PTFE for Lubricating Greases

Consultant to Shamrock Technologies, Inc.
Newark, NJ 07114 USA

Abstract
The purpose of this study was to investigate the
performance of seven experimental PTFE powders in
grease and evaluate effects of particle size and dispersion
on consistency, stability, friction and wear. Laboratoryscale batches of simple lithium greases were prepared
and milled with a high-pressure homogenizer. All PTFE
greases were comparable to controls in appearance (off
white), NLGI grade (2), texture (smooth, buttery) and
storage stability (<5% bleed after 4-5 months storage).
Four-ball friction and wear tests demonstrated the
value of specific PTFE powders as additives. Submicron PTFE powders, formulated in grease without
other extreme pressure (EP) or anti-wear additives,
reduced wear, friction and temperature during mixed!
boundary lubrication. These PTFE powders prevented
adhesive wear due to oil film failure at applied loads
up to 180 kg. Test results suggested that PTFE was
entrained in contacts and helped support loads. These
results depended upon PTFE dispersion in specific
formulations.

Introduction
The present paper takes a direct and pragmatic
approach to investigate effects of experimental PTFE
powders in simple lithium greases formulated without
other additives. This Introduction provides background
information about PTFE powders. In prior studies of
PTFE in greases and liquid lubricants, results were
situational and depended upon particle size, dispersion
and tribology test conditions such as speed, load and
roll-to-slide ratio. This study compares seven powders
with primary (fully dispersed) particles that range from

sub-micron to larger sizes. This Introduction compares
these particle sizes to nanoparticles, grease fibers and
wear scars. [1]
Grease preparation is the subject of the next section.
Each lab-scale batch of grease was blended and milled
with the same procedure to model PTFE dispersion in
manufacturing processes. Results for grease consistency
and storage stability are presented next. Then, the
heart of this paper focuses on four-ball friction and
wear testing. Data collected during tests with ASTM
D2266 [2] indicated that start-up and run-in affected
test results. A four-ball technique was customized by
gradually applying loads to lessen start-up and run-in
effects. Friction and temperature data and wear scars
were analyzed to characterize effects of PTFE powders on
lubrication.
Polytetrafluoroethylene or PTFE is a synthetic
fluoropolymer of tetrafluoroethylene (-CF2-CF2-)n
with a chemical formula that resembles ethylene (-CR2CR2-)n but with fluorine instead of hydrogen atoms.
In contrast to hydrocarbons, PTFE chains are stiffer
with a greater tendency to pack in semi-crystalline
arrangements. PTFE films and solid surfaces tend to
be white, waxy and smooth. (The most well-known
application of PTFE is non-stick coatings on cookware.)
PTFE is extremely slippery with coefficient of friction,
COF, between 0.04 (sliding friction, D1894 [3]) and 0.1.
Other useful properties include resistance to chemicals,
solvents and temperature extremes (highs and lows). [4]
Lubricants are formulated with PTFE particles that are
either synthesized or produced by milling solids. PTFE

22 VOLUME 79, NUMBER 3

powders are used as thickeners for grease and additives
to improve resistance to water, chemicals and friction.
For example, PTFE is formulated in spray-on dry film
lubricants, release agents and non-flammable lubricants
intended for use under oxidizing or chemically aggressive
conditions. [4] PTFE powders were investigated as
possible friction-reducing additives in crankcase engine
lubes and transmission fluids. Continued research
in PTFE has observed that lubricating effects and
mechanisms depend upon PTFE particle properties
(size, shape, surface polarity, etc.), dispersion and test
conditions (contact geometry, roll-to-slide ratio, speed,
load, etc.) [5-7]

PTFE powders can be manufactured with specific size
distributions of primary (non-aggregated) particles.
Sub-micron grades of PTFE contain primary particles
with average diameter ~~
PTFE Powders
0.001

0.01

0.1

Wear Scars
10

100

1000

Microns (p)
Fig. 1 Size scales of nanoparticles (
4, 6 and 14 ~sm) for dust, etc. [8], PTFE particle diameters (0.1 100 sum), and wear scar specifications
(0.5 mm or 500 sum) relative to the resolution of the unaided human eye (—‘ 40 ~tsm).

–

In other studies, PTFE powders were added to greases,
and friction, wear and load carrying capacity were
measured. Aswath [9] mixed PTFE with ZDDP (zinc
dialkyl dithiophosphate), MoS2 and FeF3 (shown to
catalyze adhesion of PTFE and MoS2 on metal) in lithium
stearate base grease. They performed block-on-ring tests
and reported that smaller PTFE (0.2 mm or 200 im)
was better than larger PTFE (1 4 mm) in terms of wear
resistance and weld load. Krawiec [10] added 6 wt.%
PTFE powder (20 40 ~tm) to all-purpose vehicle grease
and reported that PTFE increased seizure loads by 150%
and decreased wear scars by 30% relative to control grease
in four-ball tests.
–

–

Fish et al. [11, 12] compared effects of additives on
load carrying capacity of base grease with a weld load
of 160 kg (D2596, [13]). In three cases (3 wt.% PTFE; 3
wt.% MoS2; and 1.5/1.5 wt.% PTFE/MoS2), weld loads
improved to 250 kg. The weld load reached 500 kg for
grease with PTFE plus a borate additive. Their results
suggested that PTFE (particle size not specified) showed
potential as a substitute for MoS2 as an extreme pressure
(EP) additive in grease.
Ballester et al. [14] compared four PTFE powders with
average primary particle diameters of 0.8, 4.2, 20.5 and
46.8 pm. They mixed PTFE (1 wt.%) into commercial
lithium base grease (no EP additives) using an overhead
23

–

NLGI SPOKESMAN, JULY/AUGUST 2015

mixer and a Cowles blade. Four- ball friction and wear
tests were performed according to D2266. One powder
(20.5 urn) reduced COF (20% relative to control grease).
Wear scar diameters were comparable for control and test
greases.

were comparable for control and PTFE greases. This
indicated that lithium fibers determined consistency,
and PTFE particles were dispersed among lithium fibers.
Consistency did not trend with weight percent PTFE-2
(appropriate for HX-1 registration) and PTFE-3, Fig. 4.

Grease Consistency and Stability [15]

Twenty-five samples of control and PTFE greases were
stored in the laboratory with no exposure to light. After
four to five months, there were no changes in appearance.
Less than 5% of the oil separated but mixed easily into
23 samples. Two samples (control grease, PTFE-2 grease)
softened and released more oil during storage.

Greases prepared from PTFE, Li12OHSt and PAO
(4/8/88) and control grease (0/8/92) were off-white in
color with smooth, buttery texture and NLGJ grade
2. In most cases, working grease slightly increased
penetration depth, i.e., softened grease by dispersing
solids and entrained air, Fig.3. Cone penetration data

Unworked and worked cone penetation data (0.1 mm)
310
300

Fig. 3 Control
and PTFE
greases were
NLGI grade 2

290
280

270
260

JI]

250
‘z~
00~’

‘0

l~h]]]l111
v~

Cl?

~ Worked

~ <P

c)d~’
Worked Cone Penetration (0.1 mm) vs. %
PTFE

310
Fig. 4 Worked cone
penetrationfor
greases prepared
with different levels
ofPTFE-2 and
PTFE-3

300

290
280
270
260
250

HI~ii~
HI ~I II F~
0

1.5

24 VOLUME 79, NUMBER 3
–

PTFE2
PTFE3

NLGI

Four-Ball Friction and Wear [15]
Greases were evaluated using a four-ball friction and wear machine and D2266 (75°C, 40 kg load,
1,200 rpm, sliding speed 20 rps x 2 rt x 3.8 mm 0.5 mIs). This test is used to make preliminary
assessments of anti-wear properties of greases in sliding contacts. In practice, fully-formulated
lubricants are tested relative to specifications for qualification and quality control. In this study, friction
data indicated that contacts were damaged during start-up and run-in, e.g., Fig. 5 for control grease.
Damage can form wear particles, and a few wear particles can initiate a cascade of more damage and
wear particles [18, 19]. Insufficient oil bleed can contribute to wear. [20] Thus, start-up and run-in can
affect performance during four-ball tests and compromise test results.
Control Grease, D-2266
2.5
2,0

Fig. 5 Friction data (kg)
during D2266for two
samples of control grease.

1.5
1.0
0

0.5
0.0
0

Time (mm)

laboratory was maintained at 2 1-23°C (70-74°F). Wear
scars were measured and photographed with commercial
light microscopes and cameras or calipers. The sample
cup was loaded with a consistent mass of grease or
volume of liquid. A torque wrench (54.2 N.m or 40 lb-ft)
was used to lock balls in the cup.

To reduce damage during start-up and run-in, grease
could be reformulated with EP additives. However, those
additives could have unanticipated effects on testing
PTFE. Instead, the four-ball test procedure was modified
to reduce the severity of start-up and run-in. The load
was increased gradually with a ramp. This approach
resembles gradually polarizing a working electrode and
measuring current flow to perform a potentiodynamic
scan, which is an electrochemical
technique.

To develop a load ramp procedure, simple lithium
grease formulated with PTFE-3 was
tested in four consecutive stages,
Table 1. Friction fluctuations were
not observed at the start of this test
(Fig. 7). Start-up and run-in effects
were significantly smaller with load
ramps than D2266 Friction increased
with load and possibly plateaued,
corresponding to mixed and
boundary lubrication, respectively.
Temperature increased smoothly. The
steepness of these load ramps did not
affect friction or temperature.

Fig. 6 Four-ballfriction and wear
machine (circa 1982) used in this
study.
In these four-ball tests, friction was
measured with a load cell and force
gauge, and temperature was measured
with a thermocouple at the perimeter
of the cup. Time series were collected
at intervals of 2 using an AD converter
and data acquisition software. The
–

–

NLGI SPOKESMAN, JULY/AUGUST 2015

NLGI~
Table 1 Load ramp stages in the initialfour-balifriction and wear test
Stage

[

Load

Ramp

Elapsed Time

(kg)

load increment / time interval

(mm)

31o30

3kg/5min

0to45

Maintain load

45 to 72

30to45

3kg/—2min

72to86

45 to 120

15 kg 1 10 to 20 mm

Simple statistics were calculated from data at each
load, excluding two features discussed below. Average
(mean) friction and temperature trended with load, Fig.
8. Friction fluctuations increased during the test, possibly
due to wear or more asperity contacts at higher loads, Fig.
7.

Three more statistics were calculated: standard deviation
divided by the mean, skew and kurtosis. The statistics
showed that friction fluctuations were large (relative to
the mean) and not random during run-in up to a 15 kg
load, Fig. 9. These results are consistent with shearing
individual asperities [21], churning grease [20], etc.
during run-in. Also, sounds were shrill up to 15 kg and

1.2

[

86 to 160

then less shrill at higher loads. After run-in, friction
fluctuations were relatively small ( 0.4 and
Ikurtosisi > 0.8 for 150 data) below 15 kg but not at higher
loads.]
Much larger friction fluctuations were observed at loads
of 27 and 90 kg. Temperature was steady, and welding did
not occur during these events.

—

STAGE I

STAGE 2

STAGE 3

STAGE4

1.0
0.8
0.6

—

~~—~Tenip (°C)!100

:::~Z~~
0

80
100
Time (mm)

120

140

Friction (kg)

160

Fig. 7 Four-balifriction and temperature data for simple lithium greaseformulated with PTFE-3
– 26 VOLUME 79, NUMBER 3

NLGI

Fig. 8 Load ramp and average friction and temperaturefor lithium grease
formulated with PTFE-3
1.4
1.2

Load (kg)Il00

0.8

0,6

Friction {kg)
Mean

80
100
Time (mm)

120

140

160

Fig. 9 Statistics calculatedfrom friction data for lithium grease with PTFE-3 versus load (kg)

H
StQevlMean

1.0

QSkow
cDKurtosis

::~~

______

–

NLGI SPOKESMAN, JULY/AUGUST 2015

PAO Base Oil
In this study, four-ball tests with load ramps were
applied to PAO base oil used to formulate greases. In test
PAO-a, the load was ramped from 3 to 180 kg at 3 kg!
mm, held for 15 mm and ramped down at the same rate.
Friction and temperature initially increased smoothly
with load, Fig. 10. At 72 kg load, friction increased from
0.19 to 1.59 kg and then decreased to 0.52 kg. After this
event, the friction shift (friction after friction before
event) was 0.52 0.19 kg or 0.33 kg. The friction ratio
(friction after!friction before) was 0.52!0.19 or 2.7.
Ten minutes later, the temperature was 10°C above the
—

–

Fig. 10 Friction (FF, kg,
~) and temperature
(T, °C/100, -)for
test PAO-a. The arrow
indicates thefriction
shift (0.33 kg) after the
event.
– –

temperature projected from data prior to the event, Fig.
12. PAO continued to lubricate these contacts after the
friction event. FTIR spectra of tested and untested oil
were nominally identical (not shown), i.e., PAO did not
oxidize. The average wear scar diameter was 2.70 mm,
and wear particles were visible in the oil.
The duration of this friction event for PAO was
approximately 36 s. Friction events lasted longer in some
replicate tests, e.g., PAO-j shown in Figs. 15 and 16. In
some cases, tests were stopped when friction was near the
load cell limit (2.27 kg or 5 pounds) or welding occurred.

FF (kg) and T(°C)1100, PAO-a
1.8
0

~ 1.4
p1.2
~1.0
0.8
~ 0.6
~0.4
0.2 ~
0.0
0

120

150

Time (mm)

FF (kg) and T(°C)!100, PAO.~a
o .a
Fig. 11 The arrow
indicates the shift in PAO
temperature (grey line)
ten minutes after the
event, which was 10°C
above the projection
from data prior to the
event (straight black
line).

0
0

0•
0
I

~O.5

0.2
20

Time (mm)

NLGI
particles in oil were observed at the bottom of the cup.
Grease was in good condition, although perhaps slightly
softer.

PTFE-3 Grease
Grease prepared with sub-micron PTFE-3 was tested
with a load ramp at 3 kg/mm. In test PTFE-3-a (Fig. 12),
the load was ramped up to 150 kg (50 mm), maintained
for 10 mm and ramped down to 3kg (110 mm total).
At first, friction and temperature trended smoothly
with load. When the applied load was 66 kg (21 22
mm), friction increased from 0.19 to 1.06 kg and then
decreased to 0.25 kg during 20 s. The friction shift was
only 0.25 0.19 = 0.06 kg, the friction ratio was 0.25/0.19
= 1.3, and no temperature shift was observed. A sequence
of additional friction events took place on the up ramp
and at 150 kg, but not on the down ramp. The average
wear scar diameter was 1.41 mm, and the scar appearance
was consistent with abrasive wear (see below). Some wear

In Fig. 13, data from the down ramp are plotted in
reverse order and compared with data from the up
ramp. Temperature was higher on the down ramp due
to heat stored by the grease and apparatus. As the load
was reduced on the down ramp, friction decreased and
approached data on the up ramp. This behavior (plus
wear scars, below) suggested that these friction events
were related to solids entrained in contacts [5-7] and not
adhesive wear [21] or starvation of contacts [20].

–

FF (kg) and T(°C)I100, PTFE-3~a Grease
1.4

Fig. 12 Friction (FF, kg,
and temperature
(T, °C/100, -)for test
PTFE-3-a. The load was
increasedfrom 3 to 150
kg at 3 kg/mm, main
tainedfor 10 mm and
decreased to 3 kg.
____)

1.1

– –

C-)

•~ 0.7
C

— —— — ~

U..

~ 0.4

0.0

60
Time (mm)

–

NLGI SPOKESMAN, JULY/AUGUST 2015

120

-~iF-~

FF (kg) and T(°C)l100, PTFE-3-a Grease
1.4
C
O 1.2

Fig.13 Friction (FF, kg) and
temperature (T, °C/1 00) for
test PTFE-3-a during up
(black) and down (grey)
ramps. Data from the down
ramp were plotted in reverse
orderfor comparison.

~1.0
0

~ 0.8

~0.6

go.4
Z 0.2
LJ~
0.0
0

lime (mm)
Figure 14 shows that friction data for PAO and PTFE-3
grease initially were very similar. This similarity indicated
that PAO bled from PTFE-3 grease. As the load was
increased, friction became higher for grease than oil. This
difference may be the result of resistance from entrained
solids or an artifact. Friction events occurred at 66 and 72
kg loads for grease and PAO, respectively.

Fig. 14. Maximum values of friction (excluding event
peaks) and temperature for PAO-a were 0.84 kg and
149°C, respectively, versus 0.75 kg and 63°C for PTFE
3-a. The Li12OHSt and/or PTFE solids were responsible
for overall lower friction and temperature (excluding
event peaks) during four-ball tests and prevented shifts
in friction and temperature. Control grease was tested
to identify contributions from Li12OHSt thickener in
absence of PTFE.

A single friction event in the case of PAO was more
influential than a sequence of events for PTFE-3 grease,

FE (kg> and T(°C)1100, PAO-a and PTFE-3-a
0.8
0.7
0

Fig. 14 Friction (FF, kg,
and temperature (T, °C/100,
– – -) datafor PAO-a (black)
and PTFE-3-a grease (grey).

0.6

p0.5
.t 0.4
.~

0.3
0.2
0,1

0.0
0

Time (miii)

–

VOLUME 79, NUMBER 3

NLGI
Control Grease
Control grease, 8/92 Li12OHSt/PAO, was tested with
increasing load ramps (where significant features were
observed for PTFE-3 and PAO). Figure 15 shows data for
test Control-b, where the first friction event occurred at a
75 kg load. Friction increased from 0.18 to 1.54 kg, stayed
high for over 2 mm, and then decreased to 0.47 kg. This
friction shift was 0.47 0.18 = 0.29 kg, and the friction
ratio was 0.47/0.18 = 2.6. Approximately 5 mm after
the start of this event, the temperature was almost 10°C
higher than projected from data before the event, Fig.
16. This temperature shift was similar to the temperature
–

shift observed for PAO, Fig. 11. The average wear scar
diameter was 3.31 mm.
Figure 16 compares four-ball data for Control-b grease,
PTFE-3-c grease and PAO-j. The dominant features
friction event peaks and duration, and friction and
temperature shifts were similar for control grease
and PAO. In contrast, these features were much smaller
for PTFE-3 grease. Thus, PTFE-3 powder (and not
Li12OHSt) was primarily responsible for less friction,
temperature and wear observed for PTFE-3 grease.
—

—

Comparison of PAO-j, Control-b and PTFE-3-c greases
1.6
1.4

Fig. 15 Friction (FF, kg,
and temperature
(T, °C/100, -) data
for (left to right): PAO-j
(black), Control-b grease
(dark grey), and PTFE
3-c grease (light grey).

0
C

1.2

1,0

Control Grease

____)

– –

I..

0.8
0.6

U
U-

0.4
—~

0.2
0.0
23

26
27
Time (mm)

Comparison of PAO-j, Control-b and PTFE4-c greases
0.6
0.5

Fig. 16 Friction (PP. kg,
and temperature
(T, °C/100, – -) data
for (left to right): PAO-j
(black), Control-b
grease (dark grey), and
PTFE-3-c grease (light
grey). Arrows indicate
temperature shifts.
____)

–

—

0.4
a

0.3
0.2
0.1
0.0
23

Time (mm)

-31

–

NLGI SPOKESMAN, JULY/AUGUST 2015

PTFE-2 Grease

In test PTFE-2-a, the friction event occurred at a 90
kg load, which was higher than Control-a (51 kg) and
Control-b (75 kg). This suggested that PTFE-2 powder
somewhat improved the load-carrying capacity relative to
control grease. However, friction and temperature shifts
were larger for PTFE-2 grease than PTFE-3 grease. Only
one four-ball test was performed with PTFE-2 grease in
this study.

PTFE-2 grease was formulated with sub-micron
PTFE-2, suitable for HX- 1 registration and food grade
applications. Figure 17 shows data from test PTFE-2-a
where friction increased from 0.30 to 1.98 kg, remained
above 1.50 for 5 mm, and decreased to 0.57 kg. The
friction shift was 0.57 0.30 = 0.27 kg, and the friction
ratio was 0.57/0.30 = 1.9. The temperature shift was
similar to those for PAO and control grease, Fig. 18. The
average wear scar diameter was 2.90 mm, comparable to
control grease.
—

Overall, these results for PTFE-2 and PTFE-3 greases
are consistent with other results from the literature where
specific PTFE powders significantly enhanced loadcarrying capacity of certain greases. [3-5]

Comparison of Control-b, PTFE-3-c and PTFE-2-a greases
2.0

Fig. l7Friction (FF,
kg, _~) and tempera
ture (T, °C/100, – – -)
data for (left to right):
Control-b grease (dark
grey), PTFE-3-c grease
(light grey) and PTFE
2-a (black) greases.

PTFE-2
Grease

Control Grease

a
1~

1,0
U
U-

0.5

0.0
25

Time (mm)

Comparison of Control-b, PTFE-3-c and PTFE-2-a greases
Fig. 18 Friction
(FF, kg,
and
temperature (T,
°C/100, – – -) data
for (left to right):
Control-b grease (dark
grey), PTFE-3-c grease
(light grey) and PTFE
2-a (black) greases.
Arrows indicate
temperature shifts.

0.6

___)

4
0
0

0.4

~0,3
0.2
0.1
0.0
25

Time (mm)
– 32 VOLUME 79, NUMBER 3

NLGI
Wear Scars
A simple experiment was performed to identify wear
mechanisms during these tests. Four-ball tests with 3
kg/mm load ramps were stopped at 45, 75, 78 and 81
kg loads. Wear scars were measured and photographed
under a light microscope at 5x magnification, and
photographs were enlarged to examine the appearance
of the scars. Figure 19 shows a representative scar from
PAO-f (final load 45 kg, average scar diameter 0.41 mm).
These scars were round with smooth edges. For PAO-g
(75 kg, 0.55 mm), scars were similar but larger than PAO
f. Both tests PAO-f and —g were stopped before friction
events occurred, and the appearance of these scars was
consistent with abrasive wear. [211
Figure 20 shows a representative wear scar from PAO-h
(78 kg, 2.3 mm), which was stopped immediately after the
first friction event occurred. PAO-h scars were larger and
more irregular and rugged” than PAO-f and PAO-g scars.
Scars for PAO-h and -i (81 kg, 2.7 mm) were similar in
appearance. Black solids were observed on one side of
each scar. It appeared that these deposits were metal that
softened and was displaced by the rotating ball during the
test. These observations were consistent with temperature
shifts following friction events for PAO and adhesive
wear. [21]

The oval shape and uneven edges of PTFE-3-a scars
suggested that pressure and stress concentrations
may have redistributed (and/or contacts may have
shifted) during this test. PTFE particles, aggregates or
agglomerates can produce effects that are more localized
than films. For example, Palios et al. [4] studied two
types of PTFE particles (diameters —1 and 5 rim) in
EHD (elastohydrodynamic) lubrication test rigs. They
observed individual PTFE particles in contacts and
elastic impressions made by particles on contact surfaces,
and they concluded that PTFE particles helped support
applied loads.
In the present study of PAO, a single friction event
corresponded to adhesive wear and failure of the oil film
to support the applied load. Before an event occurred,
abrasive wear gradually increased diameters of round
scars, Fig. 21. A friction event corresponded to adhesive
wear, which was relatively brief (—30 s) but substantially
increased scar diameters. For PTFE-3 grease, wear scars
and no temperature shifts showed that PTFE-3 powder
prevented adhesive wear and helped support loads. A
reasonable explanation for friction events is that PTFE-3
was entrained in contacts on up ramps (as loads increased
and gaps narrowed) but not down ramps, consistent with
observations.

In test PTFE-3-a, the load was ramped up to 150 kg and
back down to 3 kg; this test was longer (2 h) than PAO-f
through —i. For scars from PTFE-3-a, there was no visual
evidence of softened metal or adhesive wear, Fig. 20. The
average wear scar diameter was 1.41 mm. In the direction
parallel to sliding, scar size (0.93 mm) was similar to
abrasive wear in PAO-f and -g, Fig.21.

Fig. 19 Enlarged photograph of a magnified
scarfrom four-ball test PAO-f This test
was stopped before thefirstfriction event.
Average wear scar diameter was 0.41 mm.

– 33 NLGI SPOKESMAN, JULY/AUGUST 2015

NLGI
Fig. 20 Enlarged photograph of a
magnified wear scarfrom four-ball
test PAO-h. This test was stopped
immediately after the firstfriction
event. Average wear scar diameter
was 2.4 mm.

Fig. 21 Enlarged photograph of a
magnified wear scarfrom PTFE
3-a (grease). Load was ramped up
to 150 kg, maintained and ramped
back down (about 2 h). A sequence
offriction events occurred, which
caused smalifriction shifts but
no observable temperature shifts.
Average wear scar diameter was
1.41 mm.

Fig. 22 Average wear scar
diameter (mm) versusfinal
loadfor PAO before (PAO-f
and -g, grey circles) and
after (PAO -h and -i, black
triangles) the firstfriction
event (adhesive wear). For
PTFE-3-A grease, squares
are average scar diameter
(middle) and components
perpendicular (top) and
parallel to sliding (bottom)
after a longer, more severe
four-ball test (2 h, maximum
load 150 kg).

Wear Scar Diameter (mm) vs Final Load (kg)
A

E
~2.5

~2.0
E

(~1.0

~15

~ 0.5

~0.0
0

Final Load (kg)
– 34 VOLUME 79, NUMBER 3

120

160

NLGI
PTFE-2 Liquid Dispersion
Food grade PTFE-2 powder was dispersed (20 wt.%)
in white mineral oil with a polymeric dispersant.
Samples from a stability test at room temperature (after
approximately 6 m) and base oil were evaluated with the
four-ball procedure used for greases and PAO.
For test Mineral oil-A, Fig. 23, a friction event was
observed at a 99 kg load. Friction increased from 0.37
to 2.13 kg and then decreased to 0.42 kg. The friction
shift was 0.42 0.37 = 0.05 kg, and the friction ratio was
0.42/0.37 = 1.1. A temperature shift was observed, Fig. 24.
This event plus friction and temperature shifts resembled
data for PAO that were caused by adhesive wear.
—

In test 20% PTFE-2 dispersion-A, no events were
observed on the up ramp, Fig. 23. The largest fluctuation
occurred on the down ramp, and it did not shift friction
or temperature, Fig. 25. The average scar diameter was

1.04 mm, and scars resembled abrasive wear. There was
no evidence of adhesive wear. Photographs in Fig. 26
show that wear particles blackened mineral oil but not
PTFE-2 dispersion. According to FTIR data (not shown),
mineral oil did not oxidize during this test.
In this liquid dispersion, PTFE-2 prevented adhesive
wear and helped support applied loads.PTFE-2 did
not provide these benefits in grease, Figs. 17 and 18. A
reasonable explanation is that the dispersion and mobility
of PTFE-2 were better in the liquid than in the grease.
Better PTFE-2 dispersion and fewer solids entrained
in contacts could also explain observations of friction
fluctuations instead of events, Fig. 23. Different results
for PTFE-2 in these liquid and grease formulations
demonstrate that dispersion is critical to performance of
PTFE additives in lubricants and greases.

of Mineral Oil and 20% PTFE Dispersion
2.0
0
0

Fig. 23 Friction (kg, _J and
temperature (°C//100, – – -)
datafor mineral oil (black)
and 20% PTFE-2 in mineral
oil with dispersant (grey).

Mineral Oil

0.0

Time (mm)

– 35 NLGI SPOKESMAN, JULY/AUGUST 2015

100

120

140

Comparison of Mineral Oil and 20% PTFE Dispersion
0.6

Mineral Oil

0.5

Fig. 24 Friction (kg, __) and
temperature (°C/100, – – -)
data for mineral oil (black)
and 20% PTFE-2 in mineral
oil with dispersant (grey).
Arrow indicates temperature
shift afterfriction eventfor oil.

0
0

~0~3

20% PTFE-2

~0.2
L1~

0,1
0.0

33
34
Time (mm)

Comparison of Mineral Oil and 20% PTFE Dispersion
0.8

Fig. 25 Friction (kg, ~J and
temperature (°C/1 00, – – -)
data for mineral oil (black)
and 20% PTFE-2 in mineral
oil with dispersant (grey).
Arrow indicates temperature
shift afterfriction eventfor oil.

0,6

LL
Li.

20% PTFE-2

0.3

75
Time (mm)

Fig. 26 Wear particles black
ened mineral oil (right) but not
PTFE-2 liquid dispersion (left).

– 36 VOLUME 79, NUMBER 3

NLGI
Discussion

Values of the friction shift (friction after friction
before an event) in Fig. 28 show benefits of PTFE. For
PAO, friction shifts were between 0.3 and 0.5 kg because
adhesive wear damaged contact surfaces. For PTFE-3
grease, friction shifts were much smaller because PTFE
protected contacts from damage. Values of the friction
ratio (friction after/friction before) convey the same
information, Fig. 29. Results for control grease and PTFE
2 grease were similar to PAO and not PTFE-3 grease.
Additional replicate tests are needed to better compare
mineral oil with this PTFE-2 liquid dispersion.
—

Friction events for PTFE and control greases, a liquid
dispersion and two base oils differed clearly from artifacts
of start-up and run-in (from 3 to 15 kg load). Figure 27
shows the applied load at the start of each event. Loads
were smaller (26 90 kg) for PTFE-3 grease than for PAO
(72 104 kg). In the case of PTFE-3 grease, events were
caused by entrained PTFE that helped support loads. For
PAO, events corresponded to film failure and adhesive
wear.
—

—

LOAD (kg)
160

Fig. 27 Applied load (kg)
that coincided with the first
friction event in each test

140
120
100

1~]

80
60
40
20
0

~fl~A

I~I

I!I ~

‘9

~
‘0

FrIction Shift (kg)
0.5
0.4
0.3
Fig. 28 Friction shift (friction
after minus friction before the
event) in each test.

0.2
0.1

0.0

0~ O~O~O~J~

37 NLGI SPOKESMAN, JULY/AUGUST 2015

NLGI
Friction Ratio
7
6
5
Fig. 29 Friction ratio (friction
after /friction before the
event) in each test.

4
3

~
~

2[1Lj~
1
0~ o~°

PTFE is useful as an anti-wear additive in grease
provided it is dispersed adequately. Performance depends
on PTFE grade, grease formulation, milling and tribology
test procedures. In four-ball friction and wear tests,
benefits of PTFE were most easily observed when run-in
effects were avoided by gradually applying the load.
Seven experimental PTFE powders were formulated
and milled in simple lithium greases based on PAO oil.
All seven powders (4% by weight) slightly softened grease
relative to control grease but did not alter NLGI grade (2)
or detract from storage stability.

Food grade PTFE powder improved friction,
temperature and wear in a liquid dispersion in mineral
oil. In simple lithium grease based on PAO, this
particular grade of PTFE was less effective, possibly due
to dispersion. Specific PTFE powders provide significant
lubrication benefits that depend upon dispersion as well
as lubricant formulation.

~fliiiu~

Conclusions

Sub-micron PTFE powder significantly improved
lithium grease lubrication by preventing adhesive wear
and limiting friction, temperature and abrasive wear
at loads up to 180 kg. Results indicated that PTFE was
entrained in contacts and helped support applied loads.

Four-ball friction and wear tests provided useful
information about grease lubrication. The most
meaningful results were obtained using load ramps to
avoid run-in effects. Monitoring friction and temperature
data was used to distinguish between abrasive and
adhesive wear, and between film failure versus
entrainment of solids in sliding contacts.

Acknowledgements
The author gratefully acknowledges Shamrock
Technologies, Inc. for supporting this experimental
project and providing the opportunity to present this
paper at NLGI’s 81st Annual Meeting, June 14-17, 2014,
Palm Beach Gardens, FL.

References
[1] NLGI, Lubricating Grease Guide, Fourth and Fifth
Editions, National Lubricating Grease Institute, Lee’s
Summit, MO, USA, 1996 and 2009.
[2] ASTM D2266 Standard Test Method for Wear
Preventive Characteristics of Lubricating Grease
(Four-Ball Method), ASTM International, West
Conshohocken, PA, USA.
[3] ASTM D1894, Standard Test Method for Static and
Kinetic Coefficients of Friction of Plastic Film and

38 VOLUME 79, NUMBER 3

NLGI
Sheeting, ASTM International, West Conshohocken,
PA, USA.
[4] Rudnick, L.R., Ed., Lubricant Additives: Chemistry
and Applications, Second edition, CRC Press, Taylor
and Francis Group, Boca Raton, FL, USA, pp. 18 1-2.
[5]

Palios, S., Cann, P. M. and Spikes, H. A., Behaviour
of PTFE Suspensions in Rolling/Sliding Contacts,
The Third Body Concept, D. Dowson et al., Editors,
Elsevier Science B.V., 1996.

[6] Wan, G.T.Y. and Spikes, H.A., The Behaviour
of Suspended Solid Particles in Rolling Sliding
Elastohydrodynamic Contacts, STLE Transactions,
31, pp. 12-24, 1987.
[7] Jia, Z. and Yang, Y., Composites Part B:Engineering,
Vol 43, Issue 4, June 2012, pp. 2072-2078.
[8] International Organization for Standardization,
ISO 4406:1999(E) Hydraulic fluid power Fluids
Method for coding the level of contamination by
solid particles.
—

—

Annual General Meeting, Paris, France, 2011 and
published in Eurogrease 1, 36-48, 2012.
[13] ASTM D2596 Standard Test Method for
Measurement of Extreme-Pressure Properties of
Lubricating Grease (Four-Ball Method), ASTM
International, West Conshohocken, PA, USA.
[14] Ballester, E., Sui, Manshi and Fillion, C., Effect
of PTFE Particle Size on Wear and Coefficient of
Friction, Paper #0025, Presented at the NLGI 68th
Annual Meeting, Asheville, NC, USA, 2000.
[15] Moon, M., PTFE for Lubricating Grease, Extended
Abstract, Presented at the Society of Tribologists
and Lubrication Engineers National Meeting,
Orlando, FL, USA, May, 2014.
[16] NSF International, Ann Arbor, MI, USA, www.nsf.
org.
[17] ASTM D217 Standard Test Methods for Cone
Penetration of Lubricating Grease, ASTM
International, West Conshohocken, PA, USA.

[9] Aswath, P., Patel, K., Munot, S. and Elsenbaumer,
R.L., Development of a High Performance Low
Molybdenum Disulfide Grease, Paper #0530,
Presented at the NLGI 72nd Annual Meeting, San
Antonio, TX, USA, 2005.

[18] Sayles, R. S. and Macpherson, P. B., Influence of
Wear Debris on Rolling Contact Fatigue, in Rolling
Contact Fatigue Testing of Bearing Steels, STP771,

[10] Krawiec, ST., The influence of polymers fillers in a
grease lubricant, Archives of Civil and Mechanical
Engineering, Vol. VII, No. 4, pp. 121-128, 2007.

[19] Moon, M., Taking Lubricant Cleanliness to the
Next Level, Gear Solutions Magazine, June, 2009.
(available at www.gearsolutionsonline.com)

[11] Fish, G. and Ward Jr., WC., Extreme Pressure
Performance of Greases: Testing and Additive
Solutions, Paper #1002, Presented at the NLGI
77nd Annual Meeting, Bonita Springs, FL, USA,
2010.

[20] Lugt, P.M., Grease Lubrication in Rolling Bearings,
John Wiley & Sons Ltd, West Sussex, UK, 2013.

J. J. C. Hoo, Editor, ASTM International, West
Conshohocken, PA, USA, 1982.

[21] Bayer, R. G., Mechanical Wear Fundamentals and
Testing, Second edition, Marcel Dekker, Inc., New
York, NY, USA, 2004.

[12] Herrero, F., Fish, G. and Ward Jr., WC., Extreme
Pressure Performance of Greases: Testing and
Additive Solutions, Presented at the 23rd ELGI
– 39 NLGI SPOKESMAN, JULY/AUGUST 2015

PTFE for Lubricating Greases

ABOUT NLGI

HELPFUL LINKS

CONTACT

injecting the cart for now, hidden but triggers it