Wear of Self-Mated Molybdenum Disulphide Tribological Interfaces

steffens_j.pdf

600 (a)

S200
0.--
1 -200 original
a) surface
_c *worn surface
-600 wear scar
limits


-00 100 300


500 700 900


6001 (d)


APPENDIX B
SURFACE PROFILES


(b)
*' *. 5* .<


(c)
a "


IV*^;l s.Arttt RAP% .I%*IM


0 100 300 500 700 9C

(e)


I.


)0 0 100


300 500 700 900


(f)


)nn I


N 4t


'' t. I t "
-600

-600


I*


p ..

ax


0 100 300 500 700 900 0 100 300 500 700 900 0 100 300 500 700 900
position (pm) position (pm) position (pm)
Figure B-1. Surface profiles of the MoS2/Ni composite coating: (a) original surface, (b)
cycle 2000, (c) cycle 4000, (d) cycle 6000, (e) cycle 8000, (f) cycle 10,000.


500


- 300
c

S100
C


-100

-0 50 150 250 350 450

500
(d)

300

100
7 ----^.z ....


0 50 150 250 350 450

(e)





tl/ ^


1 50 150 250 350 450

(f)





3-- i-:


SI


0 50 150 250 350 450 0 50 150 250 350 450 0 50 150 250 350
position (pm) position (pm) position (pm)
Figure B-2. Surface profiles of the MoS2/Ti composite coating: (a) original surface, (b)
cycle 2000, (c) cycle 4000, (d) cycle 6000, (e) cycle 8000, (f) cycle 10,000.


S-


C.
a-


(a) original
surface
*worn surface
wear scar
limits


^^^^


c









CHAPTER 2
EQUIPMENT CHARACTERISTICS

2.1 General Tribometer Design

When investigating tribology, there are two standard tribometer designs; a pin-on-

disk (POD) where the disk rotates with an eccentrically located pin/ball, and a pin-on-flat

where the flat is translated using a reciprocating stage. General tribometer design

philosophy follows a need to characterize the normal and friction forces, the path of

travel for each cycle, and the path characteristics.

2.1.1 Force Characterization

To accurately record force measurements, a force transducer must be placed

inside the path of contact to ground. In addition, there should be no other moving

assemblies, such as bearings, bushings, or gimbals in this path to introduce parasitic

forces. Figure 2-1 illustrates that one of the contacting bodies (in this case, the ball) is

held rigidly, and when its path is traced back to ground, there aren't any moving

systems while the tribometer is performing an experiment. Moving systems between

the load cell and the contacting body would introduce parasitic forces measured by the

load cell which leads to inaccurate data. Additionally, moving systems between the load

cell and ground would introduce inertial forces, which would also be measured by the

load cell leading to inaccurate data.

To contrast good tribometer design, a standard POD tribometer is illustrated in

Figure 2-2. Following the logic presented for following the path from ground in Figure 2-

1, it can be determined that the tribometer in Figure 2-2 has a moving mechanical

assembly in the path from contact to ground, in this case a gimbal. A tribometer such





















For Nick Kate









CHAPTER 7
CONCLUSION

A tribometer to measure in-situ wear rates of tribological interfaces was designed

and constructed. It was demonstrated that in a low humidity environment that MoS2

containing coatings exhibit low friction and low wear, and that the wear was reduced by

one to two orders of magnitude with the inclusion of Sb203. Additionally, in a low

humidity environment, it is necessary to reciprocate for tens to hundreds to thousands

of cycles to remove just one layer of MoS2. Of the coatings tested, the MoS2/Sb203/C

coating exhibited the lowest friction and the MoS2/Sb203/Au exhibited the least amount

of wear.









WEAR OF SELF-MATED MOLYBDENUM DISULPHIDE TRIBOLOGICAL
INTERFACES




















By

JASON GEOFFREY STEFFENS


A THESIS PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
MASTER OF SCIENCE

UNIVERSITY OF FLORIDA

2010









as the one present in Figure 2-2 has a higher uncertainty in the friction force

measurement.


ball sample
point of contact


ground

Figure 2-1. Illustration of the path required for the force to be reacted from the point of
contact to ground.


housed
gimbal


known dead-weight
normal load
'-An


single axis load cell


Figure 2-2. Standard POD tribometer illustrating a moving mechanical assembly in the
path of contact to ground.

2.1.2 Path and Path Characteristics

The use of a stepper motor to control the motion is instrumental to reproduce the

same path for each consecutive cycle. A stepper motor breaks up the total signal of the

output motion into small discrete pulses that turn a central iron gear tenths to one









S 6 axis load cell
leaf-type flexure -

micrometer
stage


'- linear stage
pin sample


(b) adjust stage down
to apply load

1I


Figure 2-3. Tribometer and interferometer schematic, a) tribometer schematic, b)
illustration of loading mechanism, c) tribometer sitting on interferometer stage.

Due to the geometry of the loading, high contact pressures are achieved with very

low loads, where a load of just 5 N would yield a Hertzian contact stress of 1 GPa for a

self-mated steel contact. Because a low loads induce contact pressures of this order, a

load cell with the ability to accurately measure such loads is instrumental. To measure

these loads a JR3 50M31A (JR3, Woodland, CA) was chosen as the load cell. It is a

six-axis load cell that has an axial load capacity of 220 N, while in the other two

directions the maximum load is 110 N. The load cell is oriented in such a manner that

the two most sensitive axes are used to measure the normal load and the friction force.

This load cell has a very high resolution of 28 mN in the axial direction and 14 mN in the

axes used in making the force measurements.

The loading of this tribometer makes use of a simple leaf-type flexure. The

flexures are connected to a linear micrometer stage with a maximum capacity of 10 N

that translates vertically to bring the ball in and out of contact. The flexures are steel









Abstract of Thesis Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Master of Science

WEAR OF SELF-MATED MOLYBDENUM DISULPHIDE TRIBOLOGICAL
INTERFACES

By

Jason Geoffrey Steffens

August 2010

Chair: W. Gregory Sawyer
Major: Mechanical Engineering

Engineers designing mechanisms intending to go into orbit around the earth need

to know reliably how geometries of tribological interfaces change as they perform. The

driving force behind changing interfaces is wear. As interfaces change, precision

designed assemblies may end up not being so precise after all. So, design engineers

need to know when a surface will wear, how much it will wear, and at what rate it will

wear.

Molybdenum disulphide (MoS2) has been known as a low friction solid lubricant for

vacuum applications as early as 1941. Friction studies on MoS2 are extensive

throughout the literature, and they continue today. The main focus appears to be on the

friction response of MoS2, and these studies dominate over the number of studies

where wear is the predominant characteristic. When wear is quantified it is often

discussed in terms of the number of cycles experienced with low steady state friction, or

in terms of the distance traveled, instead of providing a useful tribological wear rate.

Even in the instances where the total wear volume is measured, a single-point wear rate

measurement is performed, but it is often a poor representation of how the specimen

actually performs through different regimes. The intention of this study was to provide









LIST OF FIGURES


Figure page

2-1 Path required for the force to be reacted from the point of contact to ground..... 16

2-2 Standard POD tribometer .......... .... ............... ........... .... 16

2-3 Tribom eter and interferom eter schem atic............................... ..................... 18

2-4. Schem atic of test equipment................................................. ................ 19

4-1 Illustration of the reciprocation path................ ............................. 24

4-2 Surface scan location along the wear track........ ..... .. .................... ............... 26

4-2 Comparison between worn and virgin surfaces.......................... ... ............... 26

6-1 Illustration of different wear rate analyses ........ ............. ........ .... .............. 31

A-1 MoS2/Ni average friction coefficient per cycle ............... .................... 34

A-2 MoS2/Ni total volume lost as a function of sliding cycle................ ................ 34

A-3 MoS2/Ti average friction coefficient per cycle............................. ... ................ 35

A-4 MoS2/Ti total volume lost as a function of sliding cycle. ............... ................ 35

A-5 MoS2/Sb203 average friction coefficient per cycle....... ....... .................. 36

A-7 MoS2/Sb203/graphite average friction coefficient per cycle .............................. 37

A-8 MoS2/Sb203/graphite total volume lost as a function of sliding cycle............... 37

A-9 MoS2/Sb203/Au average friction coefficient per cycle...................................... 38

A-10 MoS2/Sb203/Au total volume lost as a function of sliding cycle ..................... 38

B-1 Surface profiles of the MoS2/Ni composite coating .................. .................. 39

B-2 Surface profiles of the MoS2/Ti composite coating. .................................... 39

B-3 Surface profiles of the MoS2/Sb203 composite coating .................. ......... 40

B-4 Surface profiles of the MoS2/Sb203/graphite composite coating ........................ 40

B-5 Surface profiles of the MoS2/Sb203/Au composite coating .............................. 41









very uniform wear, where every 1,000 cycles 5-13% of the total lost material was

removed except for the last 1,000 cycles where 20% of the total lost material was

removed (Figure A-4).

5.4 MoS2/Sb203 Composite Coating

The MoS2/Sb203 composite had an average friction coefficient of 0.062 (Figure A-

5). The maximum wear scar depth was 468 nm and the maximum track width was 123

pm (Figure B-3). Its steady state wear rate was 5.05E-08 mm3/Nm. This coating

achieved steady state almost immediately and all cycles were used in determining its

steady-state wear rate. Like the MoS2/Ni coating, nearly half of the total wear occurred

within the first 1,000 cycles with 47% being removed (Figure A-6). At cycle 8,000 there

was an increase in the volume of material in the wear track likely from the coating

having been worn from the ball surface and being deposited into the wear track.

5.5 MoS2/Sb203/graphite Composite Coating

The MoS2/Sb203/graphite composite had the lowest average friction coefficient of

0.010 (Figure A-7). The maximum wear scar depth was 423 nm and the maximum track

width was 35 pm (Figure B-4). This coating had two different wear regimes which had

similar wear rates. For cycles 0-5,000 steady-state wear rate was 4.65E-08 mm3/N m

and for cycles 8,000-10,000 the steady-state wear rate was 2.82E-08 mm3/Nm.

Between cycles 7,000 and 8,000 there was a gain in volume in the wear track likely

from the ball coating being worn and depositing onto the wear scar. Over half of the

total wear occurred within the first 1,000 cycles of sliding where 52% of the total lost

material was removed (Figure A-8).












300 )


(b)



c--^^^


S 100 200


*1


300 400 500 600 700 0 100 200 300 400 500 600 700


5~L1~- *+ftmqh..DPCm.s tqm.~whg imb..IUCr~IC~~L


0- 100 200 300 400 500 600 700 0 100 200 300 400 500 600 700 0 100 200 300 400 500 600 700
position (pm) position (pm) position (pm)
Figure B-3. Surface profiles of the MoS2/Sb203 composite coating: (a) original surface,

(b) cycle 2000, (c) cycle 4000, (d) cycle 6000, (e) cycle 8000, (f) cycle 10,000


-"' (a)
300
E original
100 surface
. 0 worn surface
'-100 wear scar
limits
-300

-ann


0 50 100 150 200


250


(d)
300
E
100

0 -100

-300


Figure


(b)









0 50 100 150 200 25

(e)






V


(c)




i


0 5 100 150 200 250


50 100 150 200 250 50 100 150 200 250 0 50 100 150 200 2
position (pm) position (pm) position (pm)
B-4. Surface profiles of the MoS2/Sb203/graphite composite coating: (a) original
surface, (b) cycle 2000, (c) cycle 4000, (d) cycle 6000, (e) cycle 8000, (f)
cycle 10,000.


'znn


S100
S n


l)
'

-1300

-300

_0nn


uu


1


^


(f)



~~e ~lc

i


- UU0









entire tribometer with the interferometer objective aligned with a viewing hole in line with

the wear track. A seal was made between the objective and the chamber using an

unlubricated Magnum condom (Church and Dwight, Co., Princeton, NJ). Dry nitrogen

gas was bled into the chamber until the relative humidity was less than 1%. The stage's

speed, acceleration, and deceleration were set to 10 mm/s, 100 mm/s2, and 100 mm/s2

respectively. An initial scan of the surface was made representing cycle 0, so that the

subsequent profiles could be compared to the native surface. A 5 N normal load was

applied, and the test began. Every 1000 cycles (equivalent of 10 m) the test was

stopped, and a surface measurement was performed on the flat counterface. This

continued until 10,000 cycles was reached.

The data for the friction coefficients were collected at 500 Hz and were computed

by the technique described by Dickrell et al (13), where the friction coefficient reported

is the half the difference between the forward and reverse directions of a cycle.

4.2 Determination of Wear Rates

To determine the volume of material lost, the scan of cycle zero was compared to

the scan of interest (i.e. cycle 1000 was compared to cycle 0 and cycle 2000 was also

compared to cycle 0) as shown in Figure 4-2. The scan was performed at middle of the

wear scar (Figure 4-3), not the middle of the reciprocating path. The wear scar area

was calculated as the difference between the two surfaces, and was then extruded to

the total length of the reciprocating path, 10 mm. This technique was introduced by

Williamson and Hunt in examining the asperity persistence even after plastic

deformation occurred (14). Sayles then expanded upon this technique when he applied

it to surfaces (15). To determine the wear rate uncertainties, a Monte Carlo simulation

was performed as previously described by Schmitz et al (16).









All of these coatings exhibited very low wear. To illustrate this point the wear of

the coating can be thought of in terms of how many layers lubricant, in this case MoS2,

are removed per cycle. Knowing the wear rates of the coatings, the wear scar width,

and that one layer of MoS2 is approximately 6 A in depth (17), each coating on average

takes tens to hundreds of cycles to remove one atomic layer. The number of cycles

calculated to remove one layer of MoS2 for all the composites are located in Table 6-2.

Table 6-2. Number of cycles to remove one layer of MoS2 based on different wear rate
calculations.
Cycles
Composite Depth (nm) Width(pm)
ksp kCD kss
MoS2/Ni 906 206 21 46 67
MoS2/Ti 246 155 117 117 117
MoS2/Sb203 468 123 148 293 293
91
MoS2/Sb203/graphite 423 35 90 184 1
149
MoS2/Sb203/Au 177 69 206 448 2733


After the MoS2/Ni, MoS2/Sb203, and the MoS2/Sb203/Au coatings wore-in, the

number of cycles required to remove one atomic layer of MoS2 increased by 2-10 times.

The MoS2/Ti exhibited extremely steady wear resulting in the number of cycles to

remove one layer of MoS2 being independent of the type of wear rate used. There was

no wear-in regime for this coating, and as stated previously, every 1,000 cycles

removed 5-13% of the total volume.









CHAPTER 5
EXPERIMENTAL RESULTS

5.1 Average Temperature and Humidity

While all of the experiments were performed in a temperature controlled room, the

temperature of the experiments was not controlled, but instead was monitored. Table 5-

1 provides a summary of the temperature and humidity of the experiments.

Table 5-1. Average temperatures and humidities for the duration of the experiments.
Composite Average Temperature (C) Average %RH

MoS2/Ni 29.8 0.10
MoS2/Ti 29.5 0.14
MoS2/Sb203 27.3 0.10
MoS2/Sb203/C 29.3 0.10
MoS2/Sb203/Au 29.6 0.11


5.2 MoS2/Ni Composite Coating

The MoS2/Ni composite had the highest average friction coefficient of 0.100

among all of the samples (Figure A-1). The maximum wear scar depth was 906 nm,

and the maximum track width was 206 pm (Figure B-1). Its steady state wear rate was

3.70E-07 mm3/N- m, steady-state wear was considered be at 3,000 cycles. Nearly half

of the total wear occurred within the first 1,000 cycles of sliding where 49% of the total

lost material was removed (Figure A-2).

5.3 MoS2/Ti Composite Coating

The MoS2/Ti composite had an average friction coefficient of 0.051 (Figure A-3).

The maximum wear scar depth was 246 nm, and the maximum track width was 155 pm

(Figure B-2). Its steady state wear rate was 1.71 E-07 mm3/N m. Steady-state wear was

achieved quickly and was analyzed beginning at 2,000 cycles. This coating very had
































2010 Jason Geoffrey Steffens








CHAPTER 4
METHODS
4.1 Test Method

Each counterface described in Chapter 3 can fit six 5 mm long wear tracks and

each ball can fit multiple contact points depending on the contact area (dependent on

contact stress). The total distance traveled in one cycle is 10 mm. Relative to the

counterface, the ball began at one end of the track and traveled 5 mm to the other end,

where it stopped and then returned to the original starting point to begin a new cycle

(Figure 4-1).


S5 mm 0

( 1. cycle begin
S2. cycle reverse
3. cycle end

O 0

Figure 4-1. Illustration of the reciprocation path.
Before the experiment was performed, a calibration step was required to align the

optics of the interferometer to the desired image location. A short experiment of a few

cycles using an uncoated aluminum ball and an uncoated steel substrate was

performed to create a wear scar. Then the optics of the interferometer were aligned to

the scar.

To perform the experiment a coated stainless steel substrate was bolted to the

reciprocating stage. Then coated aluminum ball was fastened into a PEEK holder, and

the holder was bolted to the load cell. The environment chamber was placed over the









CHAPTER 6
DISCUSSION AND ANALYSIS OF RESULTS

6.1 Single Point, Complete Data, and Steady-State Wear Rates

Wear rate, k, is defined as the total volume in millimeters cubed divided by the

quantity of the normal load in Newtons multiplied by the sliding distance in meters.

Different ways exist to interpret wear curves. Here three methods are presented: single

point wear rate (ksp), complete data wear rate (kcD), and steady-state wear rate (kss).

Typically, mass or volume loss measurements are performed during a test. However,

due to the setup or conditions, this is sometimes not possible. In this situation the final

volume loss would be determined and using the single data point, a single point wear

rate is determined. This is not the best representation of the data; because, it does not

take into the consideration of the wear-in period which may dominate over the steady-

state wear. The kcD is reported when intermittent wear measurements are performed

during a test, a least squares regression is fit through the entire data set. This is the

most commonly reported wear rate, but like ksp, it does not take into consideration the

wear-in period or the steady-state wear. The kss is achieved the intermittent wear

measurements are performed and the data is broken up into different regions of wear-in

and steady-state. A material couple can exhibit multiple steady-state wear rates during

operation. This most often occurs during environment changes.

6.2 Summary of Friction and Wear Results

The temperature inside of the chamber was higher than ambient due to the

electrical components, mostly the linear stage, heating the environment.

Using the wear rates defined in Section 6.1, the tribology data from Sections 5.2-

5.6 are summarized in Table 6-1. It follows that kss should be the smallest or equal to









LIST OF ABBREVIATIONS


A Angstrom

AFRL Air Force Research Laboratory

Au gold

DC direct current

kcD complete data wear rate

ksp single point wear rate

kss steady-state wear rate

LEO low earth orbit

MoS2 molybdenum disulphide

Ni nickel

PEEK polyetheretherketone

PLD pulsed laser deposition

POD pin-on-disk

Ra average surface roughness

RH relative humidity

Sb203 antimony trioxide

SWLI scanning white light interferometer

Ti titanium

UDCMS unbalanced direct current magnetron sputtering

UV ultraviolet






















50 100 150 200 250


50 (d)I
50
" w /--


-100
-150
- 00


0


Figure


S 50 100 150 200 250

(e)
--_ ^e


(c)








50 100 150 200 250


50 100 150 200 250 0 50 100 150 200 250 0 50 100 150 200 250
position (pm) position (pm) position (pm)
B-5. Surface profiles of the MoS2/Sb203/Au composite coating: (a) original
surface, (b) cycle 2000, (c) cycle 4000, (d) cycle 6000, (e) cycle 8000, (f)
cycle 10,000.


original
surface
*worn surface -
wear scar
limits


E

'

0I









TABLE OF CONTENTS

page

A C KNOW LEDG M ENTS .......... ..................... ....... .. ......................................... 4

LIST O F TA B LES .......... ..... ..... .................. ............................................. ...... .. 7

LIS T O F F IG U R E S .................................................................. 8

LIST OF ABBREVIATIONS ....... ............ .................... .......... 9

A BST RA C T ............... ... ..... ......................................................... ...... 10

CHAPTER

1 INTR O D U CT IO N ............................................................................................. 12

1.1 Motivation of Research ................................. .. ............ 12
1.2 Background....................................... ............... 13
1.3 Experim ent O verview .................................................................... ........ 14

2 EQUIPMENT CHARACTERISTICS.................... ............. 15

2.1 General Tribometer Design..................... ....... .................... 15
2.1.1 Force Characterization ............... ........................ ............. 15
2.1.2 Path and Path Characteristics ......................................... .. ...... ..... 16
2.2 In Situ Tribometer Design ..................... .................... 17
2.3 Interferom eter ..... .............................................. .............................. 20
2.4 Environment.............................. .... ...... ............ 20
2.5 Summary of Equipment and Tribometer Design .................. ........... 20

3 SA M PLE D ESC R IPT IO N S ................................ ......... .............. ............... 22

3.1 Substrate and Pin Descriptions........................ ...................... 22
3.2 Coating Descriptions........................ ....... .......... ............... 22
3.2.1 MoS2/Ni ..... ........................................ 22
3.2.2 MoS2/Ti ............. ................................ 22
3.2 .3 M oS 2/S b20 3 ........................................................................................ 23
3.2.4 M oS2/Sb20 3/graphite .................................... ...... ......... ............... 23
3.2.5 MoS2/Sb203/Au .......................... ........ .. ............ ......... 23

4 METHODS...................................................... 24

4.1 Test M ethod ................................................................. ........ .. ...... .......... 24
4.2 Determ nation of W ear Rates........................................................ ..... 25
4.3 Experiment Summary ................ ............................... 26









2.3 Interferometer

Many tribological coatings exhibit very low wear rates where thousands of cycles

on average are needed to achieve the removal of just one nanometer of the film. To

accurately capture the effect of the material removal, an interferometer with a very high

feature height resolution is required. The Zygo New View 5010 (Zygo, Middlefield, CT)

was chosen because it has a reported feature height resolution of 1 A, which would be

sufficient to capture the desired any topographical data; however, practical experience

suggests that uncertainties are on the order of nanometers. The entire tribometer

mounts to the stages of the interferometer, where its motorized stages can manipulate

the sample orientation in situ.

2.4 Environment

The tribometer and SWLI were located in a temperature controlled room used for

metrology studies where the temperature of the room was kept at 20 2C. The entire

tribometer was fit inside of a small, acrylic chamber with feedthroughs for electrical

components, as well as, a feedthrough for the introduction of different gas species.

Mounted to the acrylic chamber was the DY5 moisture probe with the MMY2650

HygroGuard 2650 system (GE Sensing, Billerica, MA) which measures temperature and

relative humidity.

2.5 Summary of Equipment and Tribometer Design

A tribometer was designed with a multiaxis force sensor placed in the path of

contact to ground. The multiaxis force sensor measures the normal and friction forces

simultaneously. An acrylic environment chamber was constructed which houses the

entire tribometer. The experiments can be performed in any non-corrosive gaseous

environment where the temperature and relative humidity can be monitored The









quantitative wear rates for five self-mated MoS2 composite coatings in a low humidity

environment. Used in these experiments is a custom built and instrumented

reciprocating tribometer with in situ scanning white light interferometry capability which

was used to make surface topographic measurements on a cycle-by-cycle basis.










0.16

0.14

0.12

0.1

0.08

0.06

0.04


0.02 1


0 2000 4000 6000 8000 10000

cycle


Figure A-3. MoS2/Ti average friction coefficient per cycle as a function of sliding cycle.


9.0E-05
8.0E-05
7.0E-05
6.0E-05
5.0E-05
4.0E-05
3.0E-05
2.0E-05
1.0E-05
0.OE+00


2000


4000


6000


8000


10000


cycle


Figure A-4. MoS2/Ti total volume lost as a function of sliding cycle.


9*










I I I I I









LIST OF TABLES

Table page

5-1 Average temperatures and humidities for the duration of the experiments......... 27

6-1 Summary of results for the steady state wear testing in dry nitrogen ................. 31

6-2 Number of cycles to remove one layer of MoS2............... .... .............. 32









APPENDIX A
FRICTION AND WEAR PLOTS


0.16

0.14

0.12

0.1

0.08

0.06

0.04

0.02

0


0 2000 4000 6000 8000


cycle


Figure A-1. MoS2/Ni average friction coefficient per cycle as a function of sliding cycle.


7.0E-04

6.0E-04

5.0E-04

4.0E-04

3.0E-04

2.0E-04

1.0E-04

0.OE+00


2000


4 ,


- *---------------


4000


6000


8000


10000


cycle


Figure A-2. MoS2/Ni total volume lost as a function of sliding cycle.


10000










0.16

0.14

0.12

0.1

0.08

0.06

0.04


0.02


0 2000 4000 6000 8000 10000

cycle


Figure A-5. MoS2/Sb203 average friction coefficient per cycle as a function of sliding
cycle.

6.0E-05 T


5.0E-05

4.0E-05

3.0E-05

2.0E-05

1.0E-05

0.OE+00


Ve
,- .


2000


4000


6000


8000


10000


cycle


Figure A-6. MoS2/Sb203 total volume lost as a function of sliding cycle.









ACKNOWLEDGMENTS

I would like to thank my advisor for providing this opportunity and for his guidance

through this accomplishment. He exposes his students to many new facets of science,

and he provides TO all of his students a fantastic and unique environment from which

they will undoubtedly go on to great success. If it had not been for the time is spent in

his lab, I would not have gained the understanding and knowledge that I have acquired.

Additionally, I would like to thank all of my friends and lab mates; it has definitely been

an interesting ride! I extend a very special thank you to Matt for listening when I needed

to think out loud and providing great friendship.









CHAPTER 3
SAMPLE DESCRIPTIONS

3.1 Substrate and Pin Descriptions

Each experiment required a flat substrate and a ball with the same surface

coating. The flat substrates were 304 stainless steel rectangular coupons with

dimensions of 38 mm x 25 mm x 4.75 mm. They were mechanically polished to a

surface roughness below 50 nm Ra. The ball samples had a diameter of one-quarter

inch, and the material was 6061-T6 aluminum without any surface preparation or

polishing. The balls and counterfaces were distributed to commercial companies and to

the Air Force Research Laboratory (AFRL) to have MoS2 based coatings deposited onto

them. The commercial coatings were MoS2/Ni, MoS2/Sb203/Au, and MoS2/Ti. The

AFRL provided an MoS2/Sb203 coating and one of their "chameleon" coatings of

MoS2/Sb203/graphite.

3.2 Coating Descriptions

3.2.1 MoS2/Ni

The MoS2/Ni coating is commercially available and was made by direct current

(DC) magnetron sputtering. A thin layer of nickel was first deposited to provide an

adhesion layer between the steel or aluminum surface and the MoS2. The makeup of

the coating is 95% MoS2 and 5% nickel.

3.2.2 MoS2/Ti

The MoS2/Ti coating is commercially available and is multilayered with a 100 nm

titanium layer, then a 200 nm MoS2/Ti layer, and then a 50 nm pure MoS2 layer. The

buildup of the layers continues until the coating is approximately 1 pm thick. This

coating is made by unbalanced direct current magnetron sputtering (UDCMS) using one









hundredths of one degree. Each discrete pulse is considered to be a "step". By

discretizing the total output signal to very small steps, very accurate control is achieved.

To be certain that the path characteristics are consistent for every cycle an

encoder is used to examine the movement of the stage. Using the information from the

encoder, the position along the path is measured. Using the first and second

derivatives of the path data, the velocity and the acceleration of the path is also

monitored or used as an input. The force data is synchronously collected along with

position data to be used to compare friction and normal force measurements along the

track.

2.2 In Situ Tribometer Design

An in situ wear tribometer was designed to examine the evolution of the wear track

of a flat counterface, where the contact geometry is that of a stationary ball and a flat

reciprocating counterface (it was recognized that the ball would wear, but this was not

examined). The tribometer used to run the series of experiments is shown

schematically in Figure 2-3. It makes use of a linear stage, which provides a

reciprocating contact between material pairs. To move the counterface the Parker 401-

XR (Parker, Cleveland, OH) ball screw driven linear stage was used in conjunction with

a Parker HV172 stepper motor (Parker, Cleveland, OH). This system allows for a

maximum normal load of 480 N and a maximum transverse load of 150 N with speeds

up to 50 mm/s. The stage has a linear encoder mounted to its side which optically

counts vertical lines in a gold plated iron strip to determine position. Using an encoder

such as this versus one which mounts to the motor shaft alleviates any positional errors

that may be caused by dead zones when the ball screw reverses its direction. This

encoder gives the stage a positional repeatability of 5 pm.









titanium target and three MoS2 targets. During deposition the substrate rotated among

the target (12).

3.2.3 MoS2/Sb203

The MoS2/Sb203 coating provided by the AFRL and was deposited by pulsed laser

deposition (PLD) using a krypton-fluoride laser from a single composite target with a

composition of 70% MoS2 and 30% Sb203.

3.2.4 MoS2/Sb203/graphite

The AFRL "chameleon" coating was deposited using the same PLD technique as

the MoS2/Sb203 coating. It was made from a single composite target with the

composition being 50% MoS2, 30% Sb203, and 20% carbon. The carbon phase is

distinguished as graphite so as not to be confused with other possible phases.

3.2.5 MoS2/Sb203/Au

The specific details regarding the MoS2/Sb203/Au coating are not available. It is

manufactured using DC magnetron sputtering from one composite target as specified in

MIL-STD-3071660.









5.6 MoS2/Sb203/Au Composite Coating

The MoS2/Sb203/Au composite had an average friction coefficient of 0.024 (Figure

A-9). The maximum wear scar depth was 177 nm and the maximum track width was 69

pm (Figure B-5). This coating did not achieve steady-state wear until 5,000 cycles, and

the steady state wear rate was 3.03E-09 mm3/N m (Figure A-10). After the coating was

run in, it experienced almost zero wear. This coating had the highest initial percentage

of its total lost volume removed in the first 1,000 cycles with 57%. Only 1-3% was lost

during the steady-state regime. Cycles 8,000 and 10,000 resulted in an increase in

volume from the previous cycle (Figure A-10). This can be attributed to wear from the

ball surface.







0.16
0.14
0.12
0.1
0.08
0.06
0.04
0.02
0


2000


4000


6000


8000


10000


cycle


Figure A-9. MoS2/Sb203/Au average friction coefficient per cycle as a function of sliding
cycle.

2.5E-05

2.0E-05 --

S1.5E-05
E
o 1.0E-05
I-
5.0E-06

0.OE+00 ii


2000


4000


6000


8000


10000


cycle


Figure A-10. MoS2/Sb203/Au total volume lost as a function of sliding cycle.


~z~a~









(a) (b)
500-




c.
E -500- *

I -1000-

". t .'o undeformed
-1500 surface wear scar area
e deformed
surface
-2000- 1.
0 170 340 510 680 850 170 340 510 680 850
Position (pm) Position (pm)

Figure 4-2. (a) An example comparison between a scan of an original surface with a
scan of a surface after 6000 sliding cycles, (b) the wear scar area used in
determining the wear rate.



0 5 mm 0


surface scan location

-<--------

0 one cycle0


Figure 4-3. Surface scan location along the wear track.

4.3 Experiment Summary

The summary of the experiment is as follows: five different MoS2 based coatings

were deposited on to five 6061-T6 aluminum balls and five 304 stainless steel

substrates. The balls and substrates were reciprocated, self-mated, with a normal load

of 5 N at a sliding speed of 10 mm/s for 10,000 cycles (100 m) in a nitrogen gas

environment with a relative humidity of 1%. The wear profile was characterized using a

SWLI which lead to the characterization of the wear rate.









shims that are 0.50 mm thick. They act together as an elastic spring in the loading

direction, but due to their width they are extremely stiff in the direction of motion (50

MN/m), and they are able to resist the friction force without gross rotations.

The balls are drilled and tapped to have a #4-40 machine screw holding it in place,

and it is mounted to the flexure system. The axis of the tapped hole is held 600 from

vertical so that the ball may be rotated and reused for a total of up to six experiments

(Fig. 2-4). The counterface can also have up to six experiments of 5 mm in length.


(b)













0 0


O -
(c) 0 0
0 O

Figure 2-4. Schematic of samples, a) the in-situ tribometer, b) illustration of how the ball
is rotated to yield six different experiments, c) illustration of the six wear
tracks on the flat counter face.

A custom program written in LabVIEW (National Instruments, Austin, TX) was

used to record all the data. The National Instruments data acquisition card had a 216 bit

resolution with full scale voltages of 10 V.









LIST OF REFERENCES

1. J.H. Han, C.G. Kim, Composite Structures 72, 218 (2006).

2. M. E. Bell, J. H. Findlay, Physical Review 59, 33 (May 1-3, 1941).

3. C. Donnet, J. M. Martin, T. L. Mogne, M. Belin, Tribology International 29, 123
(1996).

4. M. B. Peterson, R. L. Johnson, NACA TN, (1953).

5. A. J. Haltner, C. S. Oliver, Industrial & Engineering Chemistry Fundamentals 5,
348 (1966).

6. W. 0. Winer, Wear 10, 422 (1967).

7. M. S. MIL-M-7866B. (1965).

8. C. C. Baker, J. J. Hu, A. A. Voevodin, Surface & Coatings Technology 201, 4224
(Dec 20, 2006).

9. D. G. Teer, Wear 250, 1068 (Oct, 2001).

10. A. A. Voevodin, J. S. Zabinski, Wear 261, 1285 (Dec 20, 2006).

11. J. S. Zabinski, M. S. Donley, N. T. Mcdevitt, Wear 165, 103 (May 1, 1993).

12. N. M. Renevier, N. Lobiondo, V. C. Fox, D. G. Teer, J. Hampshire, Surface &
Coatings Technology 123, 84 (Jan 10, 2000).

13. P. L. Dickrell etal., Tribology Letters 18, 59 (Jan, 2005).

14. J. Williams, R. T. Hunt, Proceedings of the Royal Society of London Series A-
Mathematical and Physical Sciences 327, 147 (1972).

15. R. S. Sayles, Tribology International 34, 299 (May, 2001).

16. T. L. Schmitz, J. E. Action, D. L. Burris, J. C. Ziegert, W. G. Sawyer, Journal of
Tribology-Transactions of the ASME 126, 802 (Oct, 2004).

17. R. Holinski, Gansheim.J, Wear 19, 329 (1972).










0.16

0.14

0.12

0.1

0.08

0.06

0.04

0.02

0


0 -


0 2000 4000 6000 8000


cycle


Figure A-7. MoS2/Sb203/graphite average friction coefficient
sliding cycle.


2.5E-05


2.0E-05


1.5E-05


1.0E-05


5.0E-06


0.OE+00


2000


4000


6000


per cycle as a function of


.___''..'
*
*
' *
*


8000


10000


cycle


Figure A-8. MoS2/Sb203/graphite total volume lost as a function of sliding cycle.


fp


10000









1.2 Background

One of the most popular solid lubricants for commercial use today is molybdenum

disulphide. Molybdenum disulphide (MoS2), as "molybdenite", has been known as a low

friction solid lubricant for vacuum applications as early as 1941 when the Westinghouse

Lamp Division needed a lubricant with a low vapor pressure in designing a bearing for a

rotating x-ray tube. However, in that report it was also commented that, "The use of

molybdenite as a lubricant is not necessarily restricted to a vacuum" (2). While it is

generally recognized that molybdenum disulphide demonstrates higher friction in air or

lower vacuum levels (3), it wasn't until 1953 that it was discovered that it wasn't

necessarily the vacuum environment that allowed low friction; rather the absence of

water vapor allowed low friction. In that study Peterson and Johnson showed that as

the relative humidity increased up to from 6% to 65%, both the friction and the wear

increased between contacts lubricated with MoS2 (4). This trend was confirmed by

Haltner and Oliver as they expanded this research to go down to 0.5% RH (5). After the

Westinghouse study, there was such an increase in the interest of MoS2 that by 1966

there were over 100 pertinent studies on its friction and wear properties (6). Even the

military became interested, and a military specification was created for MoS2. It was

stated that, "The molybdenum disulfide is intended to function as a lubricant contained

in grease and solid films by reducing friction and wear under low and high sliding

velocities where boundary lubricants exist" (7). Haltner and Oliver's study also proved

to be important by making the connection that a dry nitrogen environment could

simulate a vacuum environment when examining the frictional response of MoS2, which

today is very popular when vacuum testing is either not possible or impractical. Already

by the late 1960s it had been well established that MoS2 was a good solid lubricant in









5 EX P E R IM E N TA L R E S U LT S .......................................................... ... .. ............... 27

5.1 Average Temperature and Humidity ............... ........... ...... .. ...... ....... 27
5.2 MoS2/Ni Composite Coating .................. ........... ........... ... ...... ........ 27
5.3 M oS 2/T i C om posite C oating ..................................................... ... ................. 27
5.4 M oS2/Sb203 Com posite Coating .............. ................. ................................. 28
5.5 MoS2/Sb203/graphite Composite Coating ............ ...................................... 28
5.6 MoS2/Sb203/Au Composite Coating................... .. ................ 29

6 DISCUSSION AND ANALYSIS OF RESULTS ................................................. 30

6.1 Single Point, Complete Data, and Steady-State Wear Rates ........................ 30
6.2 Summary of Friction and W ear Results ............ ......................... ....... 30

7 CO NCLUSION ............... ................................ ............. ......... 33

APPENDIX

A FRICTION AND WEAR PLOTS............. .. ................... ............... 34

B SURFACE PRO FILES ...................... ..................... ... ............................. 39

LIST OF REFERENCES ................................. .................... 42

B IO G RA PH IC A L S KETC H ...................... .. ............. .. ......................... ............... 43


























6









both vacuum and ambient environments. There seems to be a lull in the research of

MoS2 in the 1970s and then somewhat of an increase in the 1980s. But the 1990s

brought about a large resurgence in the interest of MoS2, and by the 2000s even more

studies were being performed. Mostly, the research has mainly focused on combining

MoS2 in with other species to create composite coatings, and how to make coatings

these coatings perform better (8-11). The topic of friction for these composite coatings

dominates the literature, and what seems to be lacking are good, quantitative results of

wear rates. Most studies have discussed wear in terms of the number of cycles

experienced with low steady state friction, or in terms the distance traveled. Even in the

instances where the total wear volume is measured, a single-point wear rate

measurement is usually performed, but it is often a poor representation of the steady

state wear rate, since the single-point wear rate captures the wear-in phase of the

material. The intention of this study is to provide quantitative wear rates for some very

common self-mated MoS2 composite coatings in a low humidity, nitrogen environment.

1.3 Experiment Overview

To determine these wear rates a linear reciprocating tribometer was enclosed

inside of an environment chamber with a scanning white light interferometer (SWLI)

located above the wear track. The chamber was flooded with dry nitrogen gas to

reduce the relative humidity (RH) to 1%, and tribology experiments were performed

while interferometric measurements of the wear surface were being quantified.









kcD wear with the ksp having the highest value. Figure 6-1 illustrates this schematically.

Table 6-1. Summary of results for the steady state wear testing in dry nitrogen.
Wear Rates, k (mm3/N-m) kss
Composite p .
ksp kcD kss uncertainty
MoS2/Ni 0.100 1.23E-06 5.46E-07 3.70E-07 1.10E-08
MoS2/Ti 0.051 1.59E-07 1.60E-07 1.60E-07 1.00E-07
MoS2/Sb203 0.062 9.98E-08 5.05E-08 5.05E-08 1.90E-08
4.65E-08
MoS2/Sb203/C 0.010 4.67E-08 2.29E-08 2.82E-08 4.40E-09
2.82E-08
MoS2/Sb203/Au 0.024 4.03E-08 1.85E-08 3.03E-09 5.50E-09



E -
E o o1 o
wear-in
P> k, regime

F,-d (N-m) F -d (N-m) F,-d (N-m)

Figure 6-1. Illustration of different wear rate analyses.

All of the composites that contained an Sb203 phase had nearly an order of

magnitude lower wear rate than the composites that only had metal inclusions. It

should come as no surprise that the composite with the highest wear rate was MoS2/Ni

since it was 95% MoS2, and it is known to not be a low wear material. A result that is of

interest is that the coating with the lowest wear, MoS2/Sb203/Au, did not have the lowest

friction coefficient. Table 6-1 shows that low friction is not necessarily indicative of low

wear. The MoS2/Sb203/Au coating and the MoS2/Sb203/graphite coating had very

nearly the same wear rate, with MoS2/Sb203/Au being the lowest, but the

MoS2/Sb203/graphite coating had friction that was over two times lower than the

MoS2/Sb203/Au coating.









CHAPTER 1
INTRODUCTION

1.1 Motivation of Research

Low earth Orbit (LEO) is defined as the altitude of 160 km to 2000 km above the

earth's surface. This is an extreme environment which is difficult for mechanisms to

operate. The pressure is 10 5 Torr (high vacuum) at an altitude of 200 km and even

lower as the altitude increases (1). At this altitude, the ozone layer is not present to

absorb ultraviolet radiation (UV) which has enough energy to break chemical bonds

which leads to decomposition of the materials, often creating free radicals. Atomic

oxygen is also present which binds with the free radicals to create new, undesirable

often degraded compounds. This list of hazards in not complete, and unfortunately this

is the environment in which the United States' Space Shuttle and many satellites

perform. All of the moving mechanical assemblies of these machines have m have

moving surfaces in contact. This research was motivated by the fact that design

engineers on earth need to reliably understand how geometries of tribological interfaces

will change as they perform. The driving force behind changing interfaces is wear.

Wear is simply material removal from the surface, and if enough material is removed

from two surfaces in contact, precision designed assemblies may end up not being so

precise after all. The variable of outer space complicates the issue further by being

inaccessible to frequent repairs like an automobile on earth with multiple service

stations per city. Very few mechanics live on the moon, and up to this point aliens have

been unreliable. So, design engineers need to know when a surface will wear, how

much it will wear, and at what rate it will continue to wear with further use.









BIOGRAPHICAL SKETCH

Jason Steffens was born in Winter Park, FL in 1979. When he was eleven years

old he moved to Eustis, FL where he was exposed to radiation that prevented his head

from developing to a mature adult's size and subsequently lengthened his arms. He is

making lemonade by attempting to use these characteristics to his advantage by

pursuing a black belt in Brazilian Jiu-Jitsu. His small head is instrumental in escaping

triangle chokes and his long arms are ideal for securing anaconda and D'Arce chokes.









tribometer was located on an interferometer with a reported feature height resolution of

1 A to investigate the evolution of the wear scar on a cycle-by-cycle basis.

PAGE 1

1 WEAR OF SELF MATED MOLYBDENUM DISULPHIDE TRIBOLOGICAL INTERFACES By JASON GEOFFREY STEFFENS A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2010

PAGE 2

2 2010 Jason Geoffrey Steffens

PAGE 3

3 For Nick Kate

PAGE 4

4 ACKNOWLEDGMENTS I would l ike to thank my advisor for providing this opportunity and for his guidance through this accomplishment. He exposes his students to many new facets of science, and he provides TO all of his students a fantastic and unique environment from which they will undoubtedly go on to great success. If it had not been for the tim e is spent in his lab I would not have gained the understanding and kn owledge that I have acquired. Additionally, I would like to thank all of my friends and lab mate s; it has definitely been an interesting ride I extend a very special thank you to Matt for listening when I needed to think out loud and providing great friendship.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ ............ 7 LIST OF FIGURES ................................ ................................ ................................ .......... 8 LIST OF ABBREVIATIONS ................................ ................................ ............................. 9 ABSTRACT ................................ ................................ ................................ ................... 10 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .... 12 1.1 Motivation of Research ................................ ................................ ..................... 12 1.2 Background ................................ ................................ ................................ ....... 13 1.3 Experiment Ove rview ................................ ................................ ........................ 14 2 EQUIPMENT CHARACTERISTICS ................................ ................................ ........ 15 2.1 General Tribometer Design ................................ ................................ ............... 15 2.1.1 Force Characterization ................................ ................................ ............ 15 2.1.2 Path and Path Characteristics ................................ ................................ 16 2.2 In Situ Tribometer Design ................................ ................................ ................. 17 2.3 Interferometer ................................ ................................ ................................ ... 20 2.4 Environment ................................ ................................ ................................ ...... 20 2.5 Summary of E quipment and Tribometer Design ................................ ............... 20 3 SAMPLE DESCRIPTIONS ................................ ................................ ..................... 22 3.1 Substrate and Pin Descriptions ................................ ................................ ......... 22 3.2 Coating Descriptions ................................ ................................ ......................... 22 3.2.1 MoS 2 /Ni ................................ ................................ ................................ ... 22 3.2.2 MoS 2 /Ti ................................ ................................ ................................ .... 22 3.2.3 MoS 2 /Sb 2 O 3 ................................ ................................ ............................. 23 3.2.4 MoS 2 /Sb 2 O 3 /graphite ................................ ................................ ............... 23 3.2.5 MoS 2 /Sb 2 O 3 /Au ................................ ................................ ....................... 23 4 METHODS ................................ ................................ ................................ .............. 24 4.1 Test Method ................................ ................................ ................................ ...... 24 4.2 Determination of Wear Rates ................................ ................................ ............ 25 4.3 Experiment Summary ................................ ................................ ....................... 26

PAGE 6

6 5 EXPERIMENTAL RESULTS ................................ ................................ ................... 27 5.1 Average Temperature and Humidity ................................ ................................ 27 5.2 MoS 2 /Ni Composite Coating ................................ ................................ ............. 27 5.3 MoS 2 /Ti Composite Coating ................................ ................................ .............. 27 5.4 MoS 2 /Sb 2 O 3 Composite Coating ................................ ................................ ....... 28 5.5 MoS 2 /Sb 2 O 3 /graphite Composite Coating ................................ ......................... 28 5.6 MoS 2 /Sb 2 O 3 /Au Composite Coating ................................ ................................ .. 29 6 DISCUSSION AND ANALYSIS OF RESULTS ................................ ....................... 30 6.1 Single Point, Complete Data, and Steady State Wear Rates ........................... 30 6.2 Summary of Friction and Wear Results ................................ ............................ 30 7 CONCLUSION ................................ ................................ ................................ ........ 33 APPENDIX A FRICTION AND WEAR PLOTS ................................ ................................ .............. 34 B SURFACE PROFILES ................................ ................................ ............................ 39 LIST OF REFERENCES ................................ ................................ ............................... 42 BIOGRAPHICAL S KETCH ................................ ................................ ............................ 43

PAGE 7

7 LIST OF TABLES Table page 5 1 Average temperatures and humidities for the duration of the experiments. ........ 27 6 1 Summary of results for the steady state wear testing in dry nitrogen ................. 31 6 2 Number of cycles to remove one layer of MoS 2 ................................ .................. 32

PAGE 8

8 LIST OF FIGURES Figure page 2 1 P ath required for the force to be reacted from the point of contact to ground ..... 16 2 2 Standard POD tribometer ................................ ................................ ................... 16 2 3 Tribometer and interferometer schematic ................................ ........................... 18 2 4. Schematic of test equipment ................................ ................................ ................ 19 4 1 Illustration of the reciprocation path ................................ ................................ .... 24 4 2 Surface scan location along the wear track ................................ ........................ 26 4 2 C omparison between worn and virgin surfaces ................................ .................. 26 6 1 Illustration of different wear rate analyses ................................ .......................... 31 A 1 MoS 2 /Ni average friction coefficient per cycle ................................ ..................... 34 A 2 MoS 2 /Ni total volume lost as a function of sliding cycle ................................ ...... 34 A 3 MoS 2 /Ti a verage friction coefficient per cycle ................................ ..................... 35 A 4 MoS 2 /Ti total volume lost as a function of sliding cycle. ................................ ..... 35 A 5 MoS 2 /Sb 2 O 3 average friction coefficient per cycle ................................ .............. 36 A 7 MoS 2 /Sb 2 O 3 /graphite average friction coefficient per cycle ................................ 37 A 8 MoS 2 /Sb 2 O 3 /graphite total volume lost as a function of sliding cycle .................. 37 A 9 MoS 2 /Sb 2 O 3 /Au average friction coefficient per cycle ................................ ......... 38 A 10 MoS 2 /Sb 2 O 3 /Au total volume lost as a function of sliding cycle .......................... 38 B 1 Surface profiles of the M oS 2 /Ni composite coating ................................ ............. 39 B 2 Surface profiles of the MoS 2 /Ti composite coating. ................................ ............ 39 B 3 Surface profiles of the MoS 2 /Sb 2 O 3 composite coating ................................ ...... 40 B 4 Surface profiles of the MoS 2 /Sb 2 O 3 /graphite composite coating ........................ 40 B 5 Surface profiles of the MoS 2 /Sb 2 O 3 /Au composite coating ................................ 41

PAGE 9

9 LIST OF ABBREVIATION S Angstrom AFRL Air Force Research Laboratory Au gold DC direct current k CD complete data wear rate k SP sing le point wear rate k SS steady state wear rate LEO low earth orbit MoS 2 m olybdenum disulphide Ni nickel PEEK polyetheretherketone PLD pulsed laser deposition POD pin on disk Ra average surface roughness RH relative humidity Sb 2 O 3 a ntimony trioxide SWLI scanning white light interferometer Ti t itanium UDCMS u nbalanced direct current magnetron sputtering UV ultraviolet

PAGE 10

10 Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirement s for the Degree of Master of Science WEAR OF SELF MATED MOLYBDENUM DISULPHIDE TRIBOLOGICAL INTERFACES By Jason Geoffrey Steffens August 20 10 Chair: W. Gregory Sawyer Major: Mechanical Engineering Engineers designing mechanisms intending to go into orbit around the earth need to know reliably how geometries of tribological interfaces change as they perform. The driving force behind changing interfaces is wear. As interfaces change, precision desig ned assemblies may end up not being so precise after all. So, design engineers need to know when a surface will wear, how much it will wear, and at what rate it will wear. Molybdenum disulphide (MoS 2 ) has been known as a low friction solid lubricant for vacuum applications as early as 1941. Friction studies on MoS 2 are extensive throughout the literature, and they continue today. The main focus appe ars to be on the friction response of MoS 2 and these studies dominate over the number of studies where wear is the predominant characteristic When wear is quantified it is often discussed in terms of the number of cycles experienced with low steady stat e friction, or in terms of the distance traveled instead of providing a useful tribological wear rate Even in the instances where the total wear volume is measured, a single point wear rate measurement is performed but it is often a poor representation of how the specimen actually performs through different regimes The intention of this study was to provide

PAGE 11

11 quantitative wear rates for five self mated MoS 2 composite coatings in a low humidity environment Used in these experiments is a custom built and instrumented reciprocating tribometer with in situ scanning white light interferometry capability which was used to make surface topographic measurements on a cycle by cycle basis.

PAGE 12

12 CHAPTER 1 INTRODUCTION 1.1 Motivat ion of Research Low earth Orbit (LEO) is defined as the altitude of 160 km to 2000 km above the difficult for mechanisms to operate. The pressure is 10 5 Torr (high vacuum) at an altitude of 200 km and even lower as the altitude increase s ( 1 ) At this altitude, the ozone layer is not present to absorb ultra violet radiation (UV) which has enough energy to break chemical bonds which leads to decomposition of the materials, often creating free radicals Atomic oxygen is also present whic h binds with the free radicals to create new, undesirable often degraded compounds. This list of hazards in not complete, and u nfortunately this is the environment in which the Space Shuttle and many satellites perform All of the moving mechanical assemblies of these machines have m have moving surfaces in contact. This research was motivated by the fact that design engineers on earth need to reliably understand how geometries of tribological interfaces will change as they perform The driving force behind changing interfaces is wear. Wear is simply material removal from the surface, and if enough material is removed from two surfaces in contact, precision designed assemblies may end up not being so precise after all. The variable of o uter space complicates the issue further by being inaccessible to frequent repairs like an automobile on earth with multiple service stations per city. Very few mechanics live on the moon, and up to this point aliens have been unreliable. So design engi neers need to know when a surface will wear, how much it will wear, and at what rate it will continue to wear with further use

PAGE 13

13 1.2 Background One of the most popular solid lubricants for commercial use today is molybdenum disulphide. Molybdenum disulphid e (MoS 2 has been known as a low friction solid lubricant for vacuum applications as early as 1941 when the Westinghouse Lamp Division needed a lubricant with a low vapor pressure in designing a bearing for a rotating x ray tube. Howeve r, in that report it was also commented that, The use of ( 2 ) While it is generally recognized that molybdenum disulphide demonstrates higher friction in air or lower vacuum levels ( 3 ) it necessarily the vacuum environment that allowed low friction; rather the absence of water vapor allowed low friction. In that study Peterson and Johnson showed that as the relative humidity increased up to from 6% to 65%, both the friction and the wear increased between contacts lubricated with MoS 2 ( 4 ) This trend was confirmed by Haltner and Oliver as they expanded this research to go down to 0.5% RH ( 5 ) After the Westinghouse study, there was such an increase in the interest of MoS 2 that by 1966 there were over 100 pertinent studies on its friction and wear properties ( 6 ) Even the military became interested, and a military specification was created for MoS 2 It was stated that in grease and solid films by reducing friction and wear under low and high sliding ( 7 ) Haltner and to be important by making the connection that a dry nitrogen environment could simulate a vacuum environment when examining the frictional response of MoS 2 which today is very popular when vacuum testing is either not possible or impractical. Already by the late 1960s it had been well established that MoS 2 was a good solid lubricant in

PAGE 14

14 both vacuum and ambient environments. There seems to be a lull in the research of MoS 2 in the 1970s and then somewhat of an increase in the 198 0s. But the 1990s brought about a large resurgence in the interest of MoS 2 and by the 2000s even more studies were being performed. Mostly the research has mainly focused on combining MoS 2 in with other species to create composite coatings and how to m ake coatings these coatings perform better ( 8 11 ) The topic of f riction for these composite coatings dominate s the literature, and what seems to be lacking are good, quantitative results of wear rates. Most studies have discussed wear in terms of the number of cycles experienced with low steady state friction, or in terms the distance traveled. Even in the instances where the total we ar volume is measured, a single point wear rate measurement is usually performed but it is often a poor representation of the steady state wear rate since the single point wear rate captures the wear in phase of the material The inten tion of this study is to provide quantitative wear rates for some very common self mated MoS 2 composite coatings in a low humidity, nitrogen environment 1.3 Experiment Overview To determine these wear rates a linear reciprocating tribometer was enclosed in side of an environment chamber with a scanning white light interferometer (SWLI) located above the wear track The chamber was flooded with dry nitrogen gas to reduce the relative humidity (RH) to 1%, and tribology experiments were performed while interferometric measurements of the wear surface were being quantified.

PAGE 15

15 CHAPTER 2 EQUIPMENT CHARACTERISTICS 2. 1 General Tribometer Design When investigating tribology, there are two standard tr ibometer designs; a pin on disk (POD) where the disk rotates with an eccentrically located pin /ball and a pin on flat where the flat is translated using a reciprocating stag e. General tribometer design philosophy follows a need to characterize the normal and fri ction forces the path of travel for each cycle, and the path characteristics. 2. 1 .1 Force Characterization T o accurately record force measurements a force transducer must be placed inside the path of contact to ground. In addition, there should be no other movi ng a ssemblies, such as bearings, bushings, or gimbals in this path to introduce parasitic forces. Figure 2 1 illustrates that one of the contacting bodies (in this case, the ball) is oving systems while the tribometer is performing an experiment. Moving systems between the load cell and the contacting body would introduce parasitic forces measured by the load cell which leads to inaccurate data. Additionally, moving systems between t he load cell and ground would introduce inertial forces, which would also be measured by the load cell leading to inaccurate data To contrast good tribometer design, a standard POD tribometer is illustrated in Figure 2 2. Following the logic presented fo r following the path from ground in Figure 2 1, it can be determined that the tribometer in Figure 2 2 has a moving mechanical assembly in the path from contact to ground, in this case a gimbal. A tribometer such

PAGE 16

16 as the one present in Figure 2 2 has a hig her uncertainty in the friction force measurement. Figure 2 1. Illustration of the path required for the force to be reacted from the point of contact to ground. Figure 2 2. Standard POD tribometer illustrating a moving mechanical assembly in the path of contact to ground 2. 1 .2 Path and Path Characteristics The use of a stepper moto r to control the motion is instrumental to reproduce the same path for each consecutive cycle A stepper motor breaks up the tota l signal of the output motion in to small discrete pulses that turn a central iron gear tenths to one

PAGE 17

17 hundredths of one degree By discretizing the total output signal to very small steps, very accurate co ntrol is achieved. To be certain that the path characteristics are consistent for every cycle an encoder is used to examine the movement of the stage. Using the information from the encoder, the position along the path is measured. Using the first and second derivatives of the path data, the velocity and the acceleration of the path is also monitored or used as an input. T he force data is synchronously collected along with position data to be used to compare friction and normal force measurements along the track. 2. 2 In Situ Tribometer Design An in situ wear tribometer was designed to examine the evolution of the wear track of a flat counterface, where the contact geometry is that of a stationary ball and a flat reciprocating counterface ( it was recognized that the ball would wear, but this was not examined). The tribometer used to run the series of experiments is shown schematically in Figure 2 3 It makes use of a linear stage, which provides a reciprocating contact between material pairs. T o move the counterface the Parker 401 XR (Parker, Cleveland, OH) ball screw driven linear stage was used in conjunction with a Parker HV172 stepper motor (Parker, Cleveland, OH) This system allows for a maximum normal load of 480 N and a maximum transver se load of 150 N with speeds up to 50 mm/s. The stage has a linear encoder mounted to its side which optically counts vertical lines in a gold plated iron strip to determine position. Using an encoder such as this versus one which mounts to the motor sha ft alleviates any positional errors that may be caused by dead zones when the ball screw reverses its direction. This encoder gives the stage a positional repeatability of 5 m.

PAGE 18

18 Figure 2 3 Tribometer and interferometer schematic, a) tribometer schema tic, b) illustration of loading mechanism, c) tribometer sitting on interferometer stage. Due to the geometry of the loading, high contact pressures are achieved with very low loads, where a load of just 5 N would yield a Hertzian contact stress of 1 GPa f or a self mated steel contact. Because a low loads induce contact pressures of this order, a load cell with the ability to accurately measure such loads is instrumental To measure these loads a JR3 50M31A (JR3, Woodland, CA) was chosen as the load cell. It is a six axis load cell that has an axial load capacity of 220 N, while in the other two directions the maximum load is 110 N. The load cell is oriented in such a manner that the two most sensitive axes are used to measure the normal load and the fri ction force This load cell has a very high resolution of 28 mN in the axial direction and 14 mN in the axes used in making the force measurements. The loading of this tribometer makes use of a simple leaf type flexure. The flexures are connected to a li near micrometer stage with a maximum capacity of 10 N that translates vertically to bring the ball in and out of contact. The flexures are steel

PAGE 19

19 shims that are 0.50 mm thick. They act together as an elastic spring in the loading direction, but due to the ir width they are extremely stiff in the direction of motion (50 MN/m), and they are able to resist the friction force without gross rotations. The balls are drilled and ta p ped to have a #4 40 machine screw holding it in place, and it is mounted to the fle xure system. The axis of the tapped hole is held 60 from vertical so that the ball may be rotated and reused for a total of up to six experiments (Fig. 2 4 ). The counterface can also have up to six experiments of 5 mm in length. Figure 2 4 Schematic of samples, a) the in situ tribometer, b) illustration of how the ball is rotated to yield six different experiments, c) illustration of the six wear tracks on the flat counter face. A custom program written in LabVIEW (National Instruments, Austin, TX) was used to record all the data. The National Instruments data acquisition card had a 2 16 bit resolution with full scale voltages of 10 V.

PAGE 20

20 2. 3 Interferometer Many tribological coatings exhibit very low wear rates where thousands of cycles on average are needed to achieve the removal of just one nanometer of the film. To accurately capture the effect of the material removal, an interferometer with a very high feature height resolution is required. The Zygo New View 5010 (Zygo, Middlefield, CT) was chosen because it has a reported feature height resolution of 1 which would be sufficient to capture the desired any topographical data ; however, practical experience suggests that uncertainties are on the order of nanometers The entire tribometer mounts to the stages of the interferometer, where its motorized stages can manipulate the sample orientation in situ 2. 4 Environment The tribometer and SWLI were located in a temperature controlled room used for metrology studies where the temperature of the room was kept at 20 2 C. The entire tribometer was fit inside of a small, acrylic chamber with feedthroughs for electrical components, as well as, a feedthrough for the introduction of different gas species. M ounted to the acrylic chamber was the DY5 moistu re probe with the MMY2650 HygroGuard 2650 system (GE Sensing, Billerica, MA) which measures temperature and relative humidity 2.5 Summary of Equipment and Tribometer Design A tribometer was designed with a multiaxis force sensor placed in the path of contact to ground. The multiaxis force sensor measures the normal and friction forces simultaneously. An acrylic environment chamber was constructed which houses the entire tribometer. The experiments can be performed in any non corrosive gaseous environ ment where the temperature and relative humidity can be monitored The

PAGE 21

21 tribometer was located on a n interferometer with a reported feature height resolution of 1 to investigate the evolution of the wear scar on a cycle by cycle basis.

PAGE 22

22 CHAPTER 3 S AMPLE DESCRIPTIONS 3.1 Substrate and Pin Descriptions Each experiment required a flat substrate and a ball with the same surface coating. The flat substrates were 304 stainless steel rectangular coupons with dimensions of 38 mm x 25 mm x 4.75 mm. They we re mechanically polished to a surface roughness below 50 nm Ra The ball samples had a diameter of one quarter inch, and the material was 6061 T6 aluminum without any surface preparation or polishing The balls and counterfaces were distributed to commer cial companies and to the Air Force Research Laboratory (AFRL) to have MoS 2 based coatings deposited onto them. The commercial coatings were MoS 2 /Ni, MoS 2 /Sb 2 O 3 / Au and MoS 2 /Ti. The AFRL provided an MoS 2 / Sb 2 O 3 coating and one MoS 2 /Sb 2 O 3 /graphite. 3.2 Coating Descriptions 3.2.1 MoS 2 /Ni The MoS 2 / Ni coating is commercially available and was made by direct current ( DC ) magnetron sputtering. A thin layer of nickel was first deposited to provide an adhesion layer between the steel or aluminum surface and the MoS 2 The makeup of the coating is 95% MoS 2 and 5% nickel. 3.2.2 MoS 2 /Ti The MoS 2 /Ti coating is commercially available and is multilayered with a 100 nm titanium layer, then a 200 nm MoS 2 /Ti l ayer, and then a 50 nm pure MoS 2 layer. The buildup of the layers continues until the coating is approximately 1 m thick. This coating is made by unbalanced direct current magnetron sputtering (UDCMS) using one

PAGE 23

23 titanium target and three MoS 2 target s During deposition the substrate rotated among the target ( 12 ) 3.2.3 MoS 2 /Sb 2 O 3 The MoS 2 /Sb 2 O 3 coating provided by the AFRL and was deposited by pulsed laser deposition ( PLD ) using a kr y pton fluoride laser from a single composite target with a composition of 70% MoS 2 and 30% Sb 2 O 3 3.2.4 MoS 2 /Sb 2 O 3 /graphite deposited using the same PLD technique as the MoS 2 /Sb 2 O 3 coating It was made from a single composite target with the composition being 50% MoS 2 30% Sb 2 O 3 and 20% carbon. The carbon phase is distinguished as graphite so as not to be confused with other possible phases. 3.2.5 MoS 2 /Sb 2 O 3 /Au The specific details regarding th e MoS 2 /Sb 2 O 3 /Au coating are not availa b le I t is manufactured using DC magnetron sputter ing from one composite target as specified in MIL STD 3071660.

PAGE 24

24 CHAPTER 4 METHOD S 4.1 Test Method E ach counter face described in Chapter 3 can fit six 5 m m long wear tracks and each ball can fit multiple contact points depending on the contact area ( dependent on contact stress ) The total distance trav eled in one cycle is 10 mm. Relati ve to the counterface, the ball began at one end of the track and traveled 5 mm to the other end, where it stopped and then returned to the original starting point to begin a new cycle (Figure 4 1). Figure 4 1. Illustration of the reciprocation path Before the experiment was performed, a calibration step was required to align the optics of the interfero meter to the desired image location A short experiment of a few cycles using an uncoated aluminum ball a nd an uncoated steel substrate was performed to create a wear scar. Then the optics of the interfero meter were aligned to the scar. To perform the e xperiment a coated stainles s steel substrate was bolted to t he reciprocating stage The n coated aluminum ball was fastened into a PEE K holder, and the holder was bolted to the load cell. The environment chamber was placed over the

PAGE 25

25 entire tribometer with the interferometer objective aligned with a viewing hole in line with the wear track A seal was made between the ob jective and the chamber using an unlubricated Magnum condom (Church and Dwight, Co., Princeton, NJ). Dry nitrogen gas was bled into the chamb er until the relative humidity was less than 1% T speed, acceleration, and deceleration were set to 10 mm/s, 100 mm/s 2 and 100 mm/s 2 respectively. An initial scan of the surface was made representing cycle 0, so that the s ubsequent profiles could be compared to the native surface A 5 N normal load was applied, and the test began Every 1000 cycles (equivalent of 10 m) the test was stopped, and a surface measurement was performed on the f lat counterface. This continued un til 10,000 cycles was reached. The data for the friction coefficients were collected at 500 Hz and were computed by the technique described by Dickrell et al ( 13 ) where the friction coefficient reported is the half the difference between the forward and reverse direc tions of a cycle. 4. 2 Determination of Wear Rates To determine the volume of material lost the scan of cycle zero was compared to the scan of interest (i.e. cycle 1000 was compared to cycle 0 and cycle 2000 was also compared to cycle 0) as shown in Figure 4 2 The scan was performed at middle of the wear scar (Figure 4 3), not the middle of t he reciprocating path. The wear scar area was calculated as the difference between the two surfaces, and was then extruded to the total length of the reciprocating path 10 mm. This technique was introduced by Williamson and Hunt in examining the asperit y persistence even after plastic deformation occurred ( 14 ) Sayles then expanded upon this technique when he applied it to surfaces ( 15 ) To determine the wear rate uncertainties, a Monte Carlo simulation was performed as previously described by Schmitz et a l ( 16 )

PAGE 26

26 Figure 4 2 (a) An exam ple comparison between a scan of an original surface with a scan of a surface after 6000 sliding cycles, (b) the wear scar area used in determining the wear rate. Figure 4 3. Surface scan location along the wear track. 4.3 Experiment Summary The summary of the experiment is as follows: five different MoS 2 based coatings were deposited on to five 6061 T6 aluminum balls and five 304 stainless steel substrates. The balls and substrates were reciprocated, self mated, with a normal load of 5 N at a s liding speed of 10 mm/s for 10,000 cycles (100 m) in a nitrogen gas environment with a relative humidity of 1%. The wear profile was characterized using a SWLI which lead to the characterization of the wear rate.

PAGE 27

27 CHAPTER 5 EXPERIMENTAL RESULTS 5.1 Average Temperature and Humidity While all of the experiments were performed in a temperature controlled room, t he tem perature of the experiments was not control led, but instead was monitored. Table 5 1 provides a summary of the temperature and humidity of the e xperiments. Table 5 1. Average temperatures and humidities for the duration of the experiments. 5 2 MoS 2 / Ni Composite Coating The MoS 2 / Ni composite had the highest averag e friction coefficient of 0.100 among all of the samples (Figure A 1). The maximum wear scar depth was 906 nm, and the maximum track width was 206 m (Figure B 1) Its steady s tate wear rate was 3.70E 07 mm 3 /Nm steady state wear was considered be at 3,000 cycles Nearly half of the total wear occurred within the first 1 000 cycles of sliding where 49% of the total lost material was removed (Figure A 2). 5 3 MoS 2 / Ti Composite Coating The MoS 2 / Ti composite had an average friction coefficient of 0.051 (Figure A 3). The maximum wear scar depth was 246 nm, and the maximum track width was 155 m (Figure B 2) Its steady state wear rate was 1.71 E 07 mm 3 / Nm Steady state wear was achieved quickly and was analyzed beginning at 2,000 cycles. This coating very had Composite Average Temp erature (C) Average %RH MoS 2 /Ni 29.8 0.10 MoS 2 /Ti 29.5 0.14 MoS 2 /Sb 2 O 3 27.3 0.10 MoS 2 /Sb 2 O 3 /C 29.3 0.10 MoS 2 /Sb 2 O 3 /Au 29.6 0.11

PAGE 28

28 very uniform wear, where every 1 000 cycles 5 13% of the total lost material was removed except for the last 1 000 cycles where 20% of the total lost material was removed (Figure A 4) 5. 4 MoS 2 /Sb 2 O 3 Composite Coating The MoS 2 /Sb 2 O 3 composite had an average friction coefficient of 0.062 (Figure A 5). The maximum wear scar depth was 468 nm and the maximum track width was 123 m (Figure B 3). Its steady state wear rate was 5.05E 08 mm 3 /Nm. This coating achieved steady state almost immediately and all cycles were used in determining its steady state wear rate. Like the MoS 2 /Ni coating, nearly half of the total wear occurred within the first 1 000 cycles with 47% being removed (Figure A 6) At cycle 8 000 there was an increase in the volume of material in the wear track likely from the coating having been worn from the ball surface and being deposited into the wear track. 5 5 MoS 2 /Sb 2 O 3 /graphite C omposite Coating The MoS 2 /Sb 2 O 3 /graphite composite had the lowest average friction coefficient of 0.010 (Figure A 7) The maximum wear scar depth was 423 nm and the maximum track width was 35 m (Figure B 4 ) This coating had two different wear regimes which had similar wear rate s For cycles 0 5 000 steady state wear rate was 4.65E 08 mm 3 /Nm and for cycles 8,000 10,000 the steady state wear rate was 2.82E 08 mm 3 /Nm Between cycles 7 000 and 8 000 there w as a gain in volume in the wear track likely from the ball coating being worn an d depositing onto the wear scar. Over half of the total wear occurred within the first 1 000 cycles of sliding where 52% of the total lost material was removed (Figure A 8).

PAGE 29

29 5 6 MoS 2 /Sb 2 O 3 /Au Composite Coating The MoS 2 /Sb 2 O 3 /Au composite had an average friction coefficient of 0.024 (Figure A 9) The maxi mum wear scar depth was 177 nm and the maximum track width was 69 m (Figure B 5). This coating did not achieve steady state wear until 5,000 cycles, and t he steady state wear rate was 3.03E 09 mm 3 /Nm (Figure A 10 ). After the coating was run in, it experienced almost zero wear This coating had the highest initial percentage of its total lost volume removed in the first 1,000 cycles with 57%. Only 1 3% was lost during the steady state regime Cycles 8 000 and 10,000 resulted in an increase in volume from the previous cycle (Figure A 10 ). This can be attributed to wear from the ball surface.

PAGE 30

30 CHAPTER 6 DISCUSSION AND ANALYSIS OF RESULTS 6.1 Single Point, Complete Data, and Steady State Wear Rates Wear rate, k, is defined as the total volume in millimeters cubed divided by the quantity of the normal load in Newtons multiplied by the sliding distance in meters. Different ways exist to interpret wear curves. Here three methods are presented: single point wear rate ( k SP ), complete data wear rate ( k CD ), and steady state wear rate ( k SS ). Typically, mass or volume loss measurements are performed during a te st. However, due to the setup or conditions, this is sometimes not possible. In this situation the final volume loss would be determined and using the single data point, a single point wear rate is determined. This is not the best representation of the d ata; because, it does not take into the consideration of the wear in period which may dominate over the steady state wear. The k CD is reported when intermittent wear measurements are performed during a test, a least squares regression is fit through the entire data set. This is the most commonly reported wear rate, but like k SP it does not take into consideration the wear in period or the steady state wear. The k SS is achieved the intermittent wear measurements are performed and the data is broken up into different regions of wear in and steady state. A material couple can exhibit multiple steady state wear rates during operation This most often occurs during environment changes. 6. 2 Summary of Friction and Wear Results The temperature inside of the chamber was higher than ambient due to the electrical components mostly the linear stage, heating the environment Using the wear r ates defined in Section 6.1, t he tribology data from Sections 5. 2 5. 6 are summarized in Table 6 1 It follows that k SS should be the smallest or equal to

PAGE 31

31 k CD wear with the k SP having the highest value. Figure 6 1 illustrates this schematically. Table 6 1 Summary of results for the steady state wear testing in dry nitrogen. Composite Wear Rates, k (mm 3 /Nm) k SS uncertainty k SP k CD k SS MoS 2 /Ni 0.100 1.23E 06 5.46E 07 3.70E 07 1.10E 08 MoS 2 /Ti 0.051 1.59E 07 1.60E 07 1.60E 07 1.00E 07 MoS 2 /Sb 2 O 3 0.062 9.98E 08 5.05E 08 5.05E 08 1.90E 08 MoS 2 /Sb 2 O 3 /C 0.010 4.67E 08 2.29E 08 4.65E 08 2.82E 08 4.40E 09 MoS 2 /Sb 2 O 3 /Au 0.024 4.03E 08 1.85E 08 3.03E 09 5.50E 09 Figure 6 1. Illustration of different wear rate analyses. All of the composites that contained an Sb 2 O 3 phase had nearly an order of magnitude lower wear rate than the composites that only had metal inclusions. It should come as no surprise that the composite with the highest wear rate was MoS 2 / Ni since it was 95% MoS 2 and it is known to not be a low wear material. A result that is of interest is that the coating with the lowest wear MoS 2 /Sb 2 O 3 /Au did not have the lowest friction coefficient. Table 6 1 shows that low friction is not necessarily indicative of low wear. The MoS 2 /Sb 2 O 3 /Au coating and the MoS 2 /Sb 2 O 3 /graphite coating had very nearly the same wear rate, with MoS 2 /Sb 2 O 3 /Au being the lowest, but the MoS 2 /Sb 2 O 3 /graphite coating had friction that was over two times lower than the MoS 2 /Sb 2 O 3 /Au coating.

PAGE 32

32 All of these coatings exhibited very low wear. To illustrate this point the wear of the coating can be thought of in terms of how many layers lubricant, in this case MoS 2 are removed per cycle. Knowing the wear rates of the coatings, the wear scar width, and that one layer of MoS 2 is approximately 6 in depth ( 17 ) each coating on a verage takes tens to hundreds of cy cles to remove one atomic layer. The number of cycles calculated to remove one layer of MoS 2 for all the composites are located in Table 6 2 Table 6 2. Number of cycles to remove one layer of MoS 2 based on different we ar rate calculations Composite Depth (nm) Width(m) Cycles k S P k CD k S S MoS 2 /Ni 906 206 21 46 67 MoS 2 /Ti 246 155 117 117 117 MoS 2 /Sb 2 O 3 468 123 148 293 293 MoS 2 /Sb 2 O 3 /graphite 423 35 90 184 91 149 MoS 2 /Sb 2 O 3 /Au 177 69 206 448 2733 After the MoS 2 /Ni MoS 2 /Sb 2 O 3 and the MoS 2 /Sb 2 O 3 /Au coatings wore in, the number of cycles required to remove one atomic layer of MoS 2 increased by 2 10 times. The MoS 2 /Ti exhibited ex tremely steady wear resulting in the number of cycles to remove one la yer of MoS 2 being indep endent of the type of wear rate used. There was no wear in regime for this coating, and as stated previously, every 1,000 cycles removed 5 13% of the total volume.

PAGE 33

33 CHAPTER 7 CONCLUSION A tribometer to measure in situ wear rates of t ribological interfaces was designed and constructed. It was demonstrated that in a low humidity environment th at MoS 2 containing coatings exhibit low friction and low wear, and that the wear was reduced by one to two orders of magnitude with the inclusion of Sb 2 O 3 Additionally, i n a low humidity environment, it is necessary to reciprocate for tens to hundreds to thousands of cycles to remove just one layer of MoS 2 Of the coatings tested the MoS 2 /Sb 2 O 3 /C coating exhibited the lowest friction and the MoS 2 /Sb 2 O 3 /Au exhibited the least amount of wear.

PAGE 34

34 APPENDIX A FRICTION AND WEAR PL OTS Figure A 1. MoS 2 /Ni average friction coefficient per cycle as a function of sliding cycle. Figure A 2. MoS 2 /Ni t otal volume lost as a function of sliding cycle

PAGE 35

35 Figure A 3. MoS 2 /Ti average friction coefficient per cycle as a function of sliding cycle. Figure A 4. MoS 2 /Ti total volume lost as a function of sliding cycle.

PAGE 36

36 Figure A 5. MoS 2 /Sb 2 O 3 average friction coefficient per cycle as a function of sliding cycle. Figure A 6. MoS 2 /Sb 2 O 3 total volume lost as a function of sliding cycle.

PAGE 37

37 Figure A 7. MoS 2 /Sb 2 O 3 /graphite average friction coefficient per cycle as a function of sliding cycle. Figure A 8. MoS 2 /Sb 2 O 3 /graphite total volume lost as a function of sliding cycle.

PAGE 38

38 Figure A 9. MoS 2 /Sb 2 O 3 /Au average friction coefficient per cycle as a function of sliding cycle. Figure A 10. MoS 2 /Sb 2 O 3 /Au total volume lost as a function of sliding cycle.

PAGE 39

39 APPENDIX B SURFACE PROFILES Figure B 1. Surface profiles of the MoS 2 / Ni composite coating: (a) original surface, (b) cycle 2000, (c) cycle 4000, (d) cycle 6000, (e) cycle 8000, (f) cycle 10,000. Figure B 2. Surface profiles of the MoS 2 / Ti composite coating: (a) o riginal surface, (b) cycle 2000, (c) cycle 4000, (d) cycle 6000, (e) cycle 8000, (f) cycle 10,000.

PAGE 40

40 Figure B 3. Surface profiles of the MoS 2 /Sb 2 O 3 composite coating: (a) original surface, (b) cycle 2000, (c) cycle 4000, (d) cycle 6000, (e) cycle 8000, (f) cycle 10,000 Figure B 4. Surface profiles of the MoS 2 /Sb 2 O 3 /graphite composite coating: (a) original surface, (b) cycle 2000, (c) cycle 4000, ( d) cycle 6000, (e) cycle 8000, (f) cycle 10,000.

PAGE 41

41 Figure B 5. Surface profiles of the MoS 2 /Sb 2 O 3 /Au composite coating: (a) original surface, (b) cycle 2000, (c) cycle 4000, (d) cycle 6000, (e) cycle 8000, (f) cycle 10,000.

PAGE 42

42 LIST OF REFERENCES 1. J.H. Han, C.G. Kim, Composite Structures 72 218 (2006). 2. M. E. Bell, J. H. Findlay, Physical Review 59 33 (May 1 3, 1941). 3. C. Donnet, J. M. Martin, T. L. Mogne, M. Belin, Tribology International 29 123 (1996). 4. M. B. Peterson, R. L. Johnson, N ACA TN (1953). 5. A. J. Haltner, C. S. Oliver, Industrial & Engineering Chemistry Fundamentals 5 348 (1966). 6. W. O. Winer, Wear 10 422 (1967). 7. M. S. MIL M 7866B. (1965). 8. C. C. Baker, J. J. Hu, A. A. Voevodin, Surface & Coatings Technology 201 4224 (Dec 20, 2006). 9. D. G. Teer, Wear 250 1068 (Oct, 2001). 10. A. A. Voevodin, J. S. Zabinski, Wear 261 1285 (Dec 20, 2006). 11. J. S. Zabinski, M. S. Donley, N. T. Mcdevitt, Wear 165 103 (May 1, 1993). 12. N. M. Renevier, N. Lobiondo, V. C Fox, D. G. Teer, J. Hampshire, Surface & Coatings Technology 123 84 (Jan 10, 2000). 13. P. L. Dickrell et al. Tribology Letters 18 59 (Jan, 2005). 14. J. Williams, R. T. Hunt, Proceedings of the Royal Society of London Series A Mathematical and Physical Sciences 327 147 (1972). 15. R. S. Sayles, Tribology International 34 299 (May, 2001). 16. T. L. Schmitz, J. E. Action, D. L. Burris, J. C. Ziegert, W. G. Sawyer, Journal of Tribology Transactions of the ASME 126 802 (Oct, 2004). 17. R. Holi nski, Gansheim.J, Wear 19 329 (1972).

PAGE 43

43 BIOGRAPHICAL SKETCH Jason Steffens was born in Winter Park, FL in 1979. When he was eleven years old he moved to Eustis, FL where he was exposed to radiation that prevented his head making lemonade by attempting to use these characteristics to his advantage by pursuing a black belt in Brazilian Jiu Jitsu. His small head is in strumental in escaping triangle chokes and his long arms are ideal for securing a


xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID EVS343LOP_N00UBP INGEST_TIME 2012-12-07T20:37:13Z PACKAGE UFE0021830_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES