Highly Exfoliated Reduced Graphite Oxide ... - Wiley Online Library

www.MaterialsViews.com

www.advmatinterfaces.de

Yingru Li, Jun Zhao, Cheng Tang, Yongyong He, Yongfu Wang, Ji Chen, Junyuan Mao, Qinqin Zhou, Baoyuan Wang, Fei Wei, Jianbin Luo,* and Gaoquan Shi* specific surface area (SSA), good physical and chemical stability, and excellent mechanical and thermal properties.[11–20] It is expected to be an ecofriendly effective additive for antifriction and/or antiwear in lubricant systems without releasing SAPS (sulfated ash, phosphorous and sulfur; SAPS would cause air pollution such as acid rain and haze weather).[21–27] Actually, a variety of graphene-based lubricant oil additives have already been reported and achieved some achievements;[7,28–39] nevertheless, most of them showed unsatisfactory performances, requiring more in-depth studies. In this paper, we report the preparation of a highly exfoliated reduced graphite oxide (heRGO) with a laminar porous microstructure and a large SSA, and its application as an additive of a poly(α-olefin) lubricant oil (type-6, PAO6). This additive greatly decreased the coefficient of friction (COF) of PAO-6 for 44% and the wear depth by more than 90%, implying that nearly wear-free rubbing surfaces was successfully realized under boundary lubrication.

Highly exfoliated reduced graphite oxide (heRGO) is prepared by thermal reduction of crude graphite oxide (GO) powders with the assistance of KOH. By using the mixtures of KOH and GO with different weight ratios (n) as the starting materials, the resulting heRGO-n powders show different microstructures and lubrication properties. The optimized heRGO-4 powders have an average particle size around 1 μm, a “loose book”-like layered microstructure, and a large specific surface area of 829.5 m2 g−1. The unique microstructure and tiny particle sizes have made this graphene material to be an extraordinary high-performance additive of poly(α-olefin) lubricant oil (type 6, PAO-6). By adding 0.5% (by weight, wt%) heRGO-4, the coefficient of friction of PAO-6 is decreased by 44% and the wear scar depth is reduced by more than 90%. The lubrication effect of heRGO-4 is much superior to that of thermally reduced graphite oxide. Given its facile, cheap, and scalable preparation process, heRGO-4 has a great potential for industrial applications.

1. Introduction Graphitic materials have been extensively explored as additives of lubricant oils or greases to improve their tribological properties and to enhance the thermal conductivities of the lubrication systems.[1–10] Among them, graphene has attracted a great deal of attention in recent years because of its carbon only composition, unique atom-thick 2D structure, a large

2. Results and Discussions 2.1. Synthesis of heRGO-n

Dr. Y. R. Li, Dr. J. Chen, Q. Q. Zhou, B. Y. Wang, Prof. G. Q. Shi Department of Chemistry Tsinghua University Beijing 100084, P. R. China E-mail: [email protected] J. Zhao, Prof. Y. Y. He, J. Y. Mao, Prof. J. B. Luo State Key Laboratory of Tribology Tsinghua University Beijing 100084, P. R. China E-mail: [email protected] C. Tang, Prof. F. Wei Department of Chemical Engineering Tsinghua University Beijing 100084, P. R. China Y. F. Wang State Key Laboratory of Solid Lubrication Lanzhou Institute of Chemical Physics Lanzhou 730000, P. R. China

The achknowledgement section of this manuscript was revised November 18th. DOI: 10.1002/admi.201600700 Adv. Mater. Interfaces 2016, 3, 1600700

The procedures of preparing heRGO-n powders are schematically illustrated in Figure 1. Graphite oxide (GO) powders were synthesized from natural graphite by a modified Hummers’ method developed by us.[40,41] The oxidizing condition was more severe than traditional Hummers method to ensure that graphite powders were fully oxidized and intercalated by sulfuric acid. This GO sample did not need to be further exfoliated and purified by sonication, centrifugation and dialysis, greatly saving time, energy and cost. This crude GO powders were mixed with certain amounts of KOH and ethanol, and then milled to form a highly viscous black paste, followed by heating treatment in a tube furnace at 700 °C for 4 h under an Ar atomsphere. During this process, the GO powders were reduced and highly exfoliated to crude heRGO. After washing with deionized water to remove residual KOH and impurities inherited from GO powders, the resulting powders were dried and smashed into small and uniform tiny particles by ball milling. The milled products are nominated as heRGOn; where n is the weight ratio of KOH and GO powders. For

© 2016 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

wileyonlinelibrary.com

(1 of 8) 1600700

full paper

Highly Exfoliated Reduced Graphite Oxide Powders as Efficient Lubricant Oil Additives


full paper


Figure 1. Schematic illustration of the preparation of heRGO-n powders.

comparison, we also prepared heat treated GO without KOH activation under the same condition and it is named tRGO. These procedures did not involve using expensive agents or instruments, and KOH activation is a widely used method to produce porous carbon materials in industrial production; thus, the technique of heRGO-n preparation is readily to be scaled up into industrial levels.

2.2. Morphological and Structural Studies heRGO-4 was chosen as a representative example to study the structures and morphology of heRGO-n. heRGO-4 is a black powdery material (Figure 2a). Its tap density was measured to be 0.368 g cm−3, slightly lighter than that of graphite powders (1200 meshes, 0.457 g cm−3). However, the tap density of tRGO was measured to be only 0.062 g cm−3. heRGO-4 consists of irregular particles with diameters around 1 μm (Figure 2b). High-resolution Scanning electron micrograph (SEM) shows that their graphene sheets are partially exfoliated (Figure 2c), looser than that of graphite powders (Figure S1, Supporting Information). They do not have observable macrosized pores, while tRGO powders showed many large pores with diameters in the range of several hundred nanometers to several micrometers (Figure 2d). The nitrogen adsorption isotherm of heRGO-4 is a typical Type-I isotherm with a narrow hysteresis loop, indicating it is a microporous material (Figure 3a). The nitrogen adsorption isotherm of tRGO belongs to Type-II with a Type-B hysteresis loop, suggesting it has a meso-/macroporous 1600700 (2 of 8)


structure (Figure 3b). On the basis of BET method, the SSA of heRGO-4 was measured to be 829.5 m2 g−1, over twice that of tRGO (384.5 m2 g−1, this value is comparable to those of other thermally reduced GO materials.[42]) (Figure 3e). According to the theoretical SSA of graphene (2630 m2 g−1), the average layer numbers of heRGO-4 and tRGO were evaluated to be about 3 and 7, indicating that heRGO-4 is a highly exfoliated RGO. The pore sizes of heRGO-4 are mainly distributed around 2 nm with a small ratio of pores in the range of 5 to 10 nm (Figure 3c). The pore size distribution curve of tRGO has a sharp peak around 4 nm with a broad shoulder distribution in the range of 4 to over 100 nm (Figure 3d). In fact, the average pore size of heRGO-4 is 2.8 nm with a porosity of 0.585 cm3 g−1 and the average pore size of tRGO is 14.8 nm with a porosity of 1.428 cm3 g−1 (Figure 3f). Accordingly, heRGO-4 possesses a microporous layered structure, making it has a lower porosity than that of tRGO. The unique microstructure of heRGO-4 is mainly attributed to KOH activation: 1) at a high temperature, KOH reacted with the gaseous decomposition products of GO including CO2 and SOx to form solid salts, avoiding the dramatic volume expansion caused by releasing gases (usually occurred at temperatures below 300 °C); 2) KOH was decomposed into metallic K and OH⋅ radicals at high temperatures, and they etch carbon atoms to create pores. It should be pointed out that the etching effect of metallic K is mild because the temperature used here for heating treatment (700 °C) was lower than those (800−1000 °C) applied for preparing KOH activated carbonaceous materials.[43–45] Therefore, the etching of carbon atoms mainly occurred at the defective sites of graphene sheets, and the basal planes of graphene sheets were kept intact. As shown in the Transmission electron micrographs (TEMs) (Figure 2e,f), heRGO-4 sheets still have a lamellar structure. More importantly, there are not any nanoscale holes observed from graphene sheets. This morphology is different from those of other graphene materials activated by KOH at higher temperatures.[44,45] We further investigated the structures of other heRGO-n carefully to optimize their synthesis conditions. The particle sizes of tRGO and heRGO-n were mainly determined by ball milling treatment, and all of them are around 1 μm. However, the morphologies and pore structures of heRGO-n powders are remarkably influenced by the feeding ratio (n) of KOH/GO used for preparing these materials. The particles of heRGO-1 are partly like those of heRGO-4 and partly similar to those of tRGO with macro-pores (Figure S2a,b, Supporting Information). However, the morphologies of heRGO-2, heRGO-6, heRGO-8 powders are nearly identical to those of heRGO-4 (Figure S2c–h, Supporting Information). The yield of heRGO-n decreases significantly with the increase of n caused by etching effect (Figure 3e). Moreover, the pore structures of


Adv. Mater. Interfaces 2016, 3, 1600700



heRGO-n have subtle differences. heRGO-1 and heRGO-2 powders possess intermediate microstructures between tRGO and heRGO-4; while heRGO-4, heRGO-6, and heRGO-8 powders have nearly the same microstructures (Figure S3, Supporting Information, Figure 3e,f). The adsorption isotherm of heRGO-1 is similar to that of tRGO, while the area of hysteresis loop of the former sample is narrower (Figure S3a, Supporting Information). The SSA of heRGO-1 was measured to be only about 226.8 m2 g−1, even smaller than that of tRGO. A sharp peak around 4 nm is also observed from the pore size distribution curve of heRGO-1 (Figure S3b, Supporting Information), while its intensity is weaker than that of tRGO. The yield of heRGO-1 (the weight ratio of heRGO-1/GO) was measured to be 39.0%, and micropores with size smaller than 2 nm were formed by KOH activation (Figure S3b, Supporting Information). Its porosity decreased to 0.509 cm3 g−1 with an average pore size of only 8.966 nm. The yield of heRGO-2 was further decreased to 36.8% and its pore structure combines the features of tRGO and heRGO-4: the first half adsorption isotherm is similar to that of heRGO-4 with a smaller adsorbed N2 volume. This isotherm still has a steep increase of absorbed N2 volume at




(3 of 8) 1600700

full paper

Figure 2. a) Typical digital photograph of heRGO-n powders; b,c) SEM images of heRGO-4 powders with different magnifications; d) SEM images of tRGO powders; e,f) TEM images of heRGO-4 powders with different magnifications.

high P0/P (0.9–1.0), while the increment is smaller than that of tRGO and the area of hysteresis loop is further reduced (Figure S3c, Supporting Information). Thus, micropores, mesopores and macropores co-exist in its pore size distribution (Figure S3d, Supporting Information). Its SSA was measured to be 419.1 m2 g−1, only half that (829.5 m2 g−1) of heRGO-4. Its porosity (0.288 cm3 g−1) is the smallest among heRGO-n. However, its average pore size is 2.749 nm, nearly the same with that of heRGO-4, making both samples have similar morphologies according to their SEM images. On the other hand, when n was controlled to be larger than 4, the microstructures of heRGOn are nearly the same to that of heRGO-4, while their yields decreased dramatically (heRGO-6: 16.1%; heRGO-8: 13%). As can be seen in Figure 3a,b and Figure S3e–h (Supporting Information), heRGO-n (n = 4, 6, or 8) share the similar isotherms and pore size distributions with comparable SSAs (heRGO-4: 829.5 m2 g−1, heRGO-6: 911.7 m2 g−1, heRGO-8: 874.9 m2 g−1), porosities (heRGO-4: 0.585 cm3 g−1, heRGO-6: 0.653 cm3 g−1, heRGO-8: 0.625 cm3 g−1), and average pore sizes (heRGO-4: 2.819 nm, heRGO-6: 2.865 nm, heRGO-8: 2.859 nm). Therefore, to achieve a stable highly exfoliated RGO microstructure and a relatively high yield, heRGO-4 is the best choice among heRGO-n for practical applications. Although heRGO-n and tRGO possess different microstructures, their chemical structures are nearly identical. Each Raman spectrum of GO, tRGO, or heRGO-n has two remarkable bands at around 1340 and 1580 cm–1 and they are assigned to the D- and G-bands of carbon (Figures S4 and S5, Supporting Information).[46,47] The G-band is related to graphitic carbon and D-band is associated with the structural defects or partially disordered structures of graphitic domains. The intensity ratio of D- and G-bands for GO was calculated to be 0.92. However, each Raman spectrum of tRGO and heRGO-n shows a stronger D-band (the ID/IG intensity ratios for these samples are listed as follows; tRGO: 1.27, heRGO-1: 1.34, heRGO-2: 1.44, heRGO-4: 1.41; heRGO-6: 1.29, heRGO-8:1.35). These results indicate that the GO sheets were reduced to RGO and their conjugated structures were partially restored upon thermal treatment. The X-ray photoelectron spectral (XPS) examination indicates that the C/O atomic ratio of GO is about 1.94. In its C 1s XPS spectrum, the total area of the bands associated with oxygenated groups (CO: 286.6 eV, CO: 287.8 eV, and OCO: 289.0 eV) is much stronger than that of the band related to intact CC/CC carbons (284.6 eV).[40,48] These results confirm that our GO was highly oxidized (Figure S6, Supporting Information). After the thermal treatment, most of the oxygen atoms were removed. As shown in Figure S7 and Table S1 (Supporting Information), tRGO and heRGO-n have similar elementary compositions. Their atomic ratios of carbon and oxygen are close to 90% and 10%, suggesting that GO was reduced to tRGO or heRGOn with comparable carbon or oxygen contents. The C 1s XPS spectra of these graphene materials are also nearly identical; each spectrum shows a strong CC/CC band and three weak bands associated with oxygenated groups (Figure S8, Supporting Information). These results can be explained as follows. The TGA curve of GO (Figure S9, Supporting Information) indicates that its most oxygenated groups can be removed at temperatures lower than 200 °C. However, the etching of GO by KOH usually occur at temperatures above 500 °C. At these


full paper


Figure 3. N2 adsorption-desorption isotherms of a) heRGO-4 and b) tRGO; BJH desorption pore size distributions of c) heRGO-4 and d) tRGO; e) Comparison of the BET SSAs and yields between RGO and heRGO-n (n = 1, 2, 4, 6, 8); f) Comparison of the porosity and average pore sizes between tRGO and heRGO- n (n = 1, 2, 4, 6, 8).

high temperatures, the oxygen content of tRGO was very low. Moreover, KOH etching removed the carbon atoms together with its bonded oxygen atoms. Thus, KOH activation induced negligible change in the oxygen content of tRGO, making the chemical compositions of tRGO and heRGO-n are all similar.

2.3. Triological Tests In a metaphorical expression, a graphite particle is a “closepacked book,” while heRGO-4 powder is a “loose book” with a lot of narrow and small cracks between its “papers” according to its microstructure. The “loose book” structure reduces the Van der Waals’ force between RGO sheets, resulting in a weaker resistance for sliding them under a shear force. The “papers” with low cavities and defects can improve the friction-reduction and anti-wear capability.[49] Accordingly, heRGO-4 is expected to have great potential in lubricant applications. PAO-6 [poly(α-olefin), type 6] was chosen as the base oil and it was uniformly mixed with graphene-based powders to form a dispersion containing 0.5 wt% graphene additive by violent stirring for 3 h (Figure S10, Supporting Information). However, the dispersion of tRGO is unstable and the graphene additive started to separate from PAO-6 after aging for 2 d. In contrast, the dispersion of heRGO-4 remained stable in the same period, indicating that heRGO-4 possesses an excellent self-dispersion characteristic. Moreover, the dispersion containing 0.5 wt% heRGO-4 remained stable after keeping it undisturbed for 4 d (Figure S11, Supporting Information). However, a weak phase separation was observed after aging it for 6 d, while its phase separation was still less severe than that of PAO-6 containing 0.5% tRGO after aging for only 2 d. 1600700 (4 of 8)


A reciprocating sliding tribometer was used to study the tribological properties of different graphene-based additives in PAO-6 (Figure 4a) at room temperature with a contact pressure of 1 GPa and sliding velocity of 2.4 mm s−1 for the consideration of boundary lubrication. As can be seen in Figure 4b, for the base oil PAO-6, the curve of COF increases quickly at the beginning, then it is stabilized at a high value. The average COF of PAO-6 was measured to be 0.151 ± 0.014 (Figure 4c). The curve of PAO-6 containing 0.5 wt% tRGO or heRGO-4 is below that of pure PAO-6 and it is stable in the whole testing process, implying that either tRGO or heRGO-4 has antifriction effect. The average COF of PAO-6 containing 0.5 wt% tRGO was tested to be about 0.109 ± 0.025, reducing the COF by about 27.8% (Figure 4c), and this value is comparable to those of previously reported other graphene-based additives in similar lubricant systems[8,29,31,50,51] (Table S2, Supporting Information). It is amazing to us that the average COF of PAO-6 with 0.5 wt% heRGO-4 additive was found to be only 0.084 ± 0.005, reducing the COF of PAO-6 by about 44.3% (Figure 4c). Such a large COF reduction has never been reported before for graphene-based additives in similar lubricant systems. After the reciprocating sliding tribological test, we examined the wear scar (washed by sonication to remove the lubricant oil on the surface of wear scar) by Raman spectra. Characteristic peaks of graphene (D-band at ≈1340 cm−1 and G-band at ≈1580 cm−1) were detected from different sites of the wear scar (Figure S12, Supporting Information). The overall features of these Raman spectra are similar to those of heRGO-4. This observation suggested that heRGO-4 sheets were physically adsorbed on the surface of steel substrate, forming a protective boundary film to induce an extraordinary lubricantion effect. For practical applications, the content of additive should be carefully optimized





full paper Figure 4. a) Schematic illustration of tribological test, b) COF curves of PAO-6, PAO-6 containing 0.5 wt% heRGO-4 or 0.5 wt% tRGO additive in a 40 min tribological test, c) Comparison of average COFs and average WSDs of different lubrication systems after 40 min tribological tests; d, e) SEM images of wear scars on the steel plate after a 40 min tribological test of PAO-6; f) white light interfering image and g) wear scar depth distribution along the line AB in (f); h, i) SEM images wear scars on the steel plate after the tribological test of PAO-6 containing 0.5 wt% heRGO-4, j) white light interfering image of wear scars and k) wear scar depth distribution along line CD in (j).

and the additive should maintain a stable performance in a wide range of concentrations, especially in the case of occurring concentration fluctuation caused by compounding lubricant oil or a loss of additive under working condition. As can be seen in Figure S13 (Supporting Information), when the content of heRGO-4 increased from 0.0 to 0.5 wt%, COF decreased gradually and then it kept at a stable value in a wide range between 0.5 and 1.0 wt%. Accordingly, heRGO-4 is an excellent additive of lubricant oil, showing a stable performance in a wide concentration range. After the tribological tests for 40 min, we analyzed the wear scars on the steel plates to study the antiwear effects of different graphene-based additives. The raw steel plate has a flat surface with some processing traces (Figure S14, Supporting Information). In the case of using pure PAO-6, deep wear scars were observed from the steel plate (Figure 4d–g), and their average depth (WSD) was measured to be 461 ± 33 nm (Figure 4c). In comparison, as PAO-6 containing 0.5 wt% tRGO


or 0.5 wt% heRGO-4 was applied, quite shallow wear scars were induced. In these cases, even the processing traces of steel plates could be found among the wear scars (Figure S15, Supporting Information, Figure 4h–k). Actually, the average WSD for tRGO is 75.0 ± 21.8 nm and that for heRGO-4 is only 31.6 ± 12.6 nm (Figure 4c). Taking the morphology of the raw steel plate into consideration, the antiwear effect of heRGO-4 is excellent. Milled GO powders via the same process of treating heRGO-n and tRGO have also been used as a lubricant oil additive for comparison. As the precursor of heRGO-4, milled GO showed negligible antifriction or antiwear effects. Although the COF of PAO-6 containing 0.5 wt% milled GO was tested to be lower than that of pure PAO-6 at the very beginning, the COF of this system increased significantly after about 10 min and approached the value of pure PAO-6, reflecting the unstability of this lubrication system (Figure S16, Supporting Information). The average COF of this system was calculated to be as high as 0.134 ± 0.025, close to that of pure PAO-6. As a result,



(5 of 8) 1600700

full paper



deep wear scars with an average WSD of 386 ± 11 nm were also observed from the steel plate (Figure 4c, Figure S17, Supporting Information). Actually, GO is thermally unstable; it decomposes at temperatures higher than 150 °C to release a large amount of gases, leading the lubricant system to be unstable. Comparing with tRGO and heRGO-4, milled GO is a much worse lubricant oil additive (Figure S18, Supporting Information). On the basis of the tribological analysis described above, it is reasonable to conclude that heRGO-4 is an excellent lubricant oil additive. Given the facile and economical process of preparing heRGO-4 and its eco-friend characteristics, we believe it has a great potential for practical applications in mechanical industry. The ultrahigh performances of heRGO-4 can be attributed to its unique microstructure: (1) heRGO-n have similar chemical structures, while they have different pore structures. As can be seen in Figure S19, all the heRGO-n show some antifriction effects compared with pure PAO-6. The COF-curves of heRGO-1 and heRGO-2 are above that of heRGO-4, indicating the former two additives have weaker antifriction effects. The performances of heRGO-6 and heRGO-8 are nearly the same to that of heRGO-4. The average COFs of PAO-6 with 0.5 wt% heRGO-1, heRGO-2, heRGO-6, and heRGO-8 were measured to be 0.128 ± 0.018, 0.098 ± 0.008, 0.088 ± 0.004, and 0.083 ± 0.004, respectively (Figure 5a). heRGO-1 did not show any antiwear effect: the wear scars (average WSD = 483 ± 76 nm) are even deeper than those induced by pure PAO-6 (Figure 5a,

Figure 5. a) Comparison of average COFs and average WSDs for PAO-6 containing 0.5 wt% heRGO-n (n = 1, 2, 4, 6, 8); b) comparison of average COFs and average WSD for PAO-6 with 0.5 wt% heRGO-4, unmilled heRGO-4, CVD-G, commercial graphene 1, and commercial graphene 2.

1600700 (6 of 8)


Figure S20, Supporting Information). For heRGO-2, the depths of wear scars were remarkably reduced, showing an average WSD of 42.8 ± 2.5 nm (Figure 5a, Figure S21, Supporting Information). As expected, the wear scars for the lubricant oil with 0.5 wt% heRGO-6 or heRGO-8 are similar to that induced by the oil containing 0.5 wt% heRGO-4 (Figure 5a, Figures S22 and S23, Supporting Information); the average WSDs = 32.3 ± 6.8 nm or 27.3 ± 2.5 nm, reducing the average WSD induced by pure PAO-6 for more than 90%. Therefore, taking tRGO into consideration, we believe that the layered microporous structures of heRGO-n (n = 4, 6, 8) provided them with excellent antifriction and antiwear effects. A chaotic pore structure (e.g., heRGO-1) showed a negative effect on the antifriction and antiwear properties of graphene based additives. (2) Under boundary lubrication regimes, the asperities of worn surfaces directly contact with each other. Thus, to protect the sliding surface, the particles of additives should be able to enter the friction interfaces. Therefore, if the additives are large, they could not enter the boundary lubrication regimes, providing negligible antifriction and antiwear effects. We chose the unmilled heRGO-4 as a control sample. The sizes of unmilled heRGO-4 particles are mainly several to tens micrometers, much larger than those of milled heRGO-4 powders (Figure S24, Supporting Information). The antifriction performances of PAO-6 with (average COF = 0.163 ± 0.009) or without 0.5 wt% unmilled heRGO-4 were tested to be comparable (Figure 5b, Figure S25, Supporting Information). Moreover, deep wear scars with an average WSD of 630 ± 43 nm were observed from the steel plate (Figure 5b, Figure S26, Supporting Information), much deeper than those induced by pure PAO-6. (3) The morphology of graphene particles also has a strong effect on their lubrication performances. A powdery graphene material prepared by chemical vapor deposition (CVD-G) using porous MgO sheets as the template and substrate has also be used as an additive of PAO-6 lubricant oil.[52] As can be seen in Figure S27 (Supporting Information), CVD-G particles have wrinkle graphene sheets; obviously different from those of heRGO-4 (flat sheets). Furthermore, their chemical structures are also different from that of heRGO-4: a strong and sharp G-band was observed from its Raman spectrum, reflecting it possesses more graphitic regions and fewer structural defects than those of heRGO-4 (Figure S28, Supporting Information). Thus, it is reasonable to conclude that CVD-G sheets have a higher structural integrity. However, as shown in Figure S29 (Supporting Information), the lubrication effect of PAO-6 with 0.5 wt% CVD-G was even worse than that of pure PAO-6, and the average COF was measured to be 0.165 ± 0.019 (Figure 5b). The wear scars on the steel plate were deep with an average WSD was 505 ± 7 nm (Figure S30, Supporting Information, Figure 5b). Therefore, the particles with flat graphene sheets are superior to the counterpart with wrinkle sheets (e.g., CVD-G) for the applications as additives of lubrication oils. On the basis of the observations described above, the lubrication mechanism of graphene-based additives is proposed as follows. First, graphene additive with small particle sizes (e.g., heRGO-4) can form a stable dispersion facilitates the permeation of the particles to the friction interfaces and work synergistically with the base oil during friction. Second, the graphene-based additive can form a protective boundary film





3. Conclusions A powdery graphene material with particle sizes around 1 μm and a porous lamellar microstructure (heRGO-4) has been successfully prepared by thermal reduction of crude GO powders with the assistance of KOH. This method is facile, cheap and readily to be scaled up into industrial levels. More importantly, heRGO-4 exhibited excellent lubrication and antiwear properties. This additive can reduce the COF of PAO-6 by 44% and decrease the average WSD by more than 90%, realizing the formation of nearly wear-free rubbing surfaces. The small particle sizes and highly exfoliated lamellar structure with a large SSA of heRGO-4 made its lubrication performance much superior to those of tRGO, heRGO-2, CVD-G, and commercial graphene materials, showing a great potential for practical applications.

4. Experimental Section Synthesis of GO Powders: Graphite oxide powder was prepared by the oxidation of natural graphite powder (1200 mesh, Qingdao Huatai Lubricant Sealing S&T Co. Ltd., Qingdao, China) using a modified Hummers’ method developed by our group.[40,41] In detail, graphite powders (1200 mesh, 10.0 g) were added to concentrated sulfuric acid


(300 mL) under stirring with ice-water bath. Under vigorous agitation, potassium permanganate (45.0 g) was added slowly to ensure that the temperature of the suspension was lower than 5 °C. Successively, the reaction system was transferred to a 40 ± 5 °C water bath and stirred for about 1.5 h, forming a thick paste. Then, 500 mL of water was added, and the solution was stirred for 15 min at 90 ± 5 °C. Additional 1500 mL water was added and followed by a slow addition of 50 mL H2O2 (30%), turning the color of the solution from dark brown to yellow. The mixture was filtered and washed with 1:10 HCl aqueous solution (500 mL) to remove metal ions followed by washing with 1000 mL water to remove the acid. The resulting mixture was then dried by freeze drying. The dried graphite oxide powder was then smashed by ball milling (The diameter of the Al2O3 milling ball is 20 mm). Before adding into PAO-6, the GO powders were further smashed into smaller particles by ball milling (The diameter of the ZrO2 milling ball is 3 mm). Synthesis of heRGO-n Powders: 1 weight equiv. GO powder, n weight equiv. KOH (n = 1, 2, 4, 6, 8), and n/2 weight equiv. ethanol were mixed by ball milling (the diameter of the ZrO2 milling ball is 6 mm) for 3 h to form a black paste-like mixture. This mixture was put in a tube furnace lined with a nickel pipe, and then it was heated to 700 °C at a rate of 5 °C min−1 and kept at this temperature for 4 h under the protection of an Ar flow (300 mL min−1). Successively, the heating treated sample was washed by deionized water to remove residual KOH and other impurities. The wet product was dried at 80 °C in a drying oven to get heRGO-n. Finally, heRGO-n was smashed into small particles by ball milling (the diameter of the ZrO2 milling balls is 3 mm). Synthesis of tRGO Powders: The tRGO powders were prepared by directly thermal-treating GO powders in the absence of KOH and ethanol under the same conditions of synthesizing heRGO-n. The tRGO powders were also smashed by ball milling (the diameter of the ZrO2 milling ball is 3 mm). Synthesis of CVD-G Powders: CVD-G was prepared by the method reported in ref. [52]. Porous MgO sheets were used as templates for growing graphene. CVD-G was synthesized by CH4 cracking on the surfaces of MgO meshes at 900 °C in a vertical quartz reactor. Characterizations: Raman spectra were recorded on a microscopic Raman spectrometer (LabRAM HR Evolution, HORIBA Jobin Yvon, France) with a 532 nm laser at a power density of 4.7 mW. X-ray photoelectron spectra were taken out by using an ESCALAB 250XI photoelectron spectrometer (Thermo Fisher Scientific, USA). SEMs were performed on a field-emission scanning electron microscope (Sirion-200, Fei, Japan). TEMs were obtained by using a transmission electron microscope (Tecnai G2 F20 S-Twin, Fei, Japan); to prepare the sample, the heRGO powders were sonicated in isopropanol for 3 h. X-ray diffraction was carried out on a D8 Advance X-ray diffractometer with Cu Kα radiation (λ = 0.15418 nm, Bruker, Germany). N2 absorptiondesorption isotherm of the samples at liquid nitrogen temperature (78 K) adsorption was performed by the use of a gas absorption analyzer (Autosorb iQ Station 2, Quantachrome Instruments, USA). Tribological Tests: PAO-6 (poly (α-olefin)-6) was chosen as the base oil. heRGO-n, GO, tRGO, CVD-G, and commercial graphene materials were added into PAO-6 separately as the only additive by stirring for 3 h to form a homogenous lubricant oil. The tribological tests were carried out on a reciprocating tribotester (UMT-3 CETR, USA) with a ball-on-disk mode. The friction pairs including a ball (4 mm in diameter) and a disk (surface roughness = 15 nm) were both made of bearing steel (AISI 52100), cleaned by petroleum ether and acetone for 20 min before test, respectively. The tribological tests were performed under a load of 2 N (1 GPa) and a sliding velocity of 2.4 mm s−1 for boundary lubrication regime at room temperature. The morphology of the wear scars was measured by a white light interfering profilometer (MICROXAM-3D, America).

Supporting Information Supporting Information is available from the Wiley Online Library or from the author.



(7 of 8) 1600700

full paper

physically adsorbed on the rubbing surfaces to prevent the direct contact of the friction pairs, improving the tribological property of lubrication oil.[53] Furthermore, the “loose book” microstructure of graphene particles makes them to be able to adaptively form a sliding-oriented arrangement on the friction interface and trend to graphitization on the wear scars (Figure S31, Supporting Information). The high-resolution transmission electron (HRTEM) image of heRGO-4 after a tribological test shows a highly ordered lamellar structure. The formation of protective film with an ordered lamellar structure on the wear scars can improves the tribological property of the additive because the lamellar layers can be easily shear deformed by overcoming their weak interlayer Van der Waals forces. At last, for the consideration of practical applications, we also compared our heRGO-4 with two commercial graphene-based additives (commercial graphene 1 from The Sixth Element Materials Technology Co. Ltd., Changzhou, China and Commercial graphene 2, and XFNANO Co., Ltd., Nanjing, China). The commercial graphene materials were directly added into PAO-6 followed by violent stirring for 3 h to form stable dispersions. Regrettably, they showed unsatisfactory antifriction or antiwear effect. The average COF of PAO-6 with 0.5 wt% commercial graphene 1 was measured to be 0.116 ± 0.020 (Figure S32a, Supporting Information). Deep wear scars were observed from the steel plate (Figure S33, Supporting Information) and their average WSD was measured to be 386 ± 11 nm (Figure 5b). The PAO-6 with 0.5 wt% commercial graphene 2 exhibited similar performances (average COF = 0.130 ± 0.036; average WSD = 283 ± 40 nm (Figure 5b, Figure S34, Supporting Information). Accordingly, heRGO-4 is a much better additive of lubrication oil with superior properties in both antifriction and antiwear, reflecting its great practical importance.

full paper



Acknowledgements Y.L. and J.Z. contributed equally to this work. This work was supported by the National Basic Research Program of China (2014CB046404, 2013CB933001), MOST (2016YFA0200200), Natural Science Foundation of China (51433005, 51321092 and 51275263), and the key special project of nanotechnology of China (2016YFA0200200). Received: July 22, 2016 Revised: August 21, 2016 Published online: October 6, 2016

[1] P. J. Blau, R. L. Martin, Tribol. Int. 1994, 27, 413. [2] L. Joly-Pottuz, B. Vacher, N. Ohmae, J. M. Martin, T. Epicier, Tribol. Lett. 2008, 30, 69. [3] P. L. Perez, A. L. Boehman, J. M. Perez, Proc. Inst. Mech. Eng., Part J 2008, 222, 593. [4] C. G. Lee, Y. J. Hwang, Y. M. Choi, J. K. Lee, C. Choi, J. M. Oh, Int. J. Precis. Eng. Manuf. 2009, 10, 85. [5] P. Samyn, P. De Baets, G. Schoukens, Tribol. Lett. 2009, 36, 135. [6] B. K. Prasad, S. Rathod, M. S. Yadav, O. P. Modi, J. Mater. Eng. Perform. 2011, 20, 445. [7] G. Y. Bai, J. Q. Wang, Z. G. Yang, H. G. Wang, Z. F. Wang, S. R. Yang, RSC Adv. 2014, 4, 47096. [8] B. Schluter, R. Mulhaupt, A. Kailer, Tribol. Lett. 2014, 53, 353. [9] K. K. Mistry, V. G. Pol, M. M. Thackeray, J. G. Wen, D. J. Miller, A. Erdemir, Tribol. Trans. 2015, 58, 474. [10] Z. C. Zhang, Z. B. Cai, J. F. Peng, M. H. Zhu, J. Phys. D: Appl. Phys. 2016, 49, 045304. [11] K. S. Novoselov, A. K. Geim, S. Morozov, D. Jiang, Y. Zhang, S. a. Dubonos, I. Grigorieva, A. Firsov, Science 2004, 306, 666. [12] K. S. Novoselov, D. Jiang, F. Schedin, T. J. Booth, V. V. Khotkevich, S. V. Morozov, A. K. Geim, Proc. Natl. Acad. Sci. USA 2005, 102, 10451. [13] A. K. Geim, K. S. Novoselov, Nat. Mater. 2007, 6, 183. [14] A. A. Balandin, S. Ghosh, W. Bao, I. Calizo, D. Teweldebrhan, F. Miao, C. N. Lau, Nano Lett. 2008, 8, 902. [15] C. Lee, X. Wei, J. W. Kysar, J. Hone, Science 2008, 321, 385. [16] S. V. Morozov, K. S. Novoselov, M. I. Katsnelson, F. Schedin, D. C. Elias, J. A. Jaszczak, A. K. Geim, Phys. Rev. Lett. 2008, 100, 016602. [17] R. R. Nair, P. Blake, A. N. Grigorenko, K. S. Novoselov, T. J. Booth, T. Stauber, N. M. Peres, A. K. Geim, Science 2008, 320, 1308. [18] M. J. Allen, V. C. Tung, R. B. Kaner, Chem. Rev. 2009, 110, 132. [19] Y. Zhu, S. Murali, M. D. Stoller, K. J. Ganesh, W. Cai, P. J. Ferreira, A. Pirkle, R. M. Wallace, K. A. Cychosz, M. Thommes, D. Su, E. A. Stach, R. S. Ruoff, Science 2011, 332, 1537. [20] K. S. Novoselov, V. Fal, L. Colombo, P. Gellert, M. Schwab, K. Kim, Nature 2012, 490, 192. [21] X. F. Feng, S. Kwon, J. Y. Park, M. Salmeron, ACS Nano 2013, 7, 1718.

1600700 (8 of 8)


[22] T. Filleter, J. L. McChesney, A. Bostwick, E. Rotenberg, K. V. Emtsev, T. Seyller, K. Horn, R. Bennewitz, Phys. Rev. Lett. 2009, 102, 086102. [23] L. Xu, T. B. Ma, Y. Z. Hu, H. Wang, Carbon 2012, 50, 1025. [24] K. S. Kim, H. J. Lee, C. Lee, S. K. Lee, H. Jang, J. H. Ahn, J. H. Kim, H. J. Lee, ACS Nano 2011, 5, 5107. [25] H. Spikes, Lubr. Sci. 2008, 20, 103. [26] R. B. Rastogi, J. L. Maurya, V. Jaiswal, Tribol. Trans. 2013, 56, 592. [27] S. Bagi, R. Bowker, R. Andrew, SAE Int. J. Fuels Lubr. 2016, 9, 239. [28] J. Lin, L. Wang, G. Chen, Tribol. Lett. 2011, 41, 209. [29] J. S. Lin, L. W. Wang, G. H. Chen, Tribol. Lett. 2011, 41, 209. [30] Y. He, Y. Liu, T. Wu, J. Ma, X. Wang, Q. Gong, W. Kong, F. Xing, Y. Liu, J. Gao, J. Hazard Mater. 2013, 260, 796. [31] D. Berman, A. Erdemir, A. V. Sumant, Mater. Today 2014, 17, 31. [32] H. Li, L. Chen, Y. Zhang, X. Ji, S. Chen, H. Song, C. Li, H. Tang, Cryst. Res. Technol. 2014, 49, 204. [33] L. Liu, Y. W. Chen, P. Huang, RSC Adv. 2015, 5, 90525. [34] L. Zhang, Y. He, S. W. Feng, L. Zhang, L. Zhang, Z. L. Jiao, Y. Q. Zhan, Y. J. Wang, Ceram. Int. 2016, 42, 6178. [35] T. Filleter, J. L. McChesney, A. Bostwick, E. Rotenberg, K. V. Emtsev, T. Seyller, K. Horn, R. Bennewitz, Phys. Rev. Lett. 2009, 102, 086102. [36] K. S. Kim, H. J. Lee, C. Lee, S. K. Lee, H. Jang, J. H. Ahn, J. H. Kim, H. J. Lee, ACS Nano 2011, 5, 5107. [37] H. J. Song, N. Li, Appl. Phys. A: Mater. Sci. Process. 2011, 105, 827. [38] S. Liu, J. Ou, Z. Li, S. Yang, J. Wang, Appl. Surf. Sci. 2012, 258, 2231. [39] E. J. Sandoz-Rosado, O. A. Tertuliano, E. J. Terrell, Carbon 2012, 50, 4078. [40] Y. R. Li, J. Chen, L. Huang, C. Li, G. Q. Shi, Adv. Electron. Mater. 2015, 1, 1500004. [41] J. Chen, Y. R. Li, L. Huang, C. Li, G. Q. Shi, Carbon 2015, 81, 826. [42] C. M. Chen, Q. Zhang, M. G. Yang, C. H. Huang, Y. G. Yang, M. Z. Wang, Carbon 2012, 50, 3572. [43] S. Murali, J. R. Potts, S. Stoller, J. Park, M. D. Stoner, L. L. Zhang, Y. W. Zhu, R. S. Ruoff, Carbon 2012, 50, 3482. [44] L. L. Zhang, X. Zhao, M. D. Stoller, Y. W. Zhu, H. X. Ji, S. Murali, Y. P. Wu, S. Perales, B. Clevenger, R. S. Ruoff, Nano Lett. 2012, 12, 1806. [45] Y. Zhu, S. Murali, M. D. Stoller, K. J. Ganesh, W. Cai, P. J. Ferreira, A. Pirkle, R. M. Wallace, K. A. Cychosz, M. Thommes, D. Su, E. A. Stach, R. S. Ruoff, Science 2011, 332, 1537. [46] J. Chen, Y. Zhang, M. Zhang, B. Yao, Y. Li, L. Huang, C. Li, G. Shi, Chem. Sci. 2016, 7, 1874. [47] M. Zhang, L. Huang, J. Chen, C. Li, G. Shi, Adv. Mater. 2014, 26, 7588. [48] Y. R. Li, J. Chen, L. Huang, C. Li, J. D. Hong, G. Q. Shi, Adv. Mater. 2014, 26, 4789. [49] Y. T. Peng, Z. G. Wang, K. Zout, Langmuir 2015, 31, 7782. [50] X. Dou, A. R. Koltonow, X. L. He, H. D. Jang, Q. Wang, Y. W. Chung, J. X. Huang, Proc. Natl. Acad. Sci. USA 2016, 113, 1528. [51] Z. B. Cai, L. Zhao, X. Zhang, W. Yue, M. H. Zhu, PloS One 2016, 11, e0152143. [52] G. Ning, Z. Fan, G. Wang, J. Gao, W. Qian, F. Wei, Chem. Commun. 2011, 47, 5976. [53] D. Berman, A. Erdemir, A. V. Sumant, Carbon 2013, 54, 454.

