Dynamic performance of newly developed environmentally friendly greases containing polysaccharide gums in rolling bearing

The present study explores the performance of novel vegetable oil-based greases containing polysaccharide gums as additives. Two series of greases with varying concentrations of gum acacia (GA) and guar gum (GG) are evaluated in a rolling bearing at various speeds and loads. The vibrations and shock pulse measurements (SPM) quantified the tribo-dynamics of the developed greases against two commercial greases used as the benchmark. The vibration and SPM assessment suggest that the newly developed eco-friendly greases are at par with the commercial ones; the performance varies with the concentration and type of gum. Where a cumulative effect of rheology and gum size seems to influence the vibrations, the entrainment of gums in the contact zone governs the SPM carpet values. The developed greases show great prospects as a sustainable, potential, and commercial alternative to existing environmentally unfriendly greases.


Introduction
Bearings are an essential and indispensable component of all rotating machinery, capable of support and transmission functionalities.The health of the bearings is often monitored by the vibrations they generate as a common industrial practice. 1,2Excessive bearing vibrations can reduce the efficiency and functional capacities of the machine and increase its downtime by reaching premature failure.The vibration of bearings is governed by many factors, such as raceway design, cage type, surface roughness, seal type, lubricant, etc.However, approximately 80% of bearing failures occur due to lubricant unsuitability, insufficiency, aging, and/or contamination. 3 Grease is the most commonly used lubricant in rolling bearings, accounting for lubricating more than 90% of all rolling bearings. 4It provides a thin film between the rolling elements and the raceway, enabling a cushioning (or damping) effect between the metal-metal contact, thus reducing vibrations. 5Greases are a semi-solid blend of three ingredients: base oil(s), thickener(s), and additive(s).Base oil is primarily responsible for lubrication.Thickener acts like a solid binding network that supplies base oil to the tribo-contacts on applying stresses and reabsorbs it after their removal.
Additives are purposefully added to customize the targeted characteristics.It is well established that the tribological performance of grease is significantly influenced by the proportion of various ingredients. 68][9] Hence, selecting the right combination of ingredients in the right proportion is crucial to develop successful grease formulations so as to extend the life and enhance the performance of bearings and machines.
Unfortunately, the grease industry is saturated with products based on environmentally-unfriendly ingredients.1][12] Synthetic oils ('11%) are emerging as an alternative to mineral oils; however, they are still non-renewable, toxic, and expensive. 13,14On the contrary, the global production of environmentally friendly bio-based oils is less than 1%. 15The commonly-used thickeners (such as soaps, complex soaps, polyurea, etc.) [16][17][18] and additives 19,20 further contribute to the harmfulness of the existing greases.At the same time, the acknowledgment of environmentally friendly alternatives to thickeners and additives is limited only to the boundaries of the research.The recurrent production and consumption of greases based on such environmentally unfriendly ingredients pose high risks to the land and aquatic life, where the greases eventually end up. 21The rising environmental concern worldwide and the stiffening government regulations are pressing the need for ecofriendly lubricants and greases.Apart from the environmental crisis, commercial greases also significantly contribute to the depletion of non-renewable petroleum resources at a rapid rate.An obsession with an escalating performance at the cost of sustainability over the years is the root cause behind this low environmental quotient of greases used today.
The researchers seem to be quite active in recent years in discovering eco-friendly alternatives to existing greases that can balance the equation of performance and sustainability.Vegetable oils have shown the finest prospects as alternative base oil.3][24][25] Among thickeners, modified cellulose [26][27][28][29] and organoclays [29][30][31] have been the two extensively tried and successful options.Although both have performed well, organoclays have additional benefits such as very high surface area (due to nano-dimensional particle size), layerlattice structure, 32 and the unique ability to intercalate/exfoliate in the presence of polymers. 33,34More recently, bio-based polyester has also been tried as an eco-friendly thickener and has shown decent performance. 35Among additives, not much has been established to date, and the research is in the beginning phase only.Cellulose nanocrystals, 36 a-zirconium phosphates, 37 talc nanosheets, 38 calcium carbonate nanoparticles, [39][40][41] etc., are among the few names that have shown good prospects.However, the abovementioned additives have been synthesized using sophisticated techniques incurring huge costs.3][44][45][46][47] The presence of reactive functional groups (carboxyl, hydroxyl, amino) in their structure facilitates easy anchoring over metallic surfaces to form protective films and is the reason behind their superior tribology. 43Besides better tribology, they are biodegradable, renewable, non-toxic, economical, and widely available. 48,49 is worth mentioning that despite several attempts, only a few eco-friendly greases developed in the past have shown acceptable tribological performance comparable to commercial greases. 35,39,42,47urther, the performance of such greases is hardly ever evaluated on actual bearings.This should be noted as it is well-accepted 5,50 that the complex rolling-sliding contact behavior of the rolling bearings cannot be characterized by any individual assessment of friction, wear, vibration, rheology, or any other grease characteristic.In previous studies, 42,47 we have explored polysaccharide gums as additives in the greases based on soybean oil and organoclay, and the developed greases have performed impressively well under the tribological tests.The synergism between organoclay and polysaccharide gums was explored to formulate high-performing greases.This study evaluates the performance of such greases in a rolling bearing.Gum acacia (GA) and guar gum (GG) are used in varying proportions to develop greases.The GA is an anionic gum due to the presence of glucuronic acid (carboxylic acids/ions), 51 whereas the GG is a nonionic gum without any uronic acid content. 52The prime focus is to evaluate the bearing performance of greases using two standard techniques: root mean square (RMS) vibration and shock pulse method (SPM) carpet value.4][55] The performance of developed greases is also benchmarked with that of two commercial greases.

Materials
Soybean oil was purchased from a supermarket.The basic physicochemical characteristics of the oil are determined and outlined in Table 1.The FTIR spectra of soybean oil (Figure 1) display the typical peaks and bands as elucidated in Supplemental File 1.The bentonite clay (primarily montmorillonite) modified with organocations was procured from Bee Chems, India, to serve as the thickener.The ultrapure powders of gum acacia and guar gum were procured from Sisco Research Laboratories (SRL) Pvt. Ltd., India.To conduct experiments, standard open ball bearings (6205) with metal cages and normal clearance were purchased from H.M. Doyal and Co., New Delhi, India.Two multi-purpose commercial greases with lithium and calcium bases were procured from renowned manufacturers.The global market dominance of these greases led to their selection to benchmark the performance of newly developed greases.The general characteristics of the commercial greases and the formulated grease (without gum) are enlisted in Table 2. Due attention was paid to using commercial greases with similar consistencies and base oil viscosities (at operating temperatures) to that of the formulated greases.The names of bearing and grease manufacturers are not disclosed because of copyright issues.

Grease formulation and development
The proportion of ingredients was selected (Table 3) so as to develop greases with bearing-grade consistency (i.e.NLGI 2).Preliminary testing was done to determine the required soybean oil to organoclay ratio (i.e.75:25) to develop a grease of such consistency.This ratio was kept constant while formulating additive-based greases to achieve similar consistencies as the additive concentration varies.This was done for a fair performance comparison among formulated greases.The concentration of additives (GA and GG) was varied in a broad range of 1-10 %w/w (1%, 4%, and 10%) to explore their behavior comprehensively.The maximum doping was constrained to 10 %w/w as the consistency of the developed greases appeared to change beyond that.The methodology adopted to develop greases is reported is illustrated in Figure 2. The basic characterization of the developed greases was reported previously by us. 42The consistencies and dropping points of newly developed greases are included in Table 3 for reference.

Morphological characterization
The morphological attributes of the additives, the newly developed greases, and the commercial greases were investigated using a FESEM microscope (JSM-7800F Prime, JEOL Japan, Spatial Resolution 0.5 nm).The particles of GA and GG were identified in the grease skeleton using the back-scattered electrons (BSE) detector.The methodology to prepare the grease samples for microscopy is elucidated in Supplementary File 2.

Rheological characterization
The yield stress of greases was evaluated on a rheometer (Anton Paar GmbH, Austria MCR 102).Shear stress was measured under rotational shear using parallel-plates configuration (diameter = 19.951mm, relative roughness = 0.4, separation = 1 mm) for a shear cycle comprising of three steps: (i) increasing shear rate from 0 to 100 s 21 , (ii) constant shear rate of 100 s 21 , and (iii) decreasing shear rate from 100 to 0 s 21 .The measurements were taken at 25°C and repeated thrice for accuracy.The shear stressshear rate curves determined the yield stress which was used to compare the greases.

Dynamic performance assessment in rolling bearing
Figure 3 shows the actual photograph and the schematic diagram of the experimental setup used to evaluate the performance of greases.It consists of a shaft supported on two deep groove ball bearings (each with a 55 mm bore).The housings of support bearings are mounted on rigid supports.Locating rings are also provided inside these housings to limit the shaft movement.A D.C. motor drives the shaft via belt drive.The motor is mounted on a separate rigid structure to minimize the transmission of vibrations to the test bearing.Additionally, the motor vibrations are expected to be well isolated by the belt pulley, being a flexible link piece.Therefore, the influence of the vibrations of other components over the test bearing's vibrations can be neglected.The shaft is extended beyond the right support bearing, and the extended portion is stepped into five diameters to mount different-sized bearings.The test bearing containing the test grease was mounted on the step of 25 mm diameter.The specifications of the test bearing are mentioned in Table 4.The exact same bearing was used to test different greases to avoid the effect of different baseline vibrations while comparing performance.An open ball bearing without a sealing ring was used as the test bearing as it needed to be regreased several times.If a sealed bearing had been used, the repeated dismounting and mounting of the sealing ring could have led to its deformation and a change in the bearing vibration levels.A split housing was used for the test bearing to ease this unloading and loading  process.The split housing is held in place by locknuts.
The housing fixes the outer race of the bearing while the inner ring rotates with the shaft.For cleaning, the test bearing was immersed in boiling 1,1,1-trichloroethane for a few minutes to facilitate the removal of the grease. 56,57After cleaning, the new grease was applied on both sides to fill approximately 30% of the bearing's free space 56,58 ('2.5 ml) and was evenly distributed by manually rotating the outer race back and forth.For each grease, measurements were taken on a set of three similar bearings to ensure the repeatability of findings.The greases were observed at an array of radial loads and speeds, as mentioned in Table 5, to understand their behavior comprehensively.Many researchers [59][60][61] have utilized a similar approach in their studies.The operating limits of the experimental rig decided the ranges of loads and speeds.A non-contact tachometer measured the shaft speed, whereas hanging dead loads provided a radial load to the test bearing.
Vibration measurements.A piezo-electric accelerometer (MMF KS 94B-10, voltage sensitivity = 1.060 mV/m/s 2 , resonant frequency ' 50 kHz) was mounted over the top surface of the test bearing housing using a magnetic attachment (Figure 3) to capture vibrations in the zone of maximum load.A supply module connected the accelerometer to data acquisition hardware (National Instrument's compact cDAQ-9171).The signals were then sent to a computer for recording and analysis via LABVIEW software.The sampling rate was set to 30 kHz.In each test, the vibrations were acquired after 2 min of running to ensure the uniform dispersion of grease in the bearing and the heat dissipation balance.This was done to record the steady-state behavior.The signals were recorded as vibration acceleration instead of velocity and displacement, as the lubricant primarily influences the high-frequency range. 62All vibrations were recorded as RMS values.
Shock pulse measurements.The Shock pulse method (SPM) involves monitoring high-frequency compression (or shock) waves of an ultrasonic nature created by a rolling bearing. 63The signal captured through accelerometers includes information from rotational forces, shocks, and friction.The shocks contain \ 1% of the total vibration energy and need specially designed transducers (tuned to resonant frequency of 32 kHz) to effectively capture this small signal.A shock pulse tester (SPM Instrument AB T2000, Sweden) and its transducer SPM 10777 were employed for shock pulse measurements.Similar to the accelerometer, the shock pulse transducer was also pressed over the top surface of the split housing to take measurements in the maximum load zone (Figure 3).The SPM readings were obtained immediately after recording the vibrations to report the steady-state behavior.The instrument takes the bearing bore diameter and shaft RPM as input parameters and gives decibel maximum value (dBm) and decibel carpet value (dBc) as outputs.As the shocks ring the transducer, it outputs electric pulses proportional to the shock magnitude, which are processed to provide dBm and dBc.The dBm is associated with the mechanical state of the bearing surfaces, that is, roughness, damage, stress, etc.In contrast, dBc is associated with lubricant quality and film thickness.
In the present study, only dBc is reported to compare the performance of greases.It can be noted that the average particle size of GG is larger than that of GA. Figure 4(c)-(h) compare the microstructures of the two commercial greases and the formulated greases without gums (bare grease).The bare grease exists as the dense entanglements of the organoclay.5][66] Figure 4(i) and (j) display the microstructures of the formulated greases containing gums.In both GA-based and GG-based greases, the   particles are observed embedded in the entanglements of organoclay.

Effect of polysaccharide gums on the grease rheology
All the newly developed greases displayed ''characteristic yielding'' and ''thixotropic behavior'' typical of lubricating grease (Figure 5).The ''yield stress'' values for developed greases are determined and outlined in Figure 6.In both cases, the results indicate a weakening of the grease structure at lower gum concentrations, followed by a gradual reinforcement as the concentration increases.However, the structure reinforces at different rates for the two gums.The experimental error observed does not exceed 10% of the average values, which ensures the repeatability of the findings.The vibration signals are divided into three frequency bands: 50-300 Hz (low-frequency), 300-1800 Hz (medium-frequency), and 1800-10,000 Hz (high-frequency).This division is recommended as per the standard method (ISO15242-1:2015) for rolling bearing vibration measurement.Figure 8 shows the averages (bars) and standard deviations (error bars) of the RMS vibration acceleration in different frequency bands at varying radial loads for different greases.Literature suggests 62 that lubricant primarily influences the high-frequency (HF) vibrations, as the latter is caused by surface roughness.The lowfrequency (LF) and medium-frequency (MF) vibrations are caused by bearing geometry and waviness, respectively, 67,68 and thus are negligibly affected by the lubricant.Similar behavior is observed in Figure 8.Therefore, the HF band signals are compared in the subsequent text to comment on the vibration performance of different greases; the smaller, the better.Figure 8 shows no significant change in the vibrations with an increase in radial load.Greases display more or less similar vibrations at different radial loads.In general, the lowest vibrations are observed for lithium grease, whereas the highest for calcium grease and GG10 grease.Except for GG10 grease, the formulated greases display intermediate vibrations to the two commercial greases.The influence of additives (GA and GG) on the vibration is also observed.The vibrations are hardly influenced by adding GA at any concentration and are almost comparable to the bare grease.However, the vibrations increase upon adding GG from GG1 to GG10 greases.The GG1 grease reports comparable vibrations to that of bare grease.

Vibration behavior of greases in rolling bearing
Figure 9 shows the averages and standard deviations of the RMS vibration acceleration in different frequency bands at varying speeds for different greases.In general, the vibrations increase with increasing speed in each case.Lithium grease reports the lowest vibrations, whereas calcium, GG4, and GG10 greases show the highest vibrations.The formulated greases, except for GG4 and GG10 greases (i.e.bare, GA1, GA4, GA10, and GG1 greases), display intermediate vibrations.In the series of GAbased greases, the vibrations are hardly influenced by GA concentration and are almost comparable to the bare grease at individual speeds.In the case of GGbased greases, the vibrations increase with the GG concentration at all speeds.

SPM carpet values of greases in rolling bearing
Figure 10 shows the averages and standard deviations of the SPM carpet values at varying radial loads for different greases.In general, lithium grease displays the lowest dBc, whereas calcium grease displays the highest dBc.The formulated greases display intermediate values.
Figure 11 shows the averages and standard deviations of the SPM carpet values at varying speeds for different greases.Similar to the case of radial loads, lithium grease reports the lowest dBc, whereas calcium grease reports the highest dBc.The formulated greases display intermediate values.

Effect of polysaccharide gums on the bearing vibration
Figure 12(a) compares the RMS vibration of greases while operating under a radial load of 5 kg and a speed of 1000 RPM.During the vibration measurement, the radial vibration signal received by the sensor from the bearing housing is primarily composed of three components: friction-induced vibrations, the dynamic vibrations caused by the collision between bearing elements (i.e.balls, raceway, and cage), and the run-out of spindle rotation.Assuming the other two vibration sources to be similar in each case, the vibration trend in Figure 12(a) is expected to be governed by friction-induced vibrations.The total friction acting in the bearings is the sum of four components: churning friction, rolling friction, sliding friction, and seal friction. 69To compare the frictioninduced vibrations for different greases, the behavior of each friction component should be first understood, as simplified by Lugt. 5 The churning friction is the friction between grease-grease and thus depends on the channeling (or rheological) properties of the grease.A grease with high yield stress is a channeling grease, that is, where the grease primarily flows unidirectionally from the rolling track toward the sides, not vice-versa, leaving stable channels.The better the channeling, the lower the churning friction.However, the too-high yield stress value is also undesirable as it may lead to lower replenishment rates during the later phases of operation, where the bearing operates for most of its life. 5The rolling friction is mainly a function of grease base oil viscosity.Lower the viscosity, the lower the rolling friction.The sliding friction is governed by the base oil viscosity in the presence of full film and the boundary friction coefficient in the absence of full film. 5Since the testing was conducted for a small duration, the sliding friction component is also expected to be primarily governed by the base oil viscosity.Lower the viscosity, the lower the sliding friction.Further, the seal friction is assumed to be identical in each case due to the same set of bearings being utilized.Thus, it is evident that the total bearing friction (or friction-induced vibration) differs for different greases primarily because of two governing factors: (i) the base oil viscosity (Table 2) and (ii) the grease yield stress (Figure 6) at operating temperatures.
Figure 12(a) shows that the formulated greases generally display intermediate vibration performance compared to the two commercial greases.This trend can be attributed to the cumulative role of both the governing factors discussed above.The superior performance of lithium grease can be attributed to a combination of higher yield stress and lower base oil viscosity.Even with a lower viscosity base oil, the higher yield stress is expected due to the fibrous thickener network (Figure 4(c) and (d)) holding the base oil with comparatively stronger forces than other greases.On the other hand, the inferior performance of calcium grease can be attributed to the higher viscosity of the base oil, neutralizing the positive contribution of yield stress.Further, the intermediate performance of bare grease can be attributed to the coexistence of lower yield stress and lower base oil viscosity.
Figure 12(a) also compares the vibration performance of gum-greases.The vibrations are hardly altered upon adding GA to the bare grease.On the other hand, the vibrations increase with the GG concentration in GG-based greases, such that GG10 grease registers even higher vibrations than calcium grease.The role of gums over the vibrations of the bare grease is expected to depend solely on the rheological behavior of the gum-greases, as the base oil viscosity is identical for all the formulated greases.However, a third factor comes into play in  gum-greases.As the size of GA and GG is larger (Figure 4(a) and (b)) than the film thickness of the grease, 59 the particles are expected to cause a high local hertzian pressure while traveling through the contact zone. 55This effect contributes to an increase in the rolling friction component.Therefore, the role of gums over the vibrations of bare grease can be attributed to a cumulative effect of churning (or rheology) and rolling friction.Literature suggests that the rolling friction increases with the size and quantity of the additives present in grease. 55This explains the higher vibrations of GG-based greases compared to GA-based greases.This further explains the increase in vibrations with the GG concentration (i.e.RMS GG1 \ RMS GG4 \ RMS GG10 ).A variation does occur in the yield stresses of GG1 (decreases), GG4 (increases), and GG10 (increases) compared to the bare grease (Figure 6); however, the change in the churning friction seems to be inferior in front of the rolling friction effect.In the case of GA-based greases, the increase in rolling friction with the GA concentration is comparatively small due to the smaller size of GA.This small increase seems to be overcome by a significant increase in yield stress at all GA concentrations (Figure 6), which considerably lowers the churning friction component.Hence, all the GA-based greases display similar vibrations to bare grease.

Effect of polysaccharide gums on the SPM carpet value
The SPM carpet values can be translated as the grease film thickness formed between the rolling elements and raceways. 63,70 However, the behavior of gum-greases is associated with the entrainment of gums in the contact zone. 50,55,68,71In the case of GA-based greases, a scanty supply of GA particles at lower concentrations (i.e. in GA1 grease) is supposed to be behind the high carpet values, thus indicating patchy film (Figure 13(a)).An optimum supply is expected in GA4 grease, forming a coherent film (Figure 13(b)) and the lowest dBc among all GA-based greases, even lower than bare grease.The carpet value increases with the surplus presence of particles, as observed in the case of GA10 grease.This can be attributed to the agglomerations among GA particles at high concentrations, thus making it difficult for the particles to reach the contact zone, forming uneven films (Figure 13(c)).The agglomerated particles are further expected to block the oil supply resulting in insufficient lubrication.However, GA10 grease shows comparatively lower dBc than GA1 grease.A similar entrainment effect is expected in the case of GG- based greases.However, the optimum supply of particles to attain a coherent film is reached at a lower concentration, that is, in GG1 only (Figure 13(d)).This may be due to the larger particle size of GG compared to GA.Thus, GG1 grease displays the lowest dBc among the three formulations of GG, even lower than the bare grease.An agglomeration effect is expected in the case of GG4 and GG10 greases, resulting in uneven films (Figure 13(e) and (f)) and higher dBc.It should be noted that GG4 and GG10 greases display similar carpet values, perhaps due to the saturation reached in the particle entrainment at 4% concentration.Further, it should be noted that the trends of SPM carpet values and the RMS vibrations are not in agreement for gum-greases.[74][75]

Conclusions
This study is a step ahead in finding a sustainable, potential, and commercial alternative to the longestablished environmentally unfriendly greases.The formulated greases have displayed impressive tribological performance under mixed and boundary regimes in our previous studies.In the present study, the performance of environmentally friendly greases is recorded in a real bearing using a standard condition monitoring procedure.Based on the study, the following salient conclusions are made: 1.The newly developed greases generally display acceptable bearing performance in terms of RMS vibration and SPM carpet value, intermediate to the lithium-based and calcium-based commercial greases in the order of lithium .formulated .calcium. 2. The RMS vibration gives insight into the frictional behavior of greases in the bearing.The findings generally indicate a stable performance of GA-based greases at different GA concentrations, whereas a decline in the performance of GG-based greases with an increase in GG concentration.A cumulative effect of particle size and rheology is ascribed to the distinct behavior in the two cases.3. The SPM carpet value gives insight into the grease's film thickness between the rolling elements and raceways.The findings generally indicate the performance of GA-based greases in the order GA4 .GA10 .GA1, whereas that of GG-based greases in the order GG1 .GG4 ' GG10.The distinct entrainment behavior of gums due to different sizes is ascribed to the distinct trends in the two cases.Except for GA1, every additive-based grease performed either equivalent (GA10, GG4, GG10) to or better (GA4 and GG1) than the grease without additive (i.e.bare grease).4.There is a lack of proportionality between the RMS vibration and the SPM carpet value of the formulated greases. 5.The performance of the developed greases is summarized above.However, the proposed approach may have the following limitations: i.The present study does not estimate the bearing life of the developed greases.ii.The greases are evaluated on a laboratory rig instead of an industrial setup.iii.The present study reports the performance of freshly developed greases only.No comments on shelf-life have been made.

Figure 1 .
Figure 1.FTIR spectrum of the soybean oil used for the study.

Figure 4
Figure 4 displays a collage of the microstructures of the additives and the skeletons of different greases.Microscopically, the GA exists as a distribution of irregular particles (Figure 4(a)) with varying sizes of '5-50 mm.On the other hand, the GG exists as entanglements of ribbon-like structures (Figure 4(b)), with lengths '35-125 mm and diameters '1-10 mm.It can be noted that the average particle size of GG is larger than that of GA.Figure4(c)-(h) compare the microstructures of the two commercial greases and the formulated greases without gums (bare grease).The bare grease exists as the dense entanglements of the organoclay.In contrast, the lithium and calcium commercial greases display a fibrous network (wrinkled microstructure) of different entanglement densities representative of different structural strengths.[64][65][66]Figure4(i) and (j) display the microstructures of the formulated greases containing gums.In both GA-based and GG-based greases, the

Figure 3 .
Figure 3. (a) Actual photograph and (b) schematic representation of the experimental setup of rolling bearing.

Figure 7
Figure7shows a typical frequency spectrum of the vibration signals obtained during the experiments.The vibration signals are divided into three frequency bands: 50-300 Hz (low-frequency), 300-1800 Hz (medium-frequency), and 1800-10,000 Hz (high-frequency).This division is recommended as per the standard method (ISO15242-1:2015) for rolling bearing vibration measurement.Figure8shows the averages (bars) and standard deviations (error bars) of the RMS vibration acceleration in different frequency bands at varying radial loads for different greases.Literature suggests62 that lubricant primarily influences the high-frequency (HF) vibrations, as the latter is caused by surface roughness.The lowfrequency (LF) and medium-frequency (MF) vibrations are caused by bearing geometry and waviness, respectively,67,68 and thus are negligibly affected by the lubricant.Similar behavior is observed in Figure8.Therefore, the HF band signals are compared in the subsequent text to comment on the vibration performance of different greases; the smaller, the better.Figure8shows no significant change in the vibrations with an increase in radial load.Greases display more or less similar vibrations at different radial loads.In general, the lowest vibrations are observed for lithium grease, whereas the highest for calcium grease and GG10 grease.Except for GG10 grease, the formulated greases display intermediate vibrations to the two commercial greases.The influence of additives (GA and GG) on the vibration is also observed.

Figure 5 .
Figure 5. Variation of shear stress with the shear rate demonstrating the ''characteristic yielding'' and ''thixotropic behavior'' of formulated greases.

Figure 7 .
Figure 7. Vibration frequency spectrum for the bare grease (without additive) at 5 kg radial load and 1000 RPM speed illustrating the vibration peaks and the three frequency bands of analysis.LF: low frequency; MF: medium frequency; HF: high frequency.

Figure 8 .
Figure 8. Variation of RMS vibration acceleration with radial loads at a constant speed of 1000 RPM for different greases.LF: low frequency; MF: medium frequency; HF: high frequency.

Figure 9 .
Figure 9. Variation of RMS vibration acceleration with speed at a constant radial load of 5 kg for different greases.LF: low frequency; MF: medium frequency; HF: high frequency.

Figure 12 .
Figure 12.A comparison of: (a) RMS vibration and (b) SPM carpet value of different greases at 5 kg radial load and 1000 RPM speed.
Figure 12(b) compares the SPM carpet values of different greases operating at 5 kg radial load and 1000 RPM speed on a rolling bearing.The lithium grease displays the minimum carpet value, whereas the calcium grease displays the maximum; the formulated greases display intermediate values.The presence of distinct film thicknesses for these three kinds of greases explains the different carpet values.The greases have different microstructures (Figure 4(d), (f), and (h)), which indicates that they have different binding forces between the base oil and the thickener in each case and, thus, the different film thicknesses 59,60 at the contact zone.

Table 1 .
Characteristics of soybean oil.

Table 2 .
Characteristics of commercial greases used as the benchmark and formulated grease (without gum).

Table 3 .
Nomenclature and characteristics of formulated greases, along with the proportion of ingredients.
Figure 2. Flow chart illustrating the development methodology of greases containing polysaccharide gums.

Table 4 .
Specifications of the test bearing.

Table 5 .
Design of experiments to evaluate the grease performance in a rolling bearing.