e,

E' T N

TUNGSTEN IN VACU )

esemcb Center

NATIONAL AERONAUTICS A N D S P A C E ADlVllNlSTRATlON e WASH NGTON. D. C. e APRIL 1969

I

I

NASA TECHNICAL NOTE

-...r-..... "'"' I Q :---

Z I-

< V")

<t Z

NASA TN D-5141 C!- ~ I

LOAN C-OPY: RETURN TO AF\"':L [WUL-2)

KfRTlAI\IO AFB. N ME)(

THE'THERMAL STABILITY AND FRICTION

OF THE DISULFIDES, DISELENIDES, AND DITELLURIDES OF MOLYBDENUM AND

TUNGSTEN IN VACUUM (10-9 TO 10-6 TORR)

by,Jpllliam A. Brainard

Leu/is Research Center

Clevellt:1Zd, Ohio

NATIONAL AERONAUTICS AND SPACE ADMINISTRATION • WASHINGTON, D. C. • APRil 1969

https://ntrs.nasa.gov/search.jsp?R=19690013627 2018-11-02T14:08:54+00:00Z

THE THERMAL STABILITY AND FRICTION OF THE DISULFIDES,

DISELENIDES, AND DITELLUHDES OF MOLYBDENUM AND

TUNGSTEN IN VACUUM TO TORR)

By William A. Brainard

Lewis Research Center Cleveland, Ohio

T

..

NASA TN D-5141

-

TECH LIBRARY KAFB, NM

111111111111111111111111111111111111111111111 0069465

THE THERMAL STABILITY AND FRICTION OF THE DISULFIDES,

DISELENIDES, AND DITELLURIDES OF MOLYBDENUM AND

TUNGSTEN IN VACUUM (10-9

TO 10 -6 TORR)

By William A. Brainard

Lewis Research Center Cleveland, Ohio

NATIONAL AERONAUTICS AND SPACE ADMINISTRATION

For sale by the Clearinghouse for Federal Scientific and Technical Information Springfield, Virginia 22151 - CFSTI price $3.00

- I

ABSTRACl

Thermal stability and friction experiments were conducted with the disulfides, diselenides, and ditellurides of molybdenum and tungsten in vacuum (lo-' to Weight loss rates were plotted a s a function of temperature: thermal dissociation tem- peratures were observed by mass spectrometry. Friction experiments showed the materials were limited a s high temperature lubricants by evaporation, dissociation, and wear. Generally, a s lubricants, the diselenides provided the best performance (fk = 0.04 to 0.08) up to 1400' F (760' C), whereas the ditellurides provided the poorest because of a low thermal dissociation temperature.

torr).

ii

I 1 I 1 1 1 1 1 I1 1111 I II II I I 111 II 11111 II

ABSTRACT

Thermal stability and friction experiments were conducted with the disulfides,

diselenides, and ditellurides of molybdenum and tungsten in vacuum (10-9 to 10-6 torr).

Weight loss rates were plotted as a function of temperature; thermal dissociation tem­

peratures were observed by mass spectrometry. Friction experiments showed the

materials were limited as high temperature lubricants by evaporation, dissociation,

and wear. Generally. as lubricants, the diselenides provided the best performance

(fk = 0.04 to 0.08) up to 14000 F (7600 C), whereas the ditellurides provided the poorest

because of a low thermal dissociation temperature.

ii

IIlilllllll" • 1.11 •• -11111 II III1II II 1111111 III I III II 1111111

THE THERMAL STABILITY AND FRICTION OF THE DISULFIDES, DISELENIDES, AND

DITELLURIDES OF MOLYBDENUM AND TUNGSTEN IN VACUUM (10-9 TO 10-6 TORR)

by W i l l i a m A. B r a i n a r d

Lewis Research Center

SUMMARY

6 Thermal stability and friction experiments were conducted in vacuum (lo-' to 10- torr) with the disulfides, diselenides, and ditellurides of molybdenum and tungsten in order that useful operating temperature ranges could be established. Thermal stability experiments were conducted in vacuum by heating small pressed compacts of each of the six inorganic compounds and monitoring weight loss ra tes and observing dissociation products on a mass spectrometer. The resul ts show the disulfides are the most ther- mally stable and the ditellurides the least stable. The temperatures at which dissociation was observed by mass spectrometry were as follows: M0S2, 2000' F (1090' C); WS2, 1900' F (1040' C); MoSe2, 1800' F (980' C); WSe2, 1700' F (930' C); MoTe2 and WTe2, 1300' F (700' C). It was observed that the weight loss rate of WS2 increased more slowly with increasing temperature above 1700' F (930' C) than M0S2.

6.35-centimeter-diameter cobalt-base alloy disk which rotated in vacuum against a 0.47 5-centimeter-radius hemispherical r ider . Tests were conducted at temperatures to 1650' F (900' C) in a pressure range of to to r r . The resul ts showed that thin, unreplenished films of the inorganic compounds were limited as effective lubricants to temperature below those expected from the thermal stability limits because of evap- oration of the compounds. 1200' F (650' C) and 1350' F (730' C), respectively. Above these temperatures the materials were lost from the disks presumably by evaporation and wear. Both disel- enides provided lubrication up to 1400' F (760' C), beyond which temperature, they were lost from the lubricated surface by evaporation, dissociation, and wear. The ditellurides generally were poor as high temperature lubricants because of their rela- tively low (1300' F or 700' C) thermal dissociation temperature. It was shown that friction increases with time at temperatures where lubricant loss occurred.

I

~

I 1

Friction tes ts were conducted using thin rubbed films of each of the compounds on a

1 Thin films of M0S2 and WS2 provided effective lubrication to

J

THE THERMAL STABILITY AND FRICTION OF THE DISULFIDES, DISELENIDES, AND

DITELLURIDES OF MOLYBDENUM AND TUNGSTEN IN VACUUM (10-9 TO 10-6 TORR)

by William A. Brainard

Lewis Research Center

SUMMARY

Thermal stability and friction experiments were conducted in vacuum (10-9 to 10-6

torr) with the disulfides, diselenides, and ditellurides of molybdenum and tungsten in

order that useful operating temperature ranges could be established. Thermal stability

experiments were conducted in vacuum by heating small pressed compacts of each of the

six inorganic compounds and monitoring weight loss rates and observing dissociation

products on a mass spectrometer. The results show the disulfides are the most ther­

mally stable and the ditellurides the least stable. The temperatures at which dissociation

was observed by mass spectrometry were as follows: MoS2, 20000 F (10900 C); WS2,

19000 F (10400 C); MoSe2, 18000 F (9800 C); WSe2, 17000 F (9300 C); MoTe2 and WTe2, 13000 F (7000 C). It was observed that the weight loss rate of WS2 increased

more slowly with increasing temperature above 17000 F (9300 C) than MoS2 .

Friction tests were conducted using thin rubbed films of each of the compounds on a

6. 35-centimeter-diameter cobalt-base alloy disk which rotated in vacuum against a

0.475-centimeter-radius hemispherical rider. Tests were conducted at temperatures to 16500 F (9000 C) in a pressure range of 10-8 to 10-6 torr. The results showed that

thin, unreplenished films of the inorganic compounds were limited as effective lubricants

to temperature below those expected from the thermal stability limits because of evap­

oration of the compounds. Thin films of MoS2 and WS2 provided effective lubrication to

12000 F (6500 C) and 13500 F (7300 C), respectively. Above these temperatures the

materials were lost from the disks presumably by evaporation and wear. Both disel­

enides provided lubrication up to 14000 F (7600 C), beyond which temperature, they

were lost from the lubricated surface by evaporation, dissociation, and wear. The

ditellurides generally were poor as high temperature lubricants because of their rela­

tively low (13000 For 7000 C) thermal dissociation temperature. It was shown that

friction increases with time at temperatures where lubricant loss occurred.

IN TROD UCTl ON

M0S2, MoTe2, MoSe2, WS2, WTe2, and WSe2 have found wide usage as solid lubri- cant materials. They possess intrinsic lubricating ability which is associated with their layer-lattice structure. These solid lubricants have been used very successfully in areas where conventional lubricants could not be used because of extremes of temperature, pressure, o r atmosphere (refs. 1 to 6). Recently, data were presented (ref. 7) on the oxidation characteristics of several of the solid lubricants derived from heavy metals. There is, however, a lack of information concerning the stability of these materials in a

consistent. widely used inorganic solid lubricant, is variously given as 1832' F ( l O O O o C) (ref. 8), 2012' F ( l l O O o C) (ref. 5), 2498' F (1370' C) (ref. 9), and 3002' to 3092' F (1650' to 1700° C) (ref. 10).

solid lubricant materials in a vacuum environment at elevated temperatures.

vacuum environment at elevated temperatures. What data a r e available a r e often not %

For example, the thermal dissociation temperature of MoS2, the most

Further, there is a lack of complete information on the lubricating ability of these

The objectives of this investigation were to examine the following: (1) The thermal stability of the molybdenum and tungsten disulfides, diselenides,

(2) The lubricating ability of these compounds at elevated temperatures in vacuum and ditellurides in vacuum

The six compounds were pressed into compacts, and heated in vacuum to determine their thermal stability.

f i lms of each of the six compounds at temperatures to 1650' F (900' C) . Friction experiments were also conducted in vacuum using thin, rubbed lubricant

APPARATUS

The r ma I S ta bi I i ty A ppa ra t u s

The apparatus shown schematically in figure 1 was used to determine the vacuum thermal stability of the inorganic solid lubricants. The apparatus is similar to that described in reference 4. A small (200-mg) pressed compact of the solid lubricant material is placed in a tantalum pan and suspended from a highly sensitive recording electronic balance. In these experiments, weight losses as small as 1. O X ~ O - ~ gram could be detected. The suspended sample is enclosed by a 1.91-centimeter-diameter tantalum tube, which was used as a thermal susceptor. Directly under the specimen in the tube was a large bead Chromel-Alumel thermocouple which was used to monitor temperatures inside the tube. This thermocouple was calibrated against a 0.0025- centimeter-diameter thermocouple embedded in the test sample. A t equilibrium tem-

2

INTRODUCTION

MoS2, MoTe2' MoSe2, WS2, WTe2, and WSe2 have found wide usage as solid lubri­cant materials. They possess intrinsic lubricating ability which is associated with their layer-lattice structure. These solid lubricants have been used very successfully in areas

where conventional lubricants could not be used because of extremes of temperature, pressure, or atmosphere (refs. 1 to 6). Recently, data were presented (ref. 7) on the

oxidation characteristics of several of the solid lubricants derived from heavy metals.

There is, however, a lack of information concerning the stability of these materials in a

vacuum environment at elevated temperatures. What data are available are often not '\..

consistent. For example, the thermal dissociation temperature of MoS2, the most

widely used inorganiC solid lubricant, is variously given as 18320 F (10000 C) (ref. 8),

20120 F (11000 C) (ref. 5), 24980 F (13700 C) (ref. 9), and 30020 to 30920 F (16500 to

17000 C) (ref. 10). Further, there is a lack of complete information on the lubricating ability of these

solid lubricant materials in a vacuum environment at elevated temperatures.

The objectives of this investigation were to examine the following:

(1) The thermal stability of the molybdenum and tungsten disulfides, diselenides,

and ditellurides in vacuum (2) The lubricating ability of these compounds at elevated temperatures in vacuum

The six compounds were pressed into compacts, and heated in vacuum to determine

their thermal stability. Friction experiments were also conducted in vacuum using thin, rubbed lubricant

films of each of the six compounds at temperatures to 16500 F (9000 C).

APPARATUS

Thermal Stability Apparatus

The apparatus shown schematically in figure 1 was used to determine the vacuum

thermal stability of the inorganic solid lubricants. The apparatus is similar to that

described in reference 4. A small (200-mg) pressed compact of the solid lubricant

material is placed in a tantalum pan and suspended from a highly sensitive recording

electronic balance. In these experiments, weight losses as small as 1. OXlO- 5 gram

could be detected. The suspended sample is enclosed by a 1. 91-centimeter-diameter

tantalum tube, which was used as a thermal susceptor. Directly under the speCimen in the tube was a large bead Chromel-Alumel thermocouple which was used to monitor

temperatures inside the tube. This thermocouple was calibrated against a 0.0025-centimeter-diameter thermocouple embedded in the test sample. At equilibrium tem-

2

// Electronic balance-.

ih Counterweight

I Liquid n i t rogen coi ls7 I \

1 Tanta lum t u b e 7 To mass .==d

spectrometer " Electron gun7

Tanta lum pan

CD-10259-15

Figure 1. - Schematic of thermal stabi l i ty vacuum apparatus.

perature, no discernible temperature difference between the two thermocouples W a s

observed. To further insure temperature measurement accuracy, the temperature measuring equipment was calibrated with the standard millivolt reference, with the maximum e r r o r being *loo R (5.5' C) at the maximum operating temperature.

gun filaments which were used to heat the tube to a maximum inside temperature of 2300' F (1260' C). The gun filament voltage was controlled by a proportional controller using the thermocouple for a signal. The tube temperature could be controlled to within *loo F (*5.5' C) .

trap any evaporated o r desorbed material. Glass slides were mounted on these surfaces so condensate could later be analyzed by X-ray techniques.

could be monitored and dissociation products observed.

Approximately 0 .5 centimeter from the outside wall of the tube were two electron

Directly above the furnace and on both sides are liquid nitrogen cooled surfaces to

A mass spectrometer was mounted adjacent to the test area so that gaseous species

3

liquid nitrogen coils7\ I \

Ther

Counterweight

rSpecimen , , Tom~:.:~

spectrometer

Toot,t"m,.. ~

CD-I0259-15

Figure 1. - Schematic of thermal stability vacuum apparatus.

perature, no discernible temperature difference between the two thermocouples was observed. To further insure temperaturb measurement accuracy, the temperature measuring equipment was calibrated with the standard millivolt reference, with the maximum error being ±10o R (5. 50 C) at the maximum operating temperature.

Approximately 0.5 centimeter from the outside wall of the tube were two electron gun filaments which were used to heat the tube to a maximum inside temperature of 23000 F (12600 C). The gun filament voltage was controlled by a proportional controller using the thermocouple for a signal. The tube temperature could be controlled to within ±100 F (±5.5° C).

Directly above the furnace and on both sides are liquid nitrogen cooled surfaces to trap any evaporated or desorbed material. Glass slides were mounted on these surfaces so condensate could later be analyzed by X-ray techniques.

A mass spectrometer was mounted adjacent to the test area so that gaseous species could be monitored and dissociation products observed.

3

D r i v e shaft

Cryopumping coil-, ,-=-Support bear ings

CD-10265-15

ii To i o n pump

F igure 2. - Vacuum- f r i c t ion apparatus.

4 4

/ /

Magnetic drive

L Electron gu n feedth rough

Drive shaft

_:'::'>-Support bearings

l~iBiETITIWilif :=--=> Sorption forepumping

CD-10265-15

jj To ion pump

Figure 2. - Vacuum-friction apparatus.

The system was evacuated by two 400-liter-per-second ion pumps and a titanium getter pump. Ultimate pressure is lo-'' t o r r but during heating, the pressure rose into

-8 the range, 10 to to r r .

Fr i c t i on Apparatus

Friction experiments with the inorganic solid film lubricants were conducted in the vacuum-friction apparatus shown in figure 2. A set of tes t specimens consisted of a 6.35- centimeter-diameter flat disk with an inorganic rubbed solid lubricant film applied to it and a 0.475-centimeter-radius bullet specimen. The disk specimen is mounted horizon- tally on a magnetically driven shaft. The rider is mounted in a holder which bolts to a bellows-sealed, gimbal-mounted load a rm. The hemispherical tip of the r ider specimen is loaded against the flat side of the rotating disk by dead weight loading. The friction force is transmitted by the a r m to a s t ra in gage assembly mounted outside the chamber. The output of the strain gage is read on a multirange potentiometer.

to rough pump the system to 1 micron. A 500-liter-per-second ion pump is used for pumping in the UHV range. A stainless steel coil of 0.955 centimeter diameter, 6 meters long can be used for cryopumping of the system. No cryopumping was used in these experiments. The pumping system is free from any organic contamination that might result from mechanical o r diffusion pumping.

A cold cathode ionization gage located adjacent to the test specimens was used to monitor the chamber pressure. Pressure varied during the experiment, depending upon tes t temperature, but never exceeded

The disk specimen w a s heated by a 1750-watt electron gun. The electron gun fila- ment was located approximately 0.63 centimeter from the edge of the disk specimen. The disk temperature was monitored by a Chromel-Alumel thermocouple located 0.158 centimeter from the disk surface near the rider-disk contact zone. The thermocouple temperature was calibrated against a contacting thermocouple while the disk w a s sta- tionary. The temperature measured by the proximity thermocouple was accurate to within approximately 25' F (14' C).

Attached to the lower end of the chamber a r e two cryosorption pumps which a r e used

t o r r .

MATERIALS

Some physical properties of the molybdenum and tungsten disulfides, diselenides, and ditellurides a r e given in table I(a) . With the exception of the orthorhombic WTe2, the materials all have the hexagonal crystal structure. The materials used in this in- vestigation were obtained from a commercial supplier; the purity levels were 99.5 percent with the exception of MoS2 which was 99.99 percent and WS2 which was 99 percent.

5

I~

The system was evacuated by two 400-liter-per-second ion pumps and a titanium

getter pump. Ultimate pressure is 10-10 torr but during heating, the pressure rose into -8 6 the range, 10 to 10- torr.

Friction Apparatus

Friction experiments with the inorganic solid film lubricants were conducted in the

vacuum-friction apparatus shown in figure 2. A set of test specimens consisted of a 6. 35-

centimeter-diameter flat disk with an inorganic rubbed solid lubricant film applied to it

and a O. 475-centimeter-radius bullet specimen. The disk specimen is mounted horizon­

tally on a magnetically driven shaft. The rider is mounted in a holder which bolts to a

bellows-sealed, gimbal-mounted load arm. The hemispherical tip of the rider specimen

is loaded against the flat side of the rotating disk by dead weight loading. The friction

force is transmitted by the arm to a strain gage assembly mounted outside the chamber. The output of the strain gage is read on a multirange potentiometer.

Attached to the lower end of the chamber are two cryosorption pumps which are used

to rough pump the system to 1 micron. A 500-liter-per-second ion pump is used for

pumping in the UHV range. A stainless steel coil of 0.955 centimeter diameter, 6 meters

long can be used for cryopumping of the system. No cryopumping was used in these

experiments. The pumping system is free from any organic contamination that might

result from mechanical or diffusion pumping.

A cold cathode ionization gage located adjacent to the test specimens was used to

monitor the chamber pressure. Pressure varied during the experiment, depending upon

test temperature, but never exceeded 10-6 torr.

The disk specimen was heated by a 1750 -watt electron gun. The electron gun fila­

ment was located approximately 0.63 centimeter from the edge of the disk specimen.

The disk temperature was monitored by a Chromel-Alumel thermocouple located 0.158

centimeter from the disk surface near the rider-disk contact zone. The thermocouple

temperature was calibrated against a contacting thermocouple while the disk was sta­

tionary. The temperature measured by the proximity thermocouple was accurate to within

approximately 250 F (140 C).

MATERIALS

Some physical properties of the molybdenum and tungsten disulfides, diselenides,

and ditellurides are given in table l(a). With the exception of the orthorhombic WTe2' the materials all have the hexagonal crystal structure. The materials used in this in­

vestigation were obtained from a commercial supplier; the purity levels were 99.5

percent with the exception of MoS2 which was 99.99 percent and WS2 which was 99 percent.

5

:ompound

MoS2 MoTe2 MoSe2

WTe2 WSe2

ws2

ii

TABLE I. - PROPERTIES AND ANALYSES OF MATERIALS

(a) Propert ies of the molybdenum and tungsten disulfides, diselenides, and ditellurides (ref. 2)

stability s t ructure

Hexagonal Hexagonal Hexagonal Hexagonal

Orthorhombic Hexagonal

Gray I I Compound I

MoS2

MoTe2

MoSe2

ws2 WTe2

WSe2

Molecular weight

160.07 351.14 253.86 247.98 439.05 341.78

7.50 3.29

9.0 3.29

(b) Analyses of mater ia ls ~~

Supplier's analysis

--- 72.00 percent tellurium

61.88 percent selenium

26.15 percent sulfur

57.46 percent tellurium

45.50 percent selenium

c-direction

12.295 13.97 12.80 12.97 _ _ _ _ _ 12.95

Stoichometric percentagc

40.06 percent sulfur

72.65 percent tellurium

62.20 percent tellurium

25.77 percent sulfur

58.12 percent tellurium

46.20 percent selenium

cm

8. 5X102 8. 7X102 1. 8x10-2 1. 4 x 1 O 1 3 . 1 ~ 1 0 - ~ 1. 1x102

OC

350 400 400 440 _ _ - 3 50

Analysis of the materials a r e given in table I(b) along with the composition for the stoichometric compounds. No attempts were made to further purify the materials.

PROCEDURE

T h e r ma I S ta bi I i ty Expe r i men t s

Thermal stability experiments were conducted with compacts pressed from each of the six solid lubricant materials. Each compact w a s suspended from the electronic balance inside the tantalum tube susceptor. After system pumpdown and bakeout (to a maximum specimen temperature of 212' F (100' C), the condensing surfaces were cooled with liquid nitrogen and specimen heating started.

The specimens were heated in steps of 100' F (59' C) from 400' to 2300' F (204' to 1260' C) or until excessive dissociation ra tes occurred. A t each temperature, weight loss readings and mass spectrometer readings were made. Large weight loss readings

6 6

TABLE I. - PROPERTIES AND ANALYSES OF MATERIALS

(a) Properties of the molybdenum and tungsten disulfides, diselenides, and ditellurides (ref. 2) -

Compound Crystal Color Molecular Density, Lattice parameter, Resistivity , Air

structure weight g/cm3 A I fl, stability,

I cm

a-direction c-direction

MoS2 Hexagonal Gray 160.07 4.80 3.16 12.295 8.5X102

MoTe2 Hexagonal

j

351. 14 7.7 3.52 13.97 8.7x102

MoSe2 Hexagonal 253.86 6.9 3.29 12.80 1. axlO-2

WS2 Hexagonal 247.98 7.50 3.29 12.97 1. 4x101

WTe2 Orthorhombic 439.05 9.4 ---- ----- 3.1xIO-3

WSe2 Hexagonal 341.78 9.0 3.29 12.95 1. Ix 102

(b) Analyses of materials .. -

Compound Supplier's analysis Stoichometric percentage

MoS2 --- 40. 06 percent sulfur

MoTe2 72. 00 percent tellurium 72.65 percent tellurium

MoSe2 61. 88 percent selenium 62.20 percent tellurium

WS2 26.15 percent sulfur 25.77 percent sulfur

WTe2 57. 46 percent tellurium 58.12 percent tellurium

WSe2 45.50 percent selenium 46. 20 percent selenium

Analysis of the materials are given in table I(b) along with the composition for the

stoichometric compounds. No attempts were made to further purify the materials.

PROCEDURE

Thermal Stability Experiments

°c

350 400 400

440 ---350

Thermal stability experiments were conducted with compacts pressed from each of

the six solid lubricant materials. Each compact was suspended from the electronic

balance inside the tantalum tube susceptor. After system pumpdown and bakeout (to a

maximum specimen temperature of 2120 F (1000 C), the condensing surfaces were

cooled with liquid nitrogen and speCimen heating started. The specimens were heated in steps of 1000 F (590 C) from 4000 to 2300

0 F (204

0 to

12600 C) or until excessive dissociation rates occurred. At each temperature, weight

loss readings and mass spectrometer readings were made. Large weight loss readings

coupled with the observation of sulphur, tellurium, o r selenium peaks on the mass spectrometer were indicative of thermal dissociation. After heating to maximum temper - ature and subsequent cooling, the chamber was opened and the condensate on the glass slides analyzed with X-ray techniques.

Fr i c t i on Experiments

Friction testing of the solid lubricant materials was carr ied out using very thin rubbed lubricant f i lms on a cobalt alloy substrate. The cobalt-base alloy (32 percent C r , 17 percent W, 2.5 percent C, 3.0 percent max Fe, 2.5 percent max Ni, and 41 percent Co) was chosen as the disk material for the following reasons:

(1) It has good hot hardness characterist ics. (2) Preliminary friction tests with bare metal combinations showed no significant

(3) The alloy has been used to make high temperature ball bearings. change in friction upon heating to the maximum testing temperatures.

The r ider was also machined from the cobalt alloy. Hardness of the r iders and disks was 55 to 57 Rockwell C.

Each lubricant film was applied in the same manner. A new set of specimens w a s used for each friction run. The disk was sandblasted to an average roughness of 40 to 45 microinches r m s . Following blasting, each disk was rinsed with acetone, and then scrubbed with moist polishing grade aluminum oxide. It was then rinsed with distilled water and finally with ethyl alcohol. The r ider specimens were cleaned in the same manner. Following cleaning, a coarse cloth w a s used to rub the solid lubricant on the disk surface in the direction of sliding until a smooth uniform film w a s achieved. No attempt was made to control o r measure film thickness. Typical surface profiles and photomicrographs are shown in figure 3.

vacuum chamber. Af te r subsequent system pumpdown and bakeout, the specimens were heated to test temperature and held there for 30 minutes before the friction test was started. Sliding was then initiated and the friction recorded.

Sliding continued for 30 minutes, and the representative friction for that test was recorded. A ser ies of runs, each with a new set of specimens and newly applied films, were made to temperatures of 1650' F (900' C) for each of the solid lubricant materials.

Following application of the lubricant -film, the specimens were mounted in the

7

coupled with the observation of sulphur, tellurium, or selenium peaks on the mass spectrometer were indicative of thermal dissociation. After heating to maximum temper­ature and subsequent cooling, the chamber was opened and the condensate on the glass slides analyzed with X-ray techniques.

Friction Experiments

Friction testing of the solid lubricant materials was carried out using very thin rubbed lubricant films on a cobalt alloy substrate. The cobalt-base alloy (32 percent Cr, 17 percent W, 2. 5 percent C, 3.0 percent max Fe, 2. 5 percent max Ni, and 41 percent Co) was chosen as the disk material for the following reasons:

(1) It has good hot hardness characteristics. (2) Preliminary friction tests with bare metal combinations showed no Significant

change in friction upon heating to the maximum testing temperatures. (3) The alloy has been used to make high temperature ball bearings.

The rider was also machined from the cobalt alloy c Hardness of the riders and disks was 55 to 57 Rockwell C.

Each lubricant film was applied in the same manner. A new set of specimens was used for each friction run. The disk was sandblasted to an average roughness of 40 to 45 microinches rms. Following blasting, each disk was rinsed with acetone, and then scrubbed with moist polishing grade aluminum oxide. It was then rinsed with distilled water and finally with ethyl alcohol. The rider specimens were cleaned in the same manner. Following cleaning, a coarse cloth was used to rub the solid lubricant on the disk surface in the direction of sliding until a smooth uniform film was achieved. No attempt was made to control or measure film thickness. Typical surface profiles and photomicrographs are shown in figure 3.

Following application of the lubricant Iilm, the specimens were mounted in the vacuum chamber. After subsequent system pumpdown and bakeout, the specimens were heated to test temperature and held there for 30 minutes before the friction test was started. Sliding was then initiated and the friction recorded.

Sliding continued for 30 minutes, and the representative friction for that test was recorded. A series of runs, each with a new set of specimens and newly applied films, were made to temperatures of 16500 F (9000 C) for each of the solid lubricant materials.

7

(Xl

0.002 cm 0.002 cm

O. OlD cm 0.010 cm

(a-l) Before appl ication. (a-2) After application.

(a) Surface profile.

(b-l) Before appl ication. (b-2) After appli cation.

(b) Photomicrograph.

Figu re 3. - Typical disk surface profile and photomicrograph before and after application of lubricant film .

-- .------------

00

0.002 cm 0.002 cm

O.OlD cm 0.010 cm

(a-ll Before appl ication . (a-2) After application.

(a) Surface prof ile.

(b-ll Before appl ication . (b-2) After application.

(b) Photomicrograph.

Figu re 3. - Typical disk su rface proli Ie and photom icrog raph before and after appl ication of I ubrica nt iii m.

L_

RESULTS AND DISCUSSION

Thermal Stabi l i ty Resul ts

Pre sed compacts of each of the solid lubricant compounds were heated in vacuum and the rate of weight loss measured. The weight loss ra te curves for the disulfides of molybdenum and tungsten are given in figure 4. No weight loss was observed for MoS2 up to 1700' F (930' C) . Beyond 1700' F (930' C) , the weight loss rate increased with temperature. A t 2000° F (1090' C), sulfur peaks were observed on the mass spectrom- eter , indicating the liberation of sulfur because of thermal dissociation. Dissociation undoubtly begins at a lower temperature but the amount of elemental sulfur liberated was not sufficient for detection. temperature of M0S2, while reference 12 gives 1922' F (1050' C). Reference 13 heated MoS2 powder in vacuum to 1832' F (lOOOo C) and reported a constant weight loss rate at 1742' F (950' C) which is in good agreement with the data of figure 4.

1 Reference 11 gives 842' F (450' C) a s the sublimation

(a) Weight loss rate of MOS2.

S u l f u r peaks I observed o n mass spectrometer

7 'I 'I77 1'1 ' IT7 'I 400 800 1200 1600 ZOO0 2400

Temperature, O F

I U 700 1000 1xwl

I I 100 400

Temperature, "C

(b) Weight loss rate of W S p

Figure 4. - Weight loss rate cu rves for MoS2 and WS2 when heated in vacuum (10-9 to 10-8 torr).

9

i

RESULTS AND DISCUSSION

Thermal Stability Results

Pressed compacts of each of the solid lubricant compounds were heated in vacuum and the rate of weight loss measured. The weight loss rate curves for the disulfides of molybdenum and tungsten are given in figure 4. No weight loss was observed for MoS2 up to 17000 F (9300 C). Beyond 17000 F (9300 C), the weight loss rate increased with temperature. At 20000 F (10900 C), sulfur peaks were observed on the mass spectrom­eter, indicating the liberation of sulfur because of thermal dissociation. Dissociation undoubtly begins at a lower temperature but the amount of elemental sulfur liberated was not sufficient for detection. Reference 11 gives 8420 F (4500 C) as the sublimation temperature of MoS2, while reference 12 gives 19220 F (10500 C). Reference 13 heated MoS2 powder in vacuum to 18320 F (10000 C) and reported a constant weight loss rate at 17420 F (9500 C) which is in good agreement with the data of figure 4.

10-6

10-7

, 1 '_"'" 'T

S u Ifu r peaks observed on mass I spectrometer 1 +

l !

(a) Weight loss rate of MoS2'

Sulfur peaks I observed on mass I + spectrometer ~ ~ +

~

10-8

1O-90~_~~~~-_---c'-------'-:~-2:-:'OOO---:!2400 400 800 1200 1600

Temperature, OF

100 400 700 1000 1300 Temperature, °C

(b) Weight loss rate of WS2'

Figure 4. - Weight loss rate curves for MoS2 and WS2 when heated in vacuum (10-9 to 10-8 torr).

9

The WS2 compact began to have a detectable loss rate at 1600' F (870' C), and sulfur peaks were observed at 1900' F (1040' C). The rate-temperature slope for WS2 is less than that of M0S2. It is of interest to note that, during oxidation studies of M0S2 and WS2 (ref. 14), it was shown that MoS2 oxidizes slower than WS2 at lower tempera- tures, but at the higher temperatures, the rate of oxidation of M0S2 is more rapid than for WS2.

The weight loss rate curves for MoSe2 and WSe2 are presented in figure 5. Both these compounds exhibit weight loss at temperatures lower than the equivalent disulfides. Selenium peaks were observed at 1800' F (980' C) for MoSe2 and 1700' F (930' C) for WSe2. Again dissociation was probably occurring at lower temperatures but could not be detected. Both diselenides have similar rate-temperature slopes. A t the lower temperatures some weight losses were detected; these losses were likely impurities.

The resul ts for the ditellurides a r e given in figure 6. Both of these compounds exhibited high weight loss rates up to 800' F (427' C) with the liberation of tellurium. It is believed that these peaks at 800' F (427' C) represent the liberation of uncombined

10-6 Probably impur i t ies

, 'I? ~ 17

Selenium peaks observed on mass

I 1

(a) Weight loss rate of MoSe2. E V

Is) - W- c m L VI

4 10-jr

10-7 I-

10-8 C I-

10-9 - 0

Selenium peaks observed on mass 4

Pro ba b I y impur i t ies

I 400 800 1200 1600 2000

Temperature, "F

I 2400

- 1 I 100 400 700 1000 1340

Temperature, "C

Figure 5. -Weigh t loss rate curves for MoSe2 and WSe2 when heated in vacuum to torr) .

10

The WS2 compact began to have a detectable loss rate at 16000 F (8700 C), and sulfur peaks were observed at 19000 F (10400 C). The rate-temperature slope for WS2 is less than that of MoS2. It is of interest to note that, during oxidation studies of MoS2 and WS2 (ref. 14), it was shown that MoS2 oxidizes slower than WS2 at lower tempera­tures, but at the higher temperatures, the rate of oxidation of MoS2 is more rapid than for WS2 .

The weight loss rate curves for MoSe2 and WSe2 are presented in figure 5. Both these compounds exhibit weight loss at temperatures lower than the equivalent disulfides. Selenium peaks were observed at 18000 F (9800 C) for MoSe2 and 17000 F (9300 C) for WSe2. Again dissociation was probably occurring at lower temperatures but could not be detected. Both diselenides have similar rate-temperature slopes. At the lower temperatures some weight losses were detected; these losses were likely impurities.

The results for the ditellurides are given in figure 6. Both of these compounds exhibited high weight loss rates up to 8000 F (4270 C) with the liberation of tellurium. It is believed that these peaks at 8000 F (4270 C) represent the liberation of uncombined

10

10-3

10-4

10-5

, Denotes (6xlO-9 (gHcm-2Hsec-1)

Selenium peaks I observed on mass. , + spectrometer ~ +

! 10-6 Probably

impurities

10-7 + + + '";"' 10-8 ~ ~ "f "~'D ~- 10-9 '-----'------'--E ~ :§ ",-

E ~ 10-3

:E .*, 10-4 os:

10-5

10-6

10-7

10-8

(a) Weight loss rate of MoSe2'

Selenium peaks observed on mass + spectrometer ! +

!

10-9'-------'------'------' I I o 400 800 1200 1600 2000

Temperature, OF

I I 100 400 700 1000

Temperature, °C

Figure 5. - Weight loss rate curves for MoSe2 and WSe2 when heated in vacuum (10-9 to 10-8 torr).

J

2400

1300

10-6

"

T e l l u r i u m peaks observed o n mass spectrometer (pre- sumed u nreacted t e l l u r i u m )

T e l l u r i u m peaks

. I 1

p 10-8 I I I I I I Q,

Y " E (a) Weight loss rate of MoTe2. - - cn -

T e l l u r i u m peaks observed

su med u nreacted t e l l u r i u m )

a- c

VI

z 10-4

2000 I U I

400 800 1200 1600 Temperature, "F

10-go t

1 100 400 700 1000

Temperature, "C (b) Weight loss rate of WTe2.

Figure 6. -We igh t loss rate curves for MoTe2 and WTe2 when heated in vacuum to IO-* t o r r ) .

tellurium rather than the reduction to some lower order telluride. X-ray diffraction and fluorescence patterns were taken of MoTe2 powder before and after heating to 800' F (427' C) in vacuum. The fluorescence patterns indicated a definite loss of tellurium, but the diffraction patterns showed no change in structure.

Upon further heating, weight loss ra tes again increased and tellurium peaks w e r e monitored on the mass spectrometer at 1300' F (700° C).

Following each run, the chamber was opened and the compact and condensate were analyzed by X-ray techniques. The results of the analyses on the condensate are given in table II. Only the condensate from the ditellurides and WS2 were thick enough for detection by X-ray analysis. The pressed compacts were analyzed by X-ray fluores- cence and the ratios of its constituent elements peaks were compared with prerun ratios. Figure 7 shows the change in the sulfur to molybdenum ratio in MoSz after heating in vacuum. There was still sulfur remaining in the compact but the ratio of intensities changed from 0.45 to 0.18 indicating the sample consisted of Mo and M0S2. In the case

b

*

11

10-6

10-7

10-8

Tellurium peaks observed on mass spectrometer (pre­sumed unreacted tellurium)

l~

Tellurium peaks

~

(a) Weight loss rate of MoTe2'

Tellurium peaks observed on mass spectrometer (pre­sumed unreacted tellurium)

l

10-9 !-------o"-::-------c:'.:-:------:cc:':-:-------c~--_:::' o 400 800 1200 1600 2000 Temperature, of

100 400 700 1000 Temperature, °C

(b) Weight loss rate of WTe2'

Figure 6. - Weight loss rate curves for MoTe2 and WTe2 when heated in vacuum (l0-9 to 10-8 torr).

tellurium rather than the reduction to some lower order telluride. X-ray diffraction and fluorescence patterns were taken of MoTe2 powder before and after heating to 8000 F (4270 C) in vacuum. The fluorescence patterns indicated a definite loss of tellurium, but the diffraction patterns showed no change in structure.

Upon further heating, weight loss rates again increased and tellurium peaks were monitored on the mass spectrometer at 13000 F (7000 C).

Following each run, the chamber was opened and the compact and condensate were analyzed by X-ray techniques. The results of the analyses on the condensate are given in table II. Only the condensate from the ditellurides and WS2 were thick enough for detection by X-ray analysis. The pressed compacts were analyzed by X-ray fluores­cence and the ratios of its constituent elements peaks were compared with prerun ratios. Figure 7 shows the change in the sulfur to molybdenum ratio in MoS2 after heating in vacuum. There was still sulfur remaining in the compact but the ratio of intensities changed from O. 45 to O. 18 indicating the sample consisted of Mo and MoS2. In the case

11

TABLE IT. - ANALYSES OF DISSOCIATION PRODUCTS CONDENSATE BY

:ompound

MoS2

MoTe2

MoSe2

ws2 WTe2

WSe2

X-RAY DIFFRACTION OR FLUORESCENCE

Deposit color

Yellow-brown

Black

Reddish brown

Yellow - brown Black

Reddish brown

A mount deposited

Very light

Heavy

Very light

Light Heavy

Very light

800

700

600

500

400

300

200

100 " al

2 -... s o E

Results of X-ray analysis

Not sufficient material for diffraction o r fluorescence

Diffraction indicates elemen- tal tellurium

Not sufficient material for diffraction o r fluorescence

Fluorescence detects sulfur Diffraction indicates elemental

Not sufficient material for tellurium

diffraction o r fluorescence

S Ka

Mqlol

2

(a) MoS2 before heating. Is K ~ / I M ~ K ~ = 0.45. 0 u

H-

5, .,- v) .- $ 700

E 6 0 0 E f i 500

200 c 9 7

Dif f ract ion angle, 28

(b) MoS after heat ing t o 2300" F (1260" C) in the rma l stabil ity apparatus. I s K~~IM~K~ = 0.18.

Figure 7. - Ratio of intensi t ies of const i tuent element peaks in X-ray fluorescence patterns of MoS2 before and after heat ing in vacuum.

12 12

TABLE II. - ANALYSES OF DISSOCIATION PRODUCTS CONDENSATE BY

Compound

MoS2

MoTe2

MoSe2

WS2 WTe2

WSe2

u ~ S V>

C :::J 0 U

0--<-

,.:; ~ V> c:

.IE

.E:

X-RAY DIFFRACTION OR FLUORESCENCE

Deposit color Amount deposited Results of X-ray analysis

Yellow-brown Very light Not sufficient material for

diffraction or fluorescence

Black Heavy Diffraction indicates elemen-

tal tellurium

Reddish brown Very light Not sufficient material for diffraction or fluorescence

Yellow-brown Light Fluorescence detects sulfur

Black Heavy Diffraction indicates elemental

tellurium

Reddish brown Very light Not sufficient material for

diffraction or fluorescence

800 ~MOKa

700

600

500 SKa

400

(a) MoS2 before heating. IS Ka/IMoKa = 0.45.

700

600

500

400

300

200

~ ~ --

Diffraction angle, 28

(b) MoS2 after h~ating to 23000 F (12600 C) in thermal stability apparatus. IS Ka/IMo Ka - 0.18.

Figure 7. - Ratio of intensities of constituent element peaks in X-ray fluorescence patterns of MoS2 before and after heating in vacuum.

0 T-

u W v)

s - v)

300

200

100

(a) MoSe2 before heating. I S ~ K ~ J I M ~ K ~ = 1.5.

ii Diffract ion angle, 28

0

(b) MoSe2 after heat ing to 2200" F (1200" C) in the rma l stabi l i ty apparatus. Ise K ~ J = 0.

Figure 8. - Ratio of intensi t ies of const i tuent element peaks in X-ray fluorescence patterns of MoSe2 before and after heat ing in vacuum.

of MoSe2 (fig. 8), no selenium was detected by fluorescence after heating to 2200' F (1200' C) indicating that the decomposition reaction was complete.

Friction Results

The molybdenum and tungsten compounds were examined a s thin lubricant films at elevated temperatures in vacuum in order that limiting temperatures for these materi- als could be established under these conditions. Figures 9 and 10 present friction data for cobalt alloy disks lubricated with the six inorganic compounds at temperatures to 1650' F (900' C). A t each temperature, a newly coated disk was used. The disk was brought to test temperature and held there for 60 minutes. During the final 30 minutes, sliding was initiated and the representative friction recorded.

For all the compounds except me2, the friction below 1200' F (650' C) was low (0.04 to 0.08). The higher friction for the WTe2 (0.20 to 0.25) lubricated disks is probably related to the orthorhombic crystal structure of W e 2 as compared with the hexagonal structure for the other compounds (ref. 8).

13

\ \

1-

MoKa

300 ! 200

100

o~-----------------------------------

(a) MoSe2 before heating. ISe KI'>J IMo Ka = 1. 5. u w on

S on C 700 :::J o U

.....; 600 ~

:e; 500 c w

C 400

300

200

100

o~~~ ____ ~~~~~O~OO~OO~O~I]~~~~~~~ __ _ Diffraction angle, 29

(b) MoSe2 after heating to 22000 F (12000 C) in thermal stability

apparatus. ISe KI'>J IMo Ka = O.

Figure 8. - Ratio of intensities of constituent element peaks in X-ray fluorescence patterns of MoSe2 before and after heating in vacuum.

of MoSe2 (fig. 8), no selenium was detected by fluorescence after heating to 22000 F (12000 C) indicating that the decomposition reaction was complete.

Friction Results

The molybdenum and tungsten compounds were examined as thin lubricant films at elevated temperatures in vacuum in order that limiting temperatures for these materi­als could be established under these conditions. Figures 9 and 10 present friction data for cobalt alloy disks lubricated with the six inorganic compounds at temperatures to 16500 F (9000 C). At each temperature, a newly coated disk was used. The disk was brought to test temperature and held there for 60 minutes. During the final 30 minutes, sliding was initiated and the representative friction recorded.

For all the compounds except WTe2, the friction below 12000 F (6500 C) was low (0.04 to 0.08). The higher friction for the WTe2 (0.20 to 0.25) lubricated disks is probably related to the orthorhombic crystal structure of WTe2 as compared with the hexagonal structure for the other compounds (ref. 8).

13

I I 1111l111111111111ll1l~.1l111111111111111l111Il I

. 4 r

t n

0 Y I

!

I I I 1000 1200 1400 1600 1800

Temperature, "F

0 400 600 800

I I I J 300 400 500 600 700 800 900

Temperature, "C

(c ) MoTep

Figure 9. - Fr ict ion coefficient for MoS2, MoSe2, and MoTe2 f i l m lubr icat ion at var ious temperatures. Exposure time, 60 minutes in vacuum to

to r r ) ; load, 100 grams; speed, 2.0 cent imeters per second.

A t the higher temperatures, the measured friction of all the lubricant films in- creased. These increases are associated with loss of the lubricant films from the disk surface by evaporation, dissociation, and wear.

For M0S2 and WS2, the mass spectrometer data shown in figure 4 indicate that dissociation is probably not the cause for lubricant loss up to 1650' F (900' C). The lubricant loss is due primarily to thermal evaporation and wear. Even though the measured weight loss ra te of both disulfides at 1400' F (760' C) is <6xlO-' (gram) (centimetere2)(second-l), the actual area of lubricant exposure is much larger than the apparent area so the amount of lubricant lost was greater than calculation based on apparent area would predict. The WS2 films provide lubrication to a slightly higher temperature than the M0S2 fi lms and this may be due to the higher molecular weight

14

1 1 111111111111111111111 __ 11111111111111111111111

~ c-o

~ b c: Q)

'u ~ Q) 0 u

.4

o

.2

o~--~~--~----~----~----~--

(a) MoS2'

.4

.2

0

(b) MoSe2'

0 400 600 800 1000 1200 1400 1600 1800

Temperature, OF

J 300 400 500 600 700 800 900

Temperatu re, °C

(c) MoTe2'

Figure 9. - Friction coefficient for MoS2' MoSe2' and MoTe2 film lubrication at various temperatures. Exposure time, 60 minutes in vacuum (10-8 to 10-6 torr); load, 100 grams; speed, 2.0 centimeters per second.

At the higher temperatures, the measured friction of all the lubricant films in­creased. These increases are associated with loss of the lubricant films from the disk surface by evaporation, dissociation, and wear.

For MoS2 and WS2, the mass spectrometer data shown in figure 4 indicate that dissociation is probably not the cause for lubricant loss up to 16500 F (9000 C). The lubricant loss is due primarily to thermal evaporation and wear. Even though the measured weight loss rate of both disulfides at 14000 F (7600 C) is <6><10-9 (gram) (centimeter-2)(second-1), the actual area of lubricant exposure is much larger than the

apparent area so the amount of lubricant lost was greater than calculation based on apparent area would predict. The WS2 films provide lubrication to a slightly higher temperature than the MoS2 films and this may be due to the higher molecular weight

14

-= . 4 r

%-

I I, 1000 1200 1400 1600 1800

I 800

Temperature, "F $00 600

I I 1 - I I --u 700 800 900 300 400 500 600

Temperature, "C

( c l WTe2.

Figure 10. - Fr ict ion coefficient for WSz, WSe2, and WTe2 f i lm lubr icat ion at var ious temperatures. Exposure time, 60 minutes in vacuum to

torr) ; load, 100 grams; speed, 2.0 cent imeters per second.

(247.98) of WS2 than MoSa (160.07). Confirming these results, reference 15 reported failure of MoS2 f i lms in vacuum above 1470' F (800' C) which the author attributes to decomposition of the MoS2.

1400' F (760' C), the increase in friction is due to evaporation and wear with the possi- bility of some dissociation at the maximum temperatures (fig. 5). MoSe2 and WSe2 provided the best lubricant performance of the compounds tested.

occurred at 1300' F (700' C) and very likely below; thus the increase in friction for lubricant f i lms of the ditellurides is due to thermal dissociation. Some slight lubricant action is probably afforded by the mixture of the metal and the remaining metal telluride.

Figures 11, 12, and 13 present photomicrographs of the disk surface with lubricant 15

The diselenides both provide lubrication over the same temperature range. Above

From the data of figure 6, it was seen that thermal dissociation of the ditellurides

~ c· 0

~ .... '0 C '" g <;:; 0 u

.4

.2

6

.4

.2

0

.4 D...--.....r---

o o .2 0

~oo ! ! ! I

600 800 1000 1200 1400 1600 1800 Temperature, OF

I I I- I 300 400 500 600 700 800 900

Temperature, °c

(c) WTe2'

Figure 10 .. Friction coefficient for W52' W5et.' and WTe2 film lubrication at various temperatures. Exposure time, 60 minutes in vacuum (10-8 to 10-6 torr); load, 100 grams; speed, 2.0 centimeters per second.

(247.98) of WS2 than MoS2 (160.07). Confirming these results, reference 15 reported failure of MoS2 films in vacuum above 14700 F (8000 C) which the author attributes to decomposition of the MoS2.

The diselenides both provide lubrication over the same temperature range. Above 14000 F (7600 C), the increase in friction is due to evaporation and wear with the possi­bility of some dissociation at the maximum temperatures (fig. 5). MoSe2 and WSe2 provided the best lubricant performance of the compounds tested.

From the data of figure 6, it was seen that thermal dissociation of the ditellurides occurred at 13000 F (7000 C) and very likely below; thus the increase in friction for lubricant films of the ditellurides is due to thermal dissociation. Some slight lubricant action is probably afforded by the mixture of the metal and the remaining metal telluride.

Figures 11, 12, and 13 present photomicrographs of the disk surface with lubricant

15

16

(a) Prerun film.

(c) moo F (6200 Cl. (d) 13250 F (720 0 C).

(e) 15500 F (8400 C).

Figure 11. - Photomicrographs of disk surface after runs at various temperatures with rubbed WS2 film lubrication. Pressure,

10-8 to 10-6 torr.

16

(a) Prerun film.

(c) moo F (6200 Cl. (d) 13250 F (720 0 C).

(e) 15500 F (8400 C).

Figure 11. - Photomicrographs of disk surface after runs at various temperatures with rubbed WS2 film lubrication. Pressure,

10-8 to 10-6 torr.

(a) Prerun film. (b) 950" F (510" C).

(d) 12%" F (680" C).

(0 15%" F (840" C).

Figure 12. - Photomicrographs of disk surface after runs at various temperatures with rubbed WSe, film lubrication. Pressure, 10-8 to 10-6 torr.

17

(a) Preri.ln film. (bl 9500 F (510 0 C).

(cl 1150" F (6200 Cl. td) 12500 F (6800 C).

(e) 1400" F (760 C). (f) 15500 F (8400 C).

Figure 12. - Photomicrographs of disk surface after runs at various temperatures with rubbed WSe2 film rubrication. Pressure, 10-8 to 10-6 torr.

17

18

(al Prerun film. (bl 950 u F (51O U Cl.

(cl 10500 F (5600 Cl. (dl 13000 F (700 0 Cl .

(el 14500 F (7900 Cl. (I) 15500 F (840 0 Cl.

Figure 13. - Photomicrographs of disk surface after runs at various temperatures with rubbed WTe2 film lubrication. Pressure,

10-8 to 10-6 torr.

_._ --"---_._-----------------

18

(al Prerun film. (bl 950 u F (51O U Cl.

(cl 10500 F (5600 Cl. (dl 13000 F (700 0 Cl .

(el 14500 F (7900 Cl. (I) 15500 F (840 0 Cl.

Figure 13. - Photomicrographs of disk surface after runs at various temperatures with rubbed WTe2 film lubrication. Pressure,

10-8 to 10-6 torr.

_._ --"---_._-----------------

fi lms of the tungsten compounds before running and after various thermal-vacuum exposures. The lubricant appears white because of the illumination. The photomicro- graphs show very definite loss of the lubricant material from the disk surface, and there is good agreement between the observed lubricant loss and the friction-temperature curve of figure 10.

disulfides and the diselenides as lubricants with exposure t imes to 5 hours. The di- tellurides were not examined because of their low dissociation temperatures. The re- sults are given in figure 14. The runs were made at 1325' F (720' C) and the friction measured during a 5-minute interval each hour. A s expected, there is a general increase in friction for all the compounds. The longer a lubricant film was exposed, the greater the lubricant loss and consequently the higher the friction.

In order to observe the effect of exposure time on friction, runs were made with the

0 MoS2

I * - c- ._ o o I V

L (a) Sulfides. ._ -

.2

WSep

0 z 1 2 3 4 5 6

Exposure time, h r

I b l Selenides,

Figure 14. - Coefficient of f r i c t i on at 1325" F 1720' C of cobalt-base alloy w i th fou r lubr icants. Pressure, to 10- 7 ) . torr, load, 100 grams; s l id ing speed, 2.0 cent imeters per second.

SUMMARY OF RESULTS

Therma l S tab i l i t y Resu l ts

Thermal stability experiments conducted in vacuum with the disulfides, diselenides, and ditellurides of molybdenum and tungsten yielded the following results:

19

films of the tungsten compounds before running and after various thermal-vacuum exposures. The lubricant appears white because of the illumination. The photomicro­graphs show very definite loss of the lubricant material from the disk surface, and there is good agreement between the observed lubricant loss and the friction-temperature

curve of figure 10. In order to observe the effect of exposure time on friction, runs were made with the

disulfides and the diselenides as lubricants with exposure times to 5 hours. The di­tellurides were not examined because of their low dissociation temperatures. The re­sults are given in figure 14. The runs were made at 13250 F (7200 C) and the friction measured during a 5-minute interval each hour. As expected, there is a general increase in friction for all the compounds. The longer a lubricant film was exposed, the greater the lubricant loss and consequently the higher the friction .

. 4

.2

.Y-c- o ."' :Q .... 0 c .~

~ .4 Q; 0 u

.2

o

0 MoS2 0 WS2

0 MoSe2 0 WSe2

(a) Sulfides .

3 Exposure time, hr

(b) Selenides.

4 6

Figure 14. - Coefficient of friction at 1325 0 F (7200 C) of cobalt-base alloy with four lubricants. Pressure, 10-8 to 10-7 torr; load, 100 grams; sliding speed, 2.0 centimeters per second,

SUMMARY OF RESULTS

Thermal Stability Results

Thermal stability experiments conducted in vacuum with the disulfides, diselenides, and ditellurides of molybdenum and tungsten yielded the following results:

19

1. The disulfides are the most thermally stable of the inorganic compounds at elevated temperatures in vacuum. The ditellurides are the least thermally stable.

2. The temperatures at which thermal dissociation in vacuum was definitely ob- served by mass spectrometry were: M0S2, 2000' F (1090' C); WS2' 1900' F (1040' C); MoSe2, 1800' F (980' C); WSe2, 1700' F (930' C); MoTe2 and WTe2, 1300' F (700' C).

above 1700' F (930' C) than the rate of MoS2. 3 . The weight loss rate of WS2 increased more slowly with increasing temperature

Friction Resu Its

Friction experiments conducted in vacuum with thin unreplenished films of each of the inorganic lubricant materials on a cobalt base alloy yielded the following results:

1. Thin, unreplenished films of the inorganic material were limited as effective lubricants not only by thermal dissociation but also by thermal evaporation and wear. The effects of evaporation lowered the temperature range where these thin inorganic films provided good lubrication.

2 . MoS2 and WS2 did not provide effective lubrication above 1200' F (650' C) for MoS2 and 1350' F (730' C) for WS2 because of evaporation of the materials.

3. The diselenides provided the best high temperature performance to 1400' F (760' C). The diselenides were lost from the disk by evaporation, by wear, and at the maximum temperature, by dissociation.

relatively low (1300' F o r 700' C) thermal dissociation temperatures.

vacuum exposure a t a temperature where lubricant losses occurred resulted in an increase in friction with time because of a decrease in the amount of lubricant affording protection.

4. The ditellurides fai1,ed as high temperature vacuum lubricants because of their

5. For the sulfides and selenides of molybdenum and tungsten, prolonged thermal-

Lewis Research Center, National Aeronautics and Space Administration,

Cleveland, Ohio, January 3, 1969, 129-03-13-09-22.

REFERENCES

1. Winer, W. 0. : Molybdenum Disulfide as a Lubricant: A Review of the Fundamental Knowledge. Wear, vol. 10, 1967, pp. 422-252.

20

1. The disulfides are the most thermally stable of the inorganic compounds at

elevated temperatures in vacuum. The ditellurides are the least thermally stable. 2. The temperatures at which thermal dissociation in vacuum was definitely ob­

served by mass spectrometry were: MoS2 , 20000 F (10900 C); WS2

. 19000 F (10400 C);

MoSe2 , 18000 F (9800 C); WSe2, 17000 F (9300 C); MoTe2 and WTe2, 13000 F (7000 C).

3. The weight loss rate of WS2 increased more slowly with increasing temperature

above 17000 F (9300 C) than the rate of MoS2 .

Friction Results

Friction experiments conducted in vacuum with thin unreplenished films of each of

the inorganic lubricant materials on a cobalt base alloy yielded the following results:

1. Thin, unreplenished films of the inorganic material were limited as effective

lubricants not only by thermal dissociation but also by thermal evaporation and wear.

The effects of evaporation lowered the temperature range where these thin inorganic

films provided good lubrication.

2. MoS2 and WS2 did not provide effective lubrication above 12000 F (6500 C) for MoS2 and 13500 F (7300 C) for WS2 because of evaporation of the materials.

3. The diselenides provided the best high temperature performance to 14000 F

(7600 C). The diselenides were lost from the disk by evaporation, by wear. and at the

maximum temperature, by dissociation.

4. The ditellurides failed as high temperature vacuum lubricants because of their

relatively low (13000 F or 7000 C) thermal dissociation temperatures.

5. For the sulfides and selenides of molybdenum and tungsten, prolonged thermal­

vacuum exposure at a temperature where lubricant losses occurred resulted in an

increase in friction with time because of a decrease in the amount of lubricant affording

protection.

Lewis Research Center,

National Aeronautics and Space Administration,

Cleveland, Ohio, January 3, 1969,

129-03-13-09-22.

REFERENCES

1. Winer, W. 0.: Molybdenum Disulfide as a Lubricant: A Review of the Fundamental

Knowledge. Wear, vol. 10, 1967, pp. 422-252.

20

2. Magie, Peter M. : A Review of the Propert ies and Potentials of the New Heavy Metal Derivative Solid Lubricants. pp. 262-269.

Lubr. Eng. , vol. 22, no. 7, July 1966,

3. Boes, D. J . : Unique Solid Lubricating Materials for High Temperature-Air Applications. ASLE Trans . , vol. 10, no. 1, Jan. 1967, pp. 19-27.

4. Buckley, Donald H. ; Swikert, Max; and Johnson, Robert L. : Friction, Wear, and Evaporation Rates of Various Materials in Vacuum to Trans . , vol. 5, no. 1, Apr. 1962, pp. 8-23.

mm Hg. ASLE

5. Amateau, Maurice F. ; Krause, Horatio H. ; Orcutt, F. K. ; Glasser, W. A . ; and Allen, C. M. : Development of Lubricants for Positive-Displacement Power Sources from 1200' F to 1500' F. Battelle Memorial Inst. (ASD-TR-61-478, DDC NO. AD-265 587), July 14, 1961.

6. Bisson, Edmond E . ; and Anderson, William J. : Advanced Bearing Technology. NASA SP-38, 1965.

7. Lavik, Melvin T . ; Medved, Thomas M. ; and Moore, G. David: Oxidation Charac- terist ics of MoS2 and Other Solid Lubricants. ASLE Trans. , vol. 11, no. 1 Jan. 1968, pp. 44-55.

8. Boes, David J. : New Solid Lubricants: Their Preparations, Properties and Potential For Aerospace Applications. IEEE Trans. on Aerospace, vol. 2 , no. 2, Apr. 1964, pp. 457-466.

9. Scholz, W. G. ; Doane, D. V. ; and Timmons, G. A . : Molybdenum by Direct Thermal Dissociation of Molybdenum Disulfide. Trans. AIME, vol. 221, no. 2. Apr. 1961, pp. 356-364.

10. Touloukian, Y . S . , ed. : Thermophysical Properties of High Temperature Solid Materials. Vol. 5. Macmillan Co., 1967, p. 691.

11. Killeffer, Do H. ; and Linz, Arthur: Molybdenum Compounds, Their Chemistry and Technology. Interscience Publ. , 1952, pp. 54-56.

12. Cannon, Peter: Melting Point and Sublimation of Molybdenum Disulfide. Nature, V O ~ . 183, 1959, pp. 1612-1613.

13. Johnston, R. R. M. ; and Moore, A . J . W. : Water Adsorption on Molybdenum Disulfide Containing Surface Contaminants. J. Phys. Chem., vol. 68, no. 11, NOV. 1964, pp. 3399-3405.

14. Sliney, Harold E. : Decomposition Kinetics of Some Solid Lubricants Determined by Elevated-Temperature X-ray Diffraction Techniques. Presented at the USA F Aerospace Fluids and Lubricants Conference, San Antonio, Texas, Apr. 16-19, 1963.

21

2. Magie, Peter M.: A Review of the Properties and Potentials of the New Heavy Metal Derivative Solid Lubricants. Lubr. Eng., vol. 22, no. 7, July 1966,

pp. 262-269.

3. Boes, D. J.: Unique Solid Lubricating Materials for High Temperature-Air

Applications. ASLE Trans., vol. 10, no. 1, Jan. 1967, pp. 19-27.

4. Buckley, Donald H.; Swikert, Max; and Johnson, Robert L.: Friction, Wear,

and Evaporation Rates of Various Materials in Vacuum to 10-7 mm Hg. ASLE

Trans., vol. 5, no. 1, Apr. 1962, pp. 8-23.

5. Amateau, Maurice F.; Krause, Horatio H.; Orcutt, F. K.; Glasser, W. A.; and

Allen, C. M.: Development of Lubricants for Positive-Displacement Power

Sources from 12000 F to 15000 F. Battelle Memorial Inst. (ASD-TR-61-478, DDC

No. AD-265 587), July 14, 1961.

6. Bisson, Edmond E.; and Anderson, William J.: Advanced Bearing Technology.

NASA SP-38, 1965.

7. Lavik, Melvin T.; Medved, Thomas M.; and Moore, G. David: Oxidation Charac­

teristics of MoS2 and Other Solid Lubricants. ASLE Trans., vol. 11, no. 1.

Jan. 1968, pp. 44-55.

8. Boes, David J.: New Solid Lubricants: Their Preparations, Properties and

Potential For Aerospace Applications. IEEE Trans. on Aerospace, vol. 2, no. 2,

Apr. 1964, pp. 457 -466.

9. Scholz, W. G.; Doane, D. V.; and Timmons, G. A.: Molybdenum by Direct

Thermal Dissociation of Molybdenum Disulfide. Trans. AIME, vol. 221, no. 2.

Apr. 1961, pp. 356-364.

10. Touloukian, Y. S., ed.: Thermophysical Properties of High Temperature Solid

Materials. Vol. 5. Macmillan Co., 1967, p. 691.

11. Killeffer, D. H.; and Linz, Arthur: Molybdenum Compounds, Their Chemistry

and Technology. Interscience Publ., 1952, pp. 54-56.

12. Cannon, Peter: Melting Point and Sublimation of Molybdenum Disulfide. Nature,

vol. 183, 1959, pp. 1612-1613.

13. Johnston, R. R. M.; and Moore, A. J. W.: Water Adsorption on Molybdenum

Disulfide Containing Surface Contaminants. J. Phys. Chem., vol. 68, no. 11,

Nov. 1964, pp. 3399-3405.

14. Sliney, Harold E.: Decomposition Kinetics of Some Solid Lubricants Determined by

Elevated-Temperature X-ray Diffraction Techniques. Presented at the USAF

Aerospace Fluids and Lubricants Conference, San Antonio, Texas, Apr. 16-19,

1963.

21

15. Rowe, G . W. : Vapour Lubrication and the Friction of Clean Surfaces. Inst. Mech. Eng. Conference on Lubrication and Wear, London, Oct. 1-3 , 1957, pp. 333-338.

22 NASA-Langley, 1969 - 15 Em4716

15. Rowe, G. W.: Vapour Lubrication and the Friction of Clean Surfaces. lnst. Mech.

Eng. Conference on Lubrication and Wear, London, Oct. 1-3, 1957, pp. 333-338.

22 NASA-LangleY,1969 - 15 E-4716

POSTAGE A N D FEES PAID NATIONAL AERONAUTICS AND SPACE A.DMINISTRATION

WASHINGTON, D. C. 20546

OFFICIAL BUSINESS

NATIONAL AERONAUTICS A N D SPACE ADMINISTRATION

FIRST CLASS MAIL

POSTMASTER: If L’ndeliverable (Section 158 * Postal Manual) Do Not Return

“The aeronnnticnl and space activities of the United Stntes shall be condncted so as t o copztribute . . . t o the expalasion of hzinian Knowl- edge of phenomena in. the ntmosphere and space. T h e Adni in is t r~ t ion shall provide for the widest prncticable and appropriate dissemination of information concerning its actiiGties and the res& thereof.’’

-NATIONAL AERONAUTICS AND SPACE ACT OF 1958

NASA SCIENTIFIC AND TECHNICAL PUBLICATIONS

TECHNICAL REPORTS: Scientific and

complete, and a lasting contribution to existing knowledge.

, technical information considered important,

TECHNICAL NOTES: Information less broad in scope but nevertheless of ihportance as a contribution to existing knowledge.

TECHNICAL MEMORANDUMS : Information receiving limited distribution because of preliminary data, security classifica- tion, or other reasons.

TECHNICAL TRANSLATIONS: Information published in-a foreign language considered to merit NASA distribution in English.

SPECIAL PUBLICATIONS: Information derived from or of value to NASA activities. Publications include conference proceedings, monographs, data compilations, handbooks, sourcebooks, and special bibliographies.

TECHNOLOGY UTILIZATION PUBLICATIONS: Information on technology used by NASA that may be of particular interest in commercial and other non-aerospace

TEchnology Utilization R~~~~~~ and N ~ ~ ~ ~ , and Technology Surveys.

- ‘ ’- CONTRACTOR REPORTS: Scientific and applications. Publications include Tech Briefs, technical information generated under a NASA contract or grant and considered an important contribution to existing knowledge.

Details on the availability of these publications may be obtained from:

SCIENTIFIC AND TECHNICAL INFORMATION DIVISION

NATlONA L AERONAUTICS AND SPACE ADMINISTRATION Washington, D.C. 20546

NATIONAL AERONAUTICS AND SPACE ADMINISTRATIqN

WASHINGTON, D. C. 20546

OFFICIAL BUSINESS FIRST CLASS MAIL

POSTAGE AND FEES PAID NATIONAL AERONAUTICS AND

SPACE ADMINISTRATION

POSTMASTER: If 11 ndeli verable ( Section 15 8 ~ Postal Manual) Do Not Return

'The aeronautical al1d space actirities of the Ul1ited States shall be cOl1dllcted so as to co~ztribute ... to the expami01z of humal1 kl1owl­edge of phenomena il1- the atmosphere and space. The Admil1istration shall provide for the widest practicable and appropriate dissemil1ation of information concerning its actit'ities al1d the results thereof."

-NATIONAL AERONAUTICS AND SPACE ACT OF 1958

NASA SCIENTIFIC AND TECHNICAL PUBLICATIONS

TECHNICAL REPORTS: Scientific and technical information considered important, compler.e, and a lasting contribution to existing knoWledge.

TECHNICAL NOTES: Information less broad in scope but nevertheless of importance as a contribution to existing knowledge.

TECHNICAL MEMORANDUMS: Information receiving limited distribution because of preliminary data, security classifica­tion, or other reasons.

CONTRACTOR REPORTS: Scientific and technical information generated under a NASA contract or grant and considered an important contribution to existing knowledge.

TECHNICAL TRANSLATIONS: Information published in.a foreign language considered to merit NASA distribution in English.

SPECIAL PUBLICATIONS: Information derived from or of value to NASA activities. Publications include conference proceedings, monographs, data compilations, handbooks, sourcebooks, and special bibliographies.

TECHNOLOGY UTILIZATION PUBLICA TIONS: Information on technology lIsed by NASA that may be of particular interest in commercial and other non-aerospace applicatiom_ Publications include Tech Briefs, Technology Utilization Reports and Notes, and Technology Surveys.

Details on the availability of these publications may be obtained from:

SCIENTIFIC AND TECHNICAL INFORMATION DIVISION

NATIONAL AERONAUTICS AND SPACE ADMINISTRATION Washington, D.C. 20546