Introduction

The study of lubricants has gained a special place among the sciences with the emergence of new debates such as optimizing the consumption and conservation of non-renewable resources, as well as meeting environmental requirements. In most cases the cost of repairs because of utilizing inappropriate lubricant are much higher than the costs of a more expensive but better lubricant.

To prepare nanolubricants, researchers have used different types of nanomaterials made of metallic, organic and inorganic materials [1,2,3,4,5,6]. The most important advantage of using nanomaterials in lubricants is their small size. In the nanometer range, complete feeding of the roller joint surface has been accomplished. Also, the synthesis of composite particles or the use of hybrid nanoparticles with various properties such as reduced friction, wear and corrosion is feasible. Nano-additives are somewhat temperature-sensitive compared to conventional additives and their friction reactions are very limited [7, 8]. The nanometer size of these materials also increase their contact surface area and is efficient at ambient temperature. The anti-wear mechanism of nano-additives is formed in two ways: the nanoparticles may melt and adhere to the friction surfaces or may form a protective layer by reacting to the surface [9,10,11]. The roller mechanism is the rolling effect of nanoparticles that can improve the movement of the surfaces connected to each other [12, 13].

Due to their high strength and durability, fullerene nanoparticles also retain their rotary-like properties under high loading conditions [14, 15]. Lee et al. [16] investigated the effect of the addition of fullerene (C60) nanoparticles on the lubricating properties of mineral oil. They measured the friction coefficient through pin-on-disk test. This test was performed under vertical loads and different volumetric concentrations of fullerene nanoparticles. Their results indicated that the presence of C60 nanoparticles caused a smaller coefficient of friction and less wear. Therefore, the lubrication properties of mineral oil have been improved by the addition of C60 nanoparticles. Besides, Ku et al. [17] studied the tribological properties of C60 nanoparticles on mineral oil. Their results revealed that a significant improvement in the lubrication properties of the oil containing these nanoparticles was obtained compared to base oil.

Ettefaghi et al. [18] evaluated various nanostructures including fullerenes in 20W50 base oil to measure thermal conductivity, viscosity, pour point and flash point with concentration of 0.1 mass%. They observed fullerene nanoparticles and carbon nanoballs had the best stability. Table 1 provides a summary of other researches done on the impact of nanoparticles on the tribological properties of oils.

Table 1 Summary of tribological study of oil-based nanofluid
Full size table

The rheological behavior of an engine oil is one of the essential indicators in determining the quality of the oil. According to the standard of engine oil, the changes of viscosity to temperature and having a certain amount of viscosity under different engine operating conditions are important. Therefore, various researchers have studied the rheological behavior of engine oil with the use of nanoparticles in recent years [28,29,30]. Hemmat Esfe et al. [31] investigated the nanofluid behavior by stabilizing magnesium oxide nanoparticles in oil at volume fraction of 0 to 2%. They reported non-Newtonian pseudoplastic behavior by measuring the nanofluid viscosity at the temperature range of 5–60 °C and in different shear rates. The highest increase in viscosity was reported by 18% in their study. Saeedinia et al. [32] studied the rheological properties of copper oxide nanoparticles with mass concentrations of 0.2 to 2%. Their results showed that as the concentration of nanoparticles increased, the thermal conductivity rose, in a way that at a concentration of 2%, the thermal conductivity enhanced by 6.2%. They also observed that the nanofluid under study had Newtonian behavior in all temperatures and concentrations. In addition, they observed that temperature had a great effect on the rheological behavior and nanofluid viscosity. Also, the specific heat capacity of the nanofluids was lower compared to the base oil and as the concentration of nanoparticles increased, this amount decreased.

Aberoumand et al. [33] also reported experimental data on thermal conductivity and viscosity of oil containing silver nanoparticles. They observed a 40% and 27% increase in thermal conductivity and viscosity, respectively. In their study, the conversion of Newtonian to non-Newtonian behavior was reported at 35 °C in all concentrations. Anoop et al. [34] suspended the silica nanoparticles with a diameter of 20 nm in mineral oil and investigated the rheological behavior of this nanofluid in two volume fractions of 1% and 2% at temperatures of 25 °C to 140 °C and in different shear rates. They reported a decrease in viscosity after the increase in temperature. Also, the non-Newtonian behavior of the nanofluid was observed. Beheshti et al. [35] studied the important parameters of engine oil by adding oxidized multi-walled carbon nanotubes to the oil at two mass concentrations of 0.01% and 0.001%. The nanofluid flash point increased by 4.6% at a concentration of 0.01%, which decreased in 0.001% compared to the base oil. In addition, their experimental results revealed that the nanofluidic behavior was similar to the base oil in these two concentrations. In another study, Afrand et al. [36] studied on SiO2-MWCNTs/SAE40 nanofluids. They discovered that by increasing the volumetric fraction at different temperatures, the dynamic viscosity will increase. In their experiment, the highest increase in viscosity was 37.4% which occurred at 1% volume fraction. They also evaluated the nanofluid behavior in all concentrations which was non-Newtonian. A review of some other researches done on nano-oils has shown that the behavior of nano-oil in all volume fractions is non-Newtonian and pseudoplastic [37,38,39,40,41,42].

Improving the thermal conductivity of engine oil by adding nanoparticles has also been a topic of great interest for recent researches. Ettefaghi et al. [43] studied the effect of multi-walled carbon nanotubes at three concentrations of 0.1%, 0.2% and 0.5% on some properties of SAE 20W50 engine oil. In their study, viscosity, pour point, flash point and thermal conductivity were investigated. Their results showed that by increasing the concentration of nanoparticles up to 0.2%, the flash point and pour point values enhanced by 13% and 3.3%, respectively. Thermal conductivity was also directly correlated with concentration. In another study, Rashidi et al. [44] added various carbon nanostructures including multi-walled carbon nanotubes, graphene nanosheets, carbon nanoball particles and fullerene nanoparticles to the SAE 20W50 engine oil. Their results showed that fullerenes nanoparticles and carbon nanoball had the highest stability rate in the base fluid, respectively. Also, the carbon nanoballs had the most positive effect on these properties with an increase in 9.3% and 18% in thermal conductivity and flash point at the concentration of 0.1%, respectively.

Previous researches have indicated that fullerene nanoparticles have a positive role in improving the tribological properties of engine oil. Silica nanoparticles are also an olephilic nanoparticle with a proper suspension in engine oil. Therefore, the goal of this study was to improve the thermal, rheological and tribological properties of 5w30 synthetic oil with the use of fullerene-silica hybrid nanoparticles at the ratio of 20% to 80%, respectively. Stability of nanofluid has an important role in the development of the use of nanofluids in industry. C60 and SiO2 nanoparticles in addition to improve the tribological and thermal properties of nanofluid, have proper stability within the base fluid. On the other hand, this type of nanofluid has not yet been investigated in any other research. Therefore, in order to study the rheological behavior, the dynamic viscosity has been measured and analyzed in different volume fractions at different temperatures. The wear rate was also measured in two volume fractions of 0.5 and 1%. Thermal conductivity coefficient of C60-SiO2/SAE 5W30 hybrid nanofluid has also been studied at different volume fractions and at ambient temperature. The results of this study can provide researchers with comprehensive information on the thermophysical and tribological properties of a hybrid nano-oil with the use of two olephilic and highly stable nanoparticles.

Preparation of nanofluids and test methods

In this section, the preparation method of nanofluid and measurement instruments is introduced. The fullerene (20%)-silicon dioxide (80%)/engine oil hybrid nanofluid was prepared in 5 volume fraction containing 0.1%, 0.25%, 0.5%, 0.75%, 1%. Experiment is done at temperatures of 5, 15, 25, 35, 45, 55 and 65 °C. The characteristics of used nanoparticles and engine oil are presented in Tables 2 and 3.

Table 2 The characteristics of nanoparticles
Full size table
Table 3 The specifications of the SAE 5W30 engine oil
Full size table

After adding nanoparticles to the base fluid, the suspension is stirred for 20 min using a digital magnetic stirrer in 1200 rpm. In the next step, the suspension is placed in an ultrasonic homogenizer for 30 min to break down the agglomeration of nanoparticles. The device was Topsonic brand with a power of 400 watt and a frequency of 20 kHz. The suspension of the nanoparticles in these oils did not require any surfactant because of using the olephilic nanoparticles. Accordingly, after the suspension process, the nanofluid behaved quite stable. Figure 1 shows the prepared nanofluids in different concentrations after 2 weeks. As can be seen, the nanofluids are completely stable and no sediment is seen in the suspensions.

Fig. 1
figure 1

The prepared nanofluid in different concentrations after 2 weeks

Full size image

After visual inspection of the sample, another way to check the stability of nanofluids is DLS method. Because the studied nanofluid had a stable behavior, only one sample was selected for DLS testing. The dynamic light scattering (DLS) was utilized to test the stability of nanofluid by using Horiba sz-100 nanoparticle analyzer. The test has been done at 90° scattering angle and 25 °C for nanofluid in 0.1 vol%. The result reveals that the average size of the sample was 37.9 and it shows that the nanoparticles were not clogged in the base fluid and its stability was adequate. Figure 2 represents the DLS result of the sample.

Fig. 2
figure 2

DLS analysis of the C60-SiO2/5W30 nanofluid in 0.1 vol%

Full size image

In order to ensure the purity and the grain size of nanoparticles, XRD analysis has been done and it is displayed in Fig. 3. By assessing the peak values of the XRD, the sizes of SiO2 and C60 nanoparticles were calculated 20 and 8 nm, respectively. There are several peaks in XRD for different angles and different intensities and each one corresponds to a particular plane of the sample. It should be noted that the angle of each peak depends on the distance between the plate and the intensity of the peaks related to the arrangement of the atoms in the plates. It is concluded that each material has its own unique X-ray pattern which is used as a fingerprint to identify it. The sharp peaks in the diagram display the crystallinity of the nanoparticles.

Fig. 3
figure 3

The XRD analysis of nanoparticles: a SiO2 nanoparticles, b C60 nanoparticles

Full size image

The thermal conductivity of the nanofluid was measured using a KD2 Pro thermal analyzer with KS-1 sensor manufactured by Decagon Devices Inc. The maximum deviation of this sensor is 5%. This sensor is capable of measuring fluid thermal conductivity in the range of 0.2–2 W m−1 K−1. The measurement for each sample was repeated at least three times and the interval between each measurement was 15 min.

For measuring the viscosity, the CAP 2000 + Brookfield viscometer was utilized. This device controls the temperature with an accuracy of 0.1 °C using a thermoelectric module. The rotation speed can be selected from 5 to 1000 rpm. A pin-on-disk device according to ASTM G99 standard was employed to measure the wear rate. The amount of wear in any system usually depends on the factors of the amount of force applied, the characteristics of the test device, the speed of rotation, the distance of rotation, the environment and the properties of the material. The wear test device and viscometer used are presented in Fig. 4.

Fig. 4
figure 4

a CAP 2000 + Brookfield viscometer, b Pin-on-disk device

Full size image

Uncertainty analysis

The accuracy of the viscometer is ± 2% and the thermal analyzer device is ± 5%. The uncertainty of the measured data is calculated from the following equation.

$$U = \frac{S}{\sqrt N }$$
(1)

where U is the standard uncertainty, N is the number of measurements, and S is standard deviation that is calculated from Eq. 2.

$$S = \sqrt {\frac{1}{N - 1}\sum\limits_{i = 1}^{N} {\left( {X_{\text{i}} - \bar{X}} \right)^{2} } }$$
(2)

Based on this equation, the uncertainty of dynamic viscosity of the base oil at temperature of 15 °C and shear rate of 7998 1 s−1 was acquired 2.12%. Also, the uncertainty of thermal conductivity of base oil at temperature of 25 °C was calculated 5.13%.

To calculate the wear rate uncertainty, the Kline–McClintock equation is used, which is according to Eq. 3.

$$\begin{aligned} & \mathop U\nolimits_{\text{R}} = \sqrt {\sum\limits_{i = 1}^{N} {\left( {\frac{\partial R}{{\partial x_{\text{i}} }}\mathop u\nolimits_{{{\text{x}}_{\text{i}} }} } \right)^{2} } } \\ & \mathop U\nolimits_{\text{R}} = \sqrt {\left( {\frac{{u_{\text{x}} }}{x}} \right)^{2} + \left( {\frac{{u_{\text{y}} }}{y}} \right)^{2} + \left( {\frac{{u_{\text{z}} }}{z}} \right)^{2} } \quad R = R(x,y,z) \\ \end{aligned}$$
(3)

Here uxi is the standard deviation in the measurement xi, and ∂ R/∂xi is the partial derivative of the function R with respect to xi. The uncertainty in measured value of wear rate is calculated from Eq. 4 as:

$$\begin{aligned} & U_{{{\text{w}}_{\text{r}} }} = \left[ {\left( {\frac{{u_{{{\Delta m}}} }}{{{{\Delta m}}}}} \right)^{2} + \left( {\frac{{u_{\uprho} }}{\rho }} \right)^{2} + \left( {\frac{{u_{\text{l}} }}{l}} \right)^{2} + \left( {\frac{{u_{\text{F}} }}{F}} \right)^{2} } \right]^{{\frac{1}{2}}} \\ & = \left[ {\left( {\frac{0.0001}{0.0046}} \right)^{2} + \left( {\frac{78.1}{7810}} \right)^{2} + \left( {\frac{0.1}{3000}} \right)^{2} + \left( {\frac{0.36}{180}} \right)^{2} } \right]^{{\frac{1}{2}}} \simeq 2.39\% \\ \end{aligned}$$
(4)

As can be seen in Eq. 4, the uncertainty of wear rate is obtain approximately 2.39% and the dominant term in wear rate uncertainty is \(\Delta m\).

Results and discussion

Changes in the thermal conductivity of the C60-SiO2/5w30 nanofluids respect to the solid volume fraction at 25 °C are presented in Fig. 5. As can be seen, all nanofluids have a higher thermal conductivity than pure oil. It is also observed that as the concentration of nanoparticles increases, the thermal conductivity coefficient enhances as well. This increasing trend was observed in many studies [45,46,47,48,49,50]. The highest increase in thermal conductivity coefficient was about 9% at volume fraction of 1%. Improving thermal properties of the oil plays an important role in the engine cooling and it rises its efficiency. The reason for the increase in thermal conductivity can be attributed to the creation of nanoparticle cluster in nanofluid and enhance the molecular transport mechanism.

Fig. 5
figure 5

Nanofluid thermal conductivity in different volume fractions

Full size image

The shear stress variations in terms of shear rates in different volume fractions are displayed in Fig. 6. As can be seen, the highest shear stress was at the volume fraction of 1% in all temperatures. Also, according to this graph, as the shear rate increases, the difference between the shear stress values in the minimum and maximum volume fractions rises. One of the most common methods for detecting nanofluid behavior is fitting the power equation on this graph, which has been used by various researchers [51,52,53,54,55,56,57,58,59].

Fig. 6
figure 6

Shear stress changes vs. shear rate in different volume fractions and various temperatures

Full size image

Table 4 represents the values of the power index (n) and the strength index (m) obtained by fitting the power equation to the shear stress graph in terms of shear rate. As can be seen, the R-squared values for all nanofluids up to 65 °C are above 0.99 and this value is greater than 0.92 for 65 °C. The high value of regression coefficient indicates the high accuracy of the fitted power equation to predict the rheological behavior of the investigated nanofluids. The values of n index in all volume fractions and temperatures were less than one, indicating the non-Newtonian shear thinning behavior of the base fluid and the studied nanofluids. It is also observed that the strength index values increase by rising the volume fraction and decreasing the temperature. This behavior is the result of an increased intermolecular force and the resistance of the fluid and nanoparticle layers at higher volume fractions and lower temperatures.

Table 4 - Evaluation of regression coefficients (R2), power index (n) and strength index (m) at different temperatures
Full size table

Figure 7 indicates the effect of shear rate on dynamic viscosity at different volume fractions and various temperatures. As can be observed, with increasing the shear rate, the nanofluid viscosity decreases in all volume fractions and temperatures. This means that the nanofluids as well as the non-Newtonian base fluid are pseudoplastic fluid. Also, this phenomenon reveals that the fluid will flow more easily if the applied shear force increases.

Fig. 7
figure 7

Viscosity changes vs. shear rate in different volume fractions and various temperatures

Full size image

The variation of the viscosity respect to the temperature at different volume fractions at constant shear rate of 7998 s−1 is shown in Fig. 8. The highest viscosity obtained at 5 °C and 1% concentration. According to this graph, viscosity decreases in all concentrations with the increase in the temperature. When the temperature rises, the oil dilutes and the movement between the nanoparticles becomes easier. Temperature increase in fluids leads to reduced intermolecular attractions (cohesive) that this phenomenon can be seen in many common liquids such as water, oil and EG.

Fig. 8
figure 8

viscosity changes in terms of temperature in different concentrations at shear rate of 7998 s-1

Full size image

Figure 9 shows the viscosity changes with the concentration in various temperatures. As can be observed, in volume fraction of 0.1% and 0.25% as the concentration of nanoparticles increases, the viscosity of the nanolubricant slightly decreases. The phenomenon is related to roller impact of spherical nanoparticles in the base oil. In contrast, in volume fraction of 0.5%, 0.75% and 1% by rising the concentration, dynamic viscosity enhances. This phenomenon occurs because the nanoparticles and their agglomerations impede on the relative movement of the oil layers. The lowest viscosity was observed at 65 °C and at the concentration of 0.25% which was approximately 8% less than the base oil. On the other hand, the highest viscosity occurs at 5 °C and 1% concentration. In general, it can be said that viscosity is inversely related to temperature. It should be noted that the effect of temperature on increasing or decreasing the viscosity is much greater than the effect of concentration.

Fig. 9
figure 9

Viscosity changes vs. volume fraction in different temperatures at constant shear rate of 7998 s−1

Full size image

In the wear test, the goal is to determine the best concentration in order to reduce friction between the components. After each step of the friction coefficient test, the mass of the disks was measured with an exact scale with the accuracy of 0.0001 gr. Mass measurement was performed to compute the wear of the components. The wear rate can be calculated by Eq. 5 [60].

$$W_{\text{r}} = \frac{\Delta m}{\rho .l.F} .$$
(5)

In this formula, Wr is the wear rate (m3 N−1 m−1), Δm is mass loss (gr), ρ is sample density (gr m−3), l is distance (m), and F is the applied load (N). The sample density is 7,810,000 (gr m−3).

It should be noted that three samples of base oil and nano-oil in volume fraction of 0.5% and 1% at a constant distance of 0 to 500 m were tested at 80 rpm and ambient temperature. The applied load on the sample was 180 N. The results are represented in Fig. 10. As can be observed, as the concentration increases, the wear rate decreases. Also, the concentrations of 0.5% and 1%, wear rate has decreased by 16.64% and 18%, respectively.

Fig. 10
figure 10

The wear rate values in different concentrations

Full size image

Figure 11 illustrates the friction coefficient diagram in terms of distance at 25 °C for three studied samples. At this temperature, the average rate of the friction coefficient was 0.23 for pure oil, and the nanofluid friction of coefficient with a volume concentration of 0.5% and 1% is 0.21 and 0.19, respectively. A 17.39% reduction in friction coefficient is visible at this temperature for concentration of 1%. Therefore, rolling mechanism is observed with the presence of nanoparticles in the base oil.

Fig. 11
figure 11

Coefficient of friction versus distance at 25 °C

Full size image

Proposed correlation

Statistical analysis was applied to model the dynamic viscosity of C60-SiO2/SAE 5W50 nanofluid based on RSM. Experimental data were used as historical data to correlate the mathematical formula. Temperature, concentration and shear rate were employed as independent input variables, and dynamic viscosity was set to be the dependent output variable. The levels and characteristic of input and response variable are provided in Tables 5 and 6, respectively.

Table 5 Input variable parameters in RSM
Full size table
Table 6 Characteristic of the response variable in RSM
Full size table

Table 7 presents the statistical results of different functions. As can be seen, two-degree polynomial function has the maximum adjusted R2 and consequently was employed as the optimal function.

Table 7 Statistical results of different functions
Full size table

The dynamic viscosity correlation which was obtained from RSM is displayed in Eq. 6.

$$\begin{aligned} & \mu^{ - 0.05} = 0.73316 \, + \, 0.00208659 \, * \, T \, + \, 0.00380303 \, *\emptyset + \, 1.82142e - 06 \, * \dot{\gamma } \\ & - 5.43636e - 05 \, * \, T \, *\emptyset + \, 6.72356e - 09 \, * \, T \, *\dot{\gamma } - 1.06601e - 07 \, * \emptyset *\dot{\gamma } \\ & - 5.22374e - 06 \, *T^{2} - 0.00690896 \, *\emptyset ^{2} - 6.19815{\text{e}} - 11 \, *\dot{\gamma }^{2} . \\ \end{aligned}$$
(6)

where T is temperature (oC), \(\emptyset\) is solid volume fraction of nanofluid (%), and \(\dot{\gamma }\) is shear rate (s−1).

The important factors of ANOVA for the predicted model are represented in Table 8. As can be seen, R2 = 0.9983, adjusted R2 = 0.9983 and predicted R2 = 0.9982; they represent conformity between the experimental data and model ones. In addition, to measure the deviation of the data from the mean value, the coefficient of variation (C.V.) has been employed. The lower the C.V., the more favorable the model would be. The C.V. gained was 0.1827 for the fit function, which is an excellent value to confirm the selected model

Table 8 The summarized results of ANOVA for predicted model
Full size table

Figure 12 displays the regression graph. As can be observed, there is an appropriate consistency between the predicted model and experimental data.

Fig. 12
figure 12

Regression graph of the model-predicted values with experimental values

Full size image

The 3D surface graphs of the model gained from the statistical analysis are presented in Fig. 13. Also, the effect of the temperature, concentration and shear rate on the response are displayed in this figure.

Fig. 13
figure 13

Model output graphs by RSM method as 3D surface. a interaction within temperature and solid volume fraction on viscosity, b interaction within temperature and shear rate on viscosity

Full size image

Conclusions

In this study, thermal conductivity, viscosity and wear rate of C60-SiO2/5w30 hybrid nanofluids were measured and evaluated. In order to study the rheological behavior, the dynamic viscosity evaluated at different volume fractions within 0–1% and temperatures from 5 to 65 °C. The wear rate was also measured in two volume fractions of 0.5 and 1% in ambient temperature. The main observation of this study summarized as follow.

  • By increasing the volume fraction, the nanofluid thermal conductivity coefficient also increased and the highest enhancement in thermal conductivity coefficient was 15% at the volume fraction of 1%.

  • The rheological behavior of nano-oil indicated that the nanofluids had a non-Newtonian shear thinning behavior.

  • The dynamic viscosity was inversely related to the temperature. In fact, an increase in temperature led to a significant decrease in the viscosity in all volume fractions.

  • As the nanofluid volume fraction increased, the viscosity of the nanolubricant rose only for the concentration of 0.5%, 0.75% and 1%.

  • The highest increment in nanofluid viscosity was 14% at the volume fraction of 1% and temperature of 5 °C.

  • Statistical approach with RSM was employed to model the dynamic viscosity of nano lubricant.

  • The wear test results showed that with the increase in concentration, the wear rate decreased. The decrease in wear rate at concentrations of 0.5% and 1% was, respectively, 16.64% and 18%.

The results of this study are considered as a part of a comprehensive research on the properties of hybrid nano-oils. To introduce an industrial nano lubricant, some factors such as analyzing the flash point, pour point, specific gravity and kinematic viscosity should be evaluated. Also, conducting economic studies, using the oil in cars for field tests, and taking the environmental considerations into account are essential.