Rudnick L. Lubricant Additives: Chemistry and Applications (Присадки, добавки к смазкам)

Подождите немного. Документ загружается.

284 Lubricant Additives: Chemistry and Applications

and physical properties of OCPs continue to evolve to achieve improvements in low-temperature

rheology, thickening ef ciency, and bulk handling characteristics.

Several excellent reviews of OCP viscosity modi ers have been published [1–3]. This chapter

serves as an update and current compilation of information relating to the chemistry, properties, and

performance characteristics of this important class of lubricant additives.

10.2 CLASSES OF OLEFIN COPOLYMERS

There are many ways to classify OCP viscosity modi ers. From a user’s perspective, OCPs are mar-

keted as either solids or liquid concentrates. The physical state of the solids depends on several fac-

tors, primarily on the ethylene/propylene (E/P) mass ratio. When E/P is in the 45/55–55/45 range,

the material is amorphous and cold  ows at room temperature. Thus, OCPs of this composition are

most commonly sold as bales, packaged in rigid boxes to maintain bale shape. When E/P is higher

than 60/40, the copolymer becomes semicrystalline in nature and does not cold  ow under ambient

conditions. Thus, both bales and pellets can be produced.

Liquid concentrates of OCP in mineral oil contain enough rubber to raise the kinematic

viscosity (KV) to 500–1500 cSt (mPa s) range at 100°C. A typical viscosity/concentration rela-

tionship is shown in Figure 10.1.

From the preceding discussion, OCPs can also be classi ed according to crystallinity, which is

measured by x-ray diffraction or differential scanning calorimetry. The in uence of crystallinity on

rheological performance will be discussed in Section 10.5.

Shear stability is another parameter by which OCP viscosity modi ers are categorized. The higher

the molecular weight of a polymer, the more prone it is to mechanical degradation when elongational

forces are imposed by the  uid  ow  eld. This subject is dealt with in detail in Section 10.5.2.2.3.

Finally, chemical functional groups can be grafted to the OCP backbone, providing added

dispersancy, antioxidant activity, and low-temperature viscosity enhancement. A number of

chemical routes for functionalizing OCPs are described in Section 10.3.3.

10.3 CHEMISTRY

10.3.1 S

YNTHESIS BY ZIEGLER–NATTA POLYMERIZATION

Although methods for synthesizing high-molecular-weight polymers of ethylene were commercial-

ized in the 1930s (the Imperial Chemical Industries (ICI) PLC, currently a division of AkzoNobel

high-pressure process), the polymers contained a signi cant number of short- and long-chain

FIGURE 10.1 Kinematic viscosity of 50 permanent shear stability index (PSSI) amorphous OCP dissolved in

100N mineral oil. (Minick, J., A. Moet, A. Hiltner, E. Baer and S.P. Chum, J. Appl. Poly. Sci., 58, 1371–1384,

1995. Reprinted with permission of John Wiley & Sons, Inc.)

1000

2000

3000

4000

0510

OCP concentration, mass %

Viscosity (cSt) at 100°C

CRC_59645_Ch010.indd 284CRC_59645_Ch010.indd 284 12/6/2008 10:10:10 AM12/6/2008 10:10:10 AM

Oleﬁ n Copolymer Viscosity Modiﬁ ers 285

branches that limited the ability to produce high-density polyethylene. The  rst commercially via-

ble synthesis of linear polyethylene at low monomer pressure was pioneered by Ziegler in 1953, and

the stereoregular polymerization of α-ole ns was demonstrated by Natta the following year [4].

The secret to their success was the discovery of catalysts (called Ziegler or Zieger–Natta catalysts),

which are molecular complexes between halides and other derivatives of group IV–VIII transi-

tion metals (Ti, V, Co, Zr, and Hf) and alkyls of group I–III base metals. A typical catalyst of this

type comprises an aluminum alkyl and a titanium or vanadium halide having the general structure

shown in Figure 10.2. Electron donors, such as organic amines, esters, phosphines, and ketones,

may be used to enhance reaction kinetics. Finally, molecular weight control is often aided by the

use of chain transfer agents such as molecular hydrogen or zinc alkyls [5], which are effective in

terminating chain growth

without poisoning the active metal center.

Ziegler–Natta polymerization is probably the best-known example of insertion, coordination,

stereoregular, or stereospeci c polymerization. This nomenclature has been adopted to describe the

mechanism(s) by which ole n monomers insert into the growing polymer chain, as directed by both

steric and electronic features of the coordination catalyst. A commonly accepted chain propagation

mechanism involves monomer insertion at the transition metal–carbon bond [4]. The main purpose

of the base metal alkyl is to alkylate the transition metal salt, thus stabilizing it against decomposi-

tion. As pointed out by Boor [4], Ziegler–Natta catalysts may be modi ed to produce copolymers

with varying degrees of randomness or, from a different perspective, blockiness of one or both

comonomers. Owing to the higher reactivity ratio of ethylene to that of propylene [4–7], the forma-

tion of long runs of ethylene is more favored than long sequences of propylene. This is substantiated

C NMR spectroscopy [8–13].

Ziegler–Natta catalysts are available in two forms—heterogeneous and homogeneous. Heteroge-

neous catalysts are insoluble in the reaction medium and are suspended in a  uidized-bed con gura-

tion. Reaction takes place at the exposed faces of the metal complex surface. Since each crystal plane

has a slightly different atomic arrangement, each will produce slightly different polymer chains in

terms of statistical monomer insertion and molecular weight distribution. Thus, they are often called

multisite catalysts. Homogeneous Ziegler–Natta catalysts are soluble in the reaction solvent and, there-

fore, function more ef ciently since all molecules serve as potential reaction sites. Since the catalyst

is not restrained in a crystalline matrix, it tends to be more “single-site” in nature than heterogeneous

catalysts. Polymers made by homogeneous polymerization generally are more uniform in microstruc-

ture and molecular weight distribution and, therefore, are favored for use as viscosity modi ers [1,4].

Nonconjugated dienes are often used in the manufacture of ethylene–propylene copolymers

(known as EPDMs-Ethylene Propylene Diene Monomer) to provide a site for cross-linking (in non-

lubricant applications) or to reduce the tackiness of the rubber for ease of manufacture and han-

dling. Certain dienes promote long-chain branching [2,5,14,15] that, in turn, increases the modulus

in the rubber plateau region. The terpolymer is then easier to handle as it is dried and packaged [1].

A disadvantage of long-chain branching is that it reduces the lubricating oil thickening ef ciency

relative to a simple copolymer of similar molecular weight and copolymer composition, although

low levels of vinyl norbornene or norbornadiene are claimed [15] to improve cold  ow without loss

in thickening ef ciency or shear stability.

FIGURE 10.2 Active center in Zieger–Natta catalysts. M

= transition metal (such as Ti); M

= base metal

(such as Al). (Adapted from Boor, J., Jr., Ziegler-Natta Catalysts and Polymerizations, Academic Press,

New York, 1979.)

Cl - M

At crystal surface

Cl - M

RÕ

Base metal complex

(

soluble s

ecie

)

CRC_59645_Ch010.indd 285CRC_59645_Ch010.indd 285 12/6/2008 10:10:10 AM12/6/2008 10:10:10 AM

286 Lubricant Additives: Chemistry and Applications

10.3.2 SYNTHESIS BY METALLOCENE POLYMERIZATION

The desire to achieve higher levels of control over stereoregularity, composition, and molecular

weight distribution led to the development of activated metallocene catalysts. Although known to

Zieger and Natta, the technology was rediscovered by Kaminsky and Sinn in 1980 and further

developed by workers such as Brintzinger, Chien, Jordan, and others [16–20]. Metallocene catalysts

consist of compounds of transition metals (usually group IVB: Ti, Zr, and Hf) with one or two

cyclopentadienyl rings attached to the metal. The most common activator is methylaluminoxane

(MAO). A large number of variants have been reported, but the highest levels of stereospeci city

have been achieved with bridged, substituted bis-cyclopentadienyl metallocenes (Figure 10.3) [21].

One of the major advantages of metallocenes over Ziegler–Natta catalysts is the ability to incorpo-

rate higher α-ole ns and other monomers into the ethylene chain.

The  rst commercial use of metallocene single-site catalysts to manufacture EPDM

elastomers was DuPont Dow Elastomers’ Plaquemine, LA facility, which began operation in

1996 using Dow’s Insite

constrained geometry catalyst [22,23]. The catalyst is described as

“mono cyclopentadienyl Group 4 complex with a covalently attached donor ligand … requiring

activation by strong Lewis acid systems [such as] MAO…” Several advantages of this technology

over conventional Ziegler–Natta processes were reported. Since the catalyst is highly ef cient,

less is needed; therefore, the process does not require a metal removal or de- ashing step. In

addition, the copolymers produced by this chemistry are reported to have narrow molecular

weight distributions for good thickening ef ciency and shear stability as well as good control

over copolymer microstructure. Metallocene-catalyzed polyole ns also differ from Ziegler–

Natta polymers in that the former contains a predominance of unsaturated ethylidene end

groups [24].

10.3.3 FUNCTIONALIZATION CHEMISTRY

Traditionally, OCPs are added to lubricating oil to reduce the degree to which viscosity

decreases with temperature, that is, to function solely as a rheology control agent. Other lubricant

additives—such as ashless succinimide dispersants, a variety of antioxidants, detergents,

antiwear agents, foam inhibitors, friction modi ers, and anticorrosion chemicals— provide other

important functions (dispersing contaminants, keeping engines clean, maintaining piston ring

performance, preventing wear, etc.). It has been recognized for many years that it is possible

to combine some of these performance and rheology control features on the same molecule.

Some report that “both dispersant and antioxidant functionality may exhibit more potent

activity when attached to the polymer backbone than in their monomeric form” [25]. Three hybrids

have been commercialized, although many more have been disclosed in the patent literature.

FIGURE 10.3 Chemical structure of generalized bridged bis-cyclopentadienyl metallocene catalyst. M is

a group IVB transition metal; X is a halogen or alkyl radical. (Rubin, I.D. and A. Sen, J. Appl. Poly. Sci.,

40, 523–530, 1990. Reprinted with permission of John Wiley & Sons, Inc.)

CRC_59645_Ch010.indd 286CRC_59645_Ch010.indd 286 12/6/2008 10:10:11 AM12/6/2008 10:10:11 AM

Oleﬁ n Copolymer Viscosity Modiﬁ ers 287

They include dispersant OCPs (DOCPs), dispersant antioxidant OCPs (DAOCPs), and grafted

OCPs (gOCPs). The addition of antiwear functionality has also been reported [25,26].

Although many grafting reactions have been described in the literature, two general classes

have received the most attention. Free radical grafting of nitrogen-containing monomers or

alkylmethacrylates onto the OCP molecule is one class. Nitrogen-containing monomers such as

vinyl pyridines, vinylpyrrolidinones, and vinylimidazoles are often cited in the patent literature

[28–33]. Free radical grafting with phenothiazine is claimed [28–33] to provide antioxidant func-

tionality as well. The grafting reaction may be conducted with the OCP molecule dissolved in

mineral oil or another suitable solvent [77,86,130,131]; alternately, solvent-free processes have been

disclosed [28–33] in which the reaction is conducted in an extruder.

Mixtures of alkylmethacrylate monomers, which are typical of those found in poly(alkyl

methacrylate) (PMA) viscosity modi ers, may be grafted to OCPs [34] to provide improved low-

temperature properties. Adding nitrogen-containing monomers to the alkylmethacrylate mixture

provides dispersancy characteristics as well. A common side reaction is homopolymerization, which

can be minimized by process optimization. Homopolymers of nitrogen-containing monomers are

usually not very soluble in mineral oils and often lead to hazy products and can attack  uoro elastomer

seals. Homopolymers of alkylmethacrylates are fully soluble in oil, however. Thus, optimizing the

grafting process is much more critical when working with nitrogen-containing monomers.

A second class of grafting reactions involves two steps [25,35–47]. In the  rst step, maleic

anhydride or a similar diacyl compound is grafted onto the OCP chain, assisted by free radical ini-

tiators, oxygen, and heat [54,77,86,130,131]. In the second step, amines and alcohols are contacted

with the anhydride intermediate to create imide, amide, or ester bonds. In many respects, this chem-

istry is very similar to that used to create ashless succinimide dispersants. An advantage of this

approach over free radical monomer grafting is that homopolymerization is avoided. The patent lit-

erature describes a related functionalization process in which free primary or secondary nitrogens

of highly basic succinimide dispersants may be used to couple preformed dispersants to the maleic

anhydride–grafted OCP molecule [49–52]. Amine derivatives of thiadiazole, phenolic [25], and

amino- aromatic polyamine [53,54] compounds have been reacted with maleic anhydride–grafted

OCP to provide enhanced antioxidant character to the additive [25].

Although maleic anhydride is the most common chemical “hook” for attaching functional

groups to OCP polymers, a number of other approaches have been reported [55–61]. Further elabo-

ration is beyond the scope of this chapter.

Another approach for attaching functionality to the OCP chain is through the non conjugated

diene in the terpolymer [26,62]. For example, 2-mercapto-1,3,4-thiadiazole is attached to the

ethylidenenorbornene site on an EPDM polymer through addition of the thio group across the

ethylidene double bond. The thiadiazole group is claimed to provide antiwear, antifatigue charac-

teristics to lubricants containing the grafted OCP.

10.4 MANUFACTURING PROCESSES

Two polymerization processes have been used for the manufacture of E/P copolymer viscosity

modi ers: solution and slurry. In the solution process, the gaseous monomers are added under pres-

sure to an organic solvent such as hexane, and the polymer stays in solution as it forms. By contrast,

the slurry or suspension process utilizes a solvent such as liquid propylene in which the resultant

copolymer is not soluble. It is reported [63] that removing the catalyst residue from the polymer is

more dif cult in the slurry process, although some contend [27] that the levels of catalyst are so low

that catalyst removal is not necessary.

Ethylene–propylene rubber was reported [64] to have been successfully manufactured in a

 uidized-bed gas-phase reactor. However, the use of  uidization aids such as carbon black is necessary

to process low-molecular-weight grades that are typical of lubricating oil viscosity modi ers. Thus,

the gas-phase process is not appropriate for manufacturing OCP viscosity modi ers.

CRC_59645_Ch010.indd 287CRC_59645_Ch010.indd 287 12/6/2008 10:10:11 AM12/6/2008 10:10:11 AM

288 Lubricant Additives: Chemistry and Applications

10.4.1 SOLUTION PROCESS

The most common method for manufacturing OCP viscosity modi ers is the solution process as

described in Figure 10.4 [65]. It is made up of four sections—polymerization, polymer isolation,

distillation, and packaging. In the polymerization section, monomers, an organic solvent such as

hexane, and a soluble catalyst are introduced into a continuously stirred polymerization reactor.

During polymerization, the polymer remains in solution and causes the bulk viscosity of the reac-

tion medium to increase. To maintain good agitation, monomer diffusion, and thermal control,

polymer concentration in the polymerization reactor is typically limited to 5–6 wt% [27]. Up to

 ve reactors arranged in series have been reported in the literature [66]. The ef uent from the last

reactor is contacted with an aqueous shortstop solution to terminate polymerization and wash away

the catalyst, although this step is often omitted when using metallocene catalysts due to their high

reactivity and, therefore, low concentration [22]. While the copolymer is still in solution, extender

oils or antioxidants can be added.

In the isolation section, three techniques have been described in the literature. In the most

common method (shown in Figure 10.4), the polymer is  occulated with steam, and the solvent and

unreacted monomers are recovered, puri ed, and recycled. The aqueous polymer slurry is mechani-

cally dewatered, granulated, and air-dried. A second nonaqueous method for isolating the polymer

has been described in which the polymer is concentrated in a series of solvent removal steps [67,68].

The  nal step may be conducted in a devolatilizing extruder. A third technique does not isolate the

polymer as a solid; rather, it mixes the polymer solution into mineral oil and distills off the solvent,

producing a  nished liquid OCP product [2].

Another type of solution polymerization process that has received a great deal of attention has

been Exxon’s tubular reactor technology. Its purpose is to generate a polymer with long blocks

differing in monomer composition for improved performance as a viscosity-improving polymer

[3,69–71]. A schematic of this process is shown in Figure 10.5. Monomers and solvent are premixed

FIGURE 10.4. Solution process for manufacturing EPDM. (Adapted from Hydrocarbon Processing, 164,

November 1981.)

Propylene

Shortstop

Waste water

Extender

oil

Steam

High-boils

Fresh

solvent

Solvent

Water

Baling

wrapping

Expeller

Expander

Off

gas

Solvent

Polymerization Dewatering

Distillation

Termonomer

Catalyst

Ethylene

CRC_59645_Ch010.indd 288CRC_59645_Ch010.indd 288 12/6/2008 10:10:11 AM12/6/2008 10:10:11 AM

Oleﬁ n Copolymer Viscosity Modiﬁ ers 289

with a highly active Ziegler–Natta polymerization catalyst and metered into a plug  ow reactor

under conditions that minimize chain transfer and termination reactions. Ethylene or propylene is

injected into the tube at different points to adjust the local monomer concentration and, thereby, the

monomer composition along the growing polymer chain. In comparing the rheological properties

of different A-B-A type block compositions, Ver Strate and Struglinski reported [12] that “chains

with high ethylene section in the center of the chain … associate at low temperature with little inter-

molecular connectivity.” When the high ethylene (crystallizable) segments are at the ends, polymer

networks can form at low temperatures, which can impart a gelatinous texture to the solution.

10.4.2 SUSPENSION PROCESS

Ethylene and a nonconjugated diene, if desired, are contacted with liquid propylene, which acts

both as a monomer and a reaction medium [27,72]. In the presence of a suitable catalyst, polymer-

ization takes place rapidly, producing a suspension of copolymer granules that are insoluble in the

reaction medium. The heat liberated during the polymerization reaction is dissipated by propylene

evaporation, thus providing a convenient mechanism for temperature control. In addition, since the

polymer is not soluble in the reaction medium, viscosity remains low. Thus, relative to typical solu-

tion processes, the polymer concentration in a suspension reactor can be  ve to six times higher.

Upon exiting the polymerization reactor, the polymer suspension is contacted with steam to strip

off unreacted propylene that is then recycled. According to Corbelli and Milani [72], the Dutral

process does not include a catalyst washing step. The copolymer product, in aqueous suspension, is

dewatered, dried, and packaged in a similar fashion to polymer made by the solution process.

10.4.3 POSTPOLYMERIZATION PROCESSES

There are two main types of packaging processes in practice today [64]. In one, the isolated poly-

mer is mechanically compressed into rectangular bales. The bales are often wrapped in a polyole n

packaging  lm to prevent the bales from adhering to one another during storage and foreign matter

from sticking to the tacky rubber surface. Typical types of polyole ns  lms include poly(ethylene-

co-vinyl acetate), low-density polyethylene, and ethylene/α-Ole n Copolymers (OCP). Another

method for packaging solid OCP rubber is to extrude the polymer, pass the melt stream into a

water-cooled pelletizer, and dry the  nal product. The pellets may be packaged in bags or boxes or

may be compressed into rectangular bales.

FIGURE 10.5 Tubular reactor process for preparation of multiblock ethylene–propylene copolymers.

(U.S. Patent 4874820, 1989; U.S. Patent 4804794, 1989; U.S. Patent 5798420, 1998.)

Catalyst

Solvent

Cocatalyst

Solvent/monomers

collection

Solvent/

monomers

Solvent/

monomers

Solvent/

monomers

Tubular reactor

Temperature-

controlled

catalyst

premixing

device

Mixing

zone

Polymerization zone

CRC_59645_Ch010.indd 289CRC_59645_Ch010.indd 289 12/6/2008 10:10:12 AM12/6/2008 10:10:12 AM

290 Lubricant Additives: Chemistry and Applications

The mechanical properties of the rubber often dictate what type of isolation and packaging pro-

cesses are the most appropriate. Amorphous Ethylene propylene (EP) copolymers are often too sticky

to successfully traverse the conventional  occulation/drying/baling process. One way to modify

these compositions to improve their handling characteristics is by introducing long-chain branching

[5,48,73] through the use of low concentrations of nonconjugated dienes or other branching agents. For

nonfunctionalized OCPs, this is the main reason that some commercial viscosity modi ers contain

dienes [2]. Copolymer compositions higher in ethylene content (>60 wt% [5]) are often semi-

crystalline and may be amenable to packaging in pellet form. In some cases, the pellets may contain

an anticaking or antiblocking agent to prevent agglomeration.

Another type of manufacturing process has been used to manufacture low-molecular-weight

OCP viscosity modi ers that are dif cult to isolate and package in conventional equipment. A

higher-molecular-weight feedstock of the appropriate composition may be fed into a masticating

extruder or Banbury mixer to break down the polymer chain to lower-molecular-weight fragments

using a combination of heat and mechanical energy [74–81]. Several patents describe the use of

oxygen [82–85] or free radical initiators [86,87] to enable this process.

10.4.4 MAKING THE OCP LIQUID CONCENTRATE

After the solid OCP viscosity modi er has been manufactured, it must be dissolved in oil before it

can be ef ciently blended with base stocks and other additives. The  rst stage entails feeding the

rubber bale into a mechanical grinder [2] and then conveying the polymer crumb into a high- quality

diluent oil that is heated to 100–130°C with good agitation. The rubber slowly dissolves, raising

the viscosity of the oil as shown in Figure 10.1. Certain high-intensity homogenizers can also be

used in which the entire rubber bale is fed directly into a highly turbulent diluent oil tank at high

temperature; this bypasses the pregrinding step.

When the solid polymer is supplied in pellet form, the rubber can be fed directly into hot oil, or,

if it is slightly agglomerated, it may  rst be passed through a low-energy mechanical grinder.

10.5 PROPERTIES AND PERFORMANCE CHARACTERISTICS

10.5.1 E

FFECT OF ETHYLENE/PROPYLENE RATIO ON PHYSICAL PROPERTIES OF THE SOLID

The comonomer composition of E/P copolymers has a profound in uence on the physical properties

of the rubber. These properties, in turn, dictate the type of containers in which the product can be

stored and how it is handled during distribution and use.

C NMR has been used extensively to characterize the sequence distribution of EP copolymers

[88–93]. As the ratio of E/P increases, the fraction of ethylene–ethylene (EE) sequences (dyads)

rises, as demonstrated by the data in Figure 10.6. Concurrently, the total fraction of E/P dyads

decreases (forward and reverse propylene insertion are designated as p and p*, respectively). Thus,

the average length of contiguous ethylene increases with ethylene content. Above ∼60 wt% eth-

ylene, these sequences become long enough to crystallize, as measured by differential scanning

calorimetry (Figure 10.7) or x-ray diffraction.

When the degree of crystallinity exceeds ∼25%, EP copolymers become unsuitable as viscosity

modi ers due to limited solubility in most mineral oils. As the propylene content approaches zero,

the copolymer takes on the physical characteristics of high-density polyethylene, which, due to its

inertness to oil, is used as the packaging material of choice for engine oils and other automotive

 uids. Microstructural investigations of metallocene ethylene/α-OCPs by Minick et al. [94] con-

cluded that the relatively short ethylene sequences of low-crystallinity (<25%) samples are capable

of crystallizing into fringed micelle or short-bundled structures (Figure 10.8). Higher-order mor-

phologies such as lamellae or spherulites are not observed. Therefore, the physical properties of

semicrystalline OCPs fall in between those of polyethylene and amorphous EP rubber.

CRC_59645_Ch010.indd 290CRC_59645_Ch010.indd 290 12/6/2008 10:10:12 AM12/6/2008 10:10:12 AM

Oleﬁ n Copolymer Viscosity Modiﬁ ers 291

Polyethylene is a rigid, high-modulus solid at room temperature. Amorphous E/P rubber is

a relatively soft material under ambient conditions, which, cold,  ows and exhibits a tacky feel.

The degree of tack is inversely proportional to its molecular weight and can be reduced by the

incorporation of long-chain branching. Solid bales of this type of rubber are easily compressed

and further densify during storage. Semicrystalline OCPs hold their shape during storage but are

slightly tacky to the touch. Higher compression pressures, longer compression times, and higher

 nishing temperatures are required to successfully produce dense bales. Typical physical properties

of E/P copolymers are summarized in Table 10.1.

FIGURE 10.6 The effect of ethylene content (wt% as measured by NMR) on ethylene–ethylene (EE) and

ethylene–propylene (EP) dyad concentration.

0.2

0.3

0.4

0.5

0.6

0.7

0.8

45 55 65 75 85

Eth

lene content

(

wt%

)

Dyad mole fraction

fEp+fEp*

fEE

FIGURE 10.7 The effect of ethylene content (wt%) on crystallinity as measured by differential scanning

calorimetry (DSC) for a range of experimental EPDM copolymers.

DSC heat of fusion (J/g)

%Eth

lene

CRC_59645_Ch010.indd 291CRC_59645_Ch010.indd 291 12/6/2008 10:10:12 AM12/6/2008 10:10:12 AM

292 Lubricant Additives: Chemistry and Applications

10.5.2 EFFECT OF COPOLYMER COMPOSITION ON RHEOLOGICAL PROPERTIES IN SOLUTION

10.5.2.1 Low-Temperature Rheology

Rubin et al. [95–101] and others [102] measured the intrinsic viscosity [η] of EP copolymers

as a function of temperature in various solvents (Figure 10.9). Intrinsic viscosity, a measure of

polymer coil size in dilute solution, is fairly insensitive to temperature for noncrystalline OCPs.

Semi crystalline copolymers undergo a precipitous drop in [η] as temperature drops below ∼10°C. In

this region, the polymer begins to crystallize, forming intermolecular associations that effectively

cause the molecule to shrink in on itself, yet remain suf ciently solvated to remain suspended in

oil solution.

The viscosity of a dilute polymer solution often follows the Huggins equation [103]



kcc⫽⫹

[]

′

[]

FIGURE 10.8 Schematic illustration of solid-state morphologies of four types of poly(ethylene-co-octene)

copolymers. # SCB/1000 C is de ned as the number of side chain branches per 1000 backbone carbon atoms.

(Minick, J., Moet, A., Hiltner, A., Baer, E., and Chum, S.P., J. Appl. Poly. Sci., 58, 1371–1384, 1995. Reprinted

with permission of John Wiley & Sons, Inc.)

SCB/1000 C

Crystallinity (%)

Density (g/cc)

20 30 40 50

0.96

0.920.910.900.890.880.87

0.86

Type I

Type II

Type III

Type IV

Fringed micelles

No lamellae

Fringed micelles

Lamellae

No fringed micelles

Lamellae

No fringed micelles

Lamellae

No spherulites Spherulites Spherulites Spherulites

TABLE 10.1

Typical Physical Properties of Ethylene–Propylene Copolymers

Property Typical Value

Density (kg/m

) 860

Heat capacity (cal/g °C) 0.52

Thermal conductivity (cal/cm s °C) 8.5 × 10−4

Thermal diffusivity, (cm

/s) 9.2 × 10−4

Thermal coef cient of linear expansion (per°C) 2.2 × 10−4

Source: Adapted from Corbelli, L., Dev. Rubber Tech., 2, 87–129, 1981.

CRC_59645_Ch010.indd 292CRC_59645_Ch010.indd 292 12/6/2008 10:10:13 AM12/6/2008 10:10:13 AM

Oleﬁ n Copolymer Viscosity Modiﬁ ers 293

where

c = polymer concentration (g/dL)

= speci c viscosity (η – η

)/η

η = solution viscosity

= solvent viscosity

k′ = constant

Thus, for a speci c lubricant composition, c and η

are  xed, and the temperature dependence of the

solution viscosity is directly related to that of the intrinsic viscosity.

Low-temperature viscosity is an important rheological feature of automotive lubricants. For the

vehicle to start in cold weather, the lubricant viscosity in the bearings should be below a critical value

as determined by low-temperature engine startability experiments [7] and de ned within SAE J300

[104] for all “W” grades. The cold cranking simulator (CCS) test, ASTM D5293, is a high-shear rate

rheometer operating at  xed subambient temperatures, designed to simulate oil  ow in automotive

bearings during start-up. After the engine starts, the lubricant must also be able to freely  ow into

the oil pump and throughout the internal oil distribution channels of the engine. This is the other

half of the low-temperature viscosity speci cation for motor oils [104]. The mini-rotary viscometer

(MRV), ASTM D4684, is a low-shear rate rheometer designed to simulate pumpability characteristics

of a multigrade oil in a vehicle that was sitting idle for about 2 days in cold weather. SAE J300 also

contains upper viscosity limits for MRV viscosity and yield stress for all W grades.

Thus, the mechanism of intermolecular crystallization, which leads to molecular size contraction

in solution, affords high ethylene OCP viscosity modi ers the opportunity to contribute less to

FIGURE 10.9 Intrinsic viscosities of EP copolymers in SNO-100 base oil at various temperatures. EPC3,

EPB1, EPA2, and EPA1 have 73, 61, 50, and 50 wt% ethylene, respectively. Differences in intrinsic viscosity

above 20°C are attributable to differences in molecular weight. (Rubin, I.D. and Sen, A., J. Appl. Poly. Sci.,

40, 523–530, 1990. Reprinted with permission of John Wiley & Sons, Inc.)

2.50

2.25

2.00

1.75

1.50

1.25

1.00

0.75

0.50

0.25

0.00

−20

−10

20 30

40 50

EPA1

EPA2

EPB1

EPC3

Intrinsic viscosity (dL/g)

Temperature (°C)

CRC_59645_Ch010.indd 293CRC_59645_Ch010.indd 293 12/6/2008 10:10:13 AM12/6/2008 10:10:13 AM