Tribo-Electric Performance of Nano-Enhanced Palm Oil-Based Glycerol Grease for Electric Vehicle Bearings

Abstract

Rolling Bearings are crucial components for induction motors and generators in electric vehicles (EVs), as their performance considerably influences the system’s operational reliability and safety. However, the commercial greases used for bearing lubrication in EV motors pose a detrimental impact on the environment. In addition, they are ineffective in mitigating the effect of electric discharges on rolling surfaces leading to premature bearing failures. This study investigates the viability of a developed eco-friendly grease from palm olein as the base oil and glycerol monostearate as the thickener, enhanced with conductive multi-walled carbon nanotubes (MWCNTs) for EV motor bearings prone to electrical currents. Chemical–physical, tribological, and electrical tests were conducted on the developed grease samples without and with MWCNTs at 1 wt.%, 2 wt.%. and 3 wt.% concentrations and results were compared to lithium and sodium greases. Palm grease samples demonstrated a lower EDM voltage range reaching 1.0–2.2 V in case of 3 wt.% MWCNTs blends, indicating better electrical conductivity and protecting the bearing surfaces from electric-related faults. These findings were further confirmed using vibrations measurement and SEM-EDX analysis of the electrically worn bearings. Bearings lubricated with palm grease blends exhibited lower vibration levels. Palm grease with 2 wt.% MWCNTs reduced vibration amplitudes by 28.4% (vertical) and 32.3% (horizontal). Analysis of bearing damaged surfaces revealed enhanced damaged surface morphology for MWCNT-enhanced palm grease as compared to surface lubricated by commercial greases. The results of this work indicate that the proposed bio-grease is a promising candidate for future application in the field of next-generation electric mobility systems.

1. Introduction

Electric vehicles (EVs) are recently introduced to the international market as a sustainable and eco-friendly mobility system in comparison with traditional internal combustion engine (ICE) vehicles. Due to technological advancements, increased environmental awareness, and supportive governmental policies [1,2,3,4], EVs are gaining broad acceptance and are increasingly viewed as a viable and desirable alternative to reduce greenhouse gas emissions and the environmental effect of transportation [5]. They possess excellent benefits, such as high reliability, high power density, high efficiency, and the capability of instant start [6]. The market share of EVs has been increasing, with countries such as China, the EU, and the United States leading the change from IC engine vehicles to electric alternatives in 2023 alone, with almost 14 million [7]. In 2023, electric vehicles were registered worldwide, recording an increase of 35% from 2022 and reaching around 40 million operational units [6]. The European Environment Agency reports that the harmful gas emissions from EVs during their lifecycle are up to 30% lower than those of ICE vehicles. It is expected that the lifecycle emissions of a typical electric vehicle could be reduced by at least 73% by 2050 [8].

EVs are equipped with almost the same motor drive system consisting of a DC/DC converter for voltage regulation, an inverter for motor control, and an electric motor. Much of the stability in these motor systems contributes to the reliability of the entire EV. Electric vehicles use various types of motors: DC motors, induction motors, permanent magnet motors, PM brushless DC motors, and switched reluctance motors [9,10]. Among these, the three-phase induction motor remains the most-used prime mover [11,12]. Yet, parasitic currents problem arises in EVs from the inherent resistances, inductances, and capacitances in components like DC-DC converters and Pulse Width Modulation (PWM) inverters [13,14]. PWM technique in variable frequency drives (VFDs), commonly used in electric vehicle (EV) inverters to control motor speed and torque, can generate high-frequency switching events that induce common-mode voltages on the motor stator windings [15,16]. These rapid voltage changes, combined with the inherent resistances, inductances, and especially the parasitic capacitances between motor windings, cables, and the vehicle chassis, result in the buildup of shaft voltages. The parasitic capacitances form an electrical pathway that allows capacitive currents to flow from the stator through the rotor and along the shaft. In this case, the lubricant used for the bearings acts as an insulator, allowing electrical charge to accumulate. When the shaft voltage across the bearing exceeds the dielectric strength of the lubricant film, the oil film can no longer insulate the bearing elements, leading to transient breakdown events. This causes a sudden discharge of current through the bearing—known as electric discharge machining (EDM) current—which passes through the thin lubricant film in the form of miniature electric arcs.

Many researchers have taken up extensive research work to explain the relationships between different influential parameters, mechanisms, and corresponding modes of failures that affect the motor bearings [17,18,19]. EDM-bearing currents occur through two main mechanisms. Under a fully lubricated condition, the insulating properties of the bearing lubricant allow it to act as a capacitor. EDM currents are generated when the common-mode voltage surpasses the dielectric strength of the lubricant, resulting in electrical discharges within the bearing [20]. Under such conditions, the lubricant shifts instantaneously from behaving as a capacitor to functioning as a resistor. Additionally, EDM currents can occur when the lubricant film fails to adequately separate the bearing surfaces, leading to contact between surface asperities [21]. This breakdown of the lubricant film creates short circuits, triggering electrostatic discharges. These currents can range between 0.5 and 3 A. Since their magnitude is generally unaffected by motor size, EDM currents pose a greater risk to smaller motors (less than 110 kW) [22,23]. The rapid energy dissipation from these discharges can cause localized temperature spikes, leading to surface melting in the bearings. Such melting often results in microscopic craters, as well as larger-scale frosting, fluting damage, white etching cracks, and lubricant degradation [24].

The common methods for mitigating this problem have been insufficient to eliminate bearing currents, and the mechanism of how these currents damage the bearings is still not well understood [25]. Several mitigation techniques have been devised to solve this problem. For instance, shaft grounding brushes are meant to create a low-impedance path for stray currents, rerouting them away from bearings. These brushes, however, need constant maintenance and replacement because of wear and contamination. Neglecting routine maintenance can lead to high impedance, thereby diminishing their performance over time [26]. Insulated bearings are made in such a way that current passage is avoided through the use of non-conductive materials [27] and coatings [28]. Despite this design, insulation breakdown can occur due to inherent material defects or presence of high-frequency voltages that allow current passage and lead to damage.

While many attempts are made to eliminate them, shaft voltages cannot be completely avoided [29]; thus, studies on their effects become more important in electric and hybrid electric vehicles. When electric discharges flow through the bearing elements, it creates microscopic arcing damage on the raceway surface. When rolling elements pass over these damaged areas, this generates vibrations that result in a thinner lubricant film, which in turn enables easier passages of subsequent current. As time passes, this process creates grooves or holes on the raceway, which gradually raise vibration levels, initiate lubrication breakdowns, leading to instability, and noise, which eventually leads to bearing mechanical failure [30,31].

The lubrication failure in electrically charged scenarios has been a frequent concern due to the altering of the chemical composition and heating of the lubricant on a localized scale by electric sparks [32,33]. The energy dissipation forms conductive particles in the lubricant, depletes critical additives, and promotes oxidation and sludge formation [34]. The presence of these particles adds more electrical pathways, increasing the chance of discharges. In addition, wear of the lubricant film reduces its capacity to keep contact surfaces apart, resulting in more friction and wear, and an increased risk of bearing mechanical failure [35,36,37]. Hence, parasitic currents have been reported to shorten the useful life of lubricants drastically, requiring more frequent maintenance and creating a reliability challenge for electric vehicles and industrial motors [38,39].

Significant efforts have been devoted to improving bearing lubricants to reduce the impact of parasitic currents. Incorporating superior additives—antioxidants, anti-wear agents, and extreme-pressure enhancers—has raised the thermal and chemical stability level in greases, even when they are subjected to rigorous working environments [34]. Nanomaterials such as activated carbon nanoparticles [38,39], carbon nanotubes, boron nitride, graphene [40], and graphene oxide have been recently found to be environmentally friendly anti-wear (AW) and extreme-pressure (EP) additives for lubricants [41,42]. The addition of nano-additives has shown promise in enhancing tribological and physical performance in lubricants [43]. Furthermore, current research focuses on the development and investigation of conductive greases formulated with ionic liquids [44]. Despite such developments, there are still insufficient studies that explored the application of advanced metallic and carbon-based nano-additives to enhance the tribo-electric behavior of bearing greases such as silver nanoparticles [45], alumina (Al2O3), and multi-walled carbon nanotubes (MWCNTs) [46].

Traditional lubricants are usually fossil fuel-based, thus highly contributing to high carbon emissions, harm to the environment due to poor disposal methods, and biodegradability problems, all of which pose significant challenges to environmental sustainability [47]. Bio-based lubricants derived from renewable sources present themselves as an alternative solution that, in addition to decreasing dependence on limited fossil fuel resources, also demonstrate improved viscosity, lower toxicity, and reduced greenhouse gas emissions during their lifecycle [48]. Vegetable oils, such as rapeseed oil [49], Soybean oil [50], Jatropha oil [51], and Cottonseed oil [52] have been receiving significant interest because of the better biodegradability compared to animal fats and esters [53]. It has also been recently demonstrated that lubricating greases based on vegetable oils provide an adequate and homogeneous full lubricant film with comparable load-carrying capacity to conventional lithium greases [54,55]. Palm oil has emerged as a potential alternative to traditional mineral oils as a lubricant due to its distinguished physical properties and high production efficiency [56]. It is composed of free fatty acids like linear carboxylic acids with carbon chains between 12% and 24%. Its main components like oleic (monounsaturated), linoleic (polyunsaturated), palmitic, and stearic acids are responsible for the efficiency of palm oil as a lubricant to minimize wear and friction [57,58].

Few tribological studies [59,60,61] have evaluated palm olein blended with mineral oils for reciprocating equipment. Nevertheless, palm oil has drawbacks concerning low oxidation stability and marginal kinematic viscosity values, further limiting its applicability in mechanical systems. Hence, researchers have tested various additives to overcome the drawbacks [62].

Based on the previous review of the literature, most of the papers focused on the impact of electrical discharge machining (EDM) currents on bearing degradation. They explored different mitigation strategies such as shaft grounding brushes, insulated bearings, and improvements in lubricant properties. These studies have highlighted the importance of addressing parasitic currents and their detrimental effects on bearing life, lubricant degradation, and motor performance in electric vehicles (EVs). However, only a few research works have addressed the impact of applying nano-additives to bearing grease on the severity and mode of damage in bearing elements due to electric faults. Furthermore, the literature lacks any effort to consider replacing conventional lithium grease with more environmentally friendly and sustainable lubricants in EV bearings as a proposed solution to mitigate or control the parasitic current damage effect.

This work investigates the effect of a palm oil-derived glycerol bio-grease on bearing tribo-electric performance under parasitic electrical currents. The emphasis is on assessing bearing electrical degradation under stepwise DC voltages (0–10 V) with palm grease in plain form and with varying concentrations of MWCNT additives. A custom test setup with an electrical circuit across the bearing rings is employed. Voltages as low as 5 V can lead to electrical bearing damage, especially in systems equipped with variable frequency drives (VFDs), due to parasitic currents and electrical discharge machining (EDM) effects [63,64]. Although AC and DC discharge characteristics are different, dielectric breakdown and EDM-induced damage mechanisms are similar. DC voltage testing here is employed to determine the threshold for EDM events and evaluate each lubricant’s ability to suppress electrical damage. Voltage and current signals are measured using sensors and an oscilloscope to characterize capacitive, EDM, and resistive responses. Bearing damage severity is assessed by vibration analysis under 100 N radial load and 1400 rpm, and surface analysis is conducted through SEM/EDX to confirm the protective role of palm grease and MWCNTs. The findings are aimed at facilitating the development of sustainable, reliable lubricants for EV applications. The manuscript is organized as follows:

Section 2—details of the test procedures, conditions, and six grease samples: (A) lithium, (B) sodium, (C) palm grease without additives, (D) palm grease with 1 wt.% MWCNTs, (E) palm grease with 2 wt.% MWCNTs, and (F) palm grease with 3 wt.% MWCNTs.

Section 3—comparative analysis of EDM damage mitigation.

Section 4—summary of findings and implications for EVs sustainable lubrication.

2. Materials and Methods

2.1. Test Rig

A customized test rig was designed and fabricated to conduct performance evaluations of ball bearings lubricated with various palm grease blends, as shown in Figure 1. The rig consists of an induction motor, flexible coupling, shaft, two supporting deep-groove ball bearings, and the test bearing mounted on a specially machined hub. A vertical radial load of 100 N is applied to the test bearing through a hinged loading mechanism that ensures constant contact. The radial load of 100 N used in this study was chosen by carefully balancing mechanical test rig constraints and relevance to actual electric vehicle (EV) motor bearing operating conditions.

Firstly, the test setup features an overhung spindle with the 6006ZZ deep groove ball bearing positioned at the shaft end and loaded via a simple hanger mechanism attached to the bearing housing. Initial trials indicated that radial loads ranging from 50 N to 300 N enable a stable bearing operation without excessive spindle deflection or misalignment, thus avoiding undesirable vibration and preserving experimental repeatability. A value of 100 N was selected within this stable range. Secondly, the 100 N load is well below the dynamic load rating (13,200 N) of the 6006ZZ bearing tested (as listed in

Table 1. Specifications of the test bearing (6006zz) provided by the NSK manufacturer.

Dimensions (mm)	d	D	B	r
	30	55	13	1
Mass (kg)	0.116
Dynamic load rating, C (N)	13,200
Static load rating, Co (N)	8300
Clearances (µm)	C3 (13–28)

), consistent with SKF catalog recommendations for moderate-duty applications. This load represents a realistic operating condition for medium-sized electric motors used in passenger EVs, where typical radial bearing loads vary widely depending on design and service conditions. According to reference [65], typical radial loads in electric motors commonly range from 100 N to 1000 N. The selected 100 N load thus reflects the lower bound of realistic radial loading encountered in motor bearings, providing a relevant and conservative test condition for tribological and electrical discharge evaluation. This configuration enables stable operation at a fixed rotational speed of 1400 rpm and operating temperatures maintained below 60 °C.

Additionally, an electrical circuit was developed across the test bearing hub using a programmable DC power supply, a digital storage oscilloscope, and precision electric terminals to monitor voltage and current across the bearing. This setup enables controlled application of DC voltages (0–10 V), with a current limit of 3 A, mimicking shaft voltage buildup and bearing current flow scenarios typically encountered in inverter-fed induction motors for EVs. Voltage increments of 0.5 V were applied every 5 min, and the onset of discharge events, indicating electrical discharge machining (EDM), was recorded to identify the transition from capacitive to resistive regimes.

While this laboratory setup does not fully reproduce the dynamic and thermally fluctuating conditions of real-world EV systems, the selected operating parameters were carefully aligned with reported ranges for electric motor bearing environments in EV applications, particularly with regard to electrical stress and moderate mechanical load [27,66,67].

The power supply is used to generate the desired DC voltage across the test bearing hub using electric connectors and wiring installed into the hub, where one terminal is to be connected to the bearing outer ring while the other terminal is attached to the shaft and inner ring. The inner ring of the test bearing is electrically insulated from the bearing hub using a polyethylene film. A two-channel digital oscilloscope is exploited to monitor and record both the applied voltage and current during the electric tests, where the first channel is used to record the voltage signal while the second channel is connected to a current sensor that monitors the current signal across the bearing raceways.

It is acknowledged that the electrical environment in electric vehicle bearings is dominated by AC and high-frequency PWM voltages, which can result in more sever discharge behaviors and surface damage mechanisms than those produced by DC excitation. In this study, DC voltage was employed to enable controlled, stepwise assessment of the dielectric breakdown and electrical discharge mitigation performance of the tested lubricants. This approach allows for direct comparison of the intrinsic electrical insulating properties of each grease formulation. DC testing simplifies the electrical environment, allowing for accurate detection of leakage current, electrical discharge machining (EDM) thresholds, and tribofilm formation without the complexities of alternating polarity and high frequency switching transients. This approach is consistent with methodologies reported in prior research aimed at isolating and benchmarking the electrical behavior of bearing lubrication systems [68,69,70].

The characteristics of the setup components are listed in

Table 2. Specifications of test rig components.

Component	Specifications
Electric Motor	GAMAK (3 hp, and 1400 rpm)
Base	C45 Carbon Steel
Shaft	SUS 420 Stainless Steel
Two Support Bearings	NU1011M Roller Bearing
Dc power supply	RD6024/RD6024-W
Digital oscilloscope	DQ7022S, two-channel (25 MHz)

. The selected bearing for this investigation is a deep-groove ball bearing of type NSK 6006ZZ (NSK distributer—Arabian Co for Electrical & Mechanical Supplies, Alexandria, Egypt) to imitate the used bearing type in the EV induction motors [71]; their main information are listed in Table 2 according to NSK bearing manufacturer (NSK Ltd., Tokyo, Japan) [72]. In this work, the motor runs at 1400 RPM while a dead weight of 100 N is hinged to the bearing hub through a suitable hook.

2.2. Characterization of Nanomaterials

Multi-walled carbon nanotubes (MWCNTs) with a diameter of 10–40 nm and a length of up to 5 µm are selected in this work as a carbonaceous nano-additive material with anisotropic characteristics. Based on the local manufacturer product specifications, they possess a purity level of 92 ± 2%.

To analyze the structural characteristics of MWCNTs, including particle dimension, particle morphology, and the amount of particles clustering, a scanning electron microscope (SEM) (JEOL JSM-IT200, JEOL Ltd., Tokyo, Japan) and a transmission electron microscope (TEM) (JEOL JEM-2100, JEOL Ltd., Tokyo, Japan) were employed. The sample preparation involved adhering the powder sample to double-sided carbon tape, then coating the sample with a platinum–palladium layer at a current of 40 mA using the JEOL JEC-3000 FC Auto Fine Coater. For TEM analysis, the preparation process included dispersing the MWCNT powder in ethanol and subjecting the mixture to ultrasonic agitation for 15 min.

2.3. Grease Synthesis

The proposed lubricant in this work is manufactured using 70% palm olein (base oil) and 30% glycerol monostearate (thickener). The preparation process is planned to ensure proper mixing, temperature control, and homogenization to achieve the desired grease properties [73,74]. Initially, a 2 L beaker filled with palm olein is placed on a magnetic stirrer with a heating function and heated to a temperature of 120 °C. The heat treatment in this step reduces palm olein’s viscosity, hence efficiently allowing the incorporation of the thickener.

As the palm olein reaches the target temperature, 20% glycerol is gradually added to the beaker while maintaining continuous stirring to ensure uniform distribution within the palm olein matrix. The temperature is sustained within 120 °C for 30 min, allowing sufficient interaction and partial integration of the components. An additional stirring period of 15 min ensures homogeneity [75]. After the initial heating and mixing, the flask is kept at 120 °C, and the remaining 10% of the glycerol is added to the mixture. This addition is critical for achieving the desired consistency and further blending the components. The mixture is stirred at 120 °C, so the new olein spreads uniformly [76]. Finally, it is left to cool to room temperature before mixing it with a commercial-grade mixer for 30 min. It will ensure the glycerol is evenly dispersed in the palm olein, creating stable and well-structured grease.

The mixing step is conducted utilizing a mechanical mixer until the output grease appearance is uniform and its color is milky yellow in the case of palm grease in its plain form and black in the cases of samples of palm grease with three different blends of 1 wt.%, 2 wt.%, and 3 wt.% MWCNTs, respectively. Samples of each blend are weighed and applied to the identical test ball bearings guided by SKF formula in Equation (1) [42].

$$G_q=0.005\left(D\right)(B)$$

(1)

where Gq is the grease quantity to lubricate the bearing (g), D is the outside diameter of the ball bearing (mm), and B indicates the bearing width (mm). Palm grease samples with the selected concentrations of multi-walled carbon nanotube (MWCNT) powder are prepared by incorporating the MWCNTs into the base grease matrix through mechanical stirring. The mixing process is carried out using a laboratory-grade mechanical stirrer to ensure homogeneous dispersion of the nanoparticles throughout the grease medium. Following preparation, the samples are labeled according to their respective MWCNT concentrations and subsequently subjected to a comprehensive series of evaluations, including chemical, physical, tribological, and electrical characterization, as summarized in

Table 3. The developed test grease blends.

Grease Sample Label	Grease Blends
Palm Grease A	Palm-oil-based glycerol grease (in its plain form)
Palm Grease B	Palm grease with 1 wt.% MWCNTs
Palm Grease C	Palm grease with 2 wt.% MWCNTs
Palm Grease D	Palm grease with 3 wt.% MWCNTs

. For benchmarking purposes, conventional lithium grease and sodium grease—each composed of 80% mineral oil and 20% thickener—are tested under identical experimental conditions. The subsequent sections detail the methodologies and procedures employed to assess the performance characteristics of the formulated palm-based greases.

2.4. Kinematic Viscosity Test

The palm oil is tested for its kinematic viscosity at 40 °C and 100 °C, according to ASTM-D445 [77]. For this purpose, a sample volume of 25–30 mL is forced to flow in a capillary tube of a viscosity meter surrounded by a temperature-controlled liquid bath. The test is repeated three times, and the average value of kinematic viscosity is recorded. Based on the literature review, the recommended ISO VG (International Organization for Standardization Viscosity Grade) for ball bearings falls between VG 68 and VG100 [78].

2.5. Dropping Point Test

To evaluate the palm grease consistency at elevated temperature in rolling bearings during operation, the dropping point is determined according to ASTM D2265 [79,80]. A 10 g sample of the prepared grease is placed in a brass cup, supported by a glass test tube, and placed in an aluminum block oven. The temperature gradually increased to 232 °C in the oven, with the temperature observed using an installed thermometer to the oven. Once an oil droplet falls from the cup into the test tube, the temperature is recorded by the thermometer and identified as the grease’s dropping point, indicating its heat resilience.

2.6. Thickener Consistency and Oil Bleeding Test

The consistency of the developed grease is evaluated to determine the resistance of the fibrous structure of the grease to a specified radial load applied in a standardized manner. A grease with adequate consistency ensures that the grease thickener possesses the required stiffness during service, while allowing base oil to be released through structure channels for lubrication in machinery components. A cone penetration test is used for this purpose according to ASTM D217 [62,78]. A metallic cup is completely filled with a grease sample size of 500 g. Then, a penetrometer of a cone shaped load is set free to drop vertically into the cup with the sample and is left for 5 s at a temperature of 25 ± 5 °C before recording the depth reading on a dial gauge in tenths of a millimeter. The output reading is compared to the National Lubricating Grease Institute (NLGI) scale to identify the consistency of the grease.

The static bleeding test is performed according to ASTM D1742 [81]. Each grease sample is placed in a sieve positioned above an aluminum crucible, with a static load of 100 g applied directly to the grease. The sample is then placed in an oven at 40 °C for 18 h. After the test period, the released oil is collected, weighed, and the amount of oil separation is calculated by dividing the weight of the released oil by the total amount of grease.

2.7. Tribological Test

The friction coefficient of palm grease samples was measured using the ball-on-disk method according to ASTM G99 [82]. The test method is designed to simulate sliding/rolling contact between lubricated surfaces with the aim of studying the frictional behavior of the grease under controlled conditions. A steel ball 10 mm in diameter rests on the surface of a rotating disk with a diameter of 200 mm as described in Figure 2. A thin, uniform layer of grease (30–40 mg) is evenly distributed over the whole disk surface prior to each test using lint-free cloth. Though ASTM G99 [82] does not mandate precise mass, we ensured consistent coverage by measuring and distributing the grease so that no dry spots formed. The disk rotates at a uniform speed, normally 600 rpm, while the rolling element (ball) is pressed against the disk under some controlled radial load of 10 N for a sliding distance of 1000 m of sliding distance. The high viscosity and consistency of the palm grease (NLGI 4) and its thickener structure (glycerol monostearate) ensured stable adhesion to the disk surface under these moderate speeds. This stability was further confirmed by the repeatability of the stable coefficient of friction (COF) measurements across three trials for each sample.

This test is performed at 25 °C to maintain consistency in the result. During testing, the friction force is generated at the contact point between the stationary pin and the rotating disk due to the resistance to relative motion. This force acts tangentially to the circular path of the pin and in the direction opposite to the surface velocity of the rotating disk. Its magnitude is proportional to the applied normal load and the coefficient of friction between the two surfaces. The friction force is instantaneously measured using a load cell attached to the pin holder. The measured frictional force is divided by the applied normal load to obtain the COF. Following every test run, the average COF will be calculated over each condition.

2.8. Electric Tests

One of the main objectives of this study is to identify the voltage ranges across the lubricant grease-separating bearing surfaces in which capacitive, electric discharge machining (EDM), and resistive currents occur. This is done to examine how the conductivity of the lubricants influences the occurrence and extent of harmful EDM-related phenomena [83,84]. To achieve this, the electrical tests are carried out in three sequential phases, as explained in Figure 3. The first phase, referred to as the run-in stage, involves continuously operating each test bearing for 16 hrs. A moderate radial load of 100 N is applied to the bearing hub, and the induction motor is set to run at a constant speed of 1400 rpm, as demonstrated in Figure 4. This stage aims to smooth out surface asperities, smoothing out the bearing raceways and rolling elements. Additionally, this step helps in reaching a full film lubrication condition, preventing direct contact between asperities that could interfere with voltage buildup [85,86]. After this stage, the electrical test sequence begins with an incremental voltage increase from 0 to 10 V in 0.5 V steps every 5 min to determine the capacitive regime and the threshold of EDM currents. Once the EDM voltage is reached, the test continues for approximately 6 h to ensure sufficient time for bearing exposure to electric discharges initiating possible damage mechanisms. By the end of this phase, the voltage increments are increased by additional 0.5 V steps every 5 min until resistive current regime is reached and confirmed by the oscilloscope monitoring of voltage and current signals. The entire test was conducted at a controlled room temperature of 25 °C. The lubricant can be viewed as a parallel RC (resistor–capacitor) circuit—accurately represents the basic electrical behavior of a lubricated rolling bearing under applied voltage, according to the literature [63,64]. When the lubricant film is intact, it separates the bearing’s rolling elements and races, acting as a dielectric between two conductors acting as a capacitor (C). When the voltage exceeds the dielectric strength of the lubricant film, the film breaks down and allows current to flow a resistive path form (R). Hence, the two paths (capacitive and resistive) exist simultaneously. Under normal (non-breakdown) conditions, only the capacitive path is active. During breakdown, the resistive path is activated, and both may be conducted (especially during transient events).

In the second phase, a DC voltage starting from 0 V and increased incrementally up to 10 V is applied to the bearing–lubricant system, with the current limit set to 3 A, values consistent with those reported in previous studies [87,88]. At the onset of this phase, the voltage is raised in 0.5 V increments every 5 min. This controlled increase enables careful monitoring of the electrical behavior across the bearing, particularly for identifying the characteristic domains of capacitive, electrical discharge machining (EDM), and resistive responses, as observed through a digital oscilloscope.

Initially, the electrical current remains negligible or near zero due to the capacitive nature of the system. However, once the applied voltage reaches a threshold sufficient to initiate EDM, sharp transient discharge events are detected. At this point, a sudden increase in current is typically observed, marking the onset of breakdown and conductive pathways through the lubricant film or surface asperities. The voltage at which these discharge events first occur is recorded as the EDM threshold voltage. Following this identification, the bearing is operated for 6 h under this constant potential difference to investigate discharge-induced degradation. Subsequently, the incremental voltage increase continues until the electrical behavior transitions into a stable resistive regime, where conduction becomes continuous rather than discharge-based, aligning with findings in prior literature [87].

2.9. Vibrations Analysis Test

The dynamic performance of each test bearing is evaluated after applying electric tests using vibration measurement and analysis as shown in Figure 5. This step is essential to stand upon the level and type of damage that took place in each electrocuted bearing, and how each grease blend lessened the harmful effect of parasitic currents on the surfaces of each bearing element. At this stage of work, the setup is set to run while the test bearing is still installed inside the hub after being disconnected from the electric circuit. Bearing vibration signals are measured using a uniaxial accelerometer mounted on the bearing hub by a magnet. The vibration signals are collected in both vertical and horizontal directions for analysis using a 2-channel data analyzer (Commtest VB5, Bentley Nevada, NV, USA). Vibration measurements are processed using special filters and analyzed in time domain and frequency domain. The amplitude of vibration levels is displayed as acceleration (mm/s²), and the mean and standard deviation of three consecutive spectra results for the same test bearing are calculated.

3. Results and Discussion

3.1. SEM and TEM Results

Figure 6 demonstrates the SEM and TEM results of MWCNT powder. Based on the examination of Figure 6a, the MWCNTs exhibit an entangled and intertwined rope-shaped structure in bundles. The TEM image in Figure 6b further confirms their curved hollow tube structures, which are typical of MWCNTs. The nanotubes have average diameters ranging from 10 to 15 nm, with lengths extending up to 5 μm. According to the supplier, the purity percentage of the supplied MWCNTs is around 92 ± 2%, which is confirmed by the SEM and TEM observations.

3.2. Physical–Chemical Results

The developed bio-grease is tested for its physical–chemical characteristics and compared with commercial lithium and sodium greases, as presented in Table 4.

The grease sample of palm oil shows an NLGI 4 with a relevant unworked penetration value of 185 dmm. This penetration value is considerably lower than the two benchmark commercial greases, indicating the higher rigidity of the palm oil-based formulation. Regarding the bleeding properties, the bio-grease sample demonstrates a higher percentage of oil release, compared to commercial greases [89]. This is illustrated by the sensitivity of the glycerol thickener consistency in particular to the temperature, which leads to the breakage of structural bonding, yielding more oil separation [90]. While the static oil bleeding rate of >6% is higher than the normal level for conventional applications, this characteristic should be evaluated within the specific context of EV bearing requirements. The higher oil release rate can be beneficial in electrically stressed applications, as it ensures continuous replenishment of the lubricating film at the contact points where electrical discharges occur. Furthermore, modern EV designs often incorporate sealed bearing units with controlled environments, where oil bleeding is less problematic. The trade-off between some conventional performance metrics and the environmental and electrical benefits represents a deliberate design choice that aligns with the growing emphasis on sustainability in the automotive sector.

Palm grease was synthesized using 30% glycerol monostearate as a thickener, which forms a robust crystalline network that traps palm olein (base oil). This proportion of glycerol in the formulation targeted increasing the dropping point and mechanical stability of the grease, allowing it to maintain its consistency well beyond 40 °C. The dropping point tests (Section 2.5, ASTM D2265) confirmed that the palm grease formulations had dropping points between 50 and 55 °C (Table 4), without significant oil separation or flow, which was also confirmed through visual inspection and stable COF readings during tests. Furthermore, similar bio-greases with glycerol monostearate thickeners have been reported to withstand temperatures up to 70 °C without failure [61] since the thickener’s high melting point (~80 °C) and hydrogen bonding with palm oil acids (e.g., palmitic/oleic acid) enhanced its thermal resilience.

Turning to the kinematic viscosity results, the grease sample based on palm oil demonstrates a viscosity value of 41 cSt at 40 °C, which is significantly lower than that of the commercial greases. However, the calculated viscosity index of palm grease is 209, which indicates more stability of palm oil properties at elevated temperatures in comparison with mineral oils used for lithium and sodium greases. Moreover, the base oil of the bio-grease sample reveals a significant value of the pour point. This behavior may be justified by the levels of saturated fatty acids, which constitute 52% of the structure [58,91]. The high saturation inhibits the grease’s ability to flow at lower temperatures, thereby increasing its resistance to movement [53].

Table 4. Physical and chemical results of test grease samples.

Test Description	Test Standard	Lithium Grease	Sodium Grease	Palm Grease A
Unworked penetration test (dmm)	ASTM D217 [62,78]	250 (NLGI 3)	265 (NLGI 2)	185 (NLGI 4)
Static bleeding value 40 °C (oil mass %)	ASTM D1742 [81]	1–3	3–4	7.1
Dropping point (°C)	ASTM D2265 [79,80]	180–200	180–200	55–60
Kinematic viscosity at 40 °C (cSt)	ASTM D445 [77]	125	135	41
Kinematic viscosity at 100 °C (cSt)	ASTM D445 [77]	12	10	9
Viscosity Index (VI)		82	20	209
Pour point (°C)	ASTM D7346 [92]	−15	−5 to −10	9

Table 4. Physical and chemical results of test grease samples.

Test Description	Test Standard	Lithium Grease	Sodium Grease	Palm Grease A
Unworked penetration test (dmm)	ASTM D217 [62,78]	250 (NLGI 3)	265 (NLGI 2)	185 (NLGI 4)
Static bleeding value 40 °C (oil mass %)	ASTM D1742 [81]	1–3	3–4	7.1
Dropping point (°C)	ASTM D2265 [79,80]	180–200	180–200	55–60
Kinematic viscosity at 40 °C (cSt)	ASTM D445 [77]	125	135	41
Kinematic viscosity at 100 °C (cSt)	ASTM D445 [77]	12	10	9
Viscosity Index (VI)		82	20	209
Pour point (°C)	ASTM D7346 [92]	−15	−5 to −10	9

3.3. Estimated Lubrication Regime and Minimum Film Thickness

It is crucial that the lubricant film generates an elastohydrodynamic lubrication condition between rolling elements and raceway surfaces to act as a capacitance and completely isolate the interfaces during the electric tests. Hence, each grease amount inside the rolling bearing was calculated. The developed oil film thickness during phase 1 of the electric test is a function of the bearing hertzian contact area, base oil viscosity, operating speed, and radial load.

To gain a deeper understanding of how the kinematic viscosity and consistency of each grease blend influence the developed film thickness, lubrication regime, and coefficient of friction (COF) of the test bearings, the minimum oil film thickness (H_min) is determined. The Hamrock–Dowson equation, recognized as the standard method for calculating the minimum film thickness in oil-lubricated rolling bearings, is utilized in this section for this purpose [93] as shown in (Equation (2)).

$${\mathrm{H}}_{\mathrm{m}\mathrm{i}\mathrm{n}}=3.63{\mathrm{U}}^{0.68}·{\mathrm{G}}^{0.49}·{\mathrm{W}}^{-0.073}(1-{\mathrm{e}}^{-0.68\mathrm{k}})$$

(2)

where U is the speed parameter, G is the material parameter, and W is the loading parameter. Notably, the material parameter G is directly proportional to the pressure–viscosity coefficient (α) of the base oil in the grease, which defines how viscosity increases under pressure, critically affecting EHL film thickness [94]. For the studied greases, the used pressure–viscosity coefficient values are as follows: commercial lithium grease: α = 2.0 × 10⁻⁸  Pa⁻¹, sodium grease: α = 0.5 × 10⁻⁸  Pa⁻¹, and palm grease: 1.0 × 10⁻⁸  Pa⁻¹ [95,96,97]. These values are obtained from published data on base oils and confirmed through comparison with reported film thickness calculations for similar lubricants [98]. To assess the lubrication regime of the grease under the selected speed and loading conditions, the λ parameter is computed (Equation (3)), representing the relationship between the minimum oil film thickness and the combined surface roughness of the interacting surfaces of the bearing raceways [99,100,101].

$$\mathsf{\lambda }={\displaystyle \frac{{\mathrm{h}}_{\mathrm{m}\mathrm{i}\mathrm{n}}}{\sqrt{{{\mathrm{R}}_{\mathrm{q}1}}^2+{{\mathrm{R}}_{\mathrm{q}2}}^2}}}$$

(3)

Rq₁ represents the surface roughness value of the rolling element and Rq₂ is the corresponding surface roughness of the inner or outer raceway in contact. For the lubricated bearing to operate in the elastohydrodynamic lubricating (EHL) regime, λ values should be above 3. However, λ values between 1 and 3 indicate an operation of bearing in a mixed lubrication regime with partial film separation and moderate wear. On the other hand, boundary lubrication condition with direct metal contact takes place in case the λ value is less than 1, leading to high mechanical wear and a resistive state of current.

In the film thickness calculation, a temperature of 40 °C was assumed, which aligns with the design limits and operating conditions specified for the bearing–lubricant system under study. This temperature was selected based on expected thermal behavior during moderate-speed and moderate-load operation, as commonly reported in elastohydrodynamic lubrication (EHL) studies [102,103]. In the actual tests, bearing temperature was measured directly using a contact thermometer placed near the outer race of the bearing housing. The average temperature is recorded during each running test and was found to be 40 ± 4 °C, depending on lubricant type, applied voltage, and test duration. These values confirm that the assumed temperature for the film thickness model reasonably represents the real thermal environment during bearing operation [104].

There are several factors that control the dielectric strength and EDM voltage of lubricants in bearings subjected to parasitic currents such as oil film thickness, lubricant relative permittivity, and applied nano-additives. The minimum oil film thickness of the hertzian contact between rolling element and each raceway is one decisive factor in determining the capacitance and the dielectric strength of a typical lubricant type. According to capacitor theory, electric capacitance is inversely proportional to the film thickness. In a previous study [105], an experimental test setup was used to establish a more accurate simulation model that correlates the capacitance of a lubricant in deep-groove ball bearing and film thickness. In a more recent study, Bader et al. [106] experimentally confirmed that the oil film thickness is inversely proportional to the capacitance of the lubricant with increasing the operating speed. On the other hand, Maruyama et al. [107] developed electrical impedance methods to simultaneously measure oil film thickness and breakdown ratio in elastohydrodynamic (EHD) contacts of practical ball bearings. The breakdown ratio was found to increase as the oil film thickness decreases, indicating that thinner films are more prone to electrical breakdown under operational conditions.

The values of H_min and λ are calculated using Equations (2) and (3) for palm grease, lithium grease, and sodium grease; the values are presented in

Table 5. Calculated minimum film thickness H_min and λ parameter for each grease type.

Lubricant Blend	Operating Speed (rpm)	Bearing Radial Load (N)	Interacting Raceway	Hmin (µm)	λ
Lithium grease	1400	100	Inner	1.03	7.28
			Outer	1.21	8.60
Sodium grease			Inner	1.20	8.52
			Outer	1.42	9.65
Palm grease A			Inner	0.50	3.53
			Outer	0.59	4.16

. All grease blends showed λ parameter values higher than 3, indicating that the test bearings operate predominantly in the EHL regime under the predefined operating speed and load. The differences between the results amongst the grease types are mainly attributed to the base oil viscosity since other parameters remained constant for all tested grease types. Test bearings lubricated with lithium grease and sodium grease showed calculated minimum oil film thickness (H_min) values of 7.28 µm and 8.52 µm at the inner raceway, respectively. In comparison, they are 8.60 µm and 9.65 µm at the outer raceway, respectively. The estimated values suggest that these greases have adequate viscosity values at the operating temperature, which are confirmed by the kinematic viscosity results in Table 4. Based on the previously mentioned works, it can be inferred that they possess the highest electric capacitance values amongst all grease blends and are expected to have a higher threshold for EDM voltage levels than palm grease samples. Palm grease A (without additives) exhibited lower (H_min) values (0.50 µm for the inner raceway and 0.59 µm for the outer raceway), corresponding to λ parameter higher than the threshold for EHL zone. Hence, the calculated (H_min) values indicate that palm grease is expected to contribute to lowering the breakdown voltage value in comparison with lithium and sodium grease when subjected to electric field across the bearing rings.

3.4. Tribological Results

According to the calculations of minimum oil film thickness, at the hertzian contact, all tested grease blends manage to operate the tested deep groove ball bearing in the EHL regime. This positively impacts reducing the friction coefficient during tribo-testing using the pin-on-disk tester. Figure 7 shows the calculated average COF values for three repeated tests for each grease type. It is observed that the COF of palm grease A (in its plain form) is lower than that of lithium grease and sodium grease by around 40% and 30%, respectively. For palm grease B, C, and D, the COF is decreased by 53%. The lowest COF value was recorded for palm grease D, reaching only 0.06.

Both constituents of palm grease play a vital role in enhancing the friction behavior between the two mating surfaces of bearing components [61]. Glycerol as a thickener has polar molecules, namely the hydroxyl (–OH) groups, which create hydrogen bonds with oxides present on steel surfaces. This interaction creates a hydration layer, which reduces direct metal-to-metal contact and acts as a boundary lubricant [108,109]. Palm oil, on the other hand, is rich with long-chain fatty acids (palmitic acid, oleic acid, and linoleic acid). It adheres to surfaces primarily through free (unsaturated) fatty acids that interact with metal surfaces via chemisorption, forming a metal soap layer (metal–carboxylate complex). Although less interactive with metal surfaces, the non-polar hydrocarbon chains in palm oil from the saturated palmitic fatty acids still adhere to metal surfaces via Van der Waals forces, forming a thin, stable hydrophobic film that repels water and enhances lubrication. Hence, the multi-layered boundary films from the base oil and thickener reduce friction by preventing direct contact between sliding surfaces.

The addition of MWCNTs in palm grease blends (B, C, and D) further enhances the COF values by creating a protective tribofilm on contact surfaces, minimizing direct metal-to-metal interaction and reducing friction [110,111,112,113]. They can possibly act as nano-rollers and their rolling effect between frictional interfaces contributes to a lower COF by transforming sliding friction into a combination of rolling and sliding motion [114]. Moreover, MWCNTs absorb onto the metal surface via weak van der Waals interactions and/or π–π stacking with metallic surfaces containing carbonaceous layers to generate a thin lubricating layer, reducing direct contact between asperities [115].

3.5. Electrical Conductivity and EDM Voltage Analysis

The main focus of this work is to study the electric behavior of palm grease as an alternative to commercial greases in rolling bearings under electric fields. Additionally, the influence of MWCNT nano-additives at different concentrations of palm grease on the threshold and range of EDM voltage is being investigated. The bearing circuit current and the corresponding input DC voltage signal are continuously recorded for each test bearing, as described in Figure 3.

Figure 8 summarizes the obtained recorded electric results for each grease blend. In the case of lithium grease, the electric circuit remained in capacitive state at voltage amplitudes below 1.8 V. This phase is manifested by the absence of flowing electric current across the lubricant sides. When the input DC voltage value reached 1.8 V, repeated electric discharges started to appear in the monitored current signal, indicating the onset of EDM voltage occurrence. Increasing the input voltage beyond the threshold results in a corresponding rise in the severity of EDM current amplitudes. By reaching voltage levels of 7.5 V, both voltage and current signals manifest significant surges with rapid fluctuations, indicating that the grease in the hertzian contacts enters a resistive state to current flow. Sodium grease exhibited similar behavior to lithium grease, indicating comparable dielectric strength.

An example of the EDM voltage and current events for lithium grease are depicted in Figure 9 under 1.7 V amplitude. It is observed from the voltage signal that the lubricant undergoes rapid transient states of charging and discharging of current. During charging phase of the lubricant as a capacitance, the bearing voltage remains at an approximately constant value, then the amplitude suddenly drops to almost zero, marking the start of the discharging phase during which the insulating limit of the grease is reached, triggering the behavior of the lubricant to alternate between capacitive fluctuation from (EDM) current [116].

Palm grease A (in its plain form) showed a different electric capacitance response, with EDM voltage ranging from 1 V to 2.2 V only before entering the resistive state. This indicates that palm grease has lower breakdown voltage (BDV) than the two benchmark greases and operates more likely in the resistive state under considerable voltage values with limited possibilities for electrical discharge events to take place. The introduction of 1 wt.% MWCNTs into palm grease further reduced the BDV range to between 0.5 V and 0.9 V. This moderate decrease compared to plain palm grease suggests that MWCNT nanoparticles suspended in the grease and also adsorbed to the metallic surface provide channels for electrical current to flow at lower applied voltages, helping to limit the voltage fluctuations originating from EDM current charges. More enhancement was observed in the case of 2 wt.% and 3 wt.% MWCNTs where the voltage range reduced to only 0.6–1 V and 0.5–0.75 V, respectively. This substantial reduction in EDM voltage ranges suggests that MWCNTs increased the electric conductivity of the grease, preventing excessive EDM current effects, without affecting tribological performance as confirmed by Figure 7.

The presence of transient voltage spikes observed during testing is more accurately attributed to electrical discharge events (EDMs) occurring across the lubricant film, rather than indicating a complete transition to a resistive conduction state. These discharges arise when the applied voltage locally exceeds the BDV of the lubricant, i.e., resulting in momentary breakdowns of the insulating film. This phenomenon should be understood as statistical in nature, where the likelihood and frequency of discharge events increase progressively with rising electric field intensity. It does not represent a binary shift from capacitive to resistive behavior, but rather a growing probability of localized dielectric failure. The lubricant film may continue to exhibit predominantly capacitive characteristics while sporadically allowing discharges under high potential gradients. This interpretation is consistent with established research on electrostatic bearing degradation and discharge mechanisms in lubricated contacts under electric stress [31,117].

Palm oil has lower electrical conductivity than other vegetable oils with higher unsaturated fatty acid content. The presence of approximately 40–45% saturated palmitic acid (C16:0) reduces the number of polar functional groups, contributing to electrical conductivity [118]. A previous study was conducted by Slita et al. [119] to measure and compare the electrical resistivity of various vegetable oils. The results showed that palm oil had higher resistivity than other oils, indicating lower electrical conductivity. Yet, palm oil is still rich in monounsaturated oleic acid (C18:1), accounting for around 39–45%, polyunsaturated linoleic acid (C18:2) representing 10–11%, and minor traces of myristic acid (C14:0), lauric acid (C12:0), and linolenic acid (C18:3), as well as moisture [120]. Hence, the dielectric behavior (i.e., relative permittivity) of palm oil is significantly controlled by its fatty acid composition, particularly the balance between saturated and unsaturated fatty acids [121]. The presence of unsaturated fatty acids along with traces of moisture contributes to the increased molecular polarizability, which can enhance the dielectric constant in comparison with mineral oils existing in lithium and sodium greases. In a previous study by Rajab et al. [122], palm oil’s dielectric constant and BDV was compared to mineral oil at different temperature ranges as an insulating liquid for transformer operations. It was found that the dielectric constant of palm oil is in the range of 3.0–3.2, while mineral oil is around 2.1–2.2 at room temperature. This is due to the higher degree of polarity in palm oil, which is defined by the high degree of unbalance in molecules’ geometrical chemical structure. Furthermore, the BDV of palm oil at temperatures below 100 °C is lower than mineral oil as an insulating oil. It was justified that the relative amount of water concentration in palm oil is higher than the case of mineral oil, which enhanced the polarity and reduced BDV. Other researchers conformed with the previous results and proved that different vegetable oils exhibit higher dielectric constant values compared to mineral transformer oils [123]. Hossain et al. [121], on the other hand, compared palm oil along with other vegetable oils to mineral oils in terms of viscosity, BDV, and relative permittivity. Palm oil showed higher BDV than mineral oil, in contrast to the results of the current work. However, this is justified by the higher viscosity results of palm oil (4 times) than the mineral oil used.

The apparent contradiction of reduced friction and electrical breakdown, in the case of palm grease samples, despite the oil film being thinner than that of lithium and sodium grease, can be explained by the formation of a robust protective tribofilm on the bearing surface. This tribofilm arises from the polar fatty acid components of palm oil and is further enhanced by the presence of multi-walled carbon nanotubes (MWCNTs) in the experimental blends. The tribofilm effectively reduces direct metal-to-metal contact, acting as a physical barrier that lowers friction and mitigates electrical discharge machining (EDM) events by providing conductive pathways that dissipate charge and prevent severe electrical breakdown. Moreover, while transient local mixed or boundary lubrication conditions may arise during electrical discharge events, the tribofilm buffers these occurrences, maintaining lubrication efficiency and protecting against wear. Thus, the lubrication regime with palm grease should be characterized as primarily EHL combined with dynamic mitigation of localized boundary interactions through tribofilm formation, resulting in superior tribo-electric performance compared to traditional lithium and sodium greases.

3.6. Vibration Analysis Results

Following the electric tests, the damage level in each lubricated test bearing is assessed in this section by operating the bearing under the same operating conditions in the test rig while switching off the electric circuit. Vibrations are measured in the time domain and frequency domain to evaluate the induced faults within the bearing elements using a radially mounted accelerometer to the bearing hub [124,125]. The electric sparks generated during EDM discharges can produce extremely high temperatures, which typically range up to several thousand degrees Celsius [117]. These intense temperatures are sufficient to locally melt and vaporize bearing surfaces, leading to surface localized faults such as spalling and micro-pitting. Frequent melting on localized areas on the surface may cause material quenching in the presence of base oil and gradually generate frosting or distributed roughing of the surface.

Each time the rolling elements pass over a localized fault, it induces recurring temporary impulses in the measured vibration acceleration signal at one of the bearing characteristic frequencies in the low-to-medium frequency range [126,127]. By this time, thousands of micro-pits are generated, resulting in distributed faults along the raceway of rough-surface-inducing harmonics and sidebands to the bearing fault frequency signal and worsening the vibration levels. An example of the time waveform and frequency spectrum for an already electrocuted test bearing with palm grease is shown in Figure 10. The time domain shows the effect of the damage on the raceways in the form of transient spikes where the time between each consecutive impulses represents the frequency of the bearing fault. The frequency domain shows peaks at 1x of the operating speed with its harmonics resembling the excessive clearance in the bearing due to wear. The bearing fault frequency and its harmonics are raised by a hump of energy, indicating broadband noise and late stage of bearing faults before failure.

Figure 11 shows the summary of the overall amplitudes of vibration spectra in RMS that are calculated for each bearing. It is observed that sodium grease exhibits the highest vibration levels among the tested lubricants, indicating severe electric-based damage on the bearing raceways and rolling element during operation. Lithium grease, on the other hand, significantly improves vibrations, with 23% and 25% drop in overall levels in the vertical and horizontal directions, respectively, when compared with sodium grease. This means less induced localized and distributed faults on the bearing raceways when using lithium grease. It also indicates that lithium grease provides better protection against electric sparks while providing better tribological behavior.

A similar improvement was detected for test bearings lubricated with palm grease. In its plain form, palm grease reduced the vibrations amplitudes by 26.5% in the vertical direction and 33% in the horizontal direction, compared with sodium grease. When MWCNTs are introduced into palm grease, the vibration levels further decrease, highlighting the impact of nanoparticles in mitigating the damage-induced vibrations. For instance, bearings lubricated with palm grease B (blended with 1 wt.% MWCNTs) reduced the vibration levels (in the vertical direction) by 23% and 41% compared to lithium grease and sodium grease, respectively.

The most significant improvement is observed with palm grease C and D, which achieve the lowest vibration levels among all tested greases. These formulations reduced vertical and horizontal vibrations by 29% and 33% compared to lithium grease, respectively. The presence of MWCNTs in palm grease contributed to lower EDM voltage threshold, leading to expected less intense sparks between rolling elements and raceways. While the EDM discharges intense heat, which tends to degrade the grease components, MWCNTs on the mating surfaces of the already electrocuted bearing raceways have the potential of creating mechanical interlocking and fill-in effects. They penetrate surface asperities, smoothing rough surfaces and filling micro-pits. This mitigates the resultant mechanical vibrations from localized faults (such as micro-pits and spalls) and distributed faults (frosting rough texture) on the surface.

Figure 12 demonstrates a section of the test bearing lubricated with lithium grease after the electric test run. The raceway of the inner ring reveals the generation of gray frosting in the middle region. In contrast, the lithium grease on the raceway, the cage, and rolling element surfaces suffered a discoloration from bright to dark yellow and gray. Test bearings lubricated with other grease blends experienced a similar damage pattern.

3.7. Surface Damage Analysis

For more profound understanding of vibrations and tribo-electric results, SEM examination is conducted on the worn surfaces of sectioned raceways of each test bearing subjected to the electric stress at a resolution of 10 μm and 1000× magnification, as shown in Figure 13. The images depict damaged raceway surface texture and faults to provide more information on the impact of each grease blend on the bearing tribo-electric performance when subjected to the parasitic currents. Further analysis is conducted using energy dispersive X-ray analysis (EDX) using (JEOL JSM-IT200, Akishima, Tokyo, Japan) to discover evidence of the deposited chemical elements on the worn surfaces of the inner and outer raceways of the bearing and to explain the damage mechanism and severity. In Figure 14a–f, significant concentrations of Fe, Cr, and Mn are detected on the surface, confirming the presence of steel alloy constituents of the bearing raceway material.

The formation of such tribofilm was confirmed via SEM-EDX examination of the bearing raceway, which revealed a smoother worn surface and the presence of a uniform film-like layer on the contact track for MWCNT-enhanced grease samples compared to unmodified grease. EDX analysis of the same regions further supported this finding by showing elevated carbon signals and reduced metal peak intensities, indicating that the surface was covered with a carbonaceous layer attributed to MWCNTs. Tribofilm involving carbonaceous layers, metal oxides, or organometallic compounds have been reported when MWCNTs were used under electrical stress.

Figure 13a shows the wear scar surface of the inner race of the ball bearing sample lubricated with lithium grease. Widely distributed micro-pits with frosting of a dull appearance and rough morphology can be observed on the examined worn surface. The micro-pits and large craters indicate severe surface damage induced by electric discharge sparks during the test duration, leading to localized melting and material removal. The repeated discharges cause surface alterations, leading to the characteristic frosted appearance. Based on the calculated film thickness and lambda values for lithium grease, the developed tribo-layer during running tests was expected to sufficiently shield the contact surfaces, leading to adequate protection against mechanical rubbing between the rolling elements and raceways. Mechanical damage in the form of deep furrow marks also occurs due to abrasive wear in the rolling direction. This is justified by the fact that micro-arcing, during EDM current phase, creates cratering and pitting, releasing hardened metallic debris into the lubricant grease and encouraging three-body wear mechanisms. In addition, the rapidly repetitive EDM currents degrade and carbonize the grease blend within the bearing, producing byproduct sludge and abrasive residues. The loss of lubricant film integrity allows contaminants to embed into surfaces, accelerating the three-body abrasion. The previously mentioned faults on the bearing raceway were manifested by the significant vibration levels (RMS) measured on the test bearing in the radial direction. According to Reddy et al. [128], commercial lithium grease possesses low electric conductivity even with lithium ions within the thickener structure.

This increased the electric impedance of the sides of the lubricant film acting as a capacitor. Gradually, voltage amplitudes will build up across the capacitor, leading to significant discharges of intensive heat that eventually evaporate the base oil and create micro-pits across the raceways by electric erosion [129]. In Figure 14a, the EDX analysis of commercial lithium grease reveals moderate carbon content (11.91%) and oxygen content (4.31%) along with traces of Na (0.05%), indicating a remarkable degradation of grease structure under electric stress into its basic elements. This conforms with the SEM findings, which show the rapid evaporation of lubricant film, leaving the raceway surfaces prone to both mechanically abrasive wear and electrical-induced spalling and frosting.

Similar worn surface results are found in the case of sodium grease-lubricated bearings as shown in Figure 13b. Rolling tracks beside narrow and deep grooves can be observed on the examined worn surface [130] caused by secondary damage mechanism from micro-pitting of electrocuted surfaces. Furthermore, obvious frosting texture is detected due to cyclic melting of the surface during the EDM voltage application. Grease components (base oil and soap) decomposed, leaving behind carbonaceous residues and sodium on the raceway surface accelerating the mechanical abrasive wear mechanism.

The elemental analysis of the worn surfaces confirms this. The EDX result revealed a carbon content of 12.9% with extremely low O and Na contents, as seen in Figure 14b. In Figure 13c, the raceway worn surface lubricated with palm grease is mainly shallow with abrasive friction tracks and irregular embossments on the surface from localized melting and spalling.

The fatty acids in palm oil contributed to the reduction in mechanical damage on the raceway surface while also improving the coefficient of friction (COF). This effect can be attributed to their long molecular chains and polar functional groups, which promote strong interactions with the rubbing contact surfaces [62]. In addition, frosting and micro-pits appeared in the case of palm grease as they showed a similar EDM voltage threshold to sodium grease. Figure 14c illustrates the EDX results for the case of palm grease. EDX measurements proved 15.32% carbon content and the highest oxygen content (19.17%). Since the palmitic and stearic acids are long-chain fatty acids bonded to a glycerol backbone, they possess higher molecular weight organic compounds with a more significant proportion of carbon, hydrogen, and oxygen. This is reflected in the increase in the carbon and oxygen content deposited on the surface compared to lithium and sodium grease. Unexpected presence of Si content appears in the EDX spectrum for palm grease samples. Since palm grease in principle does not have silicon-based additives, the detected Si is thus not from intrinsic grease components. The most plausible source of Si could be debris trapped and deposited on the bearing surface due to the thinner lubrication film formed by the palm grease.

For palm grease with MWCNT additives (Figure 13d–f), the worn surface images show that the furrows become superficial and mild with a smoother topography. From the tribological perspective, it is established that the self-lubricating characteristic of the carbonaceous material greatly reduces the COF compared to lithium, sodium, and palm grease, which is reflected on the enhanced worn surface results. Also, the surfaces lack obvious micro-pits or frosting, suggesting a significant impact of MWCNTs in reducing the surface-induced electric faults. This can be justified by the electric stress results where palm grease with MWCNTs exhibited lower EDM voltage margins than other grease types. The palm grease formulation used in this work does not contain any calcium-based thickeners. Therefore, the calcium detected in the EDX analysis of bearing surfaces lubricated with palm grease containing 2 wt.% MWCNTs (Figure 14e) is likely due to environmental contamination or wear debris from bearing components. The presence of MWCNTs may enable the adsorption or localization of Ca ions onto the nanotube surfaces, enhancing their detectability in EDX.

Previous works such as Jonjo et al. [46] proved that the addition of MWCNTs to lithium grease enhanced its electric conductivity and mitigated the surface damage during EDM discharge events. The formation of physical adsorption tribofilm from conductive nano-additives plays a crucial role in facilitating conductive paths for current through the grease structure [131]. The enhancement in electric conductivity reduces the charging time across the lubricant bearing interfaces, resulting in less intense spark to discharge and hence lower damage to the surface of bearings. EDX results (Figure 14d–f) revealed carbon content higher than 60% on the worn surfaces of tested bearings, proving the presence of large residuals of MWCNT layer, confirming the mechanical wear resistance, and less electrical fault features. Palm grease with 2% MWCNTs appears to provide the best balance between lubrication efficiency and oxidative stability among the tested samples.

The service life of the newly developed palm grease formulations containing multi-walled carbon nanotubes (MWCNTs) is dependent on their thermal, oxidative, mechanical, and electrical stability. In this study, grease samples were tested under controlled conditions, including a radial load of 100 N, an average operating temperature of 40 °C, and incremental DC voltages up to 10 V. These conditions simulated moderate stress levels over continuous 6 h test durations, during which no signs of base oil separation, hardening, or degradation were observed. This suggests that the grease maintains functional stability during medium-term operation under typical loading and temperature conditions [132].

The enhanced electrical conductivity observed in MWCNT-doped palm greases can be attributed to the formation of a percolation network, which facilitates electron transport across the lubricant film. This conductivity improvement contributes to lowering the electrical discharge machining (EDM) threshold by enhancing charge dissipation. The network formed by well-dispersed nanotubes remains stable as long as there is no significant mechanical shear degradation or thermal oxidation. Previous studies have shown that carbon-based nanofillers, when adequately dispersed, can sustain enhanced conductivity and electrostatic discharge mitigation capabilities over extended use [132,133].

Nonetheless, over long-term operation, the electrical performance of the grease may decline due to potential agglomeration of MWCNTs, thermal aging of the palm oil base, and mechanical breakdown of the percolated structure. Such effects can reduce conductivity and alter the tribo-electrical behavior. To quantify this decline, standardized long-duration testing such as ASTM D3336 (Grease Life in Ball Bearings at Elevated Temperatures) is recommended, along with extended cyclic electrical stress testing to observe discharge behavior over time [134]. These protocols would provide a more accurate assessment of the grease’s operational life and its electro-functional stability under realistic EV conditions.

4. Conclusions

A novel palm grease, enhanced with MWCNTs, is synthesized and tested against commercial greases in tribo-electric environment. Through rigorous chemical–physical, tribological, electrical analyses, vibrational, and SEM-EDX analyses, this study reveals the following key conclusions:

• The palm-based bio-grease demonstrated superior consistency (NLGI 4) and thermal stability compared to commercial lithium and sodium greases, with a high viscosity index (209) ensuring performance across temperatures.
• Palm grease reduced the coefficient of friction (COF) by 40% versus lithium grease and 30% versus sodium grease. Incorporating multi-walled carbon nanotubes (MWCNTs) further lowered COF to 0.06 (3 wt.%), reducing friction by up to 60 %. The results indicate better physical and chemical adsorption of palm grease to the surface, providing a protective layer.
• Benchmark greases resulted in higher EDM voltage thresholds and ranges, with lithium grease entering the EDM phase at 1.7 V and sodium grease at 1.2 V. Exposed bearing raceways to these frequent current discharges resulted in severe localized and distributed surface damage along with high vibration levels in subsequent bearing runs.
• Palm grease exhibited a lower EDM voltage range (1.0–2.2 V), reducing harmful bearing currents. MWCNT additives (2–3 wt.%) further decreased EDM thresholds (0.5–0.75 V), mitigating surface damage from electric discharges as evident by subsequent vibration analysis and SEM-EDX examination of damaged surface. Palm grease with 2 wt.% MWCNTs achieves optimal vibration damping, reducing vertical and horizontal vibrations by 28.41 % and 32.37 %, respectively, outperforming commercial greases.

In summary, the bio-derived palm grease is found to mitigate harmful electric erosion due to EDM voltage with no compromise in tribological performance. The palm grease with MWCNTs as nano-additives is believed to form a protective tribofilm, reducing direct metal-to-metal contact and improving lubrication efficiency. Its comparable performance to lithium grease makes it a potential lubricant for challenging applications such as electric mobility.

Future work is devoted to study the rheological and thermal behavior of palm grease under different operating conditions including a clear identification of the grease operating temperature limits and the enhancement of the oxidation stability. It is recognized that AC and PWM voltages can induce more severe discharge events, including characteristic fluting damage, which may not be fully replicated under DC conditions. Future work will extend this investigation to include AC and PWM excitation to more closely simulate real-world EV-operating environments Furthermore, a comprehensive investigation of the durability of bearings lubricated with palm-based nano-lubricants is considered for further study to accurately assess how each grease influences bearing the attainable fatigue life factors under electrical and mechanical stresses.