C60 Intercalated Graphite as Nanolubricants.

We synthesized the novel nanocomposite consisting of alternately stacked single graphene sheets and a C60 monolayer by using the graphite intercalation technique in which alkylamine molecules help intercalate large C60 molecules into the graphite. It is found that the intercalated C60 molecules can rotate in between single graphene sheets by using 13C NMR measurements. The grease with the nanocomposite materials provides a much better lubricating performance than that with other additives that have been well-known up to now. This result exhibits that a C60 monolayer intercalated between graphenes plays an important role in lubricating behavior.

1. Introduction

Nanocomposites composed of an organic polymer and inorganic layered host are a new type of composite that have been developed recently, and have unique properties. Since van der Waals forces dominantly acting between successive layers of graphite are relatively weak, it is possible for a wide range of atoms, molecules, and ions to intercalate between graphite layers, thus producing the graphite intercalation compounds (GICs) [1]. However, it is difficult to intercalate organic molecules or polymers directly into the interlayer of graphite through an ion exchange reaction to obtain the polymer/graphite nanocomposites, particularly large molecules. Here we show nanocomposite materials consisting of alternately stacked single graphene sheets and C60 monolayers, in which the C60 molecules can rotate in between single graphene sheets. We synthesized this novel nanocomposite by the graphite intercalation technique in which alkylamine molecules help intercalate large C60 molecules into the graphite. This provides a general preparation method for intercalating huge fullerene molecules into graphite, which will lead to promising materials with novel mechanical, physical, and electrical properties.

2. Experimental

First, graphite oxide (GO) was synthesized from graphite powder with an average size of 500 μm (Nippon Carbon Co., Ltd.) in accordance with the Hummers method [1] with some modifications, and the intercalation of octylamine into graphite (GO-OA) was performed as described previously [2,3]. Next, octylamine-intercalated graphite oxide (GO-OA) was added to fullerene solution (100 mg of C60 dissolved in 100 ml toluene), and after that, the toluene used was evaporated at room temperature, leaving behind in the products (GO-OA-C60). The C60 fullerene used in this experiment was purchased from Frontier Carbon Co., Ltd, in Japan. The products of GO-OA-C60 were treated with 0.1 N hydrochloric acid solution at room temperature for at least 30 minutes and dried in air at 80 °C overnight to remove the octylamine, resulting in the C60-intercalated graphite oxide (GO-C60). Finally, in order to remove C60 powders that are not intercalated into the graphite, and moreover, to remove octylamines that are intercalated into the graphite, GO-C60 was heated for at least 80 minutes at 600 °C under a high vacuum of 10−6 Torr, which results in the nanocomposite consisting alternately of a stacked single graphene sheet and a C60 monolayer. All specimens were analyzed by using X-ray diffraction (Rigaku RINT 2200/PC diffractometer: CuKα radiation at 40 kV and 30 mA), FT-IR spectroscopy (FTIR：JASCO 480 Plus FT-IR spectrometer: the samples in KBr pellets), NMR (our original 7.1 T spectrometer with a Tecmag Apollo spectrometer and a Doty SuperSonic MAS 7 mm probe head).

3. Results and Discussion

Figure 1a shows the X-ray diffraction (XRD) intensity of graphite oxide (GO) and alkylamine-intercalated graphite oxide (GO-amine) with different alkyl chain lengths (CnH2n+1NH2) (n = 3 to 8). The appearance of a peak in the GO sample of Figure 1a shows that the spacing between graphite oxide sheets is approximately 8 Å, which is identical to the published data [1]. It was found that the spacing between graphite oxide sheets in the case of the GO-amine (n = 3 to 8) increases with the increase in the length of the alkyl chain incorporated in the interlayer space of the GO. Figure 1b shows the XRD intensity of the GO-amine (n = 3 to 8) in C60 solution, which we call the C60-intercalated graphite oxide (GO-C60). It should be noted that there appear drastic changes in the XRD intensity between n = 5 and n = 6 in Figure 1b, which indicates that C60 molecules are intercalated in the interlayer space of the GO by the driving force of alkylamine situated between the graphene oxide sheets when the interlayer space is sufficiently larger than the C60 molecule. However, there exist many C60 powders which are not intercalated into the GO in these GO-C60 specimens, because the stronger XRD intensity peaks, (111), (220), (311), (222), (331), (420), (422) and (511), from C60 powders (JCPDS file No.44-0558) appear in the spectra of Figure 1b. In order to remove C60 powders that are not intercalated into the GO, and moreover, to expel the alkylamines that are intercalated into the graphite, the GO-C60 specimens shown in Figure 1b were heated for 80 minutes at 600 °C under a high vacuum of 10−6 Torr, as shown in Figure 2a.

Figure 1

XRD intensity from as-prepared samples. (a) XRD intensity from graphite oxide (GO) and alkylamine-intercalated graphite oxides (GO-amine) with different chain lengths (CnH2n+1NH2) (n = 3–8). The peaks of GO and GO-amine are indicated by the arrows in the figure; (b) XRD intensity from GO-amine with different chain lengths in C60 solution (n = 3–8).

Figure 2

XRD intensity from the GO-C60 specimens of Figure 1b after heating for 80 minutes at 600 °C under a high vacuum of 10−6 Torr.

It should be noted that n = 6, n = 7 and n = 8 have broad peaks of A and B (as shown in Figure 2) corresponding to d-spacings of 9 Å and 4.6 Å, respectively, in addition to a broad peak of d = 3.3 Å corresponding to the spacing of graphite layers. n = 3, n = 4 and n = 5 have only a single broad peak of d = 3.3 Å. Since the GO-amine reverts to the graphite layers when alkylamines leave the GO-amine host after heating at up to 100 °C, as shown in the XRD intensity of Figure 3, the graphene oxide layers in the GO-C60, which do not include C60 molecules, also revert to the graphite layers when alkylamines go out from it after heating. It is expected that the peaks of A and B (n = 6 to 8) in Figure 2 are due to the d-spacing between the graphenes intercalating the C60 monolayer and their stacking faults, respectively. However, since peak A is also widely distributed, the d-spacings between the graphenes intercalating the C60 monolayer are considered to be widely distributed.

Figure 3

XRD intensity from the alkylamine-intercalated graphite oxide (GO-amine) specimens after heating for 80 minutes at 100 ºC under a high vacuum of 10−6 Torr.

Figure 4 show the FT-IR spectra for the specimens of Figure 2. The FT-IR spectra from n = 6, n = 7 and n = 8 in Figure 2b exhibit the C60 intermolecular IR-active (F1u) modes [4], although those with the shorter alkylamine chains do not exhibit these modes. This result is consistent with the conclusion based on Figure 2 that C60 molecules can be intercalated into the GO with alkylamine chains longer than that of n = 5. However, one mode of 526 cm−1 among IR-active (F1u) modes only appears in these FT-IR spectra because the number of C60 molecules included in the specimens is rather small.

Figure 4

FT-IR spectra for the specimens of Figure 2. The C60 intermolecular IR-active (F1u) modes are indicated by the inset of each panel of the figure.

The rotational dynamics of C60 molecules between graphenes have been investigated by 13C NMR in the temperature range from room temperature to −80 °C. We prepared C60 materials 20–30% enriched in 13C in order to increase the 13C NMR signal. The present specimen was mixed with Na2SO4 in a weight ratio of 1:50 to avoid arcing in an NMR probe. The NMR experiments were performed at 75.4 MHz for 13C in an external field of 7.1 T by the pulse inversion recovery method. 13C NMR spectra were taken by Fourier transforming the signal following the π/2 pulse. The typical π/2 pulse width was 5.4 μs. It is well known that for C60 molecules in solid C60 at room temperature, large rotational motion averages out the chemical-shift anisotropy (CSA) and the 13C NMR spectra show motional narrowing of 2.5 ppm in width. In contrast, spectra broaden at low temperature and develop the CSA power pattern with a CSA tensor with the principle values δ11 = 213 ppm, δ22 = 182 ppm, and δ33 = 33 ppm [4]. Figure 5 shows the 13C NMR spectra for the specimens of Figure 2 at room temperature, where the 13C NMR spectrum at room temperature is the same as that at the temperature of −80 °C. Only one sharp line with a peak position of 144 ppm was observed, and its line shape is a Gaussian-like function with an FWHM value of 5 ppm. The positions are in good agreement with the average principle values for C60 molecules in solid C60, and the linewidth is about one-twentieth narrower than that of the powder pattern. These observations clearly demonstrate the lack of the polymerization [5] of C60 molecules in the present material. Furthermore, the observed Gaussian-like line shape means a motional narrowing and that C60 molecules rotate quasi-freely with a correlation time on the order of 10 ps. This correction time is similar to that of the same case [6]. This means that no strong bonding such as chemical bonding between the graphenes and C60 molecules is made, and C60 molecules easily rotate for outer force.

Figure 5

The 13C NMR spectra for the specimens shown in Figure 2 at room temperature. Only one sharp line with a peak position of 144 ppm [4] was observed.

Many lubricant agents contain solid particles, such as commercial colloidal dispersions of graphite, tungsten disulfide, molybdenum disulfide (MoS2), or polytetrafluoroethylene as additive to enhance the lubricating ability of the base oil or grease. Their lubricating properties are generally attributed to a layered structure on the molecular level with weak bonding between layers. Such layers are able to slide relative to each other with minimal applied force, thus giving them their good tribological properties. We added this nanocomposite to the commercially available grease and evaluated the performance of that as the additive of lubricant. The lubricating performance is tested using a four-ball tester with four balls. The machine including the four-ball tribosystem has been used to determine the lubricant properties. The four-ball test is the industrial standard test method for measuring the wear preventive characteristics of a lubricant. Placed in a bath of the test lubricant, three fixed steel balls are put into contact with a fourth ball in rotating contact at set conditions. Lubricant wear protection properties are measured by comparing the average wear scars on the three fixed balls. The smaller the average wear scar, the better the protection. The nanocomposite was added to the base grease and the lubricant wear protection properties of the grease were investigated. Similarly, the lubricating properties of the greases blended with widely used additives such as graphite and MoS2 were also studied. The lubricating tests were carried out under the lubrication of 2 g grease containing 1.0% additives. Those were evaluated by measuring the wear scar diameter (WSD) and wear volume loss of the test ball. The test was conducted at a rotating speed of 1200 rpm and a constant load of 441 N for 1 h. The results of the wear behaviors for some lubricating additives are shown in Figure 6. According to Figure 6, the WSD and wear volume loss of the test ball for the base grease were 0.83 mm and 11.7 × 10−3 mm3, respectively. The WSD and wear volume loss of MoS2 were 0.46 mm and 0.9 × 10−3 mm3, respectively. The WSD and wear volume loss of graphite were 0.68 mm and 2.9 × 10−3 mm3, respectively. It is found that the base grease with graphite did not represent better lubricating performance than that with MoS2, but it largely improved in tribological quality relative to the base grease. C60 itself is decidedly inferior in lubricating performance as compared with the generally used lubricating additives such as MoS2 and graphite. Here, it should be noted that the WSD and wear volume loss of the nanocomposite materials were 0.41 mm and 0.5 × 10−3 mm3, respectively. Thus it is found that the nanocomposite prepared shows a much better lubricating performance than that with other additives. This result suggests that a C60 monolayer between graphenes strongly influences lubricating performance. Further detailed investigations of the tribological behaviors and the lubrication mechanism of the nanocomposite are now in progress.

Figure 6

The wear scar diameter (WSD) and wear volume loss of the test ball for the grease with various kinds of additives.

4. Conclusions

At present, preliminary experiments indicate that it is possible to intercalate huge fullerene molecules such as C70 and La@C82, larger than a C60 molecule, into graphite. Up until now, we have studied C60 monolayers confined by graphite flakes [7] and C60 monolayers included among graphite [8,9,10], which, interestingly, exhibits ultralow friction, because C60 molecules act as molecular bearings. We investigated lubricating properties of nanocomposite materials in commercial grease. The grease with the nanocomposite materials provides a much better lubricating performance than that with other additives that have been well-known up to now. This result exhibits that a C60 monolayer intercalated between graphenes plays an important role for the lubricating behaviors.