Structural, Thermal, and Tribological Properties

of Intercalated Polyoxymethylene/Molybdenum
Disulfide Nanocomposites

J. Wang,1 K. H. Hu,2 Y. F. Xu,2 X. G. Hu2

1
School of Chemical Engineering, Hefei University of Technology, Hefei 230009, China
2
Institute of Tribology, Hefei University of Technology, Hefei 230009, China

Received 12 July 2007; accepted 18 March 2008

DOI 10.1002/app.28519
Published online 13 June 2008 in Wiley InterScience (www.interscience.wiley.com).

ABSTRACT: Intercalated polyoxymethylene (POM)/mo- tigated on an MQ-800 end-face tirbometer under dry fric-
lybdenum disulfide (MoS2) nanocomposites were prepared tion. The worn surfaces were observed by scanning electron
by in situ intercalation/polymerization. The structures of microscopy. The results show that POM/MoS2 presented
the composites were characterized by means of powder better friction reduction and wear resistance, especially
X-ray diffraction (XRD) and transmission electron micros- under high load. The friction mechanism of the nanocom-
copy. The XRD pattern showed that the polymer was posites is also discussed in association with X-ray photo-
inserted into the MoS2 galleries. The interlayer spacing of electron spectroscopy. VC 2008 Wiley Periodicals, Inc. J Appl

the intercalated phase increased from 6.15 to 11.18 Å. The Polym Sci 110: 91–96, 2008
thermal behavior of the composites was also investigated
through thermogravimetric analysis. The results show that Key words: inorganic materials; mechanical properties;
the heat resistance of the intercalated composites decreased nanocomposites; particle size distribution; thermal
slightly. The tribological behavior of POM/MoS2 was inves- properties

INTRODUCTION exfoliation–adsorption technique.5,6 The composites

can be obtained quite easily by treatment of the
There has considerable interest and research activity
product of the intercalation of lithium in MoS2 with
in the preparation and characterization of layered
an aqueous solution of the polymer because the
nanostructured composites containing polymers.1–4
aqueous polymer is compatible with the single-layer
Layered nanocomposites can improve mechanical
water suspension of MoS2.
and thermal properties, fire retardancy, electric con-
In this study, the polyoxymethylene (POM)/MoS2
ductivity, and so on. The intercalated technique pro-
nanocomposites were prepared by means of in situ
vides a good method for resolving the dispersion of
intercalation/polymerization. POM is one kind of
the material on a nanoscale range.
engineering plastic with self-lubrication properties.
Transition metal dihalogenides (MX2) have layered
Some studies have involved modified POM with dif-
crystal structures, which have attracted significant
ferent solid lubricant additives for self-lubrication
attention over recent years.5–9 Molybdenum disulfide
applications.12–14 As a solid lubricant, MoS2 is
(MoS2) has great technological importance as cata-
always blended with plastics directly. It is indeed
lysts (for the petrochemical hydrotreatment process),
hard to make the blend disperse uniformly for the
electrodes, solid-state lubricants, and materials with
sake of compatibility, and it would weaken the per-
unusual two-dimensional magnetic properties.10 As
formances of polymer composites. In our previous
it was recently understood, the use of single-layer
research,15 we confirmed that the mechanical and
MoS2 dispersions allowed an extension of the nature
tribological properties of POM were better when the
of guests.11 Some polymer matrix nanocomposites
nanosized particles were dispersed into the POM
with water solubility have been prepared by the
matrix uniformly. However, it is possible to obtain a
uniform blend on nanoscale via in situ intercalation/
Correspondence to: X. G. Hu (xghu@hfut.edu.cn). polymerization. It was reported that ultrasonication
Contract grant sponsor: National Natural Science could help the intercalation.16 In this experiment,
Foundation of China; contract grant number: 50475071. ultrasonication technology was also used for assis-
aaContract grant sponsor: Anhui Provincial Natural Sci- tance. The tribological performance was researched
ence Foundation; contract grant number: 070414152.
at high load values to provide some fundamental in-
Journal of Applied Polymer Science, Vol. 110, 91–96 (2008) formation for high-load applications, such as sliding
V
C 2008 Wiley Periodicals, Inc. bearing systems with higher loads.
92 WANG ET AL.

EXPERIMENTAL
Materials
Trioxane with chemical purity was purchased from
Shanghai Chemical Factory (Shanghai, China) and
was recrystallized in ethane dichloridize before use.
Dioxolane with chemical purity was from Acros
Orgaics Co., Ltd. Boron trifluoride etherate for use as
a catalyst, hexahydrobenzene for use as a solvent,
ethane dichloridize, and ammonia water were all
purchased from National Reagent Corp. (Shanghai,
China). They were further purified before use. MoS2
and n-BuLi were purchased from Anhui Metallurgy
Institute (Hefei, China) and Yangzhou Chemicals Co,
(Yangzhou, China) respectively.

Preparation of the intercalated Figure 1 Schematic illustration of the end-face friction

POM/MoS2 composites tribometer.
The preparation processing of restacked MoS2 was
performed as described in ref. 17. The cationic ring-
in Figure 1. The average surface roughness values of
opening copolymerization of the monomers trioxane
the steel and composite, determined from random
and dioxolane (5.0 wt %) proceeded at 658C in solu-
surfaces (isotropic roughness texture) with central
tion. A solution with two monomers was added to a
line average (CLA) surface roughness values, were
dry, three-necked bottle under a nitrogen atmos-
1.5 and 5.8 lm, respectively, after polishing. Their
phere, and then, the desired amount of restacked
measurements were done by means of a Taylor–Hob-
MoS2 (2.0 wt %) was added to the solution. The mix-
son model 6 profilometer (Taylor-Hobson Co., UK).
ture was stirred by a vortex and ultrasonicated for
The morphologies of the wear scar were observed
30 min. Finally, the initiator was introduced into the
with a JSM-6700 field emission scanning electron
flask. The reaction continued at 608C for 2 h and
microscope (Electron Co., Japan). The elements of the
was then treated in 5.0 wt % ammonia at 1408C for
counterpart steel surface were investigated on VG
5 h. The treated suspension was filtered, and the
model Escalab 250 X-ray photoelectron spectroscope
obtained residue was washed with deionized water
(Thermo Electron, USA).
and acetone several times. The obtained copolymers
were dried in vacuo. Pure POM copolymer was also
prepared as a comparative sample.
RESULTS AND DISCUSSION
Characterization of the POM/MoS2 nanocomposites Structure of the POM/MoS2 nanocomposites
The lattice spacing of the composites was monitored Figure 2 presents the XRD patterns of POM/MoS2
on a Philips X’Pert Pro Super diffractometer with and 2H-MoS2 (tri-prism configuration). Figure 2(a)
graphite-monochromatized Cu Ka radiation (Philips shows that the POM was crystalline. The feature dif-
Co., Holland) (k ¼ 1.54178 Å). A JEM-100SX trans- fraction peak (2y ¼ 23.068) was ascribed to the 100
mission electron microscope was used to observe the reflections of hexagonal POM.18 XRD data also dem-
morphology of the POM/MoS2 nanocomposites. A onstrated that POM was intercalated into MoS2. The
thermogravimetry/differential thermal analysis first two (001) reflections were almost covered com-
instrument (Shimadzu Co., Japan) was used to eval- pared with the strong diffraction peaks of POM. The
uate the thermal properties of the composite. Sam- two reflections were observed clearly from the mag-
ples of about 5 mg were heated from ambient nified drawing of the partial zone. In comparison
temperature to approximately 4008C at a rate of with the 002 reflection at 2y ¼ 14.48 (d-spacing ¼
108C/min under an argon atmosphere. 6.15 Å) of 2H-MoS2 [Fig. 2(b)], the peak at about 88
Friction and wear tests were conducted with an (d-spacing ¼ 11.18 Å) indicated that the spaces of
MQ-800 end-face tribometer (Jinan, China) with AISI the gallery were expanded, and the interlayer space
1045 steel sliding against the polymeric composites expansion was 5.03 Å with respect to pristine MoS2.
with a face-contacting method in air (relative humid- From the results of XRD, we deduced the interca-
ity ¼ 75%) at room temperature (208C) with a sliding lation mechanism of POM into MoS2. In this cationic
velocity of 0.8 m/s under dry friction. The schematic polymerization, when the nonpolar hexahydroben-
diagram for the MQ-800 end-face tribometer is shown zene was chosen for the solvent, the boron

Journal of Applied Polymer Science DOI 10.1002/app

INTERCALATED POM/MOS2 NANOCOMPOSITES 93

Figure 3 TEM image of the POM/MoS2 nanocomposites.

important for the tribological performance of the

POM/MoS2 nanocomposites. The layered structure
of MoS2 is the basis of its use as a solid lubricant.
The thickness of layered MoS2 in the POM matrix
was about 20 nm according to the transmission elec-
tron microscopy (TEM) image of the POM/MoS2
nanocomposites (Fig. 3). Obviously, the layers of
MoS2 became thinner compared with those of the
original MoS2, which indicated that the restacked
MoS2 had good dispersion in POM. So, we affirmed
that in situ intercalation/polymerization was a good
method for preventing the agglomeration of inor-
ganic nanoparticles.

Figure 2 XRD patterns of (a) POM/MoS2 and (b) 2H- Thermal analysis of the POM/MoS2
MoS2 (CPS ¼ counts per second). nanocomposites
The thermal stabilities of pure POM and the POM/
triflouride etherate acted as a catalyst, and water MoS2 composite powders (without any other addi-
acted as the cocatalyst. The partly dissociated reac- tives) were also studied by means of thermogra-
tion of boron trifluoride etherate in hexahydroben- vimetric analysis (TGA). Figure 4 illustrates the
zene was as follows: thermogram of pure POM and the POM/MoS2
nanocomposite powders. A difference of the interca-
BF3 Et2 O þ H2 O BF3 OH2 þ Et2 O lated composite was the loss of weight about 10% in

BF3 OH2 Hþ ðBF3 OHÞ

It was an equilibrium reaction; the Hþ from the reac-

tion could initiate the trioxane copolymerized with
dioxolane. When the single-layer MoS2 suspension
was restacked, some water molecules were com-
bined into the gallery of MoS2 simultaneously. These
co-intercalated water molecules participated in the
reaction and acted as the cocatalyst. Thus, the mono-
mers could insert into the galleries, and polymeriza-
tion could also happen here.
Figure 3 represents the microscopy of the POM/
MoS2 nanocomposites. As shown, MoS2 dispersed in
the polymer matrix still had a layered structure,
which was consistent with the XRD results men-
tioned previously. That is, the POM/MoS2 compo-
sites were intercalated compounds, which was Figure 4 TGA curves of POM and POM/MoS2.

Journal of Applied Polymer Science DOI 10.1002/app

94 WANG ET AL.

that the friction coefficients of the POM/MoS2 nano-

composites were always lower than those of pure
POM at different loads. That is, the POM/MoS2
nanocomposites had better friction reduction than
the pure POM. This was probably related to the
thinner MoS2 with its layered structure. The thinner
MoS2 affected the performance of the polymer ma-
trix on a nanoscale. POM is a viscoelastic material,
and its deformation under load is also viscoelastic.
The friction coefficients vary with the load according
to the following equation:20

l ¼ k  Nðn
1Þ

where l is the coefficient of friction, N is the load, k

Figure 5 Relationship between the friction coefficient and is a constant, and n is also a constant and varies
load. between 2 3 and 1. Thus, the coefficient of friction
decreased with increasing load. The POM/MoS2
the range 160–2208C. It was probably related to the nanocomposites exhibited better self-lubrication
degradation of the lower molecular weight polymer properties in this experiment, especially in case of
chain. During the cationic polymerization, chain ter- higher loads. Figure 6 represents the wear scar
mination could take place by a reaction with MoS2 depths of POM and the POM/MoS2 nanocompo-
acting as an impurity. However, the effect of interca- sites. It is clear that the wear scar depth of POM/
lation on the thermal stability of the polymer was MoS2 was smaller than that of POM, which indicates
not apparent compared with the results of that POM/MoS2 had better wear resistance. The
polyether/MoS2 nanocomposites.19 The difference MoS2 layer could absorb the POM molecular
between the temperature of maximum weight loss
rate of POM and the POM/MoS2 nanocomposites
was about 108C. The decompositions of both pure
POM and the intercalated composite occurred in a
relatively narrow range of temperatures.

Tribological properties of the

POM/MoS2 nanocomposites
Figure 5 represents the variation of friction coeffi-
cient with load within 30 min. These results show

Figure 6 Comparison of the wear depths of POM and Figure 7 SEM images of (a) pure POM and (b) the POM/
POM/MoS2. MoS2 nanocomposites.

Journal of Applied Polymer Science DOI 10.1002/app

INTERCALATED POM/MOS2 NANOCOMPOSITES 95

TABLE I was attributed to the deformation and adhesion of

XPS Results for the Wear Surfaces of the polymer. The resin melted and softened with the
the POM/MoS2 Nanocomposites
friction heat energy. However, the wear surface of
Name Peak binding energy Atom % POM/MoS2 was smoother and could be observed
C1s 284.57 44.02
only with abated scuffing, according to Figure 7(b).
S2p 169.1 0.2 During the friction period, a lubrication film on the
Mo3d 232.2 0.46 counterpart steel was seen on the macroscopic scale.
Fe2p 710.51 17.27 It was confirmed from the results of X-ray photoelec-
O1s 529.59 12.51 tron spectroscopy (XPS). Table I shows the XPS
O1s 531.21 25.49
Mo3d5 228.41 0.04
results of the counterpart steel surface, which rubbed
against the POM/MoS2 nanocomposites. The ele-
ments Fe, C, O, Mo, and S were detected on the
rubbed surface, and the contents of C and O were
segments, and this absorption could resist the plastic very high. As show in Figure 8, the peak at 531.2 eV
deformation and adhesion transfer of the POM resin. was well attributed to O1s of

C

O

, which was
Figure 7 shows the scanning electron microscopy different from the 533.1-eV peak of POM and was
(SEM) micrographs of the worn surfaces of POM induced by the tribological oxidization of POM. It
and POM/MoS2. As shown in Figure 7(a), the worn was indicated that the lubrication film was formed
surface of POM took on a wrinkled wavy morphol- on the counterpart surface by tribochemical reaction
ogy surface with deep furrows in some places. This rather than physical absorption. The XPS of Fe

Figure 8 XPS spectra of the wear surfaces of the POM/MoS2 nanocomposites: (a) O1s, (b) Fe2p, (c) S2p, and (d) Mo3d.

Journal of Applied Polymer Science DOI 10.1002/app

96 WANG ET AL.

showed that the binding energy was 710.5 eV. This References
was ascribed to Fe2p (electrons at 2P electronic orbit
1. Mirabal, N.; Aguirre, P.; Santa Ana, M. A.; Gonzalez, B. G.
of Fe) of Fe2O3. The two peaks at 232.2 and 228.4 eV Electrochim Acta 2003, 48, 2123.
were caused by Mo3d of MoO3 and MoS2, respec- 2. Fu, X.; Qutubuddin, S. Mater Lett 2000, 42, 12.
tively. Most of Mo4þ was changed into Mo6þ, accord- 3. Wang, T. M.; Shao, X.; Wang, Q. H.; Liu, W. M. Tribology
ing to the XPS data. The peak at 169.1 eV 2005, 25, 322.
corresponded to S2p (electron at 2s electronic orbit of 4. Bissessur, R.; White, W. Mater Chem Phys 2006, 99, 214.
5. Benavente, E.; Santa Ana, M. A.; Gonzalez, G.; Becker-Guedes,
S) of FeSO4. The previous results indicate that the tri- F.; Mello, N. C.; Panepucci, H. C.; Bonagamba, T. J.; Donoso, J.
bochemical reactions happened quite easily for the P. Electrochim Acta 2003, 48, 1997.
POM/MoS2 nanocomposites. The lubrication film 6. Gonzalez, G.; Santa Ana, M. A.; Benavente, E.; Donoso, J. P.;
was mainly composed of alkyl chain, Fe2O3, and Bonagamba, T. J.; Mello, N. C.; Panepucci, H. Solid State
MoO3. Also, it also had some small amounts of FeSO4 Ionics 1996, 85, 225.
7. Golub, A. S.; Zubavichus, Y. V.; Slovokhotov, Y. L.; Novikov,
and MoS2. The resulting lubrication film could pre- Y. N.; Danot, M. Solid State Ionics 2000, 128, 151.
vent continuous cutting and ploughing on the poly- 8. Mirabal, N.; Lavayen, V.; Benavente, E.; Santa Ana, M. A.;
mer surface by metal asperities. Gonzalez, G. Microelectron J 2004, 35, 37.
9. Zubavichus, Y. V.; Slovokhotov, Y. L.; Schilling, P. J.; Titts-
worth, R. C.; Golub, A. S.; Protzenko, G. A.; Novikov Y. N.
Inorgan Chim Acta 1998, 280, 211.
CONCLUSIONS 10. Benavente, E.; Santa Ana, M. A.; Mendizabal, F.; Gonzalez, G.
Coord Chem Rev 2002, 224, 87.
POM/MoS2 intercalated nanocomposites were syn-
11. Zubavichus, Y. V.; Golub, A. S.; Lenenko, N. D.; Novikov, Y.
thesized successfully via in situ intercalation/poly- N.; Slovokhotov, Y. L.; Danot, M. J Mol Catal A 2000, 158, 231.
merization. POM was inserted into the MoS2 12. Hu, X. G. Trib Lett 1998, 5, 313.
galleries, and the interlayer space expansion was 13. Wu, Y.; Hua, M. Y.; Long, C. G. Lubri Eng 2006, 6, 32.
5.03 Å. The MoS2 dispersed into the polymer matrix 14. Wang, H. T.; Liu, W. M.; Yang, S. R.; Xue, Q. J. Polym Mater
was still in a layer state. The heat resistance of the Sci Eng 1997, 13, 126.
15. Wang, J.; Hu, X. G.; Tian, M.; Stenger, R. Polym Plast Tech
intercalated composites decreased slightly. The Eng 2007, 46, 469.
POM/MoS2 nanocomposites exhibited better friction 16. Sukpirom, N.; Lerner, M. M. Mater Sci Eng A 2002, 333, 218.
reduction properties than pure POM, especially 17. Gee, M. A.; Frindt, R. F.; Joensen, P.; Morrison, S. R. Mater
under high load. The wear resistance of POM/MoS2 Res Bull 1986, 21, 543.
18. Yin, J. H.; Mo, Z. S. The Current Polymer Physics; Science
was also better than that of unfilled POM. The fric-
Press: Beijing, 2001.
tion mechanism was considered to be the formation 19. Mirabal, N.; Aguirre, P.; Santa Ana, M. A.; Benavente, E.;
of the lubrication film mainly containing alkyl chain, Gonzalez, G. Electrochim Acta 2003, 48, 2123.
Fe2O3, and MoO3. 20. Unal, H.; Mimaroglu, A. Mater Des 2003, 24, 183.

Journal of Applied Polymer Science DOI 10.1002/app