Effect of molecular chain length on the tribological properties of two diazomethine functionalised molecules as efficient surface protective lubricant additive: experimental and in silico investigation

Abstract In the boundary lubrication regime, the addition of long as well as straight alkyl chain containing additive exhibit a serendipitous impact on protecting various metallic surfaces or machinery components from friction and subsequent wear. In quest of environment-friendly, proficient friction-reducing, surface protective and cost-effective lubricant additives; herein, two diazomethine functionalised long-chain consisting organic molecules, namely (3E)-N-((E)-2-(octadecylimino)ethylidene)octadecan-1-amine (ODE) and (3E)-N-((E)-2-(dodecylimino)ethylidene)dodecan-1-amine (DDE) were synthesized through a one-pot condensation reaction. The surface protective film-forming, as well as wear and friction reducing properties of these organic additives on steel balls within paraffin oil (PO), were thoroughly explored using a four-ball tester with variation in time, load and speed. The addition of these additives showed a remarkable reduction of coefficient of friction (COF) and wear amount in compared to the pure base oil. ODE exhibited better performance with a ∼62% reduction in COF and ∼23% reduction in wear amount for steels in contact. Electronic level analysis for elucidating the tribofilm formation capability of the additives was rationalized using computational approaches such as density functional theory (DFT) and Fukui indices. Additionally, the molecular dynamics (MD) simulation which is an efficient computational approach was used to explore the spontaneous adsorption insight of additives; and also radial distribution function (RDF) were analysed to impressively explore the molecular-level interactions and adsorption mechanism of the additives with the metal atoms. Thereby, the mechanism of surface adhesive tribofilm formation and its metal protection capability has been explained in a comprehensive manner. Highlights Diazomethine functionalised organic molecules were synthesized with a variation of aliphatic chain length. Long aliphatic chain containing ODE additive showed enhanced surface protective performances in paraffin oil exhibiting lower COF value in comparison to DDE. Possible formation of ODE-based efficient tribofilm in metal-solution interface which exhibits enhanced surface adsorption property on the metal surface. Substantial decrease in wear upon addition of ODE in the base oil, which was confirmed by FE-SEM and 3D surface profilometer study. MD simulation has been used to investigate the adsorption competence of ODE and DDE on the iron surface.

In the boundary lubrication regime, the addition of long as well as straight alkyl chain containing additive exhibit a serendipitous impact on protecting various metallic surfaces or machinery components from friction and subsequent wear. In quest of environment-friendly, proficient friction-reducing, surface protective and cost-effective lubricant additives; herein, two diazomethine functionalised long-chain consisting organic molecules, namely (3E)-N-((E)-2-(octadecylimino)ethylidene)octadecan-1-amine (ODE) and (3E)-N-((E)-2-(dodecylimino)ethylidene)dodecan-1-amine (DDE) were synthesized through a one-pot condensation reaction. The surface protective film-forming, as well as wear and friction reducing properties of these organic additives on steel balls within paraffin oil (PO), were thoroughly explored using a four-ball tester with variation in time, load and speed. The addition of these additives showed a remarkable reduction of coefficient of friction (COF) and wear amount in compared to the pure base oil. ODE exhibited better performance with a $62% reduction in COF and $23% reduction in wear amount for steels in contact. Electronic level analysis for elucidating the tribofilm formation capability of the additives was rationalized using computational approaches such as density functional theory (DFT) and Fukui indices. Additionally, the molecular dynamics (MD) simulation which is an efficient computational approach was used to explore the spontaneous adsorption insight of additives; and also radial distribution function (RDF) were analysed to impressively explore the molecular-

Introduction
With ever-increasing technological innovations from micro/nanoelectromechanical systems (MEMS/NEMS) to industrial machinery equipment, reduction of friction and protection against wear has become a crucial factor for improving the durability and efficiency of a mechanical system [1][2][3][4]. Approximately 23% of global energy consumption takes place owing to the tribological interactions; whereas, around 20% of energy losses are incurred to conquer the frictional force and the remaining 3% to compensate the wear [5]. Lubricants play a pivotal role in minimising as well as optimising various tribological properties viz. friction, wear, etc. and also reduce energy consumption by forming a protective layer between the interfaces of two mutually sliding surfaces [6,7]. Liquid lubricants are most frequently used in various mechanical components. Lubricating oil is primarily composed of either pure base oil or a package of additives along with base oil. An additive induces added functionality to the base oil towards the improvement of tribological performances [8,9].
Another concerning issue is that more than 60% of unconsumed lubricant goes to the environment in different ways. Various types of organic and/or inorganic pollutants present in the additives cause severe health issues. In this relevance, the development of high-performance, cost-effective, environment-friendly lubricant additives and the minimization of their dosage have been a long-standing research endeavour. Over the decades, Zinc dialkyldithiophosphates (ZDDP) and its derivatives were profoundly used as anti-wear and friction-reducing agents [10,11]. In spite of the superiority of ZDDP or related compounds, the presence of zinc, sulphur and phosphorus can poison the catalytic converters and shorten the emission system lifetime of an engine. In the context of adverse health and hazardous nature of ZDDP, much effort has been levied towards the replacement of ZDDP either partially through the synergistic effect of organomolybdenum, organic borate ester, etc. or fully by the use of sulphur and phosphorus less additive [12][13][14]. There is scanty literature that reports the use of organic molecules containing hetero-atoms (like nitrogen, oxygen, etc.) as alternative lubricant additives [15][16][17].
With the aim of exploring more efficient organic additive molecules, it has been focussed on the exploration of Schiff base molecules as lubricant additives. Owing to the possession of azomethine linkage (-CH¼N-), these compounds exhibit excellent adhesion on the metal substrates which form a thin protective film. This protective organic film possesses anti-wear properties, load-bearing abilities, thermal stability and anti-corrosion properties that add extra functionality to the additive towards the formation of an efficient tribofilm on the targeted metal surfaces [18][19][20][21]. From the existing literature reports, it was observed that the molecular chain length and the molecular weight of the saturated linear hydrocarbons and their related acids and alcohols decrease the coefficient of friction in the boundary lubrication regime. With the increase in chain length, the coefficient of friction linearly decreases for the hydrocarbon compounds, i.e. from C5 to C24 for alkanes, from C1 to C16 for alcohols and from C1 to C18 for acids. In most cases, paraffin oil (PO) is used as lubricant base oil, which is composed of long-chain hydrocarbon compounds [22][23][24][25]. Keeping consistent with these factors, here two different additives have been synthesized, which are consisting of octadecyl and dodecyl alkyl chain. The aim of the present research work is to provide a deep insight underlying the efficient adhesion of organic film of two Schiff base compounds containing long hydrocarbon chain, namely, (3E)-N-((E)-2-(octadecylimino)ethylidene)octadecan-1-amine (ODE) and (3E)-N-((E)-2-(dodecylimino)ethylidene)dodecan-1-amine (DDE) as highperformance anti-friction and anti-wear lubricant additive. The lubrication performance of both the ODE and DDE additive has been evaluated by a four-ball tester instrument. The concentration of both the additives has been optimized and further tribological studies have been carried out with the variation of load, run time and rotating speed to understand the performance of the additives in a deeper way. Furthermore, the theoretical, as well as simulation approaches, have been used to explore the mechanism of adsorption or adhesion of additives onto the metal's surface. The electronic properties of the additives have been explored using density functional theory (DFT). Furthermore, In order to explore the insight of the adhesion, visualisation of the adsorbed configuration of the organic additives on the targeted metal surface and to explain the film-forming mechanism, the molecular dynamics (MD) simulations have been performed. It helped in deciphering the film formation and explaining the frictional behaviour of the developed lubricant additives at the sliding interface. Additionally, radial distribution function (RDF) analysis has been performed to envisage whether physisorption or chemisorption is the main driving force facilitating the adsorption of the additive on the targeted surface leading to organic tribofilm formation.

Materials
The paraffin oil (herein base oil) and the requisite solvents such as methanol and acetone were procured from Merck, India and used without additional purification. Octadecan-1-amine, dodecan-1-amine and oxalaldehyde were purchased from Sigma-Aldrich (now Merck, India). The paraffin oil having specific gravity of 0.84 at 25 C and kinematic viscosity 30 cSt at 40 C were used for performing the experimental works. The steel balls (AISI 52100) having a diameter of 12.7 mm were taken for tribological analysis.

Synthetic procedure of the additives
Two diazomethine functionalised additive molecules, namely, (3E)-N-((E)-2-(octadecylimino)ethylidene)octadecan-1-amine (ODE) and (3E)-N-((E)-2-(dodecylimino)ethylidene)dodecan-1-amine (DDE) were synthesised by uni-step condensation of an aliphatic amine with a dialdehyde in a 2:1 stoichiometric ratio. Initially, 2 mM of octadecan-1-amine was added into 10 mL methanol taken in a round-bottom flask with vigorous stirring using a magnetic stirrer for 20 min until complete dissolution of the amine was attained. Subsequently, 1 mM of oxalaldehyde was added dropwise, and the reaction was allowed to undergo constant stirring for 12 h at ambient conditions. The product, namely, ODE was obtained, which was then washed thoroughly with ether after the completion of the reaction. Similarly, DDE was synthesized in a similar way by using dodecan-1-amine as amine. Both these products were dried and collected into a sample container and stored in a vacuum desiccator. The schematic representation of the chemical synthesis of ODE and DDE additive is represented in Scheme 1.

Characterization of ODE and DDE
The synthesized additive molecules, ODE and DDE were characterized using FT-IR spectroscopy (Spectrum 100 FT-IR spectrometer, Perkin Elmer), ESI mass spectroscopy (expression CMS, Advion) and Nuclear Magnetic Resonance (NMR) spectroscopy. All 1 H-NMR spectra were recorded using a 500 MHz Bruker Avance spectrometer at room temperature. The samples were run using CDCl 3 as a solvent. Chemical shift values are reported in ppm and for the residual solvent CDCl 3 is 7.29 ppm, multiplicity is reported as follows: s ¼ singlet, d ¼ doublet, t ¼ triplet, q ¼ quartet and m ¼ multiplet and coupling constant J expressed in Hz. The thermal stability of the synthesized additives was analysed by a thermogravimetric analyser (NETZSCH TG 209F3, with a temperature ramp of 10 C/min).

Dispersions stability measurements
The average dimensions of the additive molecules in base oil were monitored by dynamic light scattering (DLS) technique. DLS experiments were carried out by Malvern Nano ZS instrument (k ¼ 633 nm, operating power 4 mW, at h ¼ 173 ) using a disposable cuvette at 25 C. The refractive index n (1.480), the viscosity g (34.88 cP) of base oil was taken into consideration. The measurement was always made at a fixed volume of the samples in the cuvette.

Sampling methodology
Four different sets of lubricant blends were prepared with the addition of varying concentrations (0.016, 0.023, 0.035 and 0.042 w/v%) of ODE and DDE separately with paraffin oil. Sonication was performed for the 1800 s prior to each tribological experiment.

Tribological tests
The tribological tests for different lubricant blends of ODE and DDE (0.016, 0.023, 0.035 and 0.042 w/v%) in PO were performed using a four-ball tester instrument (Ducom, India) following the ASTM D4172 standard procedure i.e. applied load of 392 N, 1200 rpm rotating speed, 75 C temperature and 3600s run time duration [26]. Four numbers of steel balls, each having a 12.7 mm diameter, were used to carry out tribological experimentations. WINDCOM software equipped with the four-ball tester instrument was used for measuring the COF values. It was observed that 0.035 w/v% of both ODE and DDE in PO showed the lowest COF compared to other concentrations. Henceforth, the optimized 0.035 w/v% additive concentration in PO was used for further investigation of the tribological properties in various applied load, rotating speed and experiment time durations. The applied loads were varied as 294, 392, 490 and 588 N by keeping the temperature fixed at 75 C and run time duration for 3600 s. Secondly, the run time duration was varied as 900, 1800, 3600 and 4500 s with 392 N applied load and was 1200 rpm at 75 C. In the third set of experiments, the rotational speed was varied as 600, 900, 1200 and 1500 rpm with 392 N applied load and at 75 C. After completion of four-ball test experiments, the magnified image analyses of these samples were performed to calculate wear scar diameter (WSD), mean wear scar diameter (MWSD) and mean wear scar volume (MWSV).

Surface morphological analysis
The steel balls were taken out from the four-ball tester after performing experimentation with different concentrations (viz, 0.016, 0.023, 0.035 and 0.042 w/v%) of ODE and DDE additive in PO. The wear scar on each worn-out steel ball was identified and its surface morphological alterations were studied using field emission scanning electron microscope (FE-SEM) and energy dispersive X-ray analysis (EDS) technique. The three-dimensional (3D) surface profile nature of these studied samples was analysed using a non-contact optical 3D surface profilometer (Talysurf CCI-lite, Taylor-Hobson, UK, resolution 0.1 Å) according to ISO 4287:1997 standard [27,28].

Theoretical methods and computational details
2.8.1. Quantum chemical aspects DFT calculations were performed employing the ORCA programme. The geometry optimizations and electronic property derivations have been performed with hybrid B3LYP/G functional. All-electron Gaussian basis set has been utilized here. A Triple-f quality basis set TZV(P) has been used in association with one set of polarization functions on N and O atoms. On the other hand, a polarised split valence SV(P) basis set of double-f quality in the valence region and a polarising set of d functions for the non-hydrogen atoms. The self-consistent field (SCF) was converged with a density change of 10 À7 Eh, energy: 10 À8 Eh and maximum element of DIIS (Direct Inversion of/in the Iterative Subspace) error vector: 10 À7 [29][30][31].
The frontier molecular orbital such as highest occupied molecular orbital (HOMO) and lowest unoccupied molecular orbital (LUMO) was determined. The overall reactivities of the molecules were analyzed by evaluating the energy of the highest occupied molecular orbital (E HOMO ), and the lowest unoccupied molecular orbital (E LUMO ), the absolute electronegativity (v), hardness (g). v and g of the additives were calculated as per the Pearson equation:

Local reactive sites analysis
The analysis of local reactive sites of the ODE and DDE additive was performed using the DMol 3 module in Material studio TM . The GGA and BLYP exchange-correlation functionals were employed along with double numerical polarisation (DNP) basis set as reported elsewhere [32][33][34]. The local reactivity of ODE and DDE additives were determined from the Fukui function (f k ) as expressed as the first derivative of electron density q r ! ð Þ for N electron(s) at constant external potential, vð r ! Þ as presented in Equation 3.
Furthermore, the finite difference approximation has been utilised to determine the Fukui functions designated for nucleophilic (f þ k ) and electrophilic attack (f À k ), as shown in Equations 4, 5, respectively: where, the q N k , q Nþ1 k and q NÀ1 k represent gross charges on neutral, cationic and anionic charges for k th atom, respectively.

Adsorption analysis using MD simulation
MD simulation has been exhaustively used as an effective tool for investigating the interaction of organic molecules on the surface of metal [35]. In the present work, the Forcite module present in Material Studio TM software (version 17.1.0.48) was used to evaluate the interaction behaviour of the additive on the constructed Fe (110) layer [36][37][38]. The simulation box is comprised of an iron slab (as the bottom layer), additive molecule and base oil in 1:18 ratio (the intermediate layer) and the topmost layer that remains vacuum. A linear chain alkanes consists of 56 atoms [C 18 H 38 ] modelled as paraffin oil [39]. The overall dimension of the modelled box was 59.67 Â 59.67 Â 108.07 Å 3 . Fe (1 1 0) plane with ten layers was chosen for revealing the adsorption pathway of lubricant additive molecule on the steel surface, which was relaxed by decreasing the energy to a minimum. Eight layers from the bottom side were frozen at the fixed Cartesian position before computation. Initially, the geometry optimization of the constructed box was performed for energy minimization of the overall system. Group-based cutoff and Ewald schemes were adopted for the van der Waals interactions and electrostatic interactions respectively. The velocity Verlet integrator was applied for solving Newton's motion equation with a time step of 1 fs. To obtain the global energy minimum, the Quench task approach was employed. Quench task calculation was performed at 348 K temperature using NVT thermodynamic ensemble, Nos e thermostat with a time step of 1.0 fs and a total simulation time of 200 ps. Compared with reactive force fields, herein, the COMPASS II force field was employed for geometry optimization and molecular dynamics simulation [40][41][42][43][44]. After the accomplishment of the simulation, the actual configuration of ODE and DDE additive adsorb on Fe (110) plane has been obtained. Subsequently, the interaction energy (E interaction ) and binding energy (E binding ) were determined. The energy obtained due to the interactions of ODE or DDE additive with the frozen surface of Fe (110) is expressed as E interaction , which is evaluated by the expression (6) where the simulation system's total energy is denoted by E total ; the energy of Fe (110) plane i.e. E Fe 110 ð Þþ PO is the total energy of Fe (110) surface and the PO in the absence of additive, (considered to be zero since the iron layer was kept fixed), and the energy of the molecule was designated as E additive : The energy of binding (E binding ) is the negative of the energy of interaction (E interaction ) Furthermore, the analysis of distances between the two physically and/or chemically interacting species is very important for explaining the mechanism of interaction. In this regard, RDF analysis is the most effective and efficient technique in doing so [45][46][47]. The RDF represented as g(r) is defined by Hansen and McDonald as follows: where, <q B > local is the particle density of B averaged over the entire shell around particle A.
If the initial peaks for interacting species obtained from RDF analysis lie within 1 Å to 3.2 Å, then it is assumed that the consequent peaks are attributed to the electron sharing between the interacting species leading to chemisorptions. Alternatively, the consequent peaks greater than 3.2 Å are attributed to physisorption [45][46][47]. Adsorption mechanistic pathway of additives onto Fe (110) plane has been evaluated through RDF analysis from MD simulation results.

Characterizations
FT-IR spectrum (of the synthesised additives viz. ODE and DDE are shown and compared with their precursors in Figure S1 and Figure S2 respectively. The appearance of characteristic CH¼N peak at 1634 cm À1 and disappearance of the peak of N-H and C¼O/C-O stretching for primary amine and aldehyde functional group is attributed to the formation of azomethine linkage between the amine group of octadecan-1-amine and the carbonyl group of oxalaldehyde yielding ODE additive. The octadecan-1-amine retains a peak at 3337 cm À1 for the primary amine functional group (N-H stretching) and its characteristic peaks at 2923 and 2850 cm À1 arising because of aliphatic hydrocarbon moiety (C-H stretching). The C¼O and C-O stretching for the aldehyde functional group in oxalaldehyde appeared at 1628 cm À1 and 1064 cm À1 , respectively [48,49]. Similarly, for the DDE additive, the functional moieties are the same however the chain length is different. So the appearance of characteristic CH¼N peak at 1632 cm À1 and disappearance of the peak of N-H and C¼O/C-O stretching for primary amine and aldehyde functional group strengthen the DDE additive formation [50]. The ESI-mass spectroscopy peak at m/z 561.9 and m/z 392.5 (vide Figure S3) confirms the formation of the ODE and DDE additive respectively. 1 H-NMR measurement was performed to validate the chemical structure of the ODE and DDE. The 1 H-NMR spectra of ODE are given in Figure S4. The ODE shows various peaks at: d 7.94 (s, 2H), 3.59 (t, J ¼ 7.0 Hz, 4H), 1.74-1.65 (m, 6H), 1.28 (s, 58H), 0.90 (t, J ¼ 6.9 Hz, 6H). The 1 H-NMR spectra of DDE are given in Figure S5. The obtained peaks are as follows: d 7.93 (s, 2H), 3.58 (t, J ¼ 7.0 Hz, 4H), 1.69 (t, J ¼ 7.1 Hz, 4H), 1.28 (s, 36H), 0.90 (t, J ¼ 6.8 Hz, 6H) [51,52].
Comparative thermal stability of ODE and DDE additives has also been studied and presented in Figure 1. From the thermal stability plot, it has been observed that ODE has better thermal stability than DDE. In the case of ODE up to 200 C, no weight loss was observed, degradation started above 200 C and $40% weight loss occurred up to 400 C. Finally, degradation of ODE was completed at $500 C with almost no residual mass. Adequate thermal stability suggested that the presently explored ODE can be utilized as an effective lubricant additive.

Dispersion stability measurements
The dispersion of ODE, as well as DDE in base oil, was studied using dispersion stability measurements. It showed that the particle diameters were in the range of 100-1000 nm (vide Figure S6). The dispersion of the additives in the PO is considered to be fairly stable. However, further addition of the additives beyond its optimized concentration causes the intensity band to slowly drift to the higher values implying slow aggregation of the particle [53].

Anti-wear properties
The lubricant formulations comprising the synthesized lubricant additive (0.016, 0.023, 0.035 and 0.042 w/v%) in PO were made and its tribological performances were analysed using a four-ball tester according to ASTM D4172 standard. The tests executed to determine the tribological properties of the synthesised additives in PO were performed thrice. The measure of the degree of friction and wear under sliding contact such as COF, MWSD and MWSV were determined. The relation between COF, MWSD and MWSV with the variation of concentration has been shown in Figures  2(a), 3 (a), 4(a) respectively. The performance of a particular additive primarily depends on its added amount in the base oil, so the optimization of the additive concentration is of utmost importance for tribological analysis. It has been revealed that the optimum concentration of newly synthesized additive molecules viz. ODE and DDE is 0.035 w/v%. Figure 2(a) depicts that the addition of ODE and DDE additive reduces the COF value significantly in PO. In the case of pure PO, the recorded COF and MWSD values are 0.081 ± 0.0005 and 0.950 ± 0.002 respectively. It has been found that, upon addition of 0.035 w/v% of additive, the COF reaches an optimum value as 0.0303 ± 0.0005 and 0.0423 ± 0.0005 for ODE and DDE respectively. The reduction percentage in COF for ODE is greater than DDE due to the presence of an extra hydrocarbon chain in ODE. Therefore, a decrease in friction was observed with the increase in the molecular chain length.
The addition of ODE or DDE additive with the PO not only decreases the COF but also reduces the MWSD value. The variation of MWSD with the additive concentration has been shown in Figure 3(a). The wear amount of steel balls decreases up to the addition of 0.035 w/v% of the additive. i.e. at optimum additive concentration, the amount of wear reaches a minimum value thereby generating less wear scar diameter. At optimum concentration the addition of ODE and DDE additive results in the MWSD values as 0.732 ± 0.0010 mm and 0.751 ± 0.0005 mm respectively. So in terms of anti-wear behaviour ODE shows better efficiency.
Another significant way to determine the efficiency of a lubricant additive is to measure MWSV which provides a more realistic measure of the wear amount. Figure 4(a) illustrates the variation of MWSV of the steel ball lubricated with pure PO and various concentrations of additives added in PO. The obtained MWSV is the lowest for ODE at optimum concentration. The calculated MWSV (vide Table S1) according to its descending order is as follows: Pure PO (58.65 ± 0.5658) > 0.035 w/v% DDE þ PO (21.19 ± 0.277) > 0.035 w/v% ODE þ PO (19.14 ± 0.1104). Therefore, it has been observed from the COF, MWSD and MWSV data that in terms of anti-wear as well as friction-reducing properties ODE shows better efficiency than DDE and pure PO. The additives form a tribofilm over the sliding steel surface and increase the separation distance of the two sliding surfaces from each other which results in a decrease in COF, MWSD and MWSV. According to the anti-wear or anti-friction efficiency of the base oil and additive are as follows: (PO þ 0.035 w/v% ODE) > (PO þ 0.035 w/v% DDE) > (pure PO).

Effect of load on the tribological properties
Both ODE and DDE show optimum COF as well as MWSD and MWSV at 0.035 w/v% concentration. Therefore, further analyses were carried out at optimized 0.035 w/v% concentration of ODE and DDE additive in PO. To explore the effect of applied load on the COF, MWSD and MWSV, the tests were performed at different applied loads 294, 392, 490, 588 N for 3600 s run time duration, at 75 C and 1200 rpm rotational speed. The variation of COF with the applied load as shown in Figure 2(b) depicts that the COF value increases with the increase of applied load. In the case of DDE, it has been found that the COF value gradually increases with an increase of load whereas in the case of ODE, the COF gradually increases up to 490 N load and after that COF value increases sharply. A sudden increase in COF implies that the tribofilm might have broken at the extreme condition. Thus, the anti-wear, as well as the anti-friction performance of the ODE additive, reduces considerably beyond 490 N load.  appreciably high during that period. As the experimentation proceeds, the additive molecules get uniformly adsorbed over the sliding surfaces. Furthermore, the plot of friction variation with time of ODE and DDE additives is depicted in Figure S7. COF and MWSD are very high when the experiments are run up to 900 s which is attributed to incomplete formation of tribofilm on the targeted metal surfaces. During the 1800 s run time, the surfaces which are partially covered by the formed tribofilm consequently results in more reduction of COF and MWSD. Again, 3600 s run time duration is the optimum run time so that the ODE additive molecule covers the whole interacting surfaces and consequently COF and MWSD reach a minimum value. Further increasing of runtime duration leads to rupturing of the formed tribofilm. As a result, the COF along with MWSD and MWSV values are increased at higher run time duration (vide Table S3).

Effect of speed on the tribological properties
The tribological properties as a result of speed variation are shown in Figure 2(d), 3(d) and 4(d). It is revealed that the optimum COF has been obtained for ODE and DDE at 1200 rpm rotating speed. It may plausibly be due to the appropriate spreading of ODE additive over the sliding surface leading to the formation of an effective tribofilm. The COF and MWSD values are higher in both conditions viz. at extremely high and low rotating speeds as shown in Figures 2(d), 3(d) respectively. Again from Figure 2(d) it is seen that the COF value reaches a minimum at an applied speed of 1200 rpm. With increasing speed beyond the optimum value (1200 rpm) the COF value increases. Therefore, from Figure 2(b,d), it can be concluded that after 392 N load and 1200 rpm speed, lubrication regimes go to the hydrodynamic region. Changes in MWSV with speed variation are present in Figure 4(d) and Table S4.

Surface morphological study
Surface morphology analysis of a worn surface gives a clear perception of the severity of the wear. After completion of the tribological test in a four-ball tester instrument, the surface morphology of the worn-out steel balls was characterized using FE-SEM, energy dispersive X-ray analysis (EDS) technique and a 3D surface profilometer. The worn surfaces in the presence of additives are smoother than the surface lubricated with base oil. The FE-SEM images of the worn-out steel balls before and after the addition of varying weight percentage of ODE and DDE in PO has been depicted in Figure 5. It is clearly visible that the wear scar diameter changes upon the addition of different concentrations of ODE. In the case of 0.035 w/v% ODE, the wear, as well as MWSD, are the lowest. Whereas, the obtained wear rate as well as MWSD, are high for other concentrations. The obtained MWSD for DDE is quite higher than ODE.
The EDS spectra analysis results are shown in Figure S8. The weight percentage content of various elements of the steel balls surface determined by EDS are also embedded in it. It can be observed from Figure S8 that the surfaces of the steel balls lubricated with pure base oil are prone to damages as evidenced from several cracks and peaks of 'N' elements are absent on these steel surfaces. On the other hand, the addition of ODE and DDE additives to PO causes less damage to the surface of the steel balls. The presence of peak of 'N' element (vide Figure S8) strongly supports the adsorption of ODE or DDE on the steel surface.
The three-dimensional topographic analysis of worn-out steel surface texture provides in-depth information regarding the damages on the surfaces. Figure 6 illustrates the 3D surface profile for the surface lubricated with the addition of ODE and DDE additive at different concentrations. The tribofilm formed by the ODE lubricant additive results in a better surface covering than DDE or pure base oil by the formation of a thin organic layer that shields the steel balls from wear. The determined surface topography values presented in Table S5 shows that the average roughness (R a ) is 0.156 mm, root means square roughness (R q ) is 0.188 mm, and an average maximum height of the profile (R z ) is 0.741 mm for 0.035 w/v% of ODE in PO, which is comparatively lower than DDE additive at the same concentration.

Theoretical exploration and simulations
3.5.1. DFT analysis DFT is one of the most appreciated theoretical methods used for determining the quantum chemical parameters such as electron density distribution of HOMO and LUMO along with its related energy values of various organic molecules. Hence, the DFT study has been performed for ODE and DDE additives. The quantum chemical parameters such as energies of frontier molecular orbitals, their corresponding energies  and Fukui indices were determined. The electrophilic attack takes place at HOMO, whereas the LUMO corresponds to the centres for nucleophilic attack. The energy of HOMO (E HOMO ) is directly related to the ionization potential. The susceptibility of the organic molecule towards electrophilic attack is expressed in terms of the E HOMO value. Unlikely, the LUMO energy (E LUMO ) indicates the electron affinity and which expresses the vulnerability of the molecule towards a nucleophilic attack.
The geometry optimized structures of the studied additives and HOMO-LUMO distributions are shown in Figure S9. It can be observed that a high electron density in both HOMO and LUMO of the additives is mostly localised on or around the vicinity of both azomethine linkage. Therefore, these moieties are the most active centres for electron transfer (either electron donation or acceptance). The obtained E HOMO and E LUMO of the synthesized additives and base oil are listed in Table 1. The energy gap (DE) values (i.e. DE ¼ E LUMO -E HOMO ) represent the interaction probability of organic molecules with metal surfaces. When, DE value decreases the adsorption performance of the organic molecule on the metal surface increases. From the DFT analysis, the DE value of ODE is comparatively lower than that of DDE suggesting it be highly capable of better adsorption. The hardness (g) and softness (r) are also two characteristic features to measure molecular stability and reactivity. The iron atom is preferentially interacts with molecules that have low g and high r value, as the soft molecule with a lower DE value can easily exchange electrons. From the obtained result (vide Table 1) it has been revealed that ODE having lower g and higher r value compared to DDE adsorb better on the iron surface. Another most important parameter is a fraction of electron transfer (DN) which is highly useful in explaining the interaction of organic molecules with metal surface atoms. Hence, DN has been determined based on Pearson's method using chemical hardness (g) and electronegativity (v) of organic molecules and iron surfaces [54][55][56]. The DN is represented by the following equation.
where, the electronegativity and hardness are denoted by v and g respectively. Here work function of Fe (110) plane i.e. / ¼ 4.82 eV/mol was used instead of v Fe for DN 110 calculation and considered as g Fe ¼ 0 [54][55][56].
For the studied molecules, ODE and DDE, the iron-additive interactions occur by transferring electrons from the additive to the metal surface and vice versa. A positive DN indicates more facile electron transfer occurs from additive molecule to metal centre. It has been observed from the obtained result (vide Table 1) the calculated DN values for these molecules are positive suggesting the occurrence of electron transfer from organic molecules to the iron surface atoms. Furthermore, DN value of ODE is  higher than that of DDE. Hence, it implies that ODE adsorbed on the iron surface better than that of DDE. Thus, it can be said that when the additive molecule approaches close to the metal substrate, electron density can be easily shared with the available vacant d-orbital of the metal substrate and may give rise to chemical bonding leading to thin film formation on the metal surface in contact. In addition, the excess electrons present in the metal d-orbital or accumulated electron density in the metals surface atoms can easily be donated back to the available p-orbitals of the azomethine linkages present in the additive molecules. In this way, the synergistic chemical bonding may be feasible which facilitates strong adsorption of additives onto metal surfaces.

Local reactive sites analysis
The organic molecules are adsorbed on the metal surface by donor-acceptor (D-A) interaction. The heteroatom(s) possessing high electronic charges can easily participate in this type of interaction. Therefore, it is obligatory to analyse the charge distribution on different donor atoms of the additive molecule. In the present investigation, the additive molecules possess azomethine linkage which is connected to long-chain aliphatic groups. Fukui indices analysis has been executed to decipher the nucleophilic and electrophilic nature of the additive molecules [28,42]. The maximum threshold value of f þ k and f À k are generally used to determine the nucleophilic and electrophilic behaviour of an organic molecule. Electron density changes are measured in terms of f þ k and f À k values according to the acceptance and donation of electrons respectively by additive molecule. Higher values of f þ k and f À k indicates respectively the high acceptance and donation capability of the molecule. The determined Fukui indices values for ODE and DDE are tabulated in Table 2. It has been observed that the N atoms of two C ¼ N bonds are the suitable centre for electron acceptance and donation. For ODE molecule, the electron acceptance centre i.e. N (19)

Adsorption capability analysis using MD simulation
From the MD simulation approach, the actual adsorption of ODE and DDE with minimum energy configuration has been obtained and presented in Figure 7. The attainment of an equilibrium state has been confirmed from the observation of energy and temperature fluctuation as shown in Figure S10. The top and side view of adsorbed additives onto the Fe (110) plane has been depicted in Figure 7; it has been observed that after the accomplishment of the simulation process, the entire modelled system has reached equilibrium which propelled the adsorption of additive onto Fe (110) surface in a horizontal fashion throughout the molecular skeleton. It suggests the occurrence of chemisorption phenomenon during the adsorption of additive onto Fe (110) plane. Furthermore, the energy due to interaction (E interaction ) between additive molecules and Fe (110) surface is determined by equation (6). Now, the E interaction energy of ODE: À1385.95 kJmol À1 and DDE: À1144.87 kJmol À1 respectively. It suggests that the adsorption of ODE additive onto the Fe (110) plane is more spontaneous than DDE. Furthermore, based on the E binding which is given by the negative magnitude of E interaction energy, it can also be said that the ODE additive possesses a strong binding propensity with the iron surface atoms than that of DDE. 0.000 0.000 C (35) 0.000 0.000 C (36) 0.000 0.000 C (37) 0.000 0.000 C (38) 0.000 0.000 C (39) 0.000 0.000 C (40) 0.001 0.001 From all these obtained results, it implies that the synthesised ODE as a lubricant additive is adsorbed in a horizontal orientation and provides more surface coverage to form an effective lubricant film that shielded the steel substrate from external wear and frictional damage.
Furthermore, the RDF analysis revealed the mechanism of adsorption of the organic additive onto the iron surface. A chemical bonding is affirmed when the first peak lies within 3.2 Å and strong Coulomb and van der Waals interactions are considered if the peak lies between 3.2 Å and 5.0 Å. Larger than 5 Å weak van der Waals force is considered. The obtained RDF plot representing g(r) vs. distances (r) in Å has been presented in Figure 8. For the adsorption of ODE additive on Fe (110) surface, the initial peaks of Fe-N (19) and Fe-N(22) curves have appeared within the distance of 3.2 Å which indicates that the nitrogen atoms chemically bind with the iron atom [44,45]. It implies that these two azomethine linkages can share their electrons with the metal surface atoms and follow chemisorptions mechanistic pathway leading to its adsorption onto Fe (110) plane. The presence of double azomethine linkages has enhanced the chemical interaction of ODE additive onto the metal surface atom. Additionally, the peaks greater than 3.5 Å has originated due to the accompanying physisorption of azomethine linkage as well as the entire ODE structure onto the iron surface. Thus, it can be said that physisorption is subsequently followed by chemisorptions of ODE additive onto metal surface take place and lead to strong tribofilm formation on the metal surface. In the case of DDE additive same RDF trend has been observed as like ODE.  additives can be easily adsorbed on the iron surface and thereby form a protective tribofilm for efficient wear reduction [57]. When the lubricant additive molecule comes closer to the metal substrate; there is a strong inclination to undergo van der Waal type of interaction which facilitates physisorption of lubricant additive onto the metal surface. The adsorption, as well as tribofilm formation, is a time taking process i.e. it takes an appreciable time to form a stable tribofilm between interfaces of the reciprocating pairs during the sliding condition. The initially physisorbed additive additionally interacts with surface atoms and is strongly adhered through chemical bonding. Now, from the molecular structure viewpoint, it can easily be said that the presence of N atom on the ODE additive leads to the formation of metal-ODE bonding. Herein, the lone pair of electrons of filled p-orbital of the nitrogen atom drifts towards the vacant d-orbital of the metal, thereby, instigating the chemisorptions phenomenon to be operative by   Steel-steel ASTM D4172 Four-ball tester Present work chemical bond formation. Again, the long-chain hydrocarbon moieties present in the additive helps to cover more surface area. The long-chain organic compounds form a tribofilm on the frictional surface to reduce the interaction or friction between two mutual sliding surfaces, just like the role of traditional organic lubricant additive. It can also form an efficient surface protective layer for reducing friction through extended surface coverage by the long chain. These double-functions may lead to higher performance of the additives in terms of anti-friction and anti-wear efficiency at a very lower concentration of organic lubricant additive, as depicted in Table 3 [13,18, 58-65].

Conclusion
Tribological properties of two diazomethine functionalised alkyl chain-containing molecules namely, ODE and DDE were synthesized, characterized, investigated. A Straight forward and facile synthetic procedure was adopted for their synthesis. Both ODE and DDE showed remarkable tribological performance at 0.035 w/v% concentration. The durability of tribological properties of the additive molecules in PO was exhaustively explored upon variation of time, load, and speed. It showed that the addition of ODE with optimum concentration resulted in a low COF value i.e. $0.03. Reaching an incredibly low COF value by adding a minuscule amount of ODE additive makes the present work unprecedented in its congener. FE-SEM and 3D surface profile studies reveal characteristics reduction in the severity of the wear. The presence of nitrogen centre, azomethine linkage along with long hydrocarbon chain facilitates adsorption as well as the formation of efficient tribofilm on the worn surface. The adsorption of these additives onto steel ball surfaces has also been ascertained by EDS. Furthermore, DFT study exhibited that the additive molecules possess highly active FMOs at the two adjacently placed azomethine linkages. These FMOs facilitate their strong adsorption on the targeted metal surface and strengthen tribofilm formation, thereby, protect the metal surface from wear. The adsorption or interaction energy values obtained from MD simulation are well corroborated with the experimental outcomes. The mechanism of interaction and subsequent adsorption of additive molecules onto metal surface atoms were elucidated through RDF analysis.
In nutshell, it can be perceived that the addition of ODE in the PO remarkably reduced the COF value to 0.030; which showed better results compared to DDE. It may be attributed to its unique geometrical structure, chain length, as well as the electronic property of the synthesized ODE additive that facilitates the spontaneous diminution of COF and wear.