Dynamic Fluid Seals for Power Transmission Due to its resistance to hydrocarbon luids and additives, combined with its enhanced thermal resistance and excellent abrasion resistance, HNBR is an ideal alternative to NBR for radial shaft seals in power transmission applications. In this application dust exclusion and abrasion resistance are vital, and HNBR has found use in the following power transmission applications: • Input and output radial shafts in automatic and manual transmission systems. • Input shafts and reciprocating racks in hydraulic power-assisted rack and pinion steering systems. • Radial differential shafts. • Hydraulic power-assisted steering pumps.

5.6 Hydraulic Fluid and Lubricant Hoses Good resistance to hydrocarbon luids, oxygen and ozone, combined with thermal stability and abrasion resistance make HNBR valuable for hoses containing hydrocarbon hydraulic luids and lubricants, for which NBR is prone to luid additive attack at high temperatures. HNBR can be used in the liner, the reinforcing layer or the cover for such hoses.

59

Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology

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: • Static seals: –

O-rings.

–

Square cross-section seals.

–

Gaskets (homogeneous rubber or composite-reinforced).

• Dynamic seals: –

Radial shaft seals for rotation or reciprocation.

5.8 Seals used in Petroleum Exploration and Drilling In common with other applications, a combination of thermal stability, resistance to aggressive chemicals and abrasion resistance make HNBR an excellent choice in petroleum handling, for example: • Packer elements and blow-out preventers. • O-rings: –

Packer element static seals.

–

Tubing seals.

–

Pump static seals.

–

Drill bit seals.

• Pump radial shaft seals.

5.9 Other applications These include:

60

Typical Applications of Hydrogenated Nitrile Rubber • Static seals for hydrocarbon luid cooler systems (e.g., engine oil, or power transmission luids): –

O-rings.

–

Square cross-section seals.

–

Gaskets.

• Flexible protective bellows on mechanical face seals. • Engine fuel static seals (away from the heat of the engine): –

O-rings for fuel injector return lines.

References 1.

R.W. Keller, Rubber Chemistry and Technology, 1985, 58, 3, 637.

61

Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology

62

6

Formulating Guidelines for Hydrogenated Nitrile Rubber

6.1 Introduction In this chapter the general guidelines for HNBR elastomer formulations are discussed, including those for vulcanisation, iller and plasticiser systems.

6.2 Vulcanisation Systems In Chapters 1 and 2 the manufacture, chemistry and the available grades of HNBR polymers were discussed. Most commercial HNBR polymers contain a small degree of residual unsaturation in the polymer backbone. This provides the opportunity for vulcanisation, either using accelerated sulphur- or peroxide-based systems. The most reactive areas of unsaturation in NBR polymers naturally provide preferential sites for catalytic hydrogenation. It follows that any residual unsaturation in a HNBR polymer is likely to be of a less reactive nature. This means that in the case of either accelerated sulfur or peroxide vulcanisation a much higher level of crosslinking agent and accelerator will normally be required for HNBR than for NBR. A general study of sulfur-cured systems has been conducted by Zeon Chemicals, and the results are summarised in Table 6.1 [1]. The Zeon study was carried out using a base polymer with approximately 9% residual unsaturation. From the available polymer grades listed in Chapter 2 it is seen that very low residual unsaturation may not be adequate for accelerated sulfur vulcanisation. Zeon has also conducted an interesting study on the peroxide plus coagent vulcanisation of a polymer of medium ACN level and 4% residual unsaturation [2]. The results of this study are summarised in Table 6.2 and are plotted in Figures 6.1 to 6.4.

63

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

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

0.5

0.3

1.0

1.0

Zetpol 1020

Sulfur (325 mesh) 0.50

4,4′-Dithiomorpholine

0.50

Dipentamethylene thiuram hexasulide Tetramethylthiuram disulide

2.0

N-cyclohexyl-2-benzothiazole sulfenamide

1.0

2-Mercaptobenzothiazole

2.0

2.0

2.0

0.7

0.7

0.7

0.7

0.7

0.7

0.5

0.5

0.5

0.5

0.5

0.5

0.7

0.5

Zinc dimethyl dithiocarbamate Zinc dibutyl dithiocarbamate

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

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

10

11

Shore A hardness

Compression set, ASTM D395 Method B, 25% compression

Recipe from above 1

64

2

8

9

70 h at 100 ºC

35

34

37

41

42

41

70 h at 120 ºC

54

46

46

56

57

57

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

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

Zetpol 2010

Ingredient

phr in recipe 7

8

9

10

11

12

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 5methylmercaptobenzimidazole

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

Zetpol 2010

Properties after vulcanisation 15 min at 170 ºC Property

Recipe 1

2

3

4

5

6

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

Shore A hardness

65

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

11

12

Property

Recipe 7

8

9

10

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: • Shore A hardness = 74.70 + 0.75P + 4.0C + 0.5P2 – 0.375C2 + 0.15CP,

(6.1)

where P = phr α,α-bis(t-butylperoxy)diisopropylbenzene peroxide, and C = phr N,N′-m-phenylene dimaleimide coagent.

R-squared, adjusted for degrees of freedom = 97.74%.

• 100% modulus = 1.774 – 0.052P – 0.034C + 0.006P2







+ 0.0187C2 + 0.0232PC



R-squared, adjusted for degrees of freedom = 99.54%.

• % elongation = 711.6 – 49.6P – 76.9C + 1.0P2 + 3.73C2 + 2.96PC

(6.2)

(6.3)

R-squared, adjusted for degrees of freedom = 97.70%.

• % compression set = 84.1 – 6.18P – 12.3C + 0.12P2 + 0.56C2 + 0.688PC (6.4)

66

R-squared, adjusted for degrees of freedom = 88.03%.

Formulating Guidelines for Hydrogenated Nitrile Rubber

78

7 6

Coagent phr

5 4

76 74

3 2 1 72

0 5

6

7 8 Peroxide phr

9

10

Figure 6.1 Contour plot of Shore A hardness as a function of peroxide and coagent level

3.5

7

4.0

6

Coagent phr

5

2.5 3.0

4 3 2 1

2.0

0 5

6

7 8 Peroxide phr

9

10

Figure 6.2 Contour plot of modulus at 100% elongation as a function of peroxide and coagent level

67

Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology

7 6 250

Coagent phr

5 4 200

3 350

2 1 300

450 400

0 5

6

7 8 Peroxide phr

9

10

Figure 6.3 Contour plot of ultimate elongation as a function of peroxide and coagent level

7 6

Coagent phr

5 4 30

3 25

2

40

1 50

35 45

0 5

6

7 8 Peroxide phr

9

10

Figure 6.4 Contour plot of compression set as a function of peroxide and coagent level

68

Formulating Guidelines for Hydrogenated Nitrile Rubber Oven post-cure processes are also effective with HNBR compounds, as illustrated in a further study conducted by Zeon (Table 6.3 and Figure 6.5) [3].

Table 6.3 Zeon study on the effect of post-cure: Recipes and properties [3] Ingredient

phr in recipe 1

Zetpol 1020

2

100.0 100.0

3 100.0

4

5

6

7

8

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 disulide

2.5

2.5

2.5

2.5

2-Mercapto benzothiazole

0.5

0.5

0.5

0.5 8.0

8.0

8.0

8.0

40% Dicumyl peroxide on an inert carrier 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

69

Compression Set (70 h at 120 °C)

Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology 60 50 40 Sulfur

30

Peroxide

20 10 0 0

2

4

6

8

10

Post-cure time (h at 150 °C)

Figure 6.5 Effect of post-cure on compression set

6.3 Carbon Black Fillers Carbon black is one of the most universally used illers 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 Zetpol 2010 Carbon black (various grades)

phr 100.0 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

70

Formulating Guidelines for Hydrogenated Nitrile Rubber The results for various carbon black levels and grades, along with nominal DBP absorption values, are illustrated in Figures 6.6 to 6.8.

Shore A

80 70 60

125

50

100 75 0

20

40 phr

DBP

50 60

Figure 6.6 Surface plot showing effect of carbon black on Shore A hardness

100% modulus

12

8

4

125 100

0

75 0

20 phr

40

DBP

50 60

Figure 6.7 Surface plot showing effect of carbon black on modulus at 100% elongation

71

Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology

ML(1+4) at 100 °C

150

100 125 100 50 0

75 DBP 20

40 phr

50 60

Figure 6.8 Surface plot showing the effect of carbon black on Mooney viscosity

The predictive equations for these properties as a function of the phr and DBP number of the carbon black are as follows: • MV = Mooney viscosity, ML(1+4) at 100 °C

= 60.16 – 0.929p + 0.746D + 0.00986p2 – 0.000540D2 + 0.0120pD,

(6.5)

where p = phr carbon black and D = DBP number of carbon black.



R-squared, adjusted for degrees of freedom = 89.66%

• H = Shore A durometer hardness

= 34.36 + 0.264p + 0.372D – 0.00071p2 – 0.00202D2 + 0.00207pD





(6.6)

R-squared, adjusted for degrees of freedom = 91.19%

• M1 = stress at 100% elongation (100% modulus) in MPa



= – 0.0804p + 0.0593D + 0.00149p2 – 0.000360D2 + 0.000808pD





R-squared, adjusted for degrees of freedom = 91.94%

72

(6.7)

Formulating Guidelines for Hydrogenated Nitrile Rubber It is appreciated that these equations bear little resemblance to the theoretical equations for iller 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 illers on viscosity and modulus (Chapter 3), the following theoretical equation was presented:

G1 = G0 (1 + Aφ + Bφ2)

(6.8)

where G0 = shear modulus of unilled material, G1 = shear modulus of illed material, and φ = volume fraction of iller in the mixture. Equation (6.6) might suggest that the level of iller was the only important factor in reinforcement, and not the nature of the iller. 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 inished 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 inal assembly. If the NBR and HNBR rubber articles were both black it would be dificult for the customer to monitor the changeover from NBR to HNBR. It might therefore be desirable to make the inished HNBR rubber article a different colour, say green, to avoid mistakes in assembly and warranty situations. Mineral illers are important for producing colours other than black in the inished rubber articles. Here again Zeon have provided a comprehensive study of non-black illers in a wide range of particle sizes and reinforcing ability [6]. The test recipe and results are summarised in Table 6.5.

73

Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology

Table 6.5 Zeon study of non-black fillers Ingredient

phr

Relative particle size

Shore A

100% modulus

50

1.1

∆ Shore A per phr

∆ Modulus at 100% per phr

None

0

Ultra-Plex

90

small

65

1.6

0.167

0.0056

Roy-Cal L

90

medium

65

1.8

0.167

0.0078

Zeeospheres 200

90

medium

66

2.4

0.178

0.0144

Imsil 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 Zetpol 2010

phr 100.0

TOTM, Plasthall

5.0

Titanium dioxide

5.0

74

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 illers 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 iller. A typical example might be in the production of a green article of 70 nominal Shore A hardness. A mineral iller with a relatively low reinforcing factor is needed, and a higher level of iller might be used to keep down the cost. We know from Table 6.5 that our base recipe without iller would give 50 Shore A hardness. By selecting a couple of mineral illers from Table 6.5 the reinforcing factor from the table can be used as follows:

Phr Nulok 390 = (70–50)/0.256 = 78 phr



Phr HiSil 233 = (70–50)/0.450 = 44 phr.

If we needed to maximise the iller loading to keep the cost down, then Nulok 390 would be a better choice than HiSil 233 as a non-black iller, assuming the other properties were satisied.

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 signiicantly higher temperature limit and this means that the more volatile plasticisers suitable for NBR may not be practical for HNBR. Zeon has conducted a study showing the effect of a variety of plasticisers used in a simple HNBR recipe [7]. Some of the data are given in Table 6.6, together with the softening factors calculated by the author for each plasticiser.

75

Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology

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

76

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

–1.4

–9.7

–8.6

–6.1

–3.4

–1.7

Heat resistance weight loss, 168 h at 150 ºC

Softening Effects ∆ Shore A per phr plasticiser

–0.4

–0.5

–0.5

–0.4

–0.5

∆ Mooney viscosity per phr plasticiser

–3.1

–3.7

–3.5

–2.8

–3.1

∆ 100% Modulus per phr plasticiser, MPa

–0.140

–0.187

–0.193

–0.133

–0.127

∆ 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(tbutylperoxy) diisopropylbenzene, 40% on inert carrier

6

77

Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology Trioctyl trimellitate (TOTM) is one of the most common plasticisers used in HNBR formulations and the reasons for this are clear from Table 6.6, which shows that TOTM offers a good balance of raw compound viscosity reduction, reduction of low temperature properties as measured by TR–10, and excellent thermal stability in terms of low weight loss at 150 °C. The softening effect listed in Table 6.6 should help in selecting the correct plasticiser and quantity, as well as determining the amount of additional iller 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-2,2,4trimethyl 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 Tackiiers 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 eficient internal lubricants in HNBR for this purpose. Commercial varieties of such oleamides include Armoslip CP and Kenamide E.

78

Formulating Guidelines for Hydrogenated Nitrile Rubber

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 lexibility and weather resistance, with some sacriice in resistance to swelling in hydrocarbon luids. Generally, 10–20 phr of EPDM elastomer, with medium to high third monomer content to assist vulcanisation, will give improvements in low temperature lexibility and weather resistance without loss of hydrocarbon luid 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 rubbercovered 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 inished seal cannot be allowed to shrink due to plasticiser leaching into the power steering hydraulic luid. With this knowledge in hand we are able to design a starting compound as follows: • Polymer: Zetpol 4310 EP (17 mole% acrylonitrile), to give the best low temperature lexibility, selecting the low-viscosity grade for injection moulding. • Cure system: Peroxide plus coagent system for compression set resistance: 10.0 phr α,α-bis(t-butylperoxy)diisopropylbenzene peroxide with 7.50 phr N,N′-mphenylene dimaleimide. • Carbon black: N550 (based on past experience with NBR seals in this application). Shore A hardness H is provided by Equation (6.6) above:

H = 34.36 + 0.264p + 0.372D – 0.00071p2 – 0.00202D2 + 0.00207pD.

The value of D for N550 carbon black is 121, so we can solve for phr knowing

79

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. • 1.5 phr 1,2-dihydro-2,2,4-trimethyl quinoline antioxidant. • Past experience with HNBR compounds has shown that 5.00 phr French process zinc oxide and 1.0 phr stearic acid provide excellent cure activation and stability. This gives the starting recipe in Table 6.7

Table 6.7 Possible recipe for reciprocating lip seal phr

Ingredient

Comments

Zetpol 4310 EP

100.0

Very good low temperature lexibility 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 iller

α,α-Bis(t-butylperoxy)diisopropylbenzene peroxide

10.0

Eficient crosslinking, low compression set

N,N′-m-Phenylene dimaleimide coagent

7.5

Eficient 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 lexibility is not required in this application. On the other hand, since the article is an O-ring compression set is a primary functional

80

Formulating Guidelines for Hydrogenated Nitrile Rubber requirement. We also need to keep the cost of the compound as low as possible to lessen the impact of the change from NBR to HNBR. We wish to select a mineral iller capable of high loadings, and will therefore use 10 phr of TOTM plasticiser to allow a higher iller content without seriously sacriicing high temperature compression set. The target Shore A durometer reading for this compound is 75. • Polymer: Therban 3446 (34 mole% ACN, 4% residual unsaturation for high crosslink density and good compression set resistance. • Heat-stable pigmentary system: 1.00 phr N990 carbon black plus 4.00 phr red iron oxide powder. • Filler: Zeeospheres 200 to allow high iller loading: We require 25 points Shore A from the polymer and a further 4 points Shore A from the iller to offset the effect of the plasticiser: Total iller loading = (25+4)/0.178 = 163 phr (Table 6.5 and Table 6.6). The resulting starting recipe is given in Table 6.8.

Table 6.8 Possible compound for light brown O-ring for a hydraulic system Ingredient Therban 3446

phr

Comments

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

Eficient crosslinking, low compression set

N,N′-m-phenylene dimaleimide coagent

7.50

Eficient crosslinking, low compression set

81

Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology These examples will give the reader some idea how to use this information to design formulations. In the chapter following we will discuss specimen formulations for speciic applications of HNBR. We shall also discuss the value of statistical experimental design and desirability functions for studying formulation variables and selecting the optimal formula for the application concerned.

References 1.

Zeon Chemicals LP, Semi-EV and EV Curing Systems for Zetpol 1020 (Z5.1.2), Louisville, KY, USA, 1999.

2.

Zeon Chemicals LP, Study of Level and Ratio of Peroxide and Co-Agent in Zetpol 2010 (Z5.1), Louisville, KY, USA, 1999, p.1.

3.

Zeon Chemicals LP, Effects of Post Curing on Zetpol 1020 (Z5.1), Louisville, KY, USA, 1999, p.1.

4.

Zeon Chemicals LP, Zetpol HNBR Elastomers: Carbon Black Study (SA–26), Louisville, KY, USA, April 2001, p.1.

5.

R.W. Keller, Hydrogenated Nitrile Rubber, in Handbook of Specialty Elastomers, Ed., R.A. Klingender, CRC Press, Boca Raton, FL, USA, 2008.

6.

Zeon Chemicals LP, Evaluation of Larger Particle Non-Black Fillers in Peroxide Cured Zetpol 2010 (Z5.3.3), Louisville, KY, USA, 1999, p.1.

7.

Zeon Chemicals LP, Low Temperature Properties of Zetpol and Influence of Plasticizers and Tackifiers (Z5.4), Louisville, KY, USA, 1999, p.1.

8.

Zeon Chemicals LP, Protection Systems for Zetpol (Z5.5), Louisville, KY, USA, 1999.

82

7

Examples of Hydrogenated Nitrile Rubber Formulae for Specific Applications

7.1 Introduction This discussion opens with an overview of statistical experimental design and the use of the desirability function. If there was ever an area where the use of statistical experimental design techniques was desperately needed, it is in the formulation of speciality elastomers. The author has personally encountered a situation in which variant 61 of an experimental formula was the one that inally 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 ictitious 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 irst of these contains a relatively brief but very adequate description of the techniques and calculations and will give the reader a quick start in these techniques. The second book provides a more comprehensive background to statistical experimental design. As mentioned above, the variables in a rubber formulation may interact, and statistical experimental design can detect and quantify these interactions, particularly when full factorial design and response surface designs are employed. These techniques are illustrated below using a theoretical example. Starting compounds and full factorial design are discussed and employed in other examples in this chapter. In addition to statistical experimental design, the desirability function concept is also useful in evaluating rubber formulations [3, 4]. The basics of the desirability function are given by Equation (7.1).

83

Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology n

n

∏ (d1, d2, d3, ... dn)

D=

d=1

,

(7.1)

where D is a composite desirability number between 0 and 1.00 (0 = completely undesirable and 1.00 = completely desirable), d1 is the desirability of property 1 in the range 0 to 1.00, d2 is the desirability of property 2 in the range 0 to 1.00 and dn is the desirability of property n in the range 0 to 1.00. The individual desirabilities can have pass/fail, linear or curvilinear functions. An example of a pass/fail property might be ozone resistance (ASTM D1171) above 168 h, any cracking being considered undesirable (0) and total absence of cracking being completely desirable (1.00). The dn relationships are usually arrived at after full involvement of the customer. The remainder is a straightforward calculation. As an example, let us imagine that we are measuring ive 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 d1

Formulation

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=

84

5

(d1 * d2 * d3 * d4 * d5)

(7.2)

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 speciic 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 speciic requirement. On an allwheel 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 lex life, resistance to the joint grease, and wear and abrasion resistance, all at this operating temperature. From the ASTM D2000 classiication chart in Chapter 4, shown in Figure 7.1, it is clear that HNBR is the correct choice.

300

200 150 100 50

FKM

Temperature resistance (°C)

250

H

VMQ FVMQ F ACM E D HNBR C NBR B

AEM EPDM CR NR

0 200

A

150

100

50

Type by test temperature

ASTM D2000 Classification % Swell in Irm-903 Oil

0

Class by Oil Swell A

B

C

D

E

F

G

H J

K

Figure 7.1 Graphical presentation of ASTM D2000 85

Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology Working with the customer, we have identiied the individual desirabilities shown in Table 7.2.

Table 7.2 Definition of desirability of individual properties, theoretical joint boot compound d1

Deinition: d1 = 1.00 if weight loss after 1,000 hours at 125 °C in dry heat is 0–5%

d2

Deinition: 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

Deinition: 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

Deinition: d4 = 1.00 if dynamic lex life rating is < 3 after 1,000,000 cycles; d4 = 0.00 if dynamic lex life rating is ≥ 4 after 1,000,000 cycles

d5

Deinition: 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

Deinition: 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)

1 0.8 d3

0.6 0.4 0.2 0

–52

–47

–42

TR-10, degrees C

A

1 0.8 d6

0.6 0.4 0.2 0 0 B

86

20

40

60

80

100

% Comp. set, 1,008 hrs. @ 125C in grease

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 lex 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 low; medium to high third monomer level for eficient 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-2,2,4trimethyl quinoline antioxidant

1.0

Optimal antioxidant and antiozonant balance

Parafin wax

2.0

Optimal antioxidant and antiozonant balance

N762 Carbon black

55.0

Reasonable reinforcement while maintaining high elongation for lex resistance

Trioctyl trimellitate plasticiser

10.0

Reduction of compound viscosity for injection moulding; improvement of low temperature lexibility; good heat stability

Dicumyl peroxide, 40% on inert mineral carrier

3.0– 5.0

Fast and eficient curing in injection moulding where cycle time is critical

Optimise level of peroxide for best balance of properties

87

Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology

phr peroxide

The reader may be daunted at the prospect of making up and testing as many as nine formulae. On the other hand, guessing a formula and hoping it will perform well in all respects will almost certainly involve making up and evaluating many more formulations than this. The experiments are illustrated graphically in Figure 7.2.

5.00 4.80 4.60 4.40 4.20 4.00 3.80 3.60 3.40 3.20 3.00 0.00

10.00

20.00

phr EPDM in polymer blend Figure 7.2 Graphical depiction of HNBR joint boot compound experimental design

It should be noted that one of the key principles is to randomise the order of the experiments to eliminate bias, in this case in either the mixing or the testing. We therefore submit the formulae to the mixing and testing laboratories in a random order, using computer software to plot the results for each of the key properties involved, as illustrated in Figures 7.3 to 7.6. All formulae are found to pass the dynamic lex life requirement and they all give a 100% quality retention rating on accelerated ozone exposure testing.

88

Examples of Hydrogenated Nitrile Rubber Formulae for Specific Applications 5.0 –2.4

–2.8

phr peroxide

4.5

4.0

–3.2

3.5

3.0

–2.6

–3.4

–3.0

0

5

10

15

20

phr EPDM

Figure 7.3 Experimental design of HNBR joint boot compound: Surface plot showing the effect of peroxide and EDPM on weight loss at 125 ºC

5.0 –0.5

0.5

0.5

–1.5

phr peroxide

4.5

4.0

3.5

–2.0

–1.0

0.0

1.0

10 phr EPDM

15

3.0 0

5

20

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

89

Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology 5.0

–46.5

–48.0

–43.5

phr peroxide

4.5

4.0

3.5

–45.0

–42.0

3.0 0

5

10 phr EPDM

15

20

Figure 7.5 Experimental design of HNBR joint boot compound: Results for low temperature retraction, TR–10

5.0

50

4.5

phr peroxide

70 4.0

3.5

80

60

3.0 0

5

10 phr EPDM

15

20

Figure 7.6 Experimental design of HNBR joint boot compound: Effect of peroxide and EDPM on compression set (Cset)

90

Examples of Hydrogenated Nitrile Rubber Formulae for Specific Applications When we analyse the nine formulae in terms of our desirability function and calculate a composite desirability value for each formula, this gives the results summarised in Table 7.4.

Table 7.4 Experimental results for HNBR flex boot experimental design Factors

Responses

phr EPDM

phr peroxide

d1

d2

d3

d4

d5

d6

D

0.00

3.00

1.00

0.00

0.05

1.00

1.00

0.15

0.00

0.00

3.00

1.00

0.00

0.11

1.00

1.00

0.14

0.00

0.00

3.00

1.00

0.00

0.08

1.00

1.00

0.13

0.00

10.00

3.00

1.00

1.00

0.35

1.00

1.00

0.32

0.69

10.00

3.00

1.00

1.00

0.30

1.00

1.00

0.32

0.68

10.00

3.00

1.00

1.00

0.32

1.00

1.00

0.31

0.68

20.00

3.00

1.00

1.00

0.55

1.00

1.00

0.45

0.79

20.00

3.00

1.00

1.00

0.56

1.00

1.00

0.44

0.79

20.00

3.00

1.00

1.00

0.61

1.00

1.00

0.45

0.81

0.00

4.00

1.00

0.00

0.12

1.00

1.00

0.25

0.00

0.00

4.00

1.00

0.00

0.15

1.00

1.00

0.24

0.00

0.00

4.00

1.00

0.00

0.16

1.00

1.00

0.24

0.00

10.00

4.00

1.00

1.00

0.46

1.00

1.00

0.40

0.75

10.00

4.00

1.00

1.00

0.50

1.00

1.00

0.40

0.76

10.00

4.00

1.00

1.00

0.50

1.00

1.00

0.40

0.77

20.00

4.00

1.00

1.00

0.72

1.00

1.00

0.52

0.85

20.00

4.00

1.00

1.00

0.72

1.00

1.00

0.52

0.85

20.00

4.00

1.00

1.00

0.75

1.00

1.00

0.52

0.86

0.00

5.00

1.00

0.00

0.20

1.00

1.00

0.40

0.00

0.00

5.00

1.00

0.00

0.25

1.00

1.00

0.40

0.00

0.00

5.00

1.00

0.00

0.21

1.00

1.00

0.39

0.00

10.00

5.00

1.00

1.00

0.58

1.00

1.00

0.50

0.81

10.00

5.00

1.00

1.00

0.58

1.00

1.00

0.49

0.81

10.00

5.00

1.00

1.00

0.60

1.00

1.00

0.50

0.82

20.00

5.00

1.00

1.00

0.81

1.00

1.00

0.61

0.89

20.00

5.00

1.00

1.00

0.81

1.00

1.00

0.60

0.89

20.00

5.00

1.00

1.00

0.85

1.00

1.00

0.61

0.90

91

Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology Table 7.4 shows that the formulae with no EPDM blended with the HNBR polymer has zero composite desirability, since these all show shrinkage on high temperature exposure to the customer’s axle grease, and shrinkage in this test is deined as a zero individual desirability. Finally, we can plot the composite desirability (D) values to ind the ideal formula from the nine under examination.

composite D

0.75 0.50 0.25

5

0.00

4 phr peroxide

0 10 phr EPDM

3 20

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.

92

Examples of Hydrogenated Nitrile Rubber Formulae for Specific Applications The following sections illustrate further examples of formulae for HNBR in speciic 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 insuficient 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 luids 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 luid resistance. The optimisation factors can then be combined in an experimental design with response surface techniques for inal development of the compound.

Responses (properties) to be measured at 125–150 ºC include: • Resistance to diesel engine oil. • Resistance to water/glycol coolant. • O-ring compression set resistance.

93

Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology

Table 7.5 Oil cooler O-ring compound experimental design Ingredient

phr

Logic/comments

HNBR polymer with 34–36% ACN; 4–5.5% residual unsaturation

100.0

Best overall balance of luid resistance and low temperature properties; the higher levels of unsaturation give the most eficient 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

Varox VC40KE

10.0

Peroxide for compression moulding at 180 ºC

Vanax MBM (VARIABLE)

4.5–7.5

Coagent for increased eficiency of cure and best compression set resistance

Optimisation: 2-factor 3-level full factorial design; 8 compounds and experiments

94

Experimental design optimisation

Evaluate two carbon black types for best overall balance of properties

Optimise coagent levels for best balance of properties

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 sulide, aggressive amines and water. Fluorocarbon elastomers have outstanding high temperature resistance but their inability to withstand aggressive amines rules them out. Conventional NBR have the required oil, amine and water resistance, but cannot operate at 125 °C with excursions to 150 °C. When the packer element seals the well very high pressure differentials may be encountered, and we need to formulate a tough material of high hardness. The packer element has a relatively simple cylindrical shape, and compression moulding is therefore suitable. Its critical properties are heat resistance and pressure, as indicated above, plus resistance to crude oil, hydrogen sulfide, steam and aggressive amines. HNBR show a smaller drop in modulus of elasticity and stiffness at increasing temperatures, and a HNBR packer retains its pressure resistance better than alternative materials at high temperature. A starting compound is shown in Table 7.6, with variables listed for further study and optimisation using experimental design, desirability functions and response surface methods. Responses (properties) to be measured include: • Heat resistance, 125–150 ºC. • Crude oil resistance. • Hydrogen sulide resistance. • Steam resistance. • Aggressive amine resistance. • Pressure extrusion resistance.

95

Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology

Table 7.6 Oil well packer compound experimental design Ingredient

phr

Logic/comments

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

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 eficient vulcanisation

Optimisation: 2-factor 2-level full factorial design with centre point; 5 compounds and experiments

96

Experimental design optimisation

Evaluate two different carbon blacks for best overall balance of properties

Optimise compound based on evaluation of two chemical types of coagent

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 lex and weather resistance, the reduced air low 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 lex life. Critical properties are therefore heat resistance over the range 100–125 °C, static and dynamic ozone resistance, and dynamic lex life. The better retention of stiffness at high temperatures offered by HNBR compounds will ensure meshing of the belt at high temperatures. A basic recipe for this application is shown in Table 7.7, including compounding variables for experimental design and optimisation. Responses (properties) to be measured include: • Heat resistance, 100–125 ºC. • Static and dynamic ozone resistance.

97

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 eficient vulcanisation

Vary levels of zinc methacrylate modiier to balance properties; total polymer phr = 100

HNBR modiied with zinc methacrylate

30-10

Added tear and tensile strength

Vary levels of zinc methacrylate modiier 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 lex 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

98

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 luid 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 lexing of the bellows, demanding lex fatigue resistance, but the lexing 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 lexibility for cold starting and dynamic lex 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 inal compound optimisation. Organic pigments are not suitable for peroxide-cured materials since the pigments can act as free-radical scavengers and may interfere with the vulcanisation reaction. In addition, the oxidising effect of peroxides may destroy the colour. Responses (properties) to be measured: • Heat resistance, 1008 h at 125 ºC. • Heat resistance, 504 h at 150 ºC. • Ozone resistance. • Coolant resistance, 1008 h at 125 ºC. • Low temperature lexibility. • Dynamic lex fatigue resistance. • Moulding performance, low and cure rates.

99

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. lexibility 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 lex; good heat stability

Nulok 321

50.0

Extending clay iller

HiSil 532EP (VARIABLE)

0–20

Reinforcing iller

Level 1 will be this grade

HiSil 233 (VARIABLE)

0–20

Reinforcing iller

Level 2 will be this grade

Vinyl-tris(ßmethoxyethoxy) silane (Silane A–172)

1.0

Coupling agent for iller dispersion

Red iron oxide

4.0

Stable pigment

Varox DCP–40C (VARIABLE)

4–8

Rapid vulcanisation on injection moulding at 170 ºC

Optimisation: 2-factor and 3-level full factorial design; 6 compounds and experiments

100

Three levels to optimise properties

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 low and cleanliness, bonding to the steel case, and shaft seal life in the simulated application extremes test programme. The starting recipe and variables for experimental design and optimisation are shown in Table 7.9. Responses (properties) to be measured are: • Resistance to gear lubricant at 125 ºC for 1008 h. • Moulding evaluations: low, cleanliness and bonding to the steel case. • Shaft seal life testing at extremes of operating environment.

101

Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology

Table 7.9 Differential shaft seal compound experimental design Ingredient

phr

Logic/comments

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 illers show best rotating shaft seal performance

N550 Carbon black

10.0

Black pigment

Varox VC40KE

5-10

Eficient vulcanisation

Experimental design optimisation

Evaluate two different illers for best overall properties

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 luid. 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 luid on the inside. The hose will be plied around the

102

Examples of Hydrogenated Nitrile Rubber Formulae for Specific Applications braided fabric reinforcement and the plies will be formed by extrusion. Due to the location of an exhaust header near the short section of hose the operating temperature of the hose will be normally at 125 °C, with an occasional excursions to 150 °C, for example when idling in heavy trafic. 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, lex fatigue resistance, smooth processing by extrusion, adhesion of the plies and the ability to withstand the customer’s pressure pulse test on the inished 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

HNBR polymer with 34–36% ACN; 1% and 5% unsaturation

100.0

Zinc oxide

2.0

Stearic acid

1.0

N550 Carbon black

60.0

Plasthall TOTM

10.0

Viscosity reduction without sacriice 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 eficient curing

Best overall balance of luid resistance

Experimental design optimisation Low level at 1% unsaturation, high level at 5% unsaturation

Variable level to optimise properties

Optimisation: 2-factor, 2- and 3-level full factorial design; 12 compounds and experiments

103

Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology Responses (properties) to be measured: • Static ozone resistance. • Dynamic ozone resistance. • Flex-fatigue resistance. • Extrusion processing. • High-temperature pressure pulse test, simulating application conditions.

7.10 Compound Example: Chemically Resistant, Low Hysteresis Roller for Paper Mills The customer for this application is a paper mill with a requirement for long-running gloss calender rolls operating at temperatures slightly above 120 ºC. The chemicals employed dictate that the roll compound must have good resistance to water and mineral-based luids. The current polyurethane rolls are not suficiently 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-modiied 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: • Dynamic hysteresis. • Heat resistance, 1008 h at 120 ºC.

104

Examples of Hydrogenated Nitrile Rubber Formulae for Specific Applications

Table 7.11 Low hysteresis paper mill roller compound experimental design Ingredient

phr

Logic/comments

Experimental design optimisation

HNBR polymer with 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 modiied with zinc methacrylate

50.0

Adds toughness, abrasion resistance and good hardness

Silane-treated calcined clay

70.0

Mineral iller

Level 1 iller evaluation

Mistron Cyprubond

70.0

Mineral iller

Level 2 iller evaluation

Varox VC40KE

7.0

Vanax MBM

4.5–7.5

Eficient 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 speciic examples. It has also introduced the reader to statistical experimental design and desirability functions for optimising HNBR formulae to give the best overall balance of properties. Using these techniques it is possible to formulate compounds to give the best overall capability for meeting customer requirements.

105

Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology

References 1.

L.B. Barrentine, in An Introduction to Design of Experiments: A Simplified Approach, American Society for Quality, Milwaukee, WI, USA, 1999.

2.

G.E.P. Box, J.S. Hunter and W.G. Hunter, Statistics for experimenters: Design, discovery and innovation, John Wiley and Sons, Hoboken, NJ, USA, 2005.

3.

E.C. Harrington, Industrial Quality Control, 1965, 21, 494.

4.

G. Derringer and R. Suich, Journal of Quality Technology, 1980, 12, 214.

106

8

Solving Hydrogenated Nitrile Rubber Processing Issues

8.1 Introduction In this inal 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 requaliication, 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 low and possible scorch (premature curing), causing unacceptable knit lines and non-ills in the inished 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 irst 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-ills. However, if we bear in mind the nature of elastomeric low we

107

Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology can see that this is unlikely to happen. From the discussion in Chapter 3, the typical effect of temperature on viscosity is shown in Figure 8.1.

Temperature effect

Viscosity (poise)

1000000 100000 100 ºC 10000 125 ºC 1000

150 ºC

100 1

10

100

1000

Shear rate (s–1)

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-ifth 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 ills 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 low and non-ills.

8.2.2 Possible Factor: Excessive Shear During Injection After feeding and preheating the rubber prior to injection, the next step is its injection into the mould. A possible factor at this point is the shear rate during injection. In production processes time is always a pressing issue and the tendency is always to inject the rubber as rapidly as possible. By elevating the injection pressure the rubber

108

Solving Hydrogenated Nitrile Rubber Processing Issues

% change in heat build-up

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 low and non-ill issues observed. A reduction in injection pressure will certainly reduce the low rate of the rubber and increase injection time, but this may be a small price to pay in preventing the non-ill and low issues caused by excessive heat build-up during injection.

350% 300% 250% 200% 150% 100% 50% 0% –50% –100% –100%

–50%

0%

50%

100%

150%

% change in parameter Viscosity effect

Shear rate effect

Figure 8.2 Typical heat build-up during injection

As discussed in Chapter 3, Equation (8.1) can be used to establish correlation and predict the effect of viscosity and shear rate increase:

∆T = 0∫t (η/ρc)γ2 dt,

(8.1)

where ∆T = temperature change in ºC s–1, η = viscosity in Pa. s and γ = shear rate in s–1.

109

Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology Using Equation (8.1) we can establish the following proportionalities to estimate the effect of viscosity and shear rate on heat build-up, as shown in Figure 8.2:

∆T ∝ η, and ∆T ∝ γ2

(8.2)

Reducing the viscosity, η, 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 low and non-ill issues:

Shear stress = (R . DP) / (2L),

(8.3)

where DP = pressure drop, R = radius of capillary and L = length of capillary.

Shear rate = (4Q) / (πR3), where Q = volumetric low rate, R = radius of capillary and ηapp = apparent viscosity = (shear stress) / (shear rate) = (DP . R4 . π) / (4Q . 2L)

110

(8.4)

Solving Hydrogenated Nitrile Rubber Processing Issues Volumetric low rate Q = (DP . R4 . π) / ( 2L . 4ηapp)

(8.5)

8.2.3 Possible Factor: Sprue and Runner Design

Net % change in flow rate

This issue may be a factor, particularly if the sporadic low and non-ill 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 (ηapp) 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 lowing through, since the low volume, Q, is proportional to the fourth power of the capillary radius. These effects are illustrated in Figure 8.3.

120% 100% 80% 60% 40% 20% 0% –20% –40% –60% –20%

–10%

0%

10%

20%

30%

% Change in factor Radius of runner

Length of runner

Viscosity of material

Figure 8.3 Effect of moulding parameters on low rate

To improve the overall low and to eradicate the sporadic low and non-ill 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 low and non-ill problem.

111

Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology

8.2.4 Solving the Injection Flow Problem using the Above Factors From Chapter 7 the reader may already have concluded that a statistical experimental design would be appropriate using the factors discussed above. Listed in Table 8.1 and illustrated in Figure 8.4 is an experimental programme designed to resolve the low and non-ill issue.

Table 8.1 Experimental design for sporadic flow and non-fill injection moulding issues Responses (dimensions) parts/100 with non-fills

Comments

Mould temperature (°C)

Injection time (s)

Injection pressure holding time (s)

180

20

5

–1, +1, –1

180

20

15

–1, +1, +1

180

10

5

–1, –1, –1

180

10

15

–1, –1, +1

190

20

5

+1, +1, –1

190

20

15

+1, +1, +1

190

10

5

+1, –1, –1

190

10

15

+1, –1, +1

185

15

10

Centre point

Again, the order would need to be randomised and a number of runs are required to determine the signiicance 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-ills. Obviously, we would require one of these combinations to consistently produce zero non-ills in successive multiple runs.

112

Solving Hydrogenated Nitrile Rubber Processing Issues

Runner size, mm

6.0

150 5.0

100 100

125

Injection feed temperature, °C

Figure 8.4 Injection moulding low: 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 irst step is to examine the mould or platen temperature, since this is by far the most critical factor in determining the dimensions of the inished 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 coeficient 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

113

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 inished dimension from 76.00 mm to 75.84 mm. Similarly, on a 127.00 mm target dimension the inished part would be reduced in size from 127.00 mm to 126.62 mm. This effect is ampliied in the case of lightly illed elastomers with a high coeficient of linear expansion, such as a HNBR elastomer of lower target hardness. A carbon black-illed HNBR compound of this type might have a coeficient 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 low conditions and any shifts in moulding. As mentioned in Chapter 3 and conirmed 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 low. 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 low, 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 inished part. Perhaps the most surprising inluence 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 speciied 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 inished part. This has been frequently observed but has never been thoroughly studied or quantiied. 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 inished parts. If the relationship between injection holding time and inal dimensions is established during the initial production approval process, it is then more likely to be monitored and controlled during routine production.

114

% Die swell

Solving Hydrogenated Nitrile Rubber Processing Issues 45 40 35 30 25 20 15 10 5 0 1

10

100

1000

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 Responses (dimensions)

Comments

Mould temperature (°C)

Injection time (s)

Injection pressure hold time (s)

180

20

5

–1, +1, –1

180

20

15

–1, +1, +1

180

10

5

–1, –1, –1

180

10

15

–1, –1, +1

190

20

5

+1, +1, –1

190

20

15

+1, +1, +1

190

10

5

+1, –1, –1

190

10

15

+1, –1, +1

185

15

10

Centre point

115

Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology

Injection time, s

Press hold time, s

15

20

5

10 180

190

Mould temperature, °C Figure 8.6 Injection dimensional shift: Graphical depiction of experimental design

Again, the order of the experiments would need to be randomised and multiple runs required to determine the relative signiicance 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 identiied for further investigation:

116

Solving Hydrogenated Nitrile Rubber Processing Issues • Slip agent used in the compound: –

Low level: parafin wax at 1.00 phr.

–

High level: fatty amide at 1.00 phr.

• Extruder barrel temperature: –

Low level: 80 °C.

–

High level: 100 °C.

• Extruder breaker plate hole size: –

Low level: 7.00 mm hole size.

–

High level: 8.00 mm hole size.

• Taper angle at entrance to the tube die: –

Low level: 30º angle.

–

High level: 45º angle.

• Extruder speed: –

Low level: 30 rpm.

–

High level: 40 rpm.

• Extruder die temperature: –

Low level: 100 °C.

–

High level: 120 °C.

A full factorial design would have entailed 64 experimental runs, which was clearly prohibitive. Initially we needed to determine which of the variables were the most signiicant, and interactions between the variables would be ignored. If we discovered two or three signiicant 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:

117

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

Signiicant variables (factors): A = Slip agent in compound: –1 = parafin wax at 1.00; +1 = fatty amide at 1.00 B = Extruder barrel temperature: –1 = 80 ºC; +1 = 100 ºC C = Extruder breaker plate hole size: –1 = 7.00 mm; +1 = 8.00 mm D = Taper angle at entrance to die: –1 = 30º; +1 = 45º E = Extruder speed: –1 = 30 rpm; +1 = 40 rpm F = 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 signiicant, 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 inished O-rings. Rather than attempting to study one variable at a time, key personnel from technical and production areas were brought together for a MECE discussion of possible variables. This discussion was conducted leaving every possible factor open for discussion, eliminating any where the data supported

118

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 iller levels at the extremes of weighing tolerance. • Compound variable: vulcanising agent (peroxide) at the extremes of weighing tolerance. • Process variable: raw rubber weights at the extremes of tolerance. • Process variable: variations in mould surface temperature. The platen temperature was set at 182 °C but the actual temperature at the mould surface was found to vary between 177 °C at the corners and 188 °C in the centre. • Process variable: mould closure. Due to the platen and mould being marginally out of parallel and latness, the mould could remain open by as much as 0.05 mm. • Process variable: oven post-cure temperature. The product underwent a 2 h postcure 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 signiicant factors ive 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.

119

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 1 2 3 4 5 6 7 8 Sum + Sum – Average E+ Average E– Effect Significant? RESULTS Run y1 1 2 3 4 5 6 7 8

96.2 96.8 92.1 92.6 96.2 100.2 96.3 101.0

A 96.3 –96.6 –92.3 92.7 –96.6 99.9 96.3 –100.7 385.2 –386.3 96.3 –96.6 –0.3 no

B 96.3 96.6 –92.3 –92.7 96.6 –99.9 96.3 –100.7 385.9 –385.6 96.5 –96.4 0.1 no

C 96.3 96.6 92.3 –92.7 –96.6 99.9 –96.3 –100.7 385.2 –386.3 96.3 –96.6 –0.3 no

D –96.3 96.6 92.3 92.7 –96.6 –99.9 96.3 –100.7 377.9 –393.6 94.5 –98.4 –3.9 YES

E 96.3 –96.6 92.3 92.7 96.6 –99.9 –96.3 –100.7 377.9 –393.6 94.5 –98.4 –3.9 YES

y2

y3

y4

y5

Y (avg.) Si2

Replicates (n)

96.3 96.5 92.0 93.0 96.8 100.1 96.0 100.8

96.5 96.2 91.9 93.1 97.1 99.5 95.8 100.2

96.1 96.8 92.8 92.1 96.5 99.6 96.8 101.1

96.5 96.9 92.8 92.5 96.5 100.3 96.5 100.5

96.3 96.6 92.3 92.7 96.6 99.9 96.3 100.7

5 5 5 5 5 5 5 5

Si2 = sum of squares = (yi–y(avg.))2 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 experiments A = carbon black iller level at maximum of weigh-up tolerance B = vulcanising agent levels at maximum of weigh-up tolerance C = preform weight levels at extremes of tolerances D = mould surface temperature: – = 175 ºC, + = 190 ºC E = mould closure: – = 0.00 mm, + = 0.05 mm F = Post-cure temperature: – = 190 ºC, + = 215 ºC MSFE: 0.387 t × S(eff) t (0.950, df) 2.04 (n–1) × (number of runs) = degrees of freedom

120

F –96.3 96.6 –92.3 92.7 96.6 99.9 –96.3 –100.7 385.9 –385.6 96.5 –96.4 0.1 no

0.0632 0.1968 0.3272 0.6008 0.4392 0.2728 0.5312 0.4512 Sum Si2

G (dummy) –96.3 –96.6 92.3 –92.7 96.6 99.9 96.3 –100.7 385.2 –386.3 96.3 –96.6 –0.3 no

2.882

Solving Hydrogenated Nitrile Rubber Processing Issues 4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 0.5 re tu

ve iv

pe

e

ra

le

gh ei w

-c

ur

e

C

te

ur

m

at

m or ef Pr

n bo ar

l

t

l ve le k ac bl

ld ou M

Po

st

C

M

ou

ld

te

m

pe

cl

ra

os

tu

ur

re

e

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 signiicant factor effect (MSFE) based on a 95% conidence limit. The only two signiicant 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 signiicant inluence 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 signiicant factors might have been missed. Following capital investment to reduce variations in mould surface temperature and mould closure, variation in ID of the moulded O-rings was virtually eliminated.

121

Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology

8.5 Conclusions Techniques have been demonstrated which will help in resolving production issues. In cases where multiple variables are involved, emphasis has been placed on statistical experimental design techniques in order to determine which are signiicant. This is intended to guard against the knee-jerk response that a change in formulation is the way to solve a production processing issue, and also attempting to examine the variables one at a time and missing the possibility of interaction between them. Statistical experimental design is a technique that can help to reduce variation in manufacture. Unfortunately many process and product improvement projects do not in practice employ these techniques properly, or fail to plan the experimental design effectively. The involvement of a wide range of personnel and disciplines is critical in planning stage and developing proper experimental design.

122

A

bbreviations

ACM

Acrylic elastomer (per D1418)

ACN

Acrylonitrile

AEM

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

123

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

124

Abbreviations ∆P

Pressure drop

Ø

Volume fraction in a mixture

Ø2

Volume fraction of polymer

η

Viscosity

τ

Shear stress

125

Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology

126

I

ndex

2-Mercaptobenzothiazole 64 100% Modulus 64, 66, 69, 72, 75 200% Modulus 69 α,α-bis(t-butylperoxy)diisopropylbenzene 65-66, 77, 79-81

A ACM 17-20, 46, 49-51, 53, 56, 58, 85 ACN 1, 3, 5-11, 13-14, 16, 18-19, 43, 45, 47-53, 63, 80-81, 94, 96, 98, 102-103, 105 Acrylonitrile 1, 3, 9, 11, 33, 79, 100 Activator 80-81, 87, 94, 96, 98, 100, 102 AEM 17-20, 46, 49-50, 53, 56, 58, 85 Aerosil 300 74 Air conditioning 57, 60 ASTM D2000 17, 43-46, 55-56, 85 Automatic transmission 17-18, 49-50, 58

B Butadiene 1-12, 14, 16, 18, 20, 22, 24, 26, 28-30, 32, 34, 36, 38, 40, 42, 44, 46, 48, 50, 52, 54, 56, 58, 60, 62, 64, 66, 68, 70, 72, 74, 76, 78, 80, 82, 84, 86, 88, 90, 92, 94, 96, 98, 100, 102, 104, 106, 108, 110, 112, 114, 116, 118, 120, 122

C Cab-O-Sil M-7D 74 Cab-O-Sil TS-720 74 Capillary 37-38 length 110 radius 111 reheometer 37 Carboxylated 16 Celite 350 74, 102 Coagent 63, 65-68, 78-81, 94, 96

127

Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology Compression modulus 33-34 Compression moulding 24, 27, 93-96, 118, 121 Compression set 34, 47, 64, 66, 68-70, 79-81, 86, 90, 93-94, 97 Contour plot 67-68, 80, 89 Conveyor 56-57 belt 16, 57 component 56 Copolymerisation 11

D DBP number 32-34, 72 Desirability function 83, 85, 91-92 Dibutoxyethoxyethyl adipate 69 Dicumyl peroxide 47-48, 69, 87 Die swell 24-25, 115 Dimensional shift 113, 116 Dioctyl phthalate 76 Dynamic ozone resistance 78, 97, 103-104

E Einstein equation 32 Elasticity 24, 29, 32, 34, 52, 95 Emulsion polymerisation 1-3, 5-9 EPDM 10, 17, 29, 31, 46, 56, 79, 85, 87-93 Ether ester 76 Ethylene 9-12, 17, 19, 29, 46, 56, 79, 87, 93 Extrusion 34, 95, 103-104, 116, 118

F Fatty amide 117-118 Filler 20, 32, 34-35, 63, 73-75, 78, 80-81, 100, 105, 119-120 FKM 17, 19-20, 36, 46, 51-53, 56, 85, 93 Flexible boots 57 Full factorial design 83, 94, 96, 98, 100, 102-103, 105, 117 FVMQ 17, 19-20, 46, 53, 56, 85

G Glass transition temperature 10, 26

128

Index

H Heat build-up 27-28, 104, 109-110, 114 HiSil 233 74-75, 100 HiSil 532EP 74, 100 HNBR 1, 7-14, 16-21, 43, 45-60, 63, 69-70, 73, 75-76, 78-82, 85, 87-105, 107, 109, 113-114, 116 Hydrogenation 1, 7-9, 11, 45-46, 48, 55, 57, 63, 79 Hysteresis 34-35, 104-105

I Imsil A-8 74 Initiator 1-3 Injection moulding 13, 24-25, 27, 34, 36, 79, 87, 99-100, 107-108, 112-113, 115 Isoprene 11-12

K Knit lines 36, 107

L Lanxess 13-16

M Manual transmission 59 Mayo-Lewis copolymer equation 3 Minimum signiicant factor effect (MSFE) 121 Mistron Cyprubond 105 Mn 30 Modulus 26, 32-34, 64-67, 69, 71-77, 95, 103-104 Mole % ACN 49 Mole fraction of monomer 5 Molecular weight 2, 6, 29-31 Molecular weight distribution 29-31 Mooney viscosity 13, 31, 72, 76 Mould low 36-37, 87 Mz 30

N N110 32 N330 65, 96 N550 32, 64, 77, 79-80, 94, 96, 102-103

129

Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology N762 46-48, 87, 94, 98 N990 32, 81 NBR 1-3, 5-11, 17, 19-20, 29, 32-33, 43-49, 51-53, 55-59, 63, 73, 75, 78-81, 85, 93, 95, 99, 101 Newtonian luid 21 Non-black illers 73-74, 82 Non-ills 107-108, 112 NR 17, 46, 56, 85 Nulok 321 74, 100 Nulok 390 74-75

O O-rings 59-61, 78, 93, 114, 118, 121 Oil cooler O-ring 93-94 Oil well packer 96 Oxidation 17, 45, 49, 55 -reduction 1 Ozone resistance 78-79, 84, 97-100, 103-104

P Packer elements 60 Paper mill roller 105 Parafin wax 87, 117-118 Peroxide 9, 43, 47-48, 52-53, 63, 65-70, 78-82, 87-92, 94, 96, 99, 102, 119 Petroleum exploration 51-52, 60 phr 32-36, 46-48, 65-70, 72, 74-81, 87-92, 94, 96, 98, 100, 102-103, 105, 117 Plackett-Burman screening design 117-120 Plasticiser 35-36, 63, 75-79, 81, 87 Polymerisation 1-9, 11, 31 Post-cure 69-70, 119-121 Pressure drop 37-38, 110 Process aids 36-37 Propylene 9-12, 29, 46, 56, 79 Pseudoplastic 22-23, 108 Pyrax B 74

R Radial shaft seals 59-60 Reactivity ratios 4-9 Recipe 33, 46-48, 64-66, 69-70, 73, 75-77, 80-81, 87, 93, 97, 99, 101, 103

130

Index Residual unsaturation 9-10, 13, 63, 81, 98, 100, 102 Response surface 34, 66, 70, 83, 93, 95 Rollers 57, 104 Roy-Cal L 74 Runners 37-38, 111

S Serpentine belt 97-98 Shaft seals 59-60, 101 Shear 13, 21-25, 27-28, 31-33, 35-38, 73, 108-110, 114-115 Shear rate 21-25, 27-28, 31-33, 35-36, 38, 108-110, 114-115 Shear stress 21, 24-25, 37-38, 110 Silene 732D 74 Sprues 37-38, 111 Static ozone resistance 104 Static seals 58-61 Statistical experimental design 19, 82-83, 85, 92-93, 99, 103, 105, 107, 112, 115, 122 Stearic acid 33, 47-48, 64-65, 69-70, 75, 80-81, 87, 94, 96, 98, 100, 102-103 Steering system hose 102-103 Sulfur 9, 13, 17, 49, 52, 58, 63-64, 69-70, 101

T Tensile strength 47, 64-66, 69 Tetramethyl thiuram disulide 69 Tg 10, 24, 26 Therban® 15-16 Time-temperature superposition 26 Transfer moulding 24, 101 Translink 77 74 Trioctyl trimellitate 78, 81, 87

U Ultimate elongation 47, 64-66, 68-69 Ultra-Plex 74 Ultrasil VN-3 SP 74

V Viscosity 2, 13-14, 16, 21-23, 26-33, 35-36, 38-39, 47, 72-73, 76-79, 87, 93, 100, 103, 108-111, 114

131

Practical Guide to Hydrogenated Nitrile Butadiene Rubber Technology VMQ 17, 46, 56, 85 Volume fraction 32, 35, 73 Volumetric flow 37-38, 110-111

W Williams-Landel-Ferry relationship 26

Z Zeeospheres 200 74, 81 Zeolex 23 74 Zeon Chemicals 13, 15-16, 20, 51, 54, 63, 78, 82 Zetpol® 15-16 Ziegler-Natta-catalyst 29 Zinc methacrylate 16, 98, 104-105 Zinc oxide 33, 47-48, 64-65, 69-70, 75, 80-81, 87, 94, 96, 98, 100, 102-103 abcd abcd abcd abcd a b c d abcd a αβχδ a

132

Published by Smithers Rapra Technology Ltd, 2012

High performance engineering plastics are used in an increasingly wide range of applications and environments. Their growth in importance is a response to the ever-increasing demand for more reliable, high performance components. This book is the product of the 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.

Shawbury, Shrewsbury, Shropshire, SY4 4NR, UK Telephone: +44 (0)1939 250383 Fax: +44 (0)1939 251118 Web: www.polymer-books.com