William R. Herguth, Thomas M. Warne, Editors-Turbine Lubrication in The 21st Century (ASTM Special Technical Publication, 1407) (2001) PDF

STP 1407
Turbine Lubrication in the

21 st Century
William R. Herguth and Thomas M. Warne, editors
ASTM Stock Number: STP1407
ASTM
PO Box C700
100 Ban" Harbor Drive
West Conshohocken, PA 19428-2959
Printed in USA
Library of Congress Cataloging-in-Publication Data
Turbine lubrication in the 21st century /William R. Herguth and Thomas M. Warne, editors.
p. cm. -- (STP ; 1407)
"ASTM Stock Number: STP 1407,"
Proceedings of a symposium held June 26, 2000, Seattle, Wash.
Includes bibliographical references
ISBN 0-8031-2885-1
1. Turbines--Lubrication--Congresses. I. Herguth, William R., 1950- II. Warne,
Thomas M . , 1939- III. ASTM special technical publication ; 1407.
TJ266 .T872 2001

621.406--dc21 00-068940
Copyright 9 2001 AMERICAN SOCIETY FOR TESTING AND MATERIALS, West Conshohocken,
PA. All rights reserved. This material may not be reproduced or copied, in whole or in part, in any
printed, mechanical, electronic, film, or other distribution and storage media, without the written
consent of the publisher.
Photocopy Rights
Authorization to photocopy items for internal, personal, or educational classroom use, or

the internal, personal, or educational classroom use of specific clients, is granted by the
American Society for Testing and Materials (ASTM) provided that the appropriate fee is paid to
the Copyright Clearance Center, 222 Rosewood Drive, Danvers, MA 01923; Tel: 508-750-8400;
online: http:l/www.copyright.comL
Peer Review Policy
Each paper published in this volume was evaluated by two peer reviewers and at least one editor.
The authors addressed all of the reviewers' comments to the satisfaction of both the technical
editor(s) and the ASTM Committee on Publications.
To make technical informaUon available as quickly as possible, the peer-reviewed papers in this
publication were prepared "camera-ready" as submitted by the authors.
The quality of the papers in this publication reflects not only the obvious efforts of the authors and
the technical editor(s), but also the work of the peer reviewers. In keeping with long-standing
publication practices, ASTM maintains the anonymity of the peer reviewers. The ASTM Committee on
Publications acknowledges with appreciation their dedication and contribution of time and effort on
behalf of ASTM.
Printed in Baltimore,MD
January2001
Foreword
This publication, Turbine Lubrication in the 21 st Century, contains papers presented at the sym-
posium of the same name held in Seattle, Washington, on June 26, 2000. The symposium was spon-
sored by ASTM Committee D-2 on Petroleum Products and Lubricants and its Subcommittee
D02.C0 on Turbine Oils. The symposium chairman was William R. Herguth, Herguth Laboratories,
Inc., Vallejo, California. The symposium co-chairman was Thomas M. Warne, Chevron G!obal
Lubricants, Richmond, California.
Contents
Overview vii
The Use of a Fire-Resistant Turbine Lubricant: Europe Looks to the F u t u r e - - -

W. DAVID PHILLIPS
Advanced High-Temperature Air Force Turbine Engine Oil Program~

LOIS J. GSCHWENDER, LYNNE NELSON, CARL E. SNYDER, JR., GEORGE W. FULTZ, AND
COSTY S. SABA 17
The Evolution of Base Oil Technology--DAWD C. KRAMER,BRENT K. LOK,AND

RUSS R. KRUG 25
Turbine Oil Quality and Field Application Requirements--STEWN T. SWIFT,

K. DAVID BUTLER, AND WERNER DEWALD 39
Performance Advantages of Turbine Oils Formulated with Group H and Group HI

Basestocks---DOUGLAS J. IRVINE 53
Improved Response of Turbine Oils Based on Group H Hydrocracked Base Oils

Compared with Those Based on Solvent Refined Base Oils---
BRUCEP. SCHWAGER, BRYANT J. HARDY, AND GASTON A. AGUILAR 71
Performance Advantages of Turbine Oils Formulated with Group II Base Oils--

MARKE. OKAZAKIAND SUSANE. MILITANTE 79
Antioxidant Analysis for Monitoring Remaining Useful Life of Turbine Fluids---

JO AMEYE AND ROBERT E. KAUFFMAN 86
Overview
This publication summarizes the presentations delivered at the "Symposium on Turbine

Lubrication in the 21 st Century," held in Seattle, Washington on June 26, 2000. The symposium was
sponsored by ASTM Committee D-2 on Petroleum Products and Lubricants and its Subcommittee
D02.C0 on Turbine Oils.
In the final years of the 20th Century, the lubrication requirements of turbines used for power gen-
eration increased significantly. In response, two trends emerged. One was the production of more sta-
ble lubricants; the second was the development of improved techniques for monitoring the condition
and suitability for use of turbine lubricants.
For some applications, users have turned to synthetic, non-hydrocarbon fluids, such as polycar-
boxylic acid esters and phosphate esters. Two of the presentations describe current and future direc-
tions for some of these fluids. Phillips describes current and future applications of Fire-Resistant
Turbine Lubricants, with particular emphasis on European actions to improve the safety of turbine
operation. Gschwender, Snyder, Nelson, Carswell, Fultz and Saba address the special case of aircraft
turbine engine lubrication and the evolution of new military specifications for Advanced High-
Temperature Turbine Engine Oils.
Conventional mineral oil lubricants, produced by solvent extraction and dewaxing of heavy
petroleum fractions, still constitute the largest volume of turbine lubricants. However, as we enter the
21 st century, petroleum refiners have developed new processing methods; these lead to more stable
hydrocarbon fluids which show great promise for the production of more stable turbine oils. One
route to these hydrocarbon base fluids is through the oligomerization of olefins; the second involves
the catalytic hydrocracking and isomerization of petroleum fractions. Kramer summarizes the history
and current state of the Evolution of Base Oil Technology.
The use of such highly paraffinic, low heterocycle hydrocarbon base stocks can lead to steam and
gas turbine lubricants with significantly improved oxidation resistance and better thermal stability.
Three papers from different lubricant suppliers address some of these performance advantages these
formulators have discovered using new technology base oil. Irvine discusses the Performance Ad-
vantages of Turbine Oils Formulated with Group II and Group III Basestocks; Schwager and Hardy
address the Improved Response of Turbine Oils Based on Group II Hydrocracked Base Oils, while
Okazaki covers the Performance Advantages of Turbine Oils Formulated with Group II Base Oils.
Regardless of the stability of lubricating fluids, successful use requires that the lubricant be regu-
larly monitored to ensure continued suitability for use. Swift, Butler, and Dewald present new infor-
mation on Turbine Oil Quality and Field Application Requirements. Kauffman and Ameye describe
the use of a new instrument for oil analysis, in Antioxidant Analysis for Monitoring the Remaining
Useful Life of Turbine Fluids.
This publication would not have been possible without the contributions of time, knowledge, and
enthusiasm from our authors; the willingness of their employers to support this effort; the reviewers
who read the papers and offered suggestions for improvement; and the ASTM personnel who pro-
vided editorial assistance and a firm hand to keep us on schedule. The co-Chairs wish to thank all who
made this Symposium a success.
William R. Herguth
Herguth Laboratories, Inc.
VaUejo, California, USA;
symposium chairman and STP editor
Thomas M. Warne
Chevron Global Lubricants
Richmond, California, USA;
symposium co-chairman and STP editor
W. David Phillips ~
The Use of a Fire-Resistant Turbine Lubricant: Europe Looks to the Future
Reference: Phillips, W. D., "The Use of a Fire-Resistant Turbine Lubricant: Europe

Looks to the Future," Turbine Lubrication in the 21st Century, ASTM STP 1407, W. R.
Herguth and T. M. Warne, Eds., American Society for Testing and Materials, West
Conshohocken, PA, 2001.
Abstract: Turbine oil fires continue to cause concern. Although not fi'equent occurrences,
a serious fire can have an enormous financial irr~act. To reduce the risk ofhydrauiic oil
fires in steam turbines, phosphate esters are now widely used, but large volumes of
inflammable mineral oil remain in the lubrication system. In order to decrease the fire risk
still further, phosphates have also been used in both steam and gas turbines as fire-resistant
lubricants. This paper reviews the need for these products and the experience in both trials
and commercial operation. It examines the reasons for their slow adoption by industry but
also why current market pressures, particularly in Europe, are likely to accelerate their
use.
Keywords: safety, fire-resistant turbine lubricants, turbine fires, fire protection, phosphate
esters, steam turbines, gas turbines, fluid conditioning, life-cycle costs
Introduction
In 1944, at a meeting of the Machines Technical Committee of the German Power

Station Association, a report was made on the operation ofa 6 MW steam turbine with a
new fire-resistant lubricant based on tricresyl phosphate. After 6000 hours the experience
was regarded as totally satisfactory [1]. This is the first known use of a phosphate ester-
based turbine lubricant. The objective then, as it remains today, was to find a way of
overcoming the main disadvantage associated with mineral turbine oils, namely their
inflammability, and to avoid the occurrence of turbine oil fires with their impact on
operator safety; the often huge cost of repairs and reduced availability of equipment.
In the intervening period much work has taken place to demonstrate the technical
feasibility of using fire-resistant turbine lubricants based on aryl phosphate ester fluids as
i Global Market Leader, Performance Additives and Fluids, Great Lakes Chemical
Corporation, Tenax Road, Tratford Park, Manchester, M17 1WT, United Kingdom.
1
Copyright92001 by ASTMInternational www.astm.org

2 TURBINELUBRICATION IN THE 21 sT CENTURY
alternatives to mineral oils. Not only have trials in both steam and gas turbines taken place
but substantial commercial use has arisen in certain market segments. Their favorable
impact on safety has also been confirmed during this period following widespread use as
turbine control fluids-particularly in large steam turbines of 250-1500 MW where steam
temperatures have risen close to 600 ~
This paper summarizes the latest position with regard to the remaining "trials" on these
fluids; their current commercial use and, particularly in Germany, the factors which are
resulting in their promotion by some large utilities, their trade association and by the
insurance industry.
Turbine Fires
To many people in the power generation industry, the idea that turbo-generator fires
are a concern comes as a surprise! Some utilities, in fact, would go as far as to suggest
that fires are unknown in their stations. It is true that large fires are not a frequent
occurrence. On closer investigation, however, the situation may be somewhat different as
fires can go unreported if they are quickly extinguished and cause neither an unscheduled
outage nor casualties. Obviously a severe fire is not good publicity and can shake
shareholder confidence but even small fires are important as they can be symptomatic of a
greater problem which could eventually lead to a more serious incident. In examining the
limited statistics available we should therefore be aware that they may not truly represent
the extent of the problem
Unfortunately few detailed investigations into the origins and frequency of turbine fires
have been undertaken. The most rigorous was published in 1983 as an Electric Power
Research Institute (EPRI) report entitled "Turbine Generator Fire Protection by Sprinkler
System" [2]. This was based on 151 responses of 210 U.S. utilities and related to 1181
turbines (principally steam turbines) of>60 MW output. Between 1930 and 1983 some
175 fires were reported of which 121 involved oil either as a primary or secondary source
of ignition. Six of these fires involved nuclear units. The study also revealed that in the
early 1980s only 285 turbines (24%) of those surveyed had any form of fire protection
around the bearings while 350 (30%) had some form of protection on the oil piping.
Recent discussions with the author of this report and other authorities in the USA suggest
that, while fire protection has improved, there are still many units which are unprotected.
The frequency of turbine fires in the EPRI report appeared to increase from about 1 in
200 unit years in the 1950s through 1 in 145 unit years in the 1960s to about 1 in 100 unit
years in the 1970s. (These are probably conservative estimates.) The increase in frequency
is thought to be due mainly to better reporting but could also be the result of~ for example,
higher steam temperatures. These figures are also to be considered against the increasing
use of fire-resistant control fluids which were introduced in the mid-1950s. Unfortunately
there does not appear to be any published data for the 80s and 90s which would confirm,
or otherwise, this trend.
Apart from the above study little detailed data on turbine fires appears publicly
available. Some reports are published by the insurance industry, for example the brokers
Marsh and McLennan have issued periodic reports [3]. While the utility industry normally
avoids publicizing information on fires, in Europe the large German-hased utility
PHILLIPS ON FIRE-RESISTANT TURBINE LUBRICANT 3
association (the Technische Vereinigung der Grosskrattwerksbetreiber-VGB) maintains a

list of major fires. This currently identifies 78 that have taken place, mainly in the USA and
Europe, since 1972 [4]. In the former Soviet Union about 140 incidents took place
between 1980 and 1986 [5] and this was atter many units had already converted to fire-
resistant control fluids!
The costs of turbine fires can, in severe cases, be enormous. One publication [6]
reported on the costs of twenty large fires occurring between 1982 and 1991 where the
total property damage was $417 miUion-an average of $22.7 million per incident. These
figures did not take outage costs into consideration which could be up to double the repair
costs. The average outage period in the cases cited was 200 days. Although these figures
are for the worst incidents, a risk-benefit analysis undertaken for the EPRI report [4]
indicated that in 1983 dollars the potential cost to the utility of operating a turbo-
generator unit without fire protection for 30 years would be $1.62 million and $0.87
million for a 600 MW and 900 MW turbo-generator respectively. Today these figures
would be closer to $6 million and $3 million (or $200 000 and $100 000/year). In the UK,
two 500 MW sets have been extensively damaged as a result of turbine oil fires in the last
four years.
While repair costs can normally be covered by insurance, outage or business
interruption costs, particularly during the commissioning of new equipment when the fire
risk is probably at its highest, may not necessarily be insured. Large utilities also tend to
carry their own insurance and to be able to rely on excess capacity in times of need, a
situation that is changing with privatisation. Even when insurance is available, one of the
results of a fire can be a substantial increase in premiums as the insurance companies
attempt to recover their losses. Significant inconvenience in the post-fire period can also
be expected as alternative power supplies are sourced and the site is cleared.
Clearly, in view of the danger to life and the high financial cost, adequate fire
protection should be a priority for the utility and for many years sprinkler systems have
been used with steam turbines and gas inerting systems with combustion turbines. These
are both forms of "active" fire protection where the fire is extinguished after ignition and
can be expensive to install and maintain. The cost of mechanical fire protection for a large
steam turbine, for example, would be in the region of $40 000-100 000.
Although they are effective when correctly installed and maintained there are occasions
when availability can be impaired e.g. as a result of incomplete maintenance [ 7]. The
possibility of false alarms may be low but they are reported [8]. Lightening, for example, is
known to have activated detection systems and resulted in the unscheduled shutdown of
gas turbines with considerable damage to the beatings. As a result there may be a
reluctance to operate these systems automatically; indeed in some stations a visual
observation of a fire is relied on more than automatic means. Even the best "active"
systems are, however, of little use in the event of a catastrophic failure of the turbine with
the expulsion of blades through the turbine casing when both oil and water lines in the
vicinity of the turbine can be destroyed.
Fire protection techniques that eliminate the possibility of fire are clearly to be
preferred. An example of a "passive" protection measure would be the use of guarded
piping but this is expensive.
4 TURBINELUBRICATION IN THE 21STCENTURY
Fire-Resistant Lubricants
An alternative approach has been to consider the use of fire-resistant lubricants. Such
products offer:
9 built-in protection
9 protection throughout the whole of the lubrication system
9 protection which is available 100% of the time the fluid is in the system and which does
not deteriorate with time.
Several types of synthetic turbine oil have been considered in the past. Due to the
necessity for operation at high temperatures and high bearing loads the focus has been on
non-aqueous fluids. Initially polychlorinated biphenyls were evaluated but, while they
possessed excellent fire resistance, lubrication problems were found when used alone [9],
To overcome this deficit they were blended with triaryl phosphates and successfully tested.
However, when pcbs were banned in the 1970s for toxicity and environmental reasons, the
subsequent development concentrated on phosphate esters. The fact that they were
already widely used in turbine control systems was an obvious advantage as it meant that
the same fluid could possibly he used for both systems.
Synthetic carboxylate esters from trimethylolpropane or pentaerythritol and short
chain acids (C5-9) are used as low viscosity base-stocks for aviation gas turbine oils, while
higher viscosity esters from trimethylolpropane and C.8 unsaturated acids are occasionally
used in turbine control systems. However, this type of product does not possess the same
level of fire resistance as the phosphate esters (see Table 1) and where it has been used in
the hydraulic systems of large steam turbines, fires have resulted. Consequently, to date,
this type of fluid has not been considered as a fire-resistant turbine lubricant.
The main advantage oftriaryl phosphate esters is undoubtedly their fire-resistance. For
example they have autoiguition temperatures in the region of 550-590 ~ and possess
inherent seN-extinguishing properties. This means that if, under severe conditions, they do
ignite the flame does not propagate once the fluid has moved away fi'om the source of
ignition. Additionally these fluids possess excellent lubricating characteristics,
demonstrated by their wide use as anti-wear additives for improving the lubricating
properties of both mineral and synthetic oils. A summary of their fire resistance properties
in comparison with mineral oils and carboxylate esters is given in Table 1.
Although the phosphate esters have some characteristics in common with mineral oils
(see Table 2), there are aspects of their performance that are quite different. These
include:
9 viscosity/temperature characteristics where phosphates normally have much lower
viscosity indices. This difference requires that tank heating be available in order to
ensure that the viscosity is low enough for pumping on start-up. This aspect of design
is, however, fairly common in conventional systems.
9 density. Phosphates have values -30% higher than mineral oil possibly necessitating
more powerful pumps and a higher static head to avoid cavitation.
9 hydrolysis of the phosphate. This is a chemical reaction of water with the phosphate
which results in the production of acidic degradation products. If not controlled this
reaction can have an adverse impact on fluid life as the acid produced has an auto-
catalytic effect on fluid breakdown as well as promoting corrosion at high levels. In
PHILLIPS ON FIRE-RESISTANTTURBINE LUBRICANT 5
order to overcome this disadvantage the fluid is normally conditioned by circulating

through an adsorbent media on a by-pass system (see below)
Table 1-.4 Comparison of the Fire-Resistance Properties of an ISO VG 46 Mineral

Turbine Oil and Triaryl Phosphate Ester and an ISO VG 68 Carboxylate Ester
Property Test Method Mineral Polyol Ester Phosphate

Oil Ester
Flash point-Open Cup (~ ISO 2592 220 266 270
point (~ ISO 2592 245 313 365
Auto-ignition temperature (~ ASTM D2155 340 430 575
Hot manifold ignition (~ AMS 3150C 350 395 >800
Wick ignition ISO 14935 Fail Fail Pass
Factory Mutual Spray test FM Std. 6930
-Persistence of burning Fail Pass Pass
-Spray Flammability Parameter Group 3 Group 2 Group 1
Persistence of burning ISO 15029-1 Fail Pass Pass
Ignitability Index ISO 15029-2 Group H Group G Group D
(worst)
Compression ignition test MIL-PERF-
-ignition ratio 19457D 10 18 >42
9 incompatibility with conventional paints and seals. Fortunately a range of suitable seals
are now available, in particular fluorocarbon elastomers which are increasingly used in
turbine appfications. The interior surfaces of systems for use with phosphates are
preferably left unpainted as rusting is not normally a problem. However, the industry
trend is towards the use of stainless steel which would obviate the need for coatings.
In the same way that mineral oils can vary in their chemical composition and
performance, so phosphate esters can also vary depending on the raw materials ~om
which they are produced. There are three basic types of product used in turbine
appfications, trixylyl phosphate, isopropylatedphenyl phosphate and tertiarybutylphenyl
phosphate ester. A comparison of the main properties which are structurally influenced is
given in Table 3 and the stability properties are seen to vary significantly. As a result TXP-
based products are preferred for applications where water contamination is likely to be a
concern, while for applications where high temperature stress is unavoidable, for example
in gas turbines, the tertiarybutylphenyl phosphates may be preferred
The differences in properties in comparison with mineral oils require that systems be
designed for use with phosphates. It is unfortunately not normally possible to convert a
system from mineral oil to phosphate by draining the oil and refilling with the new fluid,
due mainly to compatibility aspects, and system modifications may also be needed e.g. to
the pump train as a result of the higher fluid density. Contact with the turbine builder is
necessary to determine what equipment modifications are required.
Table 2-,4 Comparison of Some Typical Properties of an ISO VG 46 Mineral Turbine Oil
and Phosphate Ester
Property Mineral Oil Triaryl Phosphate

Viscosity at 100 ~ (cSt) 6.7 5.0
at 40 ~ 46 43
at 0 6C 450 1700
Pour Point (~ -9 -20
Acid number (mgKOH/g) 0.15 0.05
Specific Gravity at 20 ~ 0.870 1.135
Air release (min) 3 1
Foaming tendency/stability-24 ~ (ml) 250/0 30/0
Water content (%) 0.001 0.06
Rust prevention (distilled water) Pass Fail
Specific heat at 20 ~ (J/gK) 1.76 1.69
Thermal conductivity (W/mK) 0.134 0.132
Fire-Resistant Turbine Control Fluids
Following a number of turbine fires in the early 1950s, the concept of a split hydraulic
and lubricating oil system with the use of a fire-resistant hydraulic fluid in the control
system, was introduced. This move was based on the assumption, since shown to be
erroneous, that the great majority of turbine fires originate from the hydraulic system as a
consequence of the high hydraulic pressures and the proximity of high temperature steam
pipes. In fact the statistics show that there is little difference between the hydraulic and
lube oil systems as sources of fires-at least in steam turbines [4,10].
The earliest fluids used were blends ofpolychlorinated bipbenyls (pcbs) and
phosphates. The latter were added to reduce the viscosity and density of the pcbs without
significantly adversely affecting their fire resistance. The use ofpcbs was, however,
quickly discontinued in view of valve erosion problems as well as toxicity and
environmental concerns. Since the early 1970s, phosphate esters have been the only fire-
resistant fluids approved by the turbine builders for this application although small
amounts of more flammable carboxylate or synthetic esters have been used in refurbished
systems. It is thought that there are over 1000 large steam turbines (>250 MW output)
currently in operation worldwide with fire-resistant hydraulic fluids based on phosphate
esters and the use of these fluids has largely eliminated the risk of fire arising from ignition
of the hydraulic fluid.
Experience with Fire-Resistant Lubricants
a) Steam Turbines
As suggested earlier, while the use of a fire-resistant hydraulic fluid is effective in

reducing the fire hazard associated with the control system, the risk of fire associated with
the mineral turbine oil remains. Large volumes ofoil, possibly up to 100 m 3 in the very
largest sets, remain in the bearing lubrication system. Of particular concern in this respect
is the common use ofshafl-driven oil pumps which can continue to feed a fire with oil as
long as the shaft is turning. A leak o f fire-resistant lubricant obviously does not pose the
same threat.
Table 3-The Effect of Chemical Type on the Performance of Phosphate Esters
Chemical Type Autoignition Air Release Hydrolytic Oxidative

Temperature (mins) Stability Stability
Isopropylphenyl 540 5 Moderate Moderate
Tertbutylphenyl 580 5 Poor-Moderate Excellent
Xylyl 570-590 1 Good-Very good Good
The very earliest use of a fire-resistant lubricant appears to have occurred during the
1930s when pcbs were investigated [9] as a result of a series of fires. Although some
satisfactory operating experience was obtained, interest lapsed and nothing further was
done until the 1950s when a further series of fires "sparked" a revival o f interest. The
topic was then taken up by some U.S. utilities and also in the late 1960s by General
Electric who converted a 12.5 MW set at their Medium Steam Turbine Division. Over a
decade of valuable operating experience was obtained with phosphate esters in this
industrial set which, after initial teething problems ran very successfully [11].
Subsequently, in Western Europe, several trials were also carded out in the UK, Germany
and France [12]. In the earliest of these dating from 1974, the Technical University of
Munich began to test a phosphate ester in a small steam turbine o f 1.5 MW and this was
followed in 1979 by the conversion of another unit of 5.5 MW [13]. The larger of these
units is still in operation today and has operated on two charges of fluid in this time
period.
In 1982 in Basel, a 10 MW back-pressure turbine was converted from mineral oil to
phosphate after experiencing foaming and deposition problems. In this unit the new fluid
was used not only for the turbine control and lubrication systems but also for the gears,
the jacking system, and the air-cooled generator. Today, after about 60 000 hours of
operation involving frequent starts which place increased stress on the fluid, the turbine
continues to operate on the same fluid charge filled into the unit in 1982. The quality of
the fluid is still as good as new and it is only necessary to replace the conditioning medium
used to remove fluid degradation products, annually. Table 4 shows some of the analytical
data obtained over the period ofserviee to date. Examinations of bearings and reduction
gear have not revealed any significant wear and the occasional severe water contamination
has not caused any unscheduled shutdown or significant change to the fluid properties.
More details on this experience are available in the relevant references [12,14].
A very serious incident during operation was avoided as a result of using the fire-
resistant lubricant. A boiler failure occurred which led to a loss of steam pressure.
Unfortunately, the circuit breaker which should have automatically isolated the generator,
failed to trip. The generator therefore now operated as an "engine", motoring the turbine.
It was not immediately possible to switeh-offthe power remotely because of the impact
this would have on the rest of the manufacturing site. This resulted in the turbine
overheating and the heat generated at the blade tips began to produce local welding with
subsequent jamming of the blades in the housing. At this stage the generator was
consuming about 3 MW of power! Eventually it was possible to open the switch but the
damage cost about 1 million SFr to repair. Outage costs were about 0.5million SFr and the
equipment was out of operation for about 1 year. Although fluid escaped from the
lubricant system there was no fire and when the unit was eventually returned to use, the
same fluid charge was re-instaUed.
Table 4-Analytical Data on a Phosphate Ester Usedfor the Lubrication of a IOMW

Steam Turbine
Property Units 0 3000 14000 26000 40000 55000

hrs hrs brs hrs hrs hrs
4/83 5/99
Viscosity cSt 42.7 42.7 41.6 42.4 43.5 43.4
Acid Number mgKOH/g 0.02 0.05 0.02 0.02 0.01 0.03
Water Content % 0.05 0.08 0.16 O.11 0.05 0.07
Air release rnin 2 1 1 1 1 1
Value
Foaming ml 190/0 240/0 280/0 210/0 370/0 400/0
Cleanliness SAE Class 4 0 1 0 0 1
While the tests in Europe and the USA focussed on small industrial sets it was in the
FSU that the most significant developments took place. Following trials in a 16.5 MW
industrial unit in 1959, a 300 MW set (including the generator) belonging to the All Russia
Thermal Engineering Institute in Moscow was converted to use phosphate ester and a trial
started in the early 1970s [15]. The operating experience obtained led directly to the first
commercial use of phosphate esters in large steam turbines with the conversion of a 220
MW nuclear unit at Kolskaya in 1982.
Today, the same operating charge is still in use and the turbine continues to function
satisfactorily. This has been achieved without any off-line conditioning which is common
in Western control and lubrication system design when using phosphate esters. This is a
result of the improved hydrolytic stability of the phosphate developed by the All Russia
Thermal Engineering Institute [5]. Since 1982 other large steam turbines of up to 1000
MW have been converted or operated from new [16]. Table 5 indicates those units
currently in operation. Although it was originally planned to convert all the existing
220/800 MW sets to phosphate, capital limitations have forced the postponement o f this
programme. Most of the units to date have continued to use mineral oil for the generator
bearings in view of incompatibility with the insulants used in the hydrogen-cooled units,
but the latest generator designs also use phosphates for bearing lubrication.
Outside the FSU, interest in the lubrication of large steam turbines was limited to
detailed studies by EPRI in 1983 of the feas$ility of adapting a large steam turbine [17]. It
was concluded that such a conversion was possible and that phosphate esters would be
suitable but that the modifications would be expensive. A further review of the existing
experience with fire-resistant lubricants by EPRI some six years later [18] also concluded
that phosphate esters could safely be used to lubricate small-medium sized units with the
knowledge then available.
Table 5-Use of a Fire-Resistant Lubricant in Turbines of the CIS
Power Station Type Output Date Conversion Operating

(MW) Installed to Phosphate hours on
Phosphate
Kola Unit 1 Nuclear 220 1982 1983 110000
Zapomjie Unit 6 Thermal 800 1977 1985 77500
Zaporojie Unit 5 Thermal 800 1975 1992 61000
Perm Unit 1 Thermal 800 1986 1986 95200
Perm Unit 2 Thermal 800 1987 1987 86350
Penn Unit 3 Thermal 800 1990 1990 68200
Penn Unit 4 Thermal 800 1994 1994 42500
Rovny Unit 5 Nuclear 1000 1986 1986 95100
Khmelnitsk Nuclear 1000 1987 1987 86450
Unit 1
Yuzhno-Ukraine Nuclear 1000 1989 1909 70500
Unit 3
b) Gas Turbines
Although the experience in steam turbines with fire-resistant lubricants is probably

better known, there is far greater experience in gas turbines. The earliest known operation
with fire-resistant lubricants in this application commenced in 1958 when Texas Eastern
Corporation decided to test phosphate esters in [19]. The trials were successful and Texas
Eastern then planned the conversion of the remainder o f their turbines. In Canada, the new
pipelines that sprang up for gas transmission in the 1960s also decided to use fire-resistant
lubricants and companies such as TransCanada Pipelines. Great Lakes Gas, Alberta Gas
(subsequently the Nova Corporation), West Coast Transmission, Dome Petroleum etc.
introduced these fluids into their equipment where possible-many from their initial
installation. If industrial turbines were installed, the phosphates lubricated both the turbine
and the associated compressor but if an aero-derivative unit was the prime mover, then the
phosphate ester, if used, was only for lubrication of the compressor.
10 TURBINELUBRICATION IN THE 21 STCENTURY
In the area of power generation, phosphates were used by General Electric (GE) as the
combined hydraulic fluid and lubricant in some Frame 7 units produced in the 1970s. For a
period there was a hot middle bearing which resulted in fires with mineral oil. Before the
problem was "designed out," phosphate esters were used as lubricants and about ten units
still run on these fluids today with some turbines approaching 100000 hours operation.
The number of gas turbines/compressors currently operating on phosphates in North
America is about 150 and includes equipment from G.E., Canadian Westinghouse, Nuovo
Pignone and Solar. The introduction of dry gas seals in the 1990s encouraged some users
in Canada to retum from phosphates to mineral oils but a series of fires with these units
resulted in government pressure to move all possible equipment to phosphates-including
those units that had been supplied with mineral oil.
In Europe, by comparison, the experience in gas turbines is very limited. A 5 MW
Sulzer Type 3 unit was installed by Ciba-Geigy in Basel in 1986, in the same power station
as the steam turbine. The turbine, reduction gears and generator have operated
successfully on the same fluid charge since commissioning. Table 6 shows the results of
analyses on fluid samples taken from the turbine over this period.
Two incidents have occurred during operation with the fire-resistant lubricant which,
had mineral oil been used, would have resulted in fires. The first involved a screw coupling
failure on the line from the high pressure lifting pump which resulted in fluid being sprayed
over the hot end of the turbine. Because no fire occurred it was possible to follow normal
shutdown procedures. The second was a leakage from the power turbine shall seal and oil
escaped onto the hot housing. Instead of having to shut the unit down to replace the seal,
operation continued with the fluid being collected until a convenient shutdown occasion
occurred, thus avoiding an unwanted outage [14].
Fluid Maintenance
As with existing turbine oils it is necessary to adequately maintain phosphates in order

to ensure efficient operation of the system and an acceptable fluid life. The parameters that
are of most concern are the increase in acidity, water content and particulate
contamination level. When triaryl phosphates degrade the most common result is an
increase in acidity with little effect on viscosity change. Consequently this latter property is
not one that requires regular attention. Water and acidity, however, need monitoring. To
date there are no specifications for used fire-resistant turbine lubricants but there is a
tendency to adopt the limits already in existence for phosphate esters when used as turbine
control fluids i.r a maximum acidity of 0.2-0.3 mgKOH/g.
While the presence of some dissolved water can be tolerated if the acidity is low, free
water is to be avoided in view of the potential adverse effect on rusting, emulsification etc.
As the solubility of water in phosphates is very much higher than in oil (reaching about
2500ppm at 25 ~ this is not usually a problem and the level of fluid acidity will normally
determine the suitability of the fluid for continued use.
PHILLIPS ON FIRE-RESISTANT TURBINE LUBRICANT 11
Table 6-Analytical Data of a Triaryl Phosphate Ester Usedfor the Lubrication of a 5

MW Industrial Gas Turbine
Property Units Jan '88 '90 '92 '94 '96 '98 12/99
'86 42000 82000
0 hrs hrs hrs
Viscosity cSt 45.2 46.3 46.4 46.0 46.0 45.8 45.0 45.4
at 40 ~
Acid mgKOH/g 0.07 0.08 0.01 0.01 0.01 0.04 0.04 0.08
Number
Water % 0.03 0.03 0.03 0.06 0.06 0.08 0.06 0.07
Content
Air release min 1 2 2.5 2 2 3 2 2
Value
The key to the cost-effective use of phosphate esters is the use of conditioning media
to remove acid degradation products. Initially, fullers earth and activated alumina were
used for this purpose, but experience with these adsorbents has been unsatisfactory in
systems where the acidity was allowed to rise too high. Deposits occurred and the air
release/foaming properties deteriorated. The reason for this behavior was that the filter
media, for example fullers earth, contained calcium and magnesium carbonates. These
components reacted with the acidic decomposition products from the phosphate to
produce soluble, and eventually insoluble, metal phosphate salts [20] which had an adverse
effect on the surface-active properties of the fluid.
These treatment methods could substantially increase the life of the fluid where the
system was well maintained and a low fluid acidity level was achieved. However, in cases
where the fluid was severely stressed and it was impossible to maintain a low level of
acidity this would eventually lead to an early fluid replacement and possibly the necessity
of physically removing the deposits from internal surfaces by flushing and manual cleaning.
Fortunately this problem is now solved. With the introduction of new adsorbents based on
ion-exchange resins [21], it is now possible to keep the fluid in the system for many years.
TransCanada Pipelines for example have reported units accumulating over 200 000 hours
(25 years) on the same fluid charge. In 1998 they had over 14 million operating hours on
these fluids [22]. With this length of fluid life, the life-cycle costs become very favorable
and concerns regarding the disposal of the fluid are minimized. The resins used are
however, wet and it is necessary to use a vacuum dehydration unit downstream to remove
the displaced water. This equipment replaces the centrifuge used to remove water from
wet turbine oil.
Fire-Resistant Fluids and Lubricants-Features and Benefits
As a result of the operating experience to date the principal features and benefits
associated with phosphate ester fire-resistant fluids and lubricants
are summarized in Table 7.
Table 7-The Principal Features and Benefits of a Phosphate Ester Turbine Lubricant
Feature Principal Benefit

Fire resistance 9 Improved safety/worker protection
9 Avoidance o f fires leading to reduced
downtime/increased availability
9 Elimination of mechanical fire protection
around the turbine with space saving and
no maintenance requirements
9 Avoidance of shutdowns due to spurious
alarms
9 Reduction in insurance premiums
9 Simpler building construction e.g.
removing the need to isolate the
reservoir
9 Greater shareholder confidence
Very long life when used with state of

Favorable life-cycle costs
] the art fluid conditioning
Reduced downtime for replacement and
(in most applications) lower maintenance
requirements
9 Minimizes fluid disposal costs.
System modifications possible with
9 Single hydraulic and lube oil system
favourable cost impact
9 Reduction in tank capacity o f up to 25%
with certain fluid types
9 Possibility of higher bulk fluid
temperatures
9 Simpler line construction leading to
improved accessibility for repair etc.
Obstacles to Progress with Fire-Resistant Lubricants
With such a large amount o f satisfactory operating experience it is perhaps surprising

that industry has not adopted these fluids more widely. There are several reasons for this
as follows:
1) The power generation industry is extremely conservative.
2) As suggested earlier there is no wide acceptance that turbine fires are a continuing
problerrL The builders argue, as a result, that there is no pressure from the market. In fact
there have been a number o f enquiries in recent years in which the builders have, at least
initially, shown little interest.
3) Although the capital cost o f the equipment when using a fire-resistant lubricant is not
significantly different from when mineral oil is used, the turbine builder is understandably
reluctant to introduce "design specials". Moving to a new fluid can result in considerable
effort in modifying drawings, seeking alternative components, logistics problems with
carrying different spares etc. At a time when competition is fierce such changes are
unwelcome to the manufacturers, particularly if they have no spare capacity for their
standard designs.
4) If the builders are involved in maintenance contracts, they are insisting that these
contracts will be invalidated if phosphates are used and problems occur!
5) A fire-resistant lubricant is significantly more expensive than one based on mineral oil.
However, the cost savings associated with the elimination of fire-protection around the
turbine, favorable life-cycle costs and the safety benefits associated with the new fluids are
not necessarily understood or taken into consideration when deciding what lubricant shall
be used.
6) There is a perception that phosphate esters are noxious and require special precautions
for handling. This is not the ease. Extensive toxicity and eeotoxicity data are now available
on these fluids [23] which have failed to reveal a significant risk and the handling
precautions are the same as are recommended for mineral oil.
7) With the emphasis today on the high availability of equipment, users are not prepared to
compromise that availability. If therefore the builder shows a reluctance to offer
phosphates and to guarantee equipment availability with these fluids, the user will naturally
think twice about their adoption.
Technical and Commercial Drivers Towards Fire-Resistant Lubricants
In spite of the reluctance of some turbine builders to introduce fire-resistant lubricant

technology there are a number of factors, particularly in Europe, which are steadily
influencing the market in this directiorL For example, in the search for higher equipment
efficiency, operating temperatures, and therefore also the fire risk, are continuing to
increase. Gas turbine inlet temtg~atures have increased from 700 ~ in 1950s to 1500 ~
today [24] while steam temperatures are now at about 600 ~ with the intention to
increase still further to 650 ~ when the metallurgy permits. In an attempt to reduce
generating costs, the number of operatives in the utilities has decreased dramatically-by up
to half in some cases. Consequently there are now fewer people 'walking the lines' and
looking for leaks and signs of other potential problems. Less time is available for training
and maintenance programmes are now targeting specific pieces of equipment rather than a
'holistic' approach to the system.
With the drive to increased margins there is a tendency to keep existing plant in
economical operation for as long as possible. In the USA for example ~70% of fossil units
are currently more than 30 years old. With age comes a greater chance of metal fatigue
and an escape ofoil. The current trend of refurbishing old units may also result in
extending the life of existing pipework with the same effect.
In Europe, because of a number of fires, the German utility association (VGB) is now
promoting the use of fire-resistant lubricants in both gas and steam turbines following
advice from their technical committees [25]. Also in Europe a series of Commission
Directives aimed at improving worker protection have been published. In the UK, for
example, the Supply of Machinery (Safety) Regulations arising from the Machinery Safety
14 TURBINE LUBRICATION IN THE 21sT CENTURY
Directive state "Machinery must be designed and constructed to avoid all risk offire
posed by the machinery itself or by gases, liquids etc. produced or used by the
machinery". A similar statement appears in the legislation of all EC member states. The
legislation also contains provisions for the imposition of penalties in the event of a fire
arising as a result of a lack of implementation.
The insurance industry in Europe is very supportive of the use of the new fluids.
Allianz Insurance, for example, is offering substantial discounts on insurance premiums
and also technical support to the customer in the event that they use fire-resistant
lubricants [26]. In the USA, the NFPA (National Fire Protection Association) and Factory
Mutual already recommend fire-resistant lubricants for use in gas turbines as an alternative
to mineral oil and mechanical protection [27,28] and the scope of the latter's standard for
"Less Flammable Hydraulic Fluids" is being extended to include turbine lubricating oils.
In order to examine the likely cost-effectiveness of phosphate esters, a committee was
set up in Germany in 1998 under the chairmanship ofAlllanz GmbH and consisted of
members of some of the German utilities, European turbine builders and fluid suppliers. It
concluded that "the use of a fire-resistant fluid in the control and lubrication system of a
500 MW steam turbine offers a cost advantage against mineral oil and is at least cost
neutral in combined cycle equipment" [29]. The author's experience in industrial gas
turbines would support this conclusion.
The Future
There is no doubt that the industrial trends indicated above, particularly the regulatory
aspects, will encourage the wider use of fire-resistant lubricants. Utilities will also see their
application as a means of avoiding fires and maintaining safe working conditions under
strong competitive pressures. In fact there is already a move in this direction from some of
the larger utilities in Europe particularly when considering investment in new plant. New
industrial steam and gas turbines are also expected to be in operation in the next twelve to
eighteen months and, as the advantages become dearer and more quantifiable, the
momentum is expected to gather pace.
References
[1] Der Erfahrungaustausch der Maschinentechnischen Ausschfisse der Vereinigung der

Eleklrizitdtswerke, 1944, p.359.
[2] Turbine Generator Fire Protection by Sprinkler System, EPRI Report No. NP 4144,
1985, Electric Power Research Institute, Palo Alto, CA.
[3] Biggins, J., ed., ".4 30-Year Study of Large Losses in the Gas and Utility Industry,
1969-1998," Marsh & McLennan, December 1999.
[4] Schenck, K., Hfixtermann, E. and Hartwig, J., "Operation of Turbines with Fire-
resistant Fluids, Including the Lubricating System," VGB Kraftwerkstechnik,
1997, 77,6, pp.412-416.
[5] Vilyanskaya,E. D., Lysko,V.V., Vainstein, A.G. and Mart'yanov,V.M., "Soviet Fire-
Resistant Turbine Oils and Their Use in Power Generation," Elelar.Stn, 1989, 1,
pp.44-48.
[6] Hathaway, L .IL, "Don't Be Caught Off Guard When it Comes to Protection,"
Power, 1994,138,7, pp.50-52.
[7] "Sprinkler Impairments: Planning and Prevention," Factory Mutual Record,

Nov/Dec. 1988, 3-8, Factory Mutual Corporation.
[8] Ishack, G., "Fires and fire protection systems: feedback from operating experience,"
Nuclear Engineering and Design, 125, pp.297-308,1991.
[9] Whirl, S.F. and Knight, R.G. "Fire-Resistant Turbine Lubricating Fluids," Lubr.
Eng., 1970, 26, 5, pp.164-170.
[10] Kaspar, K., "Einsatz von synthetischen schwerbrennbaren Hydraulikflfissigkeiten im

Schmier- und Steuerkreislauf von Dampfiurbosgitzen, "Der Maschinenschaden,
1977, 50, 3, pp.87-92.
[ 11] Wolfe, G.F. and Whitehead, A., "Experience with Phosphate Ester Fluids as
Industrial Steam Turbine-Generator Lubricants," Lubr. Eng., 1978, 34, 8,
pp.413-420.
[12] Phillips, W.D., "The Use of Triaryl Phosphates as Fire-Resistant Lubricants for
Steam Turbines," ASLE Preprint, 1985, 85-AM-31-2.
[13] Sch6ner,W., "Betriebliche Bewahrung schwerbrennbarer Steuer- und

Schmierfliissigkeit in zwei Heizkraftturbosetzen," VGB Kraftwerkstechnik, 1981,
6, pp.455-459.
[14] Hartwig, J. and Phillips, W. D., "Fire-Resistant Fluids for the Control and
Lubrication of Steam and Gas turbines," ASME Paper, 1990, 90-JPGC/GT-5.
[ 15] Vilyanskaya, E. D. et al, "Experimental Full-Scale Operation of OMTI Fire-

resistant Oil in the Lubricating System of a 300MW Turbounit," Teploenergetika,
1975, 22, 8, pp.43-45.
[16] Piehugin, I, "lO00-MWDampfturbinef~rKernkraftwerke undihre

Modernisierung," VGB Kraftwerkstechnik, 1991, 71, 9, pp.842-846.
[ 17] "Fire-Retardant Lubricant for Turbo-Generators," EPRI Report NP-2843,1983.

16 TURBINE LUBRICATION IN THE 21STCENTURY
[ 18] "Evaluation of Fire-Retardant Fluids for Turbine Bearing Lubricants," EPR1

Report NP-6542, 1989.
[19] Farmer, E. P. Jr., "Fire-Resistant Lubricants in Gas Turbines," Lubr. Eng., 1971,
3, pp.83-86.
[20] Phillips, W. D., "The Conditioning of Phosphate-Ester Fluids in Turbine

Applications," Lubr. Eng., 1983, 39, 12, pp.766-778.
[21] Phillips, W. D. and Sutton, D. I., "Improved Maintenance and Life Extension of
Phosphate Esters Using 1on Exchange Treatment," Proc. 10th Internat. Coll.
Trib., 1996, Esslingen, Germany.
[22] Dut~esne, P., "14 Million Hours of Operational Experience with Phosphate Ester
Fluid in a Gas Turbine Main Bearing Application," Proc. 11th Internat. Coll.
Trib., 1998, pp. 1447-1451, Esslingen, Germany.
[23] Goode, M. J. and Phillips, W. D., "Triaryl Phosphate Ester Hydraulic Fluids- A
Reassessment of Their Toxicity and Ecotoxicity Behaviour," SAE Paper, 1998,
982004.
[24] Dodman, IC, "Efficiency will be the biggest issue," Intl. Power Gen., Septembea
1997, pp.51-53.
[25] Anon, "Brief report on the Activities of the Association," VGB PowerTech, 1998,
10, 43.
[26] Grupp, H., "Schwer enO'lammbare Hydraulilflliissigkeiten-eine

Standortbestimmung,'" Proc. 10th Internat. Coll. Trib., 1996, pp.433-443,
Esslingen,Germany.
[27] "Electric Generating Plants and High Voltage Direct Current Converter Stations,"
NFPA Standard 850-24,1996, Section 6-5.3.3.
[28] "Fire Protection for Gas Turbine Installations," Factory Mutual Loss Prevention
Data 7-79, 1997, Section 8.1, p.41, Factory Mutual Corporation.
[29] Grupp, H., "Einsatz yon schwer enOqammbaren Flassigkeiten far das Steuer-und
Schmiersystem im Turbinenbereich,'" VGB-Fachtagung, Brandschutz im
Kraftwerk, 1999, Essen, Germany.
Lois J. Gschwender, l Lynne Nelson, 2 Carl E. Snyder, Jr., l George W. Fultz3 and Costy
S. Saba3
Advanced High-Temperature Air Force Turbine Engine Oil Program
Reference: Gschwender, U J., Nelson, L., Snyder, Jr., C. E , Fultz, G. W. and Saba, C.
S., "Advanced High-Temperature Air Force Turbine Engine Oil Program,"
Turbine Lubrication in the 21 ~ Century, ASTM STP 1407, W. R. Herguth and T. M.
Warne, Eds., American Society for Testing and Materials, West Conshohocken, PA,
2001.
Abstract: Military and commercial jet aircraft engines operate hotter than in the past to
increase fuel efficiency. Also, advanced structural engine materials permit hotter
burning, more efficient engines. These trends are expected to continue. Many current
gas turbine engines experience oil system coking and other problems related to the
limited thermal and oxidative stability of the conventional, ester-based gas turbine
engine oils. This is because most engines operate on gas turbine oils limited to 150~
The best current oils have a maximum operating temperature of about 200~ We
initiated a program to develop a new class of ester-based lubricants to meet current and
future high-temperature requirements. The target temperature range of this new oil is
--40~ to 232~ Target requirements were established and communicated to industry.
This paper presents those requirements and initial results on the most promising
candidates.
Keywords: gas turbine engine oil, lubrication, antiwear additive, antioxidant, ester,
base oil
Background
Hermann Zorn, of Germany, first investigated ester oils as engine lubricants in the
1930s to replace castor oils, notorious for sludge formation and cold temperature flow
problems [1]. Esters can be tailor-synthesized to have specific desirable properties,
allowing great flexibility for a unique application. Also, esters are naturally oxidatively
T Senior Materials Engineers, Air Force Research Laboratory, WPAFB, Ohio 45433.
2 Liquid Lubricants Program Manager for PRSL, Air Force Research Laboratory,
WPAFB, Ohio 45433.
3 Assistant Research Engineer and Senior Research Chemist, University of Dayton
Research Institute, 300 College Park Dr., Dayton, Ohio 45469
17

18 TURBINE LUBRICATION IN THE 21 sT CENTURY
stable and engine oils are subjected to high airflow over thin films ofoil, making esters
a good match for aircraft engines. In 1951, United States military specification MIL-L-
7808, Lubricating Oil, Aircraft Turbine Engine, Synthetic Base, first outlined ester oil
performance requirements. Initially diesters were the base fluids. As aircraft
performance became more rigorous, thermal and oxidative demands on the engine oil
increased and more complex esters were used as engine oil base fluids. The progression
of complexity goes from diesters to trimethylolpropane esters to pentaerythritol esters to
dipentaerythritol esters. As the ester complexity increases, the cost and stability
increase. Pentaerythritol and dipentaerythritol esters have excellent low temperature
viscosity, high viscosity index (VI) and very high autogenous ignition temperature
(AIT) characteristics [1]. This high MT makes them highly desirable as engine oils,
because of fire threats from very hot engine nacelles. High MT oils allow engine
weight savings because less insulation and oil cooling is needed.
Over the years, esters used in gas turbine engines have advanced until now the US
Navy has a -40~ to 204~ gas turbine engine oil (GTO), as defined in MIL-PRF-
23699 specification, Lubricating Oil, Aircraft Turbine Engine, Synthetic Base, the HTS
(high thermal stability) type. The US Air Force has a -51~ to 204~ GTO, as defined
in MIL-PRF-7808, Grade 4 (4 cSt at 100~ Both of these higher thermal stability oils
are used where better oxidative stability is required, as they are more expensive than the
Navy STD (standard) version and the Air Force Grade 3 (3 cSt at 100~ version, both
with an upper temperature oxidative stability limit of only 177~
This oxidative stability is defined in the military specification MIL-PRF-7808 by
Corrosiveness and Oxidation Stability of Hydraulic Oils, Aircraft Turbine Engine
Lubricants, and Other Highly Refined Oils (ASTM test method D4636) but this test, as
often the case with bench tests, does not define the entire problem with GTO stability.
Another important related issue is the oil coking tendency. Engine coke is insidious
black carbonaceous deposits formed from oil oxidation and thermal decomposition.
Coke causes high engine maintenance because it interferes with heat transfer in the
engine and clogs critical engine breather tubes and nozzles that spray lubricant on the
bearings. Oil can perform well in the oxidative stability test, but produce high coke in a
coking test or in an engine. Also, conversely, low coking oil may not have bulk oil
temperature oxidation stability.
For this program, we needed a very oxidatively stable, low coking, ester-based
GTO, with the very broad target temperature range of-40~ to 232~ While more
oxidatively stable fluid classes than esters exist, esters are preferred because any new
ester fluid will be compatible with any existing engine and currently used ester oil. In
other words, it would be a "drop-in" replacement, requiring no new hardware or seals,
nor cleaning the system of the old fluid to add the new fluid.
Other performance properties are also critical as will be described. Besides the
base fluid, the performance-improving additive package must be carefully considered.
For example, the antioxidant and antiwear additives affect coke and oxidative stability.
A balance of the base fluids and additives is critical to produce a successful candidate
fully formulated GTO.
In the 1970s, the US Air Force sponsored research and development ofa GTO,
designated MlL-L-27502, Lubricating Oil, Aircraft Turbine Engine, Ester Based,
similar in target properties to the desired oil of this current program. The prototype oil
GSCHWENDER ET AL. ON AIR FORCE OIL PROGRAM 19
successfully passed a 100-h J57-P29W engine test, with the bulk oil temperature
maintained at 220~ and with 330~ hot spots. Post-test inspection found clean, very
low-coked parts [2].
That oil was never transitioned to actual use because, after several months, it
developed a haze that was the additive package agglomerating. Also at that time, the
Air Force required aircraft to operate down to -54~ and MIL-L-27502 would only
operate down to -40~ Although the GTO was never transitioned to use, the program
demonstrated the optimum thermal and oxidative stability possible with a --40~ ester-
based GTO.
Based on the experience of the MIL-L-27502 development program, the MIL-L-
27502 properties were selected as target values. This new ester-based GTO would be
the most stable ester based fluid possible.
While the GTO is being developed for primarily jet engines for aircraft, other jet
engines and engines for power generating turbines will also benefit from a better oil that
provides better fuel efficiency and less maintenance.
Advanced Ester Target Properties and Results
Target properties are established as ideal goals. In trying to meet those goals,
some may become obviously impossible to meet. Others may change because input
from the customers changes the original thinking. Pratt & Whitney, a major military
engine manufacturer, has been involved with this program from the start, providing
valuable recommendations. The final product is usually the best compromise lubricant,
where less important properties are relaxed to allow achievement of the more important
properties. Tables 1 and 2 contain the latest target properties and data from the most
promising candidate from three potential fluid suppliers. Numbers in bold fall short of
the target requirements. These are the more easy-to-conduct bench tests. The most
critical or difficult to meet among these properties are viscosity-temperature, antiwear,
static coking, bearing deposition, and corrosion-oxidation. More system-like and
expensive tests are also listed (Table 3), which will be conducted on the most promising
candidates.
Viscosity-Temperature
The higher temperature viscosity is targeted at 7.0 cSt minimum at 100~ but
down to 5.0 cSt is acceptable at this time. Higher is better because better film thickness
in bearings comes from more viscous oil. A higher film thickness means less asperity
contact and lower bearing wear. The low temperature viscosity is targeted at 20,000 cSt
maximum at --40~ to meet critical cold weather starting. Obviously the high and low
temperature requirements work against each other. It is critical to have the best
viscosity index possible. The best candidates from Sources A and B are well over the
low temperature target, but have the better higher temperature viscosity. The fluid from
source C is much closer to 20,000 cSt at -40~ but is less viscous at 100~ where
higher viscosity is highly desired. The temperatures at which the candidates are 20,000
cSt are interpolated in Table 1. We have ongoing dialogue with Pratt & Whitney
concerning all of the target properties, but especially the low temperature viscosity,
20 TURBINE LUBRICATION IN THE 21ST CENTURY
Table I - Fluid Target Properties and Candidate Data

SOURCE
TARGET TEST A B C
PROPERTY REQUIRMENT METHOD
Kinematic Viscosity (cSt) ASTM D445

at 232~ report 1.18 1.06
100~ 7.0 min target 6.45 6.32 5.48
5.0 min acceptable
40~ report 37.23 35.95 29.6
-40~ 20,000 max 30899 24950 20280
-47~176 report 129600 93710
20,000 cSt interpolated temp, ~ -36.6 -38.5
Acid Number (mg KOH/gm) 0.5 max ASTM D664 0.00 0.0 0.00
Pour Point (~ -54 max ASTM D97 -54 a -50 -57
Flash Point (~ Open Cup 230 min ASTM D92 290a 268 260
Autogeneous Ignition Temp. ~ 395 min b ASTM E659 404 a 402 399
Evaporation at 232~ (6.5 Hrs), % 15 max ASTM D972 5.3 a 12.7 20.0
Four Ball Wear Scar (mm) ASTM D4172

1 hr, 40 Kgload, 600rpm
52100, 75~ 0.7 max 0.47 0.39 0.66
M-50, 200~ 1.0 max 0.52 0.37 0.84
Static CokJng, total mg coke/gm oil 49.4 max (see text) 44.7 45.6 20.5
Shear Stability (% Viscosity Loss) 4.0 max ASTM D2603 0.8 a 1.1% +chg 1.00
a Similar formulation data - not expected to be significantly different

b Recently revised from 410~
asking if the 20,000 cSt at -40~ is a critical need. In other words, would 20,000 cSt at
-38.5~ for Source B candidate be acceptable for engine start-up capability?
Antiwear
We have selected the four ball wear test, Wear Preventive Characteristics of
Lubricating Fluid, D4172, for 52100 steel at 75~ and M-50 steel at 200~ for the GTO
development program as it is a good screening test for boundary lubrication. The
ultimate test is in the engine performance.
A good phosphorus-containing antiwear additive is required to assist in boundary
lubrication of GTOs. Higher molecular weight phosphates are commercially available
from several sources and seem to meet the requirements, when properly selected.. Ifa
phosphate is too volatile, it evaporates and is not available in the application. Antiwear
additive selection must be made carefully. We have observed that different antiwear
phosphates can negatively impact the corrosion-oxidation test, the amount of static
coking and the low temperature viscosity (Table 1) Table 4 is an example of 25% by
Table 2 - Corrosion-Oxidation Fluid Target Properties and Candidate Data

ASTM D4636
TARGET SOURCE
PROPERTY REQUIRMENT A B C
Corrosion-Oxidation Stability, 10 IJh air flow

Temperature, 220~ 48 hr, 165 ml
Viscosity Change, (%) -5 to 25 8.3 15.3 18.2
Acid Number Change (rng KOHIgm) 4.0 max 0.87 3.38 1.4
% Fluid loss 8.0 max 2.9 2.1 2.7
Metal weight Change (rng/cm=)
AI (Alloy 2024) 0.2 max 0.03 -0.01 0.00
Ag (99.9% pure) 0.2 max 0.00 -0.04 -0.05
Bz ( Si-Fe, AMS 4616, AZ31B) 0.2 max -001 -0.03 -0.01
Steel (grade 1010) 0.2 max 0.02 0.02 -0.03
M-50 (AMS 6490) 0.2 max 0.02 001 -0.02
Mg - WE 43 0.4 max 0.01 -0.04 0.O0
TI (Alloy 8% Mg) 0.2 max -0.01 0.00 0.00
Incone1718 0.2 max 0.00 0.01 0.00
Fluid Appearance Report dark brown dark brown darkbrown
Corrosion-Oxidation Stability, 10 L/hr air flow

Temperature, 232~ 48 hr
Viscosity Change, (%) -5 to 25 18 23.3 40.10
Acid Number Change (rag KOH/gm) 4.0 max 1.6 3.83 8
% Fluid loss 8 0 max 4 3.75 4.8
Metal weight Change (rng/cm2)
Al (Alloy 2024) 0.2 max 0 -0.01 0.01
Ag (99.9% pure) 0.2 max -0.03 -0.06 -0.08
Bz ( Si-Fe, AMS 4616, AZ31B) 0.2 max -0.03 -0.05 -0.01
Steel (grade 1010) 0.2 max 0.02 0.02 0.02
M-50 (AMS 64fiO) 0.2 max 0.00 0.02 0.02
TI (Alloy 8% Mg) 0.2 max 0.00 0.00 0.01
Incone1718 0.2 max -0.02 0.00 0.01
Fluid Appearance Report dark brown dark brown dark brown
weight of two of the better antiwear additives, both produced by the same company, in
otherwise the same formulation. With these, the largest difference is in the -40~
viscosity and in the post test corrosion-oxidation test acid number change.
Co/ang
Engine coke involves complicated variables such as airflow, temperature,

resonance time, chemistry of the base oil and additives, fluid film thickness, and
whether the GTO is in the vapor phase or liquid phase. Coking tendency of GTOs can
be different for different engines, i.e., a GTO may be better for coking tendency in one
engine than another engine, while a different GTO may be worse for coking in the first
engine and better in the second engine.
As a screening tool, we are using a static coking test [3]. The static coker consists
ofa coker base mounted on a hot plate, a 0.005-inch thick, 2-inch diameter steel
specimen, and a programmable temperature controller with the necessary
thermocouples for monitoring and controlling temperature. Approximately 0.5 gm of
the oil sample is loaded at the 150~C presoak temperature. The temperature is held at
150~ for 30 minutes, then ramped at 10~ per minute to the 300~ test temperature
and held for 3 hours. After cooling, the amount of deposit is weighed and reported as
mg coke/g oil This test is extremely sensitive to the volatility (and therefore, the
viscosity) of the base fluid: the higher the volatility, the lower the coking value. Only
22 TURBINE LUBRICATION IN THE 21 ST CENTURY
Table3-FluidTargetProperaesforLaterTesting
PROPERTY REQUIRMENT METHOD SOURCE B
Foaming Characteristics(static) FTM 791c

Foam volume, ml 100 max method 3213
Foam collapse time, sec 60 max
Foaming Characteristics (dynamic) FTM 791c

Foam vol, ml / collapse time, sec method 3214
80% (176~ at 1000 ml/min 100 max/60 max
80% (176~ at 1500 ml/min 150 max/60 max
80% (176~ at 2000 ml/min 200 max/60 max
110~ (176~ at 1000 ml/min 100 max/60 max
110% (176~ at 1500 ml/min 150 max/60 max
110% (176~ at2000 ml/min 200 max/60 max
Elastomer Compatibility, % swell ASTM D 471

AMS 3383 25 max 17.16
MIL-R-83248 25 max 15.718.26
MIL-R-83485 25 max 12.96
Dynamic Coking (to be determined) (to be determined)
Bearing Deposition, Oil sump, 220~ Bearing 275~ FTM 791c

Deposit Rating 45 max method 3450
Filter Deposit weight (gm) 1.0 max
Oil consumption (ml) 960 max
Viscosity change at 400C -5 to +15
Acid Number change (mg KOH/gm) 1.0 max
Metal Weight Change (mg/cm2) 0.2 max
(metals as in MIL-L-7808K)
Mini-Engine Simulator (to be determined) (to be determined)
Table 4 - Effect of Anitwear Phosphate Additive on Formulation Properties

(2.5% by weight antiwear additive in same base formulation)
Oil Additive 1 2
Wear scar, mm
52100, 75~ 0.45 0.62
M-50,200~ 0.42 0.38
Viscosity, 40~ 36,390 40,660
Static coke 31.8 38
Corrosion-Oxidation
(232~
Visc. Chg., 40~ % 5.6 7.2
Acid No. Chg., mg KOH/gm 1.36 -0.12
oils of similar viscosity (and presumably volatility) should be compared in coke

formation. The static coking value of the original MIL-L-27502 was 49.4 mg coke/g
oil. This static coking value was selected as the target for the new oils since that oil
passed a full-scale engine test with remarkably clean parts. All of the candidate GTOs
(Table 1) meet this requirement.
Since low coke is one of the primary goals of this GTO development program,
further testing will be conducted. These tests include an in-house developed dynamic
coking test, mini-turbine engine lubricant simulator test, a bearing deposition test and a
hot-tube deposition test at Pratt & Whitney. The hot-tube deposition test also
investigates vapor phase coke, besides the liquid phase coke investigated by the static
coke test.
Corrosion-Oxidation
The corrosion-oxidation test (ASTM D 4636) is called for in the military GTO
specifications. We have seen GTOs that pass the 232~ corrosion-oxidation test, but,
contrary to what one might expect, gave very high static coke results. We do know that
the performance improving additives can have a negative effect on the corrosion-
oxidation test, as with the coke test. Currently we are conducting the corrosion-
oxidation test at both 220~ and 232~ In the 220~ test, we use a corrosion-resistant
magnesium (Mg) alloy, WE-43, to represent a magnesium alloy in a modern or future
gas turbine engine. Since the Mg alloy is not used in the hottest engine sections, its
testing was limited to the 220~ corrosion and oxidation test. Mg alloys, even the
relatively stable WE-43, tend to be reactive, as we learned when some of them actually
dissolved completely in the tests. (Obviously, formulations that dissolved the Mg
coupons required more modification/improvement.) The 232~ test contains the same
coupons as the 220~ test except without the Mg alloy, as the remaining metal alloys
will likely be exposed to higher temperatures.
Sources A and B GTO candidates now meet the corrosion-oxidation test target
limits. It was, however, very challenging to meet these limits, taking many formulation
modifications, in some cases. The Source C GTO candidate is over the limit in post-test
acid number, but may be fine-tuned to meet the limit in the future.
To pass the corrosion-oxidation test, the optimum blend of base fluids and additives
must be selected without causing the low temperature flow properties to become too
viscous. Antiwear and antioxidant additives can also have a negative effect in this test,
so must be selected carefully and used in not too large of a loading. The antioxidant
additive may cause excessive coking in the coke test and also in the further coke-related
tests, to be conducted in the near future.
Summary
The US Air Force is progressing to meet target goals for an advanced high-
temperature, low-coke GTO for current higher temperature and future engines.
Significant progress has been made towards the goals established Current and future
high-temperature engines will require less maintenance using the new GTO
Acknowledgments
The authors sincerely thank Mr. Timothy Reid and Ms. Mary Borchers of
University of Dayton Research Institute for much of the test results presented here; Ms.
Susan Brown of Pratt & Whitney, A United Technologies Company, for her guidance
and participation; and the companies for supplying candidate fluids and additives.
References
[1] Gunderson, R. C. and Hart, A. W., SyntheticLubricants, Reinhold Publishing

Corp., New York, 1962.
[2] Schrand, J. B., "Turbine Engine Evaluation of an Experimental MIL-L-27502

Lubricant Code 0-77-20," AFWAL-TM-81-22 (POSL), April 1981.
13] Smith, H. A. and Saba, C. S., "Lubricant Deposition Studies Using a Small Test
Volume, Low Cost Static Coker," LubricationEngineering, ~IV, 12, 1988 pp.
983 -992.
David C. Kramer, I Brent K. Lok,2 and Russ R. Krug3
The Evolution of Base Oil Technology
Reference: Kramer, D. C., Lok, B. K., Krug, R. R., "The Evolution of Base Oil
Technology," Turbine Lubrication in the 21st Century, ASTM STP #1407, W. R.
Herguth and T. M. Warne, Eds., American Society for Testing and Materials, West
Conshohocken, PA, 2001.
Abstract: From its humble beginnings over 3000 years ago, lubrication technology has
seen many phases of evolution. As it continues to evolve at an ever-increasing rate, base
oil performance is making a larger contribution to finished lubricant performance.
Turbine oils are perhaps the most dramatic example because they typically contain over
99% base oil.
Early lubrication began with animal fats and oils and slowly evolved to
petroleum-based oils. Many generations of refining processes have since improved on
Mother Nature. Early processes such as acid treating and solvent extraction improved
the quality of base oils by removing some or most of the worst molecules from the oil.
Later processes like hydrotreating, catalytic hydrocracking, catalytic dewaxing, and
modem wax hydroisomerization transformed feed molecules into molecules with
improved lubricating qualities.
Modem wax hydroisomerization, in particular, makes base oils with very low
impurities and typically water-white appearance. Now, about one third of all base oils
manufactured in North America are of such high quality.
Looking to the future, the trend is toward even higher base oil purity, higher
viscosity index (V.I.), lower volatility, and longer life. The distinction between heavily
processed mineral oils and traditional "synthetic" oils will continue to blur.
The evolution and future of base oil technology will be discussed in more detail
in this paper.
Keywords: base oil, evolution, history, hydrocracking, solvent
ISenior Staff Engineer - Base Oil Technology, Chevron Global Lubricants, 100 Chevron
Way, Richmond, CA 94802.
2Senior Product Manager - Base Oils, Chevron Global Lubricants, 6001 Bollinger
Canyon Rd, San Ramon, CA 94583.
3Manager - Base Oil Technology, Chevron Global Lubricants, 100 Chevron Way,
Richmond, CA 94802.
25

Early Base Oil Processing
Base oil technology has undergone many phases of evolution. In the first phase,
animal fats were used as lubricants. Ancient inscriptions dating back to 1400 BC show
beef and mutton fat (tallow) being applied to chariot axles. Very little changed over the
next 3000 years except that the oils sometimes came from more exotic animals such as
whales.
In 1852 petroleum-based oils first became available. They were not widely
accepted at first because they did not perform as well as many of the animal-based
products. Raw crude did not make very good lubricant. The base oil industry was on
the very steep part of the learning curve.
But as the demand for automobiles grew, so did the demand for better lubricants.
Soon lubricant manufacturers learned which crudes made the best lubricants. Then they
improved on Mother Nature by refining the crude into narrow distillation cuts with
varying viscosity. By 1923 the Society of Automotive Engineers classified engine oils
by viscosity: light, medium, and heavy. Engine oils contained no additives and had to
be replaced every 800-1000 miles.
In the 1920s more lubrication manufacturers started "processing" their base oils
to improve their performance. Three popular processing routes were:
Clay Treating
Clay similar to kitty litter was used to soak up and remove some of the worst
components in the petroleum base oil. These compounds were usually aromatic and
highly polar compounds containing sulfur and nitrogen.
Acid Treating
Concentrated sulfuric acid was used to react with the worst components in the
base oil and convert them into a sludge that could be removed. Although this process
effectively cleaned up the oil, it was expensive. This technology has virtually
disappeared from North America due to environmental concerns about the acid and
sludge [1].
S02 Treating
SO2 treating was a primitive extraction process to remove the worst components
in the lube oil using a recyclable solvent. Unfortunately, the solvent was highly toxic.
Although it also has been virtually phased out [1], it was a useful stepping stone to
conventional solvent extraction.
Solvent Refining
By approximately 1930 solvent processing emerged as a viable technology for

improving base oil performance using a fairly safe, recyclable solvent. Most base oil
producers in the world today still use this process.
KRAMER ET AL. ON BASE OIL 27
Approximately two-thirds of the base oil in North America is currently

manufactured using this route. Solvent refined base oils are commonly called Group I
base oils which are characterized as those having less than 90% saturates (>10%
aromatics) and more than 300 ppm sulfur. Table 1 shows all the base oil groups as
defined by the American Petroleum Institute (API) Publication 1509.
Table 1 - API Base Stock Categories (API Publication 1509)
Group Sulfur, Wt % Saturates V.I.

I >0.03 and/or <90 80-119
II _<0.03 and >90 80-119
III _<0.03 and >90 >120
IV All Polyalphaolefins (PAOs)
V All Stocks Not Included in Groups I-IV (Pale Oils and Non-PAO
Synthetics)
The solvents and hardware used to manufacture solvent-refined base oils have
evolved over time, but the basic strategy has not changed since 1930. The two main
processing steps are:
1. Remove aromatics by solvent extraction.
2. Remove wax by chilling and precipitation in the presence of a different solvent.
Aromatics are removed by solvent extraction to improve the lubricating quality
of the oil. Aromatics make good solvents but they make poor quality base oils because
they are among the most reactive components in the natural lube boiling range.
Oxidation of aromatics can start a chain reaction that can dramatically shorten the useful
life of a base oil.
The viscosity of aromatic components in a base oil also responds relatively
poorly to changes in temperature. Lubricants are often designed to provide a viscosity
that is low enough for good cold weather starting and high enough to provide adequate
film thickness and lubricity in hot, high-severity service. Therefore, when hot and cold
performance is required, a small response to changes in temperature is desired. The
lubricants industry expresses this response as the viscosity index (V.I.). A higher V.I.
indicates a smaller, more favorable response to temperature. Correspondingly, many
turbine manufacturers have a minimum V.I. specification for their turbine oils. Base oil
selection is key for meeting this specification because turbine oil additives do not
normally contribute positively to the V.I. in turbine oil formulations.
Aromatics are removed by feeding the raw lube distillate (vacuum gas oil) into a
solvent extractor where it is contacted countercurrently with a solvent. Popular choices
of solvent are furfural, n-methyl pyrrolidone (NMP), and DUO-SOL TM. Phenol was
another popular solvent but it is rarely used today due to environmental concerns.
Solvent extraction typically removes 50-80% of the impurities (aromatics, polars, sulfur
and nitrogen containing species). The resulting product of solvent extraction is usually
referred to as a raffinate.
The second step is solvent dewaxing. Wax is removed from the oil to keep it
from freezing in the customer's sump or crankcase at low temperatures.
Wax is removed by first diluting the raffinate with a solvent to lower its viscosity
to improve low-temperature filterability. Popular dewaxing solvents are methyl-ethyl
ketone (MEK)/toluene, MEK/methyl-isobutyl ketone, or (rarely) propane. The diluted
oil is then chilled to -10 to -20~ Wax crystals form, precipitate, and are removed by
filtration.
Additives Improve Performance
Over the next several decades, the solvent refining process did not change very
much. Finished oil quality improved mainly due to the addition of additives. Additives
began tobe widely used in 1947 when the API began to categorize engine oils by
severity of service: regular, premium, and heavy duty. Additives were used to extend
the life only in premium and heavy-duty oils.
Regular engine oils were unadditized and were intended for 80-100 hours of low-
severity service. Premium engine oils were additized to extend the life of passenger car
engine oils. Heavy duty engine oils were more heavily additized to extend the engine oil
life in more severe applications such as commercial trucks and construction equipment.
In 1950 multigrade oils were first introduced which were additized with
polymers to enhance the V.I. of the oil which improved the hot and cold performance of
the oil.
For several more decades, the lubricants industry continued to rely heavily on
additive technology to improve the performance of finished oils. Lubricant quality
improved significantly only when the additive chemistry improved. This was the only
viable strategy until a significant improvement in base oil technology was available.
Hydrotreating
Hydrotreating was developed in the 1950s and first used in base oil
manufacturing in the 1960s by Amoco and others. It was used as an additional
"cleanup" step added to the end of a conventional solvent refining process.
Hydrotreating is a process for adding hydrogen to the base oil at elevated temperatures
in the presence of catalyst to stabilize the most reactive components in the base oil,
improve color, and increase the useful life of the base oil. This process removed some
of the nitrogen and sulfur containing molecules but was not severe enough to remove a
significant amount of aromatic molecules. Hydrotreating was a small improvement in
base oil technology that would become more important later.
Hydrocracking
Hydrocracking is a more severe form of hydroprocessing. It is done by adding

hydrogen to the base oil feed at even higher temperatures and pressures than simple
hydrotreating. Feed molecules are reshaped and often cracked into smaller molecules.
A great majority of the sulfur, nitrogen, and aromatics are removed. Molecular
reshaping of the remaining saturated species occurs as naphthenic rings are opened and
paraffin isomers are redistributed, driven by thermodynamics with reaction rates
facilitated by catalysts. Clean fuels are byproducts of this process.
A primitive version of the hydrocracking process was attempted for lube oil
manufacturing in the 1930s but was soon abandoned for economic reasons after the
solvent refining process was commercialized. But hydroeraeking technology continued
to improve [2].
After WWII, predecessors to modem hydrocracking catalyst technology were
imported from Germany. Chevron commercialized this technology for fuels production
in the late 1950's [3]. In 1969 the first hydrocracker for Base Oil Manufacturing was
commercialized in Idemitsu Kosan Company's Chiba Refinery using technology
licensed by Gulf [4]. This was followed by Sun Oil Company's Yabucoa Refinery in
Puerto Rico in 1971, also using Gulf technology [2].
Catalytic Dewaxing and Wax Hydroisomerization
The first catalytic dewaxing and wax hydroisomerization technologies were

commercialized in the 1970s. Shell used wax hydroisomerization technology coupled
with solvent dewaxing to manufacture extra high V.I. base oils in Europe. Exxon and
others built similar plants in the 1990s. In the U.S., Mobil used catalytic dewaxing in
place of solvent dewaxing, but still coupled it with solvent extraction to manufacture
conventional neutral oils. Catalytic dewaxing was a desirable alternative to solvent
dewaxing especially for conventional neutral oils, because it removed n-paraffins and
waxy side chains from other molecules by catalytically cracking them into smaller
molecules. This process lowered the pour point of the base oil so that it flowed at low
temperatures, like solvent dewaxed oils. Hydroisomerization also saturated the majority
of remaining aromatics and removed the majority of remaining sulfur and nitrogen
species.
Chevron was the first to combine catalytic dewaxing with hydrocracking and
hydrofinishing in their Richmond, California, base oil plant in 1984 [5] (Figure 1). This
was the first commercial demonstration of an all-hydroprocessing route for lube base oil
manufacturing.
In 1993, the first modem wax hydroisomerization process was commercialized
by Chevron [6]. This was an improvement over earlier catalytic dewaxing because the
pour point of the base oil was lowered by isomerizing (reshaping) the n-paraffins and
other molecules with waxy side chains into very desirable branched compounds with
superior lubricating qualities rather than cracking them away. Hydroisomerization was
also an improvement over earlier wax hydroisomerization technology, because it
eliminated the subsequent solvent dewaxing step, which was a requirement for earlier
generation wax isomerization technologies to achieve adequate yield at standard pour
points. Modem wax hydroisomerization makes products with exceptional purity and
stability due to extremely high degree of saturation. They are very distinctive because,
unlike other base oils, they typically have no color.
By combining three catalytic hydroprocessing steps (hydrocracking,
hydroisomerization, hydrotreating), molecules with poor lubricating qualities are
transformed and reshaped into higher quality base oil molecules. Pour point, V.I., and
oxidation stability are controlled independently. All three steps convert undesirable
molecules into desirable ones, rather than have one, two, or all three steps rely on
Figure I - Richmond Lube Oil Project
subtraction. Among the many benefits of this combination of processes is greater crude
oil flexibility; that is, less reliance on a narrow range of crude oils from which to make
high-quality base oils. In addition, the base oil performance is exceptionally favorable
and substantially independent of crude source, unlike solvent-refined base oil.
G r o u p II - M o d e r n C o n v e n t i o n a l Base Oils
Lube base oils made by hydrocracking and early wax isomerization technologies
showed favorably differentiated performance which prompted the API to categorize base
oils by composition (API Publication 1509) in 1993, as shown earlier in Table 1.
Table 1 shows that Group II base oils are differentiated from Group I base oils because
they contain significantly lower levels of impurities (<10% aromatics, <300 ppm S).
They also look different. Group II oils made using modem hydroisomerization
technology are so pure that they have almost no color at all. From a performance
standpoint, improved purity means that the base oil and the additives in the finished
product can last much longer. More specifically, the oil is more inert and forms less
oxidation byproducts that increase base oil viscosity and react with additives.
The modem hydroisomerization process licensed by Chevron under the name
ISODEWAXING gained acceptance rapidly since its introduction in 1993. In fact,
about one-third of all base oils manufactured in North America are now manufactured
using this process (Figure 2). A similar trend can be seen in the rest of the world as well.
Figure 2 - Greater Availability (Chevron + Licensees = 42% by 2001)
More recently, Mobil (ExxonMobil) added to this trend by commercializing an

all-hydroprocessed route for Group II production {Mobil Selective Dewaxing
(MSDWrM)} which they installed in their Jurong Refinery in Singapore in 1997. Exxon-
Mobil also added a hydroprocessing step (Raffinate Hydroconversion or RHC) to a
solvent refining plant in Baytown, Texas, to upgrade its capability to Group II.
Group II base oil technology, along with specially designed additives has already
had a tremendous impact on finished oil performance. In some applications including
turbine oils, lubricating oils formulated with Group II base oils can outlive expensive
traditional "synthetic" oils made from PAO.
G r o u p III - U n c o n v e n t i o n a l Base Oils
Table 1 shows that the API defines the difference between Group II and III base
oils only in terms of the V.I. Base oils with a "conventional" V.I. (80-119) are Group II
and base oils with an "unconventional" V.I. (120+) are Group III. Group III oils are also
sometimes called unconventional base oils (UCBOs) or very high V.I. (VHVI) base oils.
Solvent-dewaxed Group III base oils have been produced in Europe for more
than 10 years, primarily by Shell and BP [7]. Group III base oils were not widely
available in North America until a few years ago when Chevron and its licensees began
making them using the all-hydroprocessing route. These modem Group III oils have
greatly improved oxidation stability and low temperature performance relative to the
solvent-refined Group III oils. Consequently, many of these older plants are now being
upgraded to enable them to make the modem hydroisomerized Group III oils [8].
From a processing standpoint, modem Group III base oils are manufactured by
essentially the same processing route as modem Group II base oils. Higher V.I. is
achieved by increasing the temperature or time in the hydrocracker. This is sometimes
collectively referred to as the "severity." Alternatively, the product V.I. could be
increased simply by increasing the feed V.I., which is typically done by selecting the
appropriate crude.
Group III base oils are now widely available in North America because they can
be manufactured in large quantities by most of the companies that currently make
32 TURBINELUBRICATIONIN THE 21 sT CENTURY
Group II oils. Many of these companies have started adding them to their "synthetic"
product lines.
Modem Group III base oils have properties which allow them to perform at a
level that is significantly higher than "conventional" Group I and Group II base oils, and
they substantially match existing levels of performance in finished lube applications
already established by traditional synthetic oils. The most notable exception is arctic
oils, which have extremely low pour point requirements and a very small market
presence.
Group IV - Traditional "Synthetic" Base Oils (PAO)
The use of the word "synthetic" in the lubricants industry has historically been
synonymous with polymerized base oils such as poly-alpha olefins (PAOs), which are
made from small molecules. The term "synthetic" was given a special meaning by the
lubricants industry because these types of oils were the only components available for
high-performance lubricants at that time. Later in this paper it should become apparent
how PAOs directly influenced the evolution of mineral-based base oil technology which
in turn has influenced the evolution of the term "synthetic."
The first commercially viable process for making PAO was pioneered by Gulf
Oil in 1951 using an A1C13catalyst. Mobil patented an improved process using a
BF3/A1CI3 catalyst in the 1960s. Mobil first used this new base oil in specialty products
such as Mobilgrease 28, which solved a wheel beating failure problem on aircraft
carders in cold climates [9].
In the 1970s PAOs became a major consumer-sought lubricant component when
Mobil Oil commercially marketed its Mobil 1TMengine oils 25 years ago. For the next
15 years, the PAO market traveled a long and winding road battling a slow, steady
growth and criticisms of justification for the higher cost compared to conventional oils.
In the last 10 years, the PAO market significantly increased, first in Europe and then in
North America experiencing periods of double-digit growth. In part, the growth might
be attributed to the stricter lubricant specifications in Europe that created a market niche
for synthetics and semi-synthetic products [9].
As the lucrative PAO market grew, some base oil manufacturers began using
higher-V.I, feedstocks (usually byproducts from wax manufacturing) to make mineral
oils with V.I.s that matched the PAOs. These new Group III oils were not manufactured
from small molecules like traditional synthetics but they bridged most of the
performance gap at a lower cost. Therefore, some lubricant manufacturers, primarily in
Europe, began replacing PAOs with these newly available Group III base oils in their
"synthetic" engine oils. This created a controversy in the lubricants industry as some
synthetic base oil producers and lubricant manufacturers believed that polymerized base
oils were the only true synthetics.
The trend toward globalized lubricant specifications and worldwide OEM
specifications is now creating more demand for Group III base oils in North America
and other continents as well. This is particularly true in North America due to the recent
ruling by the National Advertising Department of the Better Business Bureau that
Group III base oils can be considered "synthetic" and because modem Group III base
oils, made using hydroisomerization technology, have most of the attractive performance
features of early synthetics.
Group III Versus PAO Performance
Historically, PAOs have had superior lubricating performance characteristics

such as V.I., pour point, volatility, and oxidation stability that could not be achieved with
conventional mineral oils. Now, in modem base oil manufacturing, V.I., pour point,
volatility, and oxidation stability can be independently controlled. Modem Group III
oils today can be designed and manufactured so that their performance closely matches
PAOs in most commercially significant finished lube applications.
As well-designed Group III base oils become abundant in the marketplace, the
performance gap between Group III and PAO (Group IV) is closing. Here are some key
examples:
P o u r P o i n t - Pour point is the one property where Group III oils allegedly fall
short of PAO. While it is certainly true that the pour point of the neat Group III base oil
is substantially higher than that of a PAO of comparable viscosity, it is important to
understand that the pour point of the fully formulated lubricant (base oils plus additives)
is the critical property. Base oils manufactured with modem isomerization catalysts
respond very well to pour point depressant additives. For example, turbine oils
formulated with conventional Group II base oils (-12~ base oil pour point) are available
with a formulated pour point of-36~ Fully formulated Group III based lubricants can
be made with pour points of-50~ or below.
Products such as motor oils made with the lighter-grade PAOs, on the other hand,
typically have h i g h e r pour points than the base fluid, so the gap in final product pour
point between PAO-based and UCBO-based lubricants is much smaller than in the base
fluids themselves. Moreover, it is entirely possible with modem Group III
manufacturing technology to produce base oils of even lower pour point. However, this
is not common practice in the industry, because it is more economical to meet finished
lube low temperature performance using pour point depressant additives rather than
using special Group III oils having exceptionally low pour points.
C o l d C r a n k S i m u l a t o r - Viscosity in engine journal beatings during cold
temperature startup is a key factor in determining the lowest temperature at which an
engine will start. Cold Cranking Simulator (CCS) viscosity, as measured by ASTM
Method D 5293, is determined under conditions similar to those experienced in engine
bearings during starting. For base oils, this viscosity is determined almost entirely by
viscosity and V.I. Since Group III stocks typically have V.I. comparable to that of 4 cSt
PAO, one would expect comparable CCS performance. This is demonstrated in Figure
3, where it can be seen that a 4 cSt Group III base oil, with a kinematic viscosity of 4.2
cSt at 100~ and a V.I. of 129, and PAO 4, with a viscosity of 3.9 cSt and V.I. of 123,
have similar CCS values, both about half that ofa 4 cSt Group II base stock of about 100
V.I. This performance makes the Group III stock very effective for formulating fuel-
efficient multi-viscosity engine oils in the 0W-20 to 0W-50 range, one that has
historically been achieved only with PAO-based product.
34 TURBINELUBRICATION IN THE 21st CENTURY
Figure 3 - Cold Cranking Performance, Group 111 Comparable to PAO
Noack Volatility -Noack volatility of an engine oil, as measured by ASTM

D 5800 and similar methods, has been found to correlate with oil consumption in
passenger car engines. Strict requirements for low volatility are important aspects of
several recent and upcoming engine oil specifications, such as ACEA A-3 and B-3 in
Europe and ILSAC GF-3 in North America. Figure 4 shows that from a blender's
perspective, Group III base oils are similarly effective as PAOs for achieving these low
volatility requirements in engine oil applications. The V.I. of modern Group III oils
typically match or exceed PAO, so they can match the volatility of PAOs at a reasonable
distillation cut width.
Figure 4 - Group lll Performance Versus PAO, Comparable Noack Volatility
Oxidation Stability - Oxidation and thermal stability are among the most
important advantages that "synthetics" bring to the table. Better base oil stability means
better additive stability and longer life. High stability is the key to making the premium-
quality finished oils of the future with longer drain intervals. Here Group III oils
routinely challenge PAO performance.
The stability of modem Group III stocks depends mostly on their V.I., because
V.I. is an indication of the fraction of highly stable isoparaffinic structures in the base oil
[10]. However, because modem Group III stocks also undergo additional severe
hydrofinishing after hydrocracking and hydroisomerization, they achieve an additional
boost in stability because only trace amounts of aromatics and other impurities remain in
the finished stocks. On the other hand, PAO performance seems to depend largely on
residual olefin content. Olefins are an intermediate in PAO production that contribute to
instability.
Figure 5 illustrates that base oil quality can have a big impact on the oxidation
stability in turbine oils. The Turbine Oil Stability Test (TOST), or ASTM D 943,
measures the time required for a turbine oil to oxidize to the point where the total acid
number reaches 2.0 mg KOH/g. Unadditized Group I base oil fails in about 200 hours.
A modem high-quality turbine oil formulated with Group I base oil typically fails in less
than 7000 hours. A high-quality Group II formulated oil can run more than twice as
long before it fails.
Figure 5 - Higher Base Oil Quality Extends Turbine Oil Life
The benefit of all-hydroprocessed Group III base oils in oxidation stability is

illustrated in Figure 6 for hydraulic oils formulated by using the same additive system in
four different base oils. Here, the time required to reach an acid number of 2.0 (defined
by neutralization of 2.0 mg of KOH/g ofoil) in the Universal Oxidation Test (ASTM
D 4871), a common measure of oil oxidation, was substantially longer for the Group III
formulation than for either the Group I or II products. Moreover, the performance of the
Group III product was essentially the same as that for the oil formulated with PAO.
36 TURBINELUBRICATION IN THE 21sr CENTURY
Figure 6 - Oxidation Stability, Acid Number in Hydraulic Oils
Table 2 lists a variety of North American lubricants which are based upon all-
hydroprocessed Group III base stocks. These products include engine oils, industrial
oils, and driveline fluids, and are targeted at the same performance levels achieved by
traditional synthetic formulations.
Table 2 - Synthetic Quality Products Utilizing All-Hydroprocessed

Group 111Base Stocks
Available Now Upcoming

Semi and Full Synthetic PCMO GF-3 PCMO (Semi and Full Synthetic)
Semi-Synthetic HDMO Extended Drain Gear Oil
DaimlerChrysler ATF+4| High Performance Automotive (Racing) Oils
Ford Mercon| V ATF Motor Oils
Compressor Oil Gear and Transmission Oils
Future Evolution
Looking to the future, the trend is toward lubricants and base oils with even
higher purity, lower volatility, and longer life. The molecular structure of base oils will
probably look even more like PAO as they become more concentrated in the most
favorable molecular species needed for superior lubrication performance. It is likely that
recent and ongoing developments in base oil technology will enable lubricants with
exceptional performance to be marketed in much greater volumes than was feasible
when PAO was the only stock capable of such performance levels.
There are many possible routes for improving base oil quality. Continued
evolution of the all-hydroprocessing route is one likely possibility. Selectivity toward
desired molecular compositions could be improved by improving the catalysts and the
processing technology. Improving the feedstock can also improve the product. Very
paraffinie (waxy) feedstocks such as Fischer-Tropsch wax from natural gas-to-liquids
plants can potentially be further processed into high quality base oils. Volumes and
applications are expected to grow, as ultra-waxy feedstocks become more widely
available.
Other competing technologies are likely to emerge. New routes for
manufacturing PAOs have been proposed that use cheaper feedstocks such as ethylene
and propylene rather than 1-decene [11].
Future improvements in base oil technology will assuredly lead to further
improvements in the performance of turbine oils and other sensitive applications with
low additive treat rates.
Conclusions
Lubrication technology evolved slowly from ancient times until the middle of the
20 th century. Then solvent refining technology emerged and displaced naturally
occurring petroleum distillates due to its improved refined properties. Starting in the
1960s, hydroprocessing technologies were introduced which improved base oil purity
and performance further. In the 1970s and 1980s, Group II base oils were manufactured
and recognized as a separate API category in 1993, due to their positive differentiation
over conventional stocks. Modem hydroisomerization technologies, such as
ISODEWAXING, became widely accepted and grew rapidly since it was first
commercialized in 1993. Widespread licensing of this technology has created an
abundant supply of Group II oils that have exceptional stability and low temperature
performance relative to their Group I and Group II predecessors. This technology is now
used to make about one-third of all base oils in North America.
A similar trend appears to be emerging with Group III base oils, especially those
made using modem hydroisomerization. They offer most of the performance advantages
of traditional PAO-based "synthetic" oils and can be manufactured in volumes
unachievable by PAO. Most manufacturers of modem Group II base oils can make
modem Group III base oils as well.
Selected top-tier lubricants requiring PAO should continue to coexist with
Group III oils as they have for years in Europe. But widespread availability of modem
Group II and III mineral oils is accelerating the rate of change in the finished oil
markets. New improved base oils are helping the engine and equipment manufacturers
meet increasing demands for better, cleaner lubricants. This is particularly true for
turbine oils because turbine oils usually contain more than 99% base oil. Turbine oils
made from hydroisomerized Group II base oils and the appropriate additives have
demonstrated significantly longer TOST lives than turbine oils made with Group I base
oils. In fact, they commonly outperform the traditional "synthetics" made with PAO.
As base oil technology continues to evolve and improve, consumers will enjoy
even greater protection of automobiles, trucks, and expensive machinery such as
turbines. Lubrication performance that currently can be achieved only in small-volume
niche applications, using PAO and other specialty stocks, will be more widely available
using the new generation of Group II and Group III oils.
References
[1] National Petroleum Refiners Association, "1999 Lubricating Oil and Wax
Capacities of Refiners and Re-Refiners in the Western Hemisphere," January
1999.
[2] Sequeira, A., Jr., "Lubricant Base Oil and Wax Processing," Marcel Dekker, Inc.,
Chemical Industries Series, August 1994.
[3] Stormont, D. H., "New Process Has Big Possibilities," The Oil and Gas Journal,
57, 44, 1959, pp. 48-49.
[4] Company release by Idemitsu Kosan Co, Imperial Chemical LTD, Shell
Development Co, "First Lubricant-Oil Cracker," Oil and Gas Journal, June 12,
1972.
[5] Zakarian, J. A., Robson, R. J., and Farrell, T. R., "All-Hydroprocessing Route for
High-Viscosity Index Lubes," Energy Progress, 7, 1, 1987, pp. 59-64.
[6] Wilson, M. W., Mueller, T. A., and Kraft, G. W., "Commercialization of
Isodewaxing - A New Technology for Dewaxing to Manufacture High Quality
Lube Base Stocks," FL-94-112, NPRA, November 1994.
[7] Min, P. Y., "VHVI Base Oils: Supply and Demand," 4th Annual Fuels and Lubes
Conference, Singapore, January 1998.
[8] Howell, R. L., "Hydroprocessing Routes to Improved Base Oil Quality and
Refining Economics," 6th Annual Fuels and Lubes Conference, Singapore,
January 2000.
[9] Bui, K., "Synthetics II," Lubricants World, November 1999.
[10] Kramer, D. C., Ziemer, J. N., Cheng, M. T., Fry, C. E., Reynolds, R. N., Lok, B. K.,
Krug, R. R., and Sztenderowicz, M. L. "Influence of Group II and III Base Oil
Composition on V.I and Oxidation Stability," AIChE Spring Meeting, March
1999.
[11] Heilman, W. J., Chiu, I. C., and Chien, J. C. W., "New Polyalphaolefin Base Oil,"
American Chemical Society Meeting, New Orleans, August 1999.
Steven T. Swift 1, K. David Butler2, and Werner Dewalaa
Turbine Oil Quality and Field Application Requirements
Reference: Swift, S. T., Butler, K. D., and Dewald, W., "Turbine Oil Quality and
Field Application Requirements,"Turbine Lubrication in the 21 u Century, ASTM STP
1407, W. R. Herguth and T. M. Warne, Eds., American Society for Testing and
Materials, West Conshohocken, PA, 2001.
Abstract: Power generation and other types of industrial turbines have always required
the use of high quality lubricants. This paper will discuss the evolution of turbine oil
formulation technology, and the use of different quality products in different applications.
The original "R&O" type circulating oils, which are suitable for use in many hydraulic
and steam turbine systems, have a long and successful history. Next came high
performance products, designed for increased operating life. Finally, products have been
introduced recently that provide the high stability needed for systems which operate
under severe conditions of thermal stress, such as may occur in certain gas turbine
operations.
Selecting a suitable turbine oil requires meeting the application requirements of
equipment and service environment, while balancing initial and life cycle cost
considerations and oil performance history. Field experience with different types of
equipment and different turbine oil types will be described, with examples of situations
where each oil type has performed successfully. Experience with field performance
problem-solving will also be addressed, such as a gas turbine filter plugging problem
which was resolved by selecting the appropriate turbine lubricant.
Turbine equipment manufacturers generally specify the properties and performance
requirements of turbine oils according to the type and severity of operation. Standardized
laboratory tests provide an indicator of an oil's performance, although such performance
needs to be confirmed by field experience. New testing tools are needed which will
better correlate to the performance requirements of new generation equipment.
Keywords: turbine oil, steam turbine, combustion turbine, oxidation, antioxidant,

testing, oil service life, field application, turbine severity, cyclic operation, deposits
Industrial turbines, commonly used for applications such as power generation and
natural gas transmission, have always required the use of high quality lubricants. These
are manufactured using a blend of highly refined or converted petroleum base oils of
i ExxonMobil Lubricants and Petroleum Specialties Company, Fairfax, VA 22037.

2 Imperial Oil Research Centre, Sarnia, Ontario, Canada N7T 8C8.
3 Imperial Oil Limited, Toronto, Ontario, Canada M5W IK3.
39

appropriate viscosity - generally of ISO 32 or 46 grades - with chemical additives that

control properties of the oil which are important to the successful operation of the turbine
equipment. This paper discusses the evolution of turbine oil formulation technology, and
the use of different quality products in very diverse end uses.
Uninhibited refined petroleum oils were used until the 1940's when they were
upgraded to the "R&O" type circulating oils by the addition of chemical additives. These
are suitable for use in many hydraulic and steam turbine systems, and have a long and
successful history. Such products are still available, but are now mostly used as general
purpose circulating oils. Next came higher performance products (herein referred to as
"Conventional turbine oils"), designed for longer operating life, and these started to
become available in the 1970's and are still widely used in hydraulic, steam and low
severity combustion turbines. Finally, products ("Next-Generation turbine oils") were
introduced in the late 1980"s that provide the high stability needed for turbine systems
operating under severe conditions of thermal stress, such as may occur in high efficiency
gas turbines. Land-based aircraft turbine engines are also used sometimes for industrial
applications, but their very high operating temperatures mandate the use of highly stable
and very expensive synthetic lubricants. In reality, these changes evolved continuously in
parallel with the evolution of turbine equipment. The historic data shown in Table 1
represent quality levels that are only broadly associated with time.
Table 1 - Historic Industrial Turbine Oil Formulations
"R&O" Conventional Next-Generation
Basestock Group I Group I / II Group II
Antioxidant Phenolic Phenolic + DPA, PANA,

Diphenylamine Phenolic
Supplemental -- Benzotriazole Benzotriazole,

Antioxidants Thiadiazole,
Phosphite
Rust Inhibitors SuccinicAcid / E s t e r Succinimide, Succinimide,

Succinic Acid / Ester Succinic Acid / Ester
Antifoams Silicone Acrylate, Acrylate,

Silicone Silicone
Other Pour Depressant, Pour Depressant, Pour Depressant,

Antiwear Antiwear, Antiwear,
Demulsifier Demulsifier
Turbine Oil Formulation
A combination of additives is used in turbine oils to control various important

properties that are required for reliable, relatively trouble-free operation for an extended
time period. Indeed, many low severity steam turbines have a history of requiring a full
SWIFT ET AL. ON FIELD APPLICATION REQUIREMENTS 41
lubricant changeout only every 10 to 20 years or longer, with periodic top-up with fresh
oil, of course. Antioxidants are the additives which have the strongest influence on the
useful life of turbine oils. The earliest formulations used hindered phenolics for this
purpose [I]. These molecules function by trapping reactive radical materials which are
generated by oxidation of the oil hydrocarbons by air. This prevents the propagation of
chain reactions which generate acids and sludge, making the oil unusable for further
service. Later it was found that, at temperatures below about 110~ the phenolics could
be made more effective by adding a low level of diphenylamine antioxidant which
synergistically regenerated the phenolic additive [2].
In the search for oils which could function well in turbines operating under very
severe, high temperature operating conditions, basestocks with different compositions
were investigated. So-called API Group II basestocks, which are manufactured by high
pressure, catalyzed hydrogenation refining processes, have higher levels of saturated
hydrocarbons in their composition. These are now becoming much more widely
available, and they do offer some potential advantages for lubricants used in certain
circumstances. However, they also have some limitations not experienced by Group I
basestocks, which are manufactured from distillates by solvent extraction processing.
Group II basestocks do not necessarily respond to the same antioxidants that are effective
in Group I basestocks, and carefully tailored additive combinations are needed if superior
oxidation resistance at high temperatures is to be achieved. The patent literature indicates
that aromatic amines of various kinds are being used at increasingly higher levels in such
applications, although still in combination with phenolics in some cases [3 - 6].
Phosphites have also been developed as supplementary antioxidants to boost the action of
phenolics in Group II formulations [7].
It has also been discovered that other types of additives can be used to further extend
the useful life of turbine oils by suppressing oxidation of the hydrocarbons [6, 8]. Iron
and copper, which are used in the construction of turbine equipment, may slowly dissolve
at parts-per million (ppm) levels in lubricant oils in service [9]. Solubilized metal ions
act as catalysts for oil oxidation, so additives that actively prevent metals from
dissolving, or which bind the metal ions once they are dissolved, were developed.
Benzotriazoles and thiadiazoles are now commonly used for this purpose.
Interfacial properties have always been important for the successful functioning of
turbine oils. Fast and complete spontaneous separation of oil and water in steam turbine
systems is essential so that water does not reach key elements such as bearings and gears.
If this property is not adequate, then a demulsifying additive may be employed. Air is
not a good lubricant, so it is equally important that the oil does not produce a lot of foam,
and that it also rejects air bubbles from the body of the oil quickly. This latter property,
known as air release, has become even more important in recent years since turbine
systems are being constructed with ever smaller oil reservoirs. The residence time of the
oil in the sump is now much shorter, with less chance of entrained air being adequately
released. Foam suppressant additives need to be carefully selected in order to prevent
excessive foam formation, but still retain short air release times. Antifoam additives are
moving from silicone types to acrylate types since the latter seem to give a better balance
between these opposing effects.
Other important additives commonly used in turbine oils are rust inhibitors and pour
point depressants [10]. If a turbine oil is to be used in geared equipment, then antiwear
additives are also required, such as ashless phosphorus and/or sulfur compounds. Wear
protection is generally specified in terms of four-ball wear (D2266) and FZG gear test
(D5182) performance.
Turbine Oil Oxidation Testing
Standard testing methods developed and maintained by ASTM Sub-Committee D02.C

on Turbine Oils, and other ASTM sub-committees, have made major contributions for
several decades in the science of turbine oil formulation and performance assessment.
Two particular standards have been widely used by formulators, product quality labs and
turbine operators to determine the useful life of turbine oils. These procedures use
techniques that accelerate effects that are encountered in turbine operations. They are
intended to be used for particular situations, but cannot possibly directly simulate the
environment encountered by oils in the multitude of turbine systems that operate under
very diverse conditions.
The ASTM Test Method for Oxidation Stability of Steam Turbine Oils by Rotating
Pressure Vessel (D 2272), sometimes referred to as "RBOT", was designed for
monitoring in-service turbine oils, to warn the equipment operator when the oil was near
the end of its useful life. Oil that is allowed to remain in a turbine system after
degradation begins can be very damaging. Acids will attack metal surfaces, sludge will
plug filters and tubes, and oxidation products can form varnish on hot surfaces such as
bearings. This test is run at 150~ under pure oxygen at 620 kPa pressure and in the
presence of a solid copper catalyst, in order to promote measurable oxidation in a short
time. This makes it unrepresentative of most steam turbines which operate below 100~
On the other hand, oil is mixed with water in the test, making it unrepresentative of
higher temperature combustion turbines. The test is sensitive to particular additive
chemistries so that differently formulated oils cannot be validly compared using this test.
Further, it is not so much the initial RBOT value of a fresh turbine oil which is important,
but how well the turbine oil retains its RBOT value during extended service.
Nevertheless, the RBOT has been widely used for many years for controlling a wide
variety of turbine and other mechanical equipment.
The ASTM Test Method for Oxidation Characteristics of Inhibited Mineral Oils
(D943), sometimes referred to as "TOST", was designed to give an indication of the life
of steam turbines by using a test temperature of 95~ which is closer to that experienced
in real steam turbine applications. It is run using pure oxygen, water and a combination
solid iron/copper catalyst, and is reputed to be less dependent on additive chemistries
than the RBOT. With Next-Generation Turbine Oils exceeding 10000 hours (almost 14
months) in the TOST test, it has become too long to be of practical use for formulating
new products or evaluating turbine oils from the field. Further, the test conditions do not
simulate those in high severity combustion turbines.
The formation of sludge and varnish by thermal-oxidative processes has become more
critical as operating temperatures have increased. The ASTM Test Method for
Determination of the Sludging and Corrosion Tendencies of Inhibited Mineral Oils
(D4310) test has been used to measure sludge formation, but once again it is run under
the same relatively mild conditions as the TOST and for a much shorter, 1000 hour test
duration. Recently ASTM Subcommittee D02.9 on Oxidation has begun to develop the
use of the A S T M Standard Guide for Universal Oxidation ! Thermal Stability Test
Apparatus (D4871), sometimes referred to as the "UOT", at 155~ to address the need
for a determination of the tendency of turbine oils to degrade at high temperatures. It
measures the formation of sludge, varnish and acids, and monitors viscosity increase.
As shown in Table 2, RBOT and TOST lives have increased for each succeeding
generation of turbine oils. However, the RBOT and TOST methods do not necessarily
correlate well with each other. In the same way, oils with long oxidation lives in either
RBOT or TOST do not necessarily give the lowest levels of sludge formation in either
the D4310 or UOT procedures.
Table 2 - Turbine Oil Oxidation Tests
9 Typical Oxidation Test Values
"R & O" Conventional Next-Generation
D2273 (RBOT), min 250 600 1500
D943 (TOST), hr 1500 4000 10,000
RBOT, TOST, and UOT results are not well correlated
ISO 32 RBOT TOST I)4310 U O T (155~

Turbine Oils (rain) (hr) (mg sludge) (sludge)
A (Group I) 611 6564 <10 Medium
B (Group I) 880 6228 29 High
C (Group I, AW) 875 7040 60 High
D (Group II) 1675 7000 19 Low
E (Group II) 1450 > 10,000 14 Low
F (Group II, AW) 1455 >10,000 96 High
It was noted previously that the TOST is most representative of steam turbine
operating conditions, so that claims of very long lives - sometimes as long as 20000 hours
or more - for high temperature combustion turbines, are of limited value. In addition,
consider that the end-point of this test is an oil total acid number of 2.0 mg KOH/gm of
oil. Figure 1 shows photographs of the oil/water interface of turbine oils formulated with
Group II basestock from different manufacturers in the TOST, all at 10,000 hours test
time and all with a TAN of less than 0.5 mg KOH/gm of oil. It is readily apparent that
each shows characteristics which would be highly undesirable for an oil in an operating
turbine, whether it be a high level of sludge and varnish, severe corrosion of the metal
catalyst, or partial emulsification with the water. In addition, we know from experience
in this and other tests that acidic products of oxidation often rapidly become insoluble in
the oil and may not be titrated by methods D664 or D3339. This is not to say that the oils
would not work well in turbine applications, just that TOST data may give an over-
optimistic estimate of performance.
Figure 1 - TOST (D943) at 10000 Hours
As discussed above, turbine oils formulated with Group II basestocks do offer some
benefits compared with Group I based turbine oils, but they may also have some inherent
limitations. Many turbine oil suppliers continue to offer both types of products to allow
end-users to select the one that is most suited to their application. Products formulated
with Group II basestocks offer longer RBOT and TOST lives, and are especially suited to
high severity turbine applications. At the same time, Group II basestocks are poorer
solvents than Group I, as is illustrated by their higher aniline points. This can have
important consequences in at least two ways. Firstly, highly polar additives, such as rust
inhibitors, may be more sparingly soluble in Group II basestocks, so that additive
selection will be more limited. Dispersed additives, such as antifoams and demulsifiers,
are more difficult to maintain as suspensions in Group II basestocks and have an
increased tendency to drop out. Secondly, the oxidation products which gradually
accumulate in all oils during service will have a greater tendency to separate from the oil
and form deposits.
When a turbine oil lubricating system is drained of one turbine oil and refilled with
another turbine oil, as much as 10 to 15% of the previous oil can remain in the system
due to clingage, low-spots, and piping dead-legs. Compatibility between products is an
important consideration when converting from one type of turbine oil to another, either
from different suppliers or for different types from the same supplier. This is even more
critical with Next-Generation turbine oils, where incompatibilities can result in severely
SWIFT ET AL. ON FIELD APPLICATIONREQUIREMENTS 45
downgraded turbine oil performance. Carryover from the previous product in a

lubricating system conversion was simulated by adding 15% of either a Group I or Group
II based turbine oil to Next-Generation turbine oils from different suppliers. As shown in
Figure 2, the resulting RBOT value of the blend was significantly lower than would be
expected from simple linear blending, indicating incompatibility between the additives in
the two products. Significant pre-screening is required before converting a lubricating
system from one product to another, including testing of all critical properties. Without
such care the performance of the resultant turbine oil combination may be significantly
lower (in this case, RBOT) than that anticipated from fresh oil values.
Figure 2 - Turbine Oil Cross-Contamination
Steam Turbine Applications
The lubricating oil in both steam and combustion turbines is responsible for
proper lubrication of the journal and thrust beatings, and may also serve as the lubricant
for peripheral equipment [11]. The oil must be able to remain stable during its long
operating life, retaining its viscosity, rust prevention, demulsibility, and oxidation
resistance. The ASTM Practice of In-Service Monitoring of Mineral Turbine Oils for
Steam and Gas Turbines (D4378) provides guidance on tolerable limits for changes in the
turbine oil performance properties during normal service [12]. Although laboratory
testing is useful in defining the quality of fresh turbine oil, the ultimate test of robustness
is in actual field applications.
Typical bearing temperatures in a steam turbine range from 50 - 70~ although local
temperatures may be much higher. These temperatures accelerate turbine oil oxidation,
although not to the extent found in combustion turbines. During normal operation of a
steam turbine water can enter the turbine oil through leaks in the shaft seals or cooling
coils, becoming entrained in the oil, and it is important that the oil retain its ability to
shed water (demulsibility). Water can collect on metal surfaces, especially during start-
up and shutdown, and the turbine oil needs to retain its rust protection. Particulates can
enter the system through open air-breathers or improper oil addition procedures, and the
turbine oil performance must not be impacted by typical inorganic contaminants (which
are ultimately removed by external filtration systems). Despite these many deleterious
factors, a robust turbine oil can remain in steam turbine service for several decades!
In one example, two 125 MW steam turbines have been successfully operating on a
Conventional turbine oil since 1957. The turbine oil in the two 4000 gallon reservoirs
had not been changed during the 40+ years of operation, except for normal make-up for
oil losses. During recent routine sampling the turbine oil showed very little degradation
from the original product. The viscosity remained in the ISO 32 range and the oil
showed very low levels of particulates (ISO 15/11 - 17/13) and low metals content. Low
water levels (less than 100 ppm) confirmed the retention of water shedding ability over
this period. Very little oxidative degradation had occurred during this ,period, as
confirmed by the low Total Acid Number of 0.11 mg KOH/gm, which was similar to the
original turbine oil. However, slow oxidation over this long period had begun to deplete
the oxidation inhibitors, with RBOT values at 13 - 25% of the original product level,
indicating that a turbine oil change or sweetening should be considered. This long term
reduction in oxidation life was confirmed by FT-IR analysis, which showed that the
antioxidant was near depletion. Nevertheless, it is important to note that this extremely
long life in this steam turbine was obtained with a Conventional turbine oil. It is unlikely
a Next-Generation turbine oil would have provided longer service, given the variety of
factors (temperature swings, water ingression, contamination, etc.) which can adversely
impact a steam turbine oil system.
A second example comes from a 45 MW steam turbine operated as part of a combined
cycle cogen facility. In this case, an anti-wear Conventional turbine oil was used due to
the gearing on the companion combustion turbine. The turbine oil has continued to
perform successfully in continuous operation for over 23000 hours. Samples taken at the
end of 1999 showed excellent resistance to oxidation, with a TAN of 0.11 mg KOH/gm
oil, and excellent demulsibility. Figure 3 confirms the oxidation resistance of the turbine
oil, showing the RBOT of the oil from the steam turbine sump over a two year period,
with the oil still retaining -70% of its original RBOT value.
Continuously Operated Combustion Turbines
A continuously operated, or baseloaded, combustion turbine is kept running on an

ongoing basis providing power at that location or supplying into the utility grid. Turbine
oil temperatures range from 70 - 100~ although newer, more severe combustion
turbines can produce oil temperatures of 120~ (12). In continuously operated
combustion turbines Conventional turbine oils have also shown excellent performance.
In the cogen example cited above, the companion 70 MW geared combustion turbine
also operated with an anti-wear Conventional turbine oil, with a separate turbine oil
sump. Again, after over 23000 hours of operation the turbine oil shows excellent
oxidation resistance, with a TAN of 0.11 mg KOH/gm of oil, and excellent demulsibility.
Figure 3 confirms the oxidation resistance, with the RBOT retaining -85% of its initial
fresh oil value.
700
600 -~ ........ ~-
500
._= 400
300 S team
o - ~ - Combustion
200
100
0 i i i i
0 5000 10000 15000 20000 25000

Hours of Operation
Figure 3 - Performance of Conventional Anti-Wear Turbine Oil in

Cogen Steam and Combustion Turbines
A second example is from a series of eight 35 MW combustion turbines running

continuously in a municipal utility. The units ran with a Conventional turbine oil which
had not been changed during the 20+ years of operation, except for normal make-up for
oil losses. In a recent sample the turbine oil showed very little degradation from the
original product. The viscosity remained in the ISO 32 range and the oil continued to
demonstrate good rust protection. Retention of water shedding ability over this period
was confirmed by low water levels (less than 30 ppm). Very little oxidative degradation
and sludge formation had occurred during this period, as confirmed by the low Total
Acid Number of 0.01 mg KOH / gm which was similar to the original turbine oil.
However, slow oxidation over this long period had begun to deplete the oxidation
inhibitors, with RBOT values at <15% of the original product, indicating a turbine oil
change or sweetening should be considered. Again, it is worth noting that this extremely
long service life was obtained with a Conventional turbine oil.
Cyclically Operated Combustion Turbines
One application where the enhanced oxidation stability of the next-generation turbine
oils has proven to be of value is in combustion turbines that are cyclically operated in
peak-loading operations. The turbine oil is subjected to high temperatures during turbine
operation, and then is allowed to cool during the stand-down period. When this cyclic
operation is combined with more severe, high temperature combustion turbines, the
lubricating oil is subjected to extreme thermal and oxidative stresses.
48 TURBINE LUBRICATION IN THE 21sT CENTURY
In one field example, a series of 80-90 MW combustion turbines were peak operated
on a weekly cycle. During weekday operation the turbine oil was subjected to typical
operating temperatures of 70 - 80~ with isolated peaks approaching 120~ in the main
shaft bearings, then returned to ambient temperature when idled over the weekend. In
addition to lubricating the shaft bearings, the turbine oil also provided hydraulic power to
the fuel ratio servo control valves. A Conventional turbine oil was used in the system for
approximately two years, at which time a non-start condition occurred due to loss of
hydraulic pressure to the servo-valve. The hydraulic failure was traced to plugging of the
fine pencil filter immediately upstream of the servo-valve.
Infra-red analysis of the deposits removed from the pencil filter showed oxidation by-
products, with no significant additive or metals content. The turbine oil itself (Table 3)
was in relatively good condition, with low TAN and satisfactory RBOT. The oil was
clear at normal turbine operating temperature, but slightly cloudy at ambient temperature.
When filtered through a 5 micron filter the oil contained about 90 ppm of an organic
particulate.
Table 3 - In-Service Turbine Oil Analysis
Viscosity, cSt @ 40~ (ASTM D 445) 32.8

Neutralization No., mg KOH / gm (ASTM D 664) <0.1
Rotary Bomb Oxidation Test - RBOT, minutes (ASTM D 2272) 360
Particulates, 5 ~m filtration at ambient temperature, ppm 90
It was concluded that the organic deposits were the result of normal high temperature
oxidation of the turbine oil. During turbine operation these organic particulates would
either be removed by the oil filtration system or remain in solution at the 70+~ operating
temperatures. However, when the turbine was idled the oil temperature would fall to near
ambient and the low level particulate would come out of solution, building up over time
in the pencil filter which protected the fuel ratio servo-valve. This situation is unique to
cyclic operation, since in base loaded, or continuously operating, combustion turbines the
oil temperature is not subject to wide fluctuations on a routine basis.
Next-Generation Turbine Oil
Several approaches were considered to address the pencil filter plugging issue. The
most obvious approach would be to use finer filtration media in the turbine oil circulating
system. However, since the oxidation by-products were soluble at operating
temperatures, improved filtration during operation would not remove the particulates that
formed only when the turbine was idled.
The approach selected was to utilize the enhanced oxidation stability of a Next-
Generation turbine oil to slow the build-up oil oxidation by-products. The turbine oil
utilized a Group II basestock with a balanced additive package providing good foam,
rust, and demulsibility performance. The oil showed exceptional resistance to oxidation
in both the RBOT and TOST tests, and to sludge formation.
SWIFT ET AL. ON FIELDAPPLICATION REQUIREMENTS 49
1400
1200
~. - -'- GT-2
1000 -e-- GT-3
- "'-a, " ' , \ ,t...
9~ 800
** ~ o." A..... ~ O
9t-; 600
400
200
i i i i i
0 5000 10000 15000 20000 25000 30000

Fired Hours of Operation
Figure 4 - Performance of Next-Generation Turbine Oil in Cyclically

Operated Combustion Turbine
The Next-Generation turbine oil was installed in three identical high temperature
combustion turbines in 1996. The turbines have continued to run in the weekly cycled
mode since that time, with no evidence of sludge formation, pencil filter plugging, or
non-start conditions. Table 4 shows the periodic oil monitoring for one of the three
turbines, with the other two systems showing similar results. Overall performance has
been excellent, with retention of demulsibility, foam resistance, and rust protection
properties. Over the 3+ year period, TAN has remained low and RBOT remains at 60%
of the original (see Figure 4), well above the recommended oil change-out limits.
Conclusions
Turbine oil quality, for both steam and combustion turbine applications, has continuously
evolved over the last two decades. Improvements have been made in both the additive
chemistry and basestock selection. While laboratory testing is useful in developing
turbine oil formulations, field performance validation is essential to validating the
ultimate service life. Interactions between turbine oils need to be considered when
changing out the turbine oil in the lubricating system.
Conventional turbine oils have shown excellent, cost effective field performance, with
service lives which can be in excess of 20 years for steam turbines and 10 years for
combustion turbines. Next-Generation turbine oils show benefits in cyclically operated,
high temperature combustion turbines, where the higher oxidation stability provides
resistance to the formation of oxidation by-products which can settle out when the turbine
is idled. Both types of turbine oils provide the turbine operator with the flexibility to
select a product which suits a particular application.
G'l
C:)
--4
:13
rn
T~
m
i--
c
Go
20
Table 4 - Monitoring of Next-Generation Turbine Oil in Cyclically Operated Combustion Turbine .--4
Z
Months in Combustion Turbine t 6 10 18 22 27 38 ---4

Hours of Fired Operation "7"
400 2800 4700 9200 12700 14500 21300 ITI
Number of Fired Starts 3 16 27 54 81 102 175 Ix3
Viscosity, cSt @ 40~ (ASTM D445) 32.4 32.7 32.7 32.7 32.9 32.8 33.1
Neutralization Number, m g KOH / g m (ASTM D974) 0.04 0.06 0.03 0.04 0.06 0.01 0.02 m
z
Rotary Bomb Oxidation Test - RBOT, minutes (ASTM D2272) 1207 1061 1256 803 732 847 744
Particle Count - ISO Classification 20/17 13110 14/11 18/9 12/8 15/13 !.4/11 ..<
Rust Test, Seq. A (ASTM D665) Pass Pass Pass Pass Pass Pass Pass
Foam Tendency / Stability, Seq. I1, rnL ( A S T M D892) 010 2010 20/0 30/0 30/0 60/0
Demulsibility, minutes to 37 mL H 2 0 (ASTM DI401) 15 15 20 15 30 24 30
Finally, new laboratory tests need to be developed for the formulation and evaluation
of turbine oils. The TOST test (D943) runs too long to be an effective screening tool, and
the conditions are too mild to simulate a combustion turbine. The RBOT test (D2272) is
effective as a field monitoring tool, but does not correlate well with other oxidation tests,
We encourage the ASTM Subcommittee D02.C on Turbine Oils to take on the challenge
of developing a rapid laboratory oxidation test that can be correlated to field turbine
performance, allowing turbine operators to evaluate and select robust turbine oils for their
equipment.
Acknowledgement - The authors would like to gratefully acknowledge Madam Juves and
Warren Eckert of ExxonMobil for their assistance in compiling the field application data
cited in this paper.
References
[1 ] Mastin, R. G., "Practices for Determining the Expected Life of Used Turbine Oils,"
Symposium on Steam Turbine Oils, ASTM STP 211, American Society for Testing
and Materials, West Conshohocken, PA, 1956, pp. 76-92.
[2]. Gatto, V. J. and Grina, M. A., "Effects of Base Oil Type, Oxidation Test Conditions
and Phenolic Antioxidant Structure on the Detection and Magnitude of Hindered
Phenol/Diphenylamine Synergism," Lubrication Engineering, 1999, pp.11-20.
[3] Rasberger, M., "Oxidative Degradation and Stabilisation of Mineral Oil Based
Lubricants," Chemistry and Technology of Lubricants, R. M. Mortier and S. T.
Orszulik, Ed., Blackie Academic and Professional, 1992, pp. 83-123.
[4] Dolby, G. W. and Kofke, W. A., "The Role of Base Oils and Additives in Modern-
Day Turbine Oils," Symposium on Steam Turbine Oils, ASTM STP 321, American
Society for Testing and Materials, West Conshohocken, PA, 1962, pp. 1-16.
[5] Wolf, L. R., U.S. Patent 5102567A, 1992.
[6] Turnquest, B. W., Chao, T. S., and Broman, V. E., U.S. Patent 3909420A, 1975.
[7] Cohen, S. C., U.S. Patent 4652385A, 1987.
[8] Angerer, H. F., Brady, A., Kristen, U., and Regenass, F., U.S. Patent 4701273A,
1987.
[9] von Fuchs, G. H., "Evaluation and Performance of Turbine Oils," Symposium on
Steam Turbine Oils, ASTM STP 211, American Society for Testing and Materials,
West Conshohocken, PA, 1956, pp. 76-92.
[10] von Fuchs, G. H., "Rust Inhibitors - Their Evaluation and Performance," Symposium
52 TURBINELUBRICATIONIN THE 21 sT CENTURY
on Steam Turbine Oils, ASTM STP 321, American Society for Testing and
Materials, West Conshohocken, PA, 1962, pp. 28-41.
[11] Wills, J. G., Lubrication Fundamentals, Marcel Dekker, New York, 1980, pp. 226-
231 and 241-252.
[12] Roberton, R. S., "Background and Development of ASTM D 4378: Practice for In-
Service Monitoring of Mineral Turbine Oils for Steam and Gas Turbines,"
Turbine Oil Monitoring, ASTM STP 1021, W. C. Young and R. S. Roberton,
Eds., American Society for Testing and Materials, West Conshohocken, PA,
1987, pp. 3-18.
Douglas J. Irvine I
Performance Advantages of Turbine Oils Formulated with Group II and Group III
Basestocks
Reference: Irvine, D. J., "Performance Advantages of Turbine Oils Formulated with

Group II and Group III Basestocks," Turbine Lubrication in the 21 st Century, ASTM
STP 1407, W. R. Herguth and T. M. Wame, Eds., American Society for Testing and
Abstract: High temperature gas turbine applications, such as GE Frame 7 turbines, place
high demands on the thermal and oxidative stability of turbine oils. Successful products
not only start with a superior quality basestock, but must also employ a carefully
balanced, low volatility, thermally stable additive system. The excellent properties of
Group II and Group III basestocks such as high viscosity index, low volatility, superior
oxidative resistance, and high thermal stability make them ideal choices for this service.
Turbine oils formulated with severely hydrocracked Group II and Group III basestocks
offer performance on par with those formulated with synthetic hydrocarbons.
A brief review of the role and requirements of modem turbine oils is given together
with a more extensive look at current basestock technology versus traditional solvent
refined basestocks. Additive selection for high temperature turbine oils is also discussed.
In addition, data presenting the impact of poly-cycloparaffin content on the stability of
highly saturated Group II and Group III basestocks is presented. The oxidative stability
of additized basestocks increases as poly-cycloparaffin content decreases.
Keywords: turbine lubricant, antioxidants, basestocks, high temperature, hydrocracked,

iso-dewaxed, cycloparaffins
Introduction
There have been steadily increasing demands placed on turbine lubricants over the
past few decades. As turbine power outputs have increased, oil volumes in turbine
reservoirs have decreased. Bearing temperatures have risen to over 315~ (600~ and
elaborate labyrinth air sealing systems have been developed to shield the lubricant from
these high temperatures [1]. The lubricant, however, still sees relatively high
1 Product Specialist, Specialty Products & Fluids, Petro-Canada Lubricants, Research &
Development, 2489 North Sheridan Way, Mississauga, Ontario, L5K 1A8.
53

temperatures, as it must cool the bearing and bearing housing. In addition, the lubricant
is exposed to the hot sealing air from the labyrinth seal, which tends to strip out volatile
oxidation inhibitors [2]. The end result is increased thermal stresses on the lubricant and
increased bulk operating temperatures. These demands have led to the development of
high-temperature lubricants specifically designed for heavy-duty gas turbine service [3].
General Electric's Frame 7 gas turbine oil specification (GEK 32568) highlights the
severe requirements of gas turbine oil service and the need for thermally stable, low
volatility antioxidants.
The increased occurrence of cogeneration or combined cycled turbine operation (in
which the waste heat from the gas turbine exhaust is used to drive a steam turbine) has
also placed new demands on turbine lubricants. Lubricants are now commonly required
to meet both steam and gas turbine oil performance specifications.
Historically turbine lubricants have been simple rust and oxidation inhibited solvent
refined oils. Wet seal designs led to high oil make-up rates, which served to replenish the
oil's additive system. Hence oil change outs were infrequent, and even oils of moderate
stability had long service lives.
Today's heavy-duty high temperature gas turbines use dry seals and hence have very
low oil make-up rates. This coupled with the increased severity of operating conditions
necessitates close monitoring and more frequent oil change outs.
Turbine Lubricant Role and Requirements
The three main roles a turbine lubricant must fulfill are 1) bearing lubricant, 2) heat
transfer fluid, and 3) hydraulic control fluid. Oxidation of the lubricant can change its
physical and chemical properties making it less efficient at, or preventing it altogether
from performing these key functions. Physical changes include viscosity increase, which
leads to excess friction in the bearings and an accompanying generation of heat. Higher
viscosity in itself also leads to reduced heat transfer. This heat buildup serves to further
accelerate fluid oxidation and breakdown, leading to varnish, sludge, and deposit
formation. Sludge and varnish can plug filters and prevent proper operation of hydraulic
control devices. Sludge and varnish buildup also tends to insulate the bearings, further
exacerbating the problem of poor heat transfer.
Chemical changes due to oxidation lead to formation of acids and other oxygenated
species. These oxidation products can cause rust and corrosion, attack seals, and promote
foaming and poor demulsibility. Polar oxidation products can also compete with EP/anti-
wear additives and corrosion inhibitors for surface activity leading to poor performance.
In addition to accelerating oxidation, high temperatures also lead to additive
breakdown and evaporation. Examples of heat sources that turbine lubricants are
exposed to include: the bearing oil film (93~ bearing housing inner surfaces (149~
sealing air entering the oil spaces (204~ and the hot turbine shaft (204~ Smith [3]
concluded that about 2% of the oil flow in a turbine at temperatures above 204~
accounts for more than 90% of the used oil life (i.e. high temperature exposure of small
quantities of oil dictates the bulk oil life).
IRVINE ON GROUP II AND GROUP III BASESTOCKS 55
In order to function effectively over long periods of time a turbine lubricant must,
therefore, have excellent thermal and oxidative stability. One of the key factors to
achieving this stability is the selection of a highly refined, high purity basestock. This
alone, however, will not yield a long life turbine fluid. The antioxidant system employed
must also possess excellent thermal stability and have low volatility.
Basestock Technology
Base Oil Type
Refined base oils are complex mixtures of various hydrocarbon structures including
paraffins, cycloparaffins, and aromatics. Depending on the refining process used they
can also contain significant amounts of sulfur, nitrogen, and oxygen containing
compounds. The two main types of base oils available in the marketplace today are
naphthenic and paraffinic. Naphthenic base oils are characterized by high aromatic and
cycloparaffin content, but have low wax content. They have low viscosity indexes, low
pour points, and good solvency, but suffer from poor oxidative stability due to their high
aromatic content. Because of their poor oxidative stability, naphthenic base oils do not
find their way into most lubricant applications.
Paraffinic base oils are characterized by high paraffin, cycloparaffin, and wax content.
They have higher viscosity indexes and good oxidative stability, but have poorer
solvency and pour points. Highly refined paraffinic base oils with their high degree of
saturates and low sulfur, nitrogen, oxygen, and aromatic content, have been used for
years in turbine lubricants. Recent advances in refining and dewaxing technology,
however, now offer paraffinic base oils with even higher levels of saturates, and almost
no sulfur, nitrogen, oxygen, and aromatic content.
Base oil chemical structure and its impact on lubricant performance is shown in Table
1. Careful molecular tailoring in the refining process can optimize performance by
selective removal or transformation of undesirable components via ring saturation and
ring opening processes. Desirable paraffins and cycloparaffins yield high viscosity
indexes and good oxidative stability, while undesirable aromatics and hetero-atoms lead
to instability and deposit formation.
Table 2 gives a more detailed overview of the impact of hydrocarbon structure on
viscosity index, pour point, and oxidative stability.
Removal of the undesirable sulfur, nitrogen, oxygen, and aromatic components in
paraffinic base oils can be accomplished via solvent refining or hydrocracking, although
hydrocracking is more effective. Severe hydrocraeking offers almost complete removal
of these compounds. The use of iso-dewaxing technology offers further refinements
through increasing desirable iso-paraffin content. These refining steps have been
described in further detail elsewhere [4, 5]. Severely hydrocracked and iso-dewaxed
basestocks have high viscosity indexes and respond well to pour point depressants. This
makes them ideally suited to operating environments with wide temperature ranges.
Table 1. lmpact of Chemical Composition on Lubricant Performance

Chemical Composition Performance Impact
Saturates Viscosity Index

- Paraffins (N- & ISO-) ....)
- Cycloparaffins Oxidation Stability

(Naphthenes)
Pour Point
Aromatics
Alkyl Benzenes Volatility
- Naphtheno-Aromatics
- Multi-Cyclic Aromatics Solubility
Hetero-atoms Seal Characteristics

- Sulphur, Oxygen, Nitrogen
Response to Additives
The American Petroleum Institute (API) classifies basestocks according to sulfur

content, saturates, and viscosity index as shown in Table 3.
Typical chemical compositions of 4 cSt (approximate viscosity at 100~ API Group I
through IV basestocks are shown in Table 4. As the degree of refining is increased from
solvent refining to severe hydrocracking significant reductions in aromatic content are
achieved. Moving from solvent dewaxing to iso-dewaxing yields significant increases in
iso-paraffin content, and hence viscosity index. Refining and dewaxing technology now
allows production of Group III basestocks which are closer akin to synthetic Group IV
basestocks than solvent refined Group I basestocks. Recent rulings in the United States,
in a well known dispute between Castrol and Mobil, have indicated that these new
basestocks may in fact be called "synthetic" (April 1999 decision by the National
Advertising Division of the Council of Better Business Bureaus).
Basestock Research
Increasing performance demands can only be met up to a point by increasing additive

treat rates. At some point the formulator must look at the limitations of the basestock
being employed. Solvent refined oils cannot meet today's heavy-duty high temperature
gas turbine performance requirements. Mobil [6] notes that they have used high-quality
hydrotreated basestocks in their premium turbine oils worldwide since the mid 80's.
They comment that the low sulfur and aromatic levels provide excellent oxidation and
colour stability.
Attributes of severely hydrocracked/hydrotreated basestocks include: 1) extremely
low aromatic contents, 2) very low sulfur levels, 3) water white colour, 4) low DBD/DBF
levels (dioxin precursors), 5) excellent high temperature stability, and 6) excellent water
and air separation characteristics.
Table 2. Effect of Hydrocarbon Structure on Selected Lubricant Properties

Chemical Type Structure Viscosity Pour Point Oxidation
Index Resistance
n-Paraffin (Wax) ~ Very High Solid Excellent

- 175 @ s0oc
Iso-Paraffins w i t h /N/N/N/ High Good Excellent

branched chains " ~ " -150
Iso-Paraffins with /~./,. A / Good Good Excellent

highly branched . v y v ~130
chains
Cyclo-Paraffin - Good Good Good

single ring with ~130
long chains
Naphthenes, ~ Poor Good Medium

polycondensed ~ -60
Monoaromatics, \ / N / N /-~----X, / N / N / Poor Good Medium

long chains v v ~ v v ---60
Polyaromatics ~ Very Poor Good Very Poor

<0
It is well known that aromatic and sulfur contents have a strong influence on the
oxidative stability of base oils. Unadditized solvent refined oils have better oxidative
stability than hydrocracked base oils and PAOs (polyalphaolefins), which have very low
aromatic and sulfur contents. Certain natural sulfur and aromatic compounds present in
solvent refined base oils act as oxidation inhibitors, and it has been found that there is
also an optimal aromaticity level [7]. However, it has also been noted that higher
aromatic contents lead to increased deposit formation in oxidized base oils [8]. The
presence of aromatics can also lead to the formation o f insoluble compounds if oils are
exposed to high temperatures.
Table 3. API Base Oil Categories

Base Oil Category Sulfur (%) Saturates (%) Viscosity Index
Group I > 0.03 and/or < 90 80 to 119

Group II ___0.03 _>90 80 to 119
Group III _< 0.03 _>90 > 120
Group IV All polyalphaolefins (PAOs)
Group V All basestocks not included in Groups I, II, III, or IV
Table 4. Chemical Composition of Various Classes of 4 cSt API Basestocks

Basestock A B C D E F G H
API Classification I 1 II II I1 II+ llI IV
Description Solvent Solvent Hydro- Hydro- Severely Severely Severely PAO

Refined Refined Cracked Cracked Hydro- Hydro- Hydro-
Cracked Cracked Cracked
Dewaxing Method Solvent Solvent Solvent Iso Solvent Iso Iso None
Mass Spec Analysis

Paraffins, n & iso- 25.7 29.0 23.7 30.2 32.6 51.4 76.1 100.0
Mono-Cycloparaffins 20.8 25.0 30.8 30.5 34.2 24.4 14.7 --
Poly-Cycloparaffins 27.9 31.7 39.1 35.3 32.8 23.9 9.2 --
Aromatics 24.9 14.2 6.4 4.0 0.4 0.3 ....
Thiophenes 0.7 0.1 . . . . . . . . . . . .
Paraffins + Mono- 46.5 54.0 54.5 60.7 66.8 75.8 90.8 100.0
Cycloparaffins
While natural aromatic and sulfur content in solvent refined base oils enhances their
inherent oxidative stability, it is a strong liability in their response to antioxidant
additives. Rasberger [9] noted that the oxidative stability and sludge forming tendency o f
additized hydrotreated basestocks, with low aromatic and sulfur content, are clearly
superior to that o f solvent refined oils. It was also noted that polar compounds present in
solvent refined oils (especially aromatics and nitrogen containing compounds) reduce
antioxidant efficacy through interactions. In addition, solvent refined oils contain
naturally occurring pro-oxidants, which have a further negative impact on basestock
stability.
Gatto and Grina [10] noted large differences in the response of different hydrocracked
basestocks to an optimized hindered phenol/alkylated diphenylamine ratio. A very high
viscosity index Group II basestock significantly outperformed other Group I and II
basestocks tested.
Galiano-Roth and Page [11] show the superior response of two-stage hydroprocessed
oils to antioxidants. The oxidative performance increased as total and polynuclear
aromatics decreased. Galiano-Roth and Page noted that while oxidation stability is
increased by hydroprocessing, solvency (for additives and oxidation products) is
decreased. Solubility problems were observed when an additive system typically
employed in solvent refined basestocks was examined in the hydroprocessed basestocks.
It was also noted by Deckrnan et al [6] that hydroprocessing reduces the natural
dispersancy of basestocks because aromatic levels are decreased. This can be offset by
the use of dispersants in certain classes of lubricants, but not in turbine oils where
excellent air and water separation are required. They note that the very low aromatic
levels in severely hydrocracked basestocks can lead to problems in terms of additive
solubility (especially with high-temperature antioxidants needed for gas turbine oils) and
oxidation product solubility.
Recent work comparing the performance properties of low aromatic content
hydrocracked basestocks and PAOs indicates that performance properties (including
oxidative stability of the additized basestocks) can be predicted from their chemical
composition [12]. It was observed that oils containing low levels of condensed multi-ring
naphthenic structures exhibited superior oxidative performance and lower volatility, but
had poorer solvency. The Group III basestocks tested had oxidative performance on par
with the PAOs. It was also noted, as in previous work, that these highly saturated
basestocks can cause solubility limitations with certain additives.
The comparable performance of Group III basestocks and PAOs is in agreement with
a previous study by researchers (Henderson et al) at the author's company [13]. In this
study a wide range of commercially available industrial oil additive packages, designed
for synthetic basestocks, were used to compare the performance of Group III and Group
IV basestocks. In all packages tested the Group III basestock showed very comparable
oxidative performance to the synthetic Group IV basestock (Figure 1).
Table 5 shows a more exhaustive comparison of the Group III and Group IV
basestocks in commercial circulating oil additive package A (CIRC OIL A). Again the
Group III basestock showed comparable oxidative performance to the Group IV
basestock on all tests.
Additive Selection for Turbine Oils
Additive selection is an integral part of formulating a high temperature turbine oil

using Group II and higher basestocks. The additive system must be carefully balanced,
be of low volatility, and have good thermal stability, but must also have adequate
solubility in the basestocks employed in the turbine oil formulation. This last
requirement can be one of the greatest challenges facing the formulator as performance
requirements b e c o m e m o r e d e m a n d i n g necessitating higher treat rates and/or different

additive chemistries.
Antioxidant Balance Key to Optimized Performance

RPVOT Oxidation(Duplicate),Min @ 150~
1800
m
Group III
1600
[---I Group IV
1400
1200
i000
800
600
400
200
0
HydraulicOil HydraulicOil CirculationOil CirculatingOil CompressorOil CompressorOil CompressorOil
PackageA PackageB PackageA PackageB PackageB PackageC PackageD
Figure 1. RP VOT Performance Comparison Using Various Commercial Additive

Packages (Henderson et al [13] with Permission)
Table 5. Commercial Synthetic Circulating Oil Additive Package

(Henderson et al [13] with permission)
Finished Oil Properties Group III Group IV
Kinematic Viscosity@40~ cSt 32.5 32.0

Viscosity Index 134 137
Pour Point, ~ -39 <-51
RPVOT @ 150~ Min. 1628 1492
Oxidation Stability (ASTM D-4310)
mg Sludge 31.6 791.2
CIGRE Oxidation (IP280)
Volatile Acidity, mg KOH/g 0.02 0.03
Sludge, % 0.10 0.06
Top. % 0.12 0.07
Thermo-oxidative stability (FTMS 591C5308.6)
72 Hrs. @204~ 5L AIR/HR
Vis. Inc. @40~ % 2.0 2.4
Sludge, mg/100 mL 202.4 221.7
Solids, % <0,01 <0.01
IP48 Oxidation (170~ 48HR)
Vis. Inc. @40~ % 2.8 1.7
A TAN, mg KOH/g 0.1 0.1
Wt. Loss% 3.1 1.4
Solids, % 0.03 0.03
Additive balance encompasses the concepts of additive interaction and reaction, and
optimization to take advantage of additive synergies in order to minimize treat rates.
While minimizing the additive treat rates required to achieve a given performance level
makes good economic sense, it also has other important benefits. Low treat rates help
alleviate additive solubility concerns with these highly non-polar basestoeks, which have
low aromatic levels and high aniline points. Using high treat rates of antioxidants to
achieve oxidative performance can lead to problems with sludge and deposit formation.
Current Trends
Traditional turbine lubricants consist of solvent refined oil and 2,6-di-tert-butyl-para-

cresol (DBPC), or DBPC plus an amine. This chemistry, however, cannot meet modem
high temperature gas turbine demands due to its high volatility. It has been noted that
DBPC, even in lower temperature steam turbine service, is largely consumed through
evaporation [14]. The inadequacy of DBPC based formulations is highlighted by Smith
[3], who presents data on a dozen commercial DBPC containing turbine oils, which failed
after short periods of service (< 1 year) in high temperature gas turbines. The average
bearing housing temperature in the turbines was 300~ with a range of 289~ to 311 ~
Some OEM turbine oil specifications make explicit reference to low volatility
phenolic antioxidants such as DBPC (e.g. GEK 32568), prohibiting or warning against
their use. The specifications include tests designed to preclude the use of formulations
employing volatile antioxidants. One example is the inclusion of a modified RPVOT
(Rotating Pressure Vessel Oxidation Test) test in which before being tested, the lubricant
is first stripped of any volatile additives by heating it to 121~ and blowing it with
nitrogen for 48 hours. Low volatility necessitates the use of high molecular weight
antioxidants, which typically have limited solubility in hydrocracked basestocks.
Rasberger [9] presents data highlighting the improvements that can be achieved in
turbine lubricants using new generation lower volatility difunctional antioxidants. It is
not clear, however, how strong a role decreased volatility plays.
Antioxidants
It is well known that combinations of hindered phenols and alkylated diphenylamines

have synergistic antioxidant effects. This synergy, however, is dependent on the
characteristics of the basestocks and the oxidative environment employed. The optimal
phenol/amine ratio employed in one type of basestock may not be optimal for another.
Simply taking an off the shelf additive package designed and tested for traditional solvent
refined oils and employing it in Group II and higher basestocks is likely to have sub-
optimal results. While Group II and higher basestocks will almost certainly yield
superior performance to traditional solvent refined oils using a given package, balancing
the phenol/amine ratio using individual components will yield superior results [10].
Attempting to use an off the shelf additive package also raises the distinct possibility of
incurring additive solubility problems.
Further optimizations of oxidative performance can be achieved by the inclusion of a
different class of antioxidant. Hindered phenols and alkylated diphenylamines both
represent free radical scavengers. Peroxide decomposers (traditionally organosulfur and
organophosphorus compounds) such as phosphites and selected sulfur compounds
represent this other class of antioxidant.
As noted previously, the oxidative stability of unadditized base oils is dependent on
their aromatic and sulfur contents, with solvent refined base oils outperforming PAOs.
Practical experience in the laboratory has shown that the sulfur content of additized base
oils is also very important. While the oxidative stability of additized PAOs and
hydrocracked base oils is far superior to solvent refined base oils, it can be further
optimized with the inclusion of a carefully selected synthetic sulfur carrier.
Gatto and Grina [I0] observed strong synergies between hindered phenols and
diphenylamines when hindered phenols containing sulfur and aliphatic amines were used.
These functional groups within the hindered phenols act as peroxide decomposers. The
sulfur containing hindered phenol, however, did not have any significant effect in solvent
refined basestocks which were also tested. Gatto and Grina also noted an extremely
strong synergism when a synthetic sulfur source was used in combination with an
optimized phenol/amine ratio. The sulfur carrier mimics the role of naturally occurring
sulfur found in less highly refined base oils which acts as a peroxide decomposer.
Phosphorus type chemistry can also effectively fill this role as shown in a patent [15],
which deals with the stabilization of hydrocracked basestocks and PAOs using a hindered
phenol and a phosphite. In this work it was noted that while the antioxidants offered
significant benefits in hydrocracked basestocks and PAOs, no significant improvement in
oxidative stability was realized when the antioxidants were employed in a solvent refined
basestock.
Testing and Materials
The oxidative performance of three severely hydrocracked basestocks was compared

to that of a PAO using RPVOT [ASTM Standard Test Method for Oxidation Stability of
Steam Turbine Oils by Rotating Pressure Vessel (D 2272)], and pressurized differential
scanning calorimetry (PDSC). The additive systems employed were a commercially
available turbine oil package, and two experimental antioxidant systems which exhibit
good solubility in the basestocks tested.
RPVOT test conditions used were in accordance with ASTM D 2272: 150~ 90 psi
oxygen, 50 g sample, 10 mL water, and a polished copper coil catalyst.
PDSC test conditions were 500 psi oxygen at 100 mL/min, isothermal operation at
175~ and 190~ with a 10~ initial temperature ramp, 1.5 mg sample, and open
aluminum pans.
The basestocks employed in this study were representative of Group II through Group
III basestocks. Their basic physical properties are shown in Table 6.
Table 6. Basestock Properties

Property Basestock 1 Basestock 2 Basestock 3 Basestock 4
API Category Group II Group III ~ Group III Group IV

Viscosity @ 40~ cSt 34.02 31.36 32.48 30,48
Viscosity @ 100~ cSt 5.52 5.76 5.99 5.83
Viscosity Index 97 127 132 138
Density @ 15~ kg / L 0.8674 0.8398 0.8372 0.8270
Blend of Group II and Group III basestocks.
Basestocks 2 and 3 represent blends of 100 Neutral Group II+ and Group III
basestocks, respectively, with a 325 Neutral Group III basestock. Group II+ refers to
Group II basestocks with a viscosity index close to the Group II upper limit of 120. An
ISO 32 viscosity grade (28.8-35.2 cSt at 40~ was chosen as this is a commonly used
viscosity grade for turbine oils. A chemical analysis of the basestocks as well as details
on the refining and dewaxing methods used in their manufacture is given in Table 7.
Table 7. Basestock Chemical Characteristics

Basestock 1 2 3 4
API Classification II III (blend) III IV
Description Severely Severely Severely PAO

Hydro- Cracked Hydro-Cracked Hydro-Cracked
Dewaxing Method Solvent lso lso None
Mass Spec Analysis

Paraffins, n & iso- 21.7 52.9 59.4 100.0
Cycloparaffins 78.2 47.1 40.6 --
Mono-Cycloparaffins 26.9 25.1 23.3 --
Di-Cycloparaffins 20.3 12 10.6 --
Tri-Cycloparaffins 14.9 6 4.2 --
Tetra-Cycloparaffins 10.6 2.5 2.2 --
Penta-Cycloparaffins 4.1 1.4 0.2 --
Hexa-Cycloparaffins 1.3 0.1 . . . .
l-lepta-Cycloparaffins 0.1 -- 0.1 --
Aromatics 0.1 . . . . . .
Paraffins + Mono-Cycloparaffins 48.6 78.0 82.7 100.0

Poly-Cycloparaffins 51.3 22.0 17.3 0
The severity of the hydrocracking process used in the production of these basestocks
results in negligible aromatic, sulfur, nitrogen and oxygen content. Moving from solvent
dewaxing to catalytic iso-dewaxing increases the content of the more desirable iso-
paraffins and decreases the poly-cycloparaffin content, bringing the chemical

composition closer to that of a polyalphaolefin.
Results and Discussion
A commercially available turbine oil additive package was evaluated using basestocks
1 through 4 using RPVOT and PDSC. TOST (Turbine Oil Stability Test), the ASTM
Standard Test Method for Oxidation Characteristics of Inhibited Mineral Oils (D 943),
was not performed in this study due to time constraints. Previous experience with the
commercial turbine package shows TOST performance greater than 5,000 hours in Group
II basestocks. Figure 2 shows increasing RPVOT stability as we move from basestock 1
to basestock 4. PDSC oxidation induction times at 175~ showed a similar trend (Figure
3).
R P V O T - A S T M D2272
1000
900
.=
.E
800
E
700 630 645
._E 585
p- 600 540
t-
O 500
"O
400
_=
c- 300
o
o~
t~ 200
>r
O 100
0
1 - Group II 2 - Group III (blend) 3 , Group III 4 - Group IV
Basestoek
Figure 2. RPVOT Performance Using Commercial Circulating Oil Additive Package
Both the RPVOT and DSC performance of basestock 3 (Group III) was closest to that
of basestock 4 (PAO), as expected. Figure 4 compares the observed oxidative stabilities
with the poly-cycloparaffin content. The oxidation induction times increase as the poly-
cycloparaffin content decreases.
Many low volatility antioxidant additives are solids at room temperature, and have
poor solubility in severely hydrocracked basestocks. They often require high
IRVINE ON G R O U P I1 A N D G R O U P Ill B A S E S T O C K S 65
temperatures in order to be dissolved, and are difficult to handle in manufacturing

environments. In this work, a low volatility antioxidant system was developed which
exhibited good solubility in severely hydrocracked basestocks.
PDSC at 175"C
50
~" 45 43.1
40.3
~ 40 36.7
E 34.7
~ 35
E
~. 30
e-
o 25
" 20
III
,- 15
~O
~ 10
x 5
0
1 - Group I1 2 - Group III (blend) 3 - Group III 4 - Group IV
Basestock
Figure 3. PDSC Performance at 175~ - Commercial Turbine Oil Additive Package
When the new antioxidant system was examined in the same series of basestocks as
the commercially available turbine package, significantly improved RPVOT response
was observed (Figure 5). This significant improvement in oxidative performance is
confirmed via PDSC at 190~ (Figure 6). Interestingly, while the PDSC results showed
the same trend as with the commercial turbine package, the RPVOT results did not. In
this case, basestock 3 (Group III) had superior performance to basestock 4 (PAO). As
before, TOST performance was not evaluated for the new antioxidant system, but based
on previous experience with similar chemistries this additive system is expected to yield
results greater than 15000 hours.
Further work on optimization of the sulfur carrier contained in the new antioxidant
system produced a slight increase in PDSC oxidation induction times for the Group II and
Group III(blend) basestocks. No significant effect, however, was observed in the Group
III and IV basestocks (Figure 7).
RPVOT testing with this modified antioxidant system was inconclusive. A clear
oxidative induction breakpoint w ~ only observed for the Group II basestock. In this
case, a dramatic increase in RPVOT performance was observed, with the oxidation
induction time going from the previous 2160 minutes to 3300 minutes.
66 TURBINE LUBRICATION IN THE 21 ST C E N T U R Y
660 44.0
ORBdT~
9 DSC !
640 ! 42.0
[]
r
E 620 . . . . . . . . . . . . . . . . . 40.0 'EE--
600 38.0 =i
58o 36.0 ~
560 34.0
540 . . . . . . . . . . . . . . . . . . . . . . . . . . . . [] . . . . 32.0
520 30.0
10 20 30 40 50 60
Poly-Cycloparaffin Content (wt%)
Figure 4. Oxidation Induction Times versus Poly-Cycloparaffin Content
RPVOT - ASTM D 2272
4000
3500 3300
._= 2940 2910
g 3000
r
E 2500
i= 2160
c-
O 2000
r
,-3
"o
t-
1500
c-
O 1000
"o
500
o
0
1 - G r o u p II 2 - G r o u p Ill (blend) 3 - G r o u p III 4 - G r o u p IV
Basestock
Figure 5. RPVOT Performance Using Antioxidant System 1

I R V I N E O N G R O U P 11A N D G R O U P Ill B A S E S T O C K S 67
PDSC at 190~
70
60 57.5
55.9
,=1 52.3
47.9
g 50
G)
E
I- 40
O
30
e-,
m
C
O 20
X 10
o
0
1 - G r o u p II 2 - G r o u p III (blend) 3 - G r o u p III 4 - G r o u p IV
Basestock
Figure 6. PDSC Performance Using Antioxidant System l
PDSC at 190~
70
6O
e=
"~ 50
.E
i- 40
e-
0
N
"0
3o
~ 2o
x 10
o
1 - Group II 2 - Group III (blend) 3 - Group III 4 - Group IV

Basestock
Figure 7. PDSC Performance Using Antioxidant System 2
As with the commercial turbine additive, the new antioxidant systems showed the
same general trend of increasing oxidation induction times with decreasing poly-
cycloparaffin content (Figure 8).
3500 , ............................................ ~ 60
A 3350 , . . . . . . . . . . . . . . . . . . . . . . . . . j X RPv6~-AO ~- i -ff

X inPDSC-AO1 l
3200 ,,I .............. ~ !&PDSC-AO2 i- 57 "~
E
3050 i " I i-
e-
O x
2900 54 .s
,&
I
_= 2750 ! - D-
m
c
C
O
2600 51 , mo
2450
O O
I- I
O 2300 48 o
> e~
a.
2150~ ; I a.
ne ........... •
i 7
2000 ; .......... ~ . . . . . . . . . ~ . . . . . . -~ . . . . . . . ~- ...... = ~ 45
0 10 20 30 40 50 60
Poly-Cycloparaffin Content (wt%)
Figure 8. Oxidation Induction Times versus Poly-Cycloparaffin Content for New

Antioxidant Systems
Conclusions
The demands of modem high temperature gas turbine applications require thermal and
oxidative stability performance beyond the capabilities of previous generation turbine
oils. Modem turbine lubricants require highly refined basestocks and carefully selected,
low volatility, thermally stable additive systems.
Highly refined basestocks, with very low aromatic and hetero-atom content, have been
shown to have superior oxidative stability to previous generation solvent refined oils,
when additized. Oxidative stability increases and solvency decreases as total aromatic
content decreases. When formulating with these highly non-polar basestocks one must be
cognizant of their decreased solvency.
This poorer solvency must be assessed in terms of additive and oxidation product
solubility. The formulator must carefully screen additive systems and must set firm
condemning limits for used fluids.
Condemning limits for turbine oils formulated with these basestocks should not be
exceeded. New-generation oils will take a long time to reach their condemning limits,
IRVINEON GROUPIIAND GROUPIIIBASESTOCKS 69
but once they do reach them, action must be taken. The highly non-polar nature of these
fluids means that they have very limited solubility and dispersing power for polar
oxidation products. These solvency issues are not new, however, and have been faced by
formulators using PAOs.
Among highly refined Group II and Group III basestocks, with very low hetero-atom
and aromatic content, poly-cycloparaffin content is a key indicator of oxidative
performance. Oxidative stability of additized basestocks increases as poly-cycloparaffin
content decreases. Group III basestocks with low poly-cycloparaffin content offer
oxidative stability on par with PAOs.
It was also observed that the use of sulfur-based peroxide decomposers can yield
significant improvements in the oxidative stability of hydrocracked basestocks.
The use of severely hydrocracked iso-dewaxed basestocks together with thermally
stable, low volatility, additive systems is a clear method of meeting modem gas turbine
lubrication requirements.
References
[1] Smith, A. N., "Heavy-Duty Gas Turbines," CRC Handbook of Lubrication (Theory
and Practice of Tribology), Volume 1 Application and Maintenance, E. R. Booser,
Ed., CRC Press, Boca Raton, 1983, pp. 107-119.
[2] Herder, M. J., "Mineral Oil Lubrication of Large Gas Turbines," Journal of the
American Society of Lubrication Engineers, 1977, Vol. 33, No. 6, pp. 303-307.
[3] Smith, A. N., "Turbine Lubricant Oxidation: Testing, Experience, and Prediction,"
Aspects of Lubricant Oxidation, ASTM STP 916, W. H. Stadtmiller and A.N.
Smith, Eds., American Society for Testing and Materials, Philadelphia, 1986, pp.
1-26.
[4] Samman, N., and Ashman, L. A., "Severely Hydrotreated and Iso-Dewaxed Specialty
Basestocks," National Lubricating Grease Institute Annual Meeting, 1999.
[5] Mack, P. D., Zaugg, T. D., and Cohen, S. C., "Severely Hydro-Cracked/Hydro-
Isomerized Base Oils - The Key to Enhanced Performance Levels," 1996
National Petroleum Refiners Association, National Fuels and Lubricants Meeting,
1996, FL-96-114.
[6] Deckman, D. E., Lohuis, J. R., and Murphy, W. R., "HDP Base Stocks for Industrial
Lubricants," Hart's Lubricants World, 1997, pp. 20-26.
[7] Igarashi, J., Yoshida, T., and Watanabe, H., "Concept of Optimal Aromaticity in Base
Oil Oxidative Stability Revisited," American Chemical Society, Div. Pet. Chem.,
Preprints, 1997, 42, 1, pp. 211-217.
[8] Cemy, J., Stmad, Z., Pospisil, M., Sebor, G., Vaclavickova, I., and Vinklarkova, N.,
"Oxidation Stability of Hydrocracked Base Oils," Tribology 2000 - Plus, 12th
International Colloquium, W. J. Bartz, Ed., 2000, Vol. 1, pp. 313-319.
[9] Rasberger, M., "Oxidative Degradation and Stabilization of Mineral Oil Based
Lubricants," Chemistry and Technology of Lubricants, R. M. Mortier and S. T.
Orszulik, Eds., VCH Publishers Inc., New York, 1992, pp. 98-142.
70 TURBINELUBRICATION IN THE 21sr CENTURY
[10] Gatto, V. J., and Grina, M. A., "Effects of Base Oil Type, Oxidation Test Conditions
and Phenolic Antioxidant Structure on the Detection and Magnitude of Hindered
Phenol/Diphenylamine Synergism," Lubrication Engineering, 1999, Vol. 55., No.
1, pp. 11-20.
[11 ] Galiano-Roth, A. S., and Page, N. M., "Effect of Hydroprocessing on Lubricant
Base Stock Composition and Product Performance," Lubrication Engineering,
1994, Vol. 50, No. 8, pp. 659-664.
[ 12] Gatto, V. J., Grina, M. A., and Ryan, H. T., "The Influence of Chemical Structure on
the Physical and Performance Properties of Hydrocracked Basestocks and
Polyalphaolefins, Tribology 2000 - Plus, 12tnlnternational Colloquium, W. J.
Bartz, Ed., 2000, Vol. 1, pp. 295-304.
[ 13] Henderson, H. E., and Legzdins, A., "Evaluation of Severely Hydro-treated and
Isodewaxed Specialty Base Fluids in Industrial Oil Applications," 54 th Annual
Meeting, Society of Tribologists and Lubrication Engineers, Las Vegas, Nevada,
1999.
[ 14] Yoshida, T., and Igarashi, J., "Consumption of Antioxidant of Turbine Oil in Service
Unit," Tribology Transactions, 1991, Vol. 32, No. 1, pp. 51-58.
[15] Cohen, S. C., U.S. Patent 4,652,385, 24, Mar. 1987.
Bruce P. Schwager, t Bryant J. Hardy, l and Gaston A. Aguilar I
Improved Response of Turbine Oils Based on Group II Hydrocracked Base Oils

Compared with Those Based on Solvent Refined Base Oils
Reference: Schwager, B. P., Hardy, B. J., and Aguilar, G. A., "Improved Response of
Turbine Oils Based on Group II Hydrocracked Base Oils Compared with Those
Based on Solvent Ref'med Base Oils," Turbine Lubrication in the 21st Century, ASTM
STP 1407, W. R. Herguth and T. M. Wame, Eds., American Society for Testing and
Abstract: Modem industrial equipment makes increasingly severe demands on the

lubricating oils used to keep the equipment running. Higher operating temperatures,
smaller oil reservoirs, and decreased downtime all help create a severe environment for
the lubricant. In response, base oil manufacturers have developed new technology to
provide more saturated, high-performance base oils using catalytic hydrocracking
technology. Lubricants based on these base oils and properly matched with the right
additive technology have longer oxidation life, less deposit forming tendencies, and
overall higher performance than solvent refined products.
In this paper, the performance properties of lubricants based on hydrocracked Group II
base oils are compared with those of more traditional products based on solvent refined
base stocks. Laboratory tests such as the Rotating Pressure Vessel Oxidation Test (ASTM
D2272), the Turbine Oil Oxidation Stability Test (ASTM D943), and the Water
Separability Test (ASTM D1401) were used to predict performance. The lubricants tested
were all from the general category of rust and oxidation inhibited oils (R&O oils), and
included turbine oils. In some cases, actual performance was verified in field trials. The
data presented confirm the performance advantages of a properly formulated lubricant
based on Group II hydrocracked stocks.
Keywords: turbine oil, R&O oil, hydrocracked, power generation, performance
Introduction
Lubricate the moving parts. Cool hot spots. Remove contamination. Break out water.
Prevent metal-to-metal contact. Avoid varnish and sludge. Remain fluid at low
1 Conoco Inc., Lubricants R&D, P. O. Box 1267, Ponca City, OK 74602-1267.

71
Copyright92001 by ASTMlntcrnational www.astm.org

temperatures... The job description of a lubricant in a turbine system is very demanding.

Conventional lubricants formulated with solvent refined base oils have done an effective
job of meeting the challenge over the years. However, changing conditions in the power
generation industry are making the demands even tougher.
Cogeneration plants and combined-cycle units add more heat to the system, requiring
a lubricant that can stand up to hot spots in excess of 150~ (300~ Oil reservoirs are
smaller, increasing the stress on the lubricant. Control systems are sensitive to even the
smallest amounts of contamination or varnish. Plants are often called upon to start up
quickly in response to increased demand. When this happens and it is cold outside, the
lubricant must be fluid enough to flow to the bearings and other critical moving parts [1].
Lubricants formulated with hydrocracked base oils have what it takes to respond to
these increased demands. In the traditional solvent refining process, petroleum fractions
are "washed" with solvents to remove impurities, leaving behind the desirable lubricant
fractions. Due to the nature of the process, 10 - 15 percent impurities are often left in
solvent refined oil, in the form of aromatic compounds and heteroatoms, such as sulfur
and nitrogen. In the hydrocracking process, the petroleum fractions are reacted with
hydrogen under high pressure and temperature. The aromatic compounds are chemically
reduced to cycloparaffins and rearranged to form linear paraffins. The heteroatoms are
also removed in the process. The result is a lubricant base oil that is greater than 99
percent pure saturated paraffins (Table 1) [2].
Table 1 - Properties of Light Neutral Base Oils
Solvent Refined Hydrocracked

100N Base Oil ! 00N Base Oil
Clay Gel Analysis
(ASTM D2007)
- Saturates, wt. % 85 - 90 > 99
- Aromatics, wt. % 9-15 <1
- Polars, wt. % 0-1 0
Sulfur, wt. % 0.05 - 0.11 <0.001
Nitrogen, ppm 20 - 50 <2
Color (ASTM D1500) 0.5 - 1.0 <0.5
Several early studies have shown that some aromatic compounds and sulfur-
containing material can act as "natural" antioxidants [3-6]. However, these studies looked
at base oils alone, in the absence of any additives. Saturated hydrocarbons respond
dramatically to the addition of primary antioxidants, such as amines and phenols, which
interrupt the chain reaction that leads to base oil oxidation [7]. Aromatic compounds
interfere with this mechanism and reduce the effectiveness of the additives [8]. Sulfur and
aromatic compounds have also been shown to cause the formation of insoluble material
in the oil, which leads to sludge or varnish [9]. Basic nitrogen compounds, too, have been
shown to accelerate oxidation and increase the amount of sludge [10]. Overall,
SCHWAGER ET AL. ON GROUP II HYDROCRACKED BASE OILS 73
hydrocracked base oils result in formulated products that are much more resistant to
oxidation and the problems it causes.
Aromatic compounds and non-hydrocarbons such as sulfur and nitrogen-containing
materials are all surface-active compounds, compared to saturated hydrocarbons. These
surface-active compounds degrade the demulsibility of the base oil, so that water takes
longer to separate from the lubricant. The same surface-active compounds also degrade
the air-release time of the oil. Both of these properties are important to the performance
of turbine oil [11].
The hydrocracking process is often paired with catalytic dewaxing, which uses a
catalyst to remove any paraffin wax from the oil. This process is also called
hydrodewaxing. Because this process is more effective and more selective in removing
paraffin wax, hydrodewaxed oils have improved low temperature properties [12].
In short, hydrocracked base oils give the lubricant the extra boost they need to meet
the challenges of today's power systems. This paper will present several examples
demonstrating the performance advantages of lubricants formulated with hydrocracked
base oils.
R&O Oils
Rust and oxidation inhibited oils (R&O Oils) are frequently used in less demanding
applications in power plants. These applications can include steam turbines, air
compressors, pumps and gearboxes. The data below (Table 2) shows typical properties
for two general-purpose R&O oils. Both products are made with the same R&O additive
package, at the same treat rate. The first product is made with solvent refined base oil and
is a commercial product. The additive package was optimized for use in this base oil. The
second product was made by switching to hydrocracked base oil, without optimizing the
additive package. A simple base oil substitution led to the improved properties shown in
the table.
Table 2 - Properties of R&O Oils (ISO 32 Viscosity Grade)
Solvent Refined Hydrocraeked

Viscosity, cSt at 40~ 33 31
Viscosity, cSt at 100~ 5.4 5.4
Pour Point, ~ -34 -40
Color (ASTM D1500) 1.0 0.5
Oxidation Control (ASTM D943), hours 2 500 4 000
RBOT Life (ASTM D2272), minutes 530 750
Rust Test (ASTM D665 A&B) Pass Pass
Emulsion Characteristics (ASTM D 1401)
- minutes to 40-40-0 10 5
The results from both the Standard Test Method for Oxidation Characteristics of
Inhibited Mineral Oils (ASTM D943) and the Standard Test Method for Oxidation
Stability of Steam Turbine Oils by Rotating Pressure Vessel (ASTM D2272) illustrate the
improved oxidation resistance of the product formulated with a hydrocracked base oil.
This increase may translate into a longer life for the hydrocracked product. The pour
point for the hydrocracked product is lower than that of the solvent refined product,
indicating better low temperature performance, even though 25 percent less pour point
depressant is used in the hydrocracked formulation. This product also sheds water faster,
as demonstrated by the results from the Standard Test Method for Water Separability of
Petroleum Oils and Synthetic Fluids (ASTM D1401).
A lower viscosity grade R&O oil formulated for a special application highlights the
improvement in expected life that can be obtained from hydrocracked base oils. The
typical properties of the product formulated with both solvent refined and hydrocracked
base oils (Table 3) show the same advantages discussed previously.
Table 3 -Properties of R&O Oils (ISO 15 Viscosity Grade)

Color (ASTM D1500) 1.0 0.5
Oxidation Control (ASTM D943), hours - 12 000
- minutes to 40-40-0 10 5
Both products were also tested in a proprietary bubbling tube oxidation test run at
95~ This test was developed to simulate the conditions the product experiences in
actual use. The depletion of the antioxidant is monitored with infrared spectroscopy. The
data in Figure 1 show the additive is depleted much more quickly in the solvent refined
base oil. Again, this should translate directly into longer life for the hydrocracked
product.
Steam Turbine Oil
The previous examples showed the benefits from switching to hydrocracked base oils,
while using the same additive package. Even greater advantages can be obtained by
optimizing the additive package to take advantage of the unique properties of
hydrocracked base oils.
Steam turbine oils are special grades of R&O Oils, formulated to give better oxidation
resistance and longer life in a steam turbine. ASTM D943 is the most common test used
to predict the life of these products. Water shedding ability as measured in D1401 is also
important.
Figure 1 - Loss of Antioxidant in Bubbling Tube Oxidation Test at 95~
Properties of a high quality solvent refined base oil are shown in Table 4. Also shown
are the properties of a turbine oil designed with a hydrocracked base oil. Both oils are
commercial products, optimized for the specific base oil being used.
Table 4 - Properties of Steam Turbine Oils (ISO 32 Viscosity Grade)

Color (ASTM D1500) 1.0 0.5
- minutes to 40-40-0 10 5
The dramatic increase in the D943 life points out the improved life expectancy that can
be obtained by formulating a product to take advantage ofhydrocracked base oils. The
hydrocracked product also does a better job of shedding the water which is present in
steam turbine systems.
Natural Gas Turbine Oil
Gas turbines expose the lubricant to much higher temperatures than in steam turbines.
The oil must be able to withstand these hot spots without oxidizing or forming varnish or
sludge. The high purity of hydrocraeked base oils is a natural fit for this application. A
natural gas turbine oil optimized for hydrocracked base oil can provide extremely long
life in this application (Table 5). The solvent refined oil in this example is already a
premium commercial product compared to most turbine oils, yet the hydrocracked
product provides nearly four times the life in the D943 test. (By reducing the sampling
frequency, the test was carried out beyond the 10,000-hour limit of the D943 procedure to
measure the extent of the performance improvement.)
Table 5 - Properties of Natural Gas Turbine Oils (ISO 32 ViscosityGrade)

Color (ASTM D1500) 0.5 0.5
RBOT Life (ASTM D2272), minutes 900 1 900
- minutes to 40-40-0 10 5
Both of these products were used in a GE 7EA gas turbine that is part of a
cogeneration plant in south Texas. Oil samples are taken monthly and checked for
performance. D2272 is used as the primary means to evaluate remaining oxidation
resistance. Figure 2 shows the RBOT results for both products. The premium
performance and the extended life of the hydrocracked product are obvious.
Conclusion
The high purity of hydrocracked base oils (reduced aromatics and heteroatoms) leads
to some unique properties - they are more thermally and oxidatively stable when
combined with appropriate additives, they are more fluid at low temperatures, they shed
water more quickly, and they respond better to treatment with additives. By taking
advantage of these properties, products can be formulated that provide premium

performance to turbine systems. R&O oils, steam turbine oils and natural gas turbine oils
can all take advantage ofhydrocracked technology and improve the efficiency of plant
operations.
Figure 2 - Natural Gas Turbine Oils in Service
References
[1] Schimmoller, B. K., "Hydrocracked Lubricants Rival Synthetic Oils," Power

Engineering, April, 1999.
[2] Gaw, W. J., Black, E. D., and Hardy, B. J., "Finished Lubricant Benefits from Higher
Quality Base Oils," presented at Hart's World Conference on Refining
Technology and Reformulated Fuels, March, 1997, available from the author.
[3] Denison, D. H. Jr., "Oxidation of Lubricating Oils: Effect of Natural Sulfur

Compounds and Peroxides," Industrial and Engineering Chemistry, 1944, Vol.
36, No. 5, pp. 477-482.
[4] Burn, A. J. and Greig, G., "Optimum Aromaticity in Lubricating Oil Oxidation,"
Journal of the Institute of Petroleum, 1972, Vol. 58, No. 564, pp. 346-350.
[5] Korcek, S. and Jensen, R. K., "Relation Between Base Oil Composition and
Oxidation Stability at Increased Temperatures," ASLE Transactions, 1975, Vol.
19, No. 2, pp. 83-94.
[6] Cranton, G. E., "Composition and Oxidation of Petroleum Fractions,"

Thermochemica Acta, 1976, Vol. 14, pp. 201-208.
[7] Murray, D. W., MacDonald, J. M., White, A. M., and Wright, P. G., "The Effect of
Basestock Composition on Lubricant Oxidation Performance," Petroleum Review,
February 1982, pp. 36-40.
[8] Hsu, S. M., Ku, C. S., and Lin, R. S., "Relationship Between Lubricating Basestock
Composition and the Effects of Additives on Oxidation Stability," SAE Paper
821237, 1982.
[9] Maleville, X., Faure, D., Legros, A., Hipeaux, J. C., "Oxidation of Mineral Base Oils
of Petroleum Origin: The Relationship Between Chemical Composition,
Thickening, and Composition of Degradation Products," Lubrication Science, 9-1,
November 1996.
[10] Yoshida, T., Igarashi, J., Watanabe, H., Stipanovic, A. J., Thiel, C. Y., and
Firrnstone, G. P., "The Impact of Basic Nitrogen Compounds on the Oxidative
and Thermal Stability of Base Oils in Automotive and Industrial Applications,"
SAE Paper 981405, 1998.
[11] Abou E1Naga, H. H., Bendary, S. A., and Salem, A. E. M., "Industrial Oil
Performance in Terms of Base Oil Chemical Structure," Lubrication Science, 3-1,
October 1990.
[12] Rowe, C. N. and Murphy, J. A., "Low-Temperature Performance Advantages for

Oils Using Hydrodewaxed Base Stocks," SAE Paper 831715, 1983.
Mark E. Okazaki ~and Susan E. Militante 2
Performance Advantages of Turbine Oils Formulated with Group II Base Oils
Reference: Okazaki, M. E. and Militante, S. E., "Performance Advantages of

Turbine Oils Formulated with Group II Base Oils," Turbine Lubrication in the 21~t
Century, ASTMSTP 1407, W. R. Herguth and T. M. Warne, Eds., American Society for
Testing and Materials, West Conshohocken, PA, 2001.
Abstract: Formulation of high-quality turbine oils with API Group II base oils provides
finished turbine oils with numerous performance advantages. When blended with the
proper additives, these high-purity base oils provide turbine oils with outstanding
oxidation stability, lower sludging tendencies, and improved water separability
properties. The exceptional feature of these Group II finished products is their
outstanding oxidation stability. The outstanding oxidation stability of the Group II
lubricants is due not only to the high-purity base oils, but also to the carefully chosen
additive package. The inherent sludging tendency of turbine oils is reduced when
changing from Group I to Group II oils. Also, the water separability of these oils is
improved when making the same change. Data which highlight these striking
performance differences between Group I and Group I/finished turbine oils and field
test data will be shown.
Keywords: turbine oil, turbine oil additives, lubricant, Group II, oxidation stability,
RPVOT, TOST
Base Oils
The American Petroleum Institute (API) has classified base oils [1] according to
their viscosity index and sulfur and saturates content. As seen in Table 1, Group I oils
have sulfur concentrations above 300 ppm (0.03 %) and saturates content of less than
90% while Group II oils have sulfur concentration of less than 300 ppm and saturates
level of more than 90%. The saturates content is defined as the amount of paraffins and
it can be calculated by subtracting the aromatics percentage from 100. For example, if
abase oil contained 5% aromatics, then the amount of saturates would be 95%. The
~Senior Research Chemist, Chevron Products Company, Lubrication Technology Group,

100 Chevron Way, Richmond, CA 94802.
2Chemist, Chevron Products Company, Lubrication Technology Group, 100 Chevron
Way, Richmond, CA 94802.
7.0

80 TURBINE LUBRICATION IN THE 21STCENTURY
primary difference between Group II and Group III oils is the higher viscosity index for
the Group III base oils. Poly-alpha-olefin base stocks are defined as Group IV oils and
all other base stocks not defined as Group I through IV are defined as Group V oils.
Table 1 - API Base Stock Classification
Base Oil Category Sulfur (%) Saturates (%) Viscosity Index

Group I >0.03 and/or 90 80 to 120
Group II <0.03 and/or >_90 80 to 120
Group III _<0.03 and/or >_90 >_120
Group IV All polyalphaolefins
Group V All others not included in Groups I, II, Ill or IV
Turbine Oils
Gas and steam turbines typically use lubricants in their systems as beating and gear
lubricants and as a hydraulic fluid. In most steam turbines, phosphate ester-based fluids
are used as the hydraulic fluid. Turbine OEMs (Original Equipment Manufacturers)
have lubricant specifications for turbine oils. All high quality turbine oils meet the
viscosity, load-carrying ability, oxidation stability, corrosion protection, foam resistance
and water separating ability tests. Among these tests, oxidation stability is among the
most important test specifications. A more oxidatively stable oil will provide longer oil
life for the end use customer by extending the drain interval.
The turbine OEMs typically use RPVOT (Rotating Pressure Vessel Oxidation Test,
formerly known as RBOT) (ASTM D2272) and TOST (Turbine Oil Stability Test)
(ASTM D943) as specification tests for turbine oils. The RPVOT test is a fairly high
temperature test (150~ which measures the relative oxidation stability of finished oils
and it usually takes about 1 day to complete. The OEM minimum specification for
RPVOT results range from 250 to 500 minutes. The TOST test (95~ also measures
the relative oxidation stability of oils, but it typically takes over one year to complete.
The OEM minimum specification for TOST ranges from 1000 to 4000 hours.
Outstanding Oxidation Stability
For turbine oils, the greatest advantage Group II finished oils provide over Group I
oils is superior oxidation stability. Severe hydroprocessing of feedstocks greatly
reduces the aromatic, sulfur and nitrogen-containing "impurities" to give virtually
impurity-free Group II base oils. These impurities are responsible for the lower
oxidation stability of Group I stocks because they readily oxidize. These impurities
oxidize to form benzylic, sulfuric, and nitric acids and other oxidized species. Group II
base oils typically contain only trace amounts of these impurities and the finished oils
can exhibit outstanding oxidation stability properties when formulated with the proper
OKAZAKI AND MILITANTE ON GROUP II BASE OILS 81
additives. The superior oxidation stability of Group II finished turbine oils is

demonstrated in their RPVOT and TOST values.
Table 2 - Use of One Additive Package in Different Base Oils
Test/Oil Method Base Oil BaseOil BaseOil BaseOil OEM

A B C D Specifications
Base Oil Type Group I Group I Group II Group II
Turbine Oil Stability ASTMD943 4256 6636 >10000 >10000 1000- 4000
Test, Hours minimum
Rotating Pressure ASTM D2272 980 1025 2295 1698 250-500

Vessel Oxidation Test, minimum
minutes
Oxidation and FHV179 lb 5308-7
Corrosion Test
Viscosity change, % 17.7 13.7 3.8 1.4
Acid number increase 2.8 1.47 0.98 0.06
1000 Hour TOST Test, ASTMD4310 37 36.1 25 20 100 max
Insolubles, mg
Water Separability, time ASTM D1401 20 15 10 10 30max
to 0 mL emulsion,
minutes
For gas turbines, such as the General Electric (GE) Frame 7 series, a turbine oil
with excellent oxidation stability is very desirable. The GE Frame 7 series has three
journal bearings with a center bearing that can be subjected to 500~ Since a turbine oil
also acts as a coolant, it will be subjected to high temperatures from the center bearing
and it must be oxidatively and thermally stable. A turbine oil with excellent oxidation
stability is necessary in many gas turbine applications.
Data in Table 2 show the effect of using one additive package in a number of
different base stocks. As can be seen, the RPVOT and TOST values for the Group I oils
are good; however, the Group II oils far exceed the values of the Group I oils and the
OEM specification minimums. The reasons for the superior oxidation stability of the
Group II finished oils are twofold. First, Group II oils have greatly reduced impurities
and this produces a base stock that is inherently more gable. Second, the use of
synergistic and/or complementary additives can create oils with extremely high
oxidation stability. In other words, formulation of Group II oils with the proper
additives is very important.
Another test that can be used to measure the relative oxidation stability of an oil is
FTM 791b 5308-7 (Federal Test Method). This method was used as a test in some older
specifications, but it is not presently used. However, because it measures the viscosity
change and acid number increase, the results will show the relative oxidation stability of
different turbine oils.
The FTM data indicates that the Group I finished oils had larger viscosity and acid
number increases than the Group II oils. An increase in both of these properties
indicates that oxidation has occurred. Hence, the Group I oils are less oxidatively stable
than the corresponding Group II oils with the same additive package.
Table 3 shows the effect of using different additive packages in one Group II base
stock. The total additive treat levels are within a range of 80 to 120% of a base case
level. As one can see, different additive packages have very different oxidation stability
results. This demonstrates the importance of how a finished oil is formulated. Use of a
Group II oil for any finished oil does not guarantee excellent oxidation stability results.
Table 3 - Effect of Different Additive Packages in the Same Group H Base Oil
Additive/Test RPVOT, minutes TOST, Hours

Additive Package E 705 5590
Additive Package F 1320 >10000
Additive Package G 735 7800
Additive Package H 494 8434
Additive Package I 1300 6590
Reduced Sludging Tendency

In conjunction with improved oxidation stability, Group II finished turbine oils
have reduced sludging tendency when compared against the Group I finished oils. The
sludging tendency of a turbine oil is an important performance feature. Sludging
tendency is not a turbine OEM specification, but it is a part of a Military specification
(MIL-L-17331H (SH)). A lower sludge weight will indicate a lower potential for
varnishing and deposit formation with a particular turbine oil. ASTM D4310 is a
variation of the TOST test (1000 hours) and it is a requirement for the MIL-L- 17331H
(SH) specification. After running the test for 1000 hours, the total amount of sludge is
measured. A lower sludge weight is a desirable feature and Group II finished turbine
oils provide lower sludge weights.
The use of Group II base stocks lowers the sludging tendency of finished turbine
oils. Data in Table 2 show the sludge weight (ASTM D4310 results) effect of using one
additive package in different base stocks. As can be seen, the Group I finished oils have
an overall higher sludge weight than the Group II oils.
The reduced sludging tendency of Group II finished turbine oils can be seen in a
couple of commercially available turbine oils. Turbine oils which use Group I, Group II
and Group IV (poly-alpha-olefins) were tested. The RPVOT method was used as a test
to measure the relative sludging tendency of an oil. A number of oils were subjected to
the RPVOT conditions for 600, 900 and 1200 minutes. After each time period, the acid
number of the oil and the filterable solids were measured. The solids were filtered onto
a pre-weighed filter and rinsed with heptane. The filter and solids were then oven dried
and weighed again to obtain the total amount of filterable solids. The data is
summarized in Table 4.
Table 4 - RPVOT SIudge Results of CommerciallJ Available Turbine Oils
600 Minutes 900 Minutes 1200 Minutes

Oil/Test Base Oil AN Sludge, AN Sludge, AN Sludge,
Type wt., nag wt., nag wt., mg
Commercial Oil J Group I 0.08 2 0.24 4 5.1 49.8
Commercial Oil K Group I 4.53 53.7
Commercial Oil L Group II 0.06 4.6 0.1 5.1 0.17 22.8
Commercial Oil M Group II 0.05 3.8 0.08 3.5 0.05 5.4
Commercial Oil N Group IV 0.64 19 0.92 59.8 0.69 117
l_ Based on 1998 North Americanformulations
The Group I and Group IV oils had higher sludging tendencies than the two Group
II finished oils. Because the end time for one of the Group I (Oil K) oils was less than
600 minutes, one can see the effect of oxidation. This oil had oxidized, had a very high
acid number and it produced a large amount of sludge. Both Group II finished oils had
relatively stable acid numbers and low sludge weights even at 1200 minutes. The
additive packages (Oils L and M) are not the same, but this illustrates the lower inherent
sludging tendencies of the Group 1I finished oils over the Group I products.
The Group IV finished oil exhibited poor results in both acid number and sludge
weights. Although the RPVOT end time for the Group IV oil is greater than 1200
minutes, the oil exhibited poor sludge tendency throughout the test and it had a
relatively high acid number, too. In this case, the high sludging tendency can be
explained by the formulator's choice of additives and base stocks.
Improved Water Separability
When changing from a Group I base oil to the Group II oil, the water separability
of the turbine oil is improved. The ability of the turbine oil to "shed" or separate water
is a very important property for steam turbines since water can leak into the oil
reservoir. Emulsions and/or water are very poor lubricants, so the oil must be able to
separate from water. The water separating characteristic of oils can be measured
(ASTM D1401). Data shows that Group II finished oils separate faster than the Group I
counterparts.
Data (ASTM D1401 results) in Table 2 shows the difference between Group I and
Group II finished oils which use the same additive package. On the whole, the times for
the Group II oils are shorter than the corresponding times for the Group I oils. Group II
base oils have better water separability response to this additive package than the
corresponding Group I oils. A shorter time is more desirable since the oil is separating
from water in a quicker manner.
Field Test Data
When introducing a new or reformulated turbine oil, most lubricant companies will
field test the product in a challenging application. Because of the hot center bearing
design, the General Electric Frame 7 gas turbine is a good choice to evaluate the
performance of a turbine oil. A Group II finished turbine oil was placed into a GE
Frame 7B/E peaking unit in North America. The turbine's previous oil was another
Group II product with good oxidation stability (RPVOT -- 350 minutes; TOST = 5000
hours). The previous oil was used and replaced by the plant after about 40000 hours.
The new oil was monitored over a long period of time and it is still being monitored.
The new oil was monitored for over 40000 hours and it has performed well with no
problems. A number of physical and performance properties of the oil were monitored
and the data is summarized in Table 5. Overall, the oil is still in very good condition
and suitable for continued use. As evidenced by the RPVOT value at 40353 hours, the
oil still has additives to protect against oxidation.
Color, acid number and RPVOT values are indicators of oil oxidation. Over time,
the field test oil's oxidation indicators were very stable. As an oil ages and approaches
the end of its useful life, it will turn darker in color. The darker color is due to oxidized
by-products from the additives or oil itself. The field test oil started with a light color
(L1.0) and increased to only L3.5 after over 40000 fired hours. Since the oil was not
oxidized, the color increase can be explained by the expected oxidation of the anti-
oxidants. The acid number of an oil will dramatically increase when the anti-oxidant
additives are depleted. The field test oil's acid number is almost a straight line over the
monitoring period. By itself, a stable acid number is an indication that the oil has not
started to oxidize and form acidic by-products.
The RPVOT method is a good monitoring test for in-service oils. When oils are
monitored on a regular and meaningful basis, one can essentially predict when the oil
should be drained and replaced. The field test oil's RPVOT values were tracked over
the 40,000+ hour time period. Over time, the oil maintained its oxidation stability and
the last RPVOT value (40353 hours) shows that the oil is still in good condition. ASTM
D4378 recommends consideration of oil changeout when an in-service oil has a RPVOT
value that is 25% of the fresh oil. The field test oil is still above that level and,
considering all of the other properties, it is in very good condition. Continued
monitoring of the oil's RPVOT will allow this customer to predict when the oil should
be changed.
The field test oil has performed very well with no reported varnishing, deposit and
wear issues. The Group II field test oil has already extended the oil life from about
40000 to well over 40000 fired hours. Continued monitoring of the oil's RPVOT will
allow further extension of this oil's useful life.
Table 5 - Group 11 Turbine Oilfrom a GE Frame 7B/E FieM Test
Fired Hours ASTM Method 0 500 1001 1500 2 5 0 0 3 5 0 0 4500
Color DI500 LI.0 L1.0 L1.0 1 LI.5 L 1 . 5 L1.5

Viscosity, 40C, cSt D445 30.04 30.18 30.21 30.22 30.22 30.24 30.24
Acid Number, mg D974 0.05 0 . 0 6 0 . 0 5 0.05 0.07 0 . 0 6 0.05
KOH/g
RPVOT, Minutes D2272 2020 2215 2 2 5 0 2275 2 2 6 8 1641 1497
Wear Metals) ppm ICP <5 <5 <5 <5 <5 <5 <5
Other Elements,2 ppm ICP <10 <10 <10 <10 <10 <10 <10
FiredHours 5500 6 5 0 0 7500 9500 11500 13500 40353

Color DI500 1.5 L2.0 L 2 . 0 L 2 . 0 L 2 . 0 L2.0 L3.5
Viscosity, 40C, cSt D445 30.3 30.5 30.59 3 0 . 5 30.51 30.55 32.06
Acid Number, mg D974 0.06 0 . 0 6 0 . 0 5 0 . 0 6 0 . 0 5 0.05 0.05
KOH/g
RPVOT, Minutes D2272 1538 1708 2163 1432 1652 1364 742
Wear Metals,~ppm ICP <5 <5 <5 <5 <5 <5 <5
Other Elements,2 ppm ICP <10 <10 <10 <10 <10 <10 <10
l . Copper, iron, tin
2. Aluminum, barium, boron, calcium, chromium, lead, magnesium, molybdenum, nickel, phosphorus,
silicon, sodium, titanium, zinc
Summary
Formulation o f high-quality turbine oils with API Group U base oils provides
finished turbine oils with numerous performance advantages. When blended with the
proper additives, these high-purity base oils provide turbine oils with outstanding
oxidation stability, lower sludging tendencies, and improved water separability
properties. The outstanding oxidation performance o f the Group II lubricants is due not
only to the high-purity base oils, but also to the carefully chosen additive package. With
the proper experience, a formulator can develop an oil with outstanding oxidation
stability. A decrease in relative sludging tendency and shorter water separability time o f
turbine oils are two added improvements when changing from Group I to Group II oils.
Lastly, laboratory tests notwithstanding, actual use o f a turbine oil in an application is
the ultimate test to prove its performance. A Group U finished turbine oil has been in
service for over 40000 hours and it is still giving satisfactory performance in the turbine.
References
[ 1] American Petroleum Institute Publication 1509.

Jo Ameye/Robert E. Kauffman 2
Antioxidant Analysis for Monitoring Remaining Useful Life of Turbine Fluids
Reference: Jo Ameye, Robert E. Kauffman, "Antioxidant Analysis for Monitoring

Remaining Useful Life of Turbine Fluids," Turbine Lubrication in the 21 st Century,
A S T M S T P 1407, W.R. Herguth and T.M. Wame, Eds., American Society for Testing and
Materials, West Conshohocken, PA, 2001
Abstract: In this study we measured the remaining useful life of a wide range of turbine
lubricants by detecting directly the remaining activity and performance of phenolic and
amine antioxidants, known as primary antioxidants. We applied a patented (linear)
voltammetric technique. First we evaluated the accuracy of the voltammetric response
(vs. weight %) for phenolic (BHT) and amine (PANA) types of antioxidants. The next
step of the program included antioxidant performance testing using laboratory oxidation
tests. The voltammetric test data were then correlated with other widely used analytical
methods (RBOT, FTIR, and DSC). These tests enabled us to understand the performance
of mixed antioxidant systems and their importance in the remaining useful life of the new
generation of turbine lubricants.
Finally and in order to understand the chemical role of antioxidants in real life conditions
we monitored several industrial steam and gas turbine lubricants (N. America and
Europe) with voltammetric analyses to observe the antioxidants' depletion trends.
Voltammetric analysis correlated well with the RBOT (as per ASTM D-2272), and
standard physico-chemical analysis methods (TAN, viscosity, color.... ). These results
show that antioxidant monitoring adds valuable information to RBOT and DSC
measurements for the new type of turbine lubricants. This will allow the user to enhance
the management of the oil on-site; to detect abnormal equipment/usage and consequently
to improve the planning of maintenance actions resulting in a more efficient oil
management.
Keywords: antioxidants, voltammetry, RBOT, oxidation control, amines, phenols
l InternationalSalesand MarketingManager,FluitecInternational,272 Leo Street,Dayton,OH 45404

2ResearchChemist,Universityof DaytonResearchInstitute,Dayton,OH 45469
86

AMEYE AND KAUFFMAN ON ANTIOXIDANT ANALYSIS 87
Introduction
In order to meet the demands of the industry turbine oil base stocks have continuously
improved, both through processing and synthesis. They parallel the development of
performance additive packages. Environmental and performance requirements are the
main drivers for this evolution. The direct consequence of the commercial and
environmental pressure results in a second remarkable development: the power
generating industry has significantly improved the thermal efficiency of its industrial
equipment. Combining gas and steam turbines (combined cycle operation) improved the
efficiency from an average nf30-35% for a single cycle equipment to between 50 and
60%. As a direct consequence of this advanced technology lubricant specifications
became stricter e.g. controlling oxidation under the conditions of high temperatures
representative of typical gas turbine operating conditions. During the normal operation of
gas turbines, the temperature of the oil film can exceed 140 to 150~
For both these developments the goal is to obtain a lubricant with higher thermal and
oxidative stability, or to create a lubricant with a minimal oxidative degradation. Typical
results for lubricants with poor oxidation protection/control are:
9 Sludge formation/filter blockage
* Oil thickening / increased viscosity
9 Formation of deposits/oxidation products such as lacquers/varnishes etc...
9 Increased acidity
These operating problems can be prevented by the combination of good quality base oil,
with a supplemental antioxidant package. Antioxidants added alone or in synergistic
combinations extend the operating life of the oil by improving its "oxidation stability".
During their use in the lubrication systems of turbines, the oxidation inhibitors
(antioxidants, AO) will deplete to a certain critical level at which point the fluid will start
to degrade / polymerize at an accelerated rate. At that point important changes in the
physical properties of the base stock occur. The oil or lubricant is no longer able to
protect the equipment, and its useful life is over. There is a growing interest in methods
predicting oil change measuring antioxidant concentration during the lubricant lifetime.
Monitoring the onset and the propagation of oxidation will lower operating costs (e.g.
downtime costs, equipment loss), and allow users to detect abnormally oxidized and
degraded lubricants.
The basic mechanism of oxidative degradation [1] and the role that different
antioxidants play, begins with the formation of reactive compounds, better known as free
radicals, hydroperoxides and peroxides. These reactive compounds will then propagate
into oxygenated hydrocarbons, such as alcohol, acids, and sludge. The antioxidants act in
2 ways: by removing radicals (primary antioxidancy) or by decomposing hydroperoxides
to form non-reactive products, which do not participate in further oxidation of the
lubricant (secondary antioxidancy). The current commonly used primary antioxidants are
the phenolic and amine types; the secondary antioxidants are usually sulfur or metal
containing additives.
This paper presents research on the relationship between the remaining useful life of
lubricants and antioxidant concentration, detected by using an electrochemical analytical
technique, based on cyclic voltammetry. First we will present laboratory studies
demonstrating the accuracy of the cyclic voltammetric-based method for phenol and
amines antioxidants. We will also establish the correlation between the remaining
antioxidant concentration with oxidation tests (Rotating Bomb Oxidation Test -RBOT,
Differential Scanning Calorimetry - DSC) for different types of turbine lubricants.
In the second part of the research, data collected from the field are compared with
standard physical property measurements (viscosity, total acid number, color and water)
and RBOT-values.
Experimental
Fresh lubricating oils
The fresh lubricating oils evaluated during this study were obtained from the
equipment operators (power stations), as well as commercial lubricant suppliers. The
base stock oils were polyol ester-based oils (jet engine oils) and hydrocarbon-based
turbine oils, supplemented with phenolic BHT (Butylated Hydroxy toluene also known as
DBPC) and/or aminic PANA (Phenyl-ct-NaphthylAmine) and DPA (Di-
alkylphenylAmine).
Voltammetric method
The method for evaluating the remaining useful life of the oil is mainly based on
voltammetrie analysis [2] of the lubricant sample. Voltammetric techniques are
electroanalytic methods in which a sample is mixed with an electrolyte and a solvent, and
placed in an electrolytic cell. Data are obtained by measuring the current through the cell
as a function of the applied potential. The test results are based on current, voltage and
time relationships at the cell electrodes.
The cell consists of a fluid container with a small, easily polarized microelectrode, and
a large non-polarizable reference electrode. The reference electrode should be massive
compared to the microelectrode so that its behavior remains essentially constant with the
passage of small current: it remains unpolarized during the analysis period. Additional
auxiliary electrodes can be added to the electrode system to eliminate the effects of
resistive drop for high resistance solutions. In performing a voltammetric analysis, the
potential across the electrodes varies linearly with time, and the resulting current is
recorded as a function of the potential. With increased voltage to the sample in the cell,
the various additive species under investigation in the oil oxidize electrochemically. The
data recorded during this oxidation reaction can then be used to determine the remaining
useful life of the oil type.
A typical current-potential curve produced during the practice of the voltammetric test
is illustrated in figure 1. Initially the applied potential produces an electrochemical
reaction with a rate so slow that virtually no current flows through the cell. As the voltage
is increased (figurel), the electro-active species (such as substituted phenols) begin to
oxidize at the microelectrode surface, producing an anodic rise in the current. As the
potential is increased, the decreases in the electro-active species concentration at the
electrode surface and the exponential increase of the oxidation rate lead to a maximum in
the current-potential curve (figure 1).
In this study the voltammetric method was performed with a commercially available
voltammograph equipped with a glassy carbon-working electrode, a platinum wire
electrode, and a platinum wire auxiliary electrode. The oil samples (400~tl) were diluted
in an ethanol/electrolyte mixture. This extracted the antioxidants into the solvent phase.
The voltage of the auxiliary electrode was scanned flom 0 to 1.5 V at a rate of 0.2
V/second. As the potential increased, the antioxidants oxidized at the carbon electrode
surface. The oxidation released electrons to the electrode, causing current, which was
linearly related to the additive concentration in the solution (Figure 1). The peaks
produced by the voltammetric method (the oxidation wave) were then used to evaluate
the remaining additives of the used oil samples. The peak of a
zincdialkyldithiophosphate (ZDDP) additive is followed by an amine (PANA), and then
by a phenol (BHT) (Figure 1). The new oil was used as the 100% standard and the
solvent was used as the 0% standard. The measurements of the used lubricant samples
were expressed as percentages remaining additives. The repeatability of the percent-
remaining additive measurements is between 2.5 and 4% RSD [3-5]
Compound "B" LinearRamp(Voltage)[

Phenyl-ct -naphthylamlue .~ / I
I
Voltage increases with time ]
Figure 1- Cyclic Voltammetry - Voltage vs. Current for different antioxidants e.g. ZDDP
(Zincdialkyldithiophosphate), PANA and Phenol.
When comparing the voltammetric procedure to existing oil oxidation test methods,
such as RBOT, DSC, FTIR, the following differences are apparent:
9 Ease in use: the test procedure consists of sampling the oil with a pipette, mixing it
with the 5ml prepared solvent vial, inserting the probe, and measuring;
9 Test method requires a very small oil sample volume (200 - 400 lal);
9 No chemical expertise is required;
9 Portable technology, allowing on-site analysis;
9 Quick test method (less than 5 minutes), which can be used for field analysis.
Oxidation test methods
The Rotating Bomb Oxidation Test (RBOT) in accordance with the ASTM Test
Method for Oxidation Stability of Steam Turbine Oils by Rotating Bomb (D 2272) was
used to estimate the oxidation stability of turbine oils. Differential Scanning Calorimetry
(DSC) (at atmospheric pressure with 50-ml/min oxygen, 200~ hold temperature) was
also applied to measure the oxidation stability for used turbine oils.
Fourier Transform Infrared Spectroscopy (FTIR) analysis was performed on a film of
oil (10 ~tL) between two potassium bromide plates. The FTIR analyses were performed to
monitor the depletion of the phenolic antioxidants (3650 cm l ) versus time.
Precision of the technique for detecting antioxidants in turbine lubricants
For this part of the study we used a mineral base oil (Group II) in which antioxidants
(supplied by a major additive company) were added with increasing concentration, to
determine:
a) The capability of the cyclic voltammetry to detect amine (PANA) and phenol (BHT)
antioxidants;
b) Calibration plots of the cyclic voltammetry for amines and phenols from oil samples
prepared with concentration of 0.5, 1, 1.5, 2, 2.5, 3, 3.5, 4 and 4,5 wt.% antioxidants.
No other additives were present in the oil.
c) All tests were measured in triplicate, by two independent laboratories, and the
average value of the peak height value was used.
A selected oil volume of 200~tl of the prepared standards was added to the electrolytic
test solutions, mixed, and tested by voltammetry. For the amine and phenol detection we
used 2 different electrolytic test solutions, for which we refer to the patent [2]:
9 Test solution 1: acetone based solvent with a neutral electrolyte
9 Test solution 2: ethanol based solvent with a basic electrolyte
As the previous experiment [5] has shown, both test solutions have good detection
capabilities. We included them both in this precision test. The resulting data are presented
in Figures 2, 3, 4, and 5. The antioxidant concentration by weight percentage was plotted
on the x-axis. The voltammetric response in arbitrary units was plotted on the y-axis.
AMEYE AND KAUFFMANON ANTIOXIDANTANALYSIS 91
BHT%vs solution1
1200
1000
O
800
P
. o: - 600
................ ~ 2
400
E _ _ J R== 0.9984
200
O
~>
0
2 :~ 4 5
1 BHT Concentration, wt%
Figure 2: Correlation BHT% with voltammetric peak response for test solution 1
BHT % vs Solution 2
1500
1000
500 ~x-+_28.806-
J R2 = 0.997
0
0.0 1.0 2.0 3.0 4.0 5.0
BHT Concentration,wt%
Figure 3: Correlation of BHT% with voltammetric peak response in test solution 2
The data (Fig. 2-5) show that both test solutions can be used to detect the BHT and
PANA antioxidants, with correlation factors higher than 0.98.
A similar excellent correlation was demonstrated with mixtures of aryl amines and
phenol mixtures in a previous publication [5].
PANA % vs solution I
2000
1~0--
1~0
y = 397.93x + 25.17
5~ R= = 0.98
0
0 1 2 3 4 5
PANA Concentration,wt%
Figure 4: Correlation PANA% with voltammetric peak response in test solution I
PANA% vs Solution 2
2000
J
1500
Q.
o
o~
1000
t.
500 J ~ ~ y -- 349.500x + 12.361

J R 2 = 0.995
0
0.0 1.0 2.0 3.0 4.0 5.0
PANA Concentration, wt%
Figure 5: Correlation PANA% with voltammetric peak response in test solution 2

To confirm the above graphs, an oil company blended an ISO 32 grade turbine oil,
with 2,6 di-tertbutyl phenol, with increasing concentration from 250 till 4000 ppm. There
was an excellent correlation between the voltammetric response and phenolic antioxidant
concentration (Figure 6).
In the above figure 6 voltammetric response is expressed both for the highest additive
peak number (RUL Number), as well as the total area (RUL area). The total area under
the voltammogram calculates the total antioxidant concentration, whereas the additive
peak number specifically focuses on 1 type of antioxidant.
In her research of the oxidation potentials of a wide range of additives and
antioxidants Amy Clough reported that the oxidation potential (Eox) of compounds does
not have an absolute magnitude; the value depends on the measuring conditions [7]. For
example Eox values of phenols were measured in a mixture of acids/alcohol which was a
A M E Y E A N D K A U F F M A N ON A N T I O X I D A N T A N A L Y S I S 93
particularly favorable medium for the oxidation of the phenolic functional groups.
Oxidation potential depends on the number, size and position of the groups attached to
the aromatic ring of the phenols. She also reported that Eox values of other compounds,
such as phenates and sulfurized olefins/polysulfides, were generally higher than those of
phenols, indicating that they were somewhat more difficult to oxidize. No research work
has yet been reported on voltammetric response for other secondary antioxidants such as
sulfur- or selenium-containing antioxidants.
RULER Response vs. Antloxidant Concentration
700 7000
- 4 - RULER Number
9-e- RULER Area
600 6000
RULER Number Rz = 0.999

RULER Area R 2 = 1.000
500 5000
~
Z
400 9
Yelh ~yo,va.t
A r M o ~ d a n t - 2.6 dl.tert
4000
300. 3000
200 2000
100 9 1000
0 -
0 1000 2000 3000 4000
Antloxldant Concantratlon(ppm)
Figure 6: correlation antioxidant % with voltammetric peak response in test solution 2
Correlation study between oxidation test and cyclic voltammetry
Analysis of Jet Turbine Oil using Oxidation Test (DEF STAN 05-50 Part 61 Method 9)
The oxidation method used was Defence Standard 05-50 Part 61 Method 9. This
involves two different procedures [8]. The first is referred to as "Temperature Parameters
- Oxidative Stability". The second is referred to as "Effective Life - Oxidative Stability",

We will present the test data from the second procedure. The turbine oil used for this test
is a polyol ester base oil manufactured for the latest generation of turbine engines (High
Thermal Stability-HTS oils). With its specific mixture of aromatic amines, as an
antioxidant package, this HTS turbine oil has an improved high thermal-oxidative
stability. This is a great improvement since the jet engine oils of earlier generations.
2000
1500
1000
500
o I : : , :
ljl I I I I I
0 1 2 3 4 . 5 6 7 8 9 10 11 12 13 14 15 16 17
Seconds (V Mode)
Figure 6a: voltammogram HTSjet turbine oil
Treading Oraph
$0-
~, 4O'
0
0 IO
Hours Usage
Figure 6b: trending Graph AO Depletion Turbine oil
Figure 6a shows a typical voltammogram from the HTS jet turbine oil: the main
antioxidant (aromatic amine) appears at 11.5 seconds. The graph of a used (oxidized) oil
shows the appearance of dimers & trimers (of the primary antioxidants) before the
antioxidants, at 6-8 seconds.j8] In order to determine the contribution of the dimers &
trimers to the remaining useful life of the jet oil, the voltammogram calculates the total
area under the voltammetric response (area RUL%) (Figure 6b).
Effective Life - Oxidative Stability
The procedure for Effective Life - Oxidative Stability is as following [9]: a weighed
volume of oil contained in a boiling glass tube is inserted into a heating block. Water-
saturated air is bubbled through the sample maintained at a constant temperature of
250~ After measuring the volatilisation loss, the sample is restored to its original
weight by adding fresh lubricant. Its viscosity, acidity increase and insolubles content
were then determined. Samples taken after the above tests were analyzed by voltammetry
using 100pl of test oil in the test solution 1 (Figure 6a). The time at which the same
parameters (TAN increase etc.) reached the critical points was determined: a
volatilisation loss of 15% w/w; an acidity increase of 1.0 mg KOH/g; a viscosity increase
of 15%; an insoluble content of 0.05% w/w and the formation of a gel (solidus).
The samples were heated to a 250~ for 0.5, 1.5, 2.5, 3.5, 4.5, 5.5, 6.5, 7.5, 8.5, and 9.5
hours.
Results - The RUL% of the samples falls rapidly to a minimum at 3.5 hours of testing,
(Figure 7). Make-up oil is used in order to make viscosity and TAN determinations
resulting in a gentle rise in RUL% of the samples from 3.5 to 9.5 hours. The viscosity and
TAN increases almost linearly until 9,5 hours.
80 - 70
70
60 J,
,~
~\
I ---,--Volatility
- ~-.Area%
l x TAN*10 ~
J
_/
60
5O
i
~\ 9 ViscOsity
50-
40
40
\\\\
20
i
~ 20-
lo- 10
..... o.-"-
o I I I 0
o 2 4 Timelhrs 6 8 t0
Figure 7: Aircraft Turbine Oil Effective Life with remaining antioxidant %, volatility,
viscosity (@40~ TAN vs. Test Duration at 250 ~ C test temperature
96 TURBINELUBRICATION IN THE 21sT CENTURY
Discussion - When the oil has less than 20% RUL the antioxidant becomes
ineffective, leaving the base oil open to oxidative degradation. This stage was reached
after 3.5 hours and from this point onwards the base oil was relatively unprotected. Both
the TAN and viscosity increases reached their critical limits between 2.5 and 3.5 hours.
A n a l y s i s of S t e a m T u r b i n e Oil Following Oxidation Test
In this part of the study, analysis was performed on a ISO VG 32 mineral steam
turbine oil with an R&0 -package. The new oil was analyzed by the voltammograph, and
a package of amines (diaphenylamines) and phenols was detected. The first additive
peak showed the amines, and the second additive peak the phenolic antioxidants (Figure
8).
Ti~Rur{E:.Q~-ge11:4~4,~Ald
1500
PHENOLS
1000
0oo
AMINE[S
o : ,, ! ! !
t 2 3 4 5 , , 0 , 1o ,1 ,, . 1, 15 10 1,
Silconall (~ Mads)
Figure 8: voltammogram for ISO VG 32 turbine oil, with in the x-axis the time of
analysis, and the Y-axis the voltammetrie response as arbitrary units.
Data were obtained from a RBOT test performed on the above-described ISO VG 32
steam turbine oil. Four oil samples were collected from 4 multiple RBOTs that were run
and terminated at the times shown in table 1. These samples were analyzed by
voltammetry for the detection of total antioxidant concentration. Figure 9 and Table 1
show the depletion trend of total antioxidant concentration, detected in both test solutions
(solution #1 & # 2), as well the % remaining RBOT life.
Table 1 - R B O T test results vs. Remaining AO%
Sample ref RBOT (min) Remaining % AO Remaining% AO

Test solution 1 Test solution 2
New oil 2008 100 100
1 1708 98 55
2 1408 58 23
3 1104 48 6
4 808 46 10
10o 100
--B-- Solution 2
75 75
& Solution 1
50
-'0-- RBOT %
o~ 2s 25
X RAN
0 Number *
100
0 t ~ample a 4
Figure 9: correlation of RBOT values with voltammetric data for oxidized turbine oils
100
.-_=,75
50 ~/'I --0~8,2 ! ~ + !2.5! i5
4 R 2 = 0.864
25
0 i
2O 40 60 80 100 120
Total A n t i o x i d a n t Concentration
Figure 1O: correlation graph between RBOT and total antioxidants concentration by
voltammetry in test solution I.
100
75
;o so J y = 0 . 5 5 7 x + 48. 374
25
0 20 40 60 80 100 120
Total Antioxidant Concentration
Figure 11: correlation graph between RBOT and total antioxidants concentration by
voltammetry in test solution 2.
For this mineral-based turbine oil the correlations between RBOT and total antioxidants
concentration by voltammetry in test solution 1 and 2 were 0.86 and 0.85 (Figures 10
&l 1). These values, based on previous research [6] are within expectations. The question
however, could be asked why the antioxidants have different depletion rates. As
voltammetric analysis has the capacity of differentiating antioxidant depletion rates, the
above data were analyzed again from that perspective. Also, the acidity increase
(measured by voltammetric Ruler Acid Number, expressed in mg KOH/ml oil) [10]
started accelerating when the remaining additive percentages fell below 40-50%. This
could be due to 2 phenomena:
* Consumption of basic antioxidants.
9 Increase of oxidation rate and products.
.100 100
I-.
0
O 75 7s ~
--AI-- Phenols
!.E 5o 50
I.-.E
--!1- Amines
25
X RAN
v v 0 Number
I 0 1 2 3 4 - - ~ - ~B0gT o/,
Sample
Figure 12: Individual antioxidant depletion vs. RBOTshows RBOTmore dependent on

amines than phenols.
AMEYE AND KAUFFMAN ON ANTIOXIDANTANALYSIS 99
The data in Figure 12 indicate the phenols that had become depleted in the first 10
hours.
Previous research on Dry-TOST test for hydraulic oils confirmed the same trend [5], but
the total acid number started to increase much faster once amine levels were reduced to
40% of the remaining antioxidant. Earlier experiences from field data [3] also indicate
the difference in antioxidant depletion mechanism between steam and gas turbines. BHT,
used commonly as antioxidant in steam turbine oil, is consumed, both by evaporation and
by oxidation. Aromatic amines typically used in gas turbine oil and subjected to much
higher temperatures than conventional steam turbine oil, have a different depletion rate
(Figure 12). Bearing this difference in antioxidant depletion mechanism in mind, the
capability of voltammetry to show the difference in depletion rates between antioxidants
represents a complementary advantage over RBOT and DSC.
Research study data on 1S0 VG 32 turbine oil- RULER vs. DSC
A DSC oxidation test was performed on all the collected oil samples from the RBOT
test. It was expressed as the Oxidation Induction Time (OIT minutes). The DSC test was
performed at atmospheric pressure, with an oxygen flow of 50 ml/min, and at a hold
temperature of 200~ The temperature ramp rate was set at 20.0 C/min. First the RBOT
results were correlated to the DSC results. This resulted in a correlation factor of 0.923
for the data shown below in table 2.
Table 2 - Results of RBOT vs. DSC for oxidized oil samples
Sample RBOT % DSC OIT

New oil 100 10.04
1 85 6.63
2 70 5.53
3 55 4.53
4 40 3.23
For the correlation with voltammetry and RBOT, we needed to determine the OIT of
the base stock, which is 0.5 minutes. Comparison of the DSC OIT with the depletion
trends of the antioxidants showed that the amine depletion rate has 2 different trends
(Figure 13):
6 For concentrations of phenols > 20%; the amines and DSC values have the same
trend.
6 Once phenol concentration is below 20%; the DSC values show more severe
oxidation.
It is interesting to note that the acid number started to increase for oil samples with low
phenol concentration. This is probably the result of lack of radical catchers.
As for the RBOT values, the individual antioxidant depletion rates have been correlated
to the DSC values (Figures 14 and 15).
70 8
60
c 50 6 (n
ix 4o o
0Oo
I~ 2 - m - Amines
10
0 . . . . 0 x RAN
0 1 2 3 4 = ---v..-~n~'rn)
Sample
Figure 13: _IndividualAntioxidant depletion rate (R UL%) vs. DSC OIT (minutes)
12 I
10 ~ - - ~ • + 0~0948 . . . . . . . . . . . . . . . .
i 8 /
/
p2 = N 7RqR ~,
...... ~.~,~,~,a.~**'*- e
4
2
0
0 20 40 60 80 100
Total Antioxidant Concentration,%
Figure 14: correlation graphsfor DSC OIT and Total Antioxidant Concentration by
voltammetryfor test solution 1
12
10
"~ 8
-----6
0 . ~ 7 x + 3.5185
04 Rz = U . ~ 3 ~
a t
2
0 20 40 60 80 100
Total Antioxidant Concentration, %
Figure 15: correlation graphsfor DSC OIT and Total Antioxidant Concentration by
voltammetryfor test solution 2
In order to compare the depletion trends of amine and phenolic antioxidants all the oil
samples analyzed by RBOT, voltammetry and DSC, were also analyzed by FTIR.
FTIR data showed the phenol depletion (3650em'l), and oxidation increase (1750 - 1650
em"1) for the new oil and the used oil. Combined spectra of the 4000-3000 cm "1 band (Fig
16) for the four oil samples illustrate the depletion of the phenol additive (3650 cm "1)
versus time. Earlier publication [11] also described the good correlation between
voltammetry and amine antioxidants by FTIR.
-u
5O
46
4O
35
3O
2O
16
tO
!i12..
S
4OOO
Oete:W ~ k ~ 01 C~2~42 ~C~O 1
SC~S: 286
R ~ 4.000
Figure 16: FTIR spectrafor the series of 4 used oil samples
Used oil analysis from field oil samples
Field practical case 1
In this part of the program, measurements were taken from a power station in Europe,
with Frame-9 Gas turbines for a total output of 2000 MW. Each turbine is equipped with
a 35,000 liter oil reservoir and uses a mineral based turbine oil. The power station
measures every 6 months the following parameters: water, viscosity, RBOT,
voltammetrie analysis for antioxidants, color and ISO cleanliness level.
The turbine fluids are now more than 2 years in service. Their
characteristics/specifications for the voltammogram, detecting broad peak of aromatic
amines as antioxidants are shown in Figure 17.
102 TURBINELUBRICATION IN THE 21st CENTURY
2001
O' 1 1 1 ',
0 1 2 3 4 5 6 7 8 9 10 11
Seconds (A Mode)
Ruler Numbers Ruler Am~e Additive RULs

Standard: 370 Standard: 3112 #1: 76%
Sample: 281 Sample: 2399
RUL: 76% RUL: 77%
Figure 17: Voltammograph of industrial gas turbine lubricant

The voltammogram shows no phenolic antioxidant. This is a logical consequence of
volatility and reactivity of phenols in the higher temperature environment. Table 3
summarizes the characteristics of the new oil.
Table 3: characteristics new gas turbine oil 1S0 VG32
Measurements New Oil

TAN (mg KOH/g) 0.07
Color L 2.0
RBOT (mins.) > 1500
Viscosity (40~ mm2/s 32,1
Water (%) < 0.05
RUL % Amines 100%
Every two months the power station analyzed samples with the voltammetric
equipment, to determine their remaining % antioxidant. They complemented their 6
months analysis with the antioxidant data, Figure 18 summarizes all the analytical data
for one of the five turbines, during the last two years of service. Figure 19 presents only
the trending graphs of the antioxidants for the 5 gas turbines.
1600
100
1400 z,<
1200 80
X
0
D
= 1000 1-
6O z.<
800
..I
I=~ soo 40 rv
x ~ X X X X
400 2~ " ~ R.B.O.T.

200 . ~ i -B- RUL%
--e- TAN ('100)
0
I e------e e-----e
l
0 / --x--Vlsc.40*C
| -e- Color
Operating Hours
Figure 18: analytical data for gas turbine oil in service on GT 2
Trending AO % for GT 3,4, 5, 6, 7
1-'-Ec4
~'~ ~ "~ I.-~-EC5
/e t-x- EC6
4380 8066 10119 12534 13116 14799 16260 17869 21368 24193
Operating Hours
Figure 19: trending graphs for antioxidants on 5 operational gas turbines

104 TURBINELUBRICATIONIN THE 21STCENTURY
It is important to note that antioxidant trending by voltammetry and RBOT (as well
DSC) gives similar results. From the operator's (power station) point of view, the
voltammetric field results could be used to screen oils for RBOT. When the antioxidant
concentration stays above 50 % and the total acid number remains below the oil company
specifications RBOT is not needed. This will save the end user money and time.
The 2-years data showed that the parameters (viscosity, color and water) remain
relatively stable. The AO concentrations suggest that the fluctuations are probably caused
by top-ups of fresh oil.
Field practical case 2
To finalize this research regular oil samples were assembled from different North-
European power stations over a period of 6 months. The oil samples from different oil
suppliers are used in various steam and gas turbines. For the 35 oil samples, RBOT and
the remaining antioxidant were measured by the power station or a central laboratory.
The laboratory supplied the fresh oil RBOT data. A good correlation can be observed
between the antioxidants remaining and the remaining RBOT life (Figure 20). The
summary of the information on all oil samples of the North-European power stations is
listed in the appendix.
lOO
.o
9~ 60
~ 4o
# 2o
0 , = ,
0 25 50 75 100
RBOT %
Fig. 20: correlation graph for field oil samples between RBOT and remaining antioxidant
concentrations
The correlation holds well for a wide variety ofoil manufacturers and different
turbines. We emphasize the added value ofvoltammetry for the new generation of
turbine fluids:
9 If the oil has a high RUL, the oil is good.

9 IfRUL % is low, confirmation by RBOT measurement is necessary.
There is a slight difference between the RBOT values and the remaining antioxidant
concentration. This can be explained by the fact that RBOT detects the total antioxidant
capacity of the lubricant, including the natural antioxidancy of the base stock. The above
data show however, that two important factors must be considered:
9 New hydrotreated base stocks, as well as synthetic base stocks, have a much lower
natural antioxidancy. This will be of less influence on their remaining useful life.
9 Existing procedures e.g. ASTM Test Method for In Service Monitoring of
Lubricating Oil for Auxiliary Power Plant Equipment (D 6224), advise to consider oil
change when the warning limit for turbine fluids reaches the value of 25% remaining
RBOT. This is in line with voltammetric analysis.
Conclusions
Laboratory and field data both show there is a good correlation between RBOT, DSC
and antioxidant analysis by voltanunetrie technique. Voltammetrie analysis has the
additional benefit to differentiate the depletion rates for different antioxidant. This is very
valuable information for the new generation of turbine fluids (e.g. re-additization).
The antioxidant detection by voltammetry proves its highest value for jet turbine
lubricants, which operate at the highest temperatures. It is a predictive technique that
detects abnormally operating engines before the viscosity and total acid number increase.
In the case of industrial turbine lubricants, field experience shows the complementary
features of voltammetry with standard analysis programs. For the new generation of
turbine fluids, in combination with higher load factors of equipment, the voltammetrie
analysis of antioxidants will result in a quick and easy understanding of oxidation
processes.
Voltammetry complemented by RBOT (DSC) also provides the user information on
which antioxidant depletes faster. Voltammetric analysis can be performed on site, with a
minimal time and sample size. Therefore it can be used to monitor the oil more frequently
and thereby detect abnormally operating conditions (accelerated rate of antioxidant
depletion) at an earlier stage.
The results show that there is good correlation between voltammetrie analyses and
RBOT measurements. Further research will improve the correlation between
voltammetry, RBOT and other field tests. It will give a better insight of the effects of
different types of antioxidants on the test results of the various techniques.
References
[1] Rasberger, M; "Oxidative degradation and stabilization of mineral oil based

lubricants; Chemistry and Technology of Lubricants", Ed. Mortier and Orszulik, Blackie
and Son Ltd, 1992, 83 - 123
[2] Kauffman, R.E., "Method for Evaluating the Remaining Useful Life of a
Hydrocarbon Oil" U.S. Patent n ~ 4, 764, 258 (1988)
[3] Kauffman, R.E.," Remaining Useful Life measurements of Diesel Engine Oils,
Automotive Engine oils, Hydraulic fluids, and Greases using cyclic voltammetrie
methods", Lubrication Engineering, STLE, Volume 51, 3,223 -229
[4] Jefferies, A and Ameye J., "RULER" and Used Engine Oil Analysis Programs",
World Tribology Congress, London, UK, September 1997, Published in Lubrication
Engineering Magazine, STLE, Volume 54, 5, 29-34
[5] Van Leeuwen W.; "Use of RULER" for the Determination of Critical Antioxidant
Concentration in Ester Based Hydraulic Fluids", Quaker Chemical, Presented at the
STLE '98, Detroit Annual Meeting, and AISE '98, September, Annual Meeting
[6] Herguth W.R. and Phillips, S., "Comparison of Common Analytical Techniques to
Voltammetrie Analysis of Antioxidants in Industrial Lubricating oils", paper presented at
Condition Monitoring Section STLE Annual Meeting Cincinnati, May 1996.
[7] Clough A.E., "The Relative Strengths of Oxidation Inhibitors Used in Lubricants and
the Monitoring of Their Consumption - an Electroanalytical Approach", Recent Res.
Devel. Oil Chemistry, Vol. 1 (1997): 57-67', published by Transworld Research
Network.
[8] M. Hergeth, "Die Bestimmung der relativen Antioxidantienkonzentration von

Triebwerkrlen mittels zyklischer Voltammetrie", DaimlerChryslerAerospaee and
Fachhochschule Miinchen University of Applied Sciences, March 2000, pp.34
[9] Ameye J. and Steve Lee, "Experiences with RULER'" Oil Analysis Instrument for
Quick Determination of Remaining Useful Life on Jet Diesel Engine Lubricants",
International Conference on Condition Monitoring, University of Wales Swansea, April
1999, proceedings pp.51 - 64.
[10] Kauffman, R.E., "Rapid, Portable Voltammetric Techniques for Performing

Antioxidant, Total Acid Number (TAN) and Total Base Number (TBN)
Measurements9 Presented at the 52ndAnnual STLE Meeting, Kansas City, Missouri,
May 1997, Published in Lubrication Engineering, January 1998, p 39-46
A M E Y E AND K A U F F M A N ON ANTIOXIDANT AN AL YSIS 107
[11] Wurzbach, R.N., "Oxidation Stability and Strategies for Extending Lubricant Life" -
Practicing Oil Analysis '99, Noria Conference, Tulsa, OK, October 1999, Conference
proceedings p. 264-273.
Appendix
1) DSC printouts
~"-"''-"''.'-. ONSET OF OXIOATION

PUe B ' ~ e : 7137111.I)00
~mmyle Name: FLUI~C ~ 3@~
DSC Temp
InW OIT- 10.04 miu
)0.00
~o. ooI /
)0.00
-10"00,[I i J J I, 1 q - , i , I J t r r I I , r i , ~, I I .00
0.00 10.00 20.00 30.00 40.00 50.00
,m,
Rate
P..,..,~
Hold Tlmp Hold Time
lime [min] ,e~ht:
CoB;
,o.,,,,,t.,,,,]
Copp.,n"
~{~ni n] [C]
200,0
[mini
&go.O
Atmosphere: Kitr
Rate Flow: 25.wlml/'Ir
Operator:
DSC test new oils
~=~"-"~""~".~ ONSEO
I FOXD
I ATO
IN
P/do Hlme: 718717,DO0
S ~ p l e H ~ e : l~/3rrEc 9o4
OSC Temp
o r r . e.e.~ mtn.
mW C
0.00 f . t100.00
-10'00~1 , , , i I , r , i I , , t i I r L , i I ]?'00
OO
.0 IO.O0 20.00 30.00 40,00
~..~ ~.~..
Rats Hold Tomp Hold Time
Time [rain] ,,~,=:
Cell:
,o,ot,,o,,.l
Copt~er
[C/mini [c] [miu] h t m ~ p h a ~ : Nltrorosuzys en
200 zoo.o 4eoo r.~ ~: ~ oo[~a/mla]
Operator:
DSC test used oil

(sample 1)
JournallD Date RBOT % RUL% Oil

1918.9.07QFA 21-01-99 29 26 Castrol
2093.9.37QFA 02-09-99 46 49 Castrol
2003.9.23QHI 07-06-99 60 63 Shell
1912.9.06QSAE 04-02-99 63 66 Shell
2069.9.30 QSIE 26-07-99 64 55 Shell
1985.9.20 SH 11-05-99 21 21 Texaco
2145.9.45 SH 09-11-99 20 30 Texaco
2140.9.44 RFV 02-11-99 69 69 Shell
2041.9.29 HKV 03-05-99 28 27 Mobil
2208.9.52 SVHO 16-12-99 24 25 Mobil
2002.9.23 RKV 04-06-99 30 37 Texaco
2143.9.45 SH 05-11-99 30 35 Texaco
2122.9.43QHO 15-10-99 92 86 Shell
1990.9.20 DIV 20-05-99 89 87 Texaco
2073.9.34 SVSB3 23-08-99 82 73 Texaco
2209.9.52 DIV 24-12-99 49 47 Shell
2121.9.43 SH 12-10-99 34 32 Texaco
2171.9.50 SVSB3 08-12-99 69 76 Texaco
2172.9.50 SVSB3 08-12-99 79 70 Texaco
2047.9.29 NWB3 21-07-99 67 75 Castrol
2079.9.35 NWB3 03-09-99 38 35 Castrol
2080.9.35 NWB3 03-09-99 74 80 Castrol
1981.9.19 SVSB3 07-05-99 99 85 Texaco
2048.9.29 NWB3 21-07-99 52 45 BP
2168.9.49 NWB3 01-08-99 77 56 BP
2049.9.29 NWB3 21-07-99 46 42 BP
2169.9.49 NWB3 01-08-99 42 44 BP
2050.9.29 NWB3 21-07-99 48 31 BP
2090.9.37 DIV 16-09-99 57 55 Shell
2091.9.37 DIV 16-09-99 78 65 Shell
2092.9.37 DIV 16-09-99 67 59 Shell
2043.9.29 HKV 15-07-99 60 65 Texaco
2046.9.29 HKV 03-05-99 68 64 Shell
2005.9.23 DIV 25-05-99 Texaco

William R. Herguth, Thomas M. Warne, Editors-Turbine Lubrication in The 21st Century (ASTM Special Technical Publication, 1407) (2001) PDF

Uploaded by

Copyright:

Available Formats

William R. Herguth, Thomas M. Warne, Editors-Turbine Lubrication in The 21st Century (ASTM Special Technical Publication, 1407) (2001) PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

William R. Herguth, Thomas M. Warne, Editors-Turbine Lubrication in The 21st Century (ASTM Special Technical Publication, 1407) (2001) PDF

Uploaded by

Copyright:

Available Formats

STP 1407

Turbine Lubrication in the

William R. Herguth and Thomas M. Warne, editors

ASTM Stock Number: STP1407

TJ266 .T872 2001

Authorization to photocopy items for internal, personal, or educational classroom use, or

Peer Review Policy

The Use of a Fire-Resistant Turbine Lubricant: Europe Looks to the F u t u r e - - -

Advanced High-Temperature Air Force Turbine Engine Oil Program~

The Evolution of Base Oil Technology--DAWD C. KRAMER,BRENT K. LOK,AND

Turbine Oil Quality and Field Application Requirements--STEWN T. SWIFT,

Performance Advantages of Turbine Oils Formulated with Group H and Group HI

Improved Response of Turbine Oils Based on Group H Hydrocracked Base Oils

Performance Advantages of Turbine Oils Formulated with Group II Base Oils--

Antioxidant Analysis for Monitoring Remaining Useful Life of Turbine Fluids---

This publication summarizes the presentations delivered at the "Symposium on Turbine

The Use of a Fire-Resistant Turbine Lubricant: Europe Looks to the Future

Reference: Phillips, W. D., "The Use of a Fire-Resistant Turbine Lubricant: Europe

In 1944, at a meeting of the Machines Technical Committee of the German Power

Copyright92001 by ASTMInternational www.astm.org

association (the Technische Vereinigung der Grosskrattwerksbetreiber-VGB) maintains a

order to overcome this disadvantage the fluid is normally conditioned by circulating

Table 1-.4 Comparison of the Fire-Resistance Properties of an ISO VG 46 Mineral

Property Test Method Mineral Polyol Ester Phosphate

Property Mineral Oil Triaryl Phosphate

Fire-Resistant Turbine Control Fluids

Experience with Fire-Resistant Lubricants

As suggested earlier, while the use of a fire-resistant hydraulic fluid is effective in

Table 3-The Effect of Chemical Type on the Performance of Phosphate Esters

Chemical Type Autoignition Air Release Hydrolytic Oxidative

Isopropylphenyl 540 5 Moderate Moderate

Tertbutylphenyl 580 5 Poor-Moderate Excellent

Xylyl 570-590 1 Good-Very good Good

Table 4-Analytical Data on a Phosphate Ester Usedfor the Lubrication of a IOMW

Property Units 0 3000 14000 26000 40000 55000

Table 5-Use of a Fire-Resistant Lubricant in Turbines of the CIS

Power Station Type Output Date Conversion Operating

Although the experience in steam turbines with fire-resistant lubricants is probably

As with existing turbine oils it is necessary to adequately maintain phosphates in order

Table 6-Analytical Data of a Triaryl Phosphate Ester Usedfor the Lubrication of a 5

Fire-Resistant Fluids and Lubricants-Features and Benefits

Feature Principal Benefit

Very long life when used with state of

Obstacles to Progress with Fire-Resistant Lubricants

With such a large amount o f satisfactory operating experience it is perhaps surprising

Technical and Commercial Drivers Towards Fire-Resistant Lubricants

In spite of the reluctance of some turbine builders to introduce fire-resistant lubricant

[1] Der Erfahrungaustausch der Maschinentechnischen Ausschfisse der Vereinigung der

1997, 77,6, pp.412-416.

[7] "Sprinkler Impairments: Planning and Prevention," Factory Mutual Record,

[10] Kaspar, K., "Einsatz von synthetischen schwerbrennbaren Hydraulikflfissigkeiten im

[13] Sch6ner,W., "Betriebliche Bewahrung schwerbrennbarer Steuer- und

[ 15] Vilyanskaya, E. D. et al, "Experimental Full-Scale Operation of OMTI Fire-

[16] Piehugin, I, "lO00-MWDampfturbinef~rKernkraftwerke undihre

[ 17] "Fire-Retardant Lubricant for Turbo-Generators," EPRI Report NP-2843,1983.

[ 18] "Evaluation of Fire-Retardant Fluids for Turbine Bearing Lubricants," EPR1

[20] Phillips, W. D., "The Conditioning of Phosphate-Ester Fluids in Turbine

[26] Grupp, H., "Schwer enO'lammbare Hydraulilflliissigkeiten-eine

Advanced High-Temperature Air Force Turbine Engine Oil Program